Technology TransferJ
EPA/62571-89/023
Design Manual

Fi n e Po r e Ae r a t ion
Systems

-------

-------
                                      EPA/625/1-89/023
                                       September 1989
        Design Manual
Fine  Pore Aeration Systems
      U.S. Environmental Protection Agency
      Office of Research and Development
  Center for Environmental Research Information
      Risk Reduction Engineering Laboratory
            Cincinnati, OH 45268

-------
                                       Notice
This document has  been reviewed in accordance with the U.S. Environmental  Protection
Agency's peer and administrative review policies and approved for publication. Mention of trade
names or commercial products does not constitute endorsement or recommendation for use.

This document is  not intended to be a guidance or support document for a specific regulatory
program. Guidance documents  are  available from EPA and must be  consulted to address
specific regulatory issues.

-------
                                     Contents


Chapter                                                                        Page

1   INTRODUCTION AND OVERVIEW			.   1

    1,1 Historical Overview	   1
    1.2 Objectives of the Manual	 ,	   2
    1.3 References 	,	   2

2   FINE PORE DIFFUSER CHARACTERISTICS  	   3

    2.1 Introduction   	,	   3
    2.2 Types of Fine Pore Media	   3
       2.2.1 Ceramics    . . .	   3
       2.2.2 Porous Plastics  	   4
       2.2.3 Perforated Membranes	   5
    2.3 Types of Fine Pore Diffusers  	   6
       2.3.1 Plate Diffusers	   6
       2.3.2 Tube Diffusers  	   6
       2.3.3 Dome Diffusers	  10
       2.3.4 Disc Diffusers   	  10
    2.4 Diffuser Layout    	  14
       2.4,1 Plate Diffusers  	  14
       2.4.2 Tube Diffusers  	  14
       2.4.3 Disc and Dome Diffusers	  16
    2.5 Characteristics of Fine Pore Media	 .16
       2.5.1 Physical Description  	  16
       2.5.2 Dimensions	  17
       2.5.3 Weight and Specific Weight	  17
       2.5.4 Permeability   	'.	  17
       2.5.5 Perforation Pattern	  18
       2.5.6 Strength   	  18
       2.5.7 Hardness    	  18
       2.5.8 Environmental  Resistance	  18
       2.5.9 Miscellaneous  Physical Properties  	  19
       2.5.10 Oxygen Transfer Efficiency	  19
       2.5.11 Dynamic Wet Pressure	  20
       2.5.12 Bubble Release Vacuum  	  20
       2.5.13 Uniformity	  22
    2.6 Clean Water Performance  	  24
       2.6.1 Introduction  	  24
       2.6.2 Clean Water Data Base	,  25
    2.7 References   	  33

3   PROCESS WATER  PERFORMANCE  	  37

    3.1 Introduction	  37
    3.2 Factors Affecting Performance   	  40
                                          Hi

-------
                                Contents (continued)


Chapter                                                                         Page

    3,3 Diffuser Fouling   	  40
       3.3.1 Background	  40
       3.3.2 Types of Fouling   	  41
       3.3.3 Foulant Characteristics	  42
       3,3.4 Fouling Rates	  53
    3.4 Process Water Data Base	  57
       3.4.1 General Data Summary   	  57
       3.4.2 Selected Variables Affecting Process Water Performance	  58
    3.5 References	  73

4   OPERATION AND MAINTENANCE	  77

    4.1 Introduction   	  77
    4.2 Operation	  77
       4.2.1 Start-up   	  77
       4.2.2 Shutdown	  77
       4.2.3 Normal Operation	  78
    4.3 Maintenance	  85
       4.3.1 Blowers   	  85
       4.3.2 Air Systems	  85
       4.3.3 Diffusers	  85
       4.3.4 Troubleshooting	  95
    4.4 References	  95

5   SYSTEM DESIGN AND INSTALLATION	  97

    5.1 Introduction	  97
    5.2 Process and O&M Considerations	  97
       5.2.1 Process-Related Considerations 	  97
       5.2.2 O&M-Related Considerations	   100
    5.3 Process Oxygen  and Mixing Requirements	   101
       5.3.1 Process Oxygen Requirements		   101
       5.3.2 Process Mixing Requirements	   113
    5.4 Air Diffusion System	   115
       5.4.1 Diffuser Selection  	   115
       5.4.2 Basin Arrangement	   118
       5.4.3 Airflow Distribution  	   118
       5.4,4 Diffuser Cleaning and Maintenance	   119
       5.4.5 Diffuser installation	   120
       5.4.6 Specifications	 .   121
       5.4.7 Retrofit Considerations	 . .	,	   121
       5.4.8 Air Diffusion System Design Example  	   122
       5,4.9 Flexibility of Design  	   122
    5.5 Air Supply System	   122
       5.5.1 Air Piping	   132
       5.5.2 Air Filtration	   134
       5.5.3 Blowers   	   134
       5.5.4 Design and Installation	   136
       5.5.5 Retrofit Considerations	   137
       5.5.6 Air Supply System Design Example	   138
    5.6 Summary of Aeration System Design Procedure   	   138
       5.6.1 Outline of Approach   	   143
       5.6.2 Steps in Design	   143
    5.7 References	   146
                                          IV

-------
                               Contents (continued)


    Chapter                                                                     Page

6   AERATION CONTROL		   149

    6.1 Introduction   	,		   149
    6.2 Benefits of Aeration Control  .  ,	,	   149
       6.2.1 Process Implications	   149
       6.2.2 Economic Considerations	   149
    6.3 Control Strategy Development	, . .	   151
       6.3.1 Degree of Control   	   151
       6.3.2 Control Systems	   152
       6.3.3 DO Control Strategies	   154
    6.4 Control System Components		   156
       6.4.1 Instrumentation   	   156
       6.4.2 Final Control Elements	   158
       6.4.3 Controller	   159
       6.4.4 Software		   159
    6.5 Aeration Control Example   .	   160
       6.5.1 Air Delivery Control	   160
       6.5.2 Air Distribution Control	   162
       6.5.3 DO Probe Location and DO Set-Point	   168
    6.6 Experiences with Automated Aeration Control	   169
       6.6.1 Piscataway, MD   	,	   169
       6.6.2 Madison, Wl	   171
    6.7 Summary	   173
    6,8 References	   174

7   ECONOMIC ANALYSIS	   177
                                                              \
    7.1 Introduction	   177
    7.2 Cost Components	   177
    7.3 Economic Analysis Procedure	   177
       7.3.1 Calculate Initial Costs	   179
       7.3.2 Calculate Energy Costs	   179
       7.3.3 Calculate Maintenance Costs	   181
       7.3.4 Calculate Diffuser Cleaning Costs	   181
       7.3.5 Calculate Replacement Costs	   181
       7.3.6 Calculate Total Present Worth  Cost   	   181
       7.3.7 Determine Lowest Total Present Worth Cost  	   181
    7.4 Sample  Desktop Economic Analysis	   181
       7.4.1 Fine Pore System Design	   182
       7.4.2 Coarse Bubble System Design	   183
       7.4.3 Comparison of Present Worth Costs  	   184
       7.4.4 Sensitivity Analysis  	   184
       7.4.5 Retrofit Comparison  	   184
    7.5 Lotus Spreadsheet	 , ,	   184
    7.6 Compendium of Empirical Cost Data	   186
       7.6.1 Basin Cleaning and Preparation Costs  	   186
       7.6.2 Diffuser Costs	   186
       7.6.3 Air Filtration Costs	   187
       7.6.4 Blower Costs	 .   188
       7.6.5 Ceramic  Diffuser Gas Cleaning Costs  	   192
       7.6.6 Other Diffuser Cleaning Methods and O&M Costs   	   194
       7.6,7 Power Charges	   198
    7.7 References	   202

-------
                               Contents (continued)


Chapter                                                                       Page

8   CASE HISTORIES  	   205

    8.1 Introduction	   205
    8.2 Performance Evaluation by Off-Gas Testing   	   207
       8.2.1 Frankenmuth Wastewater Treatment Plant   	   207
       8.2.2 Glastonbury Water Pollution Control Plant  	   212
       8.2.3 Green Bay Wastewater Treatment Plant  . .	   217
       8.2.4 Hartford Water Pollution Control Plant   	   225
       8.2.5 Jones Island Wastewater Treatment Plant  	   235
       8.2.6 Nine Springs Wastewater Treatment Plant   	   239
       8.2.7 Ridgewood Wastewater Treatment Plant    	   247
       8.2.8 Whittier Narrows Water Reclamation Plant   	   254
    8.3 Performance Evaluation by Means Other Than Off-Gas Testing  	   260
       8.3.1 Cleveland Wastewater Treatment Plant	   260
       8.3.2 Plymouth Wastewater Treatment Plant  	   262
       8.3.3 Renton Wastewater Treatment Plant  	   265
       8.3.4 Ripon Wastewater Treatment Plant 	   267
       8.3.5 Saukville Wastewater Treatment Plant  	   269
    8.4 Sources of Information   	   271

APPENDIX A  Abstracts of Contractor Studies Completed Under
                 EPA/ASCE Fine Pore Aeration Project  	   273

APPENDIX B  Selected Diffuser Characterization and Cleaning Methods	   275

APPENDIX C  Selected Physical and Chemical Tables and Graphs 	   285

APPENDIX D  Economic Analysis Spreadsheet  	   293

APPENDIX E  Symbols, Terms, and Acronyms Used in this Manual  	   301

APPENDIX F  Conversion Factors 	   305
                                         VI

-------
                                        Figures


Number                                                           ,               Page

2-1    Typical ceramic plate diffusers  	>.	,	  7
2-1    (continued)	:	  8
2-2    Typical ceramic or rigid porous plastic tube diffuser   '.	  8
2-3    Typical perforated membrane tube diffusers	  9
2-4    Typical nonrigid porous plastic tube diffuser	   10
2-5    Typical ceramic dome diffusers	   11
2-6    Typical ceramic and rigid porous plastic disc diffusers	   12
2-7    Typical perforated membrane disc diffusers  .'	  .	   13
2-7    (continued)	   14
2-7    (continued)		   14
2-7    (continued)	 . . ....   15
2-8    Typical diffuser layouts   	   15
2-9    On-line device for monitoring DWP of fine pore diffusers	 .   21
2-10   Removable test header	   22
2-11   Apparatus for measuring  DWP in the laboratory  	   23
2-12   BRV relationships  ,	: ,	 .	   23
2-13   Effect  of diffuser submergence on C*w20 for three diffuser types  	   26
2-14   C*o,20 vs- diffuser submergence for perforated membrane disc
           and tube diffusers	   26
2-15   Effect  of unit airflow rate  on SOTE for fine pore tube diffusers	   29
2-16   Effect  of unit airflow rate  on SOTE for ceramic dome/disc diffusers  	   29
2-17   Effect  of unit airflow rate  on SOTE for rigid porous plastic disc diffusers .  . .	   30
2-18   Effect  of diffuser density on SOTE for ceramic disc/dome
           grid configurations	 .  .	   30
2-19   Effect  of unit airflow rate  on SOTE for perforated membrane
           disc diffusers	   31
2-20   Clean  water test data - Monroe, Wl 	   31
2-21   Effect  of water depth on SOTE for three diffuser types	   32
2-22   Effect  of water depth on SAE for three diffuser types  	   32
3-1    Hypothetical fouling patterns	   39
3-2    Linear fouling factor model  	   40
3-3    Schematic structure of Type I fouling	   42
3-4    Idealized plot showing effects of Type I fouling on DWP and OTE  	   42
3-5    Schematic structure of Type II fouling	   43
3-6    Idealized plot showing effects of Type II fouling on DWP and  OTE  	   43
3-7    F vs. foulant accumulation - Interplant Fouling Study   	   45
3-8    Relationship between BRV and F - Interplant Fouling Study	   45
3-9    F vs. DWP:BRV - Interplant Fouling Study  	   46
3-10   Progression of foulant accumulation - Interplant Fouling Study	   47
3-11   aF(SOTE) vs. time in service - Monroe, Wl  . . .	   48
3-12   aF(SOTE) vs. tank length before and  after load reduction - Madison, Wl   	   48
3-13   DWP fouling rates for eight wastewater treatment
           plants - Interplant Fouling Study	   54
3-14   aF(SOTE) vs. time in service for Basin 5 - Frankenmuth, Ml	   56
3-15   aF(SOTE) and airflow rate vs. time since initial liquid acid cleaning
           for Basin 1 - Whittier  Narrows, CA	   57
                                           VII

-------
                                Figures (continued)


Number                                                                        Page

3-16   EF vs. time in service - Green Bay, Wl	,	   58
3-17a  24-hr variations in aF and aF(SOTE) - Ridgewood, NJ  	   64
3-17b  Variations in aF, aF(SOTE), and COD - Whittier Narrows, CA  	........   64
3-17e  Variations in aF, aF(SOTE), and TOG at influent end of
          aeration tank - Madison, Wl	   65
3-18   aF(SOTE) and SOTE vs. applied airflow rate for
          ceramic disc diffusers - Monroe, Wl	   66
3-19   Mean oF(SOTE) vs. airflow rate per unit area of diffuser media  	•'. .   66
3-20   aF(SOTE) vs. time in service for ceramic disc diffusers
          in first pass - Madison, Wi	 .   67
3-21   aF vs. organic loading for selected plants in the Minneapolis/St. Paul area  	   67
3-22   aF(SOTE) vs. MLVSS for ceramic disc diffusers - Whittier Narrows, CA	. .   68
3-23   aF(SOTE) vs. SRT for ceramic diffuser facilities	'. .   68
3-24   aF(SOTE) vs. tank length for  plug flow and step feed
          aeration systems -  Madison, Wl	 .	   69
3-25   aF vs. time in service for three passes of a plug flow
          aeration system - Madison, Wl East Plant	   69
3-26   aF vs. time in service for three passes of a step feed
          aeration system - Hartford, CT   	   69
3-27   aF vs. basin distance for plug flow aeration
          system - Whittier Narrows, CA	   71
3-28   aF(SOTE) and  aF vs. basin position for step feed
          aeration systems -  Monroe, Wl	   71
3-29   aF and aF(SOTE) vs. basin distance for plug flow
          aeration system - Ridgewood,  NJ	   72
3-30   aF(SOTE) vs. basin distance  for plug flow
          aeration system - Milwaukee, Wl Jones Island East Plant	   72
3-31   aF(SOTE) vs. basin distance  for step feed
          aeration system - Milwaukee, Wl South Shore Plant	   72
4-1    Trend charts for aeration system monitoring,	   83
4-2    Idealized plot of optimum cleaning frequency
          to minimize  power and cleaning costs		   90
4-3    Power ratio vs. time for ADWP = +2.5 in w.g./month	   91
4-4    Schematic of Green Bay wastewater treatment plant   	,..,'..   92
4-5    EF vs. time in service for ceramic disc diffusers - Green Bay, Wl  	'. . i .  . . .   92
4-6    Power cost vs. time for ceramic disc diffusers - Green Bay, Wl  	>	   93
4-7    Effects of cleaning frequency on annual average
          F (if -15 percent/month)	 .   93
4-8    Cost tradeoff analysis for determining cleaning frequency
          based on fouling patterns  shown in Figure 4-7	   94
4-9    Cost tradeoff analysis for fp = 1 percent and 5 percent/month	   94
5-1    Schematic of a fine pore aeration system	   98
5-2    Representative (commonly-used) biological wastewater
          treatment systems using fine pore aeration   	   9B
S-3    Effect of  process loading on carbonaceous oxygen requirement  	,. .   102
5-4    Example  oxygen consumption ratios for carbonaceous oxygen demand	   103
5-5    Nitrogen  metabolism in  nitrifying activated sludge systems	   106
5-6    Example  diurnal BODg loading for a municipal wastewater
          treatment plant	   111
5-7    Activated sludge system for Example 5-8   	   113
5-8    Spatial variation in process oxygen requirements
          along the length of a biological reactor	   113
S-9    Manufacturer's SOTE data for Design Example 5-10   	   132
                                         VIII

-------
                                Figures (continued)
Number                                                                         Page

5-10   General arrangement of diffusers in in-tank air piping
           for Design Example 5-10   . . . .	   133
5-11   Air  supply system schematic   	.-. .	,....: . .-. %- r: ; .'. . :   133
5-12   Types of compressors and blowers	   135
5-13   General operating characteristics of  blowers  	,	   135
5-14   General arrangement of process air piping for Design Example 5-11  	   144
6-1     Comparative DO profiles for automated control vs. manual operation  	   150
6-2    Comparative energy eomsumption for automated control
           vs. manual operation	   151
6-3    Block diagram for feedback control loop	   153
6-4    Block diagram for feedforward -feedback control loop  	   153
6-5    Two-stage DO control system for a compartmentalized
           plug flow aeration train	   155
6-6    Block diagram of a self-tuning regulator	   155
6-7    Blower inlet guide vane control schematic	   161
6-8    Blower control schematic	,	   162
6-9    Low-complexity control schematic	   164
6-10   Moderate-complexity control schematic  	   166
6-11   High-complexity control schematic   	.	   167
6-12   SignificanceofadditionalDOmeasuringprob.es
           on interpolation of  DO profile . . . .	   168
6-13   Manual control of DO at  Piscataway	   170
6-14   Automated control of DO at Piscataway  . .	   171
6-15   Plant schematic for Nine Springs wastewater
           treatment plant - Madison, Wl	,..:....	   172
7-1     Present worth operating costs of example fine pore aeration system
           for Case 3 fouling  rate	   189
7-2    Optimal cleaning intervals and costs for example fine pore aeration system
           for alternative fouling rates  .	   190
7-3    Present worth costs generated by economic analysis spreadsheet - example
           fine pore aeration system: Case 3 fouling rate	   194
7-4    Monthly operating costs generated by economic analysis spreadsheet
           during first 5 years of operation - example fine pore
           aeration system: Case 3 fouling rate   	   195
7-5    Monthly airflow generated by  economic analysis spreadsheet
           during first 5 years of operation - example fine pore aeration system
            Case 3 fouling rate  	•-••-,	   196
7-6    Blower costs as a function of capacity   	,	'. . .   199
8-1     Aeration tank arrangement - Frankenmuth, Ml	   208
8-2    Aeration tank schematic  - Glastonbury, CT  	   213
8-3    Treatment plant schematic - Green Bay, Wl   ,	   219
8-4    OTE (off-gas method) vs. time - Green Bay, Wl	 .   222
8-5    Secondary treatment process schematic - Hartford, CT  	   226
8-6    OTE characteristics -  Hartford, CT	 .	   229
8-7    Typical diffuser layout for one aeration tank - Hartford, CT  .	   230
8-8    Oxygen transfer performance for Aeration Tank 2 - Hartford, CT 	   232
8-9    Power consumption profile for Aeration Tank 2 - Hartford, CT . . . ,'	   234
8-10   Plant layout - Madison, Wl	 .   240
8-11   aF(SOTE) profile for East plant Tanks  1-3 - Madison, Wl	   242
8-12   aF(SOTE) profile for Tank 21 (showing influence of high SRT
           and cleaning) - Madison, Wl	   244
8-13   aF(SOTE) profile for Tank 25 (showing influence of low SRT) -
           Madison, Wl	   245
                                          IX

-------
                                Figures (continued)


Number                                                                        Page

8-14   Plant flow diagram (original coarse bubble aeration
          system) - Ridgewood, NJ  	   248
8-15   Plant flow diagram (retrofitted fine pore aeration
          system) - Ridgewood, NJ  	   249
8-16   SOTE vs. airflow rate (fine pore system) - Ridgewood, NJ  	   250
8-17   Savings in power consumption - Ridgewood, NJ	   253
8-18   Aeration basin schematic - Whittier Narrrows, CA	   256
8-19   aF(SOTE) vs. time (ceramic dome diffusers)  - Whittier Narrrows, CA  	   258
8-20   Comparison of ceramic dome and disc diffuser
          performance - Whittier Narrrows, CA	   259
8-21   Air header and distribution system used to retrofit fine pore ceramic discs
          into a tank formerly equipped with surface aeration - Plymouth, Wl  	   263
8-22   Checking for leaks and uniform air distribution1- Ripon, Wl  	   268
B-1    BRV test points for ceramic dome	   276
B-2    BRV apparatus	 .   277
B-3    BRV flow calibration    	   278
B-4    Measurements of air line and diffuser pressure  ........................   279
B-5    Airflow profile apparatus	  .	   280
B-6    Steady-state clean water OTE test apparatus 	   282
C-1    Standard atmospheric  pressure for altitudes of sea level to 10,000 ft  	   289
C-2    Moody diagram for friction factor in pipes  	'.		   290
C-3    Equivalent air pressure (EAP) curves for use in blower selection  '. .  .	   291

-------
                                        Tables
Number                                                                          Page
2-1    Standard Equations for Clean Water Oxygen Transfer Tests	   26
2-2    Clean Water Oxygen Transfer Efficiency Comparison
           for Selected Diffusers	   27
2-3    SOTE vs. Airflow for Selected Fine Pore Diffusers in Clean Water  	   28
2-4    Clean Water Oxygen Transfer Efficiencies of
           Fine Pore Tube Diffuser Systems	   28
2-5    Clean Water Oxygen Transfer Efficiencies of Fine Pore
           Disc/Dome  Grid Systems 	   29
2-6    Clean Water Oxygen Transfer Efficiencies of
           Perforated Membrane Diffuser Systems  	   31
3-1    Guide to Application of Equation 3-3	   38
3-2    Results of Interplant Fouling  Study	   44
3-3    Diffuser Characteristics in 3-Pass, Step Feed Aeration Basin - Madison, Wl  .....   49
3-4    Characteristics  of Diffuser Residue Samples in 3-Pass,
           Step Feed Aeration Basin - Madison, Wl	   49
3-5    Air-side Analysis of Ceramic  Disc Diffusers from Test Headers	   50
3-6    Characteristics  of Perforated Membrane Diffusers in Service	   52
3-7    Operating Conditions for Two Food Processing Industrial
           Treatment Sites Using Perforated Membrane Tube Diffusers  	   53
3-8    Fouling Rates (fp) and Fouling Factors (F)  for Selected Treatment Plants   	   56
3-9    Fouling Rates Estimated by Clean Water Column Testing   	   58
3-10   Physical Characteristics of Wastewater Treatment Facilities
           Providing Oxygen Transfer Data  .	   60
3-11   Oxygen Transfer and Plant Process Data   	   62
3-12   24-hr aF and aF(SOTE) Variations at Selected Municipal Treatment Plants	   63
3-13   aF(SOTE) for Aeration Systems with Different SRTs at Madison, Wl West  Plant  . .   67
3-14   aF Profiles for Various Aeration Systems	   70
4-1    Cleaning Experiences with Fine Pore Diffusers	   89
4-2    Annualized Costs for a Fine Pore Aeration System Experiencing
           a DWP Increase of 1.0 in w.g./month (see Example 4-2)  	   91
4-3    Fine Pore Aeration System Troubleshooting Guide 	   94
5-1    Suggested Worksheet for List of Process Oxygen Requirements  	   101
5-2    Properties of Airflow Measurement Devices  	   119
5-3    Control Methods for Blowers  	   136
5-4    Air Piping Headless Calculations for Example 5-10 - Rrst Iteration  	   144
5-5    Air Piping Headless Calculations for Example 5-10 - Second Iteration   	   145
7-1    Information Needed to Perform Desktop Economic Analysis
           of Diffused  Aeration Systems   	   179
7-2    Design Information for Example Fine Pore Aeration System	   183
7-3    Initial Costs of Design Example Fine Pore  and Coarse
           Bubble Aeration Systems 	   183
7-4    Design Information for Example Coarse Bubble Aeration System  	   183
7-5    Sample Desktop Calculations: Case 3  Fouling Rate for
           Fine Pore Aeration Design Example (18-month Cleaning Interval)  	   185
7-6a   Sample Desktop Calculations (Case 3  Fouling Rate): Zone 1
           of Fine Pore Aeration Design Example (18-month Cleaning Interval)  	   186
                                           XI

-------
                                 Tables (continued)
Number                                                                         Page

7-6b   Sample Desktop Calculations (Case 3 Fouling Rate): Zone 2
           of Fine Pore Aeration Design Example (18-month Cleaning Interval)  	   187
7-6c   Sample Desktop Calculations (Case 3 Fouling Rate): Zone 3
           of Fine Pore Aeration Design Example (18-month Cleaning Interval)  	   188
7-7    Present Worth Costs as a Function of Cleaning Interval
           for Case 3 Fouling Rate	   188
7-8    Economic Comparison of Newly Constructed Fine Pore
           and Coarse Bubble Aeration Systems  	   189
7-9    Sample Desktop Calculations for Coarse Bubble Aeration Design Example   ....   191
7-10   Sample Desktop Calculations for Zone 1
           of Coarse Bubble Aeration Design Example	   191
7-11   Sensitivity of Fine Pore Aeration Costs to Changes in Price
           of Power and Diffuser Cleaning	   192
7-12   Economic Comparison of Retrofit Fine Pore Aeration
           and Existing Coarse Bubble Aeration Systems	   192
7-13   Input Data Form for Oxygen Requirements  for Economic
           Analysis Spreadsheet - Case 3 Fouling  Rate  	, .   192
7-14   Input Data Form for Diffuser Characteristics for
           Economic Analysis Spreadsheet - Case 3 Fouling Rate ,	 .   193
7-15   Input Data Form for Economic Factors for Economic Analysis
           Spreadsheet - Case 3 Fouling Rate (9-month Cleaning Interval)	   193
7-16   Input Data Form for Blower Characteristics  for
           Economic Analysis Spreadsheet - Case 3 Fouling Rate  	   193
7-17   Output Display from Economic Analysis Spreadsheet - Case 3
           Fouling Rates (9-month Cleaning Interval)   	   193
7-18   Initial Cost Information for Selected Fine Pore Aeration Systems	   197
7-19   Costs Associated with Gas Cleaning	 .	   200
7-20   Estimated Time Requirements for Basin and Diffuser Cleaning  	   200
7-21   Diffuser Cleaning and Repair Costs - Ridgewood, NJ  .	   200
7-22   Milwaukee Method  Diffuser Cleaning Costs - Hartford, CT	   200
7-23   Wisconsin Public Service Corporation Rate Schedule  	   201
7-24   Cost of Operating a 750-kW (1000-hp) Motor at 70 percent of Rated
           Load for Various Periods Over a Year (Demand  =  522 kW)  	   201
8-1    Summary of Case Histories	   206
8-2    Comparison of Fine Pore and Coarse Bubble Aeration
           Energy Requirements: Pre-design Estimates -  Frankenmuth, Ml  	   209
8-3    1986 Performance  Summary -  Frankenmuth, Ml  	   210
8-4    Off-Gas Test Results - Frankenmuth, Ml  	   210
8-5    Estimated Energy Cost Savings - Frankenmuth, Ml 	   211
8-6    Summary of aF (SOTE) Determinations - Glastonbury, CT  	   215
8-7    Fine Pore Aeration  System Design Criteria - Green Bay, Wl  	   219
8-8    Fine Pore Diffuser Configuration Summary - Green Bay, Wl	   220
8-9    Operating Data - Green Bay, Wl  	   221
8-10   aF(SOTE) Values from Off-Gas Testing - Green Bay, Wl	   222
8-11   aF as a Function of Time in Service - Green  Bay, Wl  	   223
8-12   Cost Summary Comparison for Alternative
           Aeration Systems - Green Bay, Wl   	   224
8-13   Tank Average Off-Gas Results (Aeration Tank 2) - Hartford, CT  	   233
8-14   Average Off-Gas Test Results Before and After May 1987 Cleaning
           (Aeration Tank 2) - Hartford, CT  	   233
8-15   Diffuser Cleaning Costs - Hartford, CT  	   233
8-16   Summary of Oxygen Transfer Tests - (Jones Island) Milwaukee, Wl   	   237
                                          XII

-------
                                Tables (continued)


Number                                                                       Page

8-17   East Plant Operating and Performance Data
          (Annual Averages) - (Jones Island) Milwaukee, Wl  	   238
8-18   Oxygen Transfer Performance (1985-1988) for Tapered, Full-Floor
          Plate Configuration (Jones Island East Plant, Basin 6) - Milwaukee, Wl	   238
8-19   Oxygen Requirements - Madison, Wl	   241
8-20   Blower Capacities - Madison, Wl	   241
8-21   East Plant aF Values - Madison, Wl	 . . .   243
8-22   Off-Gas Test Results: West Plant (Plant 3) - Madison, Wl  	   245
8-23   Air Usage - Madison, Wl	   246
8-24   Physical Characteristics of Original Coarse Bubble Aeration
          System (Tanks 1 and 2 ) - Ridgewood, NJ	   248
8-25   Projected Energy Savings - Ridgewood, NJ	 .   248
8-26   Physical Characteristics of Fine Pore Retrofit System
          (Tanks 3 and 4) - Ridgewood, NJ	   250
8-27   Estimated Yearly Average aF(SOTE) and aF - Ridgewood, NJ	   250
8-28   Average Blower Power Reduction - Ridgewood, NJ  . .	   252
8-29   Dome Diffuser System Maintenance Costs - Ridgewood, NJ 	   252
8-30   Dome System Economic Summary-Ridgewood, NJ (1983-1986)  ...........   252
8-31   Diffuser Layout Summary - Whittier Narrows, CA  	   256
8-32   Project Chronology - Whittier Narrows, CA  	   257
8-33   Monthly Average Performance - Renton, WA	   266
8-34   DWP Monitoring Results - Saukville, Wl   	   270
B-1    Example Diffuser Flux Calculations  	   281
C-1    DO Saturation Values	   285
C-2    Hydraulic Headlosses for Appurtenances	   286
C-3    Typical Air Velocities in Air Delivery Systems	 .   287
C-4    Properties of Standard Atmosphere	   288
C-5    Physical Properties of Water	   288
D-1    Cell Entries for Economic Analysis Spreadsheet .'	   294
D-2    Command Script for Economic Analysis Spreadsheet	 .   299
                                         Xlil

-------
                                Acknowledgments


This manual was prepared by the American Society of Civil  Engineers (ASCE)  Committee on
Oxygen Transfer, New York, NY,  under Cooperative Agreement No. 812167 between U.S.  EPA
and ASCE. This Committee  was chaired  by William C. Boyle, who also  served  as Principal
Investigator for the project.

A Steering Subcommittee of the  ASCE Committee on Oxygen Transfer, chaired  by Hugh  J.
Campbell, Jr., was created to coordinate the 3 years of field research that, in part, supported this
manual.  The  dedicated efforts  of the  members  of this  Subcommittee  are  gratefully
acknowledged. The manual could not have  been completed without their critical review of all  field
subcontractor reports and their patient critique of manual drafts:

     C. Robert Baillod                          James J. McKeown
     William C. Boyle                           Henry K. Melcer
     Richard C. Brenner                        Thomas C. Rooney
     Hugh J. Campbell (Chairman)               F. Michael  Saunders
     Edwin Jones                              H.  David Stensel
     Frederick K. Marotte (Vice Chairman)         Fred W. Yunt

The following manual authors reviewed the existing data base on  fine pore aeration as well as
the draft reports of the field research subcontractors. They analyzed that data and, utilizing  their
experience in the field, developed the technical base for this manual. Their  efforts are gratefully
acknowledged:

     William C. Boyle                           James J. McKeown
     Glen T. Daigger                           Brooks Newbry
     James A. Heidman                         Thomas C. Rooney
     Gregory L. Huibregtse                      Lewis A. Rossman
     James J. Marx                            H.  David Stensel

This manual represents the cooperative efforts of  over 50 members of the ASCE Committee on
Oxygen Transfer and other experts in the field. The contributions of the following individuals are
acknowledged for their time and effort in reviewing drafts of this manual:

     Thomas A. Allbaugh                       Richard J.  Lorge
     Richard Atoulikiant                         Denis J. Lussier
     William L. Berkt                           James J. Marxt
     Arthur G. Boont                           James A. Muellert
     James H. Clarkt                           George G.  Powellt
     George H. Rushtont                       Michael G. Rietht
     Lawrence A. Ernestt                       Vernon T. Stackt
     Lloyd Ewing                               Charles E.  Tharp
     R. Gary Gilbertt                           Donald J. Thiel
     James A. Heidman                         Richard Veeder
     Bengt G. Hellstromt                       James R. Wahlt
     Gregory L. Huibregtse                      Read Warriner
     John S. Hunter, lilt                        Jerome D.  Wrent
     S. Joh Kangt                              Shang Wen Yuant
     Boris Khudenkot                          R.  Bruce Zimmerman
     Paul M. Kuberat

     ^Committee Member
                                         XIV

-------
                          Acknowledgments (continued)
A substantial amount of the technical information in this manual was derived from subcontracted
research on fine pore aeration. Appendix A lists these research projects and the subcontractors
who performed the work.

Individuals  who  provided specific information, on  fine pore aeration  plant case histories are
acknowledged in Section 8.4.

Manufacturers of fine pore aeration equipment were contacted on several occasions relative  to
the contents of this manual. Their contributions to the manual are also acknowledged.

U.S. EPA Project Officers:
Richard C.  Brenner, U.S. EPA, Risk Reduction Engineering Laboratory,  Cincinnati, OH
Denis J. Lussier, U.S. EPA, Center for Environmental Research Information, Cincinnati, OH

Peer Reviewers:
William H. Busch, Illinois EPA, Springfield, IL
C. Wayne Dillard, HNTB, Orlando, FL
Mario Salazar, U.S. EPA, Office of Municipal Pollution Control, Washington, DC
Robert Polta, Metropolitan Waste Control Commision, St. Paul, MN
K. Fredrick Updegraff, Gannet Fleming Engineers, Harrisburg, PA
                                           xv

-------

-------
                                              Chapter 1
                                    Introduction and Overview
The supply of oxygen for aeration is the single largest
energy  consumer at activated  sludge  wastewater
treatment plants,  representing 50-90 percent of  total
plant  energy requirements (1). Replacement of  less
efficient  aeration  systems with  fine  pore  aeration
devices can save up to 50 percent of aeration energy
costs  and has resulted  in  typical  simple  payback
periods  of 2-6 years (see Chapter 8).  As a result  of
these very impressive cost savings,  more than 1,300
municipal and industrial wastewater treatment facilities
in  the United States and  Canada now use fine  pore
aeration.

Fine pore aeration technology remains relatively  new
in   North  America, and  new   materials   and
configurations continue to be  developed. This manual
provides designers,  end  users,  and  regulators
information  on  the nature  of  fine pore  aeration
devices, their performance, and related operation and
maintenance (O&M)  requirements  to promote the
intelligent application of fine pore aeration technology.
Standardized testing  of  oxygen  transfer  devices  in
both  clean  and process  waters is  a recent major
advancement in the field. A consensus Standard for
testing  aeration devices  in clean  water has  been
adopted by a  large  segment  of the industry  (2).
Extensive testing of  aeration equipment  using  this
Standard  has led to the development of a large  data
base  on the performance of aeration devices in clean
water.  In addition,  over the  past  10  years,  the
development of improved (more precise and accurate)
field test  methods have permitted generation of  data
that can be  used to better characterize the translation
of  clean water test results to process conditions (3).

In  1985,  The U.S.  Environmental Protection  Agency
(EPA) funded a  Cooperative Research  Agreement
(No.  CR812167)  with.the American Society  of  Civil
Engineers (ASCE)  (hereinafter  referred  to  as the
EPA/ASCE Fine Pore Aeration Project) to evaluate the
existing  data base  on  fine  pore diffused  aeration
systems in both clean and process waters, to conduct
field  studies at a  number of  municipal  wastewater
treatment facilities employing fine  pore aeration, and
to  prepare this manual. Appendix A summarizes the
field studies that were conducted. A  Summary Report
on fine  pore aeration systems (4)  was published  in
October 1985 to summarize the early findings of this
study.
This manual  has  been  assembled based on  the
evaluation and analysis of data collected over the last
4 years. It has  been  written  to  provide  engineers,
regulatory personnel, and others involved in fine pore
aeration  system design  and  installation  the most
current design and O&M information available.


1.1 Historical Overview
Experiments on wastewater aeration in England date
back to as early as 1882 (5). In these experiments, air
was introduced through open tubes or perforations in
air  delivery pipes.  In  the early years,  patents were
granted - primarily in the United Kingdom (U.K.) - tor a
variety of diffusers, including perforated metal prates,
porous tubes with fibrous materials, and nozzles' (6).
As  activated  sludge  process  investigations
progressed,  greater oxygen transfer efficiency was
sought with the production of smaller bubbles created
by passing compressed air  through porous media of
various  types. Experiments  conducted  in  the  U.K.
seeking  a better porous material included evaluations
of limestone, fire  brick,  sand and glass  mixtures,
pumice, and other materials. The first  porous  plates
were made available as early as  1915 in the U.K. In
the following  years,  several  U.S.  companies
(Carborundum Co., Ferro Corp.,  and Norton  Co.)
offered porous plates  that became the most popular
method  of aeration in this country in the 1930s and
1940s.

It became clear, shortly after the emergence of porous
diffusers  that media clogging  could be  a problem.
Work  in  Chicago between 1922 and 1924  prompted
the use  of coarse  media  to avoid  clogging  (7).
Clogging was attributed to liquid-side fouling  and air-
side clogging due  to  dirt and  oil in  the air delivery
system.  Emphasis  at that, time was on improving  air
filtration  (8-10). Substantial experimentation  was
performed to develop  effective air filtration devices
(9,10), and the results of that early work have  led to
the high-efficiency air filters  used today  in  many
porous diffused air systems (8).

Mechanical aeration was one answer to the clogging
problem.  Since  the  introduction of Archimedean
screw-type  aerators in  1916, a  multitude  of
mechanical aeration devices has been  developed and
used. Today, mechanical aeration devices serve  an

-------
important function  in many applications for treatment
of industrial and municipal wastewaters.

Another  approach  to the clogging problem emerged
with the  development  of  large  orifice-type (coarse
bubble) diffusers. First  marketed as early as 1904 in
the U.K., updated .versions evolved in the  1950s in an
effort to improve on earlier perforated pipe devices
and were designed for easy access and maintenance.
In general,  the principles  that  resulted  in  easily
maintained and  accessible aeration systems also
produced low oxygen transfer efficiencies.

With the renewed emphasis  on more energy efficient
aeration  systems in the 1970s, North America again
turned to European technology  for  more  efficient
oxygen transfer devices. As a result, newer generation
porous  diffusers,  producing fine  bubbles  with
accompanying  high oxygen  transfer efficiencies,
became popular. Yet, considerable concern has been
registered   regarding the   performance  and
maintenance of these porous diffusers owing to their
historical susceptibility to clogging.

Today,  the  aeration market  is  in a state of flux.
Emphasis on high  efficiency oxygen  transfer has led
to intensive  research   and  development programs
aimed  at providing high transfer efficiency devices
with low  maintenance requirements. The search for a
better oxygen transfer device will  continue to occupy
the time  and resources of a number of manufacturers
and researchers.
1.2 Objectives of the Manual
The term "fine  bubble"  diffused  aeration is  elusive
and difficult to define and comprises a wide variety of
diffuser types,. The term "fine pore" was  adopted in
the previous Summary Report (4) to more nearly
reflect the porous characteristics of one class of high
efficiency diffusers marketed  today.  Typically, fine
pore diffusers will produce a headless due to  surface
tension in clean  water of greater than about 5 cm (2
in) water gauge.  For the purposes of this manual, fine
pore diffusers are defined as  including the following
devices:

 *  Porous ceramic plates, discs, domes, and tubes
*  Rigid porous plastic plates, discs, and tubes
*  Nonngid porous plastic tubes
»  Perforated membrane tubes and discs

The major objective of this manual is to  provide the
technical community with the most current  information
available on fine pore aeration systems.  Because of
the rapid changes that  are  taking  place  with this
technology,  it is possible that some systems  have
unintentionally not been described.  The manual
presents  what  are  considered  the  best current
practices  for  selecting,  designing,  operating,
maintaining,  and  controlling  fine  pore aeration
systems.
Example calculations are presented  throughout the
manual to illustrate concepts and are  not intended  to
set a rigid standard in design protocol. Case histories
are presented to give the user an appreciation for the
highly site-specific nature of this technology.
1.3 References
When an NTIS number is cited
reference is available from:
in  a reference, that
  National Technical Information Service
  5285 Port Royal Road
  Springfield, VA 22161
  (703)487-4650

1.   Wesner, G.M.,  L.J. Ewing, T.S. Lineck,  Jr. and
    D.J. Hinrichs. Energy Conservation in  Municipal
    Wastewater Treatment.  EPA-430/9-77-011,  NTIS
    No. PB81-165391, U.S.  Environmental Protection
    Agency, Washington, DC, 1977.

2.   American  Society  of Civil  Engineers.  ASCE
    Standard:  Measurement  of Oxygen  Transfer  in
    Clean Water. ISBN 0-87262-430-7, New YorkJJY,
    July 1984.

3.   Mueller, J.A. and  W.C.  Boyle. Oxygen Transfer
    Under  Process  Conditions. JWPCF  60(3):342-
    341, 1988.

4.   Summary  Report:  Fine  Pore (Fine  Bubble)
    Aeration  Systems.  EPA-625/8-85-Q10,  U.S.
    Environmental Protection  Agency, Cincinnati, OH,
    1985.

5.   A.J.  Martin. The  Activated  Sludge  Process,
    MacDonald and Evans, London, England, 1927.

6.   Air Diffusion in Sewage Works. Manual of Practice
    5,  Federation  of  Sewage and  Industrial  Wastes
    Associations, Champaign, IL, 1952.

7.   Bushee, R.S. and S.I. Zach. Tests on Pressure
    Loss in Activated  Sludge  Plants. Engineering
    News Record 93:21, 1924

8.   Aeration In Wastewater Treatment.  Manual  of
    Practice 5. Water  Pollution Control  Federation,
    Washington, DC, 1971.

9.   Anderson, N.E.  Tests and Studies on Air Diffusers
    for Activated Sludge.  Sewage and  Industrial
    Wastes 22:461, 1950.

10.  Morgan,  P.F.  Maintenance   of  Fine  Bubble
    Diffusion. J. San. Eng. Div., ASCE  84(SA2): 1609,
    1958.

-------
                                              Chapter 2
                                 Fine Pore Diffuser Characteristics
2.1 Introduction
Since introduction of the activated sludge process in
the early  1900s,  many  different types  of diffused
aeration devices have been designed and developed
to dissolve  oxygen into wastewater.  These have
ranged from simple individual orifices (holes or slots)
drilled in a section of pipe to elaborate devices made
up of small diameter particles fused together.
Although  their  size,  shape, and  materials  of
construction may vary considerably,  diffused aeration
devices are  usually classified as either fine or  coarse
bubble  -  referring  to the  relative  diameter  of  the
bubble produced. The demarcation between fine  and
coarse bubbles is not well defined.

Coarse  bubble diffusers normally  produce a  bubble
diameter  of 6-10 mm  (1/4-3/8 in)  in clean  water.
Although  the actual  orifice may be much larger,  the
bubbles  produced tend  to shear  and  break  into
smaller  bubbles as they are produced and rise to the
surface. For this type of device, as long as the mixing
intensity stays roughly  the same,  bubble  size is
usually  independent of  the airflow rate  through  the
diffuser.  However,  mixing intensity  increases  with
increasing airflow,  which  most likely  shears  the
bubbles into bubbles with smaller diameters. This may
explain  the  apparent  gradual increase in  oxygen
transfer efficiency (OTE) with increasing airflow  rate
for some coarse bubble systems.

Fine bubble  diffusers,  when new,  produce bubbles
with  a diameter of approximately 2-5 mm  (0.08-0.20
in) in clean water (1-5). One class of diffusers defined
in Chapter  1   as  fine pore diffusers will  normally
generate  fine  bubbles. The  bubble size produced by
fine  pore  devices is affected  by airflow,  becoming
somewhat larger as airflow increases.
This chapter discusses the important characteristics
of fine pore aeration devices. Included are sections on
material  types,  diffuser  shapes,  typical  layouts,
physical  properties,  and  clean water  performance
(OTE) data.


2.2 Types of Fine Pore Media
Although  several  materials capable of serving  as
effective fine pore diffusion media exist, few are being
used  in the wastewater  treatment  field  because  of
cost considerations,  specific characteristics,  market
size, or other factors. Fine pore media in use today
can  be divided  into the  following  three  general
categories:

1. Ceramics

2. Porous  plastics

3. Perforated membranes

2.2.1 Ceramics
Ceramics  are the oldest, and currently most common,
porous media on the market. Ceramic media consist
of irregular or spherically shaped mineral particles that
are sized,  blended  together with bonding materials,
compressed into various  shapes,  and fired  at an
elevated temperature to form a ceramic bond  between
the particles. The result  is a network of intercon-
necting passageways through which air flows.  As air
emerges from the surface  pores, pore size, surface
tension,  and  flow  rate  interact to  produce  the
characteristic bubble size (6). Ceramic diffusers have
been  manufactured  from  alumina, aluminum silicate,
and silica.

Alumina (aluminum  oxide)  media  used for fine pore
diffusers  are produced  from  bauxite,  a naturally-
occurring   ore  consisting  primarily of  aluminum
hydroxide. Calcination followed  by  electrorefining
produces a material that exceeds 80 percent  alumina.
The refined molten mineral is allowed to solidify and is
subsequently crushed and screened  to  select the
desired sizes (7). This raw material is called regular or
brown fused aluminum oxide.

Other materials  used for fine  pore media  that are
similarly crushed and then sized  are  produced from
either  man-made or naturally-occurring aluminum
silicates such as cyanotic. AH these  minerals all
combinations of  alumina  (aluminum oxide) and silica
(silicon oxide) with an alumina content of  50 percent
to  slightly over  70 percent.  On  heating  at  high
temperatures, they  tend  to  form the  equilibrium
species of the mineral (mullite) and  a  siliceous glass.
Mullite consists  of three  parts alumina  to two parts
silica.

-------
Silica used in the manufacture of fine pore diffusers is
a mined material with a somewhat limited particle size
range.  As  a result,  the  pore  sizes  that can  be
produced are restricted to  naturally-occurring  grain
sizes.  Some sources  of silica are less angular than
the more fagged crushed alumina or aluminum silicate
particles. It  has  been  claimed that the silica material
may be more  resistant to fouling  and  more easily
cleaned (2). Although this  claim  has not been  well
documented based  on controlled experiments,  silica
plates  have been used for years in at least one facility
(see Section 8.2.5).

No studies have been published that suggest there is
a difference In process performance between diffusers
made with either alumina or aluminum silicate media.
Experience  would  seem to  indicate that the  two
materials may have  been used interchangeably.  As a
raw material, alumina is  more abrasion resistant than
either  silica  or  aluminum  silicate;  however, actual
strength and abrasion resistance also depends on the
nature of the ceramic bond itself. Silica porous media,
as manufactured, are generally considered to have the
lowest overall strength, and their thickness  is usually
increased to compensate for this weakness.

Sources of ceramic diffuser media are  comprised
primarily of companies supplying  industrial  abrasives
or refractory materials. In the past, these basic porous
media  manufacturers  have  both  marketed diffusers
directly and supplied various media shapes to aeration
equipment manufacturers.  The aeration  equipment
manufacturers,  in turn, marketed the diffuser, holder,
piping,  and other ancillary  equipment.  Today, the
basic porous media manufacturers will market finished
diffuser  assemblies  directly to  the end  user  with
warranties covering only mechanical specifications for
materials, dimensions, and  permeability. They will not
warranty oxygen transfer performance.

Ceramic diffusers have been in use for a  long period
and their advantages, as well as operational problems,
are well documented. Because  of this  longevity,
ceramic materials  have,   in  effect, become  the
standard for comparison. Each new generation of fine
pore media reportedly offers some advantage, either
in performance or cost, over ceramic diffusers. In the
past, the new devices  have often not lived up to their
expectations. As  a  result, ceramic diffusers continue
to capture a significant share of the fine pore aeration
market.

2.2.2 Porous Plastics
A recent development  in the fine pore diffuser field  is
the use ol porous plastic materials. As with ceramics,
a  medium  is  created  consisting  of  several
interconnecting  channels  or pores through which
compressed air can travel. The manufacturing process
can be  controlled  to  produce different  pore sizes.
Some  of the advantages  of plastic materials  over
ceramics are their ease of manufacture, lighter weight
(which  makes  them  especially suited  to liftout
applications),  durability,  cost effectiveness,  and,
depending  on  the  individual  material,  greater
resistance to breakage. Some disadvantages include
their  lower  strength,  greater susceptibility to creep,
and lower environmental resistance.

To  provide  further  definition, porous  plastics  are
classified as  either  rigid or nonrigid. The basic
difference between the two is that a nonrigid porous
plastic  requires  some form  of internal  structural
support.
2.2.2.1 Rigid Porous Plastics
Rigid  porous  plastics  are  made  from  several
thermoplastic  polymers  including  polyethylene,
polypropylene,  polyvinylidene  fluoride,  ethylene-vinyl
acetate,  styrene-acrylonitrile,  and   polytetra-
fluoroethylene (8).  Besides their application  in  the
aeration field, these materials are used extensively as
filtration media (air, water, chemicals). When used as
a fine pore aeration device,  the two most  common
types of plastic  media are high density polyethylene
(HOPE) and styrene-acrylonitrile (SAN).

HOPE is relatively inexpensive and easy  to process
when compared  with other thermoplastics.  In addition,
shrinkage is  low,  a  uniform  quality  product  can  be
obtained,  and small pore  sizes  can  be  produced.
HPDE diffusers  are manufactured  by   proprietary
extrusion processes. A HOPE diffuser is usually made
from  a  straight  homopolymer  (not  a  blend),  is,
nonpolar, and contains no additives or binders.

A  European  manufacturer  produces  a double-layer
HPDE material  (9). It consists of  a grainy, open-pore
structure approximately  6-mm (1/4-in)  thick covered
by a thinner [(3-mm (1/8-in) thick)], less porous layer.
The manufacturer claims that the double layer  results
in  a  filtering  effect  that  decreases  maintenance.
Allegedly, this lower maintenance would result  from a
reduction in air-side fouling of the diffuser. However,
recent studies have demonstrated that air-side  fouling
does  not appear to be an important  fine pore diffuser
operation and maintenance  (O&M) factor (10,11). The
manufacturer also  claims that the thin outer layer is
potentially beneficial in  helping  to  produce a small
diameter bubble uniformly over the diffuser surface.
The manufacturer further suggests the thin film layer
may  act  as a barrier in preventing  precipitates from
forming deep within the media.  If  the foulants  are
restricted  to  the  surface, they can  be more easily
removed (9). Any tendency  for  increased  external
fouling in such diffusers is unknown.

The major advantages of HOPE media compared with
other  plastic  media  are  their  lighter  weight
(approximately  560  kg/m3   [35 Ib/cu   ft)),  inert
composition, and resistance to breakage, even under
freezing conditions.

-------
SAN diffusers are manufactured from a copolymer.
The raw  material  is  a  mixture  of  four  different
molecules. Physically,  SAN  media are composed  of
very small  resin  spheres  fused together  under
pressure.  SAN has a density only slightly greater than
HOPE. The presence of the  styrene, however, makes
the material  brittle, and the  media  can  break  if
dropped,  even at  room  temperature.  A  major
advantage of the SAN material is that it  has  been
successfully used for about 20 years.

Although  rigid plastic  media diffusers  (especially the
HOPE  material)  were  installed in several wastewater
treatment  plants  in the early 1980s, they are no longer
as  popular.  SAN diffusers,  which seemed to  work
well, are  no  longer being actively marketed.  HOPE
diffusers,  on the other hand, were plagued by several
problems that contributed to their decline in popularity.
These  included media fouling, lack of quality control in
the manufacturing  process (diffusers  did not always
produce uniform air distribution),  and the  emerging
cost competitiveness  of  other  fine  pore  diffusion
products.

2.2.2.2 Nonrigid Porous Plastic
Only one  type of diffuser material is being marketed
that could be considered a nonrigid  porous plastic
(12).   This  material,  which  is extruded from  a
combination  of rubber  and  HOPE, is  soft and
somewhat flexible.  When air  is  applied  at normal
pressures, this  type  of  media  does  not expand.
However,  some expansion will  occur  at higher
pressures. It has  been  demonstrated  that a high
pressure water flush can be used to remove foulants
that may have become  trapped in the pores (11).

2.2.3 Perforated Membranes
Membrane diffusers differ from  the first  two groups
(ceramics  and porous  plastics)  in that  the diffusion
material  does not  contain  the  network   of
interconnecting passageways through  which air must
travel.  Instead, mechanical means are used to create
preselected  patterns  of  small,  individual orifices
(perforations) in the membrane to allow passage of air
through the material.

Membrane diffusers have been  in use for about  40
years.  They initially were  called  "sock" diffusers and
were made from materials such as plastic,  synthetic
fabric cord, or woven cloth.  A  metallic or plastic core
material was required  with  the  woven  sheaths  to
provide structural  support.  Although  sock  diffusers
were capable of achieving high oxygen transfer  rates,
fouling problems were  often severe.  Today, there  is
essentially no market for the  early sock design.

A new type of perforated diffuser has been introduced
within  the last decade. It consists of a  thin flexible
membrane made from  either a thermoplastic material
or an elastomer. Because of their physical properties,
current literature usually  describes diffusers  made
from  these materials as  "flexibles"  or  "flexible
membrane"  diffusers. However,  since  the  patterned
orifices  in  the  membrane  material  are intentionally
made during the manufacturing process,  this new
generation of membrane diffusers is referred to herein
as "perforated"  membrane diffusers.

ASTM D-883  (13)  defines  a thermoplastic as  "a
plastic that  repeatedly can be softened by heating and
hardened by cooling  through  a temperature  range
characteristic of the  plastic, and, in the softened state,
can  be  shaped  by  flow into articles  by molding or
extrusion."  The most  common thermoplastic material
is  polyvinyl  chloride (PVC).  Plasticizers are  added to
produce a soft, flexible membrane.

The  term elastomer includes the complete  spectrum
of  elastic or rubberlike polymers that are sometimes
randomly referred to as rubbers, synthetic rubbers, or
elastomers  (14). Rubber is a natural material, while all
other elastomers are synthetic. ASTM D-883 (13)
defines an elastomer as "macromolecular material that
returns rapidly to approximately the initial dimensions
and  shape  after substantial  deformation (defined as
twice its length in an earlier addition)  by a weak stress
and release of the stress."

Most elastomer  membranes  are  made  from
ethylenepropylene dimer (EPDM). Although  the main
ingredient may  be  EPDM,  proprietary  additives are
usually  included  to  enhance   the  material
characteristics.  As a  result, it may  not always  be
possible to  establish media characteristics or physical
properties simply by consulting a table in a  reference
text or handbook.

After the  membrane  material  is  produced,  air
passages are created by punching or cutting minute
holes or slits in  the membrane.  When the air is turned
on, the material expands. Each  hole acts as a variable
aperture  opening; the higher  the  airflow  rate,  the
greater the  size of the opening. When the air is turned
off, the  membrane relaxes down against the support
base  and a seal is formed between membrane  and
support in   systems  where the  membrane  area
conforms to the support.  This closing action  will
reportedly eliminate  or at least  minimize the backflow
of  liquid into the diffuser (15-20).

Perforated membrane diffusers have been developed
over the last 10-15 years in the United States  and
Europe. The most significant advantage claimed  for
the perforated  membrane diffuser is  that its smooth
surface  and apertures  may  be more  resistant to
fouling than are other types of media. Formation of
biological and chemical  foulants has been  noted  on
the surfaces of the thermoplastic media,  however
(10).  One  of the disadvantages  of  the perforated
membrane   diffusers  is that thermoplastic and
elastomer materials  can experience physical property
changes with time. These changes depend, to varying

-------
degrees,  on the  materials used,  their shape and
dimensions, and environmental conditions.
2.3 Types of Fine Pore Diffusers
There  are  four  general types  (shapes)  of fine pore
diffusers on the market:  plates, tubes,  domes,  and
discs. Each is discussed in detail below.

2.3.1 Plate Diffusers
The original fine  pore diffuser design was  a  flat
rectangular ceramic plate. These plates are usually 30
cm (12 in) square  and 25-38  mm (1-1.5 in) thick.
They are manufactured from either glass-bonded silica
or ceramically-bonded aluminum oxide and aluminum
silicate. The plales are installed  by grouting them  into
recesses  in the basin floor,  cementing them  into
prefabricated  holders, or clamping  them into metal
holders. Of  the three, the metal holders are the least
attractive because corrosion of  some  of the  holders
may result in fouling of the underside of the diffusers.
A  chamber underneath  the plates  acts as  an  air
plenum. The number of plates fixed over a common
plenum is  not standard and can  vary from a single
plate  to 500  or more.  Individual control orifices  are
normally nol provided for this original plate design.

Fine pore ceramic plates were  used almost exclusively
as  the method of air diffusion in  the  early activated
sludge planls through the 1920s.  Other than retrofits
or expansions of existing plants in  Milwaukee, WI  (21)
and Chicago,  IL  (22), the original ceramic plate design
is seldom specified today. Some possible explanations
for its decline in  popularity include:

1.   problems  obtaining uniform air  distribution with
    several plates attached to the same plenum,

2.   inconvenience of removing plates that are grouted
    or cemented in place,

3.   difficulty  in adding diffusers  to  meet  future
    increases in  plant loading, and

4.   lack  of  active  marketing  by  any  equipment
    supplier or media manufacturer.

Some  of the  advantages of plate diffusers are their
documented service  life,  high  oxygen transfer,  and
ease of in-situ cleaning.

A recently-developed plate design offers both ceramic
and porous plastic  media options (23). Both these
alternatives  are marketed in sizes of 30 cm x  61 cm
(12 in x 24  in) and 30 cm x 122 cm (12 in x  48 in),
although other specialty sizes are also available.  The
ceramic plates, made of ceramically-bonded aluminum
oxide, are 19 mm (3/4 in)  thick, whereas the  porous
plastic  (35-micron  HOPE) design  uses a  media
thickness of only 6 mm (1/4 in). Both the ceramic and
plastic media are mounted on  top of vacuum-formed,
ABS plastic plenums. The undersides of the plenums
are filled  with  concrete  at  the job site to  provide
position stability  on  the  basin floor and  adequate
ballast to prevent floating  of the diffusers. The porous
HOPE media  sheets  are sealed to  their  individual
plenums by 25-mm (1-in) wide stainless steel frames
bolted into the concrete ballast, whereas the ceramic
plates are mastic bonded to ledges  set into the inner
periphery of their plenums.

Air is  fed to each diffuser of this new generation plate
design through a separate 3.8-cm (1.75-in)  diameter
rubber hose. The hoses are  connected by saddles to
common  air delivery  headers.  Individually-drilled
orifices are inserted into the feed nipple of each plate,
promoting improved air distribution control compared
with the original plate design. Typical  plate diffusers of
both the original and new  designs are shown  in Figure
2-1.

2.3.2  Fode Diffusers
Like   plates, fine  pore  tubes  have  been  used  in
wastewater  treatment for many years (12,15,20,24-
28).  The  early tubes,  Saran  wound or  made from
aluminum  oxide,  have been  followed  by  the
introduction  of  SAN  copolymer, porous HOPE, and,
most  recently,  the  new generation  of  perforated
membranes and nonrigid porous plastic.

Most  tube  diffusers on the  market  are of the same
general shape. Usually, the media portion is 51-61  cm
(20-24 in) long and has an  outer diameter (O.D.)  of
6.4-7.6 cm (2.5-3.0 in). A tee assembly is sometimes
placed in the middle, increasing total length to nearly
100 cm (40 in).  The thickness of the  media  is
variable.  Perforated  membranes  are  very  thin,
commonly 0.6-2.5 mm (0.025-0.10 in). HOPE media
are usually supplied at a thickness of 6.4 mm (1/4  in),
SAN  media at approximately  15 mm (0.6  in),  and
fused ceramic media at 9.5-12.7 mm (3/8-1/2  in).

The dimensions of the rubber-HDPE  (nonrigid porous
plastic) tube diffuser differ from those just mentioned.
This diffuser has an O.D.  of 25 mm (1.0 in), comes in
lengths  of  up  to 91 cm  (36  in), and has  a  media
thickness of approximately 3 mm (1/8 in).

The holder designs for the ceramic and rigid porous
plastic media are very similar. Most consist of two end
caps  held together by a  connecting  rod through the
center. The rod is threaded  into the feed end of the
holder, the media and outer end cap installed, and a
hex nut placed  on the threaded rod to  secure the
assembly. When  ceramic media  are used,  the end
caps and connecting rod are  usually metallic (stainless
steel). With HOPE media, because  of their  lighter
weight,  PVC,  polypropylene, or  some  other
thermoplastic is often used.  When plastic  holders  or
pipe connectors are used, the effects of creep should

-------
 Figure 2-1.   Typical ceramic plate diffusers.  ,


                            -2       /—3
                       Legend
                     S.S. Eye-anchor
                 2.   Ceramic Plate (HOPE media also available)
                 3.   S.S. Retainer Clip
                     PVC Air Inlet
                 5.   PVC Hose Adapter
                 6.   S.S.-Hose Clamp
                     Feeder Airline    -
                     S.S. Anchor Boil
                     Air Plenum
                     Concrete Ballast
                     Optional Side Inlet    .
 1.
 4.
 7.
 8.
 9.
10.
11.
                                                            10
                                      ENVIRONMENTAL DYNAMICS
be considered. Threaded plastic connectors are also
susceptible to failure unless properly designed.

In another slightly different version,  the feed end cap
and  inner support  are  one piece with the assembly
held together by a bolt installed through the outer end
cap  and threaded  into the support  frame.  For both
designs, gaskets are placed between the media and
the end caps to provide an airtight seal. Sometimes, a
gasket or O-ring is  also used with the retaining bolt or
hex  nut. A  typical rigid porous plastic (or ceramic)
tube diffuser assembly is shown in Figure 2-2.

Because of  their flexible nature, perforated membrane
diftusers require an  internal  support structure. The
frame is made from plastic (e.g., PVC or polypropyl-
ene) and has a tubular shape.  The tube provides
support either around the .entire circumference or, in
                                       one modification,  only  the  bottom half. With  the  full
                                       tubular frame, holes in the  inlet connector or  slots in
                                       the tube  itself allow distribution of the air below  the
                                       membrane  surface. The membrane  is  usually  not
                                       perforated at the air inlet point. Thus, when the air is
                                       turned  off,  the  liquid  head collapses the  membrane
                                       against the support frame.  With the  cutaway  support
                                       frame,  an  internal flap prevents backflow into  the
                                       piping system.  The ends of the support frame  are
                                       plugged, and stainless steel hose  clamps are used on
                                       each end to secure  the membrane  tightly  to  the
                                       frame.  Typical  perforated  membrane diffusers  are
                                       illustrated in Figure 2-3.       •

                                       With  the  rubber-HDPE (nonrigid  porous  plastic)
                                       diffuser, either a permanent (crimped) stainless steel
                                       or a removable  acetal clamp  is provided an the feed
                                       end only. The opposite end of the media is bonded to

-------
Figure 2-1.   (continued)
                          MILWAUKEE, Wl, CERAMIC PLATE DIFFUSER CONFIGURATION
            • 6-in CPVC Header, typ.

                       Top of Raised Floor, 1-1/2 in Concrete
                     " Topping with Galvanized Lath Reinforcing
                         Cement Mortar Separator
                                       Cement Mortar
                                         Ledge
                  ^^!^:^'^i^^^^^>!^
                         \      \            \                         Y~
                  Epoxy Grouted
                      Pipe
_ 12-in x 12-in Ceramic
    Plate Diffuser
  Top of
Basin Floor
Figure 2-2.   Typical ceramic or rigid porous plastic tube
            diffusor.
                    Diffuser Element
                                             3/4" NP1
                                     Cast S.S.
                                    Inlet Nozzle
itself  and  requires  no  clamp. The  tubular  support
frame is made from either stainless steel or fiberglass
reinforced  plastic. To prevent liquid backflcw  into the
air distribution piping, a rubber check valve  can  be
provided. A nonrigid porous  plastic diffuser is shown
in Figure 2-4.
       To avoid corrosion problems, most components of the
       various  tube assemblies are made of either stainless
       steel or durable plastic. The gaskets are usually of a
       soft rubber material. Tube assemblies with a length of
       51 -6.1 cm (20-24 in) are generally designed to operate
       in the airflow range of 0.5-4.7 L/s (1-10 scfm)/diffuser.
       To obtain  maximum OTE,  the diffusers  are  usually
       operated near the lower end of this range  (0.5-2.4 L/s
       [1-5  scfm]/diffuser). Because of their inherent shape,
       it is sometimes difficult to obtain air discharge around
       the  entire  circumference  of the tube.  The air
       distribution pattern  will vary  with  different  types of
       diffusers.  The  extent of the inoperative area  will
       normally be a  function of airflow rate and  headloss
       across the media. As a rule-of-thumb, flow distribution
       around  the tube will improve as headloss increases.
       Laboratory- or  pilot-scale tests should  be conducted
       before selecting a  particular tube design  since dead
       areas can provide sites for foulant development.

       Some tube assemblies are  fitted with an  airflow rate
       control orifice  inserted in  the inlet nipple to aid in  air
       distribution. The orifice is normally about 13 mm (0.5
       in) in diameter,  although different sizes can be used
       for various design airflow rates.

-------
Figure 2-3.   Typical perforated membrane tube diffusers.
                            Header Pipe
                                                       PVC Perforated Membrane Tube
                                                                       Frame
                                                 PARKSON
                                                                            EPDM Perforated Membrane Tube
                                                      Air Distributors
                                                  SANITAIRE

-------
  Figure 2-4.   Typical nonrigid porous plastic tube diffuser.

                     • Diffuser Connector
             Airflow
             Control
             Orifice
             (Optional)
                                            Air Duct
                                                                   Nonrigid Porous Plastic Tube

                                                                   Media Frame

                                                                  Air Plenum
                                             Air Check Valve
                                                (Optional)
Media Frame

Diffuser Connector
                                                  AERTEC
2.3.3 Dome Diffusers
The line pore dome diffuser was developed in Europe
in the 1950s and introduced in the U.S. market in the
early 1970s  (29).  Long  considered the standard in
England and some  parts of Europe, domes  are now
installed in a large number of U.S. plants (30).   '

The dome diffuser is a circular disc with a downward-
turned edge  (31-33). These diffusers are 18 cm (7 in)
in diameter and 38 mm (1.5 in) high. Media thickness
is approximately 15  mm (5/8 in) on the edges and 19
mm (3/4 in) on the top or flat surface. Domes are now
being made predominately of aluminum oxide.

The dome diffuser is mounted on either a PVC or mild
steel saddle-type   baseplate  and attached  to  the
baseplate by a bolt through the center of  the dome.
The bolt can be made from a  number of materials
including brass,  plastics, and  stainless steel. Care
must be taken when specifying the bolt material  and
installing  the  dome to prevent over-tightening of the
center  bolt.  Applying too  much  force can  lead to
immediate diffuser breakage or future air leakage  due
to compression set  of gaskets and bolt stretching, or
failures if a plastic bolt is used. A  soft, rubber gasket
is placed  between  the  diffuser and the  baseplate.
Ideally,  the gasket material should  not take excessive
permanent compression set with the specified torque.
A washer and gasket are also used between the  bolt
head and the top  of  the diffuser. Schematics of
several dome diffusers are shown in Figure 2-5.

Usually, the PVC saddles are solvent cemented to the
air distribution piping at  the  factory.  This  minimizes
            bonding, joining,  and assembly problems that  could
            occur in the field. For one alternative design (31), the
            baseplate  includes an expansion  plug that is inserted
            into a hole drilled into the air header. It is easier with
            this design to replace damaged supports than with the
            solvent-cemented base, but  it is more difficult to level
            or align the baseplates.

            The  slope  of a headless vs. airflow rate  curve for  a
            ceramic diffuser  is  very  flat.  A  variation  from  the
            average of  ±10  percent in the specific  permeability
            (see Section 2.5.4) of  a diffuser can result in a 200-
            percent change in airflow  rate for  the same headless
            under  operating conditions  (34).  To  better distribute
            the air throughout  the system,  control  orifices  are
            placed in each diffuser assembly  to create additional
            headless and balance the airflow. The fastening bolt is
            hollowed out and a  small hole drilled  in  the side, or
            the orifice is drilled in the base of the saddle. The size
            of the orifice is normally 5 mm (0.2 in).

            Dome  diffusers are usually designed to operate at an
            airflow rate of 0.5 Us (1 scfm)/diffuser with a range of
            0.24-1.2 L/s (0.5-2.5 scfm)/diffuser.  Operation above
            1.2 L/s (2.5 scfm)/diffuser  is possible, but may not be
            economical  (see  Section 2.6).  Increasing the  airflow
            rate  above 0.9 L/s  (2  scfm) results in a continuing
            increase in backpressure  and decrease in OTE  and
            may  require a larger control orifice.

            2.3.4 Disc Diffusers
            2.3.4.1 Ceramic and Rigid Porous Plastic Media
            Disc diffusers are a recent development.  Discs are
            flat, or relatively so,  and are differentiated from  dome
                                                    10

-------
Figure 2-5.   Typical ceramic dome diffusers.
                                                        7" Dia. Aloxile
                                                        Ceramic Dome
                                                        PVC Dome Saddle

                                                        Air Flow
                                                        Control Orifice

                                                        PVC Header Pipe
                                                        Adjustable
                                                        S.S. Pipe Support and Clamp
                                                       Tank Floor
                                           PARKSON
                                 S.S. Diffuser Boll
                 S.S. Washer and Neoprene Gasket
                          4" PVC Header Pipe
     Aluminum Oxide Ceramic Dome
         Neoprene Dome Gasket
         Type 316 S.S. Worm Gear Clamp
                                           AERCOR
                                        Orifice   Ceramic
                          Expansion Plug
                                        Bolt
                                                 Dome
Support Ring
                      Strap
               (Expanded to Facilitate
               Removal Over Header)
              Integral
             Full-Face
              Support
          Footing Anchored to
          Aeration Tank Floor
          Air Header
          Adjustment
          Range

           Air
           Header
           Support

 Turn Header Support
 to Raise or Lower
 Level or Header
                                       EPCO INTERNATIONAL
                                                           11

-------
diffusers  in  that they  do  not  include  a downward-
turned  peripheral edge.  While the  dome design  is
relatively  standard,  available disc diffusers  differ  in
size,  shape,  method  of attachment,  and  type  of
material (9,35,36). Schematics of typical ceramic and
porous plastic diffusers are presented in Figure 2-6.
One design  can be equipped with either ceramic  or
glass bead media.

Figure 2-6.  Typical ceramic and rigid porous plastic disc
           dilfuscrs.

                       Nut, Washer, and Stud
                            Silicone Gaskets
                                    \
                                       304 S.S. Mount
                   FILTROS
         Porous Disc (Ceramic or Glass
             ^         Bead Media)
                                 — Media Holder

                                   Pipe Saddle
                                   Drainer Slem
                WILFLEY-WEBER
   Aluminum Oxido Disc
                  Contoured Surface
  Compressed
  Edgo
     Control Orifice
       Threaded
       Retainer Ring

    Baseplate Solvent
    Welded to Pipe
                             4-m. PVC Pipe
                   SANITA1RE
                      Polyethylene Disc
     Gaskot
   Control Ofihce and
   Chock Valvo
                                    Threaded
                                    Retainer Ring
    Baseplate
Mechanical Wedge Section
for Attaching Baseplate
                 OY AIRAM
Disc diffusers  are  available  in  diameters  of
approximately 18-24 cm (7-9.5 in)  and thicknesses of
13-19 mm (1/2-3/4 in). Except for one design, all discs
consist of two flat parallel surfaces. For this exception,
a raised ring slopes slightly downward toward both the
outer edge  and center of the disc. It is  claimed that
the  uniform  profile  aids in  producing  uniform air
discharge across the entire disc surface (37).

Two disc designs include a step on the outer edge for
the  purpose  of improving  uniformity  of  air flux and
effectiveness of the  seal at  the  diffuser edge. The
step also  facilitates the use of an  O-ring seal that is
less subject to  the adverse effects of compression set
of the seal element.

As with dome  diffusers, the  disc is  mounted on  a
plastic (usually PVC)  saddle-type baseplate.  Two
basic methods  are used to secure disc media to the
holder: a  center bolt or a  peripheral clamping  ring.
The center-bolt method is similar  to  that  used with
domes. A soft, rubber gasket is placed between the
diffuser and baseplate. The bolt assembly includes  a
washer and a  gasket. The same  precautions should
be taken for  bolt and gasket  materials,  as described
above for domes.

The more common  method of attaching the disc to
the holder is  to use  a screw-on retainer ring. Several
different  gasket  arrangements are  used with  the
threaded collar. They include a  flat  gasket  placed
below the disc,  a U-shaped gasket that covers a small
portion of the top  and bottom and  the entire  edge of
the disc, and  an O-ring gasket placed between the top
of the outer periphery of the disc and the retainer ring.
The latter arrangement provides a seal between the
vertical outer surface of the diffuser and the vertical
inner surface of the  holder.  The  baseplate normally
includes small raised ribs to aid in obtaining an airtight
seal between the gasket and the baseplate.

The retainer  ring method of attaching the diffuser to
the holder has  potential  advantages over the center-
bolt  method.  As diffusers become fouled,  excessive
amounts of air are discharged  from the edges and the
area around the center-bolt washer (38). Although not
specifically documented for  discs  under controlled
conditions, this  nonuniform  airflow  could  reduce
system  OTE. The retainer ring is  likely  to  minimize
these problems.  In  addition,  breakage of diffusers
from over-tightening the bolt, or  air leakage problems
from  gasket  compression  set or stretching of  a
nonmetalfic bolt can be eliminated.

Two methods are  used to attach disc diffusers to air
piping. The first is to solvent cement the baseplate to
the PVC header prior to shipment to the job  site. To
avoid future additional costs associated with replacing
sections of pipe, the original design  should include all
the  baseplates  needed  to  meet  future  design
requirements  for the  system.  During the early life of
the treatment plant,  not all the diffusers are installed
and plugs  are  simply  inserted  in the  unused
baseplates.

The second  disc  diffuser  attachment method uses
mechanical means.  Either a  bayonet-type holder is
                                                    12

-------
forced into a saddle on the pipe, or a wedge section is
placed around the pipe and clamps  the holder to the
pipe.  Except for  one manufacturer that employs the
wedge clamp  method of  attachment and ships units
preassembfed,  the pipe arrives at the job site  with
only the  holes drilled. This technique makes shipping
the pipe  somewhat easier (less bulky) and can reduce
damage  that may  occur  during  shipment  or
installation.  With  these types  of  designs,  holes for
additional diffusers can be  predrilled and plugged or
drilled later.

Disc diffusers  also include individual  control orifices in
each diffuser unit. Designs employing the  bolt method
of attachment  usually use a hollow bolt with an orifice
drilled  in  its side. The other designs use either an
orifice  drilled in the bottom of the diffuser holder or a
threaded inlet in  the base where  a  small   plug
containing the  desired  orifice  can be inserted.  The
diameter  of the  orifice is similar lo  that used  with
dome diffusers.

Ceramic  and plastic  media disc  diffusers usually have
a design  airflow of 0.24-1.4 Us (0.5-3 scfm)/diffuser
The most economical operating range will, however,
depend somewhat on diffuser size. The smaller 18-cm
(7-in) diameter discs are usually operated  at airflow
rates  of  0.24-0.9  Us (0.5-2 scfm)/diffuser,  similar to
those used for dome diffusers. For the larger discs,
with diameters of 22-24 cm (8.5-9.5 in), typical lower
and upper  airflow  limits  are 0.26-0.43 L/s (0.6-0.9
scfm)/diffuser  and  1.2-1.4  L/s  (2.5-3 scfm)/diffuser,
respectively. Prolonged operation at  flow rates  <0.24
L/s (0.5   scfm)/diffuser is not  desirable  with discs
because   insufficient air is available to ensure  good
distribution across the entire surface of the media. In
those applications where  operation above 0.9 L/s (2
scfm)/diffuser  is desirable, the  control orifice should
be  sized so that the  headless produced  does  not
adversely affect the economics of the system.

2.3.4.2 Perforated Membranes
Like their ceramic  and rigid porous plastic  media
counterparts,  several different sizes and  shapes of
perforated membrane  disc  diffusers  are available
(16,18-20). They range in diameter from approximately
20  to 51  cm  (8-20  in). Although one design uses a
convex shape, most are flat in the air-off position. The
convex shape  reportedly helps prevent the media from
wrinkling (19).  Figure 2-7 shows  schematics of several
perforated disc ditfusefs.

As with  perforated  membrane   tube  diffusers,
perforated  membrane  discs require support.  The
support  base  is  made from  a number  of different
types of  thermoplastic materials including  polyamide,
PVC, and polypropylene. The membrane  is usually
secured  to the  base  around  the  periphery by a
clamping  ring or wire, or a screw-on retaining ring. In
one case, however,  the elasticity of  the membrane is
enough to  secure it to the base.  When the air is
Figure 2-7.   Typical perforated membrane disc diffusers.
                               Threaded Retainer Ring

                               Perforated Membrane
                               Disc
                                Subplate with Center
                                Aperture
                                Orifice

                                 Baseplate
                                  PVC Air Distribution
                                    Piping
                       Sanitaire
applied, the membranes will flex upward approximately
6-64  mm  (0.25-2.5  in). If the  membrane  flexes  in
excess of the  manufacturer's  recommendations,
difficulty could  be experienced  in  maintaining  good
airflow distribution. Therefore, some designs  include
an  additional means of  support in the  center  to
prevent overflexing or ballooning. This  center support
can  consist  of a  piston-type   arrangement  or  a
retaining bolt. The  piston type is  designed to prevent
backflow.

It is common with the disc design to feed air through
one or several small holes in the center of the support
base. The  membrane is normally not perforated near
these holes. When the air is turned off, the liquid head
forces the unperforated section of the  membrane
down  over the air  inlet ports. This provides  an
additional means of closure to that of the perforations
that form a seal against the support member.

The  base of the support frame  is  usually threaded.
Common practice is to use a special  saddle that  is
either glued or clamped to the air header for attaching
the diffuser.  In  Europe, where  rectangular stainless
steel piping is often used for the air header, a nipple is
welded to the pipe for attaching the diffuser.

One membrane disc diffuser utilizes a holder identical
to  that  used  for a ceramic  disc (16), while another
utilizes  a  holder identical to  that  used  for a  rigid
porous  plastic disc  (9). By replacing the ceramic  or
                                                   13

-------
Figure 2*7.  (continued).
                                          Circular
                                            enures
 1.


 2.

 3,


 4,


 5.

 6.
EDPM Perforated
Membrane Disc
S,S, Wiro Fastener Assembly
Polypropylene Membrane
HoWor
Ftow Balancing Orifice
Plato Assembly
Dillusor Qaskel
PVC Mounting Saddle
ENVIREX
porous  plastic disc medium with  a membrane disc
subplate and the membrane itself, a ceramic  or rigid
porous   plastic  system  can  be  converted to  a
membrane system, or vice versa (Figure 2-7).

For  the  smaller  diameter (20-30 cm  [8-12  in]}
perforated  membrane  disc diffusers, the  recom-
mended airflow range is approximately  0.5-4.7 Us  (1-
10 scfm)/diffuser. For the  larger diameter  (51  cm [20
in)) disc, the recommended range is 1.4-9.4 Us (3-20
scfm)/diffuser.


2.4 Diffuser Layout
Figure  2-8 presents schematics of typical  diffuser
layouts. These layouts are discussed  below  as they
apply to each of the fine pore diffuser types  described
above.

2,4,1 Plate Diffusers
Fine pore ceramic plates of the original 30-cm (12-in)
square  design (21,22) are most often grouted  into the
basin floor. Downcomer pipes deliver the  air to  open
concrete channels below the plates. The channels act
as distribution manifolds.

The original plate diffusers can be installed in  either a
total floor coverage or  spiral roll pattern. Total floor
arrangements may  include closely  spaced  rows of
diffusers  running  either  the  width (transverse)  or
                                                  Figure 2-7.   (continued)


                                                         Seal Ring
                                                                    EDPM
                                                                    Perforated
                                                                    Membrane
                                                                    Disc
                                                                                              Polymid Dish
                                                                                         Note:
                                                                                         Shaded Area of
                                                                                         Membrane Non-
                                                                                         Perforated
                                                                                         - Adjustable
                                                                                          Pipe Support
                                                                                        r_
                                                                                         Bottom of
                                                                                         Aeration Tank
                                              ROEDIQER
                                         S. S. Lift Limiter and
                                           Backllow Valve
Rubber Perforated
 Membrane Disc
                                                              Support Disc Threaded
                                                                         Connection

                                                                         EIMCO
                                                                                                S.S. Clamp-
                                                                                                 ing Ring
                                                   length (longitudinal) of the basin or incorporated into a
                                                   ridge and furrow design (21). Spiral roll arrangements
                                                   include rows of plate  diffusers typically located along
                                                   one or both sidewalls of aeration  basins.  The  total
                                                   floor layout will produce a  higher  OTE,  whereas the
                                                   spiral  roll  pattern  will produce  more  effective  bulk
                                                   mixing of the mixed liquor.

                                                   The newer 30 cm x 61  cm (12 in x 24 in) and 30 cm x
                                                   122 cm (12 in x 48  in) ceramic  and  porous  plastic
                                                   plate diffusers  (23) are not attached to  the basin floor.
                                                   With the flexibility afforded by their individual  rubber
                                                   air feed hoses, they  can  be  moved, within  limits,
                                                   about  the basin floor to  form different  layout
                                                   configurations. If a change to a significantly different
                                                   configuration is desired,  new hoses may have to be
                                                   provided  to   accommodate  the  revised  diffuser
                                                   positions.

                                                   2.4.2 Tube Diffusers
                                                   Most tube  diffuser assemblies include a 19-mm (3/4-in
                                                   NPT)  threaded nipple  (stainless steel  or plastic) for
                                                   attachment to  the  air  piping system.  This design
                                                   makes the  tubes especially well suited for  retrofit or
                                                   upgrade  applications  since many coarse  bubble
                                                    14

-------
Figure 2-7.  (continued).
Figure 2-8.  Typical diffuser layouts.
                                        Threaded
                                        Retainer
                                        Ring

                                      Perforated
                                      Membrane
                                      Disc
                                       Subplate with
                                       Peripheral
                                       Aperatures
       Orifice
                                      Baseplate
                                         Gasket
                                         PVC Air
                                         Distribution
                                         Piping
                                       Mechanical
                                       Wedge Section
                                       for Attaching
                                       Baseplate
                    OY AIRAM
diffuser  systems  use  the  identical  method of
attachment.

The air headers on  which the tubes are  mounted are
usually fabricated from PVC, CPVC, stainless steel, or
fiberglass  reinforced  plastic.  Carbon  steel  is
sometimes  used  but is  less desirable  because of
corrbsionJnside the pipe. Thus,  threaded adapters or
saddles are either  glued,  welded,  or  mechanically
                                                             1
                                                                          Wastewater Flow
  Single Spiral Roll
Mid-Width
 (Center)
Dual Spiral Roll
              1
                                                                   Cross Roll
                          Total Floor Coverage
attached to the pipe at the points where the tubes are
to be  connected.  The  actual  diameter  of the  air
headers will  vary depending  on  the  number  of
diffusers to be installed and the design airflow rate.

The  depth of tube submergence  in the basin  varies
depending on the application. In new installations, the
tubes  are  usually  placed as close  to the  floor  as
possible, typically within 30 cm (1  ft) of the bottom. In
retrofit  applications, the  discharge  pressure  of the
existing  blowers  may  control   the  depth  of
submergence.  The tubes are  normally  installed  at
either  the same elevation as the original system or
possibly at a somewhat greater distance  off the floor
to compensate for  any  increase in headloss  incurred
through the fine pore media compared with the coarse
bubble  devices they  are replacing.  The  air  headers
are usually secured to the basin floor  with adjustable-
height, stainless steel pipe supports.

Tube diffusers are  often installed  down the center or
along one  or both long sides  of the  aeration basin
(mid-width, single spiral roll, or dual spiral roll pattern,
respectively).  In  some  cases,  the headers  are
mounted on  mechanical lifts  to allow removal  of  air
headers and  diffusers  for inspection and  cleaning
without dewatering the basin. On the header itself, the
tubes can be installed along either one side (narrow
band) or both sides (wide band)  of the  pipe.
                                                    15

-------
Tubes can also be Installed  in either a cross roll or
tola! door coverage pattern.  In the cross  roll design,
tha headers are placed across the basin width and the
spacing  between diffusers,  30-91 cm (12-36 in), is
small  in  comparison with  the  spacing  between
headers, 3-9 m (10-30 ft). In the total floor  coverage
pattern, in which the headers can be placed  either
across the basin width or along its length, the distance
between headers and the spacing  between diffusers
on the headers approach the same value. Total floor
coverage will  usually achieve higher  OTEs  than the
other configurations.

Spiral roll configurations  normally provide  better bulk
mixing throughout the basin  than  either total  floor
coverage or cross  roll  patterns.  Some designers
believe  that  cross-roll  patterns  may not  provide
adequate mixing. One potential disadvantage of the
cross roll and total floor coverage designs is that the
location and amount of piping required usually makes
the use  of  mechanical liftouts impractical.  There is
only one known installation in this country (39) where
a total floor coverage liftout system has been installed.

2.4.3 Disc and Dome Diffusers
Although  their shape and operating  characteristics
may differ, the typical air  piping and diffuser layout is
identical for both disc  and dome diffuser systems. The
air distribution manifold should preferably be  PVC, the
compounds  of  which  are  described in ASTM D-1784
or ASTM D-3915 (13), cell classification  12454B or
124524,  respectively. If D-1784 is specified, the pipe
malarial  should be limited  to  stress-related
compounds. The  PVC should be UV-stabiiized with 2-
percont  minimum  Ti02,   or  equivalent.  The
specifications, dimensions, and properties  of the pipe
itself conform  to either ASTM D-2241  (13) or D-3034
(13), depending on the outside pipe diameter.

The piping network usually has a nominal 10-em (4-in)
diameter, with an  actual O.D. of 10.7-11.4 cm (4.2-4.5
in). The  wall thickness is also variable, typically 3.0-
4.3 mm (0.12-0.17 in). Sections of pipe are connected
with  gusseted, mechanical expansion joints to allow
for  expansion  and  contraction of the PVC over a
temperature range, typically about  100F,  appropriate
to climatic temperature  extremes  anticipated.  Pipe
supports,  usually  made from PVC or  stainless steel,
are provided to secure the system to  the basin floor.
Tho  support  consists of  a  cradle  or saddle and a
hoiddown strap. The strap is either secured with a bolt
or snaps into  place. Pipe supports are adjustable to
compensate for variations in the basin floor elevation.
Extreme  variations in  basin floor elevation can cause
problems  in using standard  pipe supports. The pipe
support is attached to the basin floor with  one or two
stainless steel bolts and  concrete anchors. The PVC
strap and pipe  support have  experienced some
breakage  problems in the past.  To eliminate  these
problems, or in cases where  the diffusers are  to be
mounted a significant distance above  the basin floor,
i.e. 61 cm (24 in), stainless steel pipe supports can be
used.

Discs and domes are generally installed in a total floor
coverage or grid pattern.  In some  cases  where
oxygen  demand is low  and mixing may control the
design  (near the end of long narrow  basins), the
diffusers can be placed  in tightly  spaced rows  along
the side or middle of the basin to create  a spiral or
mid-width mixing pattern,  respectively  (40). The
diffusers are  usually  mounted  as  close to  the  basin
floor as possible, within 23 cm (9 in) of the highest
point of the floor being  typical. As mentioned in the
discussion   of  tube  diffusers,  the  depth  of
submergence in  some  retrofit applications  may  be
controlled by the available blower discharge pressure.
2.5 Characteristics of Fine Pore Media
Many properties can be used to characterize fine pore
media. Knowledge of these characteristics promotes a
better aeration system design for a specific set of site
conditions.   Appropriate  attention  to   these
characteristics in the design phase may also lead to
reduced O&M problems during  the life of the system.
Many of these characteristics are not routinely defined
or available for specific media.

Reasons  for defining   media  characteristics  as
thoroughly as possible include providing:

1. a means  of short-term (batch-to-batch) and long-
   term (year-to-year) quality control,

2. baseline performance and a basis for determining
   changes  in  media  properties under  process
   conditions over time,

3. quantitative  information  needed  in   design
   specifications, and

4. an indirect  indication of changes  in  field
   performance.

The following sections present a general discussion of
selected fine pore media characteristics.

2.5.1 Physical Description
For each  specific  aeration medium,  a raw  material
description  should  be  provided.  Usually,  the
equipment supplier  (in  conjunction   with  the
manufacturer) selects the optimum  materials.  By
knowing what these  materials  are, the designer can
ascertain whether  any  unique constituents  in the
wastewater or cleaning chemicals will  be incompatible
with  the  diffusion  media. Materials that  should  be
specified  include  the grit  and  type  of  binder for
ceramic media, and the. chemical composition of the
principal constituents comprising porous plastic media
and perforated membranes.
                                                  16

-------
2.5.2 Dimensions
For  each  type  of  diffuser  element,  all critical
dimensions should be stated. Other than the obvious
reason  of  assuring  compatibility with the  mounting
base or holder, these  dimensions  also  serve  to
establish  a  baseline.  After  a  certain  period  of
operation, the  dimensions  can  be  checked  to
determine  whether  changes  have  occurred.  It  is
possible that some  materials  may warp,  expand,  or
stretch  with time. Only if  baseline dimensions  have
been established can these changes be detected.

2.5.3 Weight and Specific Weight
The  weight of  the diffuser and its apparent specific
weight (calculated) should also  be determined. These
two characteristics can be used for quality control. A
wide  range  in  weight or  specific weight  between
several  apparently similar diffusers could  indicate an
unacceptable product (21). In this situation, it is likely
that  a variance  in weight will  also result in a variance
in some of the other parameters such  as dynamic wet
pressure (DWP - see Section 2.5.11), bubble release
vacuum (BRV - see Section 2.5.12), or uniformity (see
Section 2.5.13).

Initial or clean  media weight  and  specific  weight
should  also  be  determined  to establish  baseline
conditions. As  with  media dimensions, changes  in
media weight and specific weight can occur with time.
These changes can  result from absorption of liquids,
leaching of certain resins or additives, or dissolution of
the  base material.  A single  weight  measurement
should usually detect any changes. A better approach,
however, is to obtain a series of measurements over
time. Also,  if  media   weight  is  determined  in
conjunction with other physical or mechanical  testing
(e.g.,  tensile  strength),  an  indication of  any
degradation can be determined.

2.5.4 Permeability
The  permeability  test may have some application for
quality control. However,  because permeability is  such
an  inexact  parameter,  it  is  of  little  practical
significance  in  characterizing  even  ceramic-type
diffusers.  The  following   discussion  is included  to
define the term  since it is often  mentioned in regard to
fine  pore  aeration media  and point out some of the
many  shortcomings of the  test.   Other  more
meaningful parameters  are  discussed in  Sections
2.5.11 and 2.5.12.

Ceramic diffuser media  are usually  available  in a
number of different grades. Grades are distinguished
by pore size,  which is controlled by the size of the grit
material,  binder type  and   content, and  firing
temperature.  Because measurement of pore size was
not  practical  for a  large  number of  samples, the
ceramic  manufacturing industry developed  a  simple
but somewhat  arbitrary  test  to differentiate between
grades of media.  This  test was referred to as  the
permeability rating.

Permeability is a  measure  of  a  porous medium's
frictional resistance to airflow. It is  an empirical  rating
that relates flux rate to pressure loss  and pore size or
pore volume.  Permeability is usually defined as  the
amount of air at standard conditions  that will pass
through 929 cm2 (1  sq  ft) of 25-mm (1-in) thick  dry
porous media  under  a differential pressure equivalent
to 5 cm (2 in)  w.g. when tested at  room temperature.
The  airflow value obtained  (scfm)  under  these
conditions is called the permeability (perm) rating.

Unfortunately,  the permeability measurement does  not
provide a true basis  for  comparison of  media
performance. The same  permeability rating could be
obtained  from  a diffuser with  a few relatively large
pores  or  a multitude  of  small  pores.  Also,  two
diffusers  with  the  same pore structure  would have
different ratings if they have different thicknesses.

Most ceramic  and some plastic  media specifications
now include a  test for permeability. Unfortunately,  the
ceramic industry has not  "standardized"  this  test
procedure. The  early specifications  were developed
for 30 cm  x 30 cm  (12 in x 12 in) plates, 25 or 38 mm
(1 or  1.5  in) thick. Today, specifications are  needed
for products  of  various shapes,  densities,  and
thicknesses, often of poorly defined effective  area.
Attempts have been  made to apply  the principles of
the test  through  a  parameter  known  as  specific
permeability (41).  The   procedure  for  determining
specific permeability  
-------
4.  The effects of surface tension are not accounted
    for since the test is conducted on a dry basis,

2.5.5 Perforation Pattern
Each perforated membrane  diffuser manufacturer
uses a slightly different perforation pattern. In fact, the
same manufacturer  may use one type for a  disc
diffuser and a different one for a tube assembly. One
reason for changing the perforation pattern between
diffuser types is that the various  shapes  result in
different  stresses  and  resulting  strains  on  the
membrane. Thus, a pattern is developed that will best
be  able to withstand the magnitude and direction of
the loads that will be applied.

Perforations  can  differ  in regard  to  type  (holes or
slits),  size,  orientation,  and  density  (number
perlorations/diffuser surface  area). Usually,  the
specific pattern has been developed over time and is
not  a  feature that changes from  job to  job.  It is
important, however,  to detect any changes that may
occur to the membrane  over time.  Measurements or
photomicrographs  can be made and compared to
baseline conditions  to determine  if the  perforations
have either expanded or taken on  a permanent  set.
Also, in situations  where  replacement diffusers  are
purchased, the perforations should be similar to the
original diffusers to ensure compatibility.

2.5.6 Strength
Structural or physical  strength is  also an important
media  characteristic. The diffusion  media must be
strong enough to withstand: 1) the static head of the
water above the diffuser  (in cases  where  the air
supply  is  shut off), 2) the forces  applied  when
attaching  the media  to  the  diffuser  holder, and 3)
stresses and shocks of reasonable handling, shipping,
and maintenance.  Strength,  as  described in  this
section, applies only to rigid diffusion media (ceramic
and porous plastic).

For discs and  domes,  measurement of  strength
usually  involves supporting  the diffuser in  a fashion
similar to  that used  for  the final  assembly and  then
applying a load to a 25-mm (1 -in) diameter area  in the
center of  the diffuser. Using this method, developed
primarily for  ceramic diffusers, acceptable loads for
the ceramic  material  are 270-455  kg (600-1,000 Ib).
Diffusers that use a peripheral clamping method do
not require the same flexural strength as those  that
utilize the center bolt method of attachment. Slightly
different techniques have been developed to evaluate
the strength of nonrigid diffusion material.

2.5.7 Hardness
Hardness  is  an  important media characteristic for
perforated membranes because it  is an index of the
resistance of an elastomer  to  deformation.   It is
measured by pressing a ball or blunt point into the
surface of the material (13). The most commonly used
instrument to measure  hardness  is  the durorneter.
Different types  of instruments  are  available  that
produce  a  range  of  overlapping  scales.  Shore  A
durorneter readings are the  most common,  although
Shore D  readings are sometimes specified. For each
range, the higher the durorneter reading, the  harder
the material.

Hardness  can  be  used  to detect changes  in
membranes. Softening or hardening of the membrane
material may be  indicative of environmental attack. A
softening, for instance, may indicate the absorbency
of a solvent into a thermoplastic. On removal from the
wastewater, the solvent may  evaporate and the  plastic
may resume  its original hardness. Hardening, on the
other hand,  may  indicate the loss of an additive  or
plasticizer over time or attack by some type of acid. In
any regard, changes in hardness should be noted and
other characteristics (BRV, DWP, and OTE)  evaluated
to determine if media performance has been affected.
In some  cases, changes in  hardness may  denote a
change  in  media characteristics  that  will  not
necessarily shorten service life.

2,5.8 Environmental Resistance
The constituents  found in typical domestic wastewater
are usually  not  excessively harmful  to  fine pore
diffuser media. Some industrial wastes, however,  may
contain  compounds  that   will  result  in  physical
degradation  of the  media, especially for  the porous
plastics  and  perforated  membranes.  Some
compounds of potential concern  include mineral  and
vegetable oils, organic solvents, and  strong  oxidizing
agents.

It is common practice to use chemicals to clean  fine
pore diffusers when they become  fouled.  Although
hydrochloric  acid (HCI) is the most common cleaning
chemical  (either as a liquid or gas), other compounds
have  been  used.  These  include  inorganic  acids
(sulfuric,  nitric), bases (sodium hydroxide,  potassium
hydroxide, ammonium  hydroxide), carboxylic  acids
(formic, acetic), detergents, and organic solvents.

Air-phase foulants including  oxidants such  as  ozone
are also of concern. Ambient concentrations of these
oxidants  are  aggressive to a number of elastomers
used in  wastewater treatment (e.g.,  nitrite, styrene
butadiene rubber, polyisoprene, and  natural rubber).
The effect is  Intensified when the materials are under
stress.  Ozone-resistant membrane  materials should
be  developed. In  all  cases, assurance  should  be
obtained  (from the  literature, supplier, or tests)  that
the material  selected  has the requisite resistance  to
these oxidants  in the concentrations expected  in
service. The same assurances should be obtained for
those applications in which fine pore devices are used
to disperse ozone.

Most ceramic diffusers have  been marketed for many
years and, except for some slight degradation when
acids are used  for cleaning, appear to  be quite
                                                  18

-------
resistant to chemical attack.  Porous  plastics  and
perforated  membranes  have been developed in the
last  10-15  years.  Testing  has  shown that  wide
variations exist among the generic groups of plastics
in their chemical  resistance  and  physical  and
mechanical properties (14).

Even  within  a  generic  classification,  different
formulations  may  result in  a  wide   variation in
performance in a particular environment. To enhance
the physical properties  of  plastics  and elastomers,
manufacturers  often  incorporate additives into  their
membrane formulations. Therefore, the actual material
may differ  in composition from  published data.  As a
result, environmental resistance  testing should  be
conducted  with the  actual diffuser  media  in  both
stressed and  unstressed conditions. Much  of  such
testing  may  serve  as  a qualification rather  than  a
routine requirement.

In conducting  these  tests, several of the media
characteristics discussed in this section  can  be  used
to establish if any chemical degradation  is occurring.
Some of the more  important  include  changes in
dimensions, weight, strength, and DWP.

2.5.9 Miscellaneous Physical Properties
Several additional physical properties can be used to
characterize fine pore aeration media. A  partial list of
those properties that may be of importance, especially
for porous plastics and perforated membranes, is:

1. Tensile strength (stress and strain)
2. Elongation at failure
3. Modulus of elasticity
4. Creep
5. Compression set
6. Tear resistance
7. Strain corrosion
8. Solvent extraction
9. Ozone resistance

Since most of these properties are  temperature
dependent, data  should  be  developed  over the
expected  range in  temperature.  Definitions  and
procedures can be found in the ASTM Standard (13).

Since very little information exists regarding the above
six items,  it is premature  to  attempt to  place
guidelines on what would be considered acceptable. It
is possible, however, to use  these parameters  for
assessing the effects of field operation. Any  changes
over time may be an indication  of degradation of the
media or impending problems (e.g., significant creep
with a polyethylene tube diffuser may eventually lead
to air leaks at the end caps).  These properties may
also be important in evaluating the general quality of a
membrane material.  For example, the  ratio of the
tensile strength at design operating conditions and at
failure could be  calculated. The higher the ratio, the
greater the safety factor.

2.5.10 Oxygen Transfer Efficiency
The OTE that  can be achieved  with fine pore diffuser
media  over their design  life  is likely their most
important characteristic. Most projects require  a  shop
or field  oxygen  transfer test  to  verify diffuser
performance with regard to OTE. OTE information can
also  be obtained  in  using laboratory (bench-scale)
apparatus  to aid in characterizing  the  media.
Laboratory tests are not intended to be a substitute for
shop or  field  testing, or for  predicting field OTE.
Rather,  they  should  be used  to evaluate  relative
differences in performance between  two or more
diffusers.

A  typical  laboratory  setup will consist  of a small
column, 61-91  cm (2-3 ft) in diameter and 2-3 m (7-10
ft) high. The diffuser(s) to be tested is installed in the
column  and clean water OTE  is determined  over  a
range of airflows  (ideally two or three different rates).

Clean  water  OTE is usually  determined  by  the
nonsteady-state test procedure described in an ASCE
Standard (42). A second procedure that also seems to
produce excellent results involves a steady-state test
(10,43). With this technique, two diffusers are  usually
placed  in the test basin. The airflow to the first diffuser
is  set at the desired rate; a very low airflow rate (< 5
percent of the first) is set for the second diffuser.  The
basin is covered, sodium sulfite  is added at a constant
rate to achieve steady-state DO  conditions, and the
off gas is analyzed for oxygen content. After sufficient
data are obtained on one  diffuser, the air supply is
rapidly switched,  causing the second diffuser to
operate at  the specified  rate and the air to the  first
diffuser to be reduced  to maintain  only a slight
positive pressure. By repeating  the cycle  several
times, the relative oxygen transfer performance of the
two diffusers  can be obtained. Details  of this  test
procedure are  presented in Appendix B.

Both  techniques (steady-  and  nonsteady-state)
produce acceptable results. If an off-gas analyzer is
avai'dble,  the  steady-state   procedure  is  more
desirable.  Data can be generated more  quickly,  and
the ability to compare two diffusers on a side-by-side
basis is an additional advantage.

It must be emphasized that OTE values determined in
laboratory tests  are for  comparison  purposes  only.
They can  be used as a quality  control technique, for
detecting  changes  in  performance that  may occur
during  field  operation,  or  for  assessing   the
effectiveness  of various cleaning  agents.   Ideally,
laboratory OTE tests will be conducted in conjunction
with  other  media  characterization tests  (e.g., DWP
and BRV) to examine what correlations might exist.
                                                   19

-------
2.5.11 Dynamic Wet Pressure
DWP is  a very important characteristic in evaluating
and selecting  porous  media.  DWP, the  pressure
differential (headless)  across the  diffusion  element
when  operating  in  a submerged  condition,  is
expressed in cm (in) w.g. at some specified airflow
rale (38). As a general rule,  the  smaller the bubble
size,  the  higher the DWP. While  small bubbles may
increase  OTE, the  additional power required  to
overcome the  higher  headloss  may negate any
potential  savings.

DWP can be measured in the laboratory or in the
field.  Accurate  field measurements are  often more
difficult to obtain. Figure 2-9  presents a schematic of
a typical  field  setup  for in-situ monitoring  of  DWP.
This  particular arrangement  includes a  bubbler  or
static tine (Tap 3), a line to sense pressure  in the  air
header (Tap  1), and a connection in the body of the
dlffuser holder  immediately below  the media  (Tap  2).
Tap 2 is  considered optional since the physical design
of some  diffusers makes it very difficult or impossible
to locate this connection.

Using  the three taps  shown  in  Figure 2-9,  it is
possible  to measure the headloss across the  diffusion
media (Taps 1  and 3) as well as estimate the airflow
rate through the  diffuser based  on the headloss
across the control orifice (Taps 1  and 2). If Tap 2 is
omitted,  the DWP measured will include  inlet losses
as well as media losses.

Extreme  care  must be  taken  to  purge all  moisture
from both pressure sensing lines  and set the airflow
through the bubbler tube at the lowest possible rate
that will  not induce  a significant headloss. Also,
because  of the relationship  between standard and
actual airflow rates, the headloss in the field will be a
function  of  diffuser   submergence.  If   a field
measurement of DWP  is  made,  the units  used  in
describing the flow (standard  or actual) must  be
known and  corrections made before different media
are compared.

An  alternative  to  measuring DWP  in the  field  is
laboratory measurement of DWP for selected diffusers
removed  from  the  aeration basin. A  supply  of test
diffusers  may  be  provided  from  removable  test
headers.  Test  headers  having  four diffusers have
been used in several  plants (10,44,45). A typical test
header is shown in  Figure 2-10.  To reduce the weight
of the lest header assembly so that diffusers can  be
more easily removed, a two-diffuser arrangement may
be used in place of the four-diffuser assembly.

The  removable  headers can  be  equipped  with
pressure  taps  for  in-situ monitoring of DWP. These
field readings can be supplemented with more precise
measurements  of  DWP  in  the  laboratory. It  is
imperative that the test  header be operated at airflow
rates similar to the  full-scale  system. Airflow  rates to
individual diffusers can be easily determined for units
having fixed-size flow  control orifices  by measuring
the pressure drop across the orifice. Airflow rates can
also be measured using one or more rotameters. The
use of removable headers requires significant operator
attention and commitment if the data collected are to
be of value.

Although the diffusers must first be removed from the
aeration basin, laboratory measurement of DWP  will
usually be more accurate.  However,  the diffusers
must be  kept wet and rapidly transported to  the
laboratory  (ideally,  within  24  hours)  to  prevent
changes in the characteristics of the  foulant from
occurring.  In  determining DWP, it  is important that
porous diffusers (ceramic and plastic)  be allowed to
soak for several hours (plastic materials may require a
much  longer  period)  prior to testing to  completely
saturate all the pores.  Since the actual  headloss  will
be a function of the degree of water saturation in  the
diffusers, a slightly different curve will be obtained if
the air is started at a low flow rate and is increased, or
vice versa. Standard practice is to purge the  media at
the upper airflow rate for a predetermined interval (5-
10 minutes), then record subsequent headloss values
as  the airflow  rate  is  decreased. At  least  three
different airflow rates should be used to define DWP
adequately.

DWP testing  can  be  conducted  in  an  aquarium  or
similar small   basin using an  apparatus  depicted  in
Rgure 2-11. Usually, the diffuser holder will contain a
control orifice  to aid in  airflow distribution throughout
the  system.  In conducting the  DWP test, the
apparatus should be set up so that the losses across
the orifice are  not included in the DWP determination.
Orifice losses, depending on the airflow rate,  can
become  quite high.  This  may  mask the  losses
associated  with the media  that represent the main
purpose of the DWP test.

The  porous media now  in use have a DWP  of about
8-51  cm  (3-20  in) w.g. when operated  within the
typical or specified airflow ranges. The specific value
depends on the type of material, surface properties,
airflow rate, and diffuser thickness.  For ceramic and
porous  plastic materials, the headloss  vs. airflow
curve is linear  over the typical operating range and the
slope is relatively flat. For these two  types of material,
most of the  DWP is associated with the  pressure
required to form bubbles against the force of surface
tension. Only a small fraction of the DWP is required
to overcome  frictional  resistance. Because surface
tension is not  affected  significantly  by  material
thickness or airflow rate, these two  factors  are only
minor contributors to the overall DWP (38).

2.5.12 Bubble Release  Vacuum
The  BRV  test provides a means of determining the
bubble release vacuum at any point on the surface of
a diffuser  element relative  to other points on   its
                                                  20

-------
 Figure 2-9.  On-line device for monitoring DWP of fine pore diffusers.
                                                                            Air Source
                  Liquid Surface
          Diffuaer •
         Plenum •
                                                       Manometers
                                                     Tap 2
                                                                                   Bubbler Pipe
                                                                                  Tap3
                                                                                   / Bubbles
                                                                              Air Flow Control Orifice
                                            Tap 1
                                               Header
surface.  Normally,  BRV denotes the average  of  the
point  BRV values  for  a specific diffuser.  This test
procedure  is useful for assessing the  uniformity of
pores on the surface of diffusers under clean as well
as fouled conditions (46)  (see Appendix  B for test
details).

BRV,  as suggested by the  name, is a measure of the
negative pressure in cm (in) w.g. required to form and
release bubbles in tap water from a localized point on
the surface of  a  thoroughly-wetted  porous diffuser
element. This is accomplished by applying a vacuum
to a small area on  the  working  surface of a  wetted
diffuser and measuring the differential  pressure when
bubbles are released from the diffuser at the specified
air flux at the point in question. The air flux is usually
set quite low (10 L/s/m2  [2 scfm/sq ft], which is within
the range of flux values used for porous diffusers) to
minimize frictional drop in the  BRV measurement and,
in cases  where the diffusers are fouled,  to avoid
removing the fouling material during the BRV test.

The  BRV  test  was  initially  developed for use  on
ceramic and porous plastic diffusers.  It can also be
used  on  perforated membrane diffusers if  the test
equipment is modified  to account for the  greater
distance between pores - holes or slots in  this case
(10).

Both DWP and BRV provide a measure of the bubble
release pressure. For the DWP test, the measurement
is made across the entire diffuser.  In cases where the
diffuser is partially fouled, air may  simply short circuit
the fouled area. Thus, DWP may not indicate the true
condition of the media. The BRV test, on  the  other
hand,  measures  the pressure  at a  small  localized
area. For this reason, BRV has  been  found  to be the
more  sensitive test.  In  studies evaluating diffuser
fouling, the use of BRV data, averaged for  the entire
diffuser,  permits  significant shortening  of  the  test
period  required  to  obtain  definitive conclusions
regarding diffuser fouling (46).

It should  also be noted II  ,t  it may be possible  to
combine  various  measurement  parameters  in
characterizing  fine pore diffusion media. It  has  been
suggested  that  the  ratio  of  DWP  to  BRV,  when
measured at the same  air flux,  may be  a  better
indicator of diffuser fouling than  either parameter
alone (44,46,47).  DWP measures the  overall pressure
that forms bubbles at a specific air flux over  the entire
diffuser surface. BRV measures  the average negative
                                                   21

-------
Figure 2-10. Removable test header.
      Pressure Lines from
      Plenums Connect to
      Monitoring Box
    Dilluscr
                                  Air Supply for
                                  Bubbler Tube
                                Air Supply from
                                Existing Source
                             Bubbler Tube
                                  Difluser #3
                                     Diffuser #4
                                                      Pipe
                                                         Bubbler Tube
                  a. Header Layout
                                                                              Plenum


                                                                            b. Monitoring Box Detail
                                            Diffuser


                                             Orifice

                                             Header
pressure required to form bubbles at a specific air flux
over a limited region of the diffuser surface.

The ratio of DWP to BRV  is closely  related to the
fraction of the diffuser area  that is  actually  emitting
bubbles. As  the ratio decreases, less  of the diffuser
area  is  operating  for the  same airflow rate.  This
means that the active areas are operating at higher
localized air fluxes, potentially causing the formation of
larger bubbles and  resulting  in  lower  OTEs.  A
DWP:BRV of 1  (both  measured  at the same air flux)
suggests  clean uniform media,  whereas  fouled
diffusers  frequently exhibit  DWP:BRV  values
substantially  lower than 1  (see Chapter 3).

The relationship between permeability and  BRV is of
interest from specification and quality control points of
view. Many  specifications  require that  permeability
tests be  conducted  to  control  uniformity,  effective
pore size,  and backpressure  of the diffuser elements
in service. As discussed in Section 2.5.4, there is no
recognized  standard procedure  for determining
permeability. Furthermore,  since the test  measures
only an overall resistance to flow, it gives no indication
of uniformity  within an individual diffuser being tested.

On the other hand, the BRV test, which is conducted
at air (luxes  and pressure differences  comparable to
service values  and  includes the effects of surface
tension,  is  not subject  to  the  deficiencies  of  the
permeability  test.  Additionally,  the  coefficient  of
variation  of this measurement provides a measure of
diffuser uniformity (see Section 2.5.13). It  has been
suggested, therefore, that the BRV test is a far more
applicable  and  meaningful  field  test  than  the
permeability test as  presently practiced for  specifying
diffuser uniformity and effective pore size (48).

In  reasonably  uniform  diffusers of a  given  geometric
shape, a relatively good correlation exists between the
new or original BRV (BRVrj) and specific permeability.
Figure 2-12a presents data obtained from four ceramic
disc  diffusers  of   equal  size  having  a  specific
permeability of 14-50 (48). Since BRV is a measure of
diffuser pore size, it is reasonable to assume that  as
pore  size  increases and  BRV decreases, standard
oxygen transfer efficiency (SOTE) decreases. Using a
laboratory oxygen transfer test procedure described in
Section 2.5.10 (see also Appendix B),  the relationship
between  relative SOTE and specific permeability and
BRV shown in Figure 2-12b was developed (48). It is
apparent for  these  diffusers  that little difference  in
relative  SOTE  exists  for specific  permeabilities
between  14 and  38 (BRV0  = 23 to 10 cm [9 to 4 in]
w.g.).

2.5.73 Uniformity
Uniformity  of individual diffusers  and  the  entire
aeration  system  is  important if high OTEs  are to be
attained.  On an individual basis, the diffuser must be
                                                    22

-------
Figure 2-11.  Apparatus for measuring DWP in the laboratory.
                        DWP = AH - h
                      Liquid Surface
                                   7
                                                           n
                                                              AH
                                                            J
                 Manometer
                                             Header
                                                    *"».,.,-.        h
                                                                                                      Airflow
                                                                                                       Meter
                                                                                                  • Air Source
Figure 2-12.  BRV relationships.
 Specific Permeability
   80
   60
   40
   20
            I	I
                                 a.
i     i    i    i    i
                          4        6
                         BRV0, in w.g.
                                                     10
                              Relative SOTE
                                1.02
                                                                   1.00
                                                                   0.98
                                                                   0.96
                                                                   0.94
                                                                   0.92
                                             I
  Specific Permeability
I       I       I        I
                                     0      10     20      30     40     50
                                            10      7      5      4      2.7
                                                     BRV0, in W.Q.
                                                                                60
                                                           23

-------
capable  of  delivering  uniform  air distribution across
the entire surface of the media. If dead  spots  exist,
chemical or biological foulants  may form  and
eventually lead  to premature fouling of the ditfuser.
Also, if  small areas of extremely high  air flux are
present, larger  bubbles may  form  and  OTE will
decrease.

Most system designers  recognize the  importance  of
uniformity and,  as a result,  will  include a  uniformity
testing procedure as part of the job specification.  A
common practice has been to select random samples
of media from  each batch  during the  manufacturing
run. The diffusers are  placed  in water  for  a  fixed
period to saturate them  and  then tested in  a shallow
basin at a predetermined airflow rate. Usually, a visual
observation  is  the basis for the test. This type  of
qualitative method is  obviously  quite arbitrary. Two
people  are likely  to  have  somewhat  different
definilions for what  constitutes  uniform airflow.  In
other cases, a high airflow from around the periphery
may tend to mask the view of the center of a diffuser
that is completely dead. Because of these  problems, a
visual test is not considered an acceptable method  of
characterizing fine pore media for uniformity.

The use of quantitative  techniques is  a much better
approach for measuring  uniformity. These procedures
actually measure the rate of air  release from different
areas of the diffuser  (38). With the diffuser submerged
in 5-20 cm  (2-8 in) of water  and at an  airflow rate  of
approximately 10 UsfrnZ (2 scfm/sq  ft)  (or at the
recommended design  rate), the rate of air release  is
determined  by  measuring the displacement of water
from an inverted cylinder. Based on  air volume,  time,
and area of the collection cylinder, an air flux rate  is
calculated.

Air flux may be  expressed in several ways. Apparent
flux  is the  airflow rate  per diffuser divided  by the
effective  diffuser area.  For planer  diffusers,  the
exposed surface area is considered the effective area.
For dome and tube  diffusers, the projected area has
normally been used  to calculate apparent flux. Local
(lux  is the local airflow rate  per  unit area of a small
defined segment of  a  given  diffuser.  Effective flux  is
the weighted average (based on area) of the local flux
measurements  for one or more diffusers.  Additional
information on diffuser flux calculations is presented  in
Appendix B.

For diffusers that provide uniform air release (i.e., all
local  flux values are  equal), the effective flux  and
apparent flux are equal and the ratio of effective flux
to apparent  flux, the  effective flux ratio (EFR), is unity.
If air is released unevenly, the  EFR would be greater
than one. High values  of EFR may produce poor OTE
values and promote fouling (see Chapter 3).
In measuring air flux,  the  same  or varying  size
collection cylinders  can be  used. With varying  size
cylinders (sometimes referred to as  the three bucket
catch), air flux  is determined for progressively larger
rings  on  the  surface. This  method involves  a certain
degree of averaging  and as a result may not be very
sensitive. For example,  results obtained using this
technique could indicate that air flux is uniform even
though  one  quadrant  of  the  surface could   be
discharging no air at all.

A more effective procedure is to use a  relatively small
diameter cylinder and test several points on the media
surface.  These  individual  points can  be plotted  to
produce  a flow profile or simply compared.  It is also
useful to calculate s/x, where s is the sample standard
deviation and x  is  the average of  the  individual
readings. In this case, the  smaller the s/x value,  the
more  uniform the diffuser.

Besides measuring flux, BRV values can also be used
as a measure of uniformity. Based on individual point
values, s/x  can be calculated. The smaller the value,
the more uniform the specific diffuser.

The previous paragraphs  have presented  several
procedures for  establishing  uniformity.  Unfortunately,
no  well-defined guidelines  have been developed
concerning the  variations between points that can  be
tolerated before the diffuser should be rejected  as
nonuniform.  The calculated values should  be used
only for comparison purposes.
2.6  Clean Water Performance

2.6.1 Introduction
The  following discussion summarizes  clean  water
performance data for fine pore diffusion devices. Only
some of the data were generated using the current
ASCE  recommended clean  water  Standard   (42).
Thus, the oxygen transfer results summarized in this
section reflect use of the current Standard as well as
earlier  methods for  clean  water testing. The  data
bases  for all  performance  tests  were  carefully
reviewed  before incorporating  the  results  in  this
manual.  Performance data  were used  only  if  data
collection and analysis closely paralleled the current
Standard.

The clean water performance data presented  in  most
of the  tables  and  graphs  in  this section  were
generated from 1975 to 1988. Besides the changes in
data  generation and  analysis procedures  occurring
during this period, it  should also be recognized  that
some diffuser design and  production  changes also
may have occurred. Data have been screened to best
depict  the trends and  ranges  of  performance of
representative types of fine pore diffusers. It must be
emphasized that the information presented  in  this
section   provides  general   trends  in clean  water
performance: these data  shoylcI  not  be  used in final
design calculations.
                                                   24

-------
The results of clean water oxygen transfer tests are
reported in  a standardized form as  standard  oxygen
transfer rate  (SOTR),  standard  oxygen transfer
efficiency (SOTE), or standard  aeration  efficiency
(SAE) as shown in Table 2-1. The standard conditions
for  reporting clean water tests are, also delineated in
this table. All data reported in this section are given as
standard transfer values unless otherwise noted.

The performance of diffusers under clean  water test
conditions is dependent On several factors. Among the
important factors are:
   diffuser type (material, shape, and size),
   diffuser placement and density,
   gas flow rate per diffuser,
   basin geometry,
   diffuser submergence, and
   uniformity of air flux.
The information that follows  is presented to illustrate
the effects of these factors.

2.6.2 Clean Water Data Base

2.6.2.1  Steady-State DO Saturation Concentration
(C*.)
An examination of Table 2-1  indicates that one of the
critical  parameters required in  the  calculation of
oxygen transfer rate is the steady-state  DO saturation
concentration,  C*<».  For  submerged  aeration
applications, C*ro  is  significantly  greater  than the
surface  saturation  value,  C*s,  tabulated in  most
standard tables. It is necessary,  therefore, to either
measure  C'oo (42) during clean  water tests or to
calculate it based  on comparable full-scale test data.
The value of C*» is  primarily dependent on  diffuser
submergence and diffuser  type.  Typical  results of
clean  water  test  measurements are  presented in
Figures 2-13  and  2-14.   Suppliers   of  aeration
equipment should  be able to  provide appropriate clean
water test data.

Alternatively,  C'o, can  be  estimated  for  design
purposes from the surface  saturation  concentration
and effective saturation  depth, de, by:

C"oo= C*s [(Pb - PvT + 0.007ywde)/(Ps - PvT)]   (2-1)
.where,
          steady-state  DO saturation  concentration
          attained  at  infinite  time   at  water
          temperature  T  and  field  atmospheric
          pressure Pb,  mg/L

          tabular value  of DO  surface  saturation
          concentration at  water temperature T and
          standard atmospheric pressure Ps, mg/L
  de    = effective saturation depth, ft

  Ps    = atmospheric   pressure  at  standard
          conditions,  14.7  psia  or 1  atm  at 100
          percent relative humidity

  Pb    = field atmospheric pressure, psia

  PVT  = saturated  vapor, pressure  of  water  at
          temperature T, psia

  YW    = specific weight of water at temperature  T,
          Ib/cu ft

  T    = water temperature, °C

The effective  saturation  depth, de, represents the
depth  of water  under which the total  pressure
(hydrostatic plus  atmospheric)  would  produce  a
saturation concentration equal  to C'o, for water  in
contact with air at 100 percent relative humidity. The
value  of de is  usually calculated,  using Equation 2-1,
based  on a spatially-averaged  value of C"m from  a
clean  water test. For  design purposes, de  can be
estimated from  clean  water test  results on  similar
systems. For diffusers submerged  to about 90 percent
or more of basin depth, de is normally  21-44 percent
of basin liquid depth for fine pore systems (49).

2.6.2.2 Oxygen Transfer Data
Typical  SOTEs for fine pore diffused air  systems are
presented in Table 2-2 for a diffuser submergence of
4.6 m (15 ft). The effects of diffuser type, placement,
and airflow rate  per diffuser are delineated from this
summary of different clean  water studies. With  the
increasing  number  of  materials  and shapes  being
marketed as  fine  pore diffusers,  it  is becoming
increasingly difficult  to  make generalizations  about
diffuser performance. For example, Table 2-2 presents
data for perforated membrane discs that are available
in a wide range  of  diameters. This accounts for the
corresponding wide ranges in airflow rate and  SOTE
for  these devices. It would appear, based on Table 2-
2, that  there is  little difference between the oxygen
transfer performance of disc/dome and  tube fine pore
devices. However, it is evident that the clean water
oxygen transfer performance of all fine pore diffusers
shown  in Table  2-2  is superior   to that of coarse
bubble diffusers.

For a  given diffuser type, spreading the diffusers rnoie
uniformly along the basin bottom area (moving from a
single spiral  roll  to  a  dual spiral roll to a grid
configuration)  lends  to  improve  clean   water
performance  (69). The effects  of basin  and diffuser
geometry on diffuser performance have been reported
by  many investigators. One of  the early,  notable
studies (70), conducted in a 1.2 m x 7.3 m (4 ft x 24
ft) test  basin using coarse bubble spargers and fine
                                                   25

-------
TaWo 2-1.   Standard Equations for Clean Water Oxygen Transfer Tests (42)
 Siandard Oxygen Transfer Rale (SOTR), Ib/hr:
  SOTR - 8,34 Kuaao C*-20 V
 Standard Oxygen Transfer Efficiency (SOTE), percent:
  SOTE - 100 (Mass Transferred/Mass Supplied) = SOTR/W02

 Standard Aoralion Efficioncy (SAE), IWhp-hr:
  SAG » SOTWPower Input  (spscilied as delivered, brake, wire, or total wire)
  Kta » apparent volumetric mass transfer coefficient in clean waler at temperature T, t/hr
  KLa20 - Kta@20"C, 1/hr
  V « clean water volume, mil gal
  W0;. » mass rate ol oxygen, Ib/hr
  C^o m sloady-stato DO saturation concentration attained at infinite time at 20°C and 1 aim, mg/L
  T « clean water temperature,  °C

 Standard Conditions:
  DO  «• 0.0 mg/L                     Q - 1.0
  Water temperatura » 20"C           p = i.o
  Prossuro «  1,00 aim                 F = 1.0

  o » (process water KLa of a new diffuser)/(clean water KLa of a new diffusar)
  p - (process water C*J/(clean water C'^,)
  F * (process walor KLa of a diffuser after a given lime in service)/(KLa of a new diffuser in the same process waler)
Figure 2-13.   Effect  of diffuser submergence on  CUzo for    Figure 2-14.  C'^o vs. diffuser  submergence  for  perforated
              three diffuser types.                                            membrance disc and tube diffusers.
  Tank; 20 It x 20 ft
  Powcf:~ 1 hp dohvorod/1,000 cu ft for rigid porous plastic tubes
  Power. ~S lip delivered/1,000 cu ft for ceramic domes
    13
    11
    10
                                           da = 0.4 (depth)
          Rigid Porous Plastic Tubes • Dual Spiral Roll  S
                                          Coarse Bubble
                            Q, ro9/L

                          12.5

                                 Most Points Correspond Approximately to:

                                     d0 = 0.4 (depth)
                                                                    11.5
                                                                    10.5
                                 de = 0.2 (depth)
                                                                     9.5
         Coramic Domes - Grid
         	I	I
                                                                                        10             20

                                                                                       Diffuser Submergence, ft
                                                                                                                      30
J_
                   10          15          20

                      Diffuser Submergence, ft
                                                       25
                                                             26

-------
Diffuser Type and Placement
Ceramic Plates - Grid
Ceramic Discs - Grid
Ceramic Domes - Grid
Porous Plastic Discs - Grid
Perforated Membrane Discs - Grid
Rigid Porous Plasiic Tubes
Grid
Dual spiral roll
Single spiral roll
Islonriyid Porous Plastic Tubes , :
Grid
Single spiral roll
Perforated Membrane Tubes
Grid
Mid-width
Mid-width
Single spiral roll
Coarse Bubble Diffusers
Dual spiral roll :
Mid-width
Single spiral roll
Airflow Rale, scfm/diffusera
2.0-5.0 scfm/sq ft
0.4-3.4
0.5-2.5
0,6-3.5 "
0.5-20.5
2.4-4.0
3-11
2-12
1-7
2-7
1-4
2-6
2-12
2-6
• - 3.3-9.9 - -
4.2-45
10-35
SOTE at 1 5-f! Submergence, percent
26-33
25-40
• . 27-39
24-35
16-38
28-32 '
17-28
13-25 ' ' "
26-36
19-37
22-29
16-19
21-31
15:19
12-13
10-13
9-12
Ret.
50 __
51-53
53-58
59-61
61-64
65
53,58,66
59,66
67
. 67
25
25
68
25
58 •
58
58
 a Except for plates.
pore  Saran tubes, demonstrated similar effects of
placement.

The  OTE  produced by  fine pore diff users in clean
water generally decreases  as the airflow  rate  per
diffuser increases. Over the  normal range of operation
for a given diffuser submergence, type,  density,  and
geometry, the relationship between SOTE and diffuser
airflow rate,  qs, can  be  described by the following
empirical relationship:

    (SOTEa/SOTEb) = [(qa)/(qb)]m (2-2)

where,       .  "

  SOTEa  =  SOTE at a diffuser airflow rate of qa

           =  SOTE at a diffuser airflow rate of qb
   m      = a  constant  for  a given  diffuser and
            system configuration (usually a fractional
            negative number for fine pore diffusers)
It is convenient  to select qo at a rate of  0.47  Us (1
scfrri) so that:
              SOTEa =
(2-3)
where,
   SOTE-, = SOTE at diffuser airflow rate of 0.47 Us
            (1 scfm)
Data from a number of clean water studies have been
used to estimate the value of m in Equation 2-3. Table
2-3 presents the results of these clean water tests.

Table 2-4 and Figure 2-15 demonstrate the  effect of
unit airflow rate  on SOTE for tube diffusers. SOTEs
decrease with increased airflow per diffuser, whereas
coarse bubble SOTEs are relatively unaffected by unit
airflow rate with some indication of a slight increase in
SOTE at higher airflows.  Very similar patterns were.
reported in 1964 for coarse bubble spargers and fine
pore tubes  (70).

Figure 2-15 demonstrates  that there are some
overlapping regions of similar  SOTE performance for
the various  tube diffusers represented, especially in
the  typical  design  region   of  1.4-2.6  L/s  (3-6
scfm)/diffuser. These data are for a number of diffuser
placements and diffuser densities.

In  this manual,  diffuser  density  is  defined  as  the
number of installed diffusers per 100 sq ft of tank floor
area.  Another expression used  as  an  indicator  of
diffuser density is the total  projected  media surface
area of the installed diffusers (AD) divided by the area
of  the tank floor (AT).                        ;

Table 2-5  and Figures 2-16 and 2-17 illustrate  the
effect  of airflow  rate  on SOTE  for  disc/dome grid
systems.  SOTE decreases with increasing airflow rate
per  disc/dome diffuser.  At  rates  >0.9  L/s  (2
scfm)/diffuser, SOTE  is less  sensitive to airflow rate
changes.
                                                   27

-------
Table 2-3.   SOTE vs. Airflow for Selected Fine Pore Oiffusers in Clean Water
Dilliiscr Typo
Ceramic Dome
Ceramic Disc
Ceramic Disc
Rigid Porous Plas>tic Disc
Rigid Porous PldStic Tube
Nonrigtd Porous Plastic Tube
Perforated Membrane Disc
9 in Perforated Membrane Disc
EPDM Porloraled Membrane Tube
Diffuser
Layout Submergence, ft
Grid
Grid
Grid
Grid
Double Spiral Roll
Spiral Roll
Grid
Grid
Grid
14
12.3
12.3
13
13
15
14
10
10
Diffuser Density,
No./100sq ft
32
26
15
34
10.5
8.6
8.8
20«
20b
SOTE,,
percent
29.6
31.7
26.0
27.9
26.7
27.1
29.2
18.9
21.0
m
-0.150
-0.133
-0.126
-0.097
-0.240
-0.276
-0.195
-0.110
-0.150
Ref.
58
11
11
71
58
11
62
72
72
  Data lit lo: SOTE « SOTE, (q)"i (see Equation 2-3).
  a One 9-id diameter disc in a 30-in diameter column.
  b One 24-in long tube in a 30-in diameter column.
Tablo 2-4.    Clean Water Oxygen Transfer Efficiencies of Fine Pore Tube Diffuser Systems

                                                  	SOTE at Following Water Depth, percent
 Dilfusof Typo and Placement         Airflow, scfrn/diffuser
10 ft
              15ft
20 ft
                                          Ref.
Rigid Porous Plastic Tubes
Grid" . '
Dual spiral roll

StiKjto spiral roll

Purloratod Membrane Tubes
Floor cover (grid)
Quarter points
Mid-wtdlh
Mid-width
Single spiral roll
Nonrigid Porous Plastic Tubes
Grid"
Spiral roll

2.4-4.0
3-7
9-11
2-7
8-12

1-4
2-6
2-6
2-12
2-6

1-7
1-7

-
10-16
10-14
12-15
10-15

14-18
13-15
9-11
15-21
7-11

-
-

28-32
16-24
15-17
15-20
10-17

21-27
18-22
15-18
21-31
14-18

20-34
18-35

-
22-32
21-26
22-25
22

29-35
24-29
23-27
27-36
21-28

-
-

65
53,58,66
53,58,66
53,58,59,66
53,58,59,66

25
25
25
68
25

67
67
 8 Diffuser density = 13.0 tubes/100 sq ft basin Moor. Basin is 14.4 ft x 108.2 ft.
The effect of diffuser density on SOTE for disc/dome
grid configurations is shown in Table 2-5 and  Figure
2-18. Generally,  an  increase in disc/dome diffuser
density  (or AD/AT) results in an  increase  in  SOTE
(52-54,56). However, there is some indication  that a
maximum  value of AD/AT exists  above which  little
improvement  in  SOTE  occurs. The  upper limit  is
dependent on diffuser size, airflow, and spacing. For
example,  a 23-cm (9-in)  ceramic  disc diffuser at a
submergence of  4.3  m  (14.2  ft) exhibited  SOTE
values that did  not increase  at  AD/AT values  above
0.14 for airflows of 0.5 Us (1 scfm)/diffuser (51).

As  can be seen  from  Figures 2-16  and 2-17, the
oxygen  transfer  performance  of ceramic  dome/disc
diffusers and rigid  porous plastic  disc diffusers  is
similar throughout the range  of airflows presented.
However,  the ceramic diffusers demonstrate  a  higher
upper limit of oxygen transfer performance throughout
the airflow range  and are generally more efficient  at
airflows less than about 0.7 Us (1.5 scfm)/diffuser.
An  OTE  increase of 5-15 percent, varying with depth,
for  a  24-cm (9.4-in) diameter ceramic disc compared
with an 18-cm  (7-in) diameter ceramic disc has been
reported  ,(53). This increase was attributed to  a  70-
percent higher effective surface area for the 24-cm
(9.4-in) disc. A similar relationship between dome/disc
diameter and  oxygen  transfer per diffuser  has also
been  reported (29,73).

Table 2-6 and  Figure  2-19  demonstrate  the
performance  of  perforated  membrane discs.  At
airflows of 0.24-3.8 Us (0.5-8  scfm)/diffuser, SOTE
significantly decreases  with  increasing  airflow in  a
manner similar to other fine  pore diffusers. At higher
airflow rates,  SOTE  is  less  sensitive  to  airflow.
Perforated  membrane  discs  are  usually  larger  than
ceramic  and porous plastic dome/disc  units.  As  a
result, they are often operated at higher airflows  per
diffuser  than  ceramic/porous  plastic diffusers.  In
addition,  density  of perforated  membrane  discs is
normally  lower than for ceramic/plastic units.
                                                    28

-------
Figure 2-15.  Effect of unit airflow rate on SOTE for fine pore
              tube d iff users.   .
                                               Figure 2-16.  Effect of unit airflow rate on SOTE for ceramic
                                                            dome/disc diffusers.
 SOTE, percent/ft

  3   r-
Higher SOTE values for one diffuser type at
any given airflow rate indicate increased
diffuser density or dual placement.

      •  EPDM Perforated Membrane Tubes
      o  Nonrigid Porous Plastic Tubes
      •  PVC Perforated Membrane Tubes
      0  Rigid Porous Plastic Tubes
                    SOTE, percent/ft

                     3  r-
                                              Coarse Bubble
             For diffuser submergences of 10-20 ft.
                I
                          I
                 I
              Higher SOTE values for one diffuser type at
              any given airflow rate indicate increased
              diffuser density and, in some cases, larger
              surface area diffusers.
                                                                                For diffuser submergences of 10-20 ft
                          8        12        16

                    Airflow Rate, scfm/diffuser
                                                       20
                                                                 1            2            3
                                                                  Airflow Rate, scfm/diffuser
Table 2-5.*    Clean Water Oxygen Transfer Efficiencies of Fine Pore Disc/Dome Grid Systems

                           Diffuser Density,                            SOTE at Following Water Depth, percent
 Diffuser Type
          No./100sq ft
Airflow, scfm/diffuser
10 ft
              15 ft
                                                                                 20 ft
                                            Ref.
Ceramic Disc - 9.4 in


Ceramic Disc - 8.7 in




Ceramic Dome - 7 in




Porous Plastic Disc - 7 in



15.6
24.4
31.3
14.7-15.4
16.9-18.9
21.3-25.0
29.4-31.3
40.0-52.6
17.9
22.7-23.8
30.3-31.3
40.0-45.4
66.7
14.5-14.7
21.7 ,
25.6
34.5
0.9-3.0
0.8-2.9
0.7-2.6
1.5-3.2
0.6-2.5
0.6-3.4
0.4-2.8
0.7-3.1
0.5-2.0
0.5-2.5
0.5-2,5
0.5-2.5
0.5-2.5
0.6-3.5
0.6-3.5
0.5-2.3
0.4-1.5
20-22
21-24
22-25
.
-
-
-
• •
.
16-23
20-24
17-23
18-26
15-18
16-21
..
19-22
27-33
30-34
31-34
25-29
26-30
27-34
25-36
27-38
25-31
25-32
27-37
27-35
27-34
22-27
24-28
25-31
26-32
34-37
35-41
38-41
32-38
33-40
31-40
34-39
31-38
28-40 ,
30-41
31-44
33-47
' -'
-
-
-
-
69
69
70
51,52
51,52
51,52
51,52
51,52
54,57
54-56
55,56,58,69
55-57
55,56
59-61
59-61
59-61
59-61
                                                              29

-------
figure 2-17. Effect of unit airflow rate on SOTE for rigid
           porous plastic disc diffusers.

 SOTE, pofconl/lt

  3  r~
                 Higher SOTE values ai a given airflow rale
                 indicate increased diffuser density in some
                 cases.
            For diffusor submergences of 10-20 (l.
                1          2          3

                 Airllow Rale, scfm/diffuser
Figure 2-18,  Effect of diffuser density on SOTE for ceramic
            disc/dome grid configurations (56),
     45-•
     40--
    .35
     30"
     25-•
     20
           Water Doplli « 15 ft
                                 45.4 diffusers/100 sq fl
         I8.5diffusers»00sqft

H	1	1	h-
 0.5     1.0     1.5      2.0

     Airflow Rale, scfm/diffusor
                                             2,5
The  effect of density and placement of perforated
membrane discs on SOTE appears to be variable  as
indicated in Table  2-6. Results of field testing on 51-
cm  (20-in)  discs  indicate  OTE  increases  with
increasing  diffuser density  or increasing  AD/AT,  at
least  to an AD/AT of 0.26 (62). The data  suggest,
however, that, at  the same airflow rate per diffuser,
OTE  approaches a limiting value as diffuser density
increases.  A 40-percent increase in the  number  of
diffusers (producing a increase in AD/AT from 0.18 to
0.26) resulted in an increase  in OTE of  only about 5
percent.  Prior pilot-scale testing on  51-cm (20-in)
perforated  membrane discs up  to an  AD/AT of 0.59
also  indicated OTE approaches a maximum value with
increasing  density or increasing  AD/AT  at the  same
airflow rate per diffuser (74).

As discussed above for  ceramic disc/dome diffusers,
SOTE of perforated membrane disc diffusers is also
affected by airflow per diffuser, diffuser  spacing, and
diffuser density.  Thus, it is evident  diffusers  of
different size will exhibit different maximum  densities
or AD/AT  values  above which little  improvement  in
SOTE will be achieved.

As  described  in Section  2.5.12,  the  specific
permeability (or BRVo) of a ceramic diffuser  will have
some effect on   OTE; as  pore  size  increases
(permeability  usually increases), bubble size increases
and  SOTE may  decrease. Figure 2-20  depicts the
results of  a  clean water  test series  using ceramic
discs with  specific permeabilities  of 14-50  (BRV0  =
22.4-7.1 cm [8.8-2.8 in] w.g.)  (48). The results of this
study indicate that, for specific permeabilities of  14-38
(BRV0 =   22.4-10.4 cm [8.8-4.1  in]  w.g.), SOTE is
relatively  unaffected.  However,  at   a  specific
permeability of 50 (BRV0  =  7.1 cm  [2.8 in]  w.g.),
there appears to  be a  significant decline in SOTE,
especially at the lower airflow rates.

The  effects of water depth on  SOTE and  SAE  for
several types of diffusers are  illustrated  in Figures  2-
21 and 2-22,  respectively. Although these data are  for
one  specific  test  basin and  airflow  rate (58),  they
represent  the typical effects of depth on  performance.
In general, SOTE values  increase with  increasing
depth  since  mean oxygen partial pressure is higher
(thereby  resulting  in a  greater driving  force) and
opportunity is present  for  longer  bubble residence
time  in the aeration basin. SAE,  however, remains
relatively constant (or may increase  slightly [29,64])
for fine pore diffusers as depth increases since power
requirements to  drive  the  required  air  through
diffusers may increase at the  greater depths.  In
contrast, coarse bubble diffusers exhibit a  gradually
increasing  SAE  with increasing depth,  though not
reaching the  overall efficiencies demonstrated by fine
pore systems.
                                                    30

-------
Figure 2-19.  Effect of  unit  airflow rate  on  SOTE for     Figure 2-20.  Clean water test data - Monroe, Wl.
             perforated membrane disc diffusers.
  SOTE, percent/ft

  3
                  Higher SOTE values at a given airflow rale
                  indicate increased diffuser density in some
                  cases.
                For diffuser submergences of 10-20 ft.


                  I    I    !	I	I	I	!	L_
SOTE, percent
                                                                     34
                                                                     32
                                                                     30
                                                                     28
                                                                     26
                                                                     24
   22
                                                                             Tank Dimensions: 34.3 ft W x 6 ft L x 14.3 ft SWD
                                                                         -   36 9-in Disc Diffusers (AD/AT) = 0.07
                                 I
                                                                                                            Spec.
                                                                                                            Perm. BRV,,
                      8       12      16      20

                    Airflow Rate, scfm/diffusor
                                                      24
               0.5       1.0       1.5      2.0

                  Airflow Rate,, scfrn/diffuser
                                                   2.'5
Table 2-6.    Clean Water Oxygen Transfer Efficiencies of Perforated Membrane Diffuser Systems
Diffuser Type
20-irt Disc
12-13-in Discs
9-in Disc
7-8.5-in Discs
Diffuser Density,
No./1 00 sq ft
3.0
8.8
6.3
8.7-9.3
11.1
17.2
4.0
8.0
14.1-18.5
18.5-22.2
Airflow, scfrn/diffuser
3.3-20.5
2.9-19.4
1.9-12.0
2.0-12.9
1.5-10.3
2.0-5.9
0.5-7.1
0.5-6.2
0.5-6.8
0.9-4.7
*J\J I C CH i U!
10 ft
11-16
12-19
11-15
16-23
9-21
18-24
-

iiuwii ly v» cue/! ut
15fl
19-25
21-29
19-26
20-31
24-36
25-30
15-36
21-31
23-29
23-26
:j|ju i, jjcJrouiii
20 ft
24-29
27-38
28-37
34-48
27-43
' '31-36
-
-
Ref.
62
62
' 64
61,63
64
63'
61
61
6.1
61
                                                            31

-------
Figure 2-21.  Effect  of water depth on SOTE for  three     Figure 2-22.  Effect of water depth on SAE for three diffuser
             diffuser types.

 Tank: 20 ft x 20 li
 Powcr:~1 tip delivered/1,000 cu It lor rigid porous plastic lubes
 Powor:~5 hp dolivered/i ,000 cu ft for ceramic domes

 SOTE, percent

   SO
   40"
   30
   20
   10
                                  Ceramic Domes - Grid..
                                 Rigid Porous Plastic Tubes
                                     - Dual Spiral Roll
                                    Coarse Bubble
                               I
                  10
                              15

                          Water Depth, It
                                          20
                                                      25
           types.

Tank: 20 ft x 20 fl
Power: ~1  hp delivered/1,000 cu fl for rigid porous plastic lubes
Power: ~ 5 hp delivered/1,000 cu ft for ceramic domes

SAE, Ib Ou/hp-hr (wire-lo-waler)

  10
                                                                                             Ceramic Domes - Grid
                          Rigid Porous Plastic Tubes
                              - Dual Spiral Roll
                                                                                              I
                                  Coarse Bubble
                                                                                                          I
                 10         15          20

                        Water Depth, ft
                                                                                                                     25
                                                            32

-------
2.7 References

When an NTIS  number is cited in a reference, that
reference is available from:

 National Technical Information Service
 5285 Port Royal Road
 Springfield, VA22161
 (703) 487-4650
1.  Eckenfelder, W.W.,  Jr. Water Quality Engineering
    for Practicing Engineers* Professional Engineering
    Career  Development  Series, Barnes  &  Noble,
    New York, NY, 1970.

2.  Schmidt-Holthausen, H.J.  and B.C. Bievers. 50
    Years of Experience in Europe with Fine  Bubble
    Aeration.  Presented  at  the  53rd Annual
    Conference  of the  Water  Pollution Control
    Federation, Las Vegas, NV, October 1980.

3.  Filtros.  Product information  bulletin,  Ferro
    Corporation,  Refractories Division,  East
    Rochester, NY, May 1984.

4.  Hosokawa, K.  Characterization  of Various
    Diffusers and  Its Application, In: Proceedings of
    the  11th  United States/Japan Conference on
    Sewage Treatment  Technology,  EPA-600/9-
    88/010,   NTIS  No.  PB88-214986,   U.S.
    Environmental  Protection Agency, Cincinnati, OH,
    April 1988.

5.  King, H.R.  Sewage  and Industrial Wastes. 27:10,
    August 1955.

6.  Bartholomew, G.L.  Type of Aeration Devices. In:
    Aeration  of  Activated  Sludge  in   Sewage
    Treatment, Gibbon,  D.L., Editor, Pergamon Press,
    1974.

7.  Carborundum  Aloxite Porous  Products  for
    Filtration,  Aeration,  and  Diffusion.  Product
    information bulletin,  Carborundum  Company,
    Bonded Abrasives  Division, Niagara Falls, NY,
    May 1970.

8.  Porex  Porous  Plastic  Materials.   Product
    information bulletin,  Oy  Airam  AB,   Helsinki,
    Finland, 1988.

9.  Nopol  Aeration Systems. Product information
    bulletin, Nokia  Metal  Products, Vantaa, Finland
    (undated).

10. Donohue &  Assoc., Inc.  Fine Pore Diffuser
    System Evaluation for the Green Bay Metropolitan
    Sewerage  District.  Study  conducted  under
    Cooperative   Agreement  CR812167,  Risk
    Reduction  Engineering  Laboratory,  U.S.
   Environmental Protection Agency, Cincinnati, OH
   (to be published).

11. Stenstrom,  M.K. and G.  Masutani. Fine Bubble
   Diffuser Fouling: The Los Angeles Studies. Study
   conducted  under  Cooperative  Agreement
   CR812167,  Risk  Reduction  Engineering
   Laboratory, U.S. Environmental  Protection
   Agency, Cincinnati, OH (to be published).

12. AERMAX TPD  High Efficiency  Product Bulletin,
   Aeration Technologies,  Inc.,  North Andover, MA
   (undated).

13. 1988 Annual Book of ASTM Standards. Sections
   8  and 9,  Volumes  8.04  and  14.02,  American
   Society of Testing and Materials, Philadelphia, PA,
   1988.

14. Handbook of Plastics  and  Elastomers. Harper,
   C.A.,  Editor-in-chief,  Westinghouse  Electric
   Corporation, McGraw-Hill, New York, NY, 1975.

15. Sanitaire Flexible  Membrane  Tube  Diffusers.
   Product  information bulletin,  Sanitaire-Water
   Pollution Control Corp., Milwaukee, Wl, 1987.

16. Sanitaire Flexible Membrane  Disc  Diffusers.
   Product  information bulletin,  Sanitaire-Water
   Pollution Control Corp., Milwaukee, Wl, 1987.

17. Eimco Elastox-T Non Clog  Fine Bubble Rubber
   Diffuser. Product Bulletin 1335.2T, Eimco Process
   Equipment Co.,  Salt Lake City, LIT,  1986.

18. Eimco Elastox-D Non Clog  Fine Bubble Rubber
   Diffuser. Product Bulletin 1335.1, Eimco Process
   Equipment Co.,  Salt Lake City, UT,  1985.

19. Roeflex Diaphram Diffuser. Product Bulletin ROD
   100/5M, Roediger Pittsburgh, Inc., Pittsburgh, PA,
   1986.

20. Rne Bubble Membrane Diffusers for Non-Clogging
   Energy Efficient Aeration.  Product Bulletin  No.
   315-14C1, Envirex Inc., Waukesha,  Wl, 1986.

21. Ernest, L.A. Case History Report  on Milwaukee
   Ceramic  Plate  Aeration  Facilities.  Study
   conducted  under  Cooperative  Agreement
   CR812167,  Risk  Reduction  Engineering
   Laboratory, U.S. Environmental  Protection
   Agency, Cincinnati, OH (to be published).

22. Lue-Hing,  C., D.R. Zenz  and  B.  Sawyer.  Case
   History: Aeration System  Design, Operation,
   Control, and  Maintenance  at  the Metropolitan
   Sanitary District of Greater Chicago. Presented at
   the  Aeration Systems  Operations, Control  and
   Testing Conference, Georgia  Water  Pollution
   Control Association, Atlanta, GA, March  1984.
                                                33

-------
23. Reef  Aeration  Mixing  Systems.  Product
    information bulletin. Environmental Dynamics, inc.,
    Columbia, MD (undated).

24. Sanitaire  Fine Bubble  Tube  Diffuser.  Product
    Bulletin  TD  4/85, Sanitaire -  Water Pollution
    Control Corp., Milwaukee, Wl, 1985.

25. WYSS Fiex-A-Tube Diffuser. Product Bulletin WD-
    800, Parkson Corp., Ft. Lauderdale, FL (undated).

26. Engineering  Data-Endurex  Airfine  Diffusers.
    Product Bulletin 5M835, Endurex Corp., Loveland,
    OH (undated).

27. REX Fine Bubble Tube Diffusers. Product Bulletin
    31S-14A3, Envirex Inc., Waukesha, Wl, 1982.

28. Pearlcomb Air Diffusers. Product Bulletin  7824,
    FMC Corporation, Chicago, IL, 1973.

29. Houck,  D.H.  and  A.G.  Boon.  Survey and
    Evaluation of Fine Bubble Dome  Diffuser Aeration
    Equipment.  EPA-600/2-81-222,  NTIS No.  PB82-
    105578, U.S.  Environmental  Protection Agency,
    Cincinnati, OH, September 1981.

30. Houck, D.H. Survey and Evaluation of Fine Bubble
    Dome and  Disc  Diffuser Aeration  Systems in
    North  America.  EPA-600/2-88/001,  NTIS  No.
    PB88-243886, U.S.  Environmental Protection
    Agency, Cincinnati, OH, August 1988.

31. Fine Bubble  Air  Diffusers.  Product  Bulletin 106,
    EPCO International, Victoria, Australia (undated).

32. Dome  Diffuser  Aeration  System.  Product
    information bulletin,  Norton  Industrial Ceramics
    Division, Worcester, MA (undated).

33. Diffused Aeration Products: Fine  Air Ceramic
    Difluser. Product Bulletin FA1001, Parkson  Corp.,
    Ft. Lauderdale, FL (undated).

34. Wren,  J.D.  Diffused  Aeration  Types  and
    Applications.   In:   Proceedings    of
    SaminarWorkshop  on Aeration  System  Design,
    Testing, Operation, and  Control, EPA-60Q/9-85-
    005, NTIS No. PB85-173896, U.S. Environmental
    Protection Agency, Cincinnati, OH, January 1985.

35. Fixed  Fine  Bubble Aeration  System. Product
    Bulletin 315-14A4.  Envirex Inc., Waukesha,  Wl,
    1982.

36. Sanitalre Ceramic Grid Aeration System.  Product
    Bulletin CFB2-R83, Sanitaire -  Water Pollution
    Control Corp., Milwaukee, Wl, 1983.

37. Ewing,  L. and D.T.  Redmon. U.S.  Patent  No.
    4,261,933, April 14, 1981.
38. Boyle, W.C. and D.T. Redmon. Biological Fouling
    of Fine Bubble Diffusers: State-of-Art. J. Env. Eng.
    Div., ASCE 109(EE5):991-1005, October 1983.

39. Renton Plant Gets Into the Swing of Conservation.
    Monitor, January 1986.

40. Boon, A.G. and B. Chambers. Design Protocol for
    Aeration  Systems  -  UK  Perspective,  In:
    Proceedings of  Seminar/Workshop  on Aeration
    System Design,  Testing, Operation,  and Control,
    EPA-600/9-85-005,  NTIS No. PB85-173896, U.S.
    Environmental  Protection Agency, Cincinnati, OH,
    January 1985.

41. Redmon,  D.T.  Operation and  Maintenance!
    Troubleshooting.  In:  Proceedings  of
    Seminar/Workshop  on Aeration  System  Design,
    Testing, Operation,  and Control,  EPA-600/9-85-
    005,  NTIS No. PB85-173896, U.S. Environmental
    Protection Agency, Cincinnati, OH, January 1985.

42. American  Society of  Civil  Engineers.  ASCE
    Standard: Measurement of Oxygen Transfer in
    Clean Water. ISBN 0-87262-430-7,  New York, NY,
    July 1984.

43. Wren, J.D. Transcript of Biofouling Seminar. New
    York  Water Pollution  Control  Federation, New
    York, NY, January 1985.

44. Baillod, C.R. and K. Hopkins. Fouling of Fine Pore
    Diffused Aerators:  An  Interplant  Comparison.
    Study conducted under Cooperative Agreement
    CR812167,  Risk   Reduction  Engineering
    Laboratory,  U.S.   Environmental Protection
    Agency, Cincinnati, OH (to be published).

45. Winkler,  W.W.  Fine  Bubble  Ceramic Diffuser
    Maintenance. Presented at the Annual Meeting of
    the  New  England  Water  Pollution Control
    Association, Boston, MA, January 25, 1984.

46. Danly, W.B. Biological Fouling  of Fine  Bubble
    Diffusers.  MS  Thesis,  Dept.   of  Civil  and
    Environmental  Engineering,  University  of
    Wisconsin, Madison, Wl, 1984.

47. Rieth, M.G.  and R.C. Polta. A Test Protocol for
    Aeration  Retrofit  to  Fine Bubble  Diffusers.
    Presented at the 60th Annual  Conference of the
    Water Pollution Control Federation,  Philadelphia,
    PA, October 1987.
48. Ewing  Engineering  Co. The  Effect of Permeability
    On  Oxygen  Transfer  Capabilities,  Fouling
    Tendencies, and Cleaning Amenability at Monroe,
    Wl.  Study conducted  under  Cooperative
    Agreement  CR812167,  Risk   Reduction
    Engineering Laboratory,  U.S.  Environmental
                                                34

-------
    Protection  Agency,  Cincinnati,  OH  (to  be
    published).

49. Baillod, C.R., W.L Paulson,  J.J. McKeown and
    H.J.  Campbell,  Jr. Accuracy and  Precision of
    Plant  Scale and Shop  Clean Water Oxygen
    Transfer Tests. JWPCF 58(4):290-299, 1986.

50. Oxygen Transfer Performance-REEF SAM IV
    Diffusers-Richardson,  TX.   Environmental
    Dynamics, Inc., Columbia,  MO, 1988.

51. Oxygen Transfer-Ceramic Disc  Diffuser  System
    Reports. Sanitaire-Water Pollution Control Corp.,
    Milwaukee, Wl, 1976-1986.

52. Maillacheruvu, K.Y. Analysis  of Oxygen Transfer
    Performance on  Dome-Disc  Fine  Pore  Diffuser
    Systems.  MS  Thesis,  Dept. of  Civil  and
    Environmental Engineering,  University of  Iowa,
    Iowa City, IA, July 1987.

53. Huibregtse,  G.L.,   T.C. Rooney  and  D.C.
    Rasmussen.  Factors  Affecting  Fine  Bubble
    Diffused Aeration. JWPCF  55(8): 1057-1064, 1983.

54. Gilbert, R.G. and R.C. Sullivan. The Significance
    of  Oxygen  Transfer   Variables  in  Sizing  Dome
    Diffuser Aeration Equipment. In:  Scale-up of Water
    and  Wastewater   Treatment   Processes,
    Schmidtke,  N.W. and D.W.  Smith,  Editors, Ann
    Arbor Press, Ann Arbor, Ml, 1983.

55. Oxygen Transfer-Ceramic  Dome Diffuser  System
    Reports. Aeration  Engineering  Resources Corp.,
    Northboro, MA, 1976-1986.

56. Yaeger, K.A. The Effects of Tank  Geometry on
    Performance of  Fine  Pore Diffusers. MS  Thesis,
    Dept.  of  Civil  and  Environmental  Engineering,
  • University of Iowa, Iowa City, IA, May 1986.

57. Statiflo/Fine Air  System Product Bulletin. Statiflo
    International. Woodbridge,  Ontario, Canada,
    September 1987.

58. Yunt, F.W. and T.O.  Hancuff. Aeration Equipment
    Evaluation: Phase 1  - Clean  Water  Test Results.
    EPA-600/2-88/022, NTIS No.  PB88-180351,  U.S.
    Environmental Protection Agency, Cincinnati, OH,
    March 1988.

59. Fine  Bubble Diffuser Reports.  Pacific Roller Die
    Co., Inc.,  Hayward, CA, October 1986.

60. Hardy, P.J. Testing  of  a   Nokia  Nopol  Disc
    Aeration  System  at Beckton  S.T.W. Thames
    Water Authority  Publication,  London, England,
    May 1986.
61. Oxygen  Transfer  Efficiency of  Wilfley-Weber
   Diffusers. Wilfley-Weber  Inc., Englewood, CO,
   1987.

62. Huibregtse, G.L.  Evaluation of  the  IFU  Fine
   Bubble  Membrane Disc Diffuser.  Internal project
   reports, Envirex Inc., Waukesha, Wl, January and
   April 1987.

63. Evaluation of the Oxygen Transfer Capabilities of
   the  Roediger  Roeflex  Diaphragm  Diffuser.
   Roediger  Pittsburgh, Inc.,  Pittsburgh, PA,  March
   1989.

64. Evaluation of the Oxygen Transfer Capabilities of
   the Eimco Elastox-D Fine Bubble Rubber  Diffuser.
   Eimco Process Equipment Co.,  Salt Lake City,
   UT, August 1986.

65. Popel, J.H. Improvements of Air Diffusion Systems
   Applied in the  Netherlands.  In:  Proceedings of
   Seminar/Workshop on  Aeration  System  Design,
   Testing,  Operation, and Control.  EPA-600/9-85-
   005, NTIS No. PB85-173896. U.S. Environmental
   Protection Agency, Cincinnati, OH,  January 1985.

66. Paulson, W.L. and  J.K. Johnson. Oxygen  Transfer
   Study of FMC  Pearlcomb  Diffusers.  Report
   prepared for the FMC Corporation, Lansdale, PA,
   August 1982.

67. Oxygen Transfer  Performance of  AERMAX TPD
   High  Efficiency  Diffusers. Aeration Technologies,
   Inc., North Andover, MA, 1986.

68. Evaluation of the Oxygen Transfer Capabilities of
   the Eimco Elastox-T Fine Bubble Rubber  Diffuser.
   Eimco Process Equipment Co.,  Salt Lake City,
   UT, January 1987.

69. Rooney, T.C.  and  G.L. Huibregtse. Increasing
   Oxygen  Transfer  Efficiency with  Coarse Bubble
   Diffusers. JWPCF 52(9):2315-2326, 1980.

70. Bewtra, J.K. and W.R. Nicholas. Oxygenation from
   Diffused Air  in   Aeration  Tanks.  JWPCF
   36(10):1195-1224< 1964.

71. Yunt, F.W. and T.O.  Hancuff. Analysis of Shop
   Performance Tests of the Air  Diffusion Equipment
   for Valencia Water Reclamation  Plant  Stage
   Three.  Internal  report,  Los Angeles  County
   Sanitation Districts, Whittier, CA,  1986.

72. Personal  communication   from  J.D.  Wren,
   Sanitaire -  Water  Pollution  Control Corp.,
   Milwaukee,  Wl,  to R.C.  Brenner,  U.S.
   Environmental protection Agency, Cincinnati, OH,
   April 20, 1989.
                                                 35

-------
73. Yunl, F.W. and T.O. Hancuff. Relative Number of
    Diffusers  lor the Norton and Sanitaire  Aeration
    Systems to Achieve Equivalent Oxygen  Transfer
    Performance. Internal report, Los Angeles County
    Sanitation  Districts, Whittier,  CA, December 14,
    1979,

74. H.J. Popel. Oxygen Feed Capacity and Oxygen
    Yield of  the  IFU  Membrane  Aerator. Report
    submitted to K.H. Schussler, Bad Homburg, West
    Germany, November 1986.
                                                36

-------
                                             Chapters
                                   Process Water Performance
3.1 Introduction
As described in Chapter 2, a substantial data base
exists on the  performance of fine  pore  diffusers  in
clean water. In  designing aeration systems to operate
under  process  conditions, clean  water data  are
corrected to account for the influences of wastewater
characteristics,  temperature,  and  pressure.
Throughout this manual, the term "process water"  is
used  to  refer  to mixed liquor under  aeration.  The
corrections are  made using the following equations:
 OTR,  = 8.34 KLa20 aF 8 ~ C oo20 =  C g ~ C s20

  C*s    = tabular value  of DO  surface  saturation
           concentration  at  water  temperature  T,
           standard  atmospheric   pressure  Ps, and
           100 percent relative humidity, mg/L

  F     = (process  water K^a  of a diffuser  after a
           given  time  in  service) •*• (Ki_a of  a new
           diffuser in the  same process water)

Since standard oxygen transfer rate, SOTR, is:

           SOTR =  8.34 KLa20 C'a20 V       (3-2)

Equations 3-1 and 3-2 may be combined to calculate
the  process water oxygen  transfer rate, OTRf:
                                                          OTRt = aF(SOTR)6T-20 (x.SnClao - C) * CT^o (3-3)

                                                     This equation can be rearranged as follows:

                                                         aF(SOTR) = (OTR, CA«.2o 02°-T) * (tSQC'^o - C) (3-4)
                                                     The term aF(SOTR), or aF(SOTE), is often used  to
                                                     express oxygen  transfer  under  field conditions
                                                     corrected to standard temperature and pressure and a
                                                     driving force of C"a,2o (i-e., C = 0).

                                                     Further,  since  standard oxygen transfer  efficiency,
                                                     SOTE (in percent), is:
                                             - PvT)
            SOTE =  100 (SOTR/W02)
(3-5)
                                                  37

-------
where,

   W0a  = mass rate of oxygen supplied, Ib/hr

Equation 3-3  may  also  be used  to  describe  oxygen
transfer  efficiency  under  field  conditions, OTEf,  by
direct substitution.

Although  employing  clean  water SOTR  values to
estimate oxygen transfer  rates in  process water  is
conceptually straightforward, the estimate of OTRf is
subject  to considerable doubt because  of the
uncertainties  contained  in  a  and  F.  These
uncertainties are magnified  when  the  process  water
application is based on a basin  geometry and process
temperature that differ from  those of the clean  water
test.

Table 3-1  is a  guide for  applying Equation 3-3 and
indicates the source of information for the parameters
needed to  estimate OTRf.  Values of  0*0,20 and SOTR
must be  calculated  from the  clean  water  oxygen
Iransler  lest.  The average DO  value,  G,  should
represent the desired process water DO concentration
averaged  over the  entire  aeration  volume. The
temperature  correction  factor,  r,, and  pressure
correction  factor, O,  are  estimated  using the
definitions given above.

Tablo 3-1.   Guide to Application of Equation 3-3
 Parameter              Source of Information
  SOTR
clean water test results
clean water lest results
    da    clean water test results
    C    process water conditions
    T    process water conditions
     i    calculated based on tabulated DO surface saturation
          values
    ft    calculated based on site barometric pressure and
          olfoctrvo depth data
    u    estimated based on experience or on measured
          values of KLa in cloan and process waters using a
          clean dillusor
    B    calculated based on total dissolved solids
          measurements
    0    taken as 1,024 unless experimentally proven different
    F    estimated based on field experience, field
          measurements, or laboratory analyas of diftusers
          taken from the field (see Section 3.3.3)
a  is  the  ratio of the process water  K|_a of a  new
diffuser to the clean water K|_a of a new diffuser. It is
probably  the most  controversial  and  investigated
parameter used  in  translating clean  water  oxygen
transfer  data  to  anticipated  field  performance.
Variables affecting a include aerator type,  nature  of
the  wastewater  contaminants,  position within the
treatment  scheme,  process loading rate, bulk  liquid
DO, and  airflow  rate. Reliable data  on a for various
aeration  devices are limited. Much  of the reported
data were obtained from bench-scale units that did not
properly simulate mixing and K\ja levels, aerator type,
water depth, or the geometry effects of their full-scale
counterparts. Reliable full-scale  test  procedures for
use  under process  conditions, coupled with  clean
water performance data,  are  required  to  overcome
these deficiencies. Several references provide further
information on a and its measurement (1-3).

In the past, the effects of fouling  as well as process
water effects on Kta  were included in a.  The  term
apparent a, a',  was  used to indicate  the  effects of
fouling on OTE  (4). In this manual, fouling  is defined
as an impairment of diffuser performance caused by
material attached to the diffuser and  deterioration of
diffuser materials  or  appurtenances.  Examination of
the data collected during conduct of  the  EPA/ASCE
Fine Pore Aeration Project has  led to the elimination
of the term a'.  The  term aF,  where  a continues to
describe the process water effects on  K|_a as before
and  a new term,  F,  describes the  impairment of
diffuser performance  caused by  material attached to
the diffuser and/or deterioration of diffuser materials or
appurtenances, is used in its place. In most cases, it
is believed that impairment of diffuser performance is
caused by attached material. In some cases,  however,
deterioration of  diffuser  materials  or appurtenances
may have  a significant  impact  on F (see  Section
3.3.3.6).

F is the ratio of  KLa  [or aF(SOTR) or aF(SOTE)] of a
fouled diffuser to the KLa  [or aF(SOTR) or aF(SOTE)]
of a new diffuser, both measured in the same process
water.  F generally decreases from  1.0 with time of
service  in the process water.  The characteristics of
fouling dynamics are  site  and diffuser specific. Hypo-
thetical fouling patterns are depicted in Figure 3-1.

Data  on the decrease in F with service  time  are
limited, and what data are available generally consist
of only a few points along the  F vs. time profile (see
Section  3.3.4).  Although these limited  data  are
consistent with the hypothetical profiles shown (Figure
3-1), the data  points  are too  few and imprecise to
accurately define a fouling profile. For simplicity,  and
because the precision of the data does not justify  a
complex  model, the fouling profile is  approximated at
this time by the linear model shown in Figure 3-1.

The conceptual model is illustrated in Figure 3-2. The
plot of F vs. time is developed from a linearized plot of
an  oxygen  transfer  function such  as aF(SOTE),
aF(SOTR), or  K[_a vs. time (e.g., Figure  3-1). The
slope of  the F vs. time curve represents  the fouling
rate, fp,  expressed in terms of a unit decrease per
month. Note that fp is also equal  to the slope of the
linearized curve in Figure 3-1  divided by SOTE, i.e.:

 fF =(i.O-F)*t
    = [aF0(SOTE) - aF,(SOTE)] * [t(aF0){SOTE)]     (3-6)
                                                    38

-------
Figure 3-1.  Hypothetical fouling patterns.
           aF(SOTE)
     aF0(SOTE)
      f = 1.0
                                             Linear Fouling Model
                                                             Note:
                                        Curve Must be Plotted for the Same
                                        Operating Conditions and o Values;
                                        That Is:
                                        F = aFt(SOTE)/aF0(SOTE)
                                            F = F,/F0 '= F/l
     aF,(SOTE)
   aFmln(SOTE) -
                                      Time Since Diffuser was Cleaned, t
F  relationships  are  described  by  the  following
equations:
        =  1.0-t(fp)      for tFmin  (3-7)
and,
              F = Fn
fort>tr
(3-8)
This model assumes that there is a minimum value of
F, Fmjn, that occurs after some critical service period,
tc. It may be reasonable to assume Fmjn for a system
unusually susceptible to  fouling approaches the  ratio
of the oxygen transfer performance of a coarse bubble
diffuser to that of an  unfouied fine  pore diffuser in
question,  with  both diffusers  in the  same  physical
basin configuration and  operating  under the same
process loadings.
This model is applied to the design and operation of
fine  pore aeration  systems discussed  in  Sections
4.3.3.3, 5.4.8, and 7.4.  F is discussed in more detail
in Section 3.3.

B is the ratio of the average saturation concentration,
C*«,,  in process water to the corresponding value in
clean water. B can vary from approximately  0.8 to 1.0
and   is  generally  close   to  1.0  for municipal
wastewaters. 3 cannot be measured by  a membrane
probe, and many wastewaters contain substances that
interfere  with  the  Winkler  method   of  DO
measurement.  Therefore, the value of 8 for use in
Equation 3-3 may be calculated as the ratio of the DO
surface saturation  concentration in process water to
the DO  surface  saturation  concentration  in  clean
water.  The  corresponding  surface  saturation
concentrations can be interpolated from  DO  saturation
                                                   39

-------
Figure 3-2.   Linear fouling factor model.

     1.0
  3
  u.
                                      I
             1.0 - F  1 • f oF, (SOTE) / t»F0 (SOTE) |

               I                t      I
                 Time Since Ditfuser was Cleaned, t

tables (Table C-1) based on the total dissolved solids
contents of the process water and clean water.

0  is employed to  correct  Kj_a  for  changes  in
temperature. Values of 0 range from  1.008 to 1.047
and are influenced by geometry, turbulence level, and
lypo of aeration device (5). 0 is usually taken as equal
to  1.024. Little consensus  exists  regarding  the
accurate  prediction of 0, and, for this reason, clean
water  testing  for the determination  of SOTR  values
should be at temperatures close to 20°C (88°F) (6).


3.2 Factors Affecting  Performance
The performance of diffused  aeration  systems under
process conditions is affected  by  a myriad of  factors.
Some of the more important are:
   fouling, aging, fatigue, etc.,
   wastewater characteristics,
   loading conditions,
   process type and flow regime,
   basin geometry and diffuser placement,
   diffuser performance characteristics,
   mixed liquor DO control and air supply flexibility,
   mechanical integrity of the system, and
   quality of preventive maintenance.
Manual of  Practice FD-13  (7)  is  a good general
reference on the importance of the above factors. To
minimize  life-cycle costs of an  aeration system, all
these factors  must be considered during design and
some must be controlled during operation.

The areas  of greatest  concern in process  water
oxygen  transfer  performance  are  wastewater
characteristics, process type and flow regime, loading
conditions,  and  diffuser  fouling  and  material
deterioration.  In  a given case,  any combination of
these factors  can have a significant  effect on the a
profile of a  system, DO  control,  and  changes  in
aerator performance with time due to diffuser fouling.
These factors are discussed in greater detail later in
this chapter.


3.3  Diffuser Fouling
All fine pore diffusers are susceptible  to  buildup  of
biofilms and/or deposition of inorganic precipitates that
can alter the operating characteristics of the diffusers.
Porous diffuser media are also susceptible to air-side
clogging  of  pores due to particles  in the  supply  air.
Therefore,  practical  and  cost-effective  preventive
maintenance designed to keep diffusers as clean and
efficient as possible is very important. The appropriate
preventive maintenance  procedure  and  frequency
depend on the system provided, service  conditions,
and trade-offs between operating costs and  cleaning
costs.

The effects of fouling on dynamic wet pressure (DWP)
and OTE should not be confused with changes in the
air diffusion media or the integrity of the installation.
These types of changes, which include alteration  of
diffuser media properties over time and leaks in the air
piping or around diffuser gaskets, may be  caused by
poor  equipment  design, 'improper installation,  or
inadequate inspection and maintenance.

3.3.1 Background
Porous  ceramic  plate diffusers, introduced in  the
United  States  in  the  1910s,  had become  the
predominant air diffusion  device  by  mid-century  (8-
10). Various types of  foulants were  identified  by early
investigators,  and the list has  been expanded  by
recent studies to include the following (11):
Air Side
   Dust and dirt from unfiltered air
   Oil from compressors or viscous air filters
   Rust and scale from air pipe corrosion
   Construction debris from poor cleanup
   Wastewater  solids  entering  through  broken
   diffusers or pipe leaks
Liquid Side
*  Fibrous material attached to sharp edges
•  Inorganic fines entering  media at low  or zero  air
   pressure
•  Organic  solids entering  media at low  or zero  air
   pressure
•  Oils or greases in wastewater
•  Precipitated deposits, including iron and  carbonates
•  Biological growths on diffuser media
•  Inorganic and organic solids trapped by biological
   growths on diffuser media

The rate of fouling has historically been gauged by the
rise in backpressure while in service. Since significant
levels of fouling can  take place with little  or  no
                                                   40

-------
increase  in  backpressure  but with  substantial
reductions  in  OTE, this  provided  a crude and
qualitative measure at best.

It was common practice  in earlier times to operate
several diffusers  from  a  common  plenum.  This
practice resulted in  less uniformity of air distribution
than  is obtained  today  with  the use  of  restrictive
orifices on  individual diffusers. The lack  of airflow
uniformity probably exacerbated the  rate of biological
fouling experienced in the past.

In the 1950s and  1960s, the  relative cost  of fuel vs.
labor was low. As a consequence, many  of the
ceramic  plate  installations  were replaced with less
efficient, fixed-orifice coarse bubble  diffusers. By the
middle  1970s,  the  fuel  cost-to-labor  cost  ratio
increased dramatically and  many of those installations
were replaced by porous  media  diffusers  with
individual airflow control.

Better methods of measuring the degree of fouling
and  the effectiveness of  cleaning became available  in
the early 1980s, These methods include DWP, bubble
release vacuum (BRV), the ratio of one to the other,
and   chemical and microbiological analysis. The
practice  of  employing pilot diffusers  that could  be
removed from the  basin and individually analyzed also
came into use during this period (11).

Concurrently, better methods were being developed  to
measure the oxygen transfer performance of operating
aeration  systems.  These  methods  permitted  better
appraisal  of the  effects of  fouling and  facilitated
improved  preventive maintenance scheduling.  These
methods include inert gas  tracers, off-gas analysis, a
nonsteady-state  technique  that  uses  hydrogen
peroxide,  and  DO  and respiration rate profiles (12-16),
Off-gas analysis equipment has been effectively used
to evaluate the adverse effects of fouling on both full-
scale diffused air systems and individual diffusers
(13).

3,3.2 Types of Fouling
Fouling can classified as one of two general  types,
Type  I or  Type  II. The two types  have distinct
characteristics and can occur  alone or in combination,
with  variable  dominance from treatment  plant  to
treatment plant and within the  same treatment plant
from time to time.

3.3.2.1 Type I Fouling
Type  I fouling  is  characterized by  clogging of the
diffuser pores,  either on  the air side  by  airborne
particulates or on  the liquid side by  precipitates such
as  metal hydroxides and carbonates. A  schematic
representation of Type I  fouling is illustrated in  Figure
3-3.  During the process of fouling,  the areas  of the
diffusers  with  the highest  local air flux  foul  more
rapidly. This serves to reduce the flux in the high-flow
areas and increase it in low-flow areas. The combined
effect is  to improve  uniformity  of air distribution. As
fouling progresses, the BRV coefficient of variation
decreases  and the  effective flux  ratio  approaches
unity (17).

Eventually, the accumulation of foulant in  the  pores
reduces  the   po/e  size   and  DWP  rises
correspondingly. The increase  in DWP  can exceed
the capabilities of the air supply system,  and process
air delivery may fall  short of requirements.  Also, the
reduced  effective, pore diameters  produce smaller
bubbles  such that OTE does  not  decline  and can
actually increase  slightly. An  idealized  plot of how
OTE and DWP change with time during Type I fouling
is shown in Figure 3-4.

3.3.2.2 Type II Fouling
Type  II fouling is  characterized  by the formation and
accretion  of  a biofilm layer on the surface  of the
diffuser.  In  a study  of  fouled ceramic   diffusers
collected from seven  wastewater treatment plants,
scanning electron microscopy (SEM)  was used (18) to
determine the structure of biofilms with thicknesses of
at  least  1  mm (0.04 in). Thinner  films  were  not
amenable to the investigative methods used. Figure 3-
5 is a model of Type  II fouling proposed (18)  based on
SEM  data  collected  on  the biofilms  that  had
thicknesses >1 mrn  (0.04 in).  The  microscopic  work
showed that  the biofilms were composed of bacteria!
cells enmeshed in a, matrix of  their own amorphous
exopolysaccharides.  Inorganic particles were trapped
within the bacteria! matrix. This  composition  has been
observed in  other  biofilms found  in  natural  and
industrial aquatic systems (19).

The biofilms  were  not  connected  to  the  diffuser
surface at all available points, so large spaces existed
at  the diffuser-biofilm interface.  One  large (10-14 mm
[0.4-0.55 in]) air bubble  was  seen within  the  thick
biofilm on a heavily fouled diffuser. The biofilms were
traversed by  large (>0.5 mm [0.02  in]) structured air
passages that originated at the diffuser  surface and
branched towards the  biofilm  surface  where  they
terminated  in large (0.5-1.0 mm [0.02-0.04  in])  round
apertures. Although these observations were made on
fouled ceramic diffusers, similar fouling characteristics
have  also  been observed on  perforated membrane
diffusers (20).

When a  biofiim of this structure develops  on a fine
pore  diffuser, the bubble  release  surface  changes
dramatically.  The  partial  attachment allows larger
spaces to develop between the diffuser surface and
the biofilm.  It is  hypothesized  that  air  is  conveyed
from  the diffuser  pores  through  these spaces  to
openings  at the foulant  surface  where bubble
formation occurs. The bubbles  are  larger than  those
previously formed at  the diffuser surface  because the
apertures are visually or measurably larger than the
diffuser pores. Photographs of  bubbles emitted  from
clean and  heavily fouled  diffusers  have shown that
                                                   41

-------
Figure 3-3.  Schematic structure of Type I fouling.
                               Precipitated Inorganic
                                 Foulanl Layer
 Small (10.2 mm x 0.6 mm)
     Pores in Rigid Diffuser Surface
Figure 3-4.  Idealized plot showing effects of Type I fouling
           on DWP and OTE.
 ui
 s
                         Time
bubble size increases as biofilm thickness increases
(18). The dynamics of such slimes are also thought to
play a role  in bubble size and localization of flux.

Thin biofilm layers are usually not a problem, but, as
the  biofilm  thickness  increases,   BRV  and  its
coefficient  of variation will increase, DWP will increase
slightly  or not  at  all,  and   OTE   will  decrease
substantially. Values of  BRV increase  because  the
BRV test measures the pressure differential needed to
draw air through the biofilm  even in areas where there
are no established passages. Since  air passages in
the biofilm layer  are usually not uniform across  the
entire diffuser surface, individual BRV readings exhibit
more  variability (the coefficient of variation increases).

There may or may not be an increase in DWP. When
bubbles are emitted  from a  clean diffuser, most of the
pressure differential is due to  the force required to
form bubbles against the force of surface tension, and
only a small fraction of the total  gradient  is required to
overcome  frictional  resistance  (11). Also,  surface
tension forces tend to cause equal size bubbles to be
produced from orifices of a given pore size in a well-
mixed liquid. The effects of surface tension would be
essentially  eliminated in   those  areas  where  the
bubbles  are  released  to  an   air  pocket.  If  the
subsequent frictional losses associated with  flow
through the biofilm  air passages  are small, DWP
changes would  be minimal  and,  perhaps, even
negative.

The  net result of non-uniform  bubble release  and an
increase in bubble  size is a substantial reduction  in
OTE. Figure 3-6 is an idealized representation  of OTE
and  DWP changes  with time under Type  II  fouling
conditions.                                   •

3.3.3 Fouling Characteristics
Categorizing foulants as  either  Type  I or  Type  II
provides a basis for discussion of diffuser  fouling.
However, in practice, it can be difficult to distinguish
between the two types because  they occur together,
with  one or the other dominating from treatment plant
to treatment plant, and from time to time in the same
treatment  plant. For  example,  the Green  Bay
wastewater treatment plant, which  has both  contact
and  return sludge  reaeration  basins,  demonstrated
Type II fouling in its contact basins  and both  Type I
and  Type  II fouling in two  distinct  layers  in  its
reaeration basins (20). The  same fouling phenomena
were observed with both ceramic disc  and  perforated
membrane  tube  diffusers.  Ceramic  dome  diffusers
removed from  aeration basins  at the South Meadows
wastewater treatment plant in Hartford  County,  CT
were fouled in  two layers similar  to those observed  at
the Green Bay treatment plant  (21).

Fouling can be evaluated in several ways, the simplest
of which is by  visual observation. Visual observations,
however, can be very misleading. Although rarely the
case, diffusers that appear to  be heavily fouled may
not operate very differently than clean diffusers (22).
In other cases,  diffusers that  appear relatively clean
may  operate poorly. To characterize fouling properly,
physical measurements  must also be made. The  two
key  operating  parameters  are DWP,  which  affects
blower discharge pressure,  and OTE, which  affects
the volume of  air required to provide a given  oxygen
transfer rate (see Sections 2.5.10 and 2.5.11),

3.3.3.1 Characterization Tests
Changes in DWP due ,to fouling  can be  measured
directly by collecting pressure data with the apparatus
shown in  Figures 2-9  through  2-11.  The  pressure
measurements  can  be  made  by  installing  pressure
                                                   42

-------
 Figure 3-5.   Schematic structure of Type II fouling.
                                             Large (0.5 x 0.9mm) Round Apertures
                                                  in "Leathery" Biofilm
                                                                            Smal! (0.2 x
                                                                         .6 mrn) irregular
                                                                       Pores in
                                                                       Rigid
                                                                      Diffuser
                                                                      Surface
Tight Adhesion  t****
  of Biofilm
•v*  to
 ; Diffuser
     Cavernous
  as Large as 14 x 17 mm
Between Diffuser Surface
       and Biofilm
Figure 3-6.
 UJ

 O
         Idealized plot showing effects of Type it fouling
         on DWP and OTE.
                         Time
taps in  the full-scale aeration system or by obtaining
diffusers from  removable test headers  for laboratory
testing  without  having to disrupt the  process.  The
removable header method will generally provide more
precise data  than  can be  obtained  with field
measurements.

Determining the  effects that fouling has  on OTE  is
more difficult because  the  tests  used for measuring
OTE  are  more  involved. Full-scale testing  can  be
costly,  and the results can be obfuscated by other
variables  such  as a. An alternative is to evaluate
individual diffusers in  controlled laboratory or on-site
tests. This has' been  done  with  success (20,22-25),
but it requires special equipment. Other more easily
measured  parameters  such  as  DWP,  BRV,  and
foulant  accumulation have  also been investigated  to
                        determine if they  are correlated  with  loss of OTE
                        (reduced F).

                        In the interplant fouling study conducted as part of the
                        EPA/ASCE Fine Pore Aeration Project (23), fouling of
                        ceramic disc  diffusers  was  characterized  by
                        measuring DWP, BRV, and foulant  mass accumulation
                        (dry  solids  per unit  area).  The data,  which  are
                        summarized in  Table 3-2, were collected  from tests
                        conducted  on  diffusers mounted  on removable test
                        headers  equipped  with four diffusers.  One  diffuser
                        holder was isolated and operated separately so that a
                        diffuser could be removed for evaluation and replaced
                        with  a new diffuser about  every  4  months without
                        disturbing the operation of the other three diffusers. At
                        the 8-, 12-, and 16-month intervals, one of  the non-
                        isolated  diffusers was removed for evaluation. In this
                        way,  incremental and cumulative data were collected.
                        The effect that fouling had on OTE was determined by
                        testing a  fouled diffuser side-by-side with  a  new
                        diffuser.  OTE  was  measured  using  a clean water
                        steady-state test based  on  off-gas techniques  (see
                        Appendix B). The results were reported as the ratio of
                        fouled-to-new diffuser SOTE,  i.e.,  F, after  1  year in
                        service..

                        Figure 3-7 is a plot of F vs. foulant accumulation (23).
                        Plants with small accumulations of foulant had higher
                        OTEs, as indicated by higher values of F. Likewise,
                        treatment plants with higher accumulations  of foulant
                        had  lower  OTEs.  The scatter at  low  foulant
                        accumulations was probably  related to the difficulty in
                        measuring foulant accumulation. It may also  indicate
                        that the nature  of the foulant layer, in addition to the
                        quantity of foulant, affected OTE.
                                                    43

-------
Table
Silo
No.
i

2




3



4



5



6



7




8

9




3-2. Results of Interplant
Cay • Planl
Frankonmulh, MI

Groon Bay, Wl




Milwaukee, Wl;
Joiws (stand


Milwaukee, Wl;
South Shore


Madison, Wl;
Nino Springs


Mowoo, Wl



Piano, TX;
North Toxas



Los Angotos Coiinly, CA;
Whituor Narrows
Hougliton, Ml;
Portaoo Lake



Fouling Study
Elapsed
Time,
moiilbs
5.3
12.8
4.6
8.8
8.8
14.0
16.8
4.1
8.5
8.5
13.6
4.3
8.8
8.8
13.7
4.7
7.9
7.9
12.0
4.5
8.0
8.0
12.0
4.4
7.0
11.0
18.0
18.0
9.6

4.4
8.4
8.4
12.7
12.7
Time in
Service,
months
5.3
12.8
4.6
4.2
8.8
14.0
4.2
4.1
4.4
8.5
13.6
4.3
4.5
8.8
13.7
4.7
3.2
7.9
12.0
4.5
3.5
8.0
12.0
4.4
7.0
11.0
3.0
18.0
9.6

4.4
4.0
8.4
4.3
12.7
DWP@ 1
scfm/diffuser
17.3
19.0
10.1
10.5
11.0
15.0
15.8
13.0
10.8
8.7
10.0
7.6
10.1
10.6
11.5
5.7
5.4
8.6
6.2
7.2
5.2
6.5
6.9
11.0
11.9
12.4
8.7
37.0
12,1

8.9
6.5
7.5
6.4
7.5
BRV,
in w.g.
20.2
57.3
15.0
16.5
20.0
24.8
13.9
19.0
34.9
43.0
75.7
10.2
12.3
19.6
18.0
12.4
7.7
9.3
9.3
8.1
7.8
8.2
9.3
27.4
14.3
21.4
9.5
40.7
25.0

14.8
13.5 •
18.3
13.0
19.9
Foulant
Accumulation,
•F mg/cm2
95 •
0.74 96
13
10
63
0.71 30
33
100
26
107
0.56 152
6
5
7
0.99 23
7
5
8
0.99 7
50
13
81
0.98 2.6
42
8
29
0.2
18
0.90 - ;

22
Nil
4
g
0.83 2
vs,
percent
5
11
10
12
6
12
8
10
14
13
14
13
14
"33
24
50
, 42
44
52
25
55
27
•55 '
• 5 .
6
4
23
12
-

15
-
22
12
35
Figures 3-8 and 3-9 plot F as a function of BRV and
Ihe ralio  of DWP (at  0.5 Us [1  scfm]) to BRV,
respectively (23). These parameters  yield better
correlations,  probably because they  provide  a
measure of the physical nature of the foulant structure
and how  the  structure  affects OTE. No relationship
was found in  these studies between DWP and F or
DWP  and foulant accumulation. The increase  in BRV
without a corresponding increase in D.WP indicates
that the dominant fouling in these studies was Type II.

Based on  the fouling data presented  in  Table  3-2,
diffuser fouling can be divided  into three categories:
 Severe
 Moderate
 Light
BRV. in w.q.
    >40
    15-40
DWP:BRV
   <0.3
 0.3-0.6
   >0.6
F (after 1 yr)
   <0.7
   0.7-0.9
   >0.9
Note  that diffusers from  the same treatment plant
could fit into more than one category depending on
the time in  service and  other  conditions associated
with treatment plant operations.

3.3.3.2 Foulant Properties
Several  physical  and chemical  properties  of  the
foulant samples collected during the interpiant fouling
comparison  study  mentioned above were  measured
(23). The two most notable characteristics were  the
volatile solids  and  silica contents.  The volatile solids
content was often very low  (less  than 15  percent),
even when the foulant appeared to be  a biofilm. This
is  reasonable considering that biofilm is mostly water
and thus constitutes a large volumetric fraction of the
foulant even though the gravimetric fraction is small.
The highest volatile  fractions,  50-55  percent, were
less than those of a typical  mixed liquor. In most
cases, the  silica  content was high,' about 10-20
percent of  the  residual solids after  ignition.  The
                                                  :44

-------
Figure 3-7.   F vs. foulant accumulation - Interptant Fouling Study.
 F (after 1 yr in service)

           Madison
   1.0
   0.8
   0.6
   0.4
    0.2
                     > South Shore
         Monroe

         * Portage Lake
                                                                  Frankenmuth
                        • Green Bay
                                                                                          Jones Island
                   20
                              40
               60          80         100

                 Foulanl Accumulalion, mg/cm2
                                                                            120
                                                                                       140
                                                                                                   160
Figure 3*8.   Relationship between BRV and F - Interplant Fouling Study,

 F (after 1 yr in service)
                   Madison
    1.0
             ~~~~~___"Monroe
                                      • Whittier Narrows
    0.8
    0.6
    0.4
    0.2
Portage Lake

           •
       Green Bay
                                                                            Frankenmuth
                                                                                       Jones Island
                               20
                                                      40

                                                  BRV, in w.g.
                                                                             60
                                                                                                    80
combination of low  volatile solids content and high
silica content  has  been  explained  by microscopic
observations  (18),  which  revealed  that bacterial
biofilms entrap inorganic materials.

3.3.3.3 Temporal Variations
The relationship between accumulated  foulant mass
and time in service  for eight  of  the  treatment plants
covered in the interplant fouling study is shown  in
                              Figure 3-10.  Incremental  and  cumulative foulant
                              accumulations  at  most of  the  treatment  plants
                              fluctuated throughout the study period. Although some
                              of the fluctuations undoubtedly resulted from problems
                              associated  with  sample  collection  and analysis,  it  is
                              apparent that the  accumulation  of  foulant  varied
                              considerably  throughout the study period. The data
                              were  evaluated  to determine  if  treatment  plant
                              operational parameters affected the severity of fouling
                                                     45

-------
Flfluro 3-9,  F vs. DWP:BRV - Jnterplant Fouling Study.

 F (allor 1 yr in son/tce)

  1.0
 0,8
 0,6
  0,
   4


 0,2
                             South Shore
                                                 Whittier Narrows
                                                                               Madison
                                                                                          Monroe
Portage Lake


  Frankenmuth
                                                                        * Green Bay
                  Jones Island
                           0.2
                                                  0.4

                                            DWP (at 1 scfm):BRV
                                                                        0.6
                                                                                               0,8
as indicated by BRV. No correlation between BRV and
food-to-microorganism  (F/M) loading  and  solids
retention  time  (SRT)  was  found.  It  did appear,
however, that fouling was  more severe at treatment
plants  that  did  not  have  primary clarifiers or had
substantial industrial waste contributions.

The dynamic nature of fouling  and  how fouling can  be
affected  by plant operations  were observed  at the
Monroe, Wl wastewater treatment plant (22). Rgure 3-
11 plots  aF(SOTE) vs. time in service over an 18-
month  period during  which the plant was operated
with and  without an aerated flow equalization basin in
service. During  the  first 6  months, the equalization
basin was bypassed, aF(SOTE)s were low and foulant
accumulation was high  (up to  50 mg/cm2) during this
period. At  about the 7th  month,  the aerated
equalization basin was placed  in service. Not only did
aF(SOTE)  increase,  but foulant  accumulation
eventually decreased to a negligible  amount. When
the pond was again bypassed in the 18th month,
ciF(SOTE)  once again  decreased  and  foulant
accumulation increased.

A similar type of investigation was undertaken at the
Madison, Wl treatment  plant (26).  Influent  flow  to  an
aeration  basin was  stopped  for  1 week  while the
mixed  liquor continued to  be aerated.  Off-gas  tests
were conducted before  the influent flow was stopped,
2 days alter  influent feed  was resumed, and  after
another 10 days of operation (Figure 3-12). Two  days
after resumption  of flow, aF(SOTE) was significantly
higher  than before flow  was stopped. After another  10
days of  operation, however,  aF(SOTE) had  fallen.
Under  all  the circumstances described,  SRT  in this
basin  was  approximately 2  days. Observations  at
               Monroe and Madison suggest aF is a  dynamic
               parameter that changes with loading conditions  and
               wastewater characteristics, among  other variables. It
               is important to recognize this when estimating values
               of F for a given facility.

               3.3.3.4 Spatial Variations
               In a survey of fine pore aeration treatment plants in
               the United Kingdom (27), diffuser fouling was reported
               to be  more severe at the  inlet  end of  highly loaded
               plug flow aeration basins. Data from treatment plants
               in the  United States show that these spatial  variations
               in fouling are not a common occurrence.

               At  the Madison, Wl  treatment  plant  (26), diffusers
               were collected from each of  six  grids in a three-pass,
               step feed aeration basin after being in service for 14
               months  (Table  3-3).  Fouling was  fairly uniform  as
               measured by  DWPrBRV. Foulant quantities were only
               slightly higher in the inlet grid. Varying amounts  of
               inorganic materials incorporated in  the  biofilm  layer
               probably caused  the lower  volatile solids measured at
               the  basin inlet.  Table  3-4  presents the  results  of
               inorganic analyses of  foulants obtained   from  the
               samples in  each grid. Silica  content was less in  the
               samples collected  furthest  downstream  while
               precipitation of phosphorus, magnesium, calcium, and
               iron increased in  the downstream grids, likely because
               of changes in pH.

               An  evaluation of  another  basin  at  the  Madison
               treatment plant (26)  indicated little  variation in  BRV
               and DWP from grid to grid down the length  of the first
               two passes of the three-pass, plug flow system. The
               last pass was a little  less fouled. The percent volatile
               solids  content in the foulant samples did not vary from
                                                  46

-------
Figure 3-10.  Progression of Joulant accumulation - Interplant Fouling Study.
               o
                  160-
                  140-
                  120-
                  100-
                   80-
                   60-
                   40-
                   20-
               ,0
               o>
               o
               u.
A - Ffankenmulh
x - Green Bay
v - Jones Island
M — Madison
ffl - Monroe
ffl - Norlh Texas
p - Portage Lake
O - Sou I! i Shoro
Note: Numbers in Parentheses Indicate
     Actual Time in Service (or Dilfusers
     Replaced During the Study.
                                                                                             ,(4-2)
                                            I   '   «    *    •    I    r   '
                                           .5               ,    10
                                                      Elapsed Time, months
                                                          I
                                                          15
                                 I
                                20
                                                                10
                                                      Elapsed time, months
                                                            47

-------
Figure 3-11.  oF(SOTE) vs. time in service - Monroe, Wl.


 oFcSOTE)/(t, percent

                Bypass Equalization Pond
           •*: —
   1.2
                                                                                           Bypass Equalization Pond
   1.0


   0,8


   06


   0.4


   0,2


     0
             7/8-9/86
                                                                                                        4/87
                                                                                                       26 mg/cm2
                                                                                                         11/3/87
           Juno 19B6
           All Oilluscfs
             Ctoanod
                                         200
                                                                           400
                                                                                                              600
                                                   Time in Service, days
Figure 3-12,  uF(SOTE) vs. tank length before and after load reduction - Madison, Wl.

 «P(SOT6), pofcont

   13  r—
                                    2 Days After Flow Resumed
                                                        ^ ,-*•., ,J 2 Days After Flow Resumed_ _.». —
                                                   .-""        "*""•--„   	-*--"
                                         Before Flow Disconinued
                                       (flow disconUnued for 1 week)
                                                 I
                                                 4         5

                                                      Grid Position
                                                          48

-------
Table 3-3.   Diffuser Characteristics in 3-Pass, Step Feed Aeration Basin - Madison, Wl
                                                                                        Residue
Pass

1
1
2
2
3
3
Grid
Clean Dome
1
2
3
4
5
6
BRV, in w.g.
6
48
52
45
70
46
41
S/X
0.06
0.32
0.22
0.27
0.41
0.34
0.23 "
scfm/diffuser, in w.g.
6.2
17.7
18.8
17.3
21.5
20.3
15.3
DWP:BRV
1.03
0.37
0.36
0.38
0.31
0.44
0.37
Volatile, percent
-
26
47
52
55
65
72
Mass, mg/cm2
-
26
16
19
14
12
9
 * Mean values for 4 samples after 14 months in service.
Table 3-4.   Characteristics of Diffuser Residue Samples in
           3-Pass, Step Feed Aeration Basin - Madison, Wl*

Mg
Al
Si
P
S
Cl
K
Ca
Ti
Cr
Mn
Fe
Cu
Zn
Cd

Pass
Grid 1
0.7
1.2
11.5
4.1
0.8
0.4
1.1
7.5
0.4
0.1
-
2.5
0.3
0.5
-

1
Grid 2
0.9
1.3
10.5
4.9
0.3
0.8
1,1
a.9
0.7
-
-
3.1
-
0.9
-
Percent by
Pass
Grid3
1.2
2.0
10.9
4.8
0.6
0.2
1.3
9.3
0.6
-
0.2
3.6
-
0.9
-
Weighl
2
Grid 4
1.1
2.0
11.4
5.7
0.9
0.4
1:5
10.4
0.8
-
0.3
3.7
0.7
1.0
0.3

Pass
Grids
1.0
1.5
8.0
4.9
0.5
0.3
1.0
9.0
0.6
-
0.2
2.9
0.2
0.6


3
Grid6
1.3
1.8
9.4
6.7
0.5
0.7
2.0
10.7
0.7
0.1
0.2
3.8
0.7
1.6
• 0.4
   After 550 °C firing.
the  inlet to  the outlet  of  the  basin.  Silica  and
phosphorus  did  not increase downstream; calcium,
iron, and aluminum did.

In a 28-month study at the Whittier Narrows treatment
plant in  Los Angeles  County, CA  (28), ceramic disc
and dome diffusers were  removed  from each  of three
grids of plug flow aeration basins each time the basins
were  drained.  Analyses  included  DWP, BRV,
DWP:BRV,  and foulant accumulation. An analysis of
variance was performed on the data  to determine the
effects on fouling of  time in service,  treatment with
HCI  gas,  basin  number,  and grid  number.  The
analysis showed that  the grid number (inlet, middle,
and outlet) had the least significant effect on fouling.

Mean foulant accumulations were  10.0, 6.7,  and 4.7
mg/cm2 for Grids  1, 2, and  3,  respectively.  This
decrease was not statistically significant due to the
variability in  the data.  However, visual observations
tended to confirm the mean  values reported.  The
results of the variance analysis were also affected  by
the differences in mechanical integrity of the disc and
dome  diffusers evaluated during the study.

In a  similar  study at  Green Bay,  Wl  (20), foulant
accumulation varied  substantially  from diffuser  to
diffuser, but  the variations were random. The influent
consisted of  a mixture of domestic and pulp and paper
wastewater.  There was  no clear  pattern  of  more
foulant at a basin inlet  or outlet  end. The volatile
fraction  in the biofilm  layer was relatively  constant.
These observations applied to both  ceramic disc and
perforated membrane tube diffusers. The perforated
.membrane  tube  diffusers were found to  have  a
greater accumulation on  the top of the diffuser. The
top  portion  of the foulant was  very similar to  the
foulant found on the ceramic diffusers, suggesting that
the fouling mechanisms for the ceramics and the tops
of  the  membranes  were similar.  Less  foulant
accumulated on the bottom of the tube diffusers, and
it had  much  higher volatile  fractions. These
differences  indicate that  gravity settling  may  have
played a role in  the  deposition  of inert solids in  the
biofilm layer  at this facility.

3.3.3.5 Air-Side Fouling
Many  of the early installations  of  ceramic  fine pore
diffusers had significant plugging  problems resulting in
substantial DWP  increases. The clogging  problems
were often attributed to air-side  fouling (9,10). More
recent experience has shown that air-side fouling of
porous  media diffusers is not  a common  problem.
Better air filtration and use of corrosion-resistant  air
distribution materials have eliminated plugging caused
by airborne  dirt.  Clogging caused  by intrusion  of
mixed  liquor can  be  minimized by selecting and
carefully installing systems with  good mechanical
integrity  and by  providing  good  preventive
maintenance, i.e., inspecting the system on  a regular
basis  and fixing  leaks,  not operating  the aeration
system  below the recommended  minimum airflow
rate, and avoiding loss  of the air supply  due to power
outages or other causes.
                                                   49

-------
The effects of air-side fouling can be measured by
conducting  BRV  tests  on  the  air-side  diffuser
surfaces. Such measurements were made on the air
side of the ceramic disc diffusers taken from the test
headers  used  in  the  EPA/ASCE  interplant  fouling
study (23). The test header diffusers received air that
had undergone varying degrees of paniculate removal.
The diffusers were in  service for 10-14 months.  The
results of the tests are summarized in Table  3-5. Air-
side fouling was not a problem in any of the treatment
plants tested.
Tablo 3-5. Air-side Analysis of Ceramic Disc Diffusers from
Test Headers
Cay - Plain
Houyhton, Ml:
Portage Lake
Groon Bay, Wl
Madison. Wl
Milwaukee, Wl;
South Shoro
Milwaukee, Wl;
Jones Island
Los Angotos
County, CA;
Whiiiior Narrows
Time in
Service,
months
15
14
12
13
13
NA
Air-side
BRVa,
in w.g.
5.5
5.1
4.7
5.5
4.9
5.4"
Air Filtration
12 in of glass wool
Coarse roll filter only
Static air Fillers
preceded by roll filter
Precoat bag house
Electrostatic
Glass fiber prelillers
and high-efficiency
paper filters (not used
4/86-9/87}
 a Now diffusor BRV « 5.0-5.5 in w.g.
 b Two dilfusofs removed and tested in September t987.
While clean air is essential for satisfactory operation of
porous media  diffusers,  the  degree of  filtration
required is not /irmly  established.  Recommended
design standards'for porous diffusers have been 0.1-
1.0 mg dirt/1,000  cu  ft air  (8,9).  Another common
design  criterion for air filtration  is  to remove 95
percent of all particles /0.3 micron in  size (21).
Experience indicates that these  stringent  air filtration
criteria may not always be necessary (see Chapter 8).

Data from the following treatment plants suggest that
air-side fouling  did  not occur  or  its  effects were
removed by Ihe injection of acid gas:

•  At Ihe Whittier Narrows, CA treatment  plant,  BRV
   data were collected after the air filter system had
   been out  of  service for  18  months  (28).  The
   blowers lake suction from the covered space above
   the  primary clarifiers. The average air-side BRV of
   ceramic  dome diffusers  that had been  periodically
   acid gas  cleaned  was  14.2 cm  (5.6  in)  w.g.
   (essentially  equal  to  that  of  a  new  diffuser)
   compared  to  24.4 cm (9.6 in) for untreated dome
   diffusers.

»  The  Glendale,  CA treatment  plant  uses  a
   combination of ambient air and  air from  covered
   headworks and primary clarifiers (29). The ambient
   air receives  coarse  filtration  from  oil-coated
   fiberglass frame filters on the inlet to the blowers,
   whereas the primary clarifier air is not filtered. The
   air-side BRVs for ceramic domes after 10 years of
   service  were  14.0-22.4  cm  (5.5-8.8  in)  w.g.
   (essentially equal to that of a new diffuser).
»  The rated efficiency of a modified  2-stage filter at
   the Green Bay, Wl  treatment  plant is  98 percent
   removal of  all particles  >1.0 micron in size  (20).
   After  14 months in  service, the ceramic  disc
   diffuser quadrant was  cleaned  by hosing, injection
   of 45 g (0.1  Ib). HO  gas/diffuser, and  rehosing.
   DWP  measurements  on  the cleaned diffusers
   produced  values equivalent  to  those  for  new
   diffusers. Air-side BRV data on diffusers from the
   EPA/ASCE test  header showed no air-side fouling
   (23).

•  The  air filtration system  at  the  Hartford,  CT
   treatment plant is rated to remove 95 percent of all
   particles >0.3 micron  in size (21). Visual observa-
   tion of ceramic dome  diffusers from the treatment
   plant  revealed small amounts of dried-on scale  on
   the air-side. These deposits were believed to have
   originated from  mixed liquor intrusion into  the  air
   supply piping due to broken dome bolts (plastic),
   cracked  gaskets,  and  other leaks  and  did  not
   appear to be significant.

»  At the Frankenmuth,  Ml treatment plant,  blower
   inlet  air filters  were  modified  to remove  only
   particles >20 microns in size (30).  Air-side  BRV
   values for  the  ceramic  disc diffusers over a  17-
   month study period were 14-17 cm  (5.5-6.9 in) with
   acid gas cleaning performed about once  a month.

3.3.3.6 Media  Effects
Fine pore aeration systems have sometimes exhibited
reductions in  OTE that  could  be  attributed  to
mechanical failures or material changes rather than a
or fouling effects.

Two fine  pore  ceramic diffuser  systems  were
evaluated at the Whittier  Narrows treatment  plant in
Los Angeles County, CA (28). Ceramic disc diffusers
were installed in  one  aeration basin in  December
1980, and  ceramic dome diffusers were  installed in
two aeration basins in Spring  1982. The disc diffusers
were  installed in  their  holders  with  O-rings  and
peripheral retainer  rings.  Plastic hold-down bolts and
soft gaskets were used to install the dome diffusers.

At  Whittier   Narrows,   ceramic  disc   diffusers
outperformed  ceramic  dome  diffusers  based  on
numerous off-gas tests conducted over more  than 2
years, starting  in April 1986. The dome diffusers were
inspected in the fall of 1987 by draining the basins to
just a few  inches above  the  diffusers and observing
the air release pattern. In the  inlet grid of  one of the
basins,   only  26  percent of the  diffusers  were
                                                  50

-------
performing properly. In  the  middle  and outlet grids,
<15 percent of the diffusers were operating properly.
Nearly  40  percent of  the diffuser  gaskets were
leaking. Much of the  reduced  OTE measured in the
off-gas  tests could be  attributed to diffuser design
deficiencies in conjunction with fouling that resulted in
increased  DWP and  eventual gasket failure.  In
contrast, inspection of the disc diffuser system in the
summer of 1988 revealed only five leaking O-rings in
the entire basin.

Similar  gasket failures and broken plastic  bofts were
experienced with  ceramic  dome  diffusers  at the
Hartford,  CT treatment plant  (21),  Ceramic dome
diffuser systems   at the  Madison,  Wi  (26) and
Glendale, CA (29) treatment plants, which use metal
bolts and pipe  inserts, have performed  well for many
years.

Changes in diffuser media properties can  also affect
DWP and  OTE.  As described  in  Section  2.2.3,
plasticized  PVC (a  thermoplastic elastomer) and
EPDM  (a  thermoset  elastomer), the  two principal
perforated membrane materials in use, can experience
various  physical  property changes with time when
used as wastewater aeration devices. These changes,
however, may differ significantly in nature and degree
between the two materials.  Specific, comprehensive
descriptions  of the materials  employed in  media
production are therefore  essential  for  prediction  of
perforated membrane diffuser performance.

Conditions that can substantially affect  perforated
membrane  performance  and  life  include  loss  of
plasticizer, hardening or  softening of the material, loss
of  dimensional  stability  through creep,  absorptive
and/or  extractive  exchange with wastewater, and
chemical changes resulting  from environmental
exposure.

Plasticizer  migration can  cause  hardening and
reduction  in  membrane  volume,  resulting  in
dimensional changes.  Absorption by the membrane of
various  constituents, including  oils, can  result in
softening  of the membrane  with volumetric  changes
and subsequent dimensional changes.

Membrane creep, which may be  influenced by the
above factors, will reduce OTE in some cases. It may
also be accompanied  by a reduction in  DWP to lower
than the original value after cleaning. This reduction is
not recoverable by known maintenance procedures.

Reported  data on performance  and  changes  in
characteristics of perforated  membranes  under
service  conditions,  although limited,  are  becoming
increasingly available. In three  of  the  studies
conducted under the  ASCE/EPA  Fine  Pore  Aeration
Project, plasticized PVC  membrane  tube diffusers
were evaluated.   Information  on  four  types  of
. perforated  membrane diffusers (plasticized  PVC tubes
of U.S.  manufacture,  EPDM tubes  of  European
manufacture,  and  EPDM  discs  of  both U.S.  and
European manufacture) are summarized in Table 3-6.

At Green Bay, ceramic disc and perforated membrane
tube diffusers  that  had been in operation for  18
months  were tested to determine if any changes in
diffuser media had occurred (20,31). As expected,  the
dimensions  and  strengths  of  the  ceramic disc
diffusers  had not changed. After thorough  cleaning,
DWPs of these diffusers were restored to "near new"
values.
On the  other  hand,  the Green bay perforated
membrane tube diffusers made from plasticized PVC
(Site  A-1,  Table 3-6)  experienced  changes  in
dimensions, weight, and elasticity. Visual observations
indicated a widening of the slit-type perforations. The
combination  of reduced weight, length,  and thickness
and increased tensile  modulus suggested a loss of
plasticizer. Creep was also considered  to be a
contributing  factor  in  the  enlargement  of   the
perforations.  The  relatively high  mixed  liquor
temperatures (29°C [85°F])  experienced  during  the
study probably increased the rates of both creep and
plasticizer loss over what  they would  have been at
lower,  more  typical  municipal  wastewater
temperatures.

Perforated membrane tube  diffusers  of  the same
material (plasticized PVC) were installed in the Cedar
Creek treatment plant in Nassau County, NY (Site B,
Table 3-6) in 1984-1985 (37). After about 4 years of
service, 6 diffusers were  removed and tested (32).
Little or no  change was  evident in dimensions,  but
media  weight was reduced by  approximately  10
percent.  The  membranes were more rigid  after
exposure to wastewater, as indicated by the increase
in tensile modulus from about 4,140 kPa (600 psi) to
8,825-41,370  kPa (1,280-6,000  psi).  Following
laboratory cleaning,  DWP  and OTE were both  less
than for a new diffuser.

Inspection of the same  type diffusers at the  Terminal
Island treatment plant in Los Angeles County, CA after
1 year  of service  revealed that the membranes were
no longer as flexible or loose as they were when new
(28). No  significant changes in  aF(SOTE)  were
observed over the test period, however.

Certain industrial wastewaters may have a  dramatic
effect on both plasticized PVC and EPDM perforated
membrane tube diffusers. Site C is equipped with U.S.
PVC membrane  tubes  and Site  D European  EPDM
tubes. Both  sites  are industrial wastewater treatment
plants  handling food processing  wastes.  Pertinent
conditions existing at each site are shown in Table 3-
7. The apparent loss  of  plasticizer was extreme in
both cases,  as reflected  by  losses in weight,  wall
thickness, DWP, and OTE, and the increase  in tensile
modulus.  Neither   waste  was   considered
representative of typical domestic wastewaters.
                                                  51

-------
Otffuser Physical Characteristics
Diffuser
Site Type
A-ld Tube


A-2d Disc



B Tube


C Tube


D Tube

E Tube


01 F Tube
\Jl
10

G Disc


H Disc


1 Disc


J Disc


K Disc


Membrane
Type
PVC
(U.S.)

EPDM
(U.S.)


PVC
(US.)

PVC
(U.S.)

EPDM
(Eur.)
EPDM
(Eur.)

EPDM
(Eur.)

EPDM
(Eur.)

EPDM
(US.)

EPDM
(U.S.)

EPDM
(U.S.)

"EPDM
(U.S.)

Wastewater
Type
Municipal'
Industrial

Municipal/
Industrial


Municipal


Industrial


Industrial

Industrial


Municipal


Municipal


Municipal


Municipal


Municipal


Municipal


Tune in
Service,
months
7


14



48


' 13


24

8


13


12


11.5


7


3


7


Diffuser
Condition*
New
AS«
AC
New
AS
AC

New
AS«
AC6
New
AC

New
AC
New
AS
AC
New
AC

New
AS
AC
New
AS
AC
New
AS
AC
New
AS
AC
New
AS
AC
Weight, g
120

114
174

174

118

108
117
87.5

280
222






263

. 243
174

178
174

177
175

179
174

176
Thickness,
m
0.031

0.028
0.105

0.106

0.028

0.028
0.030
0.024

0,080
0.062
0.083

0.078
0.078
0.075

0.086

0.078
0.105

0.106
0.105

0.103
0.105

0.099
0.105

0.105
Duromater
Shore A
63

63
60

68

63

75
60
81









50.8

51.1
60

62
60

61
54

59
60

64
Tensile
Modulus*.
psi
630

900
612'

728'

653

3,273
600
17,000

375
1,450
360

487
375
560

334'

31 8f












Type of Test
Specimen


Pilot

Pitot
Pilot



Field

Field


Field


Pilot

Field


Field
Field

Field
Field

Pilot
Pitot

Pilot-
Pilot

Pilot
Pitot
DWP (@ 1
sefmftSffuser), in
7.3
6.1
3.2
12.0
12.7
10.7

5.5
6.2
4,3
8.8
3.8

15.3
6.3
11.78
14.58
9.659 _
11.0
11.3

8,4
15.5
7.7
12.0
16.3
11.6
12.0
19.4
11.1
10.0
15.9
11.3
12.0
12.5
11.0
SOTEe (@ 1
scfmftjiffuser), %
18,0
11.9
13,2
19,5
18.2
18.0

18.0
13.5
14.2
18.0
14.8

23.0
17.5
21.8
19.8
21,8
23.0
19.8

17.8h
16.7h
18,0h
19.5
18.5
20.0
19.5
19.5
19.4
19.5
17.9
20.5
19.5
18.9
19.5
Ref.
20,31


32



32


32


32

33


34


35


36


32


32


32


2
fa
w
0»
0
G!
3
O
ID
Sfl
it
to
O
«4»
^*
u
ID
a.

81
IP
a
ID

tr
3
ID
g
c
ID
3
5'
en
IB
i.
8












a AS - after indicated period of service; AC - after cleaning (performed after indicated period of service).
b Radial for discs; circumferential for tubes.
° Clean water test in laboratory column at 10-ft diffuser submergence.
d Not concurrent in time.
e Mean of 6 samples.
f @ 30 psi stress.
a DWP @ 0.75 scfm/diffuser.
h @ 3 scfm/diffuser.

-------
Table 3-7.   Operating Conditions for Two Food Processing
           Industrial Treatment Sites  Using Perforated
           Membrane Tube Diffusers
 Site
 Membrane Type

 Waste Type


 Average Daily Flow, mgd
 Wastewater Temperature, °C
 pH
 Air Temperature (at blower), °C
 Diffuser Submergence, ft
 Airflow, scfrn/diffuser
 ML.SS, mg/L
 BOO5 Loading, Ib/d/i ,000 cu ft
 COD Loading, lb/d/1,000 cu ft
Plaslicized
  PVC
               D"
              EPDM
Hydrolized
Protein and
Dairy Wasle
0.10
28
6.8
-
8.0
5.0
8,000
-
Potato
. Processing
Waste
2.1
29-33
• 6.9
107
20,5
3.3
5,000
42
   50
 * Diffuser characteristics arid performance are summarized in Table
   3-6.
Vegetable oils have been suspected as the causative
agents for plasticizer loss in some of the studies cited
in Table 3-6, but  definitive investigations  were  not
performed. A  laboratory  study  was  conducted,
however,  on   European  and  U.S.  EPDM  and
plasticized  PVC materials (32). The membranes were
stretched 15  percent  on a  fixture  that was then
immersed  in  vegetable  oil.  The specimens  were
periodically weighed and tested for hardness.

Over the  1,800-hr  test  period, all  three products
exhibited significant, although different, changes. The
plasticized  PVC material lost weight and  hardened
rapidly. The U.S.  EPDM material  lost  weight and
softened gradually. The European  EPDM  material
remained at a  relatively constant weight, but  also
softened gradually. One possible explanation for these
observations is  that the PVC lost plasticizer to the oil
and the two EPDMs exchanged  plasticizer with the oil.
Although  this  study  does  not  represent  field
conditions,  it  does   indicate that  constituents
commonly found in wastewater can significantly  affect
perforated  membrane diffuser characteristics and that
the  nature  of  the  changes  can differ  among  the
various diffuser  materials.

The data in Table 3-6  suggest that changes in tensile
modulus, SOTE, and DWP  may have been occurring
for Sites A-2 and E through K. If so, loss of plasticizer
is considered  to  be the most likely cause. In  any
event, the changes appear to be gradual.

Based on the data in Table  3-6, direct comparisons of
the  different membrane diffuser types and materials
are  not considered to be  quantitatively  appropriate.
The data represent a  spectrum of wastewaters, plant
operating conditions, and exposure times over which
the actual conditions are not well known.

Only at Sites  A-1  and E were conditions accurately
known and  controlled and  the different membrane
types evaluated under near  identical conditions  and
concurrent exposures. These results are inconclusive
or not universally  applicable since the  existence of
situations with comparable  wastewaters, treatment
plants,  and  modes of  operation  is unlikely.
Nevertheless,  the data clearly indicate that significant
differences in  perforated membrane performance  and
characteristics may occur after relatively short periods
of service. These  differences should  be assessed in
the  design  phase of a  project  and   allowances
provided. This assessment is the joint responsibility of
the designer, supplier, and end user.

3,3.4 Fouling  Rates
Fouling rates  and,  more importantly,  the effects  that
foulants have  on DWP and OTE constantly change in
response to  operating  conditions,  changes in
wastewater characteristics,  and  time  in  service.  The
net  result  is  that data  collected  from operating
treatment plants contain sufficient variability that  it is
usually  difficult to  obtain  precise measurements of
fouling  rates.  Nevertheless,  it is still instructive to
investigate  and  quantify the possible  extremes of
fouling rates.  The  expected  ranges can be used for
evaluating  cleaning  frequencies  and  selecting
pressure and  air volume requirements for use in the
design of fine  pore aeration systems.

3.3.4.1 Dynamic Wet Pressure
Significant increases in DWP have not been observed
in most fine pore diffuser installations. DWP vs.  time
in service is plotted in Figure  3-13 for  ceramic  disc
diffusers tested as part of  the  EPA'ASCE interplant
fouling study (23).  Results  of testing  similar  diffusers
taken from the full-scale aeration basins at  Madison
and  Green  Bay were consistent with  the interplant
study data.  At Madison, diffusers that  had  been in
service for more than 21  months had DWPs  <25 cm
(10 in) w.g. at an  airfiow of  0.5  L/s (1 scfm)/diffuser
(26).  At Green Bay,  DWPs  of  25-38 cm (10-15 in)
w.g.  at an airflow  of 0.5  L/s (1 sefm)/diffuser were
measured after 6 months in service (20).

The  highest rate of increase in DWP was measured
for the diffusers from the  Frankenmuth,  Ml treatment
plant. After only 4 months in service, the DWP  was
about 44.5 cm (17.5  in) w.g. (30). Although such an
increase in  DWP  would affect blower efficiency  and
operating cost, it would  not  exceed   the  blower
discharge pressure provided an adequate safety factor
had been included in the original blower design.

Most  of the  treatment  plants  evaluated  in  the
EPA/ASCE  interplant fouling  study  exhibited much
smaller effects of fouling on  DWP than experienced at
Frankenmuth (23). This finding was in agreement with
                                                   53

-------
Figure 3-13. DWP fouling rates for eight wastewater treatment plants - Interplant Fouling Study.

 DWP (@ 1 scfm)/di(fiisur, in w.g.

   20 r-
   16
   12
                           Frankenmuth
                                                                                         Green Bay
                                                                                             Soulh Shore

                                                                                          Jones Island
                                                        Monroe
                            ' Madison
                                       New Ceramic Disc
                                                         I
                                            6            8

                                          Elapsed Time, monihs
                                                                     10
                                                                                 12
                                                                                              14
thai of other investigators (11). Increases in DWP over
the first 4  months ranged  from  2.5  mm  (0.1  in)
w,g./month at the Madison, WI treatment plant to 5 cm
(2 in) w.g./month at the Jones Island, Milwaukee, WI
treatment plant.  Considering  the lack  of consistent
trends in the data, regular monitoring and preventive
maintenance activities, as described in Chapter 4, are
recommended  to  keep  a  system  operating as
efficiently as possible.

3.3.4.2 Oxygen Transfer Efficiency
The effect of fouling on OTE at any time is described
by F,  The rate at which F decreases is  designated as
If, which is the slope  of the F  vs. time curve  (see
Figures 3-1  and  3-2).  While F is dimensionless, the
fouling rate factor, fp,  is expressed  as the decimal
fraction of OTE lost per unit time. The time period of 1
month is used in this  manual.  For example,  if OTE
decreases as a linear function of time and F is found
to be 0.8 after 2 months, then fp would equal (1.0  -
0.8)* 2 or 0.1 /month. If OTE does  not decrease as a
linear function of time,  then f will  vary as a function of
tune and will not approximate a constant rate,

F  and its  functional relationship with  time  can be
estimated by:

*  conducting  lull-scale OTE tests  in the  process
   water over a period of time,
»  monitoring  aeration system efficiency  using
   operational data as described in Chapter 4, or
•  conducting OTE tests on fouled and  new diffusers,
   ex situ, in an aeration column.

None  of these methods is ideal;  each has advantages
and disadvantages.

Tho full-scale methods  have  the advantage of
collecting data on the actual operating system so that
temporal and  spatial variations can  be  taken  into
account. At the same time,  temporal  loading and  a
variations may mask the fouling effects  that are of
interest.  To sort  out these  various effects,  frequent
data collection over a relatively long  time  period is
required. Although full-scale OTE tests are  the  most
accurate way to determine F as a function  of time, the
required  testing can  be  quite  expensive. Monitoring
operational data offers a cost-effective alternative for
determining F, but  care  must  be  exercised  in
collecting and  analyzing the data to ensure meaningful
results.

Since the  ex-situ method involves  removal  of the
diffusers from  the  actual operating  system,  greater
control over test conditions is gained  at the expense
of handling the diffusers before testing, which -could
have some effect on foulant characteristics.

Two ex-situ testing  techniques have  been  used to
evaluate  fine  pore  diffusers  (24,25). One  method
involves conducting tests at the plant site using mixed
liquor  pumped  continuously  from  the operating
aeration  system. Off-gas techniques are  used  to
determine OTE.  A  second  method  involves  tests
conducted  in  clean  water using  sodium sulfite  to
provide  an oxygen  demand. These  tests  can be
conducted  on  site  or  at a  remote  location  (see
Appendix B). The main disadvantages of remote clean
water testing are the need to transport the  diffusers,
which could affect foulant  characteristics,  and not
conducting the tests in process water. This technique
has  been used with success,  however,  to  screen
fouled  diffusers (20,22,23).  In these  studies, the
diffusers were  sealed in plastic bags immediately after
removal  and  shipped by same-day or overnight
transport to the testing facility.
When testing diffusers that have been in  service for
any length of time,  it is also  important to  assess the
                                                   54

-------
integrity of the diffuser media. Experience has shown
that  some diffuser material  properties  can  change,
resulting in  permanent reductions  in DWP and OTE
(20).  If  diffuser material characteristics  are  not
evaluated,  the  effects of material  changes  can be
misinterpreted as fouling effects.

The following subsections present results of tests and
data evaluations to illustrate how  the  methods des-
cribed above can and  have been used to determine or
estimate F.

a. Full-Scale OTE Testing
Off-gas  testing  was conducted  as  part  of  the
EPA/ASCE Fine Pore Aeration  Project (see Appendix
A). For those studies where it was possible, OTE data
were plotted vs. time in service since the last cleaning
of the diffusers. If the data decreased as a function of
time, linear  regression was  used. The results of  the
analyses are summarized in Table 3-8.

Although the results  of  all  regression analyses  are
presented, not all the  lines of best fit were statistically
significant. These  data were collected at operating
treatment plants that  were being tested  with several
different objectives in mind.  They were not controlled
tests for F.

One of the objectives of the studies conducted at the
Frankenmuth and  Whittier Narrows treatment plants
was to determine the effectiveness of HCI gas injec-
tion  cleaning in maintaining or restoring fine pore
diffuser OTE.

At Frankenmuth (30), Basin 5  was used as  the first
basin in  a two-basins-in-series  aeration system or as
the second  basin  in a three-basins-in-series  system.
Use  of  acid gas injection in  Basin 5 did not prevent
OTE decline. Off,-gas tests before  and after HCI gas
injection showed little or no change  in aF(SOTE)
values (Figure 3-14). It is  important to note, however,
that  HCI  gas injection  did  result in a significant
decrease in DWP in  most cases. Furthermore, it is
possible that HCI  gas injection reduced the rate of
fouling,  fp.  Since  no   controlled  studies  were
conducted, it was not possible to substantiate  whether
fp was reduced  by the acid gas treatment.

Figure  3-15 plots aF(SOTE) vs. time  since  the
ceramic  disc diffusers  in Basin 1  at  the  Whittier
Narrows treatment  plant  were initially cleaned  with
liquid acid (28). Data were collected over a 26-month
period.  HCI gas injection was performed during  the
test period at intervals of 2-4  months. As shown in
Table 3-8, fp from the 3rd through the 7th period was
0.027/month. The  use of HCI gas  did  not  prevent
long-term deterioration  in  aF(SOTE).  It appears,
however, that the HCI gas injections may have slowed
the rate of decline. The cost effectiveness of the acid
gas treatments was not determined.
At the Green Bay treatment plant (20), both ceramic
disc and  perforated  membrane  tube diffusers were
tested. The  data used for the ceramic disc diffusers
were  from three operating  periods,  one of 6-months
duration, one of 6.5-months, and  one of  4-months.
Only data from the 6-month period were used for the
membrane diffuser  system  because  the  membrane
material properties changed and the observed effects
on OTE after this period could not be attributed solely
to fouling.

A relatively low fouling rate was indicated at the Jones
Island treatment plant (38) by the linear regression of
OTE  data from 15  tests conducted in each of two
passes  in one aeration basin that were  operated in
parallel. Visual inspection  of  the  data indicated that
fouling  may have  been  rapid during  the first few
months after cleaning, followed  by no additional loss
in  OTE  for the next 27  months.  Results of  the
EPA/ASCE  interplant fouling  study   (23)  also
suggested an  initial  period of  rapid  fouling at  the
Jones, island treatment plant.

b. Operational Data Analysis
In the Green Bay study (20), nine off-gas tests were
conducted over an 18-month period.  To fill in the gaps
between off-gas testing, operational data were used to
evaluate changes in  OTE. An efficiency factor,  EF,
based on BODs loading and air usage  corrected to
zero DO,  was  calculated employing monthly average
data (see Section 4.2.3.2b). The  results are shown in
Figure 3-16.

When all the off-gas data were pooled and regression
analyses performed, a linear fouling rate appeared to
be  a reasonable  assumption. The  operational data
analysis, however, suggested two  types  of fouling
curves.  The first, between May and November  1986,
demonstrated a  rapid decrease in  OTE  over a  2-
month  period followed  by  a   relatively  constant
efficiency for the next 4 months. The data could be fit
by a  logarithmic curve or two linear  segments. The
two linear segment  approach is used  later  in  an
example in Chapter 4. During the second operating
period, OTE appeared to reduce at a reasonably linear
rate. The calculated value of fp for the second  period
was 0.064/month.

A  review of these  operating  data  illustrates the
limitation of the linear fouling rate model based on  too
few data.  While the fouling rate  may  be  linear for a
period of time, a minimum OTE will  likely be reached
at  some time  beyond  which  the  effect  of additional
fouling will not be noted. Since fouling was not rapid
in  most  of  the  studies conducted  under  the
EPA/ASCE Fine Pore Aeration  Project,  the limiting
OTE  was seldom  reached. One exception was  at
Green  Bay  (20) where  the minimum  OTE was
apparently reached  between July  and  November
1986. Full-scale off-gas tests indicated the minimum
value of F was  about 0.7 in both  the contact and
                                                  55

-------
TatJto 3-8.   Fouling Rates (fF) and Fouling Factors (F) for Selected Treatment Plants

                     "T uno in      Tosl
                    Sotvtce,d    Period,     No.       fF,      Correlation,   F (@ 1
Cily • Plant
Frankciimuiti, Ml

Groan Bay, Wl
{2nd operating
potnxl)
Groan Bay, Wl
(1st 6 months
only)

Milwaukee, Wl;
Jones Island East
Milwaukee, Wl;
South Slioro
Madison, wi;
Nino Springs
Monroe, Wl

Los Aiiyulos
County. CA;
Wluniof Narrows
Rtdyowood, NJ

years
3.3

3.0


3.3



6.0

15.0

3.5

3.0

9.3


6.0

monlhs
13.0
11.7
6.5
6.5

6.0

6.0

30.0

56.7

26.7

4.5

24.0


8.0

Tests
17
IS
9
9

5

5

20

20

37

5

23


48

month'1
0.029
0=
0.064
0.067

0.028

0.046

0.002

Oc

Oc

0.043

0.027


0.074

r
-0.52b

-0.43
-0.66»>

-0.56

-0.65»>

' -0.22

.

.-

-0.81

-0.82»


-0.15

month)
0.97
1.0
0.94
0.93

0.97

0.95

1.0

1.0

1.0

0.96

0.97


0.93

Comments
Ceramic disc diffusers; Basin 5
Ceramic disc diffusers; Basin 6
Ceramic disc diffusers: contact basin
Ceramic disc diflusers; reaeratioii basin

Perforated membrane tube diffusers;
contact basin
Perforated membrane lube diffusers;
reaeration basin
Ceramic plate diffusers

Ceramic plate diffusers; various basins

Ceramic disc diffusers

Ceramic disc diffusers during bypass of
aerated flow from equalization basin
deramie disc diffusers injected with HCI
gas 5 times about every 3 months

Ceramic dome diffusers cleaned often
by hosing or brushing with HCL solution
Ref.
30

20


20



38

39

26

22

. -28"


40

 a AS 014/1/89.
 b Significant al 95 pcrconl conltdence level.
 c Zoro (Ouhflg rate based on visual inspection.
Figure 3-14.  uF(SOTE) vs. time in service for Basin 5 - Frankenmuth, Ml.

 nF(SOTg)/H, perconl

  1.0
  0,8



  0,6



  0.4



  0,2
 Stopo » 0.00072 percenWt/day
       = 0.0216 percenWmonih

 R »  52 porceni

 From Equation 3-6: IF * 0.0216 * 0.75 = 0.029 month"1

	i	I	i	L_
                                100
                                                          200

                                                   Time in Service, days
                                                                                  • Day Before Gas Cleaning
                                                                                  x Day After Gas Cleaning
                                                                                   300
                                                                                                             400
                                                            56

-------
Figure 3-1S.  aF(SOTE) and airflow rate vs. time since initial liquid acid cleaning for Basin 1 - Whittier Narrows, CA.
 uF(SOTE)/depth of submergence, percenWt

 1.0



 0,8 P
                           Airflow Rate, sofm/sq ft basin floor
                                                                 Discs Hosed
               100
200        300        400        500

   Time Since Initial Liquid Acid Cleaning, days
                                                                       600
700
800
reaeration  basins  for  this period. To estimate  an
absolute minimum  value  of  OTE that might  occur
under unusually severe fouling conditions,  one could
assume that fine pore diffusers in their most  fouled
condition would operate at least  as  efficiently as a
coarse bubble grid system.

c. Ex-Situ Column Tests
Ex-situ  clean  water  column  tests  were  used  to
compare the effects of fouling on  OTE as part  of the
EPA/ASCE interplant fouling study (23). Diffusers that

had been in service for 10-14 months were sent to a
laboratory for OTE testing. The tests  were  conducted
in a 76-cm (30-in)  diameter basin with a 3-m  (10-ft)
water depth. Steady-state conditions were achieved
by feeding  a solution of sodium sulfite to maintain DO
at 1-3 mg/L. OTE was  determined  by  off-gas analysis.
The results of the tests are summarized in Table 3-9.

A column testing program has been used to test  fine
pore ceramic dome diffusers for use  in the retrofit of
coarse  bubble  diffusion systems (24). A test column
was submerged in an aeration basin, and mixed liquor
was pumped  through the column.  Off-gas  analysis
was used to measure OTE. To evaluate the effects of
diffuser  fouling  on  OTE, tests   were  conducted
consecutively  on  a fouled and a new diffuser.  To
avoid  significant  changes  in  wastewater
characteristics, testing  was completed within 1 hour. F
values  of  0.63 and  0.69  were  measured at  two
different treatment plants.  The time in service was  not
reported.

d. Summary
F and fp were estimated for eight  operating treatment
plants using off-gas testing data generated in full-scale
aeration basins,  fp  ranged from  0  to 0.074/month,
based  on the assumption  that F was  a linear function
of time. Although some of the linear regressions were
                           not statistically  significant, the results are  useful for
                           comparison purposes.

                           Analysis of full-scale operation  data  from the Green
                           Bay, Wl treatment plant showed that OTE decreased
                           to  a minimum value equivalent to F = 0.7. Therefore,
                           when assuming that F is a linear function of time, a
                           minimum value of OTE must also be assumed.


                           3.4 Process  Water Data Base
                           3.4.1 General Data Summary
                           As described earlier, a substantial data base exists for
                           the clean water performance  of the diffused aeration
                           systems considered in this manual. The process water
                           oxygen transfer data base is more limited.  Although
                           many techniques  are  available  for  measuring in-
                           process oxygen transfer rates,  most of  the process
                           water data base reported in this manual was collected
                           using off-gas analysis techniques (13).

                           A  summary of  process water  oxygen  transfer
                           performance data from several  evaluations at  sites
                           employing a variety of aeration devices is  presented in
                           Tables 3-10 and 3-11. The term  AD/AT in Table 3-10
                           is the ratio of the total projected  media surface area of
                           the installed diffusers to the aeration  basin floor area.
                           Footnote b in Table 3-10 defines the specific diffuser
                           areas used to calculate AD/AT values. The information
                           in  Table 3-10  provides a detailed  summary  of the
                           physical  characteristics of the  various  test sites. It
                           may be used, in conjunction with Table 3-11, to better
                           evaluate the results of oxygen transfer performance at
                           the respective  sites.  In most  instances, the  data
                           reported  herein  were  generated  during  field
                           investigations conducted as a part of the EPA/ASCE
                           Fine Pore Aeration Project,

                           The  data  in Table 3-11 include the mean  weighted
                           values for aF(SOTE)  and aF. The  ranges represent
                                                   57

-------
Figure 3-16:  EF vs. time In service - Green Bay, Wl.

        EF

    0,25


    0,20


    0.15


    0.10


    0.05
   Based on the BOD5 loading and airflow rate to the cotact
   basins and correcting the airflow rate to zero DO.

   __,	I	i	I	i	I
                        100
                                       200             300

                                                Time in Service, days
                                                                      400
                                                                                     500
                                                                                                     600
  Table 3-9.   Fouling Rates Estimated by Clean Water Column Testing
   City - Plant
Time in Service,
    months
                                 1 /month
                                      Comments
   Frankonmutli, Ml

   Green Bay. Wl
     12.8


     14.0
0.74


0.71
0.021       Lower than measured by full-scale testing
           (0.029)

0.020       Lower than measured by full-scale testing
           (0.064); linear model not appropriate for
           1 st period (see text)
Milwaukee. Wl;
Jones Island West
Milwaukee, Wl;
South Shore
Madison, Wl;
Nino Springs
Monroe, Wl
Los Angeles County, CA;
Wlutdor Narrows
HoiHjIilon. Ml;
Pofiatie Lake
13.6
13.7
12.0
12.0
9.6
12.7
0.56
0.99
0.99
0.98
0.90
0.83
0.033
< 0.001
< 0.001
0.002
0.011
0.013
No comparison with full-scale data
available
Agrees with full-scale test results
Agrees with full-scale lest results
No comparison with full-scale data
available (see text)
Lower than measured by full-scale testing
(0.027)
No comparison with full-scale data
available
    • Alter indicated time in service.

temporal  variations  in these  mean  weighted  values
and not spatial variations in oxygen transfer within the
reactors.  Spatial variations are addressed in detail in
Section 3.4.2.5.  uF was estimated using clean water
lest data lor  similar reactor geometries, airflow rates
per diffuser, and  diftuser placements. Since  many of
the data were collected after the diffusers had been in
service (or significant periods of time, the term aF was
used  instead of a (see Sections 3.1 and 3.3  for more
discussion on F).  OTEf  values collected  in  the  field
were  always corrected  to  standard  conditions  of
temperature  and pressure and a basin DO of 0 mg/L
to calculate aF(SOTE) values.

It must be emphasized  that the oxygen transfer  data
presented in Table 3-11  represent the results of many
oxygen transfer tests, each conducted over  a period
                            of several hours duration. Caution should be exercised
                            in using these  data directly for design purposes. The
                            intent of this table  is  to  give the  reader a  general
                            feeling for the  range of performance of the systems
                            listed under a variety of operating conditions.
                            3.4.2 Selected Variables Affecting  Process Water
                            Performance
                            A  review of  many oxygen transfer  process  water
                            studies indicates that  several  design  and  operational
                            variables  affect the oxygen transfer  performance of
                            fine pore aeration systems. However,  the relative lack
                            of controlled studies makes it difficult to draw strong
                            conclusions regarding the impact  of  these variables.
                            The following  sections discuss the observations made
                            to date.
                                                     58

-------
 3.4.2.1 Waste water Characteristics
 The jiwrature is replete with studies on the effects of
 waslewater characteristics on oxygen transfer. Both a
, and the nature and dynamics of fouling are attributed
 to  wastewater properties. Surfactants  (surface active
 agents) are believed to play an instrumental role in the
 depression (and  occasionally  the  increase)  of a in
 wastewaters  (47-49).  The  rise  in  a  as  treatment
 progresses down  the length  of plug  flow  aeration
 basins has been attributed to surfactant removal from
 the wastewater. Other wastewater properties may also
 have an  impact on a, including  total  dissolved solids
 (1,2)  and  transition  elements such  as  iron and
 manganese (50).
 The reader is referred elsewhere (1-3) for an in-depth
 discussion of the impact of wastewater properties on
 a.  The effect of wastewater characteristics on fouling
 has been discussed in detail in Section 3.3.3.
 It is instructive to note that aF values obtained from a
 significant number of in-process studies  (Table 3-11)
 are lower than  first anticipated for fine pore aeration
 systems  in municipal  wastewater. Although aF values
 presented in this table  are dependent on several
 process  and  design  variables for the specific
 treatment plants tested, it is apparent that the average
 mean weighted  aF is usually <0.5. The impacts of
 process  loading and  flow  regime on  this  value are
 described  in   Sections  3.4.2.4 and   3.4.2.5,
 respectively.

 The  variability  of  aF is site specific.  Examples of
 typical variations over a 24-hr period are  presented in
 Table 3-12 and  illustrated in Figure 3-17. As described
 previously, aF is affected by several  factors, including
 wastewater  characteristics  and  process conditions,
 This abbreviated data base reveals that wastewater
 strength  may have affected hour-to-hour variations of
 aF, at least  at two sites. aF variations at Ridgewood,
 NJ (Figure  3-17a) appear  to correlate  with primary
 effluent soluble TOG (40). It was also demonstrated at
 Ridgewood  that aF  decreased  with an increase in
 oxygen uptake  rate. aF(SOTE) and aF were sensitive
 to influent strength, as measured by COD  (Figure 3-
 17b), at the Whittier  Narrows treatment plant (28). At
 this facility, operated at  a low average SRT (Table 3-
 11), the effect of COD on aF extended down the basin
 to all three grids.

 In contrast,  the results of a 24-hr survey at Madison,
 Wl did not  reveal a  strong relationship  between aF
 and  influent wastewater strength  (Figure 3-17c). aF
 varied from 0.22  to  0.29  at the inlet  end  of the
 aeration  basin,  yet  influent TOG load  varied by  a
 factor  of  greater  than  2.  It is likely  the low aF
 variability at  Madison was  due  primarily to the low
 strength  of the  primary effluent (average BOD5 = 90
 mg/L).
A review of 24-hr aF variations at these sites (Table 3-
12) revealed a maximum aFraverage aF of 1.21, with
a range of 1.08-1.47.  The average value of minimum
aFraverage aF was 0.86, while the range  was 0.77-
0.96.

3.4.2.2 Diffuser  Airflow Rate
OTEs achieved  by fine  pore diffusers  in clean water
normally  decrease  as airflow  rate  per diffuser
increases, as described in Section 2.6.2.2. Equations
2-2 and 2-3  are also valid in process  waters. The
values of the constants  will  change, however,  to
reflect the effects  of process waters and  diffuser
foulants on OTE.

The  results  of  several   recent  studies  (20-
22,26,28,40,46) on fine pore aeration systems indicate
that the effect of  process conditions  on the airflow
rate-aF(SOTE)  relationship is to  shift the  curves
downward from  the corresponding clean water curve
(a typical  example is shown in  Figure 3-18). The
shape of the in-process curves varies from site to site,
however. At Ridgewood (41),  the slope of the curve
was steeper than that for clean water, while at Green
Bay (20), no discernible effect of airflow on aF(SOTE)
was  reported.  At Madison  (26),  Monroe  (22),
Glastonbury (47), and Hartford (21), the  curves were
roughly  parallel  but lower than the respective clean
water curves. At  the Jones  Island East  treatment
plant, aF(SOTE)  was constant over  a range of airflow
rates  at  the  inlet  end of  the basin  and  actually
increased  at  the effluent  end with  increased airflow
rates  (38).

In-process  data  with  values of the exponent m  in
Equation 2-3 dramatically different from those typically
determined in clean water tests have  been reported
(51).  In-process  values  of m ranged from  -0.63 to--
0.82 compared with clean water values of -0.1 to -0.3,
The  apparent extreme  change in the  relationship
between diffuser airflow  rate and OTE  may have been
caused  by the configuration of the test grids  rather
than process-related changes in diffuser performance.
Three 9.1-m (30-ft) square grids of fine pore diffusers
were  interspersed between sections of coarse bubble
diffusers.  Bulk   mixing  patterns  probably  changed
when the airflow  rates  were adjusted in this study.

An  analysis  of  data  plotted in  Figure  3-19 from
selected treatment plants listed in Table 3-10 reveals
that  a trend  exists between airflow and aF(SOTE).
Plants that were  nitrifying have been highlighted since
process conditions  will greatly affect aF(SOTE) (see
Section  3.4.2.4), Diffuser density may also  play  an
important  role   in  this  relationship.   High diffuser
densities (numbers of  diffusers/unit area of basin floor)
or high AD/AT values  will usually  result in both lower
airflow/unit area  of  diffuser  media and  higher SOTE
values.  It  has  been  reported  (44) that increasing
diffuser density increased aF(SOTE) more (from 0.75
to 1.02  percent/ft) than  would have  been  predicted
                                                    59

-------
Site
No City - Ptan
1 Frankenmuth, Ml

2 Green Bay, Wl


3 Green Bay, Wl

4 Hartford, CT

5 Milwaukee, Wl; Jones
Island East

6 Milwaukee, Wl; Jones
Island West
78,8 Madison, Wl; East


0, 9 Madison, Wl; West
O
10 Ridgewood, NJ •

1 1 Milwaukee, Wl; South
Shore

12 Los Angeles County,
CA; Whittier Narrows

13 Los Angeles County,
CA; Whittier Narrows
14 Los Angeles, CA;
Terminal Island

15 Los Angeles, CA;
Terminal Island

16 Los Angeles County,
CA; Valencia
17 Monroe, W!

No
Posses
Bastn
2
1
1
1

1
1
3
1
1


2

3


3

1

1


1


1

1


1


2
1
2

Process Type
Contact
Stabilization
Contact
Stabilization

Contact
Stabilization
Contact
Stabilization
Conventional


Conventional

Conventional


Conventional

Conventional

Conventional


Conventional


Conventional

Conventional


Conventional


Contact
Stabilization
Conventional

Flow
Regime
Plug
Flow
Plug
Flow

PlUB
Flow
Step
Feed
Plug
Flow

Plug
Flow
Plug
Ftow

Plug
Flow
Plug
Flow
Step
Feed

Plug
Flow

Plug
Flow
Plug
Flow

Plug
Flow

Plug
Flow
Step
Feed
Air
Dist
Uniform

Tapered
Uniform

Tapered
Uniform


Tapered


Uniform

Tapered


Tapered

Uniform

Uniform


Tapered


Tapered

Tapered


Tapered
Spiral

Tapered

Uniform

Dimensions of Each Pass, ft
L
44
44
244
244

244
244
194
194
370


222

135


258

116

370


300


300

300


300


54.6
"25.6
102

W
22
22
73
36.3

73
36.3
20
20
21.5


22

29.5


25.7

24

30


30


30

30


30


26.5
26.5
25

SWD
16
15
20-5
22.5

20.5
22.5
15.5
15.5
14


15

15.5


17

14.;8

15.


14.3


14.3

15


15


15
15
15

. DHf.
Sub, ft
14.1
14.1
19.1
19.1

19.1
19.1
15
15
14


15

15


16.1

14

15


12.5


12.5

.


.


13
13
14.3

DiHuset
Tyre
Ceramic
Discs
Per!. PVC
Membrane
Tubes
Ceramic
Discs
Ceramic
Domes
Ceramic
Plates

Ceramic
Plates
Ceramic
Domes

Ceramic
Disos
Ceramic
Domes
Ceramic
Plates

Ceramic
Discs

Ceramic
Domes
Perf. PVC
Membrane
Tubes
Nonrigid
Porous
Plastic Tubes
Rigid Porous
Plastic Tubes
Ceramic
Discs
No,
Aeration
Zones
per Pass
1
1
3
1

3
1
7
7
3


1

2


3

4

1


3


3

3


3


2
1
2

No
Diflusers
par Basin
400
400
4,620
1,398

6,128
2,148
2,265
1,064
1,450


2,222

3,180


3,894

650

2,448


2,026


2,686

1,000


384 2'
3833'

841
257
900

Diffuser
Density,
No.'iOOsqft
41,7

21.2


30.3

21.3

18.2


22.7

26.3


19.6

23.3

22.2


22,7


29.4

11.1


8.5


30.3

17.5

AD/AT8
0.17

0.22


0.12

0.07

0.18


0,23

0,09


0.08

0.07

0.22


0.09


0.09

0.12


0.07


0.08

0.07

Ret
30

20


20

21

38


41

26


26

40

39


28


28

28


28


28

22

Table 3-10. Pi
|
o

1
O
(D
§.'
8
o
"s
m
w
(D
B
S

1
ta

3
Tl
m
o
5
(D
tfi
3
w
5
5'
IB
O
X

-------

Site
No.
18
19
20
21&
22
23
24
25
26

City - Plant
Phoenix, AZ; 23rd
Ave.
Phoenix, AZ; 91st St.
Phoenix, AZ; 91st St.
Minneapolis, MN;
Metro
Minneapolis, MN;
Metro
U.K.; Rye Meads
U.K.; Rye Meads
Glastonbury, CT

No.
Passes
Basin Process Type
4 Conventional
4 Conventional
5 Conventional
4 Conventional
4 Conventional
2 Conventional
Inci, Anoxic
2 Conventional
3 Contact
1 Stabilization

Flow
Regime
Step
Feed
Plug
Flow
Plug
Flow
Step
Feed
Step
Feed
Plug
Flow
Plug
Flow
Plug
Flow

B>
ff
ffl
£
-p-
No. §
Aeration No. Diffuser =•
^ Dimensions of Each Pass, ft ref) niMiimr 7nnm „,„„_ noncih, g
Dist L
Tapered 310
Tapered 250
Tapered 50
Uniform 375
Uniform 375
Tapered 230
Tapered 230
Uniform 82,5
(spiral)
W SWD Sub., ft Type per Pass per Basin No./100 sq ft AD/ATa Ref. ,a
25 15.8 15 Ceramic 3 7,280 23.3 0.08 42
Domes
25 15,75 13.75 Ceramic 1 7,500 30.3 0.10 43
Domes
50 15.75 13.75 Ceramic 1 3,286 26.3 0.11 43
Discs
30 15.5 13.5 Ceramic 1 7,700 17,2 0.05 44
Domes
30 15.5 13.5 Ceramic 1 19,400 43.5 0.14 44
Domes
14,1 10.5 9.75b Ceramic 3 1,377 21.3 0,07 45
Domes
14.1 10.5 9.75C Ceramic 3 1,197 18.5 0.06 45
Domes
20 15.3 12.3 Rigid Porous 1 320 48.5 0,05 46
Plastic Tubes
a Effective (total projected) surface area of diffuser media/basin floor area.
  Ceramic disc =  0.41 sq ft; ceramic dome = 0,32 sq ft; ceramic plate  = 1,0 sq ft; perforated PVC membrane tube  = 1.047 sq ft; rigid porous plastic disc  = 0,33 sq ft; nonrigid porous
  plastic tube  = 0.33 sq ft/ft tube length; rigid porous plastic tube  = 1.08 sq ft,
b Estimated.

-------
                                                                                                                                                                            t)
                                                                                                                                                                            cr
Site
No.
1

2

3
4

5


6

7
8
g
10

11

12

13

14

15

16

17
18
10
20
21
22
23
24
25
26
oFtSOTElflt, percent
Crty - Plant
Fraflkeomuth, Ml

Green Bay, Wl

Green Bay, Wl
Hartford, CT

Milwaukee, Wl; Jones
Island East

Milwaukee, Wl; Jones
Island West
Madison, Wl; East
Madison, W; East
Madison, Wl; West
Ridgewood, NJ

Milwaukee, Wl ; South
Shore
Los Angeles County, CA;
Whittier Narrows
Los Angeles County, CA;
Whittier Narrows
Los Angeles, CA;
Terminal Island
Los Angeles, CA;
Terminal Island
Los Angeles County, CA;
Valencia
Monroe, Wl
Phoenix, AZ; 23rd Ave.
Phoenix, AZ; 91 st St. -
Phoenix, AZ; 91 st St,
Minneapolis, MN; Metro
Minneapolis, MN; Metro
Minneapolis, MN; Metro
U.K.; Rye Meads
U.K.; Rye Meads .
Glastonbury, CT
Mean
0.73

0.71

0.86
0.64

1.09


0.77

0,74
1,14
0.92
0.67

1,25

0.72

0.55

0.48

0.80

0.56

0,63
0.47
0.58
0.59
0.70
0.75
1.02
1.00
0.67
0,55
Mm
0,58

0.57

0,54
0.46

0.81


0.44

0.54
0.87
0.87
0.38

0.99

0.47

0.:25

0,27

0.62

0.40

0,54
0.44
-
.-
0.56
0.64
0.92
0.90
0.57
'0.41
Max
0,95

0.91

1.19
0.84

1,36


1.04

1.04
1.36
0.98
0,94

1.48

0.94

0.89

0.67

1.19

. 0,62

0.80
0.51
-
-
0.84
0.86
1.11
1,10
0.89
OJO
Mean
0.37

0.43

0.49
0,36

.


-

0.43
0.66
0,48
0,41

-

0.31

0.24

.

0.45

0.28

0.35
0.27
0.29
0.28
0.37
0.40
0.52
-
-
0.56
nF
Mm
0,30

0,35

0,36
0,24

«


-

0.31
0.56
0.44
0.23

-

0.21

0,11

-

0.30

0.26

0.28
0.24
-
-
0.29
0.34
0.45
-
• -
0.42
scfm/sqfP
Max
0,48

0,54

0,64
0,48

.


-

0.57 ~*
0.79
0.51
0,58

•

0,40

0.39

.

0,74

0.29

0,54
0.31
-

0.45
0.46
0.59
- •-
• .
0,67
Mean
0.80

0.73

0.62
0.36

0.20


0.17

0,28
0.23
0.20
0.39

0.14

0.25

0.32

0,68

0.47

0.43

0.36
0.53
0.37
0.33
0.43
0.43
0,39
0.21
0.21
0.19
Mm
0,63

0.46

0,39
0.18

0.12


0.08

0,15
0.19
0,19
0.11

0.08

0.15

0.20

0.39

0.23

0.27

0.29
0.31
-
-
- '
-
-
0,19
0.19
0.14
Max
0,97

1.02

0.93
0.52

0.35


0,28

0.38
0.27
0.22
0.67

0.26

0.38

0.55

1.32

0.94

0.47

0.42
0.66
-
-
-
-
-
0.24
0.22
0.29
tt> BOD5/d'
1,000 CU ft
95

149

149
52

72


71

36
12
16
20

26

30

30

.

-

-

33
38
34
33
62
28
24
36
66
27
RM», day1
0.52

0.59

0.59
.0.16

0.77


0.82

0.63
0.12
0.15
0.24

0.37

0.61

0.61
-,
'• 0,27
\
0.27

-

0.40
0.76
-
-

-
-
-
-
0.19
SRT*
days
11.0

3,0

3.0
7.6

3.8


3.3

2.2
14.0
9.6
3,1

7.5

2.6

2,6

6.6

6.6

-

8.1
1.0
2.0
1.6
5
10
9
11.7
2.7
-
scfm/BODj6
0.56

0.36

0.30
0.45

0.20


0.16

0.49
1.25
0.74
1.32

0.36

0,56

0.71

.

-

-

0.73
0.75
0.70
0.70
0.45
0,99
1,04
1.32
0.73
0.46
No.
Samples
9

9

9
12

30


21

21
8
5
48

20

35

34

8

7

7

12
6
1
1
6
18
44
-
-
6
o
w
-A
O
X
S
o
3
H
3
3
SI
(D
m
3
a.
2
0)
3
"D
3

%
o
B
a




















a scfm of airftow/sq ft basin floor area.

b Based on MLSS under aeration,

c Average scfm of airflow based on measured off-gas flux rate per unit area of basin floor/average Ib BODS applied per day.

-------
Table 3-12.  24-hr aF and aF(SOTE) Variations at Selected Municipal Treatment Plants
Site
No.
4
9
10
12
12
13
17
17
City - Plant
Hartford, CT
Madison, Wl;
West
Ridgewood, NJ
Los Angeles*
County, CA;
Whiftier Narrows
Los Angeles*
County, CA;
Whittier Narrows
Los Angeles
County, CA;
Whitlier Narrows
Monroe, Wl
Monroe, Wl

Average
0.30
0.24
0.46
0.25
0.26
0.45
0.23
0.39
ur
Minimum
0.23
0.22
0.44
0.21
0.20
0.41
0.19
0.33

Maximum.
0,44
0.29
0.59
0.27
0.30
0.50
0.28
0.45

Average
8.3
8.7
10.7
7.8
8.7
12.2
-
-
ar^ou i cf
Minimum
6.4
7.7
9.5
6.4
6.6
11.1
-
-

Maximum
11.2
10.4
13.1
8.7
9.9
13.5
!-
-
Position in Basin
Influent pass
Inlet end
Entire basin weighted
Influent grid
Middle grid
Effluent grid
Influent pass
Effluent pass
 * Dala lor 6-hr period.

because of an increase in AD/AT (from 0.05 to 0.14)
alone (see Site 22 vs. Site 23 in Figure 3-23).

The envelope depicted  in Figure 3-19  follows  the
general  shape of the  SOTE  vs.  airflow curve  for
ceramic  diffusers  (Figure  2-16). It is  not realistic to
use this curve directly for design since  many other
variables affected the  oxygen  transfer  performance
shown in  this figure.  However, it does provide an
explanation for the excellent performance of two of the
three ceramic plate installations in  Milwaukee. It
should  be noted  that the performance  of  the Mil-
waukee  Jones Island West treatment plant  (Site 6 in
Figure  3-19)  falls  below that  of  the other two
Milwaukee plate  facilities  (Sites 5 and  11).  This is
likely because of the age  of the plates in this facility
and the  air distribution  problems associated with  the
condition of the plates (10). Despite the reported con-
dition of this highly-loaded facility, which is scheduled
for rehabilitation, it  is operating at a very  low and
favorable air utilization rate per pound of applied BOD.

3.4.2.3 Diffuser Layout and Characteristics
Insufficient  data  are available  to demonstrate
significant differences in aF(SOTE) values for different
fine pore diffuser types. Few controlled  studies have
been reported, and site-to-site variations obscure any
clear delineation  between  diffuser  materials, shapes,
or  sizes. Evidence  previously  presented  in  Section
3.3.3 indicates  that  fouling  of different  diffuser
materials  may   significantly  affect  long-term
performance.
At least one study (22) has demonstrated that ceramic
diffuser specific permeabilities of 26-50 (BRVo of 15-8
cm [6-3 in w.g.]) did not  have  a dominant effect on
aF(SOTE)s in a  municipal wastewater. Operating
conditions  and wastewater characteristics appeared to
have a  greater  influence on  aF(SOTE)  than  did
diffuser pore size except for values of BRV0 less than
10 cm (4 in)  w.g.  If an optimum situation from the
standpoint  of  aF(SOTE),  operating pressure,  and
cleanability exists, it would  appear to be in the range
of specific permeabilities of 20-40 (BRV0 of 18-10 cm
[7-4  in w.g.]). The  Milwaukee experience has  been
developed from slightly lower specific permeabilities  of
15-21 (10,38,39).

It is  not possible  at this  time  to predict important
differences in aF(SOTE)  as  a  function  of  diffuser
layout. All the  treatment plants reported in Table 3-10
used a  grid  layout pattern  with  two exceptions,
Glastonbury and Terminal  Island.  Generally,  clean
water test data (Section  2.6.2.2) indicate that  grid
patterns  should  produce higher  transfer efficiencies
than  single spiral roll, mid-width,  dual spiral  roll, and
cross roll configurations.

The  performance of the  three  Milwaukee  ceramic
plate diffuser  installations  - Jones Island East (JIE),
Jones Island  West  (JIW),  and South Shore  (MSS) -
are  of interest relative  to other grid  systems
(38,39,41). As described in Section  2.6.2.2, diffuser
density affects oxygen transfer; as density or AD/AT
increases, SOTE increases. The JIW and the  MSS
plate systems  have two of the highest  AD/AT values
(0.23 and 0.22,  respectively) of the treatment  plants
presented in  Table 3-9.  The JIE  treatment  plant
AD/AT is also  high at 0.18. As a result  of  their
associated  high diffuser  densities,  these  three
treatment plants also operate at very low airflow rates
per unit area of  basin floor (0.7-1.0 L/s/m2 [0.14-0.20
scfm/sq ft]) and low diffuser air  flux  rates  (3.2-5.6
L/s/m2 [0.63-1.1  scfm/sq ft]),

SOTE and aF(SOTE) typically  increase  as unit airflow
rate  decreases,  as  shown in Figure  3-18. Two  of the
three facilities  (MSS and JIE) report two of the highest
average  aF(SOTE)  values  in  Table 3-11 for  those
                                                   63

-------
Flguro 3-17a. 34-hr variations  in  otF  and  aF(SOTE) -
            Ridgewood, NJ.
                                    Figure 3-17b. Variations in aF, aF(SOTE), and COD - Whittier
                                                Narrows, CA.
      0.6
      0,5
      0.4
                 6/16/86
             I   I  I   I  I  I   I  I   I  I   I   I  I
                                             aF

                                          0.5 r-
                                                              0.4
                                                              0.3
                                                              0.2
                                                                          Basin 3 - Effluent Grid-Domes
                                                                       Basin 2 - Middle Grid-Domes
                                                     Basin 1 -
                                                  Influent Grid-DisQS
                                                  i    li
    aF(SOTE), percent

       14
       12
       10
         800
                   6/16/86
             I   I  I  I   I  I  I   I  I   I  I
1600      2400

    Time of Day
                                           1000
                                        aF(SOTE), percent

                                           14  _-
                                                               12
                                                               10
                                                      Basin 3 - Effluent Grid-Domes
                                                                                 Basin 2 - Middle Grid-Domes
                                                                         Basin 1 -
                                                                     Influent Grid-Discs
                                                                          I   .    I    .  wl    ,
treatment plants with higher process loadings (SRTs
<8  days,  neither  plant nitrifying).  Additional support
for reducing the airflow rate through each diffuser is
that fouling is believed to  be associated with the rate
of local flux through the ceramic media (see Section
3.3.2). It may also be associated wilh the overall flux
rate as well.

These  data  indicate  fine  pore  aeration   systems
configured with high AD/AT values (thereby  incurring
low  airflow  rates per unit  area of  diffuser media) will
usually produce higher OTEs than systems with lower
AD/AT values  and higher airflows.  Although  no
controlled   studies are  available  to  support this
hypothesis, the concepts presented in Section 2.6.2.2
support this general statement.

3.4.2.4 Process Loading  Effects
A review of the dynamics  of aF(SOTE) in a variety of
activated sludge  systems  suggests  that   several
process  variables  affecting  oxygen transfer  are  not
                                        Influent COD, rng/L

                                          500
                                          400
                                          300
                                          200
                                          100
                                              700      900      1100    1300

                                                           Time of Day
                                                                              1500
                                                    64

-------
Figure 3-17c. Variations in aF, oF(SOTE), and TOC at influent
           end of aeration tank - Madison, wi.
      0.3
     0,25
      0,2
               111  i  I  i  I  i  I  i  I  i  I
    aF(SOTE), percent

      14  _
      12
      10
       6  i  i  I  i  I  i  I   i  I  i
  Prim. Effluent TOC, mg/L

     200
     150
     100
      50
         1200
                     2400         1200

                     Time of Day
clearly identifiable based on our current knowledge of
the process. For example, aF(SOTE) data collected at
Madison, WI (26) over an 800-day period  (Figure 3-
20) in the first pass of a three-pass plug flow system
demonstrate a significant variability in aF(SOTE) with
time. Some of this variability may be attributed to the
properties of the wastewater (i.e., composition and
strength),  but  cannot  account for all  of  it.  Multiple
linear regressions  of the data including SRTs, F/M
loadings,  volumetric  organic loadings, MLVSS
concentrations, and airflow rates could  account for up
to about 60-70 percent of the variability.

Similar  findings were described  for  the Whittier
Narrows treatment  plant (28), where 30-74  percent of
the variability in aF(SOTE) could be accounted for by
F/M loadings, airflow rates, and time-in-service.

These studies  suggest that elements of process
loading  may  have  some  impact   on  diffuser
performance. There are theoretical reasons  why  aF
should be a function of SRT, F/M  loading,  or MLVSS
concentration  (28).  Current  biological  treatment
models developed for  activated  sludge  predict
substrate  concentration as a function  of these
parameters.  Since  the  substrates  are partially
comprised of surfactants,  lower substrate
concentrations imply   lower  surfactant levels  and
higher a values. Furthermore, biornass production is
also related to these  variables.  The  dynamics  of
biomass accumulation  and depletion on diffuser
surfaces will very likely  influence F  (see Sections
3.3.2 and 3.3.4). The impact of these process loading
factors may be short or long term  depending on their
effects on both a and F.

Studies conducted  at the Madison, WI  East treatment
plant equipped  with  dome diffusers  revealed
significant increases in aF(SOTE)  with   increasing
SRTs (26). In  1984-85 when the East plant was not
nitrifying, SRT averaged 2.4 days  (s = 1.2)  and the
average aF(SOTE) was 11.5 percent  (s  =  2.4). In
1987  when the East plant was nitrifying, the  average
SRT  was 14.0 days  (s = 3.3)  and the average
aF(SOTE) was measured at 17.1 percent (s = 2.1).

The  Metropolitan   Waste  Commission   of
Minneapolis/St. Paul (44) also conducted studies with
dome diffusers that suggest a relationship between
process loading and aF, as depicted in Figure 3-21.
These data were collected at four   sites in  the
Minneapolis/St. Paul area,

To evaluate the  impact  of  process  loading  on
aF(SOTE),  a controlled study was performed at the
Madison, WI West treatment plant in 1986  (26). Over
a 50-day period, two parallel systems  were operated
at two different SRTs: one  at about 11 days and the
other at about 6 days. aF(SOTE) was monitored in the
first pass  of each  three-pass system.  Both systems
were operated at the same volumetric BOD5 loading
                                                  65

-------
 Figure 3-18. aF(SOTE) and SOTE vs. applied airflow rate for ceramic disc diffusers - Monroe, Wl.
    aF(SOTE) or SOTE, percent

     35
     20
     10
                                                  Clean waler
                     Gild 2.2
                                  Grid 1.2
                                                                          • Grid 1.1
                                            Grid 2.1  ~~'
                                       I
                                                                     I
                                      0.2                           0.4

                                          Airflow Rate, scfm/sq ft basin floor
                                                                                                  0.6
  Figure 3-19.  Mean uF(SQTE) vs. airflow rate per unit area of diff user media.



                                = Nitrifying by design
aF(SOTE)/ll, percent

 1.5
                                              Numbers refer to plant
                                              identifications in Table 3-10.
  1.1  —
  0.7  —
  0.3
• Ceramic Discs
o Ceramic Domes
• Ceramic Plates
a PVC Perforated Membrane Tubes
* Rigid Porous Plastic Plates
x Rigid Porous Plastic Tubes
                             2345

                                      Airflow Rate, scfm/sq ft diffuser media
and  both were nitrifying.  Results  of  this study  are
presented in Table 3-13.

In every instance, aF(SOTE) for the high-SRT system
was  higher than that of the low-SRT system. At the 95
percent confidence level,  however,  a significant
difference in aF(SOTE)  could not be demonstrated for
those Iwo operating conditions. Because of the wide
variations in aF(SOTE)s observed during the  test
period,  it is difficult  to prove that SRT alone affected
oxygen  transfer  in  this facility.  It  should  be
emphasized,  however, that the study was conducted
in the first pass of a three-pass system. Other studies
at the Madison East treatment  plant indicated that the
first-pass basins were  less sensitive to variations in
operation than the subsequent passes.

It has been shown  that, for treatment plants with  low
SRTs (1-2 days),  use  of  the  "working  definition" of
                                                      SRT [e.g.,  SRT =  X(V/QW)XW,, where  X and Xw are
                                                      MLVSS  and waste  VSS,  respectively, V is process
                                                      water volume,  and Qw is waste  solids flow  rate]
                                                      significantly underestimates  the  true  SRT  when
                                                      nonsteady-state conditions occur (28). Furthermore,
                                                      the  sampling  frequency  required  for  accurately
                                                      assessing  the  true  SRT in  low-SRT systems  is far
                                                      greater than is  practical (1.5-2 times per day).

                                                      A correlation  between aF(SOTE)  and SRT ,at  the
                                                      Whittier  Narrows  treatment  plant  was   not
                                                      demonstrated because of  the  very low SRT. On the
                                                      other  hand, a  regression  of  aF(SOTE)  vs.  MLVSS
                                                      concentration  was  much  more successful  at  this
                                                      treatment plant (28), as shown in Figure 3-22. MLVSS
                                                      concentration  should be correlated  with  growth  rate
                                                      and substrate removal since it is a component  in the
                                                      calculations of SRT and F/M loading.
                                                   66

-------
Figure 3-20, aF(SOTE) vs. time in service for ceramic disc diffusers in first pass - Madison, Wl.
 oF(SOTE)/ft, percent

 1.0
 0.8
 0.6
 0.4
 0.2
                                                  JL
      _L
                100
                           200      '   300         400        500

                                            Time in Service,' days
                                                                         600
                                                                                    700
                                                                                                800
Figure 3-21.  aF vs. organic loading for selected plants in the
            Minneapolis/St, Paul area.
  0.6
  0.5
  0.4
  0.3
       MWCC -Empire
        MWCC - Metro
                  MWCC -Empire-
                -MWCC - Melro
                              MWCC -Blue Lake
                                    MWCC - Seneca
      0     20     40    '"60"     80     100   '120

            Organic Loading, Ib BOD5/1,000 cu ft/d
Data  from  Table 3-11  were  analyzed to determine
whether a relationship  between process  loading and
aF(SOTE) may exist. A plot for the ceramic  diffuser
installations in this'table,  using SRT  as  the  loading
parameter,  is  given in Figure 3-23.  Although  wide
variations in system design  and operation, as well as
wastewater characteristics, are evident at  these  sites,
it appears a trend does exist between process loading
and aF(SOTE). Nitrifying treatment plants have  been
highlighted  in  this  figure* to  indicate  their  relative
importance to  the  relationship.  Similar efforts  to
Table 3-13.  aF(SOTE) for Aeration Systems  with Different
           SRTs at Madison, Wl West Plant

                           aF(SOTE), percent
Date
9/11/86
10/9/86
10/14/86
10/16/86
10/21/86
1.0/23/86
10/28/86
10/30/86
X
s
SRT, days
MLSS, mg/L
System 1*
14.39
10.13
9.89
10.07
10.51
9.62
9.85
9.71
10.52
1.59
11.1
1,770
System 2*
13.06
9.37
7.99
9.29
9.74
9.20
8.91
9.40
9.62
1.48
5.9
1,140
 * Results are for first pass of a 3-pass system.
develop correlations between aF(SOTE) and MLVSS
concentration  and  F/M  loading at  these ceramic
diffuser installations produced less  definitive  trends
than that exhibited between aF(SOTEj and  SRT.

3.4.2.5 Flow Regime             '
Aeration  basin  flow regime affects the mixing  pattern
of the  basin  and,  therefore,  the  residence  time
distribution  of the  influent  wastewater.  Because
wastewater  components may have  an  impact  on
aF(SOTE),  it  is  reasonable  to  expect that  mixing
patterns will also  affect aF(SOTE). A study conducted
on the  Madison, Wl ceramic dome  diffuser  system
(26)  illustrates  this concept (Figure 3-24). Single-day
aF(SOTE) profiles are  compared  as a function of grid
position for  one aeration basin when  it was operated
                                                    67

-------
Flguro 3-22.  uF(SOTE) vs. MLVSS for ceramic disc diffusers
            - Whittier Narrows, CA.

 uF(SOTE) @ Hood Position 1, percent
 (uvcrayo ol all lluuo basins)
  a
                    I
   I
                              I
                                   _L
    400        600        800       1,000

                     MLVSS, mg/U
                      1,200
first in a step feed mode vs. 2  months later when it
was operated in a plug flow pattern of three passes in
series. The switch from  step  feed  to plug  flow  was
made approximately half way between the two off-gas
test days. In both cases, SRT was  approximately 2.2
days.
 The  plug flow configuration produced  an improved
 mean weighted  aF(SOTE)  compared with the  step
 feed mode of operation (9.4 vs. 7.1 percent). Addition
 of primary effluent at several points along the  step
 feed basin resulted in depressed aF(SOTE)s at each
 feed point. Apparently,  the reduced a (or aF)  values
 associated with  each feed  point  had  a significant
 impact on the mean weighted value of aF(SOTE) for
 the step feed train. This same phenomenon  is also
 demonstrated, respectively,  in Figures 3-25 and  3-26
 as a  function of time  for the same  Madison basin
 (operated for 70 days in a plug flow mode)  (26) and
 for an aeration basin at Hartford  operated more than
 500 days in a step feed flow regime (21).

 The aF  profile along the length of any aeration basin
 will depend on the degree of mixing that exists in that
 basin. Typical results for a variety of basin geometries
 are presented in Table 3-14. The  site key given in the
 first column refers to the treatment plants described in
 Table 3-10. An examination of this table  indicates that
 treatment  plants  with high  length-to-width  ratios,
 operating as plug flow basins, generate  significant aF
 gradients. Conversely, treatment plants that are short
 and wide  (such  as Green  Bay),  or that employ  step
 feed   configurations (such  as Hartford,  Milwaukee
 South Shore, and  Phoenix -  23rd Ave.)  exhibit much
' less variability in aF along the basin length. Typical aF
 and aF(SOTE) profiles for several treatment plants are
 illustrated in Figures 3-27 through  3-31.
Figure 3-23. uF(SOTE) vs. SRT for ceramic diffuser facilities.
 oF(SOTE)/lt, porcem

  1.5  r-
  1,1
  0.7
  0,3
Numbers refer to plant
identifications in Table 3-10.
                                                    o Nitrifying
                                                    • Non-nitrifying
                                                    8

                                                SRT, days
                                                              10
                                                                          12
                                                                                     14
                                                                                                16
                                                    68

-------
Figure 3-24.  aF(SOTE) vs. tank length for plug flow and step feed aeration systems - Madison, Wl.
 aF(SOTE), percent
   16  r-
   12
                                                                             J	I
                                               56       7       8       9
                                                     Grid Position
                                                                                     10     11
                                                                                                    12
Figure 3-25.  aF vs. time in service for three passes of a plug flow aeration system - Madison, Wl East Plant.
     aF
   1.2
   0.8
   0.4
                                                              Pass No. 3
                                                                                        Pass No. 2
                                                 Pass No. 1
                                                                  I
                     10
                                    20
                                                  ,30            40

                                                  Time in Service, days
                                                                                50
                                                                                               60
                                                                                                              70
Figure 3-26.  aF vs. time in service for three passes of a step feed aeration system - Hartford, CT.
     aF

   0.6  r-
   0.5


   0.4


   0.3


   0.2

   0.1
Pass No. 1
 ^^^
  Pass No. 2
                                                           I
                        100
                                         200              300
                                                    Time in Service, days
                                                                           400
                                                                                            500
                                                                                                             600
                                                           69

-------
Tablo 3-14.  aF Profiles for Various Aeration Systems3*."
                                                               aF
Sito
No.
2


3

4

5

6

7

8

g

11

12


13


18

21

22

23

26



Zone 1
City - Plant
Oreon Bay, Wl


Grcon Bay, Wl

Hartlofd, CT

Milwaukee, Wl;
Jones Island East
Milwaukee, Wl;
Jones island Wast
Madison, Wf;
East
Madison, Wl;
East
MaUiMKi, Wl,
WoM
Milwaukee, Wl;
South SI lore
Los Anflotes
County, CA;
Wlitllior Narrows
Los Angolas
County, CA;
Wlullier Narrows
Plwonlx, AZ;
23rd Avo.
Minneapolis, MN;
Motro
Minneapolis. MN;
Motro
Minneapolis, MN;
Metro
Olastonbury, CT



Mean
0.45


0.49

0.37

0.43

0.34

0.32

0.40

0.33

0.64

0.25


0.16


0.29

0,36

.

0.30

0.59



Min
0.35


0.41

0.18

0.32

0.18

0.24

0.33

0.26

0.50

0.15


0.09


0.25

0.32

-

.

0.43



Max
0.55


0.68

0.49

0.60

0.46

0.44

0.47

0.40

0.92

0.42

1
0.27


0.34

0.40

-

.

0.69



Zone 2
Mean
0.43


0.50

0.37

0.58

0.40

0.44

0.64

0.54

0.62

0.30


0.23


0.27

0.36

-

-

0.54 '



Min
0.35


0.34

0.28

0.44

0.27

0.29

0.54

0.52

0.47

0.15


0.08


0.23

0.23

-

-

0,56



Max
0.59


0.67

0.49

0.79

0.49

0.62

0.78

0.56

0.83

0.40


0.4Q


0.31

0.42

-

-

0.77



Zone3
Mean
0.40


0.46

0.35

0.60

0.43

0.52

0.92

0.55

0.64

0.38


0.31


0.25

0.37

-

"•-

0.56 "



Min
0.31


0.30

0.24

0.47

0.25

0.36

0.77

0.52

0.51

0.22


0.17


0.21

0.24

-

-

0.37



Max
0.54


0.64

0.45

0.77

0.60

0.76

1.00

0.58

0.83

0.51


0.49


,0.30

0.45

-

-

0.65



Tola) Basin
Mean
0.43


0.49

0.36

0.54

0.39

0.43

0.66

0.48

0.63

0.31


0.24


0.27

0.37

. 0.40

0.52

' '0.56 -



Min
0.36


0.36

0.24

0.44

0.23

0.31

0.56

0.44

0.51

0.21


0.11


0.24

0.29

0.34

0.45

0.42



Max
0.53


0.64

0.48

0.68

0.52

0.57

0.79

0.51

0.75

0.40


0.39


0.31

0.45

0.46

0.59

0.67



Diffuser
Type
PVC Pert
Memb.
Tubes
Ceramic
Discs
Ceramic
Domes
Ceramic
Plates
Ceramic
Plates
Ceramic
Domes •
Ceramic
Domes
Ceramic
Discs
Ceramic
Plates
Ceramic
Discs

Ceramic
Domes

Ceramic
Domes
Ceramic
Domes
Ceramic
Domes
Ceramic
Domes
Rigid
Porous
Plastic .
Tubes
Flow
Regime
Plug
Flow

Plug
Flow
Step
Feed
Plug
Flow
Plug-
Flow
• Plug
Flow
Plug
Flow
Plug
Flow
Step
Feed
Plug
Flow

Plug
Flow

Step
Feed
1 Step
Feed
Step
Feed
Step
Feed
• Plug' '
Flow


  fl Each zono represents 1/3 ol the aeration volume.
  *» Reaeralion volume not included for contact stabilization systems.
  e SOTE tor plato rJiKusers was assumed to be 2 percent/ft submergence.
                                                             70

-------
Figure 3-27.  uF vs. basin distance for plug flow aeration system - Whittier Narrows, CA.
    oF
  0,5
  0.4
  0.3
  0.2
 0.1
             Basin 1 - Discs, HCI Gas Cleaned
                                                                     0.5
                                                                     0.4
                                                                     0.3
                                                                     0.2
                                                                     0.1
                                                                            ,  Basin 2 - Domes, HCI Gas Cleaned
                                  0.5
                                  0.4
                                  0.3
                                  0.2
                                  0.1
                                      Basin 3 - Domes, No Cleaning
                                                     100            200
                                                     Distance from irtiel, ft
                                                                                    300
Figure 3-28.  aF(SOTE) and uF vs. basin position for step feed aeration systems - Monroe, Wl.
   aF   aF(SOTE), percent
   0-6  r   12 i-
   0.5

   0.4

   0.3

   0.2

   0.1

    0
   10

    8

-   6

—   4
aF(SOTE)
                                                          aF
                                            J_
                                      J_
                                                    4           5
                                                    Basin Position
                                                            71

-------
Figure 3-29,  uF and uF(SOTE)  vs. basin distance  for plug
             flow aeration system - Ridgewood, NJ.
                                Figure 3-30.  uF(SOTE) vs. basin distance  for plug flow
                                             aeration system - Milwaukee, Wl Jones Island
                                             East Plant.
  Avcrauo Grid uF(SOTE), percent

   15 r-
   13
   11
                                     I    ,    I
                                                              uF(SOTE), percent
                                  30
                                                              20
                                                              10
                                                                                 Mir).
                                                                             I
                                                                            100         200

                                                                             Distance from Intel, ft
                                                                       300
                                                                           400
 Avorago Grid oF
 O.r
 0,6
 0,4
 0.3
  0,2
                    JL
J_
_L
             20      40      60     80

                     Distance from Intel, ft
_L
J
               100     120
                                Figure 3-31.  aF(SOTE)  vs.  basin distance for  step feed
                                             aeration system - Milwaukee, Wl South Shore
                                             Plant.

                                 aF(SOTE), percent
                                                              30
                                                              20
                                                              10
                                                                                                             Avg.
                                                                            _L
                                                            _L
                                                                            100        200        300

                                                                              Distance from First inlet, ft
                                                                                                              400
                                                         72

-------
3.5 References

When an NTIS number is cited in a  reference, that
reference is available from:

       National Technical Information Service
       5285 Port Royal Road
       Springfield, VA22161
       (703) 487-4650
1.  Stenstrom, M.  and G. Gilbert.  Effects  of Alpha,
    Beta and Theta Factors on Design  of Aeration
    Systems. Water Research 15(6):643, 1981.

2.  Doyle, M. and W.C. Boyle. Translation of Clean to
    Dirty  Water  Oxygen  Transfer Rates.  In:
    Proceedings  of Seminar/Workshop on Aeration
    System Design, Testing, Operation, and Control.
    EPA 600/9-85-005, NTIS No. PB85-173896, U.S.
    Environmental Protection Agency, Cincinnati, OH,
    January 1985.

3.  Huang, H.J.  and M.K.  Stenstrom.  Evaluation of
    Fine Bubble Alpha  Factors in Near  Full-Scale
    Equipment. JWPCF 57(12): 1143-1151, 1985.

4.  Summary  Report:  Fine  Pore (Fine Bubble)
    Aeration  Systems.   EPA-625/8-85-010,. U.S.
    Environmental Protection Agency, Cincinnati, OH,
    October 1985.

5.  ASCE Oxygen  Transfer Standards Subcommittee.
    Development of  Standard   Procedures  for
    Evaluating Oxygen Transfer Devices.  EPA 600/2-
    83-102,   NTIS  No.  PB84-147438,  U.S.
    Environmental Protection Agency, Cincinnati, OH,
    October 1983.

6.  American  Society  of Civil Engineers.  ASCE
    Standard: Measurement of Oxygen Transfer in
    Clean Water. ISBN 0-87262-430-7, New York,  NY,
    July 1984.

7.  Aeration. Manual  of Practice FD-13,  Water
    Pollution Control Federation,  Washington, DC,
    1988.

8.  Air Diffusion in  Sewage Works.  Manual of Practice
    5,  Federation of Sewage and  Industrial Wastes
    Associations, Champaign, IL, 1952.

9.  Aeration in   Wastewater Treatment.  Manual of
    Practice  5,  Water Pollution Control  Federation,
    Washington,  DC, 1971.

10. Ernest, L.A.  Case History  Report  on  Milwaukee
    Ceramic Plate Aeration Facilities. Study conducted
    under Cooperative  Agreement  CR812167,  Risk
    Reduction  Engineering  Laboratory,  U.S.
   Environmental Protection Agency, Cincinnati, OH
   (to be published).

11. Boyle, W.C. and D.T. Redmon.  Biological Fouling
   of Fine Bubble Diffusers: State-of-Art. J. Env. Eng.
   Div., ASCE 109(EE5):991-1005,  October 1983.

12. ASCE Oxygen Transfer Standards Subcommittee.
   Evaluation of  Oxygen  Transfer  Test  Procedures.
   Study conducted  under Cooperative  Agreement
   CR808840,  Risk  Reduction  Engineering
   Laboratory,  U.S.  Environmental  Protection
   Agency, Cincinnati, OH (to be published).

13. Redmon, D.T., W.C. Boyle and  L. Ewing.  Oxygen
   Transfer Efficiency Measurements in Mixed Liquor
   Using Off-Gas Techniques. JWPCF 55(11):1338-
   1347, 1983.

14. Boyle, W.C. and H.  J.  Campbell, Jr. Experiences
   with Oxygen  Transfer  Testing  of Diffused  Air
   Systems Under Process Conditions. Wat. Sci. and
   Tech. 16(10/11 ):91 -106, 1984.

15. Hovis,  J.S., J.J. McKeown,  D.  Krause,  Jr.  and
   B.B.  Benson. Gas Transfer Rate  Coefficient
   Measurement of Wastewater Aeration Equipment
   by a Stable Isotope Krypton/Lithium Technique. In:
   Gas Transfer at Water  Surfaces,  Brutsaert,  W.
   and G.H. Jirka, Editors, D. Reidel Publishing  Co.,
   Dordrecht, Holland, 1984.

16. Mueller,  J.A., R.  Sullivan  and  R.  Donahue.
   Comparison of Dome and Static Aerators Treating
   Pharmaceutical  Wastes. In:  Proceedings  of  the
   38th  Industrial  Waste  Conference,  Purdue
   University, West Lafayette, IN, May  1983.

17. Rieth, M.G., W.C. Boyle and L.  Ewing. Effects of
   Selected Design  Parameters on the Fouling  of
   Ceramic Diffusers. Presented at the  61st Annual
   Conference  of  the  Water Pollution  Control
  . Federation, Dallas, TX,  October 1988.

18. Costerton, J.W.  Investigations into  Biofouling
   Phenomena in Fine Pore Aeration Devices. Study
   conducted  under  Cooperative  Agreement.
   CR812167,  Risk  Reduction  Engineering
   Laboratory,  U.S.  Environmental  Protection
   Agency, Cincinnati, OH (to be published).

19. Costerton,  J.W. et al. Bacterial Films in  Nature
   and Disease. Ann.  Rev. Microbiol.  41:435-464,
   1987.

20. Donohue  &  Assoc.,  Inc. Fine Pore  Diffuser
   System Evaluation for the Green Bay Metropolitan
   Sewerage District.  Study  conducted  under
   Cooperative  Agreement  CR812167,  Risk
   Reduction Engineering  Laboratory,  U.S.
                                                73

-------
    Environmental Protection Agency, Cincinnati, OH
    (to be published).

21. Aeration Technologies, Inc.  Off-Gas Analysis
    Results and  Fine Pore Retrofit Case History for
    Hartford, CT MDC Facility. Study conducted under
    Cooperative Agreement  CR812167,  Risk
    Reduction  Engineering  Laboratory,  U.S.
    Environmental Protection Agency, Cincinnati, OH
    (to be published).

22. Ewing Engineering Co. The  Effect of Permeability
    On  Oxygen  Transfer Capabilities,  Fouling
    Tendencies, and Cleaning Amenability at Monroe,
    Wl.   Study  conducted under  Cooperative
    Agreement  CR812167,  Risk   Reduction
    Engineering  Laboratory,  U.S.  Environmental
    Protection  Agency,  Cincinnati,  OH  (to  be
    published).

23. Baillod, C.R. and K. Hopkins. Fouling of Fine Pore
    Diffused Aerators: An  Interplant  Comparison.
    Study conducted under Cooperative  Agreement
    CR812167,  Risk  Reduction   Engineering
    Laboratory,  U.S.  Environmental Protection
    Agency, Cincinnati, OH (to be published).

24. Rteth, M.G. and R.C.  Polta. A  Test Protocol For
    Aeration  Retrofit  To  Fine  Bubble  Diffusers.
    Presented at the 60th Annual Conference of the
    Water Pollution Control  Federation, Philadelphia,
    PA, October 1987.

25. Determination of Design Criteria  For  a Ceramic
    Disc  Fine  Pore Aeration System at the Racine
    (Wl)  Wastewater  Treatment Plant.  Report
    prepared for  City of  Racine, Wl  by Donohue  &
    Assoc., Inc., Sheboygan, Wl, December 5, 1988.

26. Boyle,  W.C. Oxygen  Transfer Studies at the
    Madison Metropolitan Sewerage District Facilities.
    Study conducted under Cooperative  Agreement
    CR812167,  Risk  Reduction   Engineering
    Laboratory,  U.S.  Environmental Protection
    Agency, Cincinnati, OH (to be published).

27. Houck,  D.H.  and A.G.  Boon.  Survey and
    Evaluation of Fine Bubble Dome Diffuser Aeration
    Equipment. EPA-600/2-81-222,  NTIS  No.  PB82-
    105578, U.S. Environmental Protection Agency,
    Cincinnati, OH, September 1981.

28. Stenstrom, M.K. Fine  Pore  Diffuser Fouling: The
    Los  Angeles Studies. Study conducted  under
    Cooperative Agreement  CR812167,  Risk
    Reduction  Engineering  Laboratory,  U.S.
    Environmental Protection Agency, Cincinnati, OH
    (to b© published).

29. Glondale  (CA)  Fine  Pore  Dome  Diffuser
    Evaluation.  Report prepared for John  Carollo
    Engineers by Ewing Engineering Co., Milwaukee,
    Wl, 1989.

30. McNamee,  Porter & Seeley  Engineers/Architects.
    Fine Pore Diffuser Case History for Frankenmuth,
    Ml.  Study conducted  under  Cooperative
    Agreement CR812167,   Risk  Reduction
    Engineering Laboratory, U.S. Environmental Pro-
    tection Agency, Cincinnati, OH (to be published).

31. Marx,  J.J., L.  Ewing, W.C.  Boyle  and  P.E,
    Thormodsgard. Full-Scale Comparison of Ceramic
    Disc  and  Flexible  Membrane Tube  Diffusers.
    Presented at the 60th Annual Conference of the
    Water  Pollution  Control Federation, Philadelphia,
    PA, October 1987.

32. Ewing  Engineering  Co. Characterization of Clean
    and Fouled Perforated Membrane Diffusers. Study
    conducted under Cooperative  Agreement
    CR812167,  Risk  Reduction  Engineering
    Laboratory, U.S.   Environmental  Protection
    Agency, Cincinnati,  OH (to be published).

33. Marx, J.J.  Fine Pore Diffuser  Study -  Summary
    Report. Report prepared for  Midtec  Paper Corp.,
    Kimberly, Wl, by Donohue & Assoc., Sheboygan,
    Wl, May5,  1989.

34. Berggruen,  S.  Examination of Membrane Tubes
    for South Haven,  Ml. Report  prepared  for  D.
    Mulac, South Haven Wastewater Treatment Plant
    by Sanitaire -  Water Pollution Control  Corp.,
    Milwaukee,  Wl, January 4,  1989.

35. Guard,  S.,  D.T.  Redmon,  D.  Bryan  and  B.
    Zimmerman. Full-Scale  Comparison of  the
    Changes in Oxygen Transfer  Efficiency of Fine
    Bubble  Membrane Diffusers. Presented  at  the
    62nd  Annual Conference of  the Water  Polluwion
    Control  Federation,  San Francisco,  CA, October
    1989.

36. Winkler, W.W.  Examination  of  Membrane Disc
    Diffusers After One Year of Operation at Wooster,
    Ohio. Report prepared for City of Wooster, Ohio,
    by Sanitaire -  Water Pollution Control  Corp.,
    Milwaukee,  Wl, January 19, 1989.

37. Mueller, J.A. and P.O. Saurer. Field Evaluation of
    Wyss Aeration  System at Cedar  Creek Plant,
    Nassau County, New York, NY. Presented at New
    York WPCA Conference, January 1987.

38. Warriner, R. Oxygen Transfer Efficiency Surveys
    at the Jones Island East Plant, August 1985-June
    1988.  Study  conducted  under  Cooperative
    Agreement CR812167,   Risk  Reduction
    Engineering  Laboratory, U.S. Environmental
    Protection  Agency,  Cincinnati,   OH  (to  be
    published).
                                                74

-------
39. Warriner,  R.  Oxygen Transfer Efficiency Surveys
   at the South  Shore Wastewater Treatment Plant,
   July  1985-March  1987. Study conducted under
   Cooperative Agreement  CR812167,  Risk
   Reduction  Engineering  Laboratory,  U.S.
   Environmental Protection Agency,  Cincinnati, OH
   (to be published).

40. Mueller, J.A.  Case  History of Fine Pore  Diffuser
   Retrofit at Ridgewood, NJ.  Study conducted under
   Cooperative Agreement  CR812167,  Risk
   Reduction  Engineering  Laboratory,  U.S.
   Environmental Protection Agency,  Cincinnati, OH
   (to be published).

41. Warriner,  R. Oxygen Transfer Measurements and
   Air Distribution for  an  Aeration  Basin with Fine
   Bubble Diffusers.  Presented  at International
   Workshop on Design  and Operation of Large
   Wastewater Treatment  Plants.  IAWPRC,
   Budapest, Hungary, September 1987.

42. Report of Off-Gas Survey Conducted at 91st Ave.
   Wastewater Treatment Plant, Phoenix, AZ. Report
   prepared  for  John  Carollo Engineers by Ewing
   Engineering Co., Milwaukee, Wl,  December 1986.
43. Ullinsky, J.D., D.T.  Redmon, P.S. Hendricks and
    W.A. Ambrose. Experiences with a  Fine Bubble
    Aeration System in a Low SRT Warm Wastewater.
    Presented at the  60th Annual Conference of  the
    Water Pollution Control  Federation,  Philadelphia,
    PA, October 1987.
44. Rieth,  M.G. and  R.C.  Polta.  Process  Water
    Oxygen Transfer Comparison Between Full-Scale
    Coarse Bubble  and Fine  Bubble Aeration
    Systems.   Presented  at the 59th Annual
    Conference of the  Water  Pollution Control
    Federation, Los Angeles, CA, October 1986.
45. Robertson, P., V.K, Thomas and  B.  Chambers.
    Energy  Savings  -  Optimization  of Fine  Bubble
    Aeration.  Final  Report and  Replicators  Guide.
    Water  Research  Centre,  Stevenage, England,
    May 1984.

46. Aeration  Technologies, Inc.  Off-Gas Analysis
    Results and  Fine  Pore Retrofit Information  for
    Glastonbury,  CT Facility,  Aeration  Tank  No. 2.
    Study  conducted  under Cooperative  Agreement
    CR812167,  Risk  Reduction  Engineering
    Laboratory,  U.S.  Environmental Protection
    Agency, Cincinnati, OH (to be published).

47. Downing,  A.L. and  L. J. Scragg.   The Effect of
    Synthetic  Detergents on the  Rate  of Aeration in
    Diffused Air Activated Sludge Plants.  Water and
    Waste Treatment Journal 102, September/October
    1958.

48. Barnnart,  E.L. Transfer of Oxygen in Aqueous
    Solutions. J. San. Eng. Div.,  ASCE 95(SA3):645-
    661, June 1969.

49. McKeown, JJ. and D.A. Okun. Effects of  Surface
    Active Agents on Oxygen Bubble Characteristics.
    J. Air and Water Pollution 5(2/4): 113-122, 1963.

50. Naime, H...  S.^ Nelson  and  D.A.  McCarthy.
    Influence of  pH,  Iron  and  Manganese
    Concentrations  on the Non-steady State Clean
    Water Test for Evaluation of Aeration  Equipment.
    In: Proceedings of Workshop  Towards  an  Oxygen
    Transfer Standard, EPA 600/9-78-021,  NTIS No.
    PB-296557,  U.S.  Environmental Protection
    Agency, Cincinnati, OH, April 1979.

51. Alibaugh, T., D.J.  Benoit  and  J.  Spangler.
    Aeration System Design Using  Off-Gas  Oxygen
    Transfer Testing.  Presented at the 58th  Annual
    Conference  of  the  Water Pollution  Control
    Federation, Kansas, MO, October 1985.
                                                75

-------

-------
                                              Chapter 4
                                    Operation and Maintenance
4.1 Introduction
The  principal  objective in the design  of  fine  pore
aeration  systems is to provide an effective system
with  the lowest  possible  present  worth  cost,
maintaining a  balance  between initial  investment and
long-term operation  and  maintenance  (O&M)
expenditures.  Many  long-term O&M characteristics
are determined  by the  capabilities  and  constraints
originally designed into the system. However, several
factors under the control of the O&M  staff will have a
significant effect on long-term O&M costs.

This  chapter  provides  guidelines  for  start-up and
shutdown, normal operation,  preventive maintenance,
restorative  maintenance, and troubleshooting of fine
pore  aeration systems.  The  intent  is  to  provide
sufficient  information  for  optimizing O&M activities.
While the  sections  are written specifically for
applications involving the activated  sludge process,
the information is readily  adaptable to other process
applications.


4.2 Operation
4.2,1  Start-up
The following  steps should be  followed when placing
an empty aeration basin into service.

1.  Before starting  up any rotating equipment,  such
   as blowers,  verify  that the equipment has  been
    properly checked out  and lubricated according to
    the manufacturer's instructions.

2.  Check  the air  piping and  diffuser system, and
    repair any loose joints, cracked piping, and  other
   defects. Confirm that piping  is free from debris
   such as  rust,  mixed liquor,  and  other residual
    matter. Leaks or faulty joints that will not stand  up
   under normal operating  stresses  could result in
   wastewater solids entering the system after  start-
    up and lead  to air-side fouling. If the system is
   equipped  with  pressure taps and  tubing for
    monitoring dynamic wet  pressure (DWP),  make
    sure all tubing connectors are tight.

3.  Check to make sure the  diffusers are installed in
    accordance with the manufacturer's specifications,
    e.g., tube diffusers  are  tightened  and oriented
    properly, gaskets and O-ring seals are elastic and
    properly seated, the system is level, and the bolts
    or  other  hardware  used  to  apply an external
    sealing force are properly adjusted.

4.  Follow the manufacturer's specifications in feeding
    air to the diffuser system before the  diffusers
    become submerged. Always  feed  at  least the
    manufacturer's  minimum  recommended airflow
    rate  per  diffuser  to  prevent  backflow  of
    wastewater through  the diffusers and into the  air
    distribution piping.

5.  Fill the aeration basin to a level of about 30 cm (1
    ft) above  the  diffusers.   Observe  the  air
    distribution,  and  check  for  significant  leaks.
    Service water, if  available,  is preferable  to
    wastewater  or mixed liquor for  the  initial filling.
    Repair  leaks  before  continuing to  fill the basin.
    Use caution  during  the early  stages of filling  to
    prevent the  force  of  incoming  water  from
    damaging the air diffusion system or its supports.

6.  Continue to fill the aeration basin while monitoring
    and adjusting the air  feed  rate.  Unless  it  is
    adjusted, the  air feed rate will decrease as the
    water level in the basin rises  and the increase in
    backpressure causes  the  blower  output  to
    decrease or  the distribution   among  basins  to
    change.

7.  Operate the  condensate blowoffs one  at a  time
    until the system is free of moisture.

8.  Adjust  the  flow  rates  of wastewater,  return
    activated sludge, and air to the basin to meet the
    desired process operating conditions.

If a basin  is put into service during cold weather and
an ice layer has formed, care must be  taken to avoid
damage due to buoyant forces  exerted by ice trying to
float when the basin is filled.

4.2.2 Shutdown
If an aeration basin must stand idle  for more than 2
weeks, it should be drained and  thoroughly cleaned.
Once cleaned, the basin should  be filled with  clean
water to a  level above the  air  diffusion system.
                                                   77

-------
Although the air distribution piping should be designed
to  withstand  the  full  range of  temperatures
encountered in the region, observance  of  the above
precautions will provide  a safety  factor  from  the
standpoint of thermal  expansion  and  contraction.
Groundwater levels and basin buoyancy must also be
considered.

If a basin  must be  taken out of service during freezing
weather,  follow the  manufacturer's recommended
procedures  to  avoid  damage.  The  aeration system
can  be idled  by  continuing  to feed  at  least  the
minimum recommended airflow rate per diffuser or the
minimum  airflow  rate  required to  keep  solids in
suspension, whichever is greater.

Do not  drain aeration basins during freezing weather
unless  absolutely  necessary because ice  formation
and  frost heave  can  cause  serious damage.  Ice
formation can be alleviated by feeding some air to the
basin through the diffusion system. Covering the basin
floor with  straw or providing heat may be required to
prevent frost heave.

When taking a  basin  out of service  for more than 2
weeks,  the following actions should be taken provided
they agree with the manufacturer's recommendations:

1.  Stop the wastewaler  and  return sludge flows to
    the  basin, but continue to feed air to the diffusers
    at or above the recommended minimum rate.

2.  Open  drain lines,  and  start  drain  pumps if
    necessary. Continue to feed air to the system until
    the  water level is below  the diffusers and  the
    diffusers are washed  off. Monitor and  adjust the
    airflow rate as the water level falls.

3.  Once  the basin is drained, verify that groundwater
    pressure relief valves are operational to prevent
    damage due to uplift pressures.

4,  Wash down basin walls and floor, air piping,  and
    diffusers to  avoid  odor  problems.  Materials
    accumulated on the diffusers should  be removed
    or at least not allowed to dry.

5.  Inspect  the  air piping and  diffusers.  Check to
    ascertain that gaskets requiring  an  externally
    applied sealing force,  e.g., as with ceramic dome
    diffusers,  are adjusted  to the  manufacturer's
    requirements.

6.  For above-freezing conditions, fill the basin with
    clean  water to an elevation of  about 1 m  {3 ft)
    above  the diffusers  or until  all  plastic piping is
    covered. This will provide  additional  protection
    against  UV  light  exposure  and excessive
    temperature changes. Feed air at a flow  rate equal
    to  or  greater  than  the  manufacturer's
    recommended  minimum.  For freezing conditions,
    more  water  may  be  needed  to protect other
    normally submerged piping. Although the air being
    fed will normally prevent serious ice damage, if an
    ice layer does form, do not drain  the water from
    the basin.  Falling ice can cause serious damage
    to the air diffusion system.

While the above procedures are considered proper for
idling a basin,  continuing to feed air for a long  time,
even  at  the manufacturer's  minimum recommended
rate, may be costly. As long as the basin is filled with
relatively  clean water,  a less costly alternative would
be to shut off the airflow completely and allow the air
distribution system to fill with water.

In  warm  weather, an  algicide should be added to
prevent  algae   growth. Continuous  airflow is
recommended if  freezing is  possible.  Otherwise,
airflow can be  turned on  periodically to turn over the
water and just prior to placing the basin back  into
service as  a check on system integrity and diffuser
performance.  Appropriate  start-up  procedures  (see
Section 4.2.1) should be followed.

If a basin or several basins are to remain idle for an
extended period because the capacity  is not needed,
the diffusers  should   be removed and  stored in
accordance    with   the    manufacturer's
recommendations.

If a basin is to be  taken out of service to effect repairs
that  can  be completed in a  few hours, no  special
cleaning  is required except in  the work area. If the
work requires several days to complete, the basin and
air diffusion system should be washed down. As a
general   rule,  inspection and  housekeeping  are
recommended  whenever the  basin  is drained
regardless of the  duration of the shutdown.

4.2.3 Normal Operation
Within the constraints placed on the activated sludge
system, the primary operational objective is to achieve
acceptable effluent quality while  maximizing  aeration
efficiency.

4.2.3.1   Factors  Affecting  Aeration  System
Efficiency
As described in Chapters 2 and 3, aeration efficiency
is affected by several controllable parameters. Among
these are:
  solids retention time (SRT),
  food-to-microorganisms (F/M) loading,
  wastewater flow regime,
  diffuser airflow rate,
  dissolved oxygen (DO) concentration,
  degree of diffuser fouling, and
  blower efficiency.
SRT (or F/M loading) and wastewater flow regime will
normally  constitute  part  of  the long-term  process
                                                  78

-------
control strategy, ranging from seasonal (to respond to
changes in raw wastewater characteristics or effluent
limits)  to many years if  stable  operation  can  bo
achieved. Diftuser airflow  rate and  mixed  liquor DO
concentration  are  part of  the  short-term, day-to-day
operating strategy. An understanding of how each of
these  parameters  affects  aeration efficiency is
mandatory in  developing  optimum  short-  and  long-
term operating strategies.

Although diffuser fouling and blower efficiency affect
aeration efficiency, control  of these factors is a matter
of minimizing  adverse effects  rather than  optimizing
an operating condition.

a. SRT and F/M Loading
There  is evidence that SRT or F/M loading affects in-
process  or field oxygen  transfer efficiency (QTEf),
increasing as SRT increases (F/M loading decreases)
over a range  of values  dependent  on the  biological
treatment system (see Chapter 3). It appears  this
relationship is dependent on the degree of stabilization
achieved in the system rather than the absolute value
of SRT or F/M loading.

Other  benefits of operating at longer SRTs  include
greater biological process stability,  better settling
characteristics,  lower  sludge   yields,  and  improved
sludge thickening  properties. Limitations to  operating
at longer SRTs include the solids loading capacity of
the  secondary  clarifiers,  sometimes  unwanted
nitrification, floating  sludge in final  clarifiers,  and
higher oxygen requirements for removing a unit mass
of BOD. Other site-specific considerations  may  also
make   operation  at high  SRTs infeasible  or
undesirable.  The determination  of  whether  it is
advantageous to operate  at higher  SRTs (lower F/M
loadings) is site specific and should be evaluated over
a  period  of   time  sufficiently  long to  produce
meaningful information.

b. Wastewater Flow  Regime
Limited data  collected under field conditions indicate
that aeration basin flow regime may  affect OTEf.  This
effect  is most likely site  specific. If the facility has
flexibility to operate under different  flow  schemes
(e.g.,   step   aeration,   plug  flow,  and  contact
stabilization), it  may  be  advantageous to experiment
with them to achieve the maximum aeration efficiency.
Note that  these alterations, like  SRT changes, will
affect  other process  performance  parameters  (e.g.,
BOD  removal  and   sludge settleabiiity)  that  may
override advantages  gained in oxygen transfer.

c. Diffuser Airflow Rate
Diffuser airflow rate  affects  aeration  efficiency by
changing OTE and  system pressure. OTE  normally
decreases as  airflow rate  increases. Operating
diffusers at the lowest airflow rate possible, while not
going   below "the  manufacturer's recommended
minimum rate, achieves maximum OTE. Based on the
clean water performance data presented in Chapter 2,
fine  pore  diffuser OTE decreases by  15-25 percent
when the diffuser airflow rate is increased from 0.5 Us
(1  scfm)/diffuser to 1.4 Us (3 scfm)/diffuser.

Increasing the  diffuser  airflow rate also  adversely
affects aeration efficiency by increasing the pressure
drop across the flow control orifices and/or diffuser
elements. The pressure drop across a clean diffuser
element, as measured  by DWP,  is  relatively small
over the normal operating range. For  example, the
change in DWP for ceramic disc diffusers operating at
0.5 and 1.4 Us (1 and 3 scfm)/diffuser is only 5 cm (2
in)  water  gauge (w.g.) or  0.5  kPa (0.07 psia). The
pressure drop  across a fixed-size orifice for the same
increase in airflow rate could be substantial,  however,
because the drop increases as the square of the flow
rate.  For  a  5-mm  (3/16-in) orifice,  the increase in
pressure drop resulting from an  increase in diffuser
airflow rate from 0.5 to 1.4 Us (1  to 3  scfm)  is about
25 cm (10 in) w.g.

There are two practical ways to adjust  diffuser airflow
rate while continuing to  meet  total  process oxygen
needs:  change  the  number of  aeration  basins  in
service or change the number of diffusers installed in
the basins.

If the  long-term total airflow rate  required to satisfy
process needs results in high  diffuser airflow  rates,
more aeration  basins can  be put  into  service  to
increase the number of operating diffusers. This will
reduce the airflow rate to all portions of an in-service
basin.  If the  airflow rate  per diffuser problem  is
localized,  diffusers   may have to  be added  to  or
removed from  the  system. Diffusers  are added  to
provide more transfer capacity where oxygen demand
is   high,  and  diffusers   are  removed  to  avoid
overaeration where demand is low. To facilitate adding
diffusers after start-up, extra diffuser holders or taps
should be provided  as part of the  original installation.
If  diffusers  are removed, the  airflow  rate  to  the
affected portion of the basin must be maintained at or
above the acceptable level required for mixing (see
Chapter 5).

Adding basins to or eliminating basins from service or
changing the number of diffusers in a basin requires a
substantial effort and should be considered long-term
adjustments.  It  is  not possible to adjust  basin  or
diffuser configuration for normal  short-term (daily or
weekly) changes in  process demands. The  aeration
system must be designed  with sufficient  flexibility to
meet these anticipated operating requirements.

d.  Dissolved Oxygen Concentration
Residual DO concentration affects OTEf by changing
the  oxygen  transfer driving  force,   which  is  the
difference  between the  in-process  saturation  DO
concentration and   the  residual  mixed  liquor  DO
concentration. The   relationship  between OTEj and
                                                   79

-------
driving  force was presented  as Equation 3-1. For
illustrative purposes,  the equation can be simplified by
assuming barometric pressure  = 101.3 kPa  (1 atm)
(fi  = 1.0} and process water temperature  = 20 °C (t
and OT-20 = 1.05.

The maximum driving force (f3C"o>20 - C) is achieved
when the system  is operated with a  residual  DO  of
zero. Since  a positive DO residual is usually required
to  obtain desired  process performance, the  driving
force will be reduced, resulting in an OTEf of less than
the maximum. To  meet the process  oxygen demand
at the tower OTEf, the airflow rate must be increased.
As  shown previously, higher airflow rates will result in
additional reductions  in OTEr.

The reduction In  OTEf caused by operating  at the
mixed liquor DO dictated by process needs  must be
considered  a  normal cost of  operation. However,
operating at a DO concentration above that  needed
for  the  process should be avoided  because  power
costs wilt increase with no improvement  in process
performance. The effect of  operating at a residual DO
of 4 mg/L vs. 2 mg/L is presented in Example 4-1.
Example 4-1  -  Effect of  Residual  DO  on Air
Requirements;
The basis for comparison is as follows:

1.  Process oxygen demand is constant.

2,  Aeration system  consists of  ceramic  dome
    diffusers operating at a  submergence of 14 ft;
    basin water depth = 15 ft.

3.  Diffuser airflow rale, q, at DO of 2  mg/L is  1
    scfm/unit.

4.  Relationship between SOTE and q in clean water
    Is:
                SOTE =  0.3 q-°-15
(4-1)
5.  Relationship between OTEf and q  in  process
    water has the same exponent (-0.15).

6.  Aeration system operating parameters are:

    C"(o2o = 10-5 rngIL
    a = constant
    F = constant
    B — 0.99
    Wastewater temp. = 20°C (t and 8T-20 = 1.0)
    Barometric pressure = l.Oatm(O =  1.0)

The in-process oxygen transfer efficiency is given by:

      OTEj =  aF (SOTE) [B(CT^o) - C]/C*»20  (4-2)
         Substituting the expression for SOTE as a function of
         q yields:

            OTEf = aF [0.3 q-o.i5]  [B(C'oo20) - C]*C"«2o (4-3)

         At a  residual DO of  2 mg/L,  the  in-process oxygen
         transfer efficiency is equal to:

           OTEf2  = aF [0.3(1.0)-0-i5] [0.99(10.5) - 2] -f 10.5
                  = aF (0.24)

         To achieve a residual  DO of 4 mg/L, the airflow  rate to
         the basin must be increased. The  in-process oxygen
         transfer efficiency now becomes:

           OTEW  = aF [0.3(q4)"0.i5] [0.99(10.5) - 4] -f 10.5
                  = aF(0.183) (q4)-°-15

         Remembering that the basis for the example was the
         oxygen demand remained  constant, the field oxygen
         transfer rate (OTRf) for the two operating conditions
         must be equal:

                         OTRf2 = OTRf4             (4-4)
                                                     or;
         or:
                                                                (OTE,2) (q2) = (OTE,4)(q4)        (4-5)
      aF (0.24) (1.0) = aF (0.183) (q4)-°-15 (q4)

Solving for q4 yields the following:

                  q4o.85 =  1.31

         q4 =  1.37 scfm (0.65 L/s)/diffuser

For the aeration  system in  the example described
above, it takes 37 percent more air to operate at  4
mg/L DO than at 2 mg/L. Of  the 37 percent increase,
about 5 percent is caused  by  a decrease in OTEf
resulting from the increase in diffuser airflow rate. The
balance of the increase is the result of the lower OTEf
caused  by  the reduction  in  driving  force.  The
attendant increase in  backpressure would be  about
0.7 kPa (0.1  psi).

Assuming constant blower efficiency and ignoring the
headless through the system, power consumption for
air compression is  directly proportional to the volume
of air compressed.  Therefore, the 37-percent increase
in airflow rate would result in  a 37-percent increase in
power  usage.  If  the aeration  system  is  properly
designed  and  installed, the  increase in  power
consumption caused by  the increases in diffuser and
air piping headlosses would be small.

The  penalty paid  in increased  power  consumption
demonstrates  the importance  of incorporating
                                                  80

-------
appropriate  control strategies that minimize  power
draw. This issue is addressed in Chapter 6.

4.2.3.2 Aeration System Monitoring
The  aeration system must  be monitored to provide
data  for optimizing  system   performance  and
maintenance  schedules. Monitoring can lead  to
optimization of system aeration  efficiency  in  three
ways. First,  the optimization of DO control,  by which
most  of  the power  savings  are  achieved,  relies  on
frequently-collected DO concentration data. Second,
the effects of process operational parameters such as
SRT and F/M loading on OTEf can be better defined
for the specific application. Third, the adverse effects
of diffuser fouling on  backpressure and OTEf can  be
identified  so that  maintenance  procedures can  be
initiated.  Data should  be collected often enough to
define normal variations and to permit recognition of
long-term  changes.  In- some  cases,  however,
collection frequency  may be dictated  by  controller
requirements.

To augment full-scale monitoring, it is often  helpful to
have  a source of diffusers from the operating system
for laboratory  testing.  For fixed-grid  systems,
removable test headers (described  in Section 2.5.11)
are recommended  for this purpose. Since fouling rates
can vary with location  in plug flow aeration basins,
several test headers or monitoring stations  should  be
provided.

The   removable  headers  can   be  equipped  with
pressure taps for  in-situ monitoring of DWP. This is
not" essential, however,  since DWP can be measured
in the laboratory, eliminating problems associated with
field  measurement. It is important,  however, that the
test units be operated  at airflow rates similar to the
full-scale system.  Airflow rate to individual diffusers
can easily be determined for units having  fixed-size
flow  control orifices  by measuring the pressure  drop
across the orifice.  Airflow rate can also be measured
using one or more rotameters.

a. Air Delivery System Pressure
Air-side or liquid-side fouling may cause an increase
in diffuser  headioss  at  constant airflow  rates.
Increases  in  wet  headioss  can  be  detected  by
monitoring operating  conditions  within the  air supply
system.  Depending on  the specific design  approach,
an increase in air supply system pressure (monitored,
for example, in the blower discharge header or  by
increased opening of the flow  control  valves) can
indicate  an  increase  in diffuser  headioss.  Significant
increases in 'blower  pressure may be  indicative of
extensive fouling  of  major  portions  of  the diffuser
system.  For this reason, blower  pressure along with
airflow rate should be monitored on a daily basis.

While overall system  pressure monitoring serves as a
potential indicator of  extreme fouling,  it  does  not
provide  a  very  sensitive  indication of increased
diffuser headioss.  The  precision of this  method  is
usually inadequate because the differential  pressure
across the diffuser element is  small  compared  with
the pressure in the air main. Other  factors - such as
surface elevation, water temperature, airflow rate( and
variable line losses - further limit precision.  In addition,
fouling of only a portion of the  diffusion system  may
lead to a substantial redistribution in airflow but  little
increase  in overall system pressure.  For many fine
pore systems, use of a monitoring technique more
sensitive  to fouling than system pressure is desirable
and often necessary.

Increased headioss  sensitivity   is   provided  by
measuring  DWP  rather than system  pressure  (see
Section 2.5.11). DWP  can be  measured by fixed
pressure  monitoring  stations  located  throughout  the
aeration  system (see  Figure  2-9).  Fixed  station
monitors  require  continual maintenance  to  ensure
accurate and  precise  DWP   measurements,  as
discussed in Section 2.5,11.

DWP measurements  can  also  be  performed in a
laboratory using  diffusers taken  from  removable test
headers  (see Figure  2-11). These diffusers  can  also
be used in other analyses described later.  Removable
test headers  require careful maintenance  and
operation, however, if the data collected are to be of
value.  Furthermore, raising  the test  header out of the
mixed  liquor to  remove   the  diffusers  may be a
considerable nuisance to the operator.

DWP measurements should be made and  recorded,
weekly for the first 6  months of operation to provide
baseline  data  and determine  if DWP is  increasing
rapidly. After the first 6 months, the  frequency can be
reduced to monthly or even less often. Since DWP is
a function of diffuser airflow rate, all readings should
be made  at approximately the same airflow  rate.
b. Aeration System Efficiency
Since diffuser fouling can substantially decrease OTEf
without  significant  attendant  increases  in
backpressure,  effective process monitoring  must
include  other  parameters besides diffuser  headioss
and  system  pressure  measurements.  Rigorous
methods for measuring OTE under process  conditions
are available (1-3). One or more methods may be
appropriate for a particular aeration system. However,
these methods usually are  time  consuming  and may
be too costly for use  in day-to-day monitoring. As an
alternative, calculated  ratios  of  operating  data can
provide good indications  of overall system  efficiency.
if the monitoring program indicates a need  for more
detailed data, a more rigorous test can be performed.

A parameter based on the ratio of the rate  of oxygen
demand  removed to rate of oxygen  supplied can be
conveniently computed from operating data and used
to assess changes in  aeration system efficiency for a
                                                   81

-------
particular plant. This parameter, termed the efficiency
factor (EF), is expressed as:
where,
        EF = (WODR+W02) [C"m* (CTo, - C)]    (4-6)
         = mass rate  of  oxygen demand  removed,
           ib/hr

         = mass  rate of oxygen supplied, Ib/hr
   C*w   = steady-state DO saturation concentration
           attained  at infinite  time  at  water
           temperature  T  and field atmospheric
           pressure P&
         - fl ft t, C'o.20.
   C     = process water DO concentration, mg/L

The rate of oxygen supplied can be calculated based
on the standardized volumetric airflow rate, qs (4):
(4-7)
                      = 1-036 qs

where qs is expressed in scfm.
       can  b0 calculated  in  various ways and is
typically  expressed  as  the  carbonaceous oxygen
demand  removed  (WCARB) plus nitrogenous oxygen
demand removed
(4-8)
            WOOR  = WCARB
WCARB is usually  approximated  based  on BOD5
removed in the process:
       - 8.34 Q (Influent BODS - Effluent BOD5)    (4-9)
where WCARB is in Ib/hr, BOD5 is in mg/L, and Q is
the waslewater flow rate in mgd.

For a particular plant,  effluent BOD5 may be taken to
include  either total  BODs  or  soluble BOD5.
Alternatively, COD or ultimate BOD can be used.
       can be approximated from the nitrate nitrogen
concentration of the effluent:

WNITR » 8=34 Q (Eff. NO3-N)(4.57 Ib O2/Ib NH3-N) (4-10)

where WNITR is in toflnr ar>d NOa-N is in mg/L.
The nitrogenous  demand removed  can  also be
estimated  by  considering both  nitrification  and
denitrification  (see  Section  5.2.2.2b).  The BOD5
analysis  should  be  performed  using  a  nitrification
inhibitor  io prevent  incorporation  of nitrogenous
demand.
EF is a useful operational parameter and is generally
related to OTEf. However, it should not be interpreted
as a measure of OTEf. EF is based on the mass rates
of oxygen  demand removed  (WQDR)  and  oxygen
supplied (Wos)- Changes in SRT within the normal
range of operating values will not significantly affect
WOOR; however, they will  affect the  rate of oxygen
demand satisfied and the associated Wga- Therefore,
EF is a function of SRT as well as OTEf.

Figure 4-1  shows an EF plot based on actual monthly
average data (5).  The monthly  averages are used to
smooth  daily variations to distinguish  longer-term
trends.

Although EF is only roughly proportional to OTEf, it
should provide a reasonable indication of how OTE| is
changing provided reasonable  care is  exercised in
collecting  the pertinent  data.  The  operating  DO
concentration is  the  greatest source of error.  The
basin average DO concentration should take  into
account spatial and temporal variations. Airflow meters
should be properly calibrated and regularly checked.

A second ratio that can be used to estimate aeration
system efficiency is the oxygen demand removed per
unit  of electrical  power consumed. Since electrical
power consumption is normally measured as the input
to the  motor, this  second  ratio  includes  the
efficiencies of  the  motor and  blower and  air
distribution  system losses.  Once  again, the  mixed
liquor DO concentration correction must be made.

If  a  decreasing trend is observed in  either  EF or
oxygen demand  removed/unit electrical  power
consumed,  the oxygen requirement per  unit mass of
oxygen demand satisfied has increased  or OTEf has
decreased.  Factors  that  can  affect  the  oxygen
requirement per unit mass of BODg satisfied include
the characteristics of  the incoming  wastewater  and
operating parameters such as SRT and F/M loading.
Since  these  factors  and  equipment  operating
parameters can  also  affect OTEf,  more  data  are
required to determine  the cause when  a decreasing
trend is observed. The operating  data should  be
scrutinized  to  see if any  significant  changes  have
occurred in SRT, F/M loading, BOD loading, or degree
of nitrification. If  no  significant  changes are  found,
diffuser fouling is  suspected and the operator should
obtain diffusers for evaluation either from removable
headers, if available, or by draining the basin.

For  the data shown in Figure 4-1, no  significant
changes in either  SRT or F/M loading with time were
noted. Therefore, diffuser fouling with attendant losses
in  OTEf would  be suspected  as  the cause of
decreasing  EF values with  time. These data depict
typical results  where  diffuser fouling over  time
adversely  affects  OTEf  and OTEf  (or  EF) is
subsequently restored  by diffuser cleaning. Methods
                                                 82

-------
Figure 4-1.   Trend charts for aeration system monitoring.
       EF
   '  0.3
     0.2
     0.1
Diffusers
Cleaned
  SRT, days


     4
                                                  Diffusers
                                                  Cleaned'
   F/M, cH
     0.4
     0.2
                                                  Diffusers -
                                                  Cleaned
                                    100
            200

    Time in Service, days
                                                                                         300
                                                                                                                    400
                                                              83

-------
for evaluating diffuser  fouling  and  cleaning  are
presented In the maintenance section of this chapter.

c. Visual Inspection
Visual observation of the system aeration pattern  can
provide  useful information. For  diffused  aeration
systems with a grid layout, the surface pattern should
bo free  of localized turbulence and boiling. Boiling  can
indicate teaks in the  submerged air supply piping,  i.e.,
leaking  gaskets,  faulty  joints,  or  broken
pipes»'diffusers, A significant amount of  energy can be
wasted  by allowing  air  to  leak to  the  atmosphere
rather than being delivered to  the process.  Leaks
should  be repaired as  quickly  as  possible, both
because of the decrease in OTEf that  will occur  due
to poor distribution of air  along with  the release of
coarse bubbles and  the  possibility of further damage
to the diffuser system.

Nonuniformity of the surface pattern can  indicate  that
portions of  the  diffusion system are  becoming
plugged. For  example,  an  unusually  low degree of
surface  turbulence in  one  segment of the aeration
basin may  indicate restriction of airflow to that portion
of the basin  resulting from fouled  diffusers.  Cleaning
of the affected diffusers may be required.

Tho size of the air  bubbles evident on  the aeration
basin surface can also provide an indication of fouling.
Loosely adherent  biomass on fine pore diffu^er media
causes  the formation of  large bubbles. Some degree
of coarse bubbling is normally observed at feed points
along the  aeration basin, and the  degree of coarse
bubbling may vary throughout the day. It is believed
this  phenomenon may  be caused by  surfactants
contained in  the influent wastewater and not fouling.
These surfactant materials  are  quickly adsorbed or
degraded by  the activated sludge,  which  restricts the
size of the  coarse bubble zone.

On the  other hand,  if biofouling  occurs,  the coarse
bubble zone  can  expand until, in the worst  cases,  it
covers the entire  surface of the aeration basin. When
an aeration system  is initially placed  in  service,  but
after equilibrium has  been established (usually several
days), the  surface of  the  aeration basin should be
inspected and photographed to become  familiar with
and to record the size and appearance  of the bubbles
at the  inlet and  outlet  ends of the  basin.  Bubble
patterns should be  photographed at  various  airflow
rates and the flow rate and time of day  when  taken
should  be marked on each photograph.  This
documentation will provide a basis for  recognizing
more extreme coarse bubbling should it occur later.

Onco problems are  identified qualitatively by visual
observation,  quantitative measurements  should  be
made to confirm the type and extent of  fouling and the
type  of diffuser cleaning required.  Experience
indicates that qualitative  observations  can be  a
valuable  tool  when  used  in  conjunction  with
quantitative measures of DWP and OTEf.

d. Compensate Removal
The air piping candensate blowoff system should be
operated on  a set  schedule. A  record of  airflow,
humidity, and mixed  liquor temperature can  provide a
reliable means of calculating  the condensation to be
expected over any period of time. Since the amount of
condensate  depends on the  season and  weather
conditions, plant operating experience  will determine
the frequency needed.  Initially, weekly checks should
be made.  When  removing  condensate,  the liquid
removed should be  observed. The liquid  should be
clear.   If it contains  appreciable  quantities  of solids,
there may be a leak in  the submerged portions of the
air piping. The piping should  be  inspected  and
repaired as soon as practical.

4.2.3.3 Recordkeeping
The following operations data  should be collected and
recorded  on  a regular basis. Daily  collection  and
recording are recommended at least initially.  A change
in the  frequency to more or less often can be made
once operating experience is gained.

«  Weather  conditions.  Temperature,  barometric
   pressure,  humidity,  wind  direction,   and
   precipitation.

•  Aeration basins. Visual observations of the mixed
   liquor, especially the surface pattern and bubble
   size. Look for boils,  and listen  for air leaks in  the
   distribution piping.  OTEf measurements periodically
   if  appropriate  equipment is  available  and  if
   efficiency ratios warrant them.

•  Process operation. Note any changes in  operation
   such as shutdowns, solids wasting rates, unusual
   wastewater characteristics, etc.

•  Blower  operation.  Discharge airflow  rate, pressure
   and  temperature,  time  in  service, oil pressures,
   vibration,  bearing  temperatures,  and  power
   consumption.  The  power drawn should  be
   measured  using a multi-phase  wattmeter.  When
   using multiple blowers,  power factors should  also
   be measured.

»  Air filter conditions. Time in service and differential
   pressure.

*  Air   distribution  system.  Total  airflow rate, airflow
   rate  to each basin and/or grid, and line pressures at
   the  aeration  basins  and other  key points in  the
   system.  DWP should be  monitored if appropriate
   equipment is installed.

»  Condensate blowoffs. Date operated, estimate of
   volume removed, and clarity.
                                                  84

-------
•  Process records.  Wastewater,  return sludge, and
   waste sludge flow rates. Influent and effluent BOD5
   concentrations.  Mixed liquor, return  sludge,  and
   waste sludge  SS  concentrations. Mixed liquor DO
   concentration  at several locations  in the aeration
   basin and mixed liquor temperature. Special events
   such as power  outages  and their duration should
   be documented.

«  Record of cleaning frequencies and methods.

In addition  to  recording  operations  data,  all
maintenance work  should be documented.  Suggested
maintenance  records  are detailed  in  Manual ,of
Practice OM-3 (6).

The  results of all special studies such as wastewater
characterizations, diffuser cleaning tests, and oxygen
transfer tests  should always  be  documented  in a
concise  report and  filed for  future reference.  The
report  should include  the  purpose  of the study,
methods used, results, and conclusions. All  raw data
and calculations should be appended to the report.


4.3  Maintenance
There  are two kinds of maintenance,  preventive and
corrective. Preventive  maintenance is performed  to
keep equipment operating at  an  acceptable level  of
performance,  prolong  equipment life,  and  avoid
emergency situations.  When  equipment  no longer
operates at an acceptable level of performance  or a
breakdown  occurs, corrective  .maintenance  is
performed. This chapter only  addresses  preventive
maintenance  since corrective  action is  usually
equipment specific and is more appropriately covered
by the  equipment manufacturer's literature.

Preventive maintenance is  necessary  to keep a fine
pore aeration  system  in proper  working  order and
maintain an optimum level of performance, as well as
minimize the need  for  emergency  corrective
maintenance  due to  system   failure.  Proper
maintenance  procedures  will also  decrease the
frequency of interruptions in the  air supply  that can
lead to the flow  of solids  into  the  air distribution
system. The  deposition of  solids  on the wastewater
side of the diffusers and subsequent penetration into
the upper pores  will also decrease with a decrease in
air supply interruptions.  Operation at airflow rates per
diffuser  equal to  or  greater than  the minimum
recommended value will help  prevent deposition  of
solids on diffuser media (7).

A major finding of  a  survey   (7)  on  dome  diffuser
plants  in the United Kingdom and Holland was that the
historically excellent O&M performance of these grid
systems was due  to  diligent care   exercised by
knowledgeable treatment plant operators.  Routine
draining, basin and grid washdown,  and  hardware
inspection were  standard operating procedures  at all
plants  surveyed. Operators were also aware of the
symptoms  of  problems in the diffuser system  and
were quick to respond.

4.3.1 Blowers
The  manufacturer's  recommended  maintenance
requirements should  be followed to minimize blower
problems.

4.3.2 Air Systems
Air systems include filtration equipment, air distribution
piping, and airflow measuring  instrumentation.
Filtration  equipment maintenance consists of cleaning
and changing filter   media and  cleaning  of ionizer
elements  in  electrostatic  filtration  units.  The
manufacturer's recommendation  for  maximum
headioss or hours of  operation should be  used to
gauge  when filter  elements  should  be  cleaned or
replaced. Preventive  maintenance on the air filtration
and supply systems can virtually eliminate air-side
dust and particulate fouling of fixed fine pore diffusers.

Air  distribution  piping  normally  requires  very little
maintenance.  Inspection  and  repair  of  protective
coatings  and joint gaskets are  usually all  that are
required. The  entire  system  should  be checked  for
leaks  at  least annually. A significant deterioration in
aeration efficiency can  result  from leaks  that rob air
from the process.

Since  accurate  airflow  and  DO measurements are
essential for monitoring aeration systems, calibration
and/or  zeroing of meters is an important maintenance
task (see Chapter 6).

The  airflow calibration data provided  by  the
manufacturer should  be reviewed after  start-up to
ensure that the specified base conditions for the
flowing air  temperature  and pressure are  appropriate.
If the  actual airflow  conditions  are  other  than the
specified values, the  meter will not read  in the units
shown  on  the  meter,  usually  scfm.  Although the
temperature of  the flowing air will  undergo seasonal
changes,  reasonable accuracy will  be  achieved  by
using the annual average airflow temperature.

4.3,3 Diffusers

4.3.3.1 Cleaning Methods
A variety of fine pore diffuser  cleaning techniques are
available. They can  be broadly classified as process
interruptive or  process  noninterruptive.  Process
interruptive cleaning techniques  require  that  the
aeration  basin  be  taken out of service to provide
access to the diffusers,  while  process noninterruptive
techniques do  not require such  access. A further
distinction  in  cleaning techniques  can  be  made
between  those that  do not require  removal of the
diffusers from the basin (in situ) and those that do (ex
situ). All  ex-situ techniques are process  interruptive,
while   only  some  in-situ  techniques  are  process
                                                  85

-------
fnterruptive. Ex-situ cleaning techniques are generally
not cost  effective  compared with  in-situ techniques
because the time and effort required to remove and
replace the diffusers are substantial.

a. Ceramic Diffusers
Some  of the cleaning procedures  that  have  been
developed, identified, and applied to cleaning ceramic
fine pore diffusers include (8-10):

Ex Situ
•  Reflring
•  Silicate-phosphorus washing
•  Alkaline washing
•  Acid washing
»  Detergent washing
»  High-pressure water jetting

In Situ - Process Interruptive
•  Acid washing
»  Flaming
•  High- and low-pressure water hosing
*  Withholding  influent   (creating  endogenous
   conditions)
*  Sandblasting
•  Chlorine washing
•  Steam cleaning
•  Gasoline washing
»  Drying
•  Ultrasonic

In Situ • Process Noninterruptive
*  Acid injection
•  Air bumping (air turned off and on)

Reflring is an  expensive  cleaning  technique.  It
involves removal  of  the  diffuser  from the  aeration
basin,  placing it in  a  kiln, and  heating it in the same
fashion originally used in its manufacture. The result is
removal of most foulants from,  or incorporated in, the
diffusar element  and restoration  of  the  element  to
essentially its original condition. Other ex-situ methods
are  usually  less  expensive  because costs for
transporting the diffusers to a kiln are eliminated. The
jet washing technique has also  been used for cleaning
ceramic tube  diffusers. Labor  costs are  reduced by
using an automatic  feed/jetting machine (11).

Among the  important in-situ  process  interruptive
cleaning methods  in use  today are  water hosing,
stoam  cleaning, and liquid acid cleaning.  Hosing with
either  high-pressure  [>415 kPa  (60  psia)] or  low-
pressure  water sprays and/or steam  cleaning will
effectively  dislodge loosely  adherent,  wastewater-side
biological growths.  The application of 14-percent HCI
(a 50-percent solution of 18 Baume inhibited muriatic
acid) with a portable spray applicator to each ceramic
diHuser following  hosing or steam cleaning  and then
rehosing the spent acid is effective in removing both
organic and inorganic foulants (12-14),
The  in-situ process noninterruptive acid gas injection
method is  accomplished by  injecting  an aggressive
gas (HCI or formic acid) into the air feed to the fouled
diffusers   (15,16),  Gas cleaning  requires  that
operations personnel understand  the  hazardous
properties  associated with  exposure  to  such gases
and become fully trained in the methods available for
their  safe  use.  With gas  cleaning  systems,  the
cleaning agent is transported to the diffuser by the air
stream where  it  may dislodge  most  foulants.
Specifically, the  gas  injection  procedure includes
increasing  the  airflow  rate per diffuser  to  near the
maximum design value to provide  good distribution
and to get  as many pores operating as possible. The
cleaning  agent can  then be injected into  the air
stream, usually until DWP stops decreasing. With the
HCI  cleaning system, the gas  reacts with the  water
entrained in the diffuser element to  form liquid  HCI.
The  acid, which can reach a concentration of  about
28 percent, dissolves acid soluble deposits. A system
is  normally cleaned  one grid at a  time,  each grid
taking about 30 minutes. This procedure is covered by
a U.S. patent (17).

Acid injection systems are most effective on Type  I
fouling (see Section  3.3.2.1)  involving inorganic acid
soluble foulants, such as iron hydroxides, and calcium
and  magnesium  carbonates.  Acid  injection  for
controlling  Type II fouling (see Section 3,3.2.2) has
been  demonstrated to  be  effective  in a  laboratory
study  (18)  but  not completely  effective  in  full-scale
evaluations  (5,19).  Acid injection will not  remove
atmospheric dust deposited on the  air  side  of the
diffuser or  granular material  such as  silica deposited
on, or incorporated in, Type II foulants adhering to the
wastewater side of the diffuser.

Air bumping of ceramic diffusers is  accomplished by
increasing  the  airflow  rate  per diffuser to a  value
recommended  by the  manufacturer  for  about 15
minutes, then returning to the normal  operating range.
This can  be accomplished  when the blowers  are
rotated by  starting the blower being  placed in service
before the  blower being taken out of service is shut
down.  Caution  must be exercised to avoid incurring
costly electrical demand charges.  Some success has
been  reported  when  air bumping of ceramic  disc
diffusers was combined with HCI gas injection (20).
b. Rigid Porous Plastic Diffusers
Most  of  the  procedures used for  cleaning  ceramic
diffusers have also been applied to rigid porous plastic
diffusers.  The  exceptions are kilning, flaming, and
sandblasting.

c. Perforated  Membrane Diffusers
The  use  of  ex-situ  techniques to  clean perforated
membrane diffusers  is generally not  cost  effective
compared to  in-situ techniques because the  time and
                                                   86

-------
effort required to remove and replace the diffusers are
substantial.

An  in-situ, process  interruptive method  comprising
hosing  with  a 415-kPa  (60-psia) water spray,
scrubbing with a stiff bristled brush, and rehosing was
effective in cleaning membrane  tube  diffusers  in
Green Bay, WI (5).

The air bumping  procedure has been recommended
by  the  manufacturers of  perforated  membrane
diffusers as a method for preventing or minimizing the
effects  of  fouling. The  air bumping  procedure,  also
called flexing, varies  depending on  the  style  of
diffuser. It is usually  accomplished by shutting off the
air to the diffusers allowing them to collapse onto the
frame. The air is  then  reintroduced and  the airflow
rate increased to 2-3 times normal for several minutes
before returning to  the  desired operating  condition.
Under  no  circumstances,  however,  should the
manufacturer's recommended maximum  airflow rate
be  exceeded.  The procedure is performed on a  grid-
by-grid  basis  to  minimize  the  air required.  The
manufacturers' recommended  flexing  frequency  is
normally 1-4  weeks. Documented  improvement  in
diffuser performance  due to  flexing,  either by
decreasing  DWP  or  increasing oxygen demand,  is
limited.  In  one  study (5),  flexing  did  not  remove
loosely  attached  foulant from  the  membranes.
Additional  experience is  necessary to adequately
characterize  the  effects  of  flexing  on  maintaining
oxygen  transfer performance under conditions  that
promote fouling,

4.3.3.2 Cleaning  Method Selection
Since fouling of fine  pore diffusers is site specific and
all the variables that  affect  fouling are not understood,
selection of a cleaning method requires appropriate
experimentation and  analysis. Therefore, the aeration
system  should be  designed to allow  evaluation of the
most likely procedures and frequencies after  initial
start-up. The  effectiveness  of  a diffuser  cleaning
program should be  judged on  its ability to control
increases in DWP  and losses in OTE, as both factors
affect overall aeration efficiency. Because wastewater
characteristics usually  change  with time,  the
preventive  maintenance  program  should  be
reevaluated every few  years  or  as specific  plant
experiences dictate.

The effectiveness of in-situ process noninterruptive
cleaning methods can be determined in several ways.
Full-scale testing can be conducted if more  than one
basin is in  service. The  preventive maintenance
technique can be applied to the diffusers in one basin
while a  second basin is  used as a control, or different
cleaning frequencies can be applied to the diffusers in
two or more basins. The oxygen transfer performance
of  the   two  basins can then  be compared by
monitoring  the air volume used per basin provided the
two  basins  are  operated in  parallel with equal
loadings, SRTs, and DO  concentrations.  Process
water  OTEs of the  two  basins  can  provide the
comparative performance  directly,  but at somewhat
higher cost. The test period should  be long enough to
separate real  long-term  trends  from  short-term
fluctuations. If the fouling rate  is  rapid,  1-2  months
should be  sufficient.  If the fouling  rate is slow,  6
months to a year may be necessary.

Laboratory testing can  be  conducted  whenever
diffusers  are  available from removable headers (i.e.,
those  mounted on  swing arms), removable test
headers,  or  fixed  grid systems  when  a basin  is
drained  (see  Section  2.5).  The  advantage  of
laboratory testing  is that  DWP and OTE can  be
measured under controlled conditions,  allowing the
generation  of more  precise  data  than  is normally
possible  in the field. Disadvantages  are  that only a
small number of diffusers can be economically tested
and  special care must be exercised  in handling and
transporting the diffusers to ensure  that their condition
remains  essentially the same as those  remaining  in
the aeration basin. Laboratory tests recommended for
.the examination of diffusers are DWP, bubble release
vacuum  (BRV) (porous diffusers  only),  OTE,  and
foulant analysis. Air-side  fouling can  and should  be
tested by checking the BRV of the air-side  of several
diffusers.

Small-scale clean  water  OTE  tests  can  also  be
conducted to evaluate the  relative  difference  in OTE
between  two diffuser elements (see Section 2.5.10 for
details of test method). The following procedure can
be used to evaluate fouled diffuser cleaning methods:

*  Obtain a  representative  sample  of new  and used
   diffusers.

•  Characterize the diffusers to determine what effect
   the foulant has had on DWP, BRV,  and  OTE.

»  Analyze the foulant.

•  Evaluate cleaning  methods using DWP,  BRV, and
   OTE as indicators of effectiveness.
A  staged  laboratory testing process, employing
several  fouled  diffusers, can  be used to test the
effectiveness of various  cleaning methods.  In  each
stage, a method is tested, and, if visual observations
indicate the method was effective, the effectiveness is
quantified  by measuring DWP, BRV, and clean water
OTE. The simplest cleaning methods should be tested
first. More involved methods can be  tested  later, if
necessary. Biological slimes can usually be removed
by  mechanical  means such  as  hosing with  a  low-
pressure  [<415  kPa  (60 psia)]  or  high-pressure
[>415 kPa (60 psia)] water spray, steam cleaning,  or
brushing.  Inorganic precipitates firmly attached to the
diffuser  element may yield to chemical treatment.
                                                  87

-------
If  time or process  constraints  will  not allow such a
rigorous  diffuser  testing  program but system
monitoring  indicates that cleaning  is necessary, a
staged approach can  be  applied  to the full-scale
system. A portion  of the grid  can be used to test
various cleaning methods. Cleaning using one or more
of the following steps,  in  the  order  presented, has
been effective in several installations (5,13,14).

*  Hosing to remove loosely attached materials

•  Brushing to loosen residual materials and rehosing

»  Acid treatment  to  remove scale  deposits  and
   rehosing

Between each step,  the  treated  units  should be
evaluated to determine if more cleaning  is needed.
The evaluation  can  take  the  form  of  visual
observations  or  a selected test  to   determine
quantitatively  the  effectiveness  of  the  cleaning
procedure.  On rigid fine  pore  diffusers  such as
ceramics and porous plastics,  the  BRV test can be
performed on in-situ  units.  If  the  initial hosing
effectively removes all the foulant  except scale,  the
second step  can  be skipped. Before treating any
diffuser element with  acid, it should  be  ascertained
that tho materials of construction will not be adversely
affected.

It is clear that every fine pore  installation  will require
some form of diffuser  cleaning on a periodic basis.
The need for, type of,  and frequency of cleaning for
fine pore installations are highly equipment- and site-
specific.  As  such,  no one  approach  can  be
recommended. Manual  of  Practice FD-13 (10)  is a
good reference on experiences with cleaning fine pore
diffusers. Additional diffuser cleaning  data generated
during  the EPA/ASCE Fine Pore Aeration  Project are
summarized in Table 4-1 and Chapter 7.

4.3.3.3 Estimating Cleaning  Frequency
Diffuser cleaning may be accomplished according to a
regular  preventive maintenance  schedule  that
balances the  cost  of  diffuser  cleaning against  the
power  cost  savings  resulting from lower  system
pressure or  higher system OTEf.  The relationship
between  O&M  costs  and cleaning  frequency  is
illustrated in   Figure 4-2.  With infrequent cleaning,
cleaning  costs are   low but  power  costs  are
substantially  higher than  those  associated with a
cleaner diffuser system. As  cleaning frequency  is
increased, power costs decrease but cleaning  costs
increase. Usually, the total of the  power and cleaning
costs will decrease initially as  cleaning frequency is
increased. The  optimum cleaning  frequency occurs
where the total cost reaches a minimum value.

Although Figure 4-2 is an idealized  plot, it presumes,
among other  things,  that  the  fouling rate and its
effects remain constant with  time and that cleaning
can  restore OTEf  and DWP  to  like-new conditions.
The  concept is useful for estimating optimum cleaning
frequency.  The following examples show how fouling
rate  data  can  be  used to determine  the  optimum
cleaning frequency for Type Land Type II fouling.

Chapter 7  (Section 7,3) presents a detailed method
for determining  an  optimal  diffuser cleaning frequency
based on fouling rate, power cost, and cleaning cost.
Section 7.4 gives an  example  using  this  method;
Section 7.5 describes a Lotus spreadsheet  based on
this  method.  The  following examples  are  consistent
with  the method described in Section 7.3.  However,
these examples compare equivalent annual  costs for
various cleaning frequencies,  whereas the  Chapter 7
example compares  present worth  costs.  Other
differences are that the Chapter 7 example includes
initial cost, uses  different power and cleaning cost
data, and allows for a higher fouling rate at the head
end of the aeration basin.
a. Example 4-2 (Type I Fouling)
This example  is  hypothetical  and  is  presented to
illustrate Type  I fouling, which adversely affects DWP
but not OTE.  The operating  conditions  and costs,
which  are similar  to  those  for Example  4-3  for
comparative purposes, are as follows:

Operating  Conditions:

1.  Field atmospheric pressure  = 14.7psia

2.  Blower discharge pressure  = 21,7 psia

3.  Blower inlet air temperature  = 20°C (528°R)

4.  Rate of DWP increase =  1.0 in w.g./mo
                        =  0.036 psia/mo

5.  Combined  blower/motor efficiency = 0.7

6.  Field standardized volumetric airflow rate
    = 16,775 scfm

7.  Power cost = $0.04/kWh

In this case, fouling causes the operating  pressure to
increase,  which  results  in  higher  blower discharge
pressure. The  blower wire power consumption can be
related to  the  discharge pressure using the equation
for adiabatic compression  of air  (4):

WP =AP/e
    = {(wRTa/eK)[(Pd/Pb)K- 1]}/2.655X1Q6    (4-11)
where,

   WP   =  wire power consumption, kW
   AP   =  adiabatic power consumption, kW
   w    =  mass  rate of  air, Ib/hr  [=  0.0750 Ib/cu
          ftx60 min/hrxqs = 4.5 qs1
                                                  88

-------
Table 4-1.    Cleaning Experiences with Fine Pore Diffusers

 Diffuser Type/Plani/Localion            .            Cleaning Procedure
                                               Results and Observations
 Ceramic domes and discs
 Whiltier Narrows Plant
 Los Angeles County, CA
 Ceramic domes and discs
 Nine Springs Plant
 Madison, Wl
 Ceramic disc diffusers (specific
 permeabilities ol 14, 26, 38, and 50
 @ 2-in w.g.; BRV0 of 9, 6, 4, and 3)
 Monroe, Wl
 Ceramic disc diffusers
 Green Bay MSD Plant
 Green Bay, Wl
 .PVC Perforated membrane lube diffusers
 Greers Bay MSD Plant
 Green Bay, Wi
 Ceramic dome diffusers
 Hartford, CT
                                        1. DWP and BRV were less for gassed
                                          diffusers but higher than for new units.
                                          Less effective on domes due to
                                          numerous gasket leaks.
                                        2. DWP and BRV were reduced
                                          substantially, but  like-new conditions
                                          were not achieved. Less effective on
                                          domes, probably  due to gasket leakage
                                          problems.
                                        3. DWP and BRV were reduced
                                          substantially, bul  like-new conditions
                                          were not achieved.
4. Milwaukee Method in a laboratory test.   4. Restored  DWP and BRV on disc
                                          diffusers to near-new conditions late in
                                          the study. Earlier laboratory tests were
                                          not as successful.
1. Acid (HCI) gas injection about every 3
  months during study.
                                        2. Modified Milwaukee Method (low-
                                           pressure hose from basin top, hose
                                           from 6-in range, apply 14-% HCI acid,
                                           soak for 30 minutes,  rehose as above).

                                        3. Hose in a laboratory  test.
1. Milwaukee Method (hose @ 120 psig
  for about 1 minute from 2-ft range while
  feeding air @ 1 scfnVdiffuser, apply 50-
  mL 18° Baume-inhibited muriatic acid
  [14-% HCI], soak for 30 minutes,
  rehose as above).

2. High-pressure hose (120 psig from 2-ft
  range).
1. Milwaukee Method (hose @ 60 psig
  while feeding air @ 3 scfm/diffuser, turn
  off air and apply 50-mL 20° Baume-
  inhibiled muriatic acid diluted 1:1 with
  water, soak for 20-30 minutes, rehose
  as above).

2. Milwaukee Method (same as above).
1. Hose from basin top, fill with service
  water to 3 ft above diffusers, inject HCI
  gas into airflow to dose diffusers @ 0.1
  Ib each, drain basin, rehose as above.

1. High-pressure  hose from basin floor,
  scrub with stiff brush, reliose as above.
1. Milwaukee Method (hose from basin
  top while feeding air @ 1 scfm/diffuser,
  apply 50-mL of «Zep" [22-% HCI plus
  surfactants], soak for 30 minutes, scrub
  with brush as necessary, rehose as
  above).
                                         . Restored DWP and BRV of dome
                                          diffusers to like-new conditions.
                                       2. Reduced BRV of ceramic disc diffusers
                                          from 15 in w.g. to < 10 in w.g., but still
                                          above like-new conditions. DWP
                                          reduced only slightly.

                                       1. Diffusers having various BRV0 values
                                          were obtained from  test headers and
                                          evaluated after 4r 12, and 16 months of
                                          operation. Based on DWP and BRV
                                          measurements, all diffusers were
                                          restored to like-new conditions. No air-
                                          side fouling occurred.
                                       2. Diffusers having BRV0 values of 6 and
                                          4 in w.g. were field cleaned after 24
                                          months in operation. Moderately-fouled
                                          diffusers were restored to like-new
                                          conditions based on DWP, BRV, and
                                          OTE measurements.

                                       1. Restored DWP and BRV to like-new
                                          conditions. OTE restored, to 95% of
                                          new conditions.
                                        1. DWP reduced to less than new
                                          conditions. OTE not restored
                                          (membrane material properties were
                                          changed).

                                        1. Dirty diffusers were severly fouled with
                                          some areas completely plugged.
                                          Cleaning greatly improved airflow
                                          uniformity. BRV was  reduced
                                          substantially,  bul not  to like-new
                                          conditions.
 Ceramic dome diffusers
 Ridgewood, NJ
 Ceramic plate diffusers
 Jones Island East Plant
 Milwaukee, Wl
i. Hose from basin lop or brush with 10-
  % HCI.
1. Milwaukee Method (high-pressure hose
  from basin floor, apply liquid HCI, soak
  for 20-30 minutes, start airflow and
  rehose as above)
                                       1. Individual diffusers were not tested.
                                          Visual inspection of OTE, data
                                          suggested regular cteaniing would
                                          maintain efficiency of the aeration
                                          system.

                                       1. Off-gas testing of full-scale aeration
                                          system indicated improved OTE, after
                                          cleaning.
                                                          89

-------
Figure 4-2.  Idealized plot of optimum cleaning frequency to
           minimize power and cleaning costs.
  Annual Cost
                                    Optimum
                                    Cleaning
                                    Frequency
             Zp =  720 (WP')(nrn){Ep)
(4-14)
                  Cleaning Frequency
  qs    = field standardized  volumetric  airflow  rate,
          scfm
  R    = ideal gas constant  = 53.3 ft-!b/Ib-°R
  Ta    = blower inlet air temperature, °R
  P(j    = blower discharge pressure, psia
  PD    = field atmospheric pressure, psia
  K    = (k - 1)/k = 0.283 for air
  k    = ratio of specific heats for air, Cp/Cv =1.4
  o    = combined blower/motor efficiency, fraction
  and 2.655X106 js in units of ft-Ib/hr-kW.

Assuming DWP  increases at  a constant  rate, the
average power consumption for operating the blowers
between cleaning events is given by:
                 '  = (WP0 + WP,)/2        (4-12)

          WP' = WP0 (1  + WP1/WP0)/2     (4-13)

whore,

  WP'   = average wire power consumption, kW

  WPr>  = wire power consumption  for the aeration
           system when  operating  with  clean
           diffusers, kW

  WPt   = wire power consumption  for the aeration
           system  when  operating  with  fouled
           dttfusers just before cleaning, kW

  and WPt/WP0 is  termed the power ratio.

The  cost of power for an  operating period of any
length in months is:
                                                      where,
  Zp = period power cost, $/operating period
  nm = number of months between cleanings
  Ep = unit power cost, $/kWh
  720 — conversion factor, hr/month

The period power cost can be put on an annual basis:

                EACp = 12Zp/nrn           (4-15)

where,

  EACp = equivalent annual power cost, $/yr
  12 = conversion factor, months/yr

  Likewise,  the period diffuser cleaning cost can also
  be put on an annual basis:
                                                                      EAC0  - 12 Z0/n
                                                                                    rn
                                           (4-16)
                                                      where,
   EACp  = equivalent annual diffuser cleaning cost,
           $/yr
   Z0  = period diffuser  cleaning  cost,  $/operating
        period

Figure  4-3 illustrates how wire power  in  terms  of
power  ratio  (WP-i/WPo)  changes  with  time  for
operating  periods of 1,  2, 3, 4, 6, and  12 months.
Table 4-2  lists annualized costs  as a  function  of
operating  period length  and cleaning frequency.  For
the conditions  listed  above, the  optimum time  of
operation  between  cleaning  of the  diffusers is  9
months. However, the  annualized  total  cost curve
near the optimum point is relatively flat and extending
the operating period to 12 months  only increases  the
cost by $227.

b. Example 4-3 (Type II  Fouling)
This example is based on fouling rate data collected
at the Green  Bay  wastewater treatment  facility
between May 1986 and  October 1987 (5). This plant
is comprised of four pairs of contact and return sludge
reaeration  basins (Figure 4-4).  Quadrant  2  was
retrofitted with ceramic disc fine pore diffusers in early
1986. The contact basin was  outfitted  with 6,128
diffusers,  and the reaeration basin received 2,148
diffusers.  The quadrant  treated  approximately  1.5
million Ib BODs/month during the test period.

Fouling  at the  Green Bay plant  was  severe  and
resulted in relatively rapid and  substantial losses in
OTEf  without significant changes  in  DWP  (Type  II
fouling). The cost of power to operate the system with
clean  diffusers was about $9/1,000 Ib BOD5  treated,
or $13,500/month to treat 1.5 million Ib BOD5.
                                                   90

-------
Figure 4-3.   Power  ratio vs. time  for  ADWP=+2.5  in
            w.gVmonth.
 Power Ratio (WP,/WP0)

   1 ~g   a. Operating Period = 12 months (1 cleaning/yr)

   1.04

   1.02

   1.00
                                                 Table 4-2.    Annualized Costs  for  a  Fine  Pore Aeration
                                                             System Experiencing a  DWP Increase of 1.0 in
                                                             w.gVmonth (see Example 4-2)
   1.06

   1.04

   1.02

   1.00
   1.06

   1.04

   1.02

   1.00
   1.06

   1.04

   1.02

   1.00
   1.06

   1.04

   1.02

   1.00
   1.06

   1.04

   1.02

   1.00
    b. Operating Period = 10 months (1.2 cloanings/yr)
    c. Operating Period = 8 months (1.5 cloanings/yr)
   d. Operating Period = 6 months (2 cleanings/yr)
    e. Operating Period = 4 months (3 cleanings/yr)
r—  f. Operating Period  = 2 months (6 cleanings/yr)
Operating
Period1,
months
1
2
3
4
5
6
7
8
9
10 '
11
12
Cleaning
Frequency2, ,
times/yr
12.0
6.0
4.0
3.0
2.4
2.0
1.7
1.5
1.3
1.2
1.1
1.0
Annualized Cost,
Power3
162,370
162,730
163,090
1 63,460
163,280
164,190
164,550
164,920
165,280
165,650
165,010
166,370
Cleaning4
31,200
15,600
1 0,400
7,800
6,240
5,200
4,460
3,900
3,470
3,120
2,840
2,600
$/yr
Total
193,570
178,330
1 73,490
171,260
170,060
169,390
169,010
168,820
168,750
168,770
168,850
"168,970
                     Time, months
                                               12
                                                          1 Time of operation between cleanings.
                                                          2 Cleaning frequency = 12 * operating period.
                                                          3 Based on power cost = $0.04/kWh.
                                                          4 Based on cleaning cost =  $2,600/cleaning.
Figure 4-5 shows EF (see Equation 4-6) as a function
of time in service for the Green  Bay  ceramic  disc
diffusers.  Since  EF is roughly proportional  to OTEf,
the rate of decrease  in  EF can  be  interpreted as a
fouling rate. Over the first 2 months  of record  shown
in Figure  4-5, EF decreased by about 15 percent per
month. This is indicative of an extremely high fouling
rate  (see Tables cj-8 and 3-9). However, no further
loss  in EF was  observed after  the first 2  months.
Laboratory testing showed that a  three-step  cleaning
procedure of hosing  the diffusers  from the basin
walkway,  followed by  treatment  with  HCI  gas  and
rehosing was effective in restoring OTEf to a near-new
condition. Approximately 0.1  Ib HCI/diffuser was used.
The   diffusers   were  cleaned  by  the  three-step
procedure after being in service  for  6  months. OTEf
then decreased at a relatively constant rate of about 5
percent/month.

The  associated power costs are depicted in  Figure 4-
6. The cost of electrical power over the course of the
study (May 1986 through October  1987) was  constant
at $0.04/kWh.

The  costs  associated  with   cleaning  the  diffusers
include  the  labor  for  hosing  the diffusers  and
performing the acid  gas treatment and the  HCI gas
itself. Hosing of the  6,128 contact basin diffusers  took
two  operators about 8 hours  (5-10 sec/diffuser)  - 16
labor-hr.  The  reaeration basin  diffusers could be
hosed off by two operators in about 4 hours - 8 labor-
hr. The labor required for the acid gas cleaning  step
was  about 2 hr/grid; there are  10 grids in the contact
basin and 6 in the reaeration basin -  32 labor-hr.  At a
labor cost of  $30/hr, the labor cost for cleaning  was
                                                      91

-------
Flguro 4-4.   Schematic of Green Bay wastewater treatment plant.
         SecoiKlary Clanfiors
       Aeration Basin Complex
              RAS
Primary Clarifiers
Reaeration
Quadrant 2
Contact
ML,M V

1
J
' 1
I 1
r
Reaeration
Quadrant i
Contact
Li ML
1 M (J
  Secondary
  Ellluctit
                                      ML
                                                                  Decant
                                                                           ML
— .0 .
Contact
Quadranl 4
Reaeration
*•
1
i

1
i
L|J
^-_

i o«—
Contact
Quadrant 3
Reaeralion
                                                                    Metro
                                                                   Waste
                                                                   Mill
                                                                   Waste
     Legend:
     RAS • Return Activated Sludge
     ML • Mixed Liquor
     Decani - Dacam Liquor from Sludge
            Hoal Treatment Process
                                          RAS
Flguro 4-5.   EF vs. time In service for ceramic disc diffusers - Green Bay, Wf.
     EF
 0.3
 0,2
 0,1
                May 1986
                                 Based on BOD5 loading and airflow rate to the contact
                                 basins and correcting the airflow rate to zero DO.
                   100
200            300

         Time in Service, days
                                                                 400
                                                                                500
                                                             600
S1|680. The chemical cost of acid gas cleaning was
8,276 diffusers x 0.1 Ib  HCI/diffuser x $2.00/lb  HCi =
S1.655,  The  total cost was  about  $3,300,  or
approximately $0.40/diffuser.
It should be emphasized that the  study  performed at
Green Bay was designed to compare  two different
diffuser systems. During the study, every effort was
made to maintain performance, perhaps to the point
where cleaning  costs  may not represent  the more
average case that would occur at another plant.
                     Figure 4-7 illustrates  the  relationship between F and
                     time for the case of  severe  fouling where  OTEf was
                     reduced by 15 percent/month to a  maximum  loss of
                     30 percent (F  = 0.70), similar to the first 2  months at
                     Green  Bay.  Fouling  factors  for   six  cleaning
                     frequencies between  1 and  12 times/yr are plotted.
                     Annual average F values range from 0.725  for annual
                     cleaning to 0.925 for monthly cleaning.

                     The  annual cost of electrical power is equal to the
                     clean diffuser power cost ($162,000/yr)  divided by the
                     annual average fouling factor. Figure 4-8 plots annual
                                                     92

-------
Figure 4-6.   Power cost vs. time in service for ceramic disc diffusers - Green Bay, Wl.
 Power Cost, $/i ,000 Ib BOD5 treated

  20
  15
          ,	  May 1986
    Plant
    Upset
                               Diffusers
                               Cleaned •
                                                                  Diffusers
                                                                  Cleaned
                                                    I
                    100
                                    200             300 .

                                            Time in Service, days
                                                                  400
                                                                                  500
                                                                                                 600
 Figure 4-7.   Effects of cleaning frequency on annual average F (fF = 15 percent/month).

    F                                                     F
  1.0

  0.8

  0.6
         a. Operating Period = 12 months (1 cieaning/yr)
           Annual Average F = 0.725
  1.0

  0.8

  0.6
        d. Operating Period = 3 months (4 cleanings/yr)
           Annual Average F = 0.8
   1.0

   0.8

   0.6
         b.  Operating Period .= 6 months (2 cleanings/yr)
            Annual Average F = 0.75
                                                          1.0
  0.6
        e. Operating Period = 2 months (6 cleanings/yr)
           Annual Average ,F = 0.85
   1.0

   0.8

   0.6
         c.  Operating Period = 4 months (3 cieanings/yr)
            Annual Average F = 0.775
  1.0

  0.8

  0.6
                                                                 f.  Operating Period =  1 month (12 cleanings/yr)
                                                                   Annual Average F = 0.925
                    Time, months
                                              12
O&M cost (electrical power plus cleaning) and annual
average F as  a function of cleaning frequency. For
this case of severe fouling, cleaning 6 times/yr {every
2 months) would have been cost effective.

For  comparison,  the cost trade-off  analyses  for  a
moderate  and a  low fouling  rate  (both hypothetical
linear  rates - the  higher of which is similar to that
observed during a portion of the test period at Green
Bay)  are  shown  in Figure  4-9.  For the moderate
fouling  rate of  5 percent  loss  in  OTEf/month  (to  a
maximum  loss, of 30 percent), the optimum cleaning
frequency would be 4  times/yr (every 3 months) but
very little extra cost would be incurred by decreasing
the frequency to 3 times/yr (every 4 months). For the
low fouling rate of  1  percent loss in OTEf per month,
the optimum  cleaning frequency  would  be  1 to  2
times/yr. These results are comparable to  those of the
example presented in Section 7.4

These  two  examples  demonstrate the merit  of
investigating  the  fouling  tendencies  of installed
diffusers. In the first  example, Type I fouling  affected
DWP  but  not  OTE  and   the effect   on   power
consumption was small and could be controlled with a
minima!  cleaning frequency.  In the second example,
Type  II  fouling  significantly  affected  OTE and
considerable  power  savings  were  achieved  with
frequent cleaning under  severely fouling conditions.  A
cost-effective  analysis showed that moderate to  low
                                                     93

-------
  Figure 4-8.   Cost tradeoff  analysis  for determining
                cleaning  frequency  based on  fouling
                patterns shown in Figure 4-7.
  Annual Cost, $1,000

    240  i-
    160
     80
                                             Total
                                   Power
                                            Cleaning
                                           Figure 4-9.   Cost tradeoff  analysis for  fF  =   1
                                                        percent and 5 percent/month.
                                         Annual Cost, $1,000

                                           240
                                                                         160
                                                                          80
                                                                                                               Tolal
                                                                                                        Power
                                                                                Cleaning
  Annual Average F

     1.0
     0,7
                        J_
  JL
           New Diffuser
I
                  2      4       6       8      10

                    Cleaning Frequency, times/yr
                                                       12
                             Annual Average F

                               1.0
0.9


0.8
                                                       1        1
	5 percent/month

         1 percent/month

I        I	I        I
                                                                                    2       4       6       8       10

                                                                                       Cleaning Frequency, times/yr
                                                                                                                            12
Table 4-3.    Fine Pore Aeration System Troubleshooting Guide
           Problem                         Probable Cause
                                                                   Action
 Blower discharge pressure     Fouled dilfusers
 increase at constant volume
                             Automatic flow control valve not operating
                             properly

                             Clogged air filler downstream of blower
                                           Evaluate condition of diffusers (see Section 4.2.3.2); clean
                                           diffusers It necessary

                                           Check operation of automatic valves in the main air
                                           distribution system

                                           Check pressure drop across filter medium; clean or replace
                                           filter elements per manufacturer's recommendation
 Air usago increase but BOD  Residual DO may be increasing
 loading remains constant
                             Air leaking from distribution system

                             Change in air requirement/mass BOD treated

                             Fouled dillusors
                                           Check calibration of DO probes

                                           Check air distribution piping for leaks; repair as necessary

                                           System may be running at longer SRT; reevaluate operating
                                           parameters; adjust sludge wasting rates if appropriate

                                           Evaluate condition of diffusers (see Section 4.2.3.2); clean
                                           diffusers if necessary
 Continuous large bubbles or  Diffusers partially clogged
 clumps of bubbles
                             Increase in surfactant concentration
                                           Inspect and clean dtffusers

                                           Evaluate oxygen transfer capacity; adjust basin loadings if
                                           necessary
 Dead spots
Not enough airflow


Diffusers completely clogged


High airitow/diffuser
                                Check that airflow is above manufacturer's minimum; check
                                calibration of airflow meters

                                Evaluate condition of diffusers (see Section 4.2.3.2); clean
                                diffusers if necessary

                                Decrease airflow or install additional diffusers
 Continuous boils
Broken diffuser or air distribution pipe
                                Dram basin; inspect diffusers and piping; make repairs as
                                necessary
                                                               94

-------
levels of fouling  could  be controlled  with  a much
reduced cleaning frequency.

4,3.4 Troubleshooting
 Some  of  the  more  common  operating problems,
 causes of trouble, and  corrective actions associated
 with  fine  pore  aeration systems  are  tabulated in
 Table 4-3. Equipment-specific  mechanical problems
 are not included because they  should be covered in
 the manufacturer's literature.
4.4 References

When an  NTIS number is cited in a reference, that
reference is available from:

 National Technical Information Service
 5285 Port Royal Road
 Springfield, VA 22161
 (703) 487-4650   '

1.   Redmon,  D.T., Boyle, W.C. and L. Ewing. Oxygen
    Transfer Efficiency Measurements in Mixed Liquor
    Using  Off-Gas Techniques. JWPCF 55{11):1338-
    1347,  1983.

2,   Mueller,  J.A. and  H.D.  Stensel. Biologically
    Enhanced Oxygen  Transfer  in the Activated
    Sludge Process - Fact or  Folly? Presented at the
    60th Annual Conference  of the Water  Pollution
    Control  Federation,  Philadelphia,  PA,  October
    1987.

3.   Campbell, H.J, Jr. Oxygen Transfer Testing Under
    Process  Conditions.   In:  Proceedings of
    Seminar/Workshop  on Aeration  System  Design,
    Testing,  Operation,  and  Control. EPA-60Q/9-85-
    005, NTIS No.  PB85-173896, U.S.  Environmental
    Protection Agency, Cincinnati, OH, January 1985.

4.   American Society of Civil  Engineers.  ASCE
    Standard:  Measurement of. Oxygen  Transfer in
    Clean  Water. ISBN 0-87262-430-7, New York, NY,
    July 1984.   .                           .

5.   Donohue  & Assoc., Inc. Fine Pore  Diffuser
    System Evaluation for the Green Bay Metropolitan
    Sewerage  District.  Study  conducted  under
    Cooperative Agreement  CR812167,  Risk
    Reduction  Engineering  Laboratory,  U.S.
    Environmental Protection Agency, Cincinnati, OH
    (to be  published).

6.   Plant  Maintenance Program. Manual of Practice
    OM-3, Water   Pollution  Control  Federation,
    Washington, DC, 1982.

7.   Houck,   D.H.  and  A.G.  Boon.  Survey  and
    Evaluation of Fine  Bubble  Dome Diffuser Aeration
    Equipment. EPA-600/2-81-222,  NTIS No.  PB82-
    105578,  U.S. Environmental Protection  Agency,
    Cincinnati, OH, September 1981.
8.  Air Diffusion in Sewage Works. Manual of Practice
   5, Federation  of  Sewage  and Industrial Wastes
   Associations, Champaign, IL, 1952.

9.  Aeration  in  Wastewater  Treatment.  Manual  of
   Practice  5,  Water Pollution  Control Federation,
   Washington, DC, 1971.

10. Aeration. Manual of  Practice  FD-13, Water
   Pollution  Control  Federation, Washington,  DC,
   1988.

11, Schreiber  Corp., Inc.  15  Years  of  Diffusur
   Maintenance  History with Schreiber  Counter-
   Current Aeration in West Germany, October 1984.

12. Rieth,  M.G., W.C. Boyle and L.  Ewing.  Effects of
   Selected Design  Parameters on .the Fouling of
   Ceramic  Diffusers. Presented at the 61st Annual
   Conference  of the  Water  Pollution  Control
   Federation, Dallas, TX, October 1988.

13. Yunt, F.W. Some Cleaning Techniques for  Fine
   Bubble Dome and Disc Aeration Systems. Internal
   report, Los  Angeles  County Sanitation Districts,
   Whittier, CA, October 1984.

14. Wu,  C. and J.A. Kellner. Restoration of East Plant
   Aeration  Tanks No.  5, 11  and  18 at  the Jones
   Island Wastewater  Treatment Plant.  Internal
   report, Milwaukee Metropolitan Sewerage District,
   Milwaukee, Wl, December 1980.

15. in-Place  Cleaning System  for Ceramic  Grid  Fine
   Bubble Aeration. Product  Bulletin' GC  1-83,
   Sanitaire - Water  Pollution  Control  Corp;,
   Milwaukee, Wl, 1983.

16. Bretscher, U.  and W.H, Hagar.  Fne Cleaning of
   Wastewater Aerators. Water and  Wastewater, Vol.
  • 6, 1983.

17. Sanitaire - Water - Pollution Control Corp.  U.S.
   Patent No. 4,382,867, May  10, 1983.

18. Wren,  J.D. Transcript of Biofouling Seminar.  New
   York  Water Pollution  Control  Federation,  New
   York, NY, January 1985.

19. Stenstrom, M.K. Fine  Pore Diffuser Fouling: The
   Los  Angeles  Studies.  Study conducted under
   Cooperative   Agreement  CR812167,  Risk
   Reduction  Engineering  Laboratory,  U.S.
   Environmental  Protection Agency,  Cincinnati, OH
   (to be published).

20. Vik, T.A., D.J.  Lamars and D.L. Roder. Full-Scale
   Operating  Experience Utilizing  Fine Bubble
   Ceramic  Aeration with  In-Place  Gas Cleaning at
   Seymour, Wisconsin.  Presented  at  the  57th
   Annual Conference of the Water Pollution Control
   Federation, New Orleans, LA, October 1984.
                                                 95

-------

-------
                                              Chapter S
                                  System Design and Installation
5.1 Introduction
A  typical  fine  pore  aeration  system  is  illustrated
schematically in  Figure 5-1. The air supply system
consists of air  filters, blowers,  air piping, and airflow
control  equipment  (flow meters and  flow control
valves). The  diffusion system consists of a series of
headers  and lateral  piping  placed  in  the aeration
basin, to which fine pore diffusers are attached. The
system is usually arranged as a series of "grids" (or
zones) to  allow  for  proper  airflow  distribution and
accommodate oxygen transfer rate  variations  within
the aeration basin.

This chapter presents  the procedures  required  to
design and install a fine  pore  aeration system. The
chapter is  divided  into  five major sections. Process
and operation and maintenance (O&M) considerations
affecting  aeration  basin and aeration system  design
are addressed  in  Section  5.2. The  determination of
process oxygen and mixing requirements is discussed
in  Section 5.3. Procedures  for  estimating process
oxygen  requirements, including  both  average  values
and their  temporal and  spatial  variations, are
presented.  Major  design  and  installation considera-
tions related  to the in-basin equipment are discussed
in  Section 5.4. Diffuser  selection,  system layout,
airflow distribution, diffuser cleaning and maintenance,
system installation, contract specifications, and retrofit
considerations  are discussed.  The  remaining  com-
ponents of the  system are considered in Section 5.5.
These include  air  filters,  blowers,  and  air piping.
Again, installation and  retrofit  considerations  are
discussed. Section  5.6  provides a  consistent and
logical approach  to the  integrated design of a fine
pore aeration system.

Extensive  use  of examples  is  made throughout
Sections 5.3, 5.4, and 5.5. These examples are not
intended  to  provide a  "cookbook"  set of design
procedures that can be followed  in every instance to
arrive at  the  "right" answer.  While component  sizing
of  a fine pore aeration system is based on a specific
set of calculations, many assumptions must be  made
and  form the  basis  for  those  calculations.  These
assumptions  are  dependent on  the specific situation
and  the  knowledge and  experience  of the system
designer. The  examples illustrate the  types  of
calculations required to  design  a fine pore aeration
system. More importantly, they  also illustrate  the
interplay  of  experience-based  assumptions  and
specific design calculations during  the design process.


5.2 Process and O&M Considerations
Several process-  and  O&M-related  considerations
affect  fine pore aeration  system  design  and  basin
configuration. These  considerations include waste-
water  characteristics  and organic  loading, nutrient
control  requirements, hydraulic flow  regime,  mixing
requirements, diffuser  fouling and  cleaning,  and
personnel  qualifications,  among others. They must be
addressed in the overall  design concept of the system
before initiating detailed  design of oxygen dissolution
and supply components.

5.2.1 Process-Related Considerations
Process-related considerations  that affect  fine pore
aeration system design include the type (or degree) of
wastewater treatment, hydraulic flow  pattern desired
for the biological  reactor, and degree of  flexibility
required for  the  aeration system.  Four  biological
reactors configured  to  provide different degrees of
wastewater treatment and satisfy different effluent
requirements are illustrated in Figure 5-2.

Biological  treatment  is  used  primarily to remove
organics {usually  quantified  as  5-day biochemical
oxygen demand [BOD5]) from  wastewater, as typified
by conventional  secondary treatment  (Figure 5-2a).
Fine  pore aeration systems   have  been  used
successfully in biological treatment systems  designed
to meet this objective. Their primary advantage in  this
application is  their  relatively low energy requirements
compared  with  other  oxygen  transfer systems  avail-
able,  which can result  in significant  operating cost
advantages. One  disadvantage in this application is
the potential for diffuser fouling in certain wastewaters,
such  as industrial  wastes containing high  strength,
readily biodegradable materials or fine colloidal solids.

As discussed  in Chapter 3, the designer should select
conservative  process  water  oxygen  transfer
efficiencies (OTEs) for  fine pore diffusers  used in
secondary  treatment systems  with  high  organic
loading rates due to the  adverse effects of these high
loadings on OTE. It can  be difficult to physically place
                                                  97

-------
Figure 5-1.   Schematic of a fine pore aeration system.
                     Diffuser
                     (Typical)'
                            Aeration Grid
                              (Typical)
                                            Aeration Tank/Basin



Influent





\















































































(
C
C
(
C
(

)
)
)
>
)
b
A

Laterals




I Header j


























,


























r






(

































Effluent
ftfe,




                                             Flow
                                            Control
                                             Valve


ftirflov
Meter


' f
4 II!
* in
To Other



Outlet
Air
Filters
As
Required



Blowers




Inlet
Air
Rlters

Air
In

                           Tanks/Basins
all  the  diffusers required  to  meet the  volumetric
oxygen  demands occurring within the influent end of
plug flow basins, where low process water OTEs  are
anticipated. This may limit the sizing and configuration
of the basin.

Ammonia and organic nitrogen may also be removed
from wastewater through their  conversion to  nitrate
nitrogen via the process of nitrification {Figure 5-2b).
Fine pore  aeration  systems have  also  been  used
successfully in this application.  Relatively  long solids
retention limes (SRTs)  are  generally required  to
achieve reliable and consistent nitrification. Reference
1  describes  the  process conditions  required  to
achieve  reliable nitrification and can  be used  as a
general   design  resource.  Because  of  their longer
SRTs,  process  organic  loadings for nitrification
systems will be relatively low and OTEs will be higher
than  those observed in   conventional  secondary
treatment  systems  (as discussed in Chapter  3).
Therefore,  the "energy  cost  advantage of fine pore
aeration  is often greater in  nitrification applications
than for conventional secondary treatment alone.

Nitrogen removal  may be  accomplished  by  the
addition of an anoxic zone to the head end of a single-
stage nitrification system (Figure 5-2c).  An  anoxic
zone is a section of the biological reactor that is mixed
but not aerated and  within which nitrate nitrogen  (not
molecular oxygen)  serves'as  the  principal terminal
electron acceptor. Fine pore aeration  systems have
been used  quite successfully in the aerobic sections
of biological  nitrogen  removal  facilities. In  fact,
significant operational advantages are  possible  with
the incorporation of anbxic'  zones  into  nitrification
systems using  fine  pore  aeration  (2).   These
advantages  include reduced  energy  requirements
resulting from the use of recycled nitrate as a terminal
                                                   98

-------
Figure 5-2.  Representative (commonly-used) biological wastewater treatment systems using fine pore aeration.
                       Influent
                               RAS
                                                                              To Clarifier
                               a.   Secondary Treatment (BOD Removal) (Design
                                   configuration can be Plug Flow, Step Feed, Complete
                                   Mix, Contact Stabilization, etc.)              !
           Influent
                 RAS
r
                                                                        To Clarifier
                                  b. Nitrification (BOD Removal and TKN Oxidation)
   Influent         _^



          RAS
   Anoxic
      o»o
         Zone
                                                                                                   To Clarifier
                                 c.   Single-Stage Nitrification with Anoxic Zone (BOD Removal,
                                     TKN Oxidation, and Partial Nitrogen Removal)
                  Influent
    RAS
                                     Anaerobic Zone
                                        Jo
                                                                                      To Clarifier
                                  d.  Biological Phosphorus Removal (BOD Removal
                                     and Enhanced Phosphorus Removal)
           RAS  = Return Activated Sludge
electron acceptor and improved aF values due to  the
metabolism of some organic  matter in the anoxic
zone. In addition, anoxic zones can effectively control
some types of sludge bulking (3).

Phosphorus can  also  be removed  from  wastewater
using a  variety  of  biological phosphorus  removal
processes. One option, illustrated in Figure 5-2d, uses
anaerobic  and  aerobic  sequencing  (4).  Biological
phosphorus removal is discussed in  detail  elsewhere
                                     (5). Considerations  for applying fine pore aeration to
                                     biological phosphorus removal are similar to those for
                                     conventional secondary treatment.  The  presence of
                                     the upstream anaerobic zone may  lead  to improved
                                     aF  values,  similar  to  the  effect seen  with anoxic
                                     zones;  however, data  confirming  this hypothesis are
                                     not currently available.

                                     The hydraulic flow pattern produced  within the aerobic
                                     section of a biological reactor must be compatible with
                                                      99

-------
the design  of the aeration system.  Hydraulic  flow
patterns  can range  from the  extreme cases  of
complete  mix  to  plug  flow,  with  intermediate
conditions characterized as the equivalent number of
basins-in-series  for  the reactor  (6).  During the late
1960s and early 1970s, the trend was toward the use
of complete mix  aerobic  reactors.  This  was  due
largely to the greater potential for  such systems  to
handle toxics  and other materials that require dilution
of the  incoming  wastewater.  However, experience
Indicated  that  complete mix  designs  can lead  to
excessive growths  of  filamentous  organisms  that
adversely  affect  the thickening characteristics  of
activated sludge mixed liquor (7).

Plug flow systems, on the other hand, discourage the
growth of filamentous organisms and encourage the
growth  of desirable floe-forming bacteria.  Plug  flow
conditions also  improve the nitrification efficiency  of
single sludge systems (1,8). As discussed in  Chapters
3 and 4,  plug flow conditions  affect the performance
characteristics  of  fine  pore diffusers. a values  vary
through  a plug flow  reactor, being  lowest at the
Influent end  and  highest at the effluent end  of the
reactor.

Diffuser fouling  rates also vary in step feed  systems,
usually being more  significant at the  influent end  of
the reactor.  Switching to a  step  feed flow regime
during periods  of high loading can minimize, this
variability by spreading out the organic load over the
full basin length.  Since coarse  bubble  diffusers are
less susceptible to  fouling than fine pore  systems,
another alternative is to use a hybrid aeration system
consisting of  coarse bubble diffusers at the influent
end and fine bubble diffusers elsewhere in the reactor.
Tho impact of variations in a and fouling factor (F) on
fine  pore  aeration system design  is  illustrated  in
Section 5.4.8.

The range of  volumetric oxygen transfer rates  that
must  be accommodated also affects aeration system
evaluation and selection. For many fine pore  systems,
Iho airflow rate per diffuser must be  maintained within
a given range. Ideally, the lower limit  is the rate that
still produces uniform air distribution across the entire
dtfluser surface. Failure to maintain this rate results in
an increased potential for diffuser fouling since foulant
deposits will accumulate on the unused portion of the
diffuser  surface.  Continued  accumulation  will
eventually decrease diffuser performance (9,10).

Upper limit airflow rates per  diffuser are those  that
cause diminishing improvements in  oxygen  transfer
rate. To illustrate the principle  of constraint on airflow
range, consider typical  ceramic  diffusers. For these
diffusers,  the allowable ratio of maximum to minimum
airflow rates  is about  5:1,  resulting  in  a  ratio  of
maximum to  minimum oxygen transfer rates of about
4:1; see Chapters 2 and 3 (11).  If a greater range  of
volumetric oxygen  transfer  rates  is  required,  the
designer should consider another diffuser type. Mixing
requirements should also be checked to confirm that
there is adequate  energy  input for mixing  at the
minimum  airflow  rate required to  satisfy the lowest
oxygen  transfer needs. Airflow and volumetric oxygen
transfer ranges vary w... sine pore diffuser type. As a
result, process requirements may  sometimes dictate
the type of fine pore diffuser that can be used.

5.2.2 O&M-Related Considerations
O&M-related  considerations that affect  the  selection
and design of a fine pore aeration system include the
degree  of treatment provided ahead of  the  biological
system, nature  of  the wastewater,  overall   O&.M
concept of the facility, physical design characteristics
of  the  system, and  level  of expertise of  the  plant
operating staff.

As  discussed in Chapter 3,  fouling rates for fine pore
aeration systems  may be more rapid in plants with no
primary treatment or with poor grit removal prior to the
biological  reactor. It appears that coarse solids not
removed in  such  systems  contribute to accelerated
fouling  of fine pore diffusers. This  effect has  been
observed  with  both porous  ceramic and perforated
membrane fine pore  diffusers (10).  As  a result, fine
pore diffuser cleaning might  need to be more frequent
in  systems   incorporating  little or  no preliminary
treatment of  the wastewater than  in systems  with
better pretreatment. If  this is  not  acceptable,  then
another oxygen transfer system should be considered.
If  fine  pore diffusers  are  used  under  these
circumstances,  the  design of  the system   must
accommodate  the potential  for  frequent diffuser
maintenance.

In  most cases,  regardless  of  the  degree of  prior
treatment provided, periodic drainage of the  biological
reactor  will  be required  to inspect, clean,  repair,
and/or replace diffusers. Multiple basins designed to
facilitate easy drainage and cleaning are  therefore
required. Sufficient treatment capacity must also be
available to allow periodic drainage of individual basins
without  adversely  affecting system performance.

Also,  as discussed in Chapter 3, wastewater type can
significantly affect fine pore  diffuser fouling rates and
required cleaning.  The presence of  high-strength,
readily-biodegradable  wastewaters  can  significantly
increase diffuser  fouling rates caused  by  biological
growths  on  the diffusers.  Fine colloidal  solids  can
accumulate on the  diffusers, again  contributing to
increased fouling  rates.  The  presence  of a  high
concentration  of  readily settleable  solids may also
interfere with  fine  pore  aeration  system  cleaning
operations. Unfavorable water chemistry can result in
the formation of precipitates within the diffusers. This
effect  can  be  accelerated by  the  addition  of
wastewater treatment chemicals such as ferrous or
ferric iron used  for  phosphorus removal.  Cleaning
methods, which should  be evaluated and  selected in
                                                   100

-------
the facility design phase, will be dictated by the nature
of  the  foulant  (refer  to Chapter  4  for more
information).

Experience  suggests  that  some  fine  pore  aeration
systems are more fragile than  other aeration systems
now available (12). Many of the problems encountered
with early fine pore systems can be avoided by careful
attention to physical design details, material selection,
and equipment  specifications.  Necessary steps
include proper design of the header and lateral system
for expansion/contraction,  proper  manufacturing
techniques,  use of stainless steel for piping supports,
and proper  selection of header  and  lateral piping
material.

Because the potential for fouling of fine pore diffusers
is greater than  for other  systems, such as coarse
bubble aeration, a higher degree of attention  may be
required to operate fine pore diffusion systems within
adequate  control ranges  and  provide necessary
maintenance. The expertise and motivation of existing
plant personnel must be assessed when selecting fine
pore  aeration  system.   In  some cases,  the
qualifications  of plant  personnel   may   not  be
appropriate  for installation  of  a fine pore system. If
possible, the designer should discuss  the advantages
and disadvantages of using  a fine   pore  aeration
system  with plant personnel. The purpose  of  this
discussion  is  to confirm  that plant  personnel are
aware of the O&M requirements of fine pore  aeration
technology  and  are  willing  to  devote the effort
necessary to operate and maintain their  system  in a
manner that minimizes  operational  problems  and
maximizes  process performance and  energy-savings
potential.

5.3 Process Oxygen and Mixing
Requirements
The first step in designing a fine pore aeration system
is to determine  the range and spatial distribution of the
process oxygen requirements  that must be met  with
the system. Sufficient energy  must also  be  imparted
into the  basin  to  maintain  biological solids  in
suspension (mixing  requirements).  This   section
describes   the  procedures  commonly  used  to
determine process oxygen and mixing requirements.

5.3.1 Process Oxygen Requirements
A suggested list of loading conditions that should be
considered in estimating process oxygen requirements
is illustrated as a worksheet  in Table 5-1.   Process
loadings  vary  over the life of the facility,  normally
being lower in initial  years  of operation, increasing  to
design conditions often projected to occur 15-20 years
in the future. Process oxygen  requirements also  vary
significantly during  any  year. Daily  values vary
significantly, as  represented  in  Table  5-1  by the
minimum,  average,  and  maximum daily  values
occurring  throughout  the year in question.  Diurnal
fluctuations occur  each day,  as illustrated  by the
minimum 4-hr,  average,  and  maximum 4-hr values
shown for  these three  loading  conditions.  Process
loadings also  vary  on  a  sustained  basis,  as
represented by  the minimum and maximum monthly
values shown in Table 5-1. Diurnal fluctuations  also
occur during these  sustained loading  periods, as
illustrated  by  the minimum 4-hr,  average,  and
maximum 4-hr values for each of these two loading
conditions.  These  represent "typical"  diurnal ranges
during the subject sustained loading period.

 Table 5-1.   Suggested Worksheet for List of Process
            Oxygen Requirements

                   	APR (kg/d)	

    Loading Condition   Initial Year  Mid-Poinl  20-yr Design
  Minimum Day
   Minimum 4-hr       	   	   	
   Average	   	   	
   Maximum 4-hr       	   	   	

  Minimum Month
   Minimum 4-hr
   Average
   Maximum 4-hr

  Average Day
   Minimum 4-hr
   Average
   Maximum 4-hr

  Maximum month
   Minimum 4-hr
   Average
   Maximum 4-hr

  Maximum day
   Minimum 4-hr
   Average
   Maximum 4-hr
It  is  not  necessary  to estimate  process oxygen
requirements corresponding to each loading condition
identified in  Table 5-1. The  range of conditions that
must  be considered depends on the process design
objectives of  the biological wastewater treatment
system, the potential  impacts of  periodic low  mixed
liquor dissolved oxygen (DO) on process performance
and  operating  characteristics, and economics.  In
general, the  objective is to meet the range of process
oxygen  requirements that occur  "most" of the  time.
However,  it  is generally not  necessary to meet the
entire range  of process oxygen requirements that can
occur under  all circumstances.

Several examples of the calculation of process oxygen
requirements for various biological  treatment systems
are presented in  this  chapter. Oxygen demands are
calculated for a variety of process  loading conditions,
                                                  101

-------
as  appropriate for the  particular system. An  overall
aeration  system  design problem is also presented,
beginning as  Example  5-8. The philosophy used  to
select  process loading  conditions in this example is
discussed in detail to illustrate the thought process.

5.3.1.1 Types of Process Oxygen Requirements
Oxygen  must be provided in  biological wastewater
treatment  systems  to satisfy several  types  of
demands. One demand (associated  with the oxidation
of organic matter or carbonaceous materials) is called
carbonaceous oxygen demand.  Carbonaceous oxygen
demand  is generated  by two cellular  functions: cell
synthesis and endogenous respiration.

Cell synthesis carbonaceous oxygen demand  occurs
when  organic  matter is  first metabolized  by
microorganisms  contained  in the mixed liquor.  It is
related to the oxygen required to oxidize a portion of
the organic matter to provide the energy necessary for
cell synthesis. Endogenous respiration carbonaceous
oxygen demand  occurs as  the synthesized organisms
are retained in the treatment system. It represents the
oxygen demand  exerted as the organisms oxidize
internal storage  reserves  to maintain essential  life
processes. The  net result is that increasing amounts
of oxygen  are  required as  lower  process organic
loadings  are used. Lower process organic loadings
are characterized by operation  at longer SRTs and
lower  food-to-mieroorganism  (F/M)  loadings. This
concept is  illustrated in Rgure 5-3. References 13-16
provide a detailed discussion of  this topic.

Figure 5-3.  Effect of  process loading  on carbonaceous
           oxygen requirement.

  Carbonaceous Oxygen Requirement
  por Unit o( Organic Mailer Removed
                        Effect of Endogenous
                           Respiration
          Syntliusis Oxygen
            Requirement
        Lower Process Loading (Longer SRT, Lower F/M)
Oxygen is also required to oxidize inorganic materials
present in Ihe influent wastewater. A good example is
hydrogen sulfide, which is oxidized chemically  when
brought in contact with DO in the biological reactor.
Other  examples are  listed  in  Section  5.3.i.2d.
Reactions of this type  can be quite  rapid and can
proceed to  complete oxidation  according to  well
established sloichiometry.
In some cases, inorganic chemical oxygen demand is
quantified  by direct measurement of the  chemical  in
the influent wastewater. In other cases, it is quantified
by  measurement of the resulting  oxygen demand,
traditionally by use of the immediate oxygen demand
(IOD) test. This test has been  eliminated from the
most recent  edition  of Standard Methods (17)
because of limitations on its accuracy.  However, the
concept  behind  the   test  is appropriate  for
understanding  the  nature  of  inorganic chemical
oxygen demand as  described above. Basically, the
oxygen consumption of a diluted sample  of waste is
measured over a period of time (15 minutes in the old
test) short enough to exclude biological oxidation but
long enough for chemical  oxidation of  all  readily
oxidizable  chemicals in the wastewater sample.

In  addition, oxygen  is required  for   biological,
nitrification of ammonia nitrogen to  nitrate  nitrogen. If
the process  is designed and operated in a nitrification
mode, the oxygen demand due to nitrification must be
included in the calculation of oxygen requirements  for
the system.  However, nitrification may also occur  in
systems where only carbonaceous  BOD  removal  is
required.  When  the  wastewater  is warm, say 20°C
(68°F) or above, it may not be possible to operate the
treatment  system at a high enough  organic loading  to
prevent nitrification  from  occurring. Under  these
circumstances, oxygen transfer capacity to meet this
additional demand must be provided.

Process  oxygen requirements  will be  reduced  if
denitrification  occurs  in   the  aeration system.
Denitrification can occur under controlled conditions if
the system  is specifically designed  with  an  anoxic
zone for nitrogen removal (see Section 5.2.1),  It also
occurs in some systems with relatively poor mixing or
in nitrifying systems  where  oxygen demand exceeds
the oxygen  supplied. These circumstances result  in
the development of  unintentional "anoxic zones"  in
the poorly  mixed region of the reactor.

5.3.1.2   Calculation  of  Process  Oxygen
Requirements
Several approaches are available  to calculate process
oxygen  requirements  for  a biological  wastewater
treatment  system.  This section presents the general
types of   procedures available  and discusses  the
differences between them.

Several  factors  are  important  in  determining  the
appropriate procedure for a particular situation. The
most important factor is the confidence the designer
has in the  accuracy of the design data base. Common
sense suggests  that little is gained in the use of  a
highly sophisticated procedure when  process loading
and  operating  conditions  are  known only  on  an
approximate  basis.  However,   the   use  of  a
sophisticated procedure  may be highly desirable  if
relatively  accurate  data  are available. The designer
must also  consider that an oxygen transfer system  is
                                                 102

-------
designed  to  meet  a  range  of  process  oxygen
requirements. In certain cases, When the actual range
of process oxygen requirements is not well known, it
may be appropriate to use a conservative approach.. If
a conservative value for the maximum oxygen transfer
requirement is used, sufficient  flexibility should  be
incorporated in the system  design to  accommodate
much lower process oxygen requirements efficiently.

a. Carbonaceous Process Oxygen Requirements
Four different  approaches are  reviewed to calculate
carbonaceous  process oxygen requirements. The first
approach is empirical in nature and is referred to here
as the oxygen consumption  ratio approach. With this
approach, a factor relating oxygen demand to  organic
loading is selected and used  for design. Most often
the organic loading is expressed as the BOD5 loading
in  units  of  mass  per  time,  while  the  oxygen
consumption  ratio factor  is expressed as mass of
oxygen required per  mass  of BOD5  applied  to  (or
removed by) the system. The regulations employed by
many  states to govern  the design  of  wastewater
treatment facilities  use this approach.  In other  cases,
this approach is incorporated into state regulations by
reference to  Ten  States  Standards  (18).  For
carbonaceous  process  oxygen requirements,  Ten
States Standards  states:  "In the absence of
experimentally determined values, the design  oxygen
requirements for all activated  sludge processes shall
be  1.1  Ib O2/lb peak BOD5 applied to the aeration
basins (1.1 kg O2/kg peak BOD5), with the exception.
of the  extended  aeration process, for which the value
shall be 1.8."
Figure 5-4.  Example oxygen consumption ratios for
           carbonaceous oxygen demand.
     1.8

  |  1.6
  o
  £  1.4

  *  12
   w  ' ~£-
  Q
  8  1-°
  a  0.8

  |  0.6
  
-------
                          Example 5-1.  Oxygen Consumption Ratio Approach
  Consider an activated sludge system with the following process loadings:


                                                          Waslewaler Temp.
                                        BOD5 Loading (Ib/d)        (°C)
Minimum monlh
Average month
Maximum monlh
Peak day
5,500
6,600
7,700
12,800
10
15
25
25
  The operating SRT is 4 days under all conditions. From Figure 5-4, the oxygen consumption ratio at a 4-day
  SRT and 10°C (50°F) is 0.65 Ib 02/lb BOD5 removed, increasing to 0.82 at 15°C (59°F) and 0.95 at 25°C
  (77° F).  Using  these factors, and neglecting the soluble  BOD5 in  the  process effluent, carbonaceous
  process oxygen requirements are calculated:

  Minimum month:

      (5,500 Ib BOD5/d)(0.65 Ib 02/lb BOD5)  =  3,575 Ib 02/d

  Annual average:

      (6,600 Ib BOD5/d)(0.82 Ib O2/lb BOD5)  =  5,412 Ib 02/d

  Maximum month:

      (7,700 Ib BOD6/d)(0.95 Ib O2/lb BOD5)  =  7,315 Ib O2/d

  Peak day:

      (12,800  Ib BOD5/d)(0.95  Ib 02/lb BOD5) = 12,160 Ib O2/d
b. Nitrification Process Oxygen Requirements
Nitrification  process  oxygen requirements  are
generally  estimated  on  a  stoichiometric  basis.
Theoretically, 4.57 Ib O2  are  required per Ib nitrate
nitrogen  produced in the  nitrification process  (1,21).
Accurate estimation of  the amount of nitrate nitrogen
that can be generated through nitrification is critical to
accurate use of this procedure.

Reduced nitrogen is generally  present in  an influent
wastewater in two primary forms:  ammonia nitrogen
(NHa-N)  and organic  nitrogen. Together,  the  two
components constitute the total  Kjeldahl  nitrogen
(TKN) of the influent wastewater. Figure 5-5 illustrates
the fate  of nitrogen in a typical  nitrifying activated
sludge system. Influent organic nitrogen consists  of
both biodegradable  and  nonbiodegradable fractions.
The nonbiodegradable organic  nitrogen can be further
subdivided  into  soluble and paniculate forms.  The
soluble nonbiodegradable  organic nitrogen will simply
pass  through  the biological  treatment system.  The
particulate  nonbiodegradable   organic nitrogen will
accumulate in the mixed liquor in the biological reactor
and will  be removed from the system in the waste
sludge.  Biodegradable  organic nitrogen  will  be
converted into ammonia nitrogen. This, combined with
the  ammonia  nitrogen  present  in  the influent
wastewater,  constitutes  the  nitrogen  in the system
available to the biomass.

Some of the available nitrogen  will  be used for cell
synthesis by the carbonaceous and nitrifying bacteria
produced in  the  treatment system.  The remainder is
available to  be  converted to  nitrate nitrogen by the
nitrifying organisms.  An  example  of the  fate  of
nitrogen in a typical wastewater  treatment system is
presented in  Example 5-3.

More   detailed  discussions  of  this  topic
(1,13,14,16,20,21) are  presented  elsewhere.  The
IAWPRC and many similar models,  as discussed in
Section  5.3.1.2a, incorporate the nitrogen cycle in the
procedures used to  calculate  nitrification process
oxygen requirements.

As with  carbonaceous process oxygen requirements,
simplifying assumptions  can  be  made in  calculating
nitrogenous  oxygen  demand for  those  situations
                                                  104

-------
                       Example 5-2,  Oxygen Demand Mass Balance Approach
  Consider the process loadings and operating conditions used in Example 5-1. Assume the following:

     The ratio of ultimate oxygen demand (BODun) to BOD5 is 1.5.
     Yg  = yield coefficient = 0.5 Ib VSS/lb BOD5 removed
     b = decay coefficient = 0.06 d-1
     The oxygen demand of cell material is 1.42 Ib/lb VSS.
     Effluent soluble BOD5 and VSS are negligible.

  For maximum month conditions,  BODuu applied to the system is:

     (7,700 Ib BOD5/d)(1.5 Ib BODull/lb BOD5) = 11,550 Ib BODu|t/d

  The biomass produced (Px) is calculated using the following equation (19):

     Px  = (BOD5)(Ya)/[(1 + b(SRT}]                                                           (5-1)

     Px  = (7,700 Ib BOD5/d)(0.5 Ib VSS/lb BOD5)/[1  + 0.06 d-i(4 d)]
         = 3,105 Ib VSS/d

  Expressed as ultimate oxygen demand, this  becomes:

     {3,105 Ib VSS/d)(1.42 Ib O2 demand/lb VSS) = 4,409 Ib 02 demand/d

  The process oxygen requirement of the system is the difference between the oxygen demand applied and
  the oxygen demand contained in the waste sludge. That value is:

     Process (>2 Requirement =  Oa Demand Applied - Oa Demand of Waste Sludge
                             =  11,550 Ib BODu|t/d - 4,409 Ib 02 demand/d
                             =  7,141 lbO2/d

  The maximum month process oxygen requirement estimated using the oxygen consumption ratio approach
  (Example 5-1) was 7,315 Ib O2/d. The close agreement between the two methods indicates that appropriate
  coefficients were assumed in the oxygen demand mass balance approach.
where precise  information  is  not  available.  For
example, estimates of  the particulate nonbio-
degradable organic nitrogen  and available nitrogen
incorporated  into  biomass  can be  combined  and
expressed in  terms of the total nitrogen  in the waste
sludge.  For  municipal  wastewater  that  has been
subjected to primary treatment (i.e., municipal primary
effluents), it  can be assumed that biologically inert
volatile matter that enters an activated sludge  system
(including  particulate  nonbiodegradable organic
nitrogen) normally contributes about  0.1  Ib  VSS/lb
BOD5 applied to the mass of waste sludge produced.
The total process yield  (biomass  produced  plus
biologically inert volatile matter that  passes through
the process)  for Example  5-3 would  be  about 0.6 Ib
VSS  produced/lb  BODs  applied  to the biological
reactor using  this assumption. The nitrogen content of
biologically inert volatile matter will generally be  less
than that of  biomass. Assuming that the nitrogen
content of the waste sludge is decreased from  10  to 9
percent  by this  inert  material, the mass of nitrogen
contained in the waste sludge can be calculated as:

   (150 mg/L BODS)(0.6 Ib VSS/ib BOD5)(0.09 Ib N/lb VSS)
    = 8.1 mg  N/L
In  other cases it may be appropriate to assume that
the influent ammonia nitrogen is equal to the nitrogen
available to  be nitrified. This  would  have  been
appropriate for the conditions described in Example 5-
3 (21.2 mg N/L influent ammonia nitrogen vs. 21,0 mg
N/L nitrogen available to be nitrified). The assumption
inherent with the use of influent ammonia nitrogen to
represent  the  amount of nitrogen  available  to be
nitrified  is  that the organic nitrogen  concentration
equals the nitrogen synthesized by the waste sludge.
This assumption is often appropriate for  municipal
wastewaters with minimal industrial components.

It is also  sometimes  assumed that all the nitrogen
present  in  the influent wastewater  (i.e., the influent
TKN) is available to be  nitrified. This is conservative
and  results in  overestimation of  the  nitrification
process  oxygen   requirement.  However, "it  is
sometimes  appropriate  to  use this  assumption,
particularly  when influent nitrogen concentrations are
not well  characterized.

c.  Denitrification Process Oxygen Credit
The occurrence of denitrification in an aeration system
reduces total  process  oxygen requirements.  This
                                                 105

-------
Figure S-S,   Nitrogen metabolism in nitrifying activated sludge systems.
                                         Influent
                                          TKN
                           Ammonia
                            Nitrogen
          Nitrogen
        Available (or
         Nitrification
                                         Nitrogen
                                        Incorporated
                                        Into Biomass
Organic
Nitrogen


Particulate
Nonbiodegradabte
Organic Nitrogen
                          Soluble
                      Nonbiodegradabte
                      Organic Nitrogen
occurs because a portion of the carbonaceous oxygen
demand  is satisfied by reduction of nitrate nitrogen to
nitrogen  gas  during the denitrification  process.
Theoretically, 2.86 Ib Oa  demand is  satisfied per Ib
nitrate nitrogen reduced to nitrogen gas (1,21).  Thus,
the denitrification  process oxygen credit is calculated
by multiplying the  mass of nitrogen denitrified by 2.86.

Denitrification occurs in a  biological  wastewater
treatment system  when  nitrate is added to an anoxic
zone (see Figure 5-2c).  The extent  of denitrification
can be limited by either of two factors:  1) the mass of
nitrate nitrogen  added to  the anoxic zone, or 2) the
denilrtfication capability of the anoxic zone. Assuming
that  sufficient capability is present in the  anoxic  zone
to denitrify the nitrate nitrogen loading, the mass of
nitrate nitrogen  denitrified can  be calculated directly
from a mass balance  on  the zone. Procedures used to
size anoxic zones are presented elsewhere (5,21).

d. Inorganic Chemical Process Oxygen Requirements
A variety of  reduced materials can be present in  a
wastewater stream that will be oxidized when brought
in contact with DO. Examples include  reduced  forms
of sulfur such  as sulfide,  sulfite,  and  thiosulfite;
ferrous  iron;  and  reduced  manganese.  For  many
wastewaters,  the process  oxygen  requirements
associated  with  oxidation of  these  materials  are
insignificant in  comparison  to  those  associated with
oxidation of organic material and ammonia  nitrogen.
This occurs because wastewater collection  systems
are ideally  designed to maintain the wastewater in an
aerobic condition. Therefore,  DO is  available in  the,
wastewater as  it  flows through the collection system
and  any reduced  inorganic materials  are  normally
oxidized  there  before  they reach the  wastewater
treatment plant.  Under certain  conditions, however,
significant  quantities of  reduced  inorganic materials
can  be  present.  These conditions include: 1)  the
development of anaerobic conditions  in the collection
system due to the use of shallow  slopes  or  force
mains, 2)  the  presence of an industrial  waste that
contains  significant  quantities  of  reduced  inorganic
materials, and 3) the addition of reduced chemicals as
part  of the treatment  process,  such as  the use  of
ferrous chloride for phosphorus removal.

Inorganic chemical  process  oxygen requirements can
be calculated directly on a stoichiometric basis.  For
example, consider the oxidation of hydrogen sulfide:
              H2S + 202 =  H2SO4
(5-2)
For this  reaction, 2  moles  of  Oa  are required  to
oxidize 1  mole of hydrogen  sulfide  (HgS)  to sulfuric
acid (HgSO^. This corresponds  to approximately 2 Ib
                                                   106

-------
                       Example 5-3.  Oxygen Demand Mass Balance Approach
  Consider  a wastewater influent  to  a biological wastewater treatment  system  with  the following
  characteristics:

      Flow = 5.3 mgd
      Ammonia nitrogen = 21.2 mg N/L
      Organic nitrogen  = 8.8 mg N/L

      [TKN of this wastewater is 21.2 mg N/L ammonia nitrogen plus 8.8 mg N/L organic nitrogen, or 30 mg
      N/L.]

  The nonbiodegradable components of the organic nitrogen are:

      Particulate nonbiodegradable organic nitrogen = 3.0 mg N/L
      Soluble nonbiodegradable organic nitrogen = 1.0 mg N/L

  The available nitrogen is the influent TKN less the nonbiodegradable nitrogen components:

      Available nitrogen = 21.2 + 8.8 - 3.0 - 1.0 = 26.0 mg N/L

  A  portion of this  nitrogen  is used  for cell  synthesis.  Cell  synthesis is a  function  of  influent  BOD5
  concentration, process loading conditions,  and process kinetics. For this example, these parameters are
  assumed to be:

      Influent BOD5 =  150 mg/L
      Yg = 0.5 Ib VSS/lb BOD5 removed
      b = 0.06 d-1
      Nitrogen content  of biomass =  0.1 Ib N/lb VSS
      Process SRT = 8 d

  Then, from Equation 5-1, the nitrogen used for synthesis of heterotrophic organisms is:

      (0.1 Ib N/lb VSS)(150 mg/L BOD5)(0.5 Ib VSS/lb BOD5)/[1  + ,0.06d-i(8d)] = 5.0 my N/L

  Theoretically, some nitrogen is also used for synthesis of the  nitrifying organisms.  However, this quantity
  can be  neglected for most practical applications and the nitrogen to be nitrified is the difference between
  the available nitrogen and the nitrogen incorporated in the biomass:

      Nitrogen to be nitrified = 26.0 - 5.0 = 21.0 mg N/L

      [This calculation  assumes either that nitrogen is not returned to the liquid process stream from solids
      processing or that this nitrogen is  included in the influent concentration.  Nitrogen recycles from solids
      processing can be substantial.]

  Assuming complete nitrification, the nitrification process oxygen requirement is:

      (5.3 mgd) (21 mg/L NO3-N generated) (4.57 Ib O2/lb N03-N generated) (8.34 Ib/mil gal/mg/L)
      = 4,242 ib O2/d                         '                           .
       gS. Similar stoichiometric calculations should
be used to estimate process oxygen requirements for
other reduced inorganic materials.

Another  approach  is  used  to  calculate  inorganic
chemical process oxygen requirements when the IOD
procedure is  employed to quantify the concentration
of  reduced  inorganic  chemicals  present  in  a
wastewater  stream. In  this   case,   since  the
concentration  of  reduced  inorganic materials  is
measured  directly as its oxygen equivalent, the
measured IOD is used  directly  as  the associated
process  oxygen  requirement.  Caution  should  be
exercised  when using the IOD procedure to ensure
that  all  inorganic  chemicals are oxidized  within the
time frame  of the  IOD  procedure. Caution  is also
required to ensure that none of the reduced inorganic
materials  are oxidized  because  of  unintentional
aeration during  sample collection  or  handling before
performing the analysis.

e. Total Process Oxygen Requirements
The  total process oxygen requirement (also called the
actual oxygen requirement, or AOR) for  a biological
wastewater  treatment system is  the sum  of the
process oxygen requirements described above:
                                                 107

-------
AOR « Carbonaceous Process Oxygen Requirement
     + Inorganic Chemical Process Oxygen Req.
     + Nitrification Process Oxygen Requirement
     - Denilrificalion Process Oxygen Credit      (5-3)

Examples 5-4 through 5-7 illustrate the calculation of
AORs for several situations.

5.3.1.3   Variations  in   Process  Oxygen
Requirements
Process oxygen requirements vary  both spatially (i.e.,
from point to point) and temporally (i.e., from time to
time)  in  biological  wastewater treatment  systems.
Spatial   variations  occur  because  the  various
components of the  total process oxygen requirement
are not uniformly distributed throughout the biological
reactor.  Temporal  variations occur  because  of
variations  in  process  loadings and  operating
conditions.  Sufficient capacity  and  flexibility  must  be
designed  into  fine  pore aeration  systems  to
accommodate these variations.

Inadequate capacity in the  highly-loaded section of a
biological  reactor  can  adversely affect  process
performance  due to  sustained   low  process  DO
concentrations. However, in some  cases, the oxygen
demand  will  be transferred down  the  basin  and  be
satisfied  without affecting effluent  quality. Poor
distribution of oxygen transfer  capacity  or inadequate
aeration  rate  turndown  capability can  result  in
overaeration and inefficient operation during periods of
lower process loading.

The magnitude  of  temporal  and  spatial  variations, •
along with  procedures  for  quantifying them,  are
discussed in the following sections.

a. Temporal Variations
The concept of temporal variations  in process oxygen
requirements was introduced  in Section 5.3.1, using
Table 5-1 to  illustrate the types of  variations that can
occur.  Temporal variations occur  because  of both
short- and long-term variations in process  loadings
and operating conditions.  Short-term variations include
both day-to-day variations  in  process  loadings and
operating conditions, as well as variations  during the
day (i.e., diurnal variations).  Figure  5-6 presents  a
typical diurnal BODs loading for wastewater treatment
plants. The peak 4-hr:average BODg loading values
reported  for four full-scale treatment plants  were 1.17,
1.30, 1.38,  and 1.50 (22).

Importantly, short-term variations in process oxygen
requirements are smaller than corresponding short-
term variations in process  loadings.  This  occurs for
three reasons: 1) the  process  does not fully adapt to
short-term  loading extremes (either high or low),  2)
storage  of  organic  matter  occurs  at peak loadings,
resulting in  attenuation of  peak process oxygen
requirements, and 3)  the aeration  basin volume has
an  equalization effect on  process loadings.  For
example,  a 3:1  diurnal  range  in  process  BOD5
loadings may result in only about a 2:1  diurnal  range
in carbonaceous process oxygen requirements.

Long-term  variations in process  loadings  and
operating conditions have a  more pronounced  effect
on process oxygen requirements because the process
fully  acclimates  to  these  changes. Long-term
variations  include weekly,   monthly, and  seasonal
variations. In reality, the impacts of  short- and long-
term  variations in process  loadings and  operating
conditions  on  process oxygen requirements  are
additive.  Long-term  variations  result in changes in
"average"  process  oxygen  requirements.
Superimposed on these  are short-term changes in
process oxygen requirements  resulting from diurnal
variations  in  process  loadings  and  operating
conditions.

Judgment must be exercised in selecting the range of
process oxygen requirements to be accommodated by
a particular design. According to Manual of Practice 8
(13), it is not unusual  for the peak organic loading to
be  5-10  times  the  minimum  hourly  load  in a
conventional  wastewater  treatment  plant.  It  is
generally not necessary, however, to  satisfy the  entire
range  of  process oxygen  requirements.  Periodic
occurrences of low process DO can  generally  be
tolerated without  compromising process integrity  or
significantly affecting process performance. However,
frequent or prolonged  periods of suppressed process
DO should be avoided.

Thus, it is  not  necessary to install sufficient oxygen
transfer capacity  to meet the peak diurnal process
oxygen requirement  occurring  on  the  peak day.
Rather, the normal  design  approach  would  be  to
install sufficient oxygen transfer capacity to  meet the
typical or average  diurnal process oxygen requirement
occurring  during  the sustained  loading  condition
selected as the basis for the process design.  Since
process designs are often based  on  maximum month
loading,  the oxygen  transfer  system  would   be
designed to meet both the peak day demands and the
typical diurnal process oxygen requirements occurring
in  the maximum  month.  Maximum diurnal  demands
during peak days  could be handled with the standby
blower capacity in  the system.

Estimation  of  temporal variations  in  total  process
oxygen requirements is illustrated in Example 5-8.

b. Spatial Variations
Figure 5-8  illustrates  an  idealized spatial distribution
of process oxygen requirements along the length of a
plug flow reactor when not limited by the process  DO
concentration.  In  this illustration,  the  carbonaceous
process oxygen requirement is divided into its  two
components,  synthesis and  endogenous  respiration.
Endogenous  process oxygen  requirements  are
associated with the biomass and tend to  be distributed
                                                 108

-------
              Example 5-4.  Conventional Secondary Treatment with Seasonal Nitrification
   Consider the activated sludge system from Example 5-1. Although not required, it is likely that nitrification
   will occur during the summer when wastewater temperatures are higher. It is assumed that wastewater
   temperatures average 20°C (68° F) during the summer months when nitrification is  occurring. Without
   adequate oxygen, the onset of nitrification can (ead to septic conditions and process upsets. Sufficient  .
   oxygen transfer capability must be provided to satisfy these periods of nitrification and to preserve
   acceptable effluent quality. However, periodic low DO episodes can be tolerated, indicating that nitrification
   process oxygen demand need not be satisfied during peak day events.

   The average ammonia nitrogen loading to the facility is about 935 Ib  NH3-N/d. Assume that the nitrogen
   available to be nitrified is equal to  the  ammonia  nitrogen loading. As a result, the nitrification process
   oxygen  requirement  is 4.57 Ib O2/Ib NO3-N generated x 935  Ib  NO3-N generateoVd  = 4,273 Ib O2/d.
   Tabulate AOR for both nitrifying and non-nitrifying conditions.  For non-nitrifying  conditions, AOR  is the
   carbonaceous  process oxygen  requirement as calculated  in Example 5-1. For  nitrifying  conditions,
   nitrification process oxygen requirement of  4,273  Ib O2/d must  be added to the  average and  maximum
   month carbonaceous process oxygen requirements. The results are:
                                                  AOR (Ib/d)
                      Loading Condilion
Non-nilrifyirtg
Nitrifying
Minimum month
Average tnonih
Maximum rnonlh
Peak day
3,575
5,412
7,315
12,160
Not Applicable
9,685
1 1 ,588
Not Applicable
                      Example 5-5.  Effect of Inorganic Chemical Oxygen Demand
   Consider the activated sludge system from Example 5-4, and evaluate the effect of wastewater septicity on
   AOR. Assume for this evaluation that the H2S concentration of the influent wastewater approaches 10
   mg/L. From  Section 5.3.1.2d, 2 Ib Oa are required to oxidize 1 Ib HaS. Therefore, the inorganic chemical
   oxygen requirement associated with the H2S in the influent wastewater is:

       (5.3 mgd)(10 mg/L H2S)(2  !b O2/lb H2S)(8.34 Ib/mil gal/mg/L) = 884 Ib/d

   Referring to the AORs listed in Example 5-4, wastewater septicity (i.e., the presence of hydrogen sulfide)
   would  increase the average month AOR by 9 percent under nitrifying conditions and by 16 percent under
   non-nitrifying conditions.
uniformly throughout  the  length of  the reactor.
Synthesis process oxygen requirements are highest in
the more highly-loaded portion of the reactor (i.e., the
initial   section).  Nitrification   process oxygen
requirements are exerted uniformly along the length of
the reactor until they become limited by the depletion
of ammonia nitrogen (23).

Use of  the  more  sophisticated  procedures for
predicting  spatial variations  in  process  oxygen
requirements  necessitates an  understanding  of the
hydraulic  flow  pattern within the biological  reactor.
Very often, flow patterns  are characterized in terms of
the equivalent number  of basins-in-series that the
basin  represents  (6).  Hydraulic flow patterns for
existing reactors may  be characterized using standard
tracer  techniques. In  other situations, experience or
            correlations from  the  literature  can be used.  For
            example,  the  Water Research Centre in the United
            Kingdom  has  developed  the  following  empirical
            relationship to estimate the  number of equivalent
            basins-in-series for a diffused air aeration  basin:
                        N = 7.4 LQ(1 + rr) * (WH)
                                      (5-4)
            where,
               N  = equivalent number of basins-in-series
               L  = aeration basin length, m
               Q  = wastewater flow, m3/s
               rr  = return  activated  sludge  recycle
                    (dimensionless)
               W = aeration basin width, m
               H  = water depth,  m
                                       ratio
                                                  109

-------
                               Example 5-6.  Nitrifying Activated Sludge
Consider the activated sludge system from Example 5-1. The system  is to be expanded to provide year-
round  nitrification. A design SRT of 12 days  is selected to allow nitrification to occur during cold weather
operation. Process BODg loadings are  as listed in  Example 5-1. An analysis of the  mass of  ammonia
nilrogen, nitrite nitrogen, and nitrate  nitrogen in the process effluent was completed  and indicated the
following masses of nitrogen were available to be nitrified:

     770 Ib/d  -   Minimum  month
     935 Ib/d  -   Average month
    1,080 Ib/d  -   Maximum month
    1,500 Ib/d  -   Peak day

Process AOR  is  calculated  as the sum of  the  carbonaceous process  oxygen  requirements and the
nitrification process oxygen requirements. Carbonaceous process oxygen requirements are calculated using
Figure 5-4:
                                                                           Carbonaceous
                                       Temp.       Ratio      BOD5 Loading   Oxygen Req.
           Loading Condition    SRT (d)      (°C)     (Ib Q.Jlb BODsj	(Ib/d)	(Ib/d) ,
            Minimum month       12       10         1.00         5,500          5,500

            Average month       12       15         1.05         6,600          6,930

            Maximum month       12       25         1.15         7,700          8,855

               Peak day          9       25         1.05        12,800         13,440
It  should be  noted  that  an "effective  SRT" of 9  days was  selected  to  calculate  process oxygen
requirements for peak day operating conditions to account for the lower unit oxygen consumption ratio that
will occur during relative short-duration,  high-loading  conditions.  In these situations, a higher proportion of
organics will be  stored rather  than oxidized by the microorganisms.  Thus, even  though  SRTs will not
change instantaneously in the field, use of an effective SRT of  9 instead of  12 days  in this example is
assumed to approximate the lower oxygen demands under these circumstances.

Nitrification process oxygen requirements are calculated directly from the mass of nitrogen available to be
nitrified:
                                                                Nitrification Oxygen
               Loading Condition   N Loading (Ib/d)   Ratio (Ib 02/lb N)        Req. (Ib/d)
                Minimum month         770            4.57              3,519

                Average month         935            4.57              4,273

                Maximum month        1,080            4.57              4,936

                   Peak day           1,500            4.57              6,855

The AOR is the sum of the carbonaceous and nitrification process oxygen requirements:


                                           Process Oxygen Requirement (Ib/d)
                                   Carbonaceous
                 Loading Condition      Demand     Nitrification Demand      AOR
                  Minimum month        5,500           3,519            9,019

                  Average month        6,930           4,273           11,203

                  Maximum month        8,855           4,936           13,791

                    Peak day         13,440           6,855           20,295

The Impact of designing for  complete,  as opposed to partial,  nitrification on process oxygen requirements
can be observed by comparing these results with those of Example 5-4. In this case, average month design
AORs are Increased by only  16 percent, but peak day AORs are increased by about 67 percent.
                                                 110

-------
                                  Example 5-7.  Impact of Anoxic Zone
  Consider the nitrifying activated sludge system from Example 5-6, and evaluate, the impact of the addition
  of an upfront anoxic zone on AOR. For the evaluation, assume that the anoxic zone size and mixed liquor
  recirculation rate (see Figure 5-2c) are adequate to allow denitrification of 60 percent of the nitrate nitrogen
  generated  in the process (see References 5 and 21 for information on the design  of  anoxic  zones).
  Denitrification satisfies 2.86 Ib Oa demand/lb NOa-N denitrified. The denitrification process oxygen credit is:
                Loading Condition
NO3-N Generated1
     (Ib/d)
 NO3-N Denitrified2
	(Ib/d)
Oxygen Credit3
    (ib/d)
                 Minimum month

                 Average month

                 Maximum month

                   Peak day
      770

      935

    1,080

    1,500
      462

      561

      648

      900
    1,321

    1,604

    1,853

    2,574
              1 From Example 5-6,
              2 60 percent o( NO3-N generated.             '                                 .       .
              3 2.86 x NO3-N denitrificatied.

  Comparing these values  to the AORs  tabulated for Example 5-6,  addition of  an anoxic zone will reduce
  AOR for the nitrifying activated sludge system by 13-15 percent. The design AORs would then be:
                                              Process Oxygen Requirement (Ib/d)
                 Loading Condition
   Carbonaceous
    Demand
  Nitrification   Denitrification
   Demand      Credit
                                                                              AOR
Minimum month
Average monlh
Maximum month
Peak day
5,500
6,930
8,855
13,440
3,519
4,273 ,,
4,936
6,855
1,321
. 1,604
1 ,853
2,574
7,698
9,599
11,938
•17,721
Figure 5-6.  Example diurnal BOD5 loading for a municipal
           wastewater treatment plant.
 Relative BOD5 Loading

    1,5
    1.0
    0.5  -
            i  I i  I i  I ,  I  i ! i  I  i I i  I  , I i |  i I
              4     8     12     16     20     24

                   Time from Midnight, hr
Several procedures can  be used  to estimate spatial
variations of  total  process  oxygen  requirements  in
biological wastewater treatment  systems.  In  some
cases,  simple rules-of-thumb based  on  experience
with similar treatment systems are  used. For example,
oxygen transfer capacity in a conventional activated
sludge reactor is often tapered as follows:
                      45-55 percent of total process air in the first one-third;
                      25-35 percent of total process air in the second one-third;
                      and
                      15-25 percent of total process air in the last one-third.

                     Another  approach involves  construction of diagrams,
                     such as Figure 5-8,  that quantitatively  distribute the
                     various  components of  the total  process oxygen
                     requirement. A numerical example of this approach is
                     presented below.

                     For  existing biological wastewater treatment systems,
                     temporal and  spatial variations  in  process oxygen
                     requirements can be measured directly. Hourly,  daily,
                     and  seasonal variations can be  measured in various
                     portions  of  the reactor  and  used to  establish ranges
                     that  must  be  accounted  for in  the  design.  This
                     approach  is  limited to  the wastewater  type  and
                     operating conditions encountered during testing. If an
                     existing  system is to be retrofitted, the possibility of a
                     change in overall reactor mixing patterns must also be
                     considered.

                     When sufficient information  is available,  process
                     models such as the IAWPRC model  described  above
                     can   be  used to  estimate spatial  and  temporal
                     variations  in  total  process oxygen  requirements.
                     Recent work (24) suggests  that the IAWPRC  model
                                                   111

-------
             Example 5-8.  Temporal Variations in Total Process Oxygen Requirements
A fine pore aeration system is being designed to serve the  wastewater treatment needs of a community.
Wastewater flows were projected 20 years into the future and are estimated to average 5,3 mgd at that
design condition. Current flows average 2.6 mgd, and wastewater flows and loadings are anticipated to
increase in an approximate linear fashion from current levels up to their 20-yr design values.

Secondary treatment is to be provided to meet discharge requirements. The average hydraulic retention
time (HRT) in the aeration basin is 6 hours.  The average operating SRT is estimated to be 4 days to limit
nitrification and its associated  process  oxygen requirements. However,  nitrification is expected to occur to
some degree during  the 5 warmest months of  the year. The design  process loadings are as  listed for
Examples 5-1, 5-2, and 5-4.

The selected activated sludge  system consists of four aeration basins, each 23 ft wide by 130 ft long with a
sidewaler depth (SWD)  of 15  ft. Diffuser  submergence is 14  ft. Four basins  may be considered an
unusually large number for a 5.3-mgd facility; some designs  would  provide only two. However, four basins
were chosen in this case considering the wide variation in process loadings from initial  operation to the 20-
yr design values. Selection of  the number of aeration basins  is an economic issue that  must be considered
for  each installation. Construction of more basins increases initial costs but is normally needed for diffuser
maintenance. In addition, operating costs are reduced since only the number of basins  necessary to satisfy
maximum process oxygen requirements need be in service at any point in the life of the facility. The trade-
off between these two factors  must be  considered for each design.  Figure 5-7 presents a schematic of the
proposed facility.

The range of total  process oxygen requirements that form the design basis for this facility are estimated in
the following  discussion.  For design purposes, total process  oxygen requirements will be estimated for five
process loading conditions:

    1.  Minimum month
    2.  Average non-nitrifying  (winter) month
    3.  Average nitrifying (summer) month
    4.  Maximum month  (including nitrification)
    5,  Peak day (non-nitrifying)

Conditions 1  through 4 represent sustained loading conditions, while condition 5 represents a short-term
peak. Short-term periods with minimum total process oxygen requirements  less than the minimum month
requirement will occur within this system. However, it  is not judged  cost effective to  design for anything
less than the minimum month  because of the low frequency of occurrence of these periods.  The option is
simply to waste some energy during those  infrequent  periods when process  demands are less than  the
minimum month value. Designing for the peak day not only accommodates  this  peak loading, but also
provides an allowance for diurnal variations during more typical average loading conditions.

Total  process oxygen requirements have already been calculated in Example 5-4 for the  specified design
loading conditions. Using the format of Table 5-1, AOR design values are:
                                                       AOR (Ib/d)
                       Loading Condi lion
Initial Year
Midpoint   20-yr Design
                        Minimum month          1,788        2,681       3,575

                   Average non-nitrifying monlh     2,706        4,059       5,412

                     Average nitrifying month       4,843        7,264       9,685

                        Maximum month          5,794        8,691       11,588

                     Peak day (non-nitrifying)       6,080        9,120       12,160
                                                                                         (continued)
                                                112

-------
         Example 5-8,  Temporal Variations in Total Process Oxygen Requirements (continued)
  The 20-yr design values are taken directly from Example 5-4, while the initial year and midpoint values were
  calculated simply as being proportional to process loadings (i.e.,  the initial-year values are one-half of the
  20-yr design values, and the midpoint values are three-quarters of the 20-yr design values). This approach
  can be used because a sufficient number of aeration basins is  available  to approximately match design
  process loadings over the design life of the facility (i.e., two basins will be operated in the initial years, three
  at the midpoint, and all four at the 20-yr design value). Note  that the  use of peak day non-nitrifying
  conditions  for  sizing purposes  will accommodate  significant  variations  in  diurnal  process  oxygen
  requirements  under average loading  conditions. Approximate  allowable diurnal  peaking  factors  are
  12,160:5,412, or 2.25:1, for non-nitrifying conditions and 12,160:9,685, or 1.25:1, for nitrifying conditions.
Figure 5-7.  Activated sludge system for Example 5-8.


                    I.                  130ft
          Primary
          Effluent
                       23 ft
                                     15ftSWD
                                          Return Activated Sludge
Figure 5-8.  Spatial  variation  in  process  oxygen
           requirements along the length of a biological
           reactor.

 Process Oxygen Requirements
       (arbitrary units)
           Synthesis
           Nitrification
               Endogenous
                                          100
               Reactor Length, percent
can  be  readily calibrated to  existing wastewater
treatment facilities.  Adjustments in  only two or three
key parameters were required  (while keeping 17 other
model  parameters at their default values) to calibrate
the model to  six  full-scale  wastewater  treatment
plants.

The  Water Research  Centre  has  also developed a
biological process mode! that it uses to predict spatial
and  temporal  variations in  total  process oxygen
requirements (25).  The use  of  process models  to
estimate variations in process  oxygen requirements is
not standard practice in North America today.

Spatial variations in  total process oxygen requirements
are estimated  for  a  hypothetical  activated sludge
system in Example 5-9.

5.3.2 Process Mixing Requirements
Sufficient aeration must  be  provided to  prevent
deposition of suspended matter. In evaluating process
mixing requirements,  different  diffuser configurations
exhibit very  different   mixing  characteristics.  Limited
information  has  been published,  however,  on
minimum mixing  requirements.  One document  (15)
indicates that the airflow rate required to ensure good
                                                   113

-------
               Example 5-9.  Spatial Variations in Total Process Oxygen Requirements
Consider  ihe  activated sludge system from Example 5-8. Experience indicates  that the hydraulic  flow
pattern wilhin  each of the basins in this example can be  approximated as three equivalent basins-in-series
(nole lhat Equation 5-4 would predict between three and four equivalent basins-in-series, depending on r).
As  a result, the diffusion system  in  each reactor will  consist of three equal size aeration  zones, as
illustrated in Figure 5-7. The number of diffusers in each  zone will be  varied in proportion to the estimated
fraction of the average total process oxygen requirement that can be satisfied in each zone.  Using this
information, estimates of the spatial distributions in total process oxygen requirements for the 20-yr design
conditions can be made.

Rrst, subdivide the carbonaceous  process oxygen  requirement  into  its synthesis  and  endogenous
components. Assume that the oxygen consumption ratio  of the synthesis component is  about 0.5 Ib O2/lb
BODs applied. At the average BOD5 loading of 6,600 Ib/d:

Synthesis Process Oxygen Requirement
    = (6,600  Ib BODs/d)(0.5 ib O2/lb BOD5) = 3,300 Ib O2/d

The tola! carbonaceous process oxygen requirement is 5,412 Ib O2/d, so the endogenous component can
be calculated  by subtraction:

    Endogenous Process Oxygen Requirement  = 5,412 - 3,300 = 2,112 Ib Oa/d.

Experience Indicates that two-thirds of the synthesis process oxygen requirement will occur in the first one-
third of the reactor, while the remainder will occur in the  middle one-third (26). If sufficient oxygen transfer
capacity is not provided in the initial and middle thirds of the reactor to satisfy these requirements, resulting
low process DOs will shift a portion of the oxygen demand load further down the reactor.

Based on the  above assumptions, the distribution in the carbonaceous process oxygen requirement among
the three zones is:
                                  Average Carbonaceous Process Oxygen
                                         Requirement (Ib O2/d)
                         Zone     Synthesis    Endogenous    Carbonaceous
1
2
3
2,200'
1,1002
_
7043
704
704
2,904
1,804
704
                       ' 2/3 x 3,300 Ib O2/d.
                       2 1/3 x 3,300 Ib (yd.
                       3 2,112 Ib Q-Jd divided equally into three zones.
This is the spatial distribution under non-nitrifying conditions. When nitrification occurs, an additional oxygen
requirement of 4,273 Ib O2/d must be satisfied. Since a 4-day  SRT is about the lowest SRT at which
reliable nitrification can be expected to occur, nitrification process oxygen requirements should occur fairly
uniformly  throughout the reactor until the nitrogen concentration  becomes substrate limiting  (23,26).
Assume, therefore, that 40  percent  occurs in the first one-third of the reactor, 40 percent in the second
one-third of the reactor, and the remainder (20 percent) in the last one-third of the reactor (23). This
distribution pattern will vary  from  plant  to  plant.  For  this  example,  the  nitrification process  oxygen
requirement will be distributed as follows:

    For the first and second zones, nitrificalion process oxygen requirement
    -  0.4(4,273 Ib O2/d) =  1,709 Ib O2/d each
    For the last zone, nitrification process oxygen requirement
    =  0.2(4,273 Ib O2/d) =  855 Ib O2/d
                                                                                         (continued)
                                                 114

-------
          Example 5-9.  Spatial Variations in Total Process Oxygen Requirements (continued)
These are added to the carbonaceous process oxygen requirement in each zone. A summary of calculations
made for each condition yields the following spatial distribution of total process oxygen requirements:
Zone
1
2
3
Total

Peak Day
6,187
4,053
1,920
12,160

Max. Month
5,430
4,147
2,010
1 1 ,588
AOR (Ib/d)
Average
Month
Nitrifying
4,613
3,513
1,559
9,685

Average
Month Non-
Nitrifying
2,904
1,804
704
'5,412

Mm. Monih
2,109
1,191
275
3,575
mixing in diffused air systems is 0.33-0.50 L/s-m3 (20-
30 scfm/1,000 cu ft). The diffuser type and layout are
not delineated in this document. Manual of Practice 8
(13) recommends a minimum  mixing  requirement  of
0.6 L/s-m2  (0.12  scfm/sq ft)  be used  for  ceramic
dome diffusers operating in a  grid configuration and
0.33 L/s-m3 (20 scfm/1,000 cu  ft) for a coarse bubble
spiral roll  configuration. Mixing  evaluations performed
on  a  ceramic  dome  diffuser grid configuration
(diffusers  61 cm  [24 in]  off  the  floor)  at  the Los
Angeles - Glendale, CA Water  Reclamation Plant (27)
revealed no solids settling problems (MLSS  =  1,500
mg/L) after 2 weeks of testing  at airflow rates as low
as 0.25 L/s-m2  (0.05 scfm/sq ft).


5.4 Air Diffusion System
The air  diffusion  system  transfers  oxygen to the
aeration basin. The transfer should occur in such a
way that  the microorganisms  are not "stressed."  A
stressed situation can occur if either the overall rate  of
oxygen transfer to the  basin  or the distribution  of
oxygen within the basin is not  adequate. This section
discusses key  considerations in the design of the air
diffusion  system,  including  diffuser  selection,
arrangement of the diffusers within the basin, airflow
distribution,  diffuser  cleaning  and  maintenance,
diffuser installation, and retrofit applications. Following
the  discussion  of  these items,  an  example  is
presented that  illustrates a typical air diffusion system
design.

5.4.1  Diffuser Selection
Several factors should be  considered in selecting the
specific fine pore diffusion device to be  used  in a
particular  application.  Cost considerations  include
initial cost of the total diffusion system, O&M costs,
and life-cycle cost.  Often the  major  concern is the
initial cost of the system.  However, this cost usually
represents a small fraction (15-25 percent) of the life-
cycle cost of the system.  The major part of the life-
cycle system cost is determined by O&M costs.  OTE
plays a major role in establishing system O&M costs.
Other factors that affect O&M costs, but that are more
difficult to quantify,  include operational flexibility,
system reliability, and maintenance requirements.

5.4.1.1 Oxygen Transfer Efficiency
One important diffuser selection criterion  is its clean
water  OTE.  Performance  and  oxygen  transfer
characteristics of fine pore diffusers in clean water are
discussed in  Chapter 2. As indicated in that chapter,
factors that affect  clean  water transfer  efficiencies
include diffuser  type  (material,  shape,  and  size),
diffuser configuration and density (number per  100 sq
ft), airflow rate per diffuser, diffuser submergence, and
air flux uniformity.

While  clean  water  OTE is  an important selection
criterion, it is not  an  accurate indicator of diffuser
performance  under process  conditions.  OTEs  are
usually substantially lower  under process conditions
than in clean water. Chapter  3  discusses  factors  that
affect  oxygen  transfer  under process  conditions.
Factors over which the designer  has control  include
process  type,  flow regime,  basin  geometry,  and
diffuser placement (including depth, configuration, and
density).

Selections of the process type, flow regime, and basin
geometry  partially determine  the average  wastewater
characteristics to which the  diffuser will be exposed.
These selections have a direct  effect on  a  and  F
profiles throughout  the basin.  Low-rate  systems (low
F/M  loadings or  high  SRTs), complete  mix systems,
and the effluent  end of plug flow  basins will normally
have higher average a values than  high-rate systems
and  the  influent  end of plug flow basins. Diffusers
operated in systems or areas within a basin with low a
values also  will usually exhibit  higher rates  of
biological fouling than diffusers operated under high a
value conditions.

Increasing the plug flow characteristics of the basin by
baffling or increasing  the  length-to-width ratio  will
cause  a to decrease and the  rate of biological fouling
to increase  at the  influent end of the  basin.  At the
same  time, the  specific  oxygen demand (oxygen
                                                   115

-------
demand per unit volume) will  increase at the influent
end of the basin.

Chapter 2 discusses the effects of diffuser depth and
density  of  placement  on  oxygen  transfer
characteristics.  Increasing diffuser depth  usually
increases OTE (as a percent of the oxygen content of
the air delivered to the basin). However, wire aeration
efficiency (measured as kg [!b] oxygen transferred per
kWh (wire hp-hr]) does not appear to vary significantly
wilh diffuser depth' over diffuser depths of 4-8 m (12-
25  ft)  (see Rgure 2-22),  For  a given airflow per
diffuser, both  OTE  and  wire aeration  efficiency
generally increase as the diffuser arrangement moves
from a spiral roll to a full floor coverage configuration.

As part of making final decisions regarding features of
the design  described above, the designer  should
evaluate  oxygen transfer characteristics of the
diffusers and  process  requirements  under various
conditions. For example,  influent zones  of plug  flow
basins may exhibit a  values of 0.3 and,  after several
months of operation, F may be reduced to 0.8. Thus,
the diffusers in these  zones might only be transferring
a  maximum of  25  percent  of their clean  water
capabilities at that time. Further, the extremes of peak
process loadings with fouled  diffusers and minimum
process loadings with  clean  diffusers  (turndown/
turnup) must be  considered.

5.4.1.2 Operational Flexibility
A  variety of activated  sludge reactor configurations
(i.e., flow regimes) can be designed. Most of these fall
into one of three groups: complete mix, plug flow, or
step feed. A step feed regime  can also be used  with a
plug flow configuration  to  offer greater  operational
flexibility.  Each of these  configurations  has
advantages and  disadvantages. As a result, treatment
facilities  are sometimes designed to allow operation
using more than one flow regime (for example, both
plug flow  and complete mix configurations  or  an
interchangeable  combination  of step  feed  and plug
flow configurations could be used).  Further,  each of
these  configurations  can  be  designed  to  operate
under high-rate (high  F/M) loading conditions, low-rate
(low F/M) loading conditions,  or somewhere between
these two extremes. Wastewater treatment plants are
normally designed anticipating growth in  the service
area, so a facility designed to operate as a high-rate
system may initially  operate  as a  low-rate  system.
Process operational flexibility increases as the number
of reactor configurations and the range  of loading
conditions the operator can use increase.

Nearly all designers  of  wasu.-waler  treatment  plants
attempt to provide some process operational flexibility.
Sufficient operational flexibility should  be included in
any aeration system to  meet  the  variable oxygen
demand  requirements  that  different reactor
configurations  and loading conditions  impose.
Methods for calculating oxygen  requirements  are
discussed in Section 5.3.1.2. The  designer needs to
estimate oxygen requirements for the entire range of
operating conditions  likely to be encountered in  the
treatment facility. As indicated previously,  oxygen
requirements  vary  both temporally and  spatially.
Variations with time may include changes that  occur
diurnally, by day of the week, by season, and between
plant start-up and when  design  loading conditions are
reached. Oxygen  demands  will be  higher at the inlet
end of a plug flow basin than at the outlet end, and
will be highest at the feed points in a step feed basin.

Having  calculated  the  range of oxygen  demand
conditions  expected, the engineer  must design  the
various  components  of  the aeration  system to meet
these conditions. These components include  the
blowers, air piping and appurtenances, and diffusers.
The first two components are discussed later in this
chapter.

All fine pore diffusers have an allowable range of unit
airflow rates that can be  applied to the diffuser. These
allowable airflow  rates  depend on  diffuser size,  as
discussed in Chapter 2.  For example, allowable airflow
rates  for perforated  membrane tube  diffusers  are
generally 0.5-4.7 Us (1-10 sefmj/diffuser. Similarly, for
ceramic discs (nominal  23-cm [9-in]  diameter) and
domes (nominal 18-cm [7-in]  diameter), the allowable
rates are generally 0.2-1.2 Us (0.5-2.5 scfm)/diffuser.
These allowable ranges of airflow rates  offer turndown
operational  flexibility of  4:1  to  10:1. For  many
systems,  these  will be adequate  to cover  the
anticipated  range in oxygen  demand.  For  systems
where significant  growth is anticipated, the diffusion
system  should be designed to meet initial diurnal and
seasonal  variations  but may  or may  not satisfy
ultimate  design oxygen  demands.  In  such cases,
provisions  should be made  for adding diffusers  or
additional basins in  the future  as  plant  loadings
increase.

Careful  consideration should be given  to the desired
airflow range during  design. Testing has shown that
OTE  is  dependent  on airflow rate  per  diffuser,
increasing as the flow rate decreases  (see Sections
2.6 and 3.4). This  performance  characteristic  may
tempt engineers  to  design  fine  pore systems  to
operate  at  very  low unit airflow  rates.  Although
favorable in terms of oxygen transfer, this practice can
lead to operational problems.

At low airflow rates, uniform air  distribution across the
entire diffuser surface may be difficult to obtain. Also,
at low diffuser airflow rates, the headless  across the
control orifice could be <25 mm (1  in) w.g., requiring
a change to different size orifices  to balance airflow
throughout the system. In any case, if either the entire
surface  or  portions  of  individual  diffusers are  not
discharging  air, foulant  deposition  can  begin,  which
                                                  116

-------
could then  lead  to  premature fouling  of  the  entire
system.

Both the type and arrangement of diffusers should be
selected based  on  anticipated operating conditions.
Generally,  the  most  cost-effective  designs
approximately match the numbers of diffusers to the
required oxygen transfer  rates  in the  basin.  For
example, more diffusers are usually provided at the
upstream end of a plug  flow  basin  than  at  the
downstream end. Diffuser arrangement is discussed in
more detail later in this chapter.
5.4.1.3 Reliability
The reliability  of fine pore  diffusion  systems  is
determined by several factors, including  maintenance
requirements and mechanical integrity.  Maintenance
requirements  are discussed in  the  next section.
Mechanical integrity  is  best determined  by  the
performance record of equipment in full-scale service.

Many  types  of  fine  pore diffusers  are currently
available.  However,  relatively  few have received
widespread use in North  America, so obtaining full-
scale reliability  information on  some  of these units is
difficult at  present. Some of these  units have been
used in other countries for several years. Transferring
this experience to North American designs should be
done  with caution,  however.   Different  design
practices,  wastewater characteristics,  and  plant
operating  methods must be considered in evaluating
true equipment  performance.

In  reviewing  a  diffusion system  for  mechanical
integrity,   each  of   the  components  should  be
considered.  Critical  components  are the diffuser
material,  diffuser supports,  diffuser  connections,
piping  supports, and  submerged  air piping.
Considerations  for  the  diffuser material include
physical and chemical resistance to the wastewater
(including, for perforated membranes, the potential for
loss of  flexibility).  Designers  should  incorporate
mounting  details  that minimize  buildup of stringy
material on diffuser piping. The diffuser supports and
connections should be able to withstand  the range of
stresses that will  occur  both  during installation and
operation.  For  example,  tube-type diffusers  will be
subject to bending and relatively high stresses at the
point  of connection  to the air piping  during normal
operation. The supports and air piping must be able to
resist  the dead  weight  of  the  equipment  during
installation as  well as the buoyant forces of  the
system in normal operation  (as  well  as keep  the
diffusers nearly level).

The designer should anticipate that some parts of the
system will eventually  fail because of  damage  or
normal  deterioration.  The design, therefore, should
permit portions  of the  system to be isolated for repairs
with minimal disruption in normal plant operation.
5.4.1.4 Maintenance Requirements
O&M requirements and procedures for fine pore
diffusion  systems  are  covered  in  Chapter 4. This
section  summarizes O&M considerations  that  the
designer  should be aware of in selecting diffusers for
a specific application.

All fine pore diffusion  systems require  maintenance.
The  amount of routine maintenance  required  will vary,
depending  on wastewater characteristics,  type  of
treatment process, and characteristics of the specific
diffuser.  Maintenance  is  required  for  two  primary
reasons:  to control diffuser fouling (and thus maximize
OTE), and  to replace diffuser components when they
deteriorate.  Maintenance requirements  are  generally
greater in high-rate systems than in  low-rate  systems.
Maintenance requirements  can  increase  as  the
fraction of industrial waste increases.

To maximize OTE and  minimize  costs, fouling must
be controlled. As fouling progresses, headless across
the  diffuser pores  may  gradually increase,  thereby
increasing  blower  energy  requirements.  Gradual
fouling of fine pore diffusers  is not uncommon and
should be considered in designing the aeration system
by allowing for  a  moderate  increase  in  headless
across the  diffusers.  Typical  designs allow  for a
headloss increase  across  fouled diffusers  prior  to
cleaning  of 3.4-10.3   kPa (0.5-1.5 psi).  Headloss
increases  beyond  the design  allowance  indicate
excessive fouling, and  periodic maintenance to control
this  condition should  also be anticipated  by  the
designer.

Provisions can be  made by the designer to manage
fine  pore diffuser fouling. Good preliminary treatment
to remove  most of the fibrous material and/or high-
density suspended solids in the influent wastewater is
appropriate  when  fine pore  diffusion   systems  are
used. This would usually include screening (maximum
of 12.7-mm  [0.5-in]  openings)  and  grit  removal.
Providing a  system that  allows routine  monitoring of
fouling should be considered.  Since fouling  normally
results in increased headloss across the diffuser,  the
ability to  monitor  changes in air  pressure in the air
distribution  piping can aid in indicating  the degree of
fouling that  has occurred  (see  Section  4.2.3.2).
Systems  that  directly  monitor  headloss across  the
diffusers have also been used (see Figure 2-9).

Short-term  increases  in  airflow through  the  diffuser
(air bumping) can remove some deposits. Should  the
designer elect to  incorporate provisions  for  air
bumping, the aeration  system  should be designed to
allow increased airflow to be applied independently to
each aeration  zone. The  aeration basins should  be
arranged  to allow  isolation and rapid  dewatering  of
each basin  to permit  the diffusers to  be physically
cleaned in  place.  Access to a source of plant water
that  can  deliver a high flow  at  reasonable  pressure
(i.e.,  approximately 415  kPa  [60  psig]) should  be
                                                  117

-------
provided  so that  ihe diffusers can  be  periodically
sprayed  to  remove  surface  deposits. Other in-basin
cleaning methods  include periodic acid gas cleaning,
acid washing, and chlorine washing.  Aeration system
materials  of construction that have adequate chemical
resistance to these  cleaners should be selected and
specified by the designer.

Fine  pore  diffusers  may  be  subject  to  gradual
deterioration, and  eventual  replacement  or
rehabilitation should  be anticipated. This deterioration
may be  due to buildup  of inorganic  materials within
the diffuser  that cannot be  removed  by moderate
cleaning procedures or to breakdown of  the diffuser
matonal  itself. The rate of diffuser  deterioration
depends on  wastewater characteristics and the type
of diffuser (see Section 3.3.3.6). The useful  service
life of a diffuser is generally considered to have been
reached when  the diffuser has deteriorated to a point
that the cost of replacing the diffuser  will be offset by
the reduced operating cost brought about  by new
diffuser performance.

Present worth  cost analyses are appropriate for both
selecting diffusers  and  evaluating   the  cost
effectiveness  of  diffuser replacement.   Economic
analysis of fine pora aeration systems is discussed in
Chapter 7,

5.4,2 Basin Arrangement
Proper arrangement of the aeration  basins  and the
dilfusers within the basins are important to provide the
degree  of wastewater treatment intended,  maximize
OTE, and facilitate  system  maintenance. Important
design considerations  include basin  inlet conditions,
wastewater  and airflow  patterns within  the  basin,
ability to isolate and  dewater individual basins, access
lo the diffusers within the basins,  and availability of
plant water.

The arrangement/configuration of the diffusers within
the basin affects wastewater flow patterns. Diffusers
should  be  arranged  to provide adequate  mixing
throughout  the basin.  Typical  basin  and  diffuser
arrangements are shown in Rgure 2-8.

The distribution of  influent  wastewater  and  return
sludge flows to the inlet end of the aeration basin(s)
should be considered  carefully. Depending on basin
Size and configuration, it  may be advisable  to
distribute these flows,  possibly through baffling,
across the entire width of the basin. Such distribution
may minimize  localized  high velocity gradients and
poor initia! mixing in this zone of the aeration basin.

Provisions should be made for partially filling the basin
without allowing the incoming flow to cascade directly
onto the diffusers and in-basin piping. Both the piping
and diffusers can  be damaged  if a large  volume of
water is allowed to drop directly onto them.
Dry  isolation  of  each  basin  will  be  necessary
periodically to allow maintenance of the in-oetsin piping
and diffusers. A drain system that permits each basin
to be  dewatered  in  a reasonable period  of  time
(normally 8-24 hr) should be provided. The basin floor
should  be  sloped adequately  to allow  complete
drainage to  occur  without ponding  and  to  facilitate
easy removal of residual solids. One arrangement that
has  been  used effectively  is to  construct  a drain
trough along one wall  of the basin, with the basin floor
sloped to drain  to the trough and the trough sloped to
drain to a sump or dewatering manhole.

Diffusers should be  arranged  within the basins ,to
allow  space for  walking and access to them. Access
to the  diffusers  is  necessary both  for installation  and
maintenance. Spacing between diffusers on adjacent
laterals, between grids, and  between each basin  wall
and  adjacent  diffusers  should  be  examined. A
minimum clear  walkway space of 51  cm (20 in) is
usually adequate.

Basin and diffuser cleaning require water at moderate
pressure (approximately 410-690 kPa [60-100 psi]) at
the nozzles. Hydrants with  38-mm  (1.5-in) hose
connections  are usually adequate. These hydrants
should  be  placed  at  frequent intervals  and  easily-
accessible  points around  the  basins; 61  m  (200 ft)
between hydrants  is  generally adequate. Either  tap
water or secondary effluent can be used.
5.4.3 Airflow Distribution

5.4.3.1 Control
Oxygen demands  vary in  the aeration basin  both
temporally and spatially, as previously discussed. Well
designed  systems provide  sufficient  flexibility  to
reasonably match  aeration rates to system oxygen
requirements. This permits the system to be operated
without stressing the  microorganisms  and  without
excessive aeration  (and the associated energy cost):

Aeration control systems can be fully manual or highly
automated,  or have  both  manual and automatic
features. Until recently, fully manual control'systems
were more common, primarily  because of the  lack of
reliable  DO  monitoring instrumentation.  With
continuing  improvements in   this area,  reasonably
reliable instrumentation is now  available and automatic
control of aeration systems is feasible.

With  a manual control system,  operators  make
periodic measurements of DO concentrations in the
aeration basins and adjust airflow rates accordingly. In
contrast, automatic control  systems monitor aeration
rates and DO concentrations  using primary  sensing
elements and  automatically adjust air  delivery rates
based on  the signals received.  Aeration  control
methods are discussed in detail in Chapter 6.
                                                  118

-------
Even with automated DO control, temporal and spatial
variations  in oxygen  demand cannot  be  exactly
matched. Further,  moderate  variations in DO within
the basin (± 0.3-0.5 mg/L) can be  tolerated without
either  adverse effects on  the treatment process  or
undue increases  in  operating costs.  Arranging the
diffusers in  grids,  as described earlier  in this chapter,
and  making  periodic automatic adjustments in airflow
rates to these  grids (at  5-  to 30-rninute intervals)
based on DO concentrations at one or two locations in
the aeration basin  normally results in a cost-effective
control system.  The designer must find a reasonable
compromise between  controlling the  aeration  rate
throughout  the  basin to minimize power costs and
maintain the biomass in a  healthy  condition,  and
minimizing the complexity, cost, and maintenance  of
the control system.

Regardless  of  the  method  of control  used, the
aeration  system should be arranged to allow aeration
zones to be isolated and airflow  to be increased  to
specific  zones  to  facilitate air bumping  if deemed
necessary  (particularly  for  perforated  membrane
systems). The designer should consider  the number
of valves to be manually operated and other changes
that will  have to be made by the operator to allow  air
bumping of  a particular aeration zone.

5.4.3.2 Airflow Measurement
The numbers and locations  of airflow measurement
points depend on  process control requirements and
energy conservation  objectives. For  most facilities,
instrumentation should be provided to allow the total
airflow to  the  aeration  basins  to  be  monitored
continuously. Provisions should also  be made to allow
the airflow to specific aeration basins  and to aeration
zones within each basin to  be monitored. This  may
include  permanent flow  meters,  access points  for
connecting  portable flow meters, or  a combination of
both.  Where  automatic   DO  control  is  used,
instrumentation is usually installed to permit airflow to
individual basins to  be monitored continuously, with
access  points to  allow  manual  measurement and
adjustment  of airflow to individual aeration grids. Even
for systems that will not have automated DO control, it
is good  practice to  provide  access points in  the  air
piping where a portable  airflow measurement device
can be attached.

As with  other mechanical components of the aeration
system,  flow meters should be sized to allow accurate
measurement of  airflows over the  entire range  of
design conditions  (including  minimum flow at  plant
start-up  and  maximum  flow  at design conditions). If
the  range  of flows is  such that a single flow meter
cannot handle the entire range of  flows,  provisions
should be made to enable system components (or the
entire system) to be replaced in the future.

Airflow  can  be  measured  with several types  of
devices. The most commonly used devices are based
on  either  differential  pressure across  a  control
element  or mass  flow.  General  features of  each
device are  described in Table 5-2.
Table 5-2,   Properties of Airflow Measurement Devices

     Device
 Description and Principles
      of Operation
Range-
 ability  Accuracy
 Differential
 Pressure

  Averaging
  Pilot Tube
 Venturi Tube
  Flow Tube
  Oriifoe Plate
Insertion probe with         3:1   2-5% of
multiple upstream ports           full scale
and one downstream port;
difference between
upstream and downstream
pressures proportional to
square root of flows; low
headless
Flow-through device with     3:1   !%of
converging and throat            actual
sections; constriction             flow
causes increase in velocity
and pressure drop
proportional to square root
of flow; moderate headloss
Similar to Venturi tube, but   3:1   1 % of
smaller in size and lower          actual
headloss; moderate              flow
headloss
Thin plate with opening "    3:1   2-5% of
placed perpendicular to           lull scale
flow; orifice causes
increase in velocity arid
pressure drop proportional
to square root of flow;
moderate headloss
Multi-blade rotor placed      10:1  0.5% o(
with axis of rotation              actual
perpendicular to flow;             flow
rotational velocity
proportional to flow; high
headloss
Heated element and        10:1  1 % of full
unheated element;              scale
lemperaiure differential
proportional to log of flow;
low headloss
In designing the air piping system, the designer should
consider the installation and operating requirements of
the specific airflow measurement devices to be used.
In particular, sufficient straight lengths of piping must
be provided  upstream  and  downstream  of  the
permanent device or portable device connection point
to  allow  accurate  measurements to  be  made. A
minimum upstream length equal to 10 pipe diameters
and a minimum downstream length equal to  5 pipe
diameters should be provided,

5.4.4 Diffuser Cleaning and Maintenance
Chapter 4 discusses diffuser O&M  requirements in
detail.  This subsection discusses  provisions that
should  be made by the designer to facilitate  diffuser
cleaning and maintenance.
  Turbine Meter
  Thermal Mass
  Flow Meter
                                                    119

-------
Because  diffusers  require  periodic  cleaning  and
maintenance, the  system  should  be  designed  to
facilitate these functions without major interruptions in
plant operation. Appropriate design provisions depend
on  the cleaning  methods  to  be  used. Types  of
cleaning methods  include  those  that  are  process
noninterruptive, process interruptive with diffusers in
place,  and  process  interruptive  with diffusers
removed.

Noninterruptive cleaning methods require the ability to
isolate a portion of the  air piping system to introduce
the cleaning  material (usually either a  high volume of
air or acid gas). If  the  diffusers are to be chemically
cleaned, materials of construction must be selected to
be  chemically resistant.  Materials  selection  is
discussed later in this chapter. The acid gas cleaning
method involves  proprietary  technology,  and special
appurtenances are required (28,29).

For the  process  noninterruptive  methods,  special
considerations  include  the  provision of local airflow
control valves and flow meters for flexing of perforated
membrane dilfusers. Ceramic  and  plastic diffuser
systems should be designed to be compatible  with
acid gas injection. To facilitate in-situ gas cleaning, a
tap, with shutoff valve, is provided on the drop pipe to
each grid. Portable acid feed  systems can be provided
for testing. A portable rig consisting of gas cylinders,
regulators, flow meters, and other  appurtenances is
then moved  throughout the system and temporarily
connected to the tap on each grid. A permanent feed
system can bs provided later if the process proves to
be functionally and  economically^  attractive.  In some
cases, the tap consists of a nozzle that vaporizes the
gas as it enters the air manifold.

Process interruptive cleaning  methods require  the
ability  to  isolate  aeration  basins.  Generally,   dry
isolation  is  required  so  that the basin  can  be
dewatered and the diffusers either cleaned in place or
removed from  the  basin for cleaning. Basin design
features to  facilitate dry  isolation  were  discussed
earlier in  this  chapter. Methods for cleaning  the
diffusers  in  place  include acid washing, steam
cleaning, and low-pressure hosing. If the diffusers are
to bo  chemically  cleaned,  chemically  resistant
materials of construction must be selected. Guidelines
for placement of  hydrants  for low-pressure hosing of
diffusers were given earlier in this chapter.

Facilities can be provided to simplify monitoring of the
condition of the diffusers by plant operators. Diffuser
fouling will result in an increase in headloss within the
aeration system. This increase in headloss may result
in an increase in system pressure of 3-5 kPa (0.4-0.7
psi). This  change  may be  imperceptible on typical
blower  discharge  pressure meters.  Systems  that
directly monitor the headloss across the diffusers (i.e.,
DWP), however, will provide  the operator with a more
sensitive measure of the degree of diffuser fouling.
As discussed in Chapter 2, systems for monitoring the
condition of the diffusers include pressure taps for
measuring  DWP  and  pilot  headers.  Pressure  taps
allow  monitoring of the headloss across the diffuser,
which can help the operator determine when cleaning
is  required. Figure 2-9  provides a  sketch  of  the
various  components of a typical headtoss monitoring
system.  The  individual pressure taps  are  usually
connected to lengths of flexible tubing  that run to the
surface. Common  practice is to  place  several  flexible
lines inside  a section  of rigid pipe that  serves  as  a
means of support. The individual lines are then  fixed
to  a board or panel attached to the  railing or basin
wall. All lines under pressure should include a  shutoff
valve. Multiple units are usually  provided so that the
headloss at  various points throughout the system can
be  monitored. Chapter  2  discusses  operational
features of these systems.

A fine pore aeration system  should include provisions
for  removing liquid that may accumulate inside the
pipe. This liquid can enter during periods when the air
is turned off or during normal operation  when moisture
in the incoming air condenses. Accumulation of liquid
within the piping may  increase headloss through the
piping and cause system performance to deteriorate.

Another approach is to  provide a separate condensate
removal system.  This  system usually  consists  of  a
short  section of pipe or tubing extending down  near
the bottom of the air distribution pipe at a low point in
the system.  In some  cases, a  collection  sump is
provided  in one  corner  of  the grid to  aid  in
accumulating the  liquid. The tubing is attached  to  a
section of rigid piping extending  to the basin surface
where a shutoff valve is placed.

5.4.5 Diffuser Installation
Installation  of fine pore diffusers  requires  special
precautions. These precautions include ensuring  that:
1)  the air supply  system is  properly prepared  before
the diffusers are installed, 2) the diffuser components
were not damaged during shipping, and 3) they are
not damaged during installation.

To  minimize the  potential for air-side  fouling  of the
diffusers, the air  header system  should  be carefully
cleaned to  remove construction  debris  (dust, metal
shavings, oil, etc.) before the diffusers are installed.
The  same  considerations  apply  whenever
modifications are  made to any part of the air  supply
system after the diffusers are in place.

Most diffusers have relatively fragile components that
can be damaged during handling, and they should be
handled accordingly. When installing the diffusers on
the   header  system,   the  manufacturer's
recommendations should  be  carefully  followed.
Over-tightening  of  certain  components,  particularly
plastics,  can lead  to  failure of these components.
Overtightening  or  undertightening  can  result  in air
                                                  120

-------
 leaks  and  maldistribution of  air,  thereby reducing
 system OTE.

 The  diffusers  must  be  accurately  leveled  after
 installation  on  the  header  system.  Differences  in
 vertical placement of the diffusers will lead to uneven
 air distribution and reduced OTE. Provisions must be
 made in  the design of the header  system to allow
 leveling,  including  adjustable  pipe supports  and
 flexible or adjustable pipe joints.

 A  manufacturer's representative and  the  design
 engineer  should be on site during initial installation.
 Both should  inspect  the work  as it  progresses  to
 ensure that proper procedures are followed, including
 connecting  the diffusers to the air  piping and  final
 leveling of the air piping and diffusers (for flat plates).
 In  addition, the manufacturer's representative  should
 certify that the completed facility was  constructed, and
 the equipment  installed,  in  accordance  with the
 manufacturer's recommendations.

 5.4,6 Specifications
 For a completed system to meet  the design intent, the
.specifications  included  in  the contract documents
 must be  thorough  and comprehensive. Some of the
 items  that should normally be addressed include the
 following:

 •  Component design  and quality assurance/quality
    control  (including diffusers,  connectors for
    attaching diffusers  to  air  piping, and  pipe
    supports)

 »  Nature  of wastewater  and  its  constituents,
    particularly  the  percentage  of  industrial
    wastewater (by loading and flow)  and specific
    types of industrial wastes

 »  Materials  of  construction (including diffusers,
    connectors, piping, and pipe supports)

 *  Shop testing  (including  component  sampling
    during  manufacture  [see  Chapter 2] and  clean
    water performance testing)

 »  Installation (including preparation  of the  air supply
    system prior to diffuser installation, assembling  of
    components to  avoid  damage, and requirements
    for leveling of diffusers)

 •  Representation of manufacturer during installation
    (including requirements to inspect,  provide quality
    control, and certify the completed installation)

 *  Operating conditions (including estimated  values
    of a and  F, maximum MLSS concentration, range
    of  wastewater and air temperatures, atmospheric
    pressure, and diffuser submergence)
*   Performance  requirements  (including  range  of
    required  SOTRs by  aeration  zone,  range  of
    allowable airflow  rates  by  aeration zone,  and
    minimum allowable number of diffusers by zone)

•   Physical  testing of  completed  installation
    (including leak testing, uniformity of air  release,
    and verification of level installation)

«   Performance  testing  of completed installation
    (including test and  data analysis, interpretation,
    and reporting  methods  to  be followed that  are
    consistent with the ASCE  Standard  (30),  and
    penalties if performance requirements are  not
    met.)  Normally,  only  clean water  performance
    testing  is required  because  of  the  potential
    variability in  wastewater  characteristics and  the
    significant impact this  can have  on diffuser
    performance.  In  addition,  the  designer  should
    consider the costs of  requiring  extensive
    compliance testing in relation to the initial cost of
    the aeration system and present  worth of future
    operating costs.

The above specification items  focus  on  clean water
performance. However, the  designer  must  be aware
that fine pore diffuser performance in process water is
normally significantly  lower than in  clean water.  The
specific diffuser,  wastewater  characteristics,  and
treatment  system  configuration  all  have significant
impact  on  a  and  F, which  together  can reduce
performance under process  operating conditions  to a
fraction of  clean  water transfer  efficiencies.  The
designer should provide sufficient operational flexibility
to enable the system to be operated efficiently over a
range of conditions, including ranges of a  and F.

5.4,7 Retrofit Considerations
An  existing  diffused  air  aeration  system  must  be
carefully evaluated when considering  replacement of
that system  with  a  alternative fine  pore  diffusion
system.  Many of the  considerations discussed
previously for diffuser selection  also apply  to retrofit
situations. In addition, the designer may be  limited  by
the  existing  blower and  air  piping  system  and basin
geometry.

Pressure requirements for the new  fine pore system
may differ from those of the existing diffusers. Again,
diffuser fouling  will  increase headless through  the
diffusers. The designer should verify that the existing
blower system has adequate capacity to deliver  the
required volume  of air to the diffusers under worst-
case, pressure loss conditions.

The potential for air-side diffuser fouling may increase
in retrofit  situations.  Deterioration of metal  air piping
may have  resulted  in  a  substantial  amount  of
corrosion products in the piping. The potential effects
                                                   121

-------
of this  material on diffuser performance  should be
considered.
because the mechanical  equipment would not  be
exposed to the weather while hot in use.
Both  diffuser  and  process  performance  can  be
significantly affected by basin geometry and basin flow
patterns.  Both  wastewater  flow patterns and  airflow
patterns  should  be considered. Fine  pore diffuser
OTEs are usually substantially higher for systems with
high diffuser  densities and  low  airflow rates per
diffuser than for systems with low diffuser densities
and high airflow rates per diffuser. However, there are
likely to  be optimum  densities  that  are  diffuser
specific.

OTE performance testing under typical  process water
operating  conditions should  be conducted  on the
existing installed aeration equipment to establish  an
accurate baseline against which the  need for a fine
pore retrofit and  the  eventual performance  of the
retrofil system can be judged. Further, the opportunity
also exists  for  evaluating   process water OTEs  for
candidate fine   pore diffusers  in  a  portion  of the
existing facility  or using the test  header apparatus
described in Chapter 2 (see Figure 2-10).
5.4.8 Air Diffusion System Design Example
Example 5-10 presents an approach to the design of a
fine pore air diffusion system. This example illustrates
some  of  Ihe  design considerations  discussed in
Sections 5.4.1 through 5.4.7. In an actual design, it is
likely that  several  unique features  and site-specific
factors  (including control  system requirements, O&M
slaff expertise, owner preferences, etc.) will have to
be  considered.  These  types  of  factors  are not
considered in this example.

5.4.9 Flexibility of Design
Tho  preceding example was based on oxygen transfer
requirements  in  the  design year (20  years  in the
future). As discussed in Example 5-8, the design must
have sufficient flexibility to handle temporal variations
in loading and oxygen demand, including hour-to-hour,
day-to-day, and year-to-year variations.

Providing  the flexibility  necessary to handle year-to-
year variations can be accomplished in several ways.
Where the design period is relatively long and  steady
growth  is anticipated (as  in  the preceding  example),
Ihe designer/owner could choose to build the facility in
phases. In this example, two  basins could be provided
in the first  construction phase, with either one  or two
additional  construction phases, as  necessary,  in the
future. Another option is to construct all facilities in the
first  phase, with provisions for operating only  two of
the four basins during the early life of the facility. A
third option is  to construct all the basins,  buildings,
and  major  yard piping in the first  phase,  and stage
construction  of the mechanical equipment (blowers,
in-basin piping, and diffusers), as necessary. This may
be the more  practical option of the last two methods
The specific approach taken  depends on  several
factors, including funding, projected growth patterns,
and owner preference. A cost-effectiveness  analysis
of several alternatives will be helpful in  choosing an
acceptable plan. In  completing this analysis,  it  is
important to take into account the higher capital cost
of facilities that are constructed in separate phases as
compared with single-stage construction.

Regardless of the approach  selected for constructing
the facilities necessary to accommodate year-to-year
variations  in  loading  and   oxygen requirements,
flexibility to allow economical  operation throughout the
design life of the plant must be provided. For example,
if  more basins  and blowers  are  installed than  are
required to handle initial loads and  oxygen demands,
capability should be provided  to operate  only as many
basins  and  blowers as needed  while  keeping  the
remaining ones out of service.

In some situations,  substantial variations  in loading
may occur on a seasonal basis. Again, the designer
should  consider providing the  ability to change  the
number of  basins  and blowers  in service  on  a
seasonal basis to handle these types  of  variations.
Providing the capability to operate in more than  one
mode, e.g., plug flow or step feed, is desirable.

Flexibility for handling  hour-to-hour  and  day-to-day
variations  in loading and oxygen  demand  should  be
accommodated by providing  the  capability to adjust
airflows to  various basins and  zones in  response to
these changes. Factors that should be considered in
selecting and  designing  DO  control  systems  are
discussed in Section 5.4.3.1 and Chapter 6.


5.5 Air Supply System
The air supply system delivers atmospheric  air or high
purity  oxygen  to  the  air  diffusion  system. While
selection and  design  of the diffusion system  often
receive more attention, care is necessary in designing
the air supply system  to ensure that overall  process
objectives  are  met  and  power  consumption is
minimized.

This  section  discusses  some of  the  key
considerations for design of  the air supply  system. It
is not  a detailed  guide for the  mechanical design of
the  various  components.  Rather,  it  presents   an
overview of the important elements of the system.

The air supply  system  consists  of  three basic
components:  air piping, blowers,  and air filters  and
other conditioning equipment (including  gas injection
diffuser cleaning systems). The air piping conveys air
from the blowers to the diffusers.  The  blowers  are
designed to develop sufficient pressure  to  overcome
                                                   122

-------
                         Example S-10.  Design of Fine Pore Aeration System
Consider the activated sludge system from Examples 5-4, 5-8,  and 5-9.  Design an air diffusion system to
meet the oxygen transfer requirements developed in these examples.

Example  5-8 addressed variations in process  oxygen  requirements from  plant  start-up through  ultimate
design loading.  This example presents calculations for the ultimate condition only. Design considerations to
facilitate operation during  initial operating years were introduced in Example 5-8 and are further discussed
later in this chapter.

A plant schematic was presented in Figure 5-7. The basic design data for the plant are as follows:


                                     BOD5 Loading  ib/d    Wastewater Temperature °C
                    Minimum month

                    Average month

                    Maximum month
 5,500-

 6,600

 7,700

12,800
10

15

25

25.
                       Peak day
                 Design Dimensions: 4 aeration basins, each 130 ft long by 23 ft wide
                . SWD = 15ft (diffuser submergence = 14 ft)
                 Wastewater Flow = 5.3 mgd
                 SRT = 4 days; HRT = 6 hr
                 Plant Elevation  = 1,000ft        •                                  •    .           -
                 Minimum Air Temperature = -9°C (15°F)
                 Maximum Air Temperature = 40°C (105"F)

A 4-step approach will be used to design the fine pore aeration system-


Step 1: Determine field oxygen transfer rates (OTRfS) by aeration zone and operating  condition. Set these
        OTRfS equal to their corresponding AORs (i.e., OTRfS must satisfy corresponding AORs).


System OTRfS were calculated by operating condition and aeration zone {or grid) in Example 5-9 and are
summarized below:

                                                 System OTR, (Ib/d)          *
             Zone   Peak Day  Max. Month   Avg. Month Nitrifying  Avg, Month Non-nitrifying    Min, Month'
               1       6,187       5,430

               2       4,053       4,147

               3       4,920       2,010

             Total    12,160      11,588
      4,613

      3,513

      1,559

      9,685
  2,904

  1,804

   704

  5,412
2,109

1,191

 275

3,575
These same OTRfS for one basin are as follows (these will be carried through the next phase of the design):


                                                  Basin OTR, (Ib/d) .             .             .
             Zone   Peak Day   Max. Month   Avg. Month Nitrifying  Avg. Month Non-nitrifying    Min. Month
               1       1,547       1,358

               2       1,013      -1,037

               3        480        503

             Total     3,040       2,898
      1,153

       878

       390

      2,421
   726

   451

   176

  1,353
 527

 298

  69

 894
                                                                                              (continued)
                                                   123

-------
                  Example S-10.  Design of Fine Pore Aeration System (continued)
Step 2: Convert OTRf values to standard oxygen transfer rate (SOTR) values to account for the effects of
        process operating conditions.

OTRj is related to SOTR as follows (see Reference 31 for a more detailed discussion of the development of
this equation and precise estimation of the various coefficients):

           OTR| = aF (SOTR)8T-2Q (O i B C"mZQ - C)/C'oa2o                                       (5-5)

where,
   OTRj = oxygen transfer rate under process conditions, Ib/hr
   a     ~ (process water Ki_a of a new diffuser)/(clean water KLa of a new diffuser)
   F     = (process water K|_a of a diffuser after a given time in serviee)/(Ki_a of a new diffuser in the same
           process water)
   aF    = used as a product  for this design example  with  design ranges for each  aeration zone (see
           detailed discussion in Chapters 3 and 4).
   SOTR = oxygen transfer rate under standard conditions (20°C, 1 aim, C = 0 mg/L), Ib/hr
   0     = 1.024
   T     * process water temperature, C
   OT-20 =Ku3/KLa20
   Ki.a   = apparent volumetric mass transfer coefficient in clean water at temperature T, 1/hr
   O     = pressure correction for Cfm ~ PyPs (approximation that excludes effect of de at relatively low
           water depths)
   PU    = field atmospheric pressure, psia
   Ps    = standard atmospheric pressure (14.7 psia or 1.0 aim at 100 percent relative humidity), psia
   t     = temperature correction for C*« = CVC'^ao = cyc*S20
   fl     = (process water CA
-------
                  Example 5-10.  Design of Fine Pore Aeration System (continued)
    0*0,20 =   10.5 mg/L (from clean water testing)

    C     =   2.0 mg/L for average month nitrifying condition and minimum month; 1.0 mg/L for average
               month non-nitrifying condition and maximum month; and 0.5 mg/L for peak day

For an elevation of 1,000 ft, the atmospheric pressure, P^, is 14.3 psi (from Appendix C, Figure C-1). Thus:

    Q  = 14.3/14.7 = 0.97

From Table  C-1, the values of i are:

    @10°C, ^ =  C"S10/C*S20 = 11-29/9.09  = 1.24
   :@15°C,i =  C"S16/C*S20 = 10.08/9.09  = 1.11
    @20°C, t =  1 (by definition)
    @25°C, i =  C"S25/C"S20 = 8.26/9.09 = 0.91

For this example, Equation 5-5 becomes:

    (OTR,/SOTR) = aF (9.98i - C)(1.024T-20)/1Q.5

The following values were selected for the design:

                                    T(°C)                i              C(mg/L)
             Peak-day

             Maximum month

             Average month,
             nitrifying

             Average month, non-
             nitrifying

             Minimum month
25 '

25'

20


15


10
0.91

0.91

1.00


1.11


1.24
0.5

1.0;, .

2.0


1.0


2.0
Assume the design operating DOs are equal in all three zones.  Lower operating DO values are sometimes
used in the first zone to reduce operating costs. However, this practice can lead to sludge bulking problems
(7). Therefore, relatively conservative values were selected. Even at these DO levels, bulking  due to low
DO may occur (7).

Using the above coefficient values, Equation 5-5 can be further simplified:
               Peak day
               Maximum month
               Average month, nitrifying
               Average month, non-nitrifying

               Minimum month
          OTR,/SOTR  =

          OTR,/SOTR  =

          OTR,/SOTR  =

          OTR,/SOTR  =

          OTRi/SOTR  =
       aF (0.920)

       aF (0.867)

       aF (0.760)

       aF (0.852)

       aF (Q.780)
                                                                                          (continued)
                                                 125

-------
                  Example 5-10.  Design of Fine Pore Aeration System (continued)
Values of OTRj/SOTR can then be computed for each zone:

                                                 Zone 1    .Zone 2  .  Zone 3
                        Peak day
                            aF                    0.20      0.30      0.60
                            OTR,/SOTR             0.18      0.27      0.55
                        Maximum month
                            oF                    0.20      0.30      0.60
                            OTR,/SOTR             0.17      0.26      0.52
                        Average month, rtiinfying
                            aF                    0.25      0.40      0.70
                            OTR|/SOTR             0.19      0.31       0.54
                        Average month, non-nitrifying
                            aF                    0.25      0.40      0.70
                            OTR,/SOTR             0.21       0.34      0.60
                        Minimum month
                            aF                    0.30      0.50-      0.80
                            OTR,/SOTR             0.23      0.39      0.62


Using the above values of OTR(/SOTR, the following SOTRs for each aeration zone and  process condition
can be generated:
                	Basin SOTR (Ib/d)	
          Zono   Peak Day   Max. Month   Avg. Month Nitrifying  Avg. Month Non-nitrifying    Min. Month
            1      8,594       7,988          6,068          -   .  3,457              2,291

            2      3,752       3,988          2,832               1,326                764

            3       873        967            722                293                111

          Tola!   13,219      12,943          9,622               5,076              3,166


Step 3: Calculate required field standardized  volumetric airflow rates (qs) by zone and the numbers of
        dlffusers necessary to handle these rates.

Al this  point,  the designer needs to determine the performance  characteristics of the  fine pore  aeration
device that was selected preliminarily. For this example, data  shown in Figure 5-9 will be used  for design
purposes.  Figure  5-9 presents transfer performance (SOTR) as a  function of  both  diffuser  density
(number/100 sq ft) and  unit airflow rate  (scfm/diffuser). These data are applicable to full  floor coverage,
ceramic disc/dome  grid  combinations.  In actual  practice,   such  data should  be  obtained  from  the
manufacturer of the specific commercial fine pore diffuser selected for the design.

Final selection of the fine pore diffuser should be  based on evaluation of several alternative devices  and
designs. Characteristics  of these devices  and designs  affect both system performance   and cost.
Accordingly, during  this  evaluation the  designer  should usually  consider equipment   costs,   equipment
compatibility/suitability, maintenance  requirements,  and equipment reliability over the  useful  life of  the
treatment system.

The following equation relates qs to SOTE and  SOTR:

    qs = (0.04 scfm/lb 02/d) (SOTR)/(SOTE)                                                     (5-6)

                                                                                          (continued)
                                                 126

-------
                  Example 5-10.  Design of Fine Pore Aeration System (continued)
Zone 1:
The first zone will need  to  satisfy the highest oxygen demands; therefore, it will necessarily  have the
highest diffuser densities and likely use the highest unit  airflow rates. Thus, as an initial attempt (this will be
an iterative process), the clean water performance for a diffuser density of 45 diffusers/100 sq ft was used
(from top curve, Figure 5-9). An airflow rate of 2.5 scfm/diffuser is selected, which is higher than typical but
considered acceptable for this application.  From Figure 5-9, an SOTE of 28 percent is estimated for Zone 1.

1.  Peak Day SOTR Requirements Control  the Design

    qs= 0.04(8,594)7(0.28) =  1,228scfm
    (1,228 scfm)/(2.5 scfm/diffuser)  = 491 diffusers

2.  Check Diffuser Density

Diffuser Density   =   (100)(491)/(43.3 ft)(23 ft)
                  =   49.3 diffuses/100 sq ft
                      [Assumed density  of 45 diffusers/100 sq ft yields  slightly lower efficiency  than will
                      occur at 49.3 diffusers/100 sq ft. Use 491 diffusers for conservative design.]

3.  Check SOTR at Minimum Airflow of 0.5 scfm/diffuser

    From Figure 5-9, SOTE = 37 percent at this airflow  rate; therefore:

    SOTRmin air = 491 diffusers(0.5 sefm/diffuser)(0.37)/(0.04) = 2,271 Ib/d

This calculated minimum SOTR is nearly equal to the minimum month oxygen requirement of 2,291 Ib/d for
this zone. Therefore, diffuser turndown should not  control the aeration rate in Zone 1 (however, if all of the
diffusers are installed  initially, during  early years of plant operation, allowable diffuser turndown is likely to
determine the minimum aeration rate. Often, only a portion of the  diffusers are  installed initially with  others
being installed on a staged basis as needed.).

4.  Check Minimum Mixing Requirements

Adequate air must be provided to maintain thorough mixing within the basin. The minimum air requirements
for mixing depend on several factors  (including basin geometry, diffuser arrangement, MLSS concentration,
settling characteristics of the sludge, and basin inlet conditions). Experience indicates airflows that  satisfy
process oxygen requirements will provide  adequate mixing through most of the  aeration basin. However,  in
some  instances, such as towards the  end of plug flow basins, mixing requirements  rather than process
oxygen requirements  may dictate minimum  acceptable aeration  rates.  Accordingly, mixing requirements
should be considered during the design.

Airflows to maintain adequate mixing in  activated sludge aeration basins are usually established from "rule-
of-thumb"  criteria,  as discussed in  Section  5.3.2. A typical  value  is 0.1 scfm/sq ft of basin area (i.e.,
approximately 7 scfm/1,000 cu ft of basin  volume).

Determine  mixing requirement:

    Area per zone  = 996 sq ft

    Mixing  airflow  = (0.1 scfm/sq ft)(996 sq ft) =  100 scfm/zone

Since  the  minimum allowable  airflow rate for Zone 1 is 491  diffusers x  0.5 scfm/diffuser (manufacturer's
recommended  minimum)  = 245 scfm, this value and not the mixing requirement is the limiting design factor
in  Zone 1.

                                                                                          (continued)
                                                 127

-------
                  Example 5-10.  Design of Fine Pore Aeration System (continued)
Zone 2:

1. Maximum Month SOTR Requirements Control the Design

    Assume a diffuser density of 30 diffusers/100 sq ft and an airflow rate of 2.0 scfm/diffuser for Zone 2.
    Then, from Figure 5-9 (second curve from top), SOTE = 27 percent.

    qs  = 0.04 (3,988)/(0.27) = 591 scfm
    (591  scfm)/(2.0 scfm/diffuser) = 296 diffusers

2. Check Diffuser Density

    Diffuser Density      = (100)(296)/(43.3 ft)(23 ft)
                        = 29.7 diffusers/100 sq ft
                          [approximately equal to assumed density]

3. Check SOTR at Minimum Airflow of 0.5 scfm/diffuser

    SOTE = 33 percent (from Figure 5-9)

    SOTR,,lin air = 296 diffusers(0.5 scfm/diffuser)(0.33)/(0.04)
               = 1,221 Ib/d

Since this is higher than the minimum month oxygen requirement of 764 Ib/d  for this zone, the  minimum
month  oxygen requirement will not control minimum airflow to Zone 2.

4. Check Minimum Mixing Requirements

Since the minimum allowable airflow rate for Zone 2 is 296 diffusers x  0.5 scfm/diffuser (manufacturer's
recommended minimum) =  148 scfm, this value and not the mixing requirement (100 scfm) is the limiting
design factor in Zone 2.

Zone 3:
1. Maximum Month SOTR Requirements Control the Design

    Assume a diffuser density of 18 diffusers/100 sq ft and an airflow rate of  1.0 scfm/diffuser for Zone 3.
    Then, from Figure 5-9 (bottom curve), SOTE = 27 percent.

    qsi=  0.04 (967)/(0.27) = 143 scfm
    (143 scfm)/( 1.0 scfm/diffuser)  = 143 diffusers

2. Check Diffuser Density

    Diffuser Density    =  (lOO)(l43)/(43.3 ft)(23 ft)
                      =  14.4 diffusers/100 sq ft
                          [less than assumed density of 18 diffusers/100 sq ft; by extrapolation in Figure
                          5-9, actual SOTE is approximately 26 percent]

    Recalculate the number of diffusers and diffuser density using SOTE  = 26 percent:

    Actual Number of Diffusers Required = 149
    Diffuser Density = 15.0 diffusers/100 sq ft
                                                                                        (continued)
                                                128

-------
                  Example 5-10.  Design of Fine Pore Aeration System (continued)
3. Check SOTR at Minimum Airflow of 0.5 scfm/diffuser

    By extrapolation  in Figure 5-9, SOTE = 28  percent (approximately) at the actual  diffuser density;
    therefore:

    SOTRminajr= 149 diffusers(0.5 scfm/diffuser)(0.28)/(0.04)
                = 522 Ib/d
                                                                          *
This is higher than the minimum month  oxygen  requirement of 111  Ib/d for  this zone. Therefore, the
minimum month oxygen requirement will not control minimum airflow to Zone 2.

4. Check Minimum Mixing Requirements

Since the minimum  allowable airflow rate  for Zone 3 is  149 diffusers x 0,5 scfm/diffuser (manufacturer's
recommended minimum)  =  75 scfm, the  mixing requirement of  100 scfm  is the limiting design factor  in
Zone 3. This equates to approximately 0.7 scfm/diffuser.

Summary:
Aeration rates were calculated for each zone based on the required number of diffusers determined above.
These aeration rates and the  number and densities of diffusers required for each zone and individual basin
and the total aeration system are summarized below for all process conditions:
                        _   .  . .. _                                                      Approx, Diffuser
                        Required Airflow (scfm)	   Approximate     Unit Airflow Rale"      Density
    Zone   Min.   Average Non-nitrifying  Average Nitrifying  Max.    No. Diffusers      (scfm/diffuser)     (No./100 sq ft)
1
2
3"
Basin
System
'245
148
100
493
1,972
432
166
100 '
698
2,792
867
405
111
1,383
5,532
1 ,228
591
149
1,968
7,872
491
296
149
936
3,744
1.8
1.4
0.7
-
-
49.3
' ' . 29.7
15.0
-
-
     For average nitrifying condition.
    This zone is mixing limited under all operating conditions.
At this point, the engineer should review the preliminary system design to identify any potential drawbacks
in terms of probable constructability and operability. Three features of this example design that the engineer
should consider are the high diffuser density in Zone 1, the fact that Zone 3 will be  mixing limited under all
operating conditions, and the wide range of unit airflow rates (scfm/diffuser) in the three zones.  Since it is
desirable for the unit airflow rate (calculated above for the average nitrifying condition) to be about the same
in each zone to minimize headless and airflow control difficulties,  the designer may want to reevaluate the
number of diffusers in each zone. The designer should also consider how the system will be operated from
start-up through ultimate capacity.

The area per diffuser in Zone 1 (2.0 sq ft) is near the minimum acceptable for 7-in diameter diffusers. The
clear walkway  between  diffusers is only about  18 in. If this is  considered inadequate by  the owner or
designer, the design will need to be modified.

Operating Zone 3 under mixing-limited conditions means that the aeration rate will exceed that necessary to
meet process requirements for a large percentage of time. This results in higher operating costs than would
occur if all zones in the basin were operated to avoid mixing limitations.

                                                                                           (continued)
                                                 129

-------
                  Example 5-10.  Design of Fine Pore Aeration System (continued)
Several options are available to address these concerns. One design option is to place fewer diffusers in
Zone 1  without changing the allowable airflow rates per diffuser. This would allow a larger diffuser spacing
in Zone 1. It would also result in more of the system oxygen demand being  passed to Zones 2 and 3. The
design could be modified so that Zone 3 could be operated to avoid mixing-limiting conditions some, or all,
of the time. A drawback to this approach is that low operating DO levels would occur in Zone  1, which
could lead to sludge bulking problems as discussed earlier,

A second design option is to provide the capability to operate the basins in  a step feed mode. This would
allow part of the influent load to be introduced into the basin in Zone 2 and/or Zone 3 as well as Zone  1. In
this  case, the designer would need to reevaluate the aF distribution down the length  of  the basin. An
additional advantage of a step feed operating mode is an improved capability to avoid solids washout during
extreme flow events. One potential disadvantage may be lower treatment efficiency during these periods.

If the relatively high diffuser density  in Zone  1  is  acceptable, a third option for avoiding  mixing-limiting
conditions in Zone 3 is to allow Zone 1 and/or Zone 2 to be  operated at low DOs. For example,  suppose
lhat during average nitrifying conditions the airflow to Zone 1 is reduced by 40 percent. This would result in
an airflow rate of  1.1 scfm/diffuser  and an SOTE of approximately  30 percent in Zone 1. Zone  1  SOTR
would drop  from 6,068 Ib/d to 4,185 Ib/d.  This  would result in  a transfer  of 1,883 Ib/d  oxygen demand
(SOTR) to Zone 3, assuming no change in  the operation  of Zone 2. The AOR transferred to Zone 3 would
be 0.19 x 1,883  [i.e., (0.768aF)(SOTR)] = 358 ib O2/d. The SOTR required  to meet this additional demand
in Zone 3 would be 358/[(0.768)(0.7)]  =  663 Ib  O2/d. The required  total SOTR in Zone 3 would be 1,385
Ib/d.  The  required airflow in Zone 3 to meet this demand would be about 1.4 scfm/diffuser, and process
oxygen  requirements (rather than mixing) would control minimum airflow to this zone. The load shift would
require a reduction in airflow of approximately 350 scfm in Zone 1. This reduction would cause depression
of DO levels in Zone 1 and could lead to bulking problems, as discussed previously.

For Ihe purposes  of this example,  the design as  presented  in the summary of required airflow rates is
assumed to be acceptable.  The next step is to configure the diffuser system.

Step 4:  Configure the diffuser system.

   Assume the following for the arrangement of the diffusers for full floor coverage of the basins:

        Number of drop legs/zone = 1
        Number of extra diffuser attachments for contingency = 20 percent

        Floor area/zone = (23 ft)(43.3 ft) =  996 sq ft

   Arrange Ihe diffusers approximately evenly within each zone. The following approximate spacings apply:

        First zone    = 996 sq ft/491 diffusers = 2.03 sq  ft/diffuser
                    = 1.4-ft spacing, center-to-center

        Second zone = 996 sq ft/296 diffusers = 3.36 sq  ft/diffuser
                    = 1.8-ft spacing, center-to-center

        Third zone   = 996 sq ft/149 diffusers = 6.68 sq  ft/diffuser
                    = 2,5-ft/spacing, center-to-center

   Assume main headers will be placed across the basin  width in the center of each zone, with laterals fed
   from that main header (half on each side of the header). Determine the number and spacing of laterals
   and  number and spacing of diffuser connectors on each lateral. Zone size is 23 ft x 43.3 ft.

                                                                                        (continued)
                                                130

-------
                   Example 5-10.  Design of Fine Pore Aeration System (continued)
First Zone:
    (23 ft)/( 1.4 ft/diffuser)  = 16.4, say 16-spaces       -     ,,    -

     ,   Use  16 laterals on each side of the main header (32 total). Arrange the laterals in pairs.  Place the
        laterals in  each pair 0.75 ft apart, center-to-center. Place adjacent  pairs 1.9 ft apart,  center-to-
        center, to maximize walking space between alternating  rows of diffusers. Place the outside laterals
        1.85 ft from center to basin wall. Even with the staggered spacing between laterals, the clear space
        between 7-in diameter diffusers on the wide-spaced laterals is only 1.32 ft (15.8 in), which is less
        than ideal.

    (491 diffusers)(1.2)/32 laterals  =  18.4, say 18 diffusers/lateral (15 installed)
                                     [Note that this results in (15)(32)  = 480 diffusers initially in  Zone 1.]

    Allow 2 ft for header in center of. zone, 2  ft at basin wall, and 1 ft at end of zone. A total of 18 diffuser
    baseplates will be installed per lateral, although 3 of them will be plugged initially. There are  17 equal-
    size spaces between the 18 baseplates.

    (43,3 - 5)/[(17)(2)]  =  1.13 ft spacings between diffusers, center-lo-center

    Use 1.13-ft nominal diffuser spacing,  2.0 ft from end  wail,  and 1.0 ft from end of zone. Arrange the
    diffusers on  each pair of laterals so that they  are not directly  across from  each other (i.e.,  stagger
    diffuser placement).

Second Zone:
'  .  (23 ft)/(1.8 ft/diffuser)  = 12.8, say 12 spaces
                                                                         i
        Use 12 laterals on each  side of the main header (24 total). Arrange  the laterals in  pairs.  Place the
        laterals in  each pair 1.25 ft apart,  center-to-center. Place adjacent pairs 2.2 ft apart,  center-to-
        center, again to maximize walking space. Place the outside laterals 2.25 ft from center to basin wall.

    (296 diffusers)(1.2)/24 laterals =   14.8, say 15 diffusers/lateral (12 installed)
                                      [Note that this results in (12)(24)  = 288 diffusers initially in  Zone 2.]

    Allow 2 ft for header in center of zone and 1 ft at each end of zone.

    (43.3 - 4)/[(14)(2)]  =  1.40-ft spacings between diffusers, center-to-center

    Use 1.40-ft nominal diffuser spacing, 1.0 ft from each end of zone.

Third Zone:
    (23 ft)/(2.5 ft/diffuser)  = 9.2, say 9 spaces '-

        Use 9 laterals on each side of the main header (18 total). Space laterals evenly at 2.25 ft apart,
        center-to-center. Place the outside laterals 2.5 ft from center to basin wall.

    (149 diffusers)(1.2)/18 laterals = 9.9, say 10 diffusers/lateral (8 installed)
                                     [Note that this results in (8)(18)  = 144 diffusers initially in Zone 3.]

    Allow 2 ft for header in center of zone, 1 ft at beginning of zone, and 2 ft  at end wall.

    (43.3 - 5)/[(9)(2)]  = 2.13-ft spacings between diffusers, center-to-center

    Use 2.13-ft diffuser spacing, 1.0 ft from beginning of zone, and 2.0 ft from basin wall.

    Figure 5-10 illustrates the diffuser arrangement described above.
                                                   131

-------
Figure 5*9.   Manufacturer's SOTE data for Design Example
            5-10.
   i
   Hi
   s
      45—
      40"
      35—
      30..
      2S"
      20
Tank Lsquid Dopih = 15 (i
Dilfusor Submergence = 14 fi
          24 dilfusors/100 Sq ft
                                 18 diffusers/100 sq ft
                                  •4-
                0,5     1,0      1.5     2.0     2,5

                   Airtow Ralo por Diffuser, scfm
(he static head and line losses  and  deliver the
required airflow lo the diffusion system. The air filters
remove participates, such  as dust and dirt, from the
blower inlet air lo protect the blowers from mechanical
damage.  These  filters may  also  be necessary  to
protecl fine  pore  diffusers  from  air-side  fouling.
Additional blower inlet air treatment may be necessary
in certain applications (for  example, when part of the
air is  taken  from  the plant  headworks  or primary
treatment areas to  help in  odor control). Also,  blower
outlet  filters  are  sometimes installed  to  provide
additional air-side protection for the diffusers.  Figure
5-11 is a schematic of a typical air supply system.


5,5,1 Air Piping
5.5.1.1 Materials
The air piping takes air from the blowers through the
droplegs  and into the  aeration  basin.  The  major
considerations in selecting  materials are  strength,
potential for deterioration due to corrosion,  attack by
HCI or other oxidants used  for  cleaning,   other
environmental factors, and  thermal effects.

Piping  materials commonly used in air supply systems
include carbon  steel, stainless  steel, ductile iron,
fiberglass reinforced plastic  (FRP), high  density
polyethylene  (HOPE), and polyvinyl chloride  (PVC).
Carbon steel, ductile  iron,  and FRP are the materials
most  commonly used for delivering  air  from the
blowQfS to the basins because of their  strength and
durability. Piping within  the basin may  be stainless
steel, PVC, or HOPE because of the resistance  of
such materials to corrosion.
Temperatures >90°C  (200°F) are not uncommon in
the blower discharge piping, it is important, therefore,
that the pipe and accessories (pipe supports,  valves,
and gaskets)  be designed  to  withstand  these high
temperatures.  Provisions  for  pipe expansion and
contraction are usually  needed since thermal stresses
can be significant. The blower discharge piping is
often  insulated to protect workers from  burns. Heat
from the supply air usually dissipates within a few feet
of the point  where the piping  goes underground or
becomes  submerged,  and  thermal  stresses  under
these conditions are normally small.  However,  during
transient  conditions  when  basins  are  down for
repair/inspection in winter or summer, these stresses
can be significant and should be evaluated.

Because of the potential for corrosion at the interface
between the  atmosphere and the liquid, a change in
piping  material from carbon  steel or ductile  iron to
stainless steel or PVC is usually made at the droplegs
into the basins. Since blower discharge pressures are
normally <100 kPa (15 psi), thin-walled stainless steel
pipe is  often used  to reduce cost. The  use of thin-
walled pipe  requires  that  the pipe  be  adequately
protected  from physical damage.  The loads the pipe
must withstand should  be considered in selecting the
material and wall thickness.

Once  inside  the basin,  the  piping  branches  into  a
system of  headers  and  manifolds.  The   choice
between stainless steel and  PVC  pipe  is   usually
dependent on the structural  requirements  of the
diffuser connection and on whether a  gas cleaning
system is being used.

Stainless  steel piping  is often  selected  for systems
that use tube diffusers  because of the cantilever load
applied to the lateral pipes by the diffusers. However,
PVC piping has also been used successfully with tube
diffusers when the connections between  the  lateral
pipes  and diffusers have been designed to withstand
the cantilever loads.

PVC piping is commonly specified when disc or dome
diffusers are  used.  These types  of  diffusers  are
usually mounted on top of the laterals and the forces
transmitted through the connection to the  laterals are
minimal.  Appropriate  PVC  pipe specifications  are
described in Section 2.4.3

When  a  gas  cleaning  system  is"  provided,  it is
essential that the piping material and the gas used for
cleaning be compatible. PVC is often the material of
choice when  an acid gas cleaning system is provided
because of its high chemical resistance.

It is not uncommon for basins to be drained  and left
empty for extended periods of time. Thus,   piping
systems designed to be submerged will be exposed, a
situation that must be considered  when selecting
materials of construction. For example, in areas where
                                                   132

-------
Figure 5-10.  General arrangement of cliff users in in-tank air piping for Design Example 5-10.
        Zone 1
        - 18 Connections on Each of 32 Laterals
        - 15 Diffusers Installed per Lateral
             (typical)
Zone 2                            Zone 3
- 15 Connections on Each of 24 Laterals -10 Connections on Each of 18 Laterals
- 12 Diffusers installed per Lateral   .    - 8 Diffusers Installed per Lateral
   (typical)                               (lypical)
                 \
      \
\
   Basin Header        /
                  Zone Header
    Zone Header

  Plan - Typical Basin (1 of 4)
                                       Zone Header
Figure 5-11.  Air supply system schematic.
                    Inlet Air Filler
                  s
   Air Supply Piping
             Flow Meter or Test Connection
                              Flow Control Valve
                                                Outlet
                                               Air Filter
                                             (As Required)
                                  Isolation
                                   Valve
                      Ooo
                      oo
                      0 «>
                                                                                Laterals
freezing occurs,  the materials need  to  be  able  to
withstand the  effects of freezing  and  thawing.  If the
header  and  lateral  piping  material  is  PVC,  it  is
important that it be formulated  to  be resistant to the
ultraviolet rays of sunlight. Titanium dioxide (minimum
TiO2  of 2 percent)  is the compound most commonly
used  in PVC piping to  provide   protection  against
ultraviolet radiation.

5.5.1.2 Design
As discussed in  previous sections, both the air piping
and aeration  basins should be  designed  to allow for
the basins to be  drained  and left  empty for extended
periods  of  time. In  cold climates,  the  effects  of
freezing  and  thawing  should be  considered  and
provisions for protecting the  system  from  damage
associated with cold weather should be made. These
provisions should include, as a minimum, designing
              the  piping  system  to  allow  for  expansion  and
              contraction,  providing  the  capability  to  drain  the
              system fully, and designing  the basin to avoid  frost
              heave.  Operational  considerations for  protecting air
              supply piping when  aeration basins are out of service
              are discussed in Chapter 4.

              Air piping should be sized so that the headless in the
              supply, header, and lateral piping is small compared
              with the headloss  across the diffusers.  Generally,  if
              losses in the air piping between the last positive flow
              split (valve or  other control  device)  and the  farthest
              diffuser are  less than  10 percent of  the  headloss
              across the diffusers,  good air distribution throughout
              the basin can be maintained.  A control valve is usually
              provided at the top of  each dropleg into  the basin
              Since  the  headloss through a  fine  pore  diffuser is
              normaliy 23-38 cm (9-15 in) w.g., including the control
                                                     133

-------
orifices,  headloss  through  the  dropleg  and  air
distribution piping within the basin should be less than
about  4 cm  (1,5  in).  Higher headlosses in the  air
distribution piping can result in uneven air distribution
within  tho basin.

Considerations in air piping design to accommodate
How measurement devices are discussed in  Section
5.4.3.2. In general, a straight piping length equal to 10
pipe diameters  should be  provided upstream  of the
device, and a straight length equal to 5 pipe diameters
should be provided downstream.

Basic  principles  of fluid  mechanics can  be  used to
determine headloss In air piping systems.  At the rates
of flow and velocities found in these systems, air can
bo treated as an incompressible  fluid within  the pipe
and the Darcy-Weisbach equation can  be  used to
determine headloss:
                  hf = f (I/O)
                                       (5-7)
where,

   hi
   v
   g
= headloss, psia
= friction factor, from the Moody diagram (see
  Appendix C, Figure C-2)
= length of pipe, ft
= inside diameter of pipe, ft
= 1.93x10-6(Ya)(v2/2g)
= velocity head in pipe, psi
= specific weight of air in pipe, Ib/cu ft
= airflow velocity in pipe, ft/min
= acceleration due to gravity = 32.2 ft/see2
Headloss calculations for straight pipe segments can
be simplified  by  using Equation 5-7  to develop
headloss charts.  Headloss  for  valves,  bends,  and
other pipe system components can be calculated as a
multiple of the velocity head:
                       = K,, Hv
                                       (5-8)
where,
      =  headloss coefficient (see Appendix C, Table
         C-2  to obtain  coefficients for  common
         appurtenances)
5.5.2 Air Filtration
5.5.2.1 Degree of Cleaning Required
The degree of cleaning or  air filtration  required  is
dependent  on both  the supply air quality and  the
particular diffusers in place.  Filtration is  required to
protect blowers from abrasion and may be needed to
control air-side fouling of fine pore diffusers.  Typical
minimum  requirements for blower inlet air filtration to
protect the blowers are  %  percent  removal  of
particles 10 microns and larger (31,32).
Standard  practice in designs using fine  pore  aeration
devices  is  to  provide blower  inlet air filtration  to
remove 90 percent of  ail particles 1  micron and larger
(31).  However,  perforated   membrane  diffuser
manufacturers  now indicate that filters designed  to
protect the blowers  will  also  provide  adequate
protection for  their diffusers  (11,31). There  is also
evidence, based on recent studies  (10,33), that this
same degree of filtration may be adequate  for ceramic
diffusers  in new installations. However,  because air-
side fouling  is  much  more  difficult  to  remove than
water-side fouling, designers should  be  cautious in
providing  less  efficient  air filtration  than  is  current
standard practice,

5.5.2.2 Types of Filters
Air filters are  commonly  grouped  into  three  broad
groups: fibrous  media  filters, renewable  media  filters,
and electronic air cleaners (34). Fibrous media filters,
which  include  viscous impingement filters and dry-
type filters, are the types most commonly used for
wastewater aeration.

Dry-type  filters  are composed  of random  mats  of
fibers.  The  size and type of  fiber used dictates the
degree of filtration achieved.  Dry-type filters are often
pleated to obtain a greater media surface  for a given
face area,  which results in  a  more  reasonable
pressure drop across the filter.
                                    ..   ' (•    •    ••'-•
Viscous impingement filters use  a high porosity  media
made  up of coarse  fibers coated  with  a  viscous
substance, such as oil.  The viscous substance acts
as an adhesive to trap particles  trying to  pass through
the media.  These  types  of filters are often  used in
series  with  dry-type filters to  minimize the possibility
of oil-laden air entering the system.

5.5.2.3 Filter Selection
Selecting  the proper filter for a particular application
should include the following considerations: efficiency,
cost, and  expected filter life. The required efficiency is
usually designated by the diffuser manufacturer. As
stated  previously, the  degree of filtration  required is
dependent on the  type of diffuser selected and the
individual diffuser manufacturer.  As  would  be
expected, the  cost of air filters  increases as  their
efficiency increases. Since filters should be cleaned or
replaced   periodically,  it is important that they be
readily accessible.


5.5.3 Blowers

5.5.3.1 Description
As  illustrated in Figure 5-12, many different types  of
blowers  are available  (35).   The  term  "blower"
generally applies to units that  deliver pressures up to
approximately  100 kPa (15 psi).  The term "fan"  is
commonly applied to units that deliver pressures up to
only about 14 kPa (2  psi). "Compressors" are units
that deliver  discharge  pressures >100 kPa (15 psi).
                                                   134

-------
Figure 5-12.  Types of compressors and blowers.
                                             Compressors and
                                                 Blowers
                            Dynamic
                             types
                 Centrifugal
                (Radial Flow)
              • Single-Stage
              « Multistage
              • Modular
              • Horizontal Split
              • Barrel
              »Interceded
   Axial Flow
» Multistage
• Mullislage with
 Variable Slator
 Valves
                                     Posilive
                                   Displacement
                                      Types
  Reciprocating
» Air Cooled
• Water Cooled
• Single-Stage
• Multistage
• Integral Gas-
 Engine Driven
• Separate 8as-
 Engine Driven
     Rotary
• Two-Lobe
• Three-Lobe
* Screw (dry)
• Screw (oil-
 flooded)
• Vane
• Liquid-Ring
Since  most  aeration  applications in  wastewater
treatment require air pressures of 70-100 kPa (10-15
psi),  blowers  are  used.  The  two  types of blowers
normally  used for  wastewater  aeration are  rotary
positive displacement units and centrifugal units  (both
single and multiple stage).

Blower capacity may be given in  several ways but, for
proper selection, the  capacity must be referenced to
inlet (or actual) conditions. Therefore, the most  useful
measure of capacity is inlet (or actual) volume per unit
time,  Us  (cu ft/min;  icfm  or acfm).  Commercially
available  rotary  positive displacement  blowers  have
capacities of 2-23,600  Us (5-50,000 acfm). Centrifugal
blowers are  usually  available in capacities of 240-
70,800 Us (500-150,000 acfm). The impellers of a
centrifugal unit can be arranged singly or in multiple
stages for higher discharge pressures.

Figure  5-13  illustrates the  general operating
characteristics of a rotary positive  displacement unit
and a centrifugal unit. As seen in the figure,  the rotary
positive displacement unit will  deliver  a  relatively
constant  airflow rate over  a  range  of  discharge
pressures. In  contrast, the centrifugal unit is capable
of  delivering  a range  of  airflow  rates  at a relatively
constant discharge pressure.

High efficiency and the ability to  operate  over a range
of  discharge  pressures are  two of  the  principal
advantages  of positive  displacement  blowers.  The
main  disadvantages  of  these units, compared  with
centrifugal  blowers, are  the inability  to effectively
throttle airflow rate (however,  the blowers can be
                      Figure 5-13.  General operating characteristics of blowers.
                                   Rotary Posilive
                                 Displacement Blower
                                      30-50% of Design
                                      ~" Point Airflow
                                                         Design Point
                                        Airflow Rate

                      speed controlled), the  usual requirement for  a more
                      substantial foundation to dampen  and resist vibration,
                      and generally noisier operation.

                      Advantages of  centrifugal  . blowers include  quieter
                      operation  and smaller foundation  requirements. Their
                      disadvantages  include  a limited  operating  pressure
                      range and a reduced volume of air delivered with any
                      backpressure buildup due to clogged diffusers.

                      5.5.3.2 Turndown Considerations
                      If the selected blowers  are to operate in  an  economic
                      range over the entire life of the project, they must  be
                      capable of supplying  appropriate volumes  of air  to
                      meet the  varying oxygen demands of the wastewater.
                      Oxygen demand variations  result from  both  diurnal
                      fluctuations and the differences between start-up and
                      design loads. Therefore, blower selection should take
                                                     135

-------
Into account  minimum air requirements at plant start-
up (which may be limited by mixing requirements) and
peak  air  requirements at design  conditions. It is
essential  that appropriate aeration  control strategies
and  equipment  be  incorporated in  the  overall   air
delivery design.  This will ensure full realization of the
potential  operating  benefits  of  reduced  power
consumption  of fine pore diffusion systems compared
with coarse bubble and mechanical  aeration systems.
This aspect of blower design assumes even greater
importance in retrofit situations.  Detailed  information
on aeration control is provided in Chapter 6.

5.5.3.3 Blower Selection
Because  different types of  blowers  have different
operating  characteristics,  it  is important to select a
blower that is compatible with the normal  operating
mode of the basins. Other factors, such as efficiency,
noise,  maintenance, and operator  preference, must
also be considered in blower selection.

As  discussed previously,  centrifugal  blowers  are
capable of supplying  a range of air volumes at a
nearly  constant  discharge pressure,  whereas rotary
positive  displacement blowers are  capable  of
supplying  a nearly constant air volume over a range of
discharge pressures. Therefore, if the  aeration system
is to be operated with a fairly  constant water depth, as
is  typical  in  activated  sludge  aeration  basins,  a
centrifugal blower is  often an appropriate selection.
Conversely, if il is to be operated over a wide range of
depths, such  as with a sequencing batch reactor, a
positive  displacement  blower  may be  a  better
selection.

Both discharge pressure and  the volumetric flow of air
necessary  to  meet  a given  process oxygen
requirement vary with inlet air temperature,  since both
the pressure and volume of  air  are  functions  of  its
density. As a result,  blower capacity  (in  inlet  L/s
[iclm]) should be based on supplying  the required  air
volume at maximum  inlet temperatures  when  the
volumetric flow required to provide a given  mass flow
will bo greatest.  Blower motor sizing should be based
on blower operation at full  capacity  under minimum
inlet temperature conditions  when the density of  air
will be highest.

5.5.3.4 Control Considerations
Chapter 6 provides  a detailed discussion of  blower
control  methods. This section summarizes some of
the key considerations.

Control of the  air volume  using a  rotary positive
displacement  blower is  normally accomplished   by
using variable speed controls. Inlet vanes or throttling
valves may be used lor control of air volume delivered
by a  centrifugal  blower. Surge limits for  centrifugal
blowers can  be  lowered to nearly 30 percent of  the
rated  capacity using  inlet guide  vanes (31,35).  Inlet
throttling valves  can  lower the surge limit to about  45
percent of rated capacity (31,35). The use of multiple
blower units is another potential  air  volume control
option.  In  all  cases,  blower  manufacturers'
recommendations should be sought.  Table  5-3
compares these three  control methods relative  to
efficiency, cost, and complexity.
Table 5-3. Control Methods for Blowers

Efficiency
O&M Costs
Capital Costs
Complexity
Positive
Displacement
Speed Control
High
Low
'High
High
Centrifugal
• Inlet Vane
Adjustment
Medium
Medium
Medium
Medium
Inlet Valve
Throttling
Low
High
Low
Low
Inlet valve throttling is often used to control the output
of a centrifugal blower because it is the least complex
method of control. Throttling the inlet valve will allow
the  air  volume to be increased or decreased. This
type of control can be  accomplished using either a
manually-operated  valve or a motorized valve  that
operates based on some other measured parameter in
the  system,  such  as  DO.  Airflow and  discharge
pressure  can  also  be  throttled, using  a  valve
downstream of the blower. However, this  method  of
control usually requires  more horsepower than if the
inlet valve is used to throttle airflow and pressure.

When varying the air volume delivered by a centrifugal
blower,  care must  be  taken  not  to decrease the
airflow rate to a point where surge occurs. Surge is an
unstable condition where rapid pulsations in  discharge
pressure and airflow  occur, producing high-frequency
reversals in the axial thrust on the blower  shaft (36).
This condition can become severe enough to damage
the  blower. To minimize  the possibility of reaching the
surge point, the  design point on  the  blower curve
should be selected so that the surge point  occurs  at
approximately 30-50 percent of the design airflow rate.
It is also  good practice to specify  a blower with a
slightly  higher surge  pressure  than  the  maximum
system  pressure  under  minimum airflow conditions.
The surge point of a typical centrifugal compressor is
shown in Figure 5-13.

5.5.4 Design and Installation
Proper  installation  of  the air  supply  system  is
necessary  to  maximize  system  efficiency and
minimize maintenance requirements. To  help achieve
proper  installation, several  items  that should  be
considered during the design of the system are listed
below:

• Provide an  adequate  number of isolation  valves.
  This  will permit  routine  maintenance without
  sacrificing the  operation of the entire  air supply
  system.
                                                  136

-------
• Avoid  low spots  in the air piping. Moisture  may
  collect at these low points causing a decrease in
  efficiency of the air supply system. If low points are
  unavoidable, provide manual or automatic drains.

* Provide  for expansion and  contraction  in the air
  piping  system  due  to  temperature  changes.
  Provisions should include  an adequate  number of
  expansion fittings  and pipe  supports  that  allow
  movement in the piping system.

• Select proper  materials for the high temperatures
  that will occur  in the piping system. Materials used
  for  gaskets,  supports, valve  seats,  and other
  miscellaneous appurtenances must be considered.

» Provide  adequate  airflow   and   pressure
  measurement  devices throughout the  system.
  These  devices  aid   the  operator  in   properly
  controlling the distribution of air in  the system.

* Locate  and size airflow and pressure indicators so
  they are easy to read from the points at which the
  operator adjusts system airflow.

» Provide  sufficient  space  around  the  blowers to
  facilitate maintenance of these units.

» Include provisions for  removing  the blowers  from
  their pads for maintenance. Depending on  the sizes
  and locations of the blowers, these provisions may
  include  overhead hatches, monorail  beams,
  mechanical hoists, access for a  front end loader,
  and others.

• Locate  the blower intake filters where it is easy for
  "operators lo  inspect  and  replace  them when
  necessary.

5.5.5 Retrofit Considerations
The retrofit  of an existing  aeration system is  site
specific. Many  of the same considerations that apply
to design  of  new aeration systems apply to design of
retrofit installations.  These  considerations  include
process oxygen requirements, diffuser selection,  and
configuration of the  air diffusion system  within  the
basins. There are some  factors, however, over which
the designer generally has little control, such as the
configuration of  the  basins. In some  cases,  for
example when replacing  a diffusion system of another
type, the designer must make decisions regarding the
use, refurbishing, or replacement of other components
of the system. The following sections address  some
of these decisions.

5.5.5,1 Air Piping
The  sizes  of  existing  air  distribution  pipes  are
generally adequate for a  retrofitted fine pore aeration
system. Because required airflow rates decrease as a
result of the improved OTE achieved by retrofitting to
fine pore diffusers,  the size of the  existing blower
discharge headers and air mains that deliver air to the
basins will usually be sufficient.  Depending  on the
type and arrangement of fine pore diffusion equipment
selected, the individual drop pipes into the basins may
also be large enough. The piping arrangement must
be checked to  determine if the  air piping  is properly
located to provide the air distribution and  flow control
capabilities required. In addition, replacement  and/or
recalibration of airflow measuring  devices should  be
considered at  this time.  It is important that the flow
measuring devices are operable and accurate over the
new airflow range.

The air  distribution  piping should  be  inspected  to
determine the condition  of the  various components.
Connections, particularly  flexible connections, should
be  checked for  leaks. Areas  subject to  corrosion
{such as  basin drop  pipes) need careful  inspection.
Existing  swing  arms  are often  unusable due   to
corrosion and  wear. Use  of these arms may require
the  addition of weights   if relatively  heavy  coarse
bubble diffusers are being replaced with light fine pore
diffusers.

As with the design of a new air piping system, control
of airflow and  distribution of  air within the aeration
basins are the keys to a successful retrofit. Control  of
airflow usually requires  a positive  control  device
(usually  a  valve)  on  each dropleg  and  each  basin
supply air main.  Considerations  for  designing  air
diffusion  systems to provide adequate air  distribution
within  the  basins, discussed earlier in this chapter,
apply to retrofit installations.
5.5.5.2 Air Filtration
Blower inlet filters will effectively remove contaminants
from the outside air, but will  not protect the diffusers'
from dirt, rust, scale, or other debris already  in the
downstream piping. This debris may  be the result  of
internal  pipe corrosion,  leaks  in  submerged  piping,
physical  damage  to  the  nonsubmerged  piping,  or
other causes. Thorough cleaning  of the  air  piping
system prior to diffuser  replacement may be required
to remove this debris.

In addition to cleaning the  air piping,  installing air
filters on the discharge side of  the blowers should be
considered.  In some cases, it  may be advantageous
to locate filters at the top  of the  drop pipes  leading
into  the  aeration  basins.  New piping  can then  be
installed in the basins, and  the  filters will minimize the
potential  for air-side fouling  of the  diffusers  due  to
debris from the existing upstream air supply piping.

Existing  piping  systems  composed  of  stainless,
galvanized,  or  coated  steel  normally  present  little
danger of diffuser  fouling from  rust or scale. Existing
painted  or uncoated steel or  iron  pipe, however,
should be reused only  with  extreme caution unless
downstream in-line filters are installed.
                                                   137

-------
5.5.5.3 Blowers
If  Ihe  aeration  system  to  be replaced  is either a
diffused  air or sparged  mechanical aeration system,
air  blowers and distribution  piping will  already  be
present.  Replacement  blowers may sometimes  be
required  due  to age or lack of  flexibility.  Blower
capacity  should  be  evaluated to  determine  if it  is
sufficient for the retrofitted  system  (keeping in mind
that replacement of coarse  bubble diffusers with fine
pore  diffusers  reduces  airflow requirements).
Turndown  capabilities should  also  be  evaluated  to
determine whether the blowers can be operated at the
lower  airflow  rates  normally used with fine pore
diffusers and whether the revised operating mode will
reduce  energy  costs.  Obviously, a  plant with
mechanical surface aerators will require installation  of
new blowers and air piping to use fine pore diffusers.

The energy savings available  with fine  pore diffusers
result from a reduction in the volume or  air required  to
transfer  required DO to the process. These savings
may be  partially offset by the "increased operating
pressure required for fine pore diffusion  systems. The
reduction in airflow, if achieved, will result in operating
fewer  blowers or  operating  the  same  number  of
blowers  at different  points  on  their  performance
curves.

As  the operating point of a centrifugal blower moves
away from its  maximum capacity, the efficiency of the
machine  decreases.  As  a   result,  the  potential
reduction in energy requirements associated with use
of a fine pore diffusion system may be  partially offset
by  lower blower efficiency if existing blowers operate
further from their  maximum  efficiency  points. It  is
important to confirm  that the  existing combination  of
blowers can operate over the required airflow range  of
the retrofitted air diffusion  system prior to replacing
the system.

Control strategies available for both centrifugal  and
positive  displacement blowers, discussed earlier  in
this chapter,  apply  to  retrofit situations.  Centrifugal
blowers not equipped with inlet guide vanes should be
throttled  on the inlet side rather than on  the discharge
side.  Discharge  throttling does  not achieve  energy
savings,  while inlet  throttling  results  in  significant
power  savings at any  operating  point.  Airflow from
positive  displacement blowers  can be  varied  by
operating more or fewer units, adjusting the speed  of
the units, or "blowing off" air to the atmosphere., The
last method is the least preferred, because it does not
reduce energy  requirements. As  with  centrifugal
blowers,  it is important to confirm that existing positive
displacement  blowers can operate over the  required
airflow range  of the retrofitted air diffusion  system
prior to installing that system.

The capability of the blowers to deliver the required air
volume at peak air  temperatures  should be checked.
The capacity of the blower motors under minimum air
temperature conditions should also be checked. The
overall blower system integrity should be evaluated 'in
terms  of  its turndown  capability  and  flexibility  of
operation.
5.5.6 Air Supply System Design Example
Consider  the  activated sludge air diffusion  system
design presented in Example  5-10. A corresponding
appropriate air supply system is designed in Example
5-11.
Several modifications to the  designs presented in
Examples 5-10  and 5-11  could  be considered to
improve  operability and  reduce operating  costs.
Possible  modifications of the  system  construction
schedule to economically satisfy temporal variations in
process oxygen  requirements  between plant start-up
and the time  when  design plant loadings  are reached
have been discussed, Another modification that could
be considered is to install a  separate main air  header
to supply each basin. This would  increase the capital
cost of the aeration  system but  would  provide the
opportunity  to reduce operating costs  by  more
accurately controlling the airflow to each basin. Airflow
control is discussed in more detail  in Chapter  6. A
third possible modification is  to supply one blower with
a capacity  of one-half the  blowers  selected for the
ultimate plant loadings (i.e.,  1,400 scfm) to provide
better  airflow  turndown capability during  the  initial
years of plant operation.
Normally, several control alternatives will be evaluated
in arriving at a final  system configuration  that meets
the  objectives  of  the  designer  and   owner.
Considerations  in  meeting  these objectives  may
include ease of operation, reliability, capital cost, and
O&M cost.
5.6 Summary of Aeration System Design
Procedure
The  design of a fine pore aeration  system  for  an
activated sludge process involves a series of steps.
Design examples that illustrate each  of  these steps
were  presented  through  the earlier sections  of this
chapter.  These  examples  illustrate  the general
approach  to  design.  Design  of  an  actual  system
requires consideration of several  site-  and system-
specific factors, not all of which  were fully discussed
in these examples.
This section briefly  summarizes the overall approach
to  the  design  of  fine  pore  diffusion  systems,
highlighting some of the key design considerations. A
summary  list of steps  is presented, followed by a
discussion of each of these steps.
                                                  138

-------
                            Example 5-11.  Design of Air Supply System
Step 1: Design the blower system.

Determine the range of aeration rates required.                  .

    Maximum  = (1,968 scfm/basin) (4 basins)  = 7,872 scfm
    Average nitrifying = {1,383 scfm/basin) (4 basins)  = 5,532 scfm
    Average nonnitrifying = (698 sefm/basin) (4 basins) = 2,792 scfm
    Minimum = (493 scfm/basin)  (4 basins) =  1,972 scfm

Set number and approximate capacity of blowers.

    Assume the following criteria for the system:

    3 blowers to provide 100 percent capacity
    4 blowers installed
    1 blower to meet minimum requirements

    Provide four 2,800-scfm capacity blowers.

Determine discharge pressure requirements.

The following piping losses were calculated using the Darcy-Weibach equation:

    Diffuser submergence =  14 ft
    Diffuser headloss = 0.70 psi
    Piping headloss  = 0.15 psi
    inlet valve and filter headloss  = 0.3 psi
    Atmospheric pressure at 1,000 ft elevation  = 14,3 psia

    Static head =  (14 ft) (0.43 psi/ft) = 6.02 psi
    System head =  6.02  + 0.70 + 0.15 + 0.3 =  7.17 psig                                   '     -
    Discharge pressure = 14.3 + 7.17 = 21.5 psia

Estimate blower horsepower.

The expression for blower wire power consumption was given in Equation 4-11. Combining the values of w,
R, and K with the conversion factor 2.655 x 1Q6 yields the following working equation:

    WP =  (4.28 x 10-4 qs ja/e) [(Pd/Pb)0-283 - 1]                                                 (5-9)

where,

    WP = wire power consumption (approximately equal to brake  horsepower), hp
    Ta =  blower inlet air temperature,  °F
    e   = combined blower/motor efficiency
    Pd = blower discharge pressure, psia
    Pt> = field atmospheric pressure, psia
    and 4.26 x 10-t js in units of hp-min/eu ft-°R. •

Using this  relationship, and assuming  a  blower inlet air temperature of 68°F and an  overall  blower/motor
efficiency of 70 percent:

    WP =  [(4.28 x 10-4) (2,800) (460 +  68)/0.70] {[(14.3 + 7.2)/(14.3)]°-283 - 1} =  111  hp
                                                                                         (continued)
                                                139

-------
                      Example 5-11.  Design of Air Supply System (continued)
Check the motor horsepower under minimum temperature conditions.                   ...

This requires determining  the relationship  between actual  airflow  (qa) and  standard  airflow under field
conditions (qs):

    (qa'Qs)  - l(Ta + 460)/(68 + 460)] (14.7/Pb)                                                (5-10)

Substituting the appropriate values for Ta and P& yields:

        (qa.'qs) = 0.925 at Trn(n =  15°F
        (qa'Qs) = 1-10atTinax =  105°F

Each blower must be sized to deliver the required qa under maximum temperature conditions:

    (qa/2,800) =  1.1; therefore, qa= 3,080 icfm

To  size  the blower motor,  assume the blower is operating at maximum capacity  (qa) under minimum
temperature conditions:

    (3,080/qs) = 0.925; therefore, qs =  3,330 scfm

Check WP requirements at maximum airflow:

    WP =  (4.28 x 10-4) (3,330) (460 +  15)/0.70]{[(14.3 + 7.2)/(14.3)]0.283 . 1}  = 117 hp '

    Use a 140-hp motor for a 2,800-scfm blower.

Published centrifugal  blower curves are applicable  to operation at  standard conditions (68°F inlet
temperature, 14.7 psia inlet  pressure, and 36  percent relative  humidity).  Where different inlet conditions
apply (as in this example), an equivalent air  pressure (EAP)  must be calculated  to allow the blower curves
to bo used.  Figure C-3 (Appendix C) can be utilized to make this correction.

Use:

    Blower discharge pressure =  7.2 psig
    Blower inlet airflow = 3,100 icfm
    Plant elevation  = 1,000ft
    Blower inlet temperature  = 105°F

    From Appendix  C,  Figure C-3, EAP = 7.9 psig

A blower with an operating  point from a published performance curve that shows 3,100  icfm airflow at a
discharge pressure of 7.9  psig will deliver  2,800  scfm at a discharge pressure of  7.2  psig at the inlet
conditions shown.

    Select 3,100-icfm (at 7.9 psig)  blowers (4 each) with 140-hp  motors.

Step 2: Design the  air piping system.

The general piping  arrangement will be  established based on site constraints,  layout of  other facilities, and
other considerations. The general arrangement assumed for the four aeration  basins  in this example  is
shown in Figure  5-14.

Air piping can be preliminarily sized using the general guidelines  for velocities given in Appendix C, Table C-
3. Accurate headless calculations  must be made,  and the pipe  sizes  (or blower discharge pressure) must
be refined as necessary to match  blower capacity and system headlosses.

                                                                                         (continued)
                                                140

-------
                      Example 5-11.  Design of Air Supply System (continued)
Using the velocity guidelines given in Appendix C, Table C-3, the following preliminary pipe sizing estimates
were made:         ;
                                Approximate Peak                     Approximate Airflow
                                  Airflow, scfm      Pipe Diameter, in       Velocity, fpm
Header Segment
i
2
3
4
:' • 5
6
7
8
Laterals
Zone i
Zone 2
Zone 3

8,000
4,000
4,000
1,600
440
1,200
600
220

40
25
12

. , 20
14
14
10
6
10
6
6

4
4
4

3,700
3,700
3,700
2,900
2,000
2,200
3,100
1,000

460
290
140
Headlosses are calculated most accurately using the Darcy-Weisbach equation (Equation 5-7), as discussed
in  Section 5.5.1.2, To use the Darcy-Weisbach equation, corrections must be made for the specific weight
of  air and  air temperature  under actual  conditions in the air piping. The  equation can be conveniently
applied in the following form:

    (hf/100) = (f/D) (v2/2g) (1.93X10-4 ya)(1/36)                                                (5-11)

where,

    hf/100  = headless per 100 ft  of pipe, psia

The specific weight of air can be determined by treating air as an ideal gas (which  is nearly correct  at
typical centrifugal blower air temperatures and  pressures):

    Ya = 144Pd/(RTp)                                                                       (5-12)

where,

    R -  ideal gas constant  = 53.3 ft-lb/lb-°R for air '
    Tp = air temperature in  pipe,  ft/'R

The temperature rise through the  blower can be estimated by assuming that compression is adiabatic:

    AT  = OVe) [(Pd/Pb)a283 - 1]                                                              (5-13)

where,

    AT  = temperature rise through blower, " R

                                                                                          (continued)
                                                 141

-------
                      Example 5-11.  Design of Air Supply System (continued)
Heal will dissipate in the air piping, so the calculated air temperature rise through the blower will not be the
temperature rise throughout the  air piping system. As an approximation, it can be assumed that  the  air
temperature in the piping is the average of the blower inlet and discharge temperatures. Then:

    Tp = Ta + (AT/2)                                         ,                              (5-14)

Combining Equations 5-13 and 5-14 yields:

    Tp = Ta{1  + (1/2e)[(Pd,/Pb)0.283.1]}                                                     (5-15)

Actual velocity in the pipe is determined by correcting for pressure, temperature, and relative humidity of the
air after compression:
    qa = qs Oyrs) [(Ps - HsPvs)/(Pd - HdPvd)]                                                  (5-16)

where,

    Ts  =  air temperature at standard conditions = 528° R
    Hs  =  relative humidity at standard conditions, fraction
    Pvs =  vapor pressure of water at standard conditions = 0.34 psia
    Ps  =  atmospheric  pressure at standard  conditions =  14.7 psia or 1 atm  at  100 percent relative
           humidity
    H0  =  relative humidity of blower discharge air, fraction
    Pvd =  vapor pressure of water in blower discharge air, psia

Normally, the vapor pressure correction terms are small and can be neglected. The velocity in the pipe is
then:

    v = (4qs/nD2) (Tj/rs) (Ps/Pd)                                '                              (5-17)
Substituting the expressions for Ya> AT, Tp, and v into Equation 5-1 1 and using the appropriate values of the
constants in these expressions yields:

    h,/100= (f/D5)(qs2/Pd)Ta{1 + (1/2e)[(Pd/Pb)0.283-i]}(i.02 x 10-8)                                (5-18)

To use this equation to determine headloss, f must be determined. This can be accomplished by calculating
the Reynolds number (Re) and using Appendix C, Figure C-2:.

    RQ = 2.37qs/Dp                                                                         (5-19)

where,

     p =  viscosity, centipoises

For temperatures encountered in aeration, ja can be estimated from:

    H  =  (32.2 +  0.28Tp) x 10-4                                                                (5-20)

                                                                                         (continued)
                                                 142

-------
                       Example 5-11.  Design of Air Supply System (continued)
  Preliminary headloss estimates can be made using a headloss chart. However, estimates determined using
  such a chart may be overly  conservative. More accurate  estimates can  be made by using the Darcy-
  Weisbaeh equation.

  Table 5-4 presents preliminary headloss calculations for this example, in these calculations, the pipe length
  for each  segment (from Figure 5-10) was increased  by 20 percent to allow for losses through fittings,
  valves, and other  appurtenances. More accurate calculations that include estimates of losses through each
  of these  elements should normally be part of the final design. Losses through these components- can  be
  estimated using the Darcy-Weisbach equation and Appendix  C, Table C-2, as discussed earlier.

  Using the data from Table 5-4, headiosses to each zone can be determined. The values obtained are:


                                                    Headloss, psia
                                               Total
            'Dropleg to Diffuser
                             Zone 1

                             Zone 2

                             Zone 3
0.079

0.134

0.107
0.014

0.054

0,008
  While these values are acceptable (<0.07 psi headloss after the last positive flow split [10 percent of the
  headloss through the diffuser orifice];  < 0.15 psi headloss between  blower and diffuser), it is prudent to
  make changes in two segments: 1  and 7. Increasing the pipe diameter to 24-in in Segment 1 and 8-in in
  Segment  7 results in the revised headloss estimates shown in  Table  5-5.  These  headiosses  are
  conservative. The design  could  be  refined  by calculating  the headiosses  through all  the piping
  appurtenances  (such  as fittings and  valves) and then  reducing  the air  piping sizes accordingly.  This
  refinement is beyond the scope of the example.
5.6.1 Outline of Approach
1.  Determine flows and loads
2.  Establish process design criteria
3.  Size basins
4.  Configure basins
5.  Determine oxygen transfer requirements
6.  Select ditfusers
7.  Determine aeration rates
8.  Configure diffuser system
9.  Design blower system
10. Review system flexibility
11. Design air piping system


5.6.2 Steps in Design
5.6.2.1 Determine Flows and Loads
Design  wastewater flows  and  loadings should  be
established for the  entire range of operating conditions
anticipated.  From these, system oxygen  requirements
can  be estimated. Accordingly, loading parameters
that describe carbonaceous oxygen demand  (BOD or
COD), nitrogenous oxygen demand (TKN and NH4-N),
and  inorganic  oxygen demand  (IOD)  are  usually
required  (see  Section   5.3.1.2  for  calculation
procedures). Waste streams include  both  the  main
plant  flow  and all sidestreams  (such as  sludge
      thickening  and dewatering  flows). Loading conditions
      to be considered generally include:

      1.  minimum month, to establish blower and diffuser
         turndown requirements (normally occurs  at plant
         start-up),

      2.  average conditions (nitrifying and nonnitrifying), to
         establish the normal operating  condition for  the
         blowers and other system components,

      3.  maximum  month, to  determine the maximum
         condition under  which  process  oxygen
         requirements  must be  met without  stressing  the
         microorganisms  (that  is, for which the design
         average DO must be maintained), and

      4.  peak  day/4-hr  peak  (considering   diurnal
         fluctuations), to establish the maximum operating
         point  for  all  system  components,  including
         diffusers, air supply piping, and blowers.
      5.6.2.2 Establish Process Design Criteria
      Several  factors  are  important in  determining system
      oxygen requirements (both under standard and actual
                                                 143

-------
Figure 5-14,  General arrangement of process air piping for Design Example 5-11.


                                             Zone Header
                                         (Laterals Not Shown)




Table 5-4.
Pipo Sea,
1
2
3
4
5
6
7
8
LI
L2
L3

l
«
r
Main Air
Supply
Header
___ •
i i i
i i i
i i i
i i i
i i i
Blower
Connection
"^® '& ®
. 0) P O
0 ! © ©
i

CO O Q
CD ui O
Dropleg
/7s PiP'n9
Vjy Segment
Identification
'
Air Piping Headless Calculations for Example 5-10 - First Iteration
Length. It D, in ks, ft
160 2u 0.00015
50 14 0.00015
40 14 0.00015
50 10 0.00015
50 6 0.00015
BO 10 0.00015
80 6 0.00015
80 6 0.00015
25 4 0.00015
25 4 0.00015
25 4 0.00015
Airflow, Air Reynolds Friction Headless,
ks/D scfm Temp., °F No. Factor , psi
0.00009 8,000 154.9 5.57E+05 0.014 0.0325
0.000128 4,000 154.9 3.98E + 05 0.015 0.0162
0.000128 4,000 154,9 3.98E + 05 0.015 0.0130
0.00018 1,600 154.9 2.23E-I-05 0-017 0.0158
0.0003 440 154,9 1.02E + 05 0.020 0.0181
0.00018 1,200 . 154.9 1.67E+05 0.016 0.0134
0.0003 600 154.9 1.39E + 05 0.020 0.0538
0.0003 220 154.9 5.10E + 04 0.022 0.0080
0.00045 40 154.9 1.39E-I-04 0.032 0.0009
0.00045 25 154.9 8.70E + 03 0.034 0.0004
0.00045 12 154.9 4.17E + 03 0.039 .0.0001
    k,, * ubsolulQ rouolwioss factor
    Field aliiiospliofic pross«iro = I4.3psia
    Olowor discliargo prossuro = 21.5 psia
    Blower mlei tompcwalurQ  = 105°F
    Blowoi olficioncy * 70 percent
                                                            144

-------
Table 5-5. Air Piping
Pipe Seg. Length ,fi
1 160
2 50
3 40
4 50
5 50
6 80
7 80
8 80
L1 25
L2 25
L3 25
Headloss
D, in
24
14
14
10
6
10
8
6
4
4
4
Calculations for Example 5-10 - Second Iteration
Airflow, Ait
ks,,ft kj/D sofm Temp, aF
0,00015
0.00015
0.00015
0.00015
0.00015
0.00015
0.00015
0.00015
0.00015
0.00015
0.00015
0.000075
0.000128
0.000128
0.00018
0.0003
0.00018
0.000225
0.0003
0.00045
0.00045
0.00045
8,000
4,000
4,000
1,600
440
. 1,200
600
220
40
25
12
154.9
154.9
154.9
154.9
, 154.9
154.9
154.9
154.9
154.9
154.9
154.9
Reynolds
No.
4.64E + 05
3.98E + Q5
3.98E + 05
2.23E + 05
1.02E + 05
1.67E + 05
1 .04E + 05
5.10E + 04
1.39E + 04
8.70E + 03
4.17E + 03
Friction
Factor
0.014
0.015
0.015
0.017
0.020
0.016
0.020
0.022
0.032
0.034
0.039
Headloss,
psi
0.0131
0.0162
, 0.0130
0.0158
0.0181
0.0134
0.0128
0.0080
0.0009
0.0004
0.0001
conditions) and the spatial  and temporal variations in
these requirements. These include:

1.   maximum  wastewater temperature  and  the
    corresponding maximum SRT, used to  determine
    the  maximum carbonaceous oxygen consumption
    ratio and  whether nitrification is  likely to  occur
    under these conditions (see Section 5.3.1.2),

2,   minimum   wastewater  temperature  and  the
    corresponding minimum SRT, used to  determine
    the  minimum carbonaceous oxygen consumption
    ratio (see  Figure 5-4) and  whether  nitrification is
    likely to  occur under  these  conditions (see
    Section 5.3.2.2),

3.   extent of denitrification  (if the system is designed
    to denitrify),  used to refine  the  system oxygen
    demand estimates (see Section 5.3.1.2), and

4.   basin  configuration,  used to  establish  the
    distribution of oxygen  demand within  the  basin
    (see Section 5.3.1.3)

Other  criteria   that are important in designing  the
aeration system include:

5.   control  of airflow  (including  process  modes
    available and  control provided)  and wastewater
    flow distribution (e.g., step feed flexibility) to'meet
    system process oxygen demand variations, and

6.   design life and process  loading growth patterns,

5.6.2.3 Size Basins
After  establishing process design criteria, the  aeration
basins can be sized. Basin sizing is beyond the  scope
of this manual.

5.6.2.4 Select Diffusers
Considerations for selecting diffusers are discussed in
detail  in  previous sections  of this  chapter. Key
considerations include OTE, O&M requirements, initial
cost,  O&M  cost,  and  compatibility with  existing
facilities (for retrofits),

5.6.2.5 Configure Basins
Basin configuration  involves converting  the required
aeration basin volume into  physical dimensions. The
depth of diffuser submergence is important because it
determines  OTE  and the  static  pressure  that the
blowers  must  overcome.  Length-to-width ratio  is
important because it establishes spatial distribution  of
oxygen demands and constrains how the air diffusion
system  can  be  arranged.  Typical  diffuser
arrangements are illustrated in Figure 2-8.

5.6.2.6 Determine Oxygen Transfer Requirements
Determination of system oxygen transfer  requirements
is  addressed in Example 5-10. Variations  with  both
time and  position  in  the aeration  basin  should
considered.
                                               be
5.6.2.7 Determine Aeration Rates
Oxygen transfer requirements  (from  Section 5.6.2.6)
are used to determine aeration rates. The procedure
for  converting  standard  oxygen  transfer rates  to
required aeration rates is illustrated in  Example 5-10.

5.6.2.8 Configure Diffuser System
After determining  required aeration rates, the diffuser
system  can be configured.  Several iterations will
normally be required to ensure  that the entire range of
oxygen  transfer requirements  can  be  met without
exceeding the recommended limits on airflow rate per
diffuser. The basin and diffuser system  also need  to
be  arranged to allow  space  for installation  and  to
facilitate  operation  (including  adequate  space  for
maintenance of the  in-basin  piping and  diffusers and
providing for isolation of portions of the system  for
maintenance).

5.6.2.9 Design  Blower System
Temporal  variations  in oxygen demands  should  be
considered in  selecting  an  appropriate number  of
blowers. Typically, the blowers are sized to allow one
                                                  145

-------
blower lo meel minimum oxygen requirements, one or
more blowers operating at full capacity to meet annual
average  requirements,  and two or more  blowers
operating  at  full capacity to meet  peak  hour
requirements.

5.6.2,10 Review System Flexibility
Sufficient flexibility should  be provided to enable  the
System to be operated cost effectively over the entire
life of the facility. This review should consider how  the
system  will be operated  at start-up and  at design
loading. Considerations should include the number of
basins constructed and operated at various points in
the design  life of the system, flexibility  to operate in
the step feed mode, diffuser  layout and ability  to
accommodate changing system oxygen  requirements,
DO and airflow control, and the  number and  capacity
of blowers  installed and operating at various points in
(he design life.
5.6.2.11 Design Air Piping System
The  air  piping system should be designed to permit
cost-effective  installation and operation.  Piping
materials should be selected to provide the degree of
durability (including resistance to mechanical damage,
corrosion, and  sunlight  degradation)  appropriate for
the facility.  Both  permanent  flow meters and  flow
points for portable meter installation  need  to be
properly located  to   allow accurate  airflow
measurements.  This  includes providing  sufficient
straight  lengths of piping upstream and  downstream
(normally, a minimum of  10  and  5 pipe diameters,
respectively). An  adequate number  of  flow  points
where portable flow measurement devices can be
inserted should be  provided.  As a minimum,  a  flow
meter for measuring  total  system airflow and  flow
points for measuring total airflow to each basin and
each zone should be included. Piping should be sized
to provide  acceptable  headlosses  at maximum
airflows, including  a headloss  between  the  last
positive  flow split and the farthest diffuser of less  than
10 percent of the loss through the diffusers. Losses
through  the  blower inlet  filter, control valves, check
valves, basin control valves, and fittings all need to be
considered  in  establishing total  blower discharge
pressure requirements.
5.7 References
When an  NTIS number is cited in a reference, that
reference is available from:

National Technical Information Service

   5285 Port Royai Road
   Springfield, VA 22161
   (70S) 487-4650
1.  Process Design  Manual for  Nitrogen  Removal.
    U.S. Environmental Protection Agency, Cincinnati,
    OH, October 1975.

2,  Robertson, P.,  V.K.  Thomas  and B. Chambers.
    Energy  Saving  - Optimisation  of Fine  Bubble
    Aeration.  Final  Report and Replicators  Guide.
    Water  Research  Centre, Stevenage  Laboratory,
    Stevenage, England, May 1984.

3.  Chambers, B. Effect of Longitudinal  Mixing  and
    Anoxic Zones on  Settleability of Activated Sludge.
    In:  Bulking of Activated Sludge - Preventative and
    Remedial  Methods, Ellis Horwood,  Chichester,
    England, 1982.

4.  Arvin, E. Biological Removal of Phosphorus from
    Wastewater. CRC  Critical  Reviews in Envir.
    Controls 15:25,  1985.

5.  Design  Manual  for  Phosphorus  Removal.
    EPA/625/1-87/001, U.S.  Environmental Protection
    Agency, Cincinnati, OH, September 1987.

6.  Levenspiel,  O.  Chemical Reaction Engineering,
    2nd Edition. John Wiley & Sons,  Inc., New York,
    NY, 1972.

7.  Summary  Report:  Causes  and   Control of
    Activated  Sludge Bulking  and Foaming. EPA-
    625/8-87-012,  U.S.  Environmental  Protection
    Agency, Cincinnati, OH, 1987.

8.  Chudoba,  J., J.S. Cech and P.  Chudoba.  The
    Effect  of Aeration  Tank Configuration  on
    Nitrification Kinetics.  JWPCF  57{11): 1078-1083,
    1985.

9.  Houck,  D.H.  and A.G.  Boon.  Survey  and
    Evaluation of Fine Bubble Dome Diffuser Aeration
    Equipment.  EPA-600/2-81-222, NTIS  No. PB82-
    105578,  U.S.  Environmental  Protection  Agency,
    Cincinnati, OH,  1981.

10.  Donohue  &  Assoc.,  Inc.  Fine   Pore Diffuser
    System Evaluation for the Green Bay Metropolitan
    Sewerage  District.  Study  conducted  under
    Cooperative  Agreement  CR812167,  Risk
    Reduction   Engineering  Laboratory,  U.S.
    Environmental Protection  Agency,  Cincinnati,  OH
    (to  be published).

11.  Summary  Report: Fine  Pore  (Fine  Bubble)
    Aeration  Systems.  EPA-625/8-85-010, U.S.
    Environmental Protection Agency,  Cincinnati,  OH,
    October 1985.

12.  Dome  Diffuser  Evaluation.  Internal  report,
    CH2M/HHI, Denver, CO, February 1981.
                                                146

-------
13.  Wastewater Treatment Plant  Design.  Manual of
    Practice  8,  Water Pollution  Control  Federation,
    Washington, DC/Manual on Engineering  Practice
    36, American  Society  of  Civil  Engineers,  New
    York, NY, 1977.

14.  Eckenfelder, W.W., Jr. Principles of Water Quality
    Management,  CBi Publishing Co., Inc.,  Boston,
   'MA, 1980.        '    '

15.  Metcalf &  Eddy, Inc.  Wastewater Engineering:
   'TreatmentfDisposal/Reuse, 2nd  Edition. McGraw-
    Hill Book Co., New York, NY, 1979.

16.  Grady,  C.P.L.,  Jr.  and  H.C.  Lim.   Biological
    Wastewater Treatment:  Theory  and  Application.
    Marcel-Dekker, Inc., New York, NY, 1980.

17.  Standard Methods for  the Examination of Water
    and Wastewater, 16th Edition.  American  Public
    Health Association, Washington,  DC, 1985.

18.  Recommended  Standards for  Sewage  Works.
    Great  Lakes - Upper Mississippi  River Board of
    State Sanitary  Engineers, 1978.

19.  Lawrence, A.W.  and P.L. McCarty. Unified Basis
    for Biological Treatment Design  and Operation. J.
    San.  Eng. Div.,  ASCE 96(SA3):757-778, June
    1970.

20.  Henze, M.,  C.P.L. Grady, Jr.,  W. Gujer, G.v.R.
    Marais and  T. Matsuo. Activated Sludge  Model
    No. 1. Scientific and Technical Reports No. 1,
  ,  IAWPRC, July  1986.

21.  Theory,  Design, and Operation  of  Nutrient
    Removal Activated  Sludge  Processes.  Water
    Research Commission, Pretoria, South Africa,
  ,  1984.

22.  Munksgaard, D.G.  and J.C.  Young. Flow  and
    Load Violations  at Wastewater  Treatment Plants.
    JWPCF 52{8):2131-2144, 1980.

23.  Boon, A.G. and B. Chambers, Design Protocol for
    Aeration, Systems  -  UK  Perspective.  In:
    Proceedings of  Seminar/Workshop  on  Aeration
    System  Design,  Testing, Operation,  and Control,
  ,,  EPA-600/9-85-005, NTIS No.  PB85-173896, U.S.
    Environmental Protection Agency, Cincinnati,  OH,
    January  1985.

24.  Baillod,  B.C.  Oxygen  Utilization in Activated
  •  Sludge Plants: Simulation  and Model Calibration.
    EPA/600/2-88/065, NTIS No.  PB89-125967, U.S.
    Environmental Protection Agency, Cincinnati,  OH,
    November 1988.
25. Chambers, B. and G.L.  Jones.  Optimisation and
    Uprating of Activated Sludge Plants  by Efficient
    Process Design. Wat Sci. and Tech. 20{4/5):121-
    132, 1988.

26. Personal  communication from  G.T.  Daigger,
    CH2M/Hill, Denver, CO,  to  R.C. Brenner,  U.S.
    Environmental Protection Agency, Cincinnati, OH,
    February 1989.

27. Yunt, F.W, Results of Mixing  Efficiency Tests With
    the Norton  Dome Aeration System  at  the L.A.-
    Glendale  Treatment  Plant.  Internal  report, Los
    Angeles County  Sanitation Districts, Whittier, CA,
    January 9, 1980.

28. Nokia  Diffusers for Wastewater  Aeration Product
    Bulletin, Nokia Metal Products,  Vantaa, Finland
    (undated).

29. Sanitaire  Flexible Membrane Tube Diffusers.
    Product information  bulletin,  Sanitaire-Water
    Pollution Control  Corp., Milwaukee, Wl, 1987.

30. American  Society of  Civil   Engineers.  ASCE
    Standard:   Measurement of Oxygen  Transfer  in
    Clean Water. ISBN 0-87262-430-7, New York, NY,
    July 1984.

31. Aeration.   Manual of  Practice FD-13,  Water
    Pollution  Control Federation, Washington, DC,
    1988.

32. Air Diffusion in Sewage Works. Manual of Practice
    5,  Federation of Sewage and Industrial Wastes
    Associations, Champaign, IL,  1952.

33. Stenstrom,  M.K.  and G. Masutani. 'Fine Bubble
    Diffuser Fouling: The  Los Angeles Studies.  Study
    conducted  under  Cooperative-  Agreement
    CR812167,  Risk  Reduction  Engineering
    Laboratory,  U.S.  Environmental   Protection
    Agency, Cincinnati, OH (to be published).

34. American  Society of Heating, Refrigerating, and
    Air  Conditioning Engineers,  Inc.  ASHRAE
    Handbook,  1983 Equipment  Volume,  Available
    from ASHRAE, Atlanta, GA.
35. Prime Movers. Manual of Practice SM-5,  Water
    Pollution Control Federation,  Washington, DC,
    1984.
36. White,  M.H. Basics  of  Surge  Control for
    Centrifugal Compressors. Chemical Engineering,
    December 25, 1972.
                                                147

-------

-------
                                             Chapter 6
                                          Aeration Control
6.1 Introduction
The  major objectives of aeration  control are to:  1)
ensure that the supply of oxygen meets the dynamic
spatial and temporal  variations  in  process biomass
oxygen  demand  discussed  in  Chapter 5,  and  2)
effectively control air delivery and  oxygen transfer to
minimize aeration energy costs.
This  chapter  specifically addresses  automated
aeration  control  in the activated  sludge  process.
Automated aeration control is the manipulation of the
aeration rate  by computer  or controller to match the
dynamic  oxygen  demand  and  maintain  a  desired
residual or set-point  mixed liquor  dissolved oxygen
(DO)  concentration.  Benefits  of  control,  control
strategies, instrumentation, and  final control  element
hardware  considerations  are  examined.  The
information  presented  will   facilitate  effective
communication  between designers, control specialists,
and operations personnel to ensure incorporation of
control considerations  in the design of the  aeration
system.


6.2 Benefits of  Aeration Control
The  benefits of  implementing  an aeration control
strategy  in a wastewater  treatment facility  are:  1)
assured  integrity  and uninterrupted operation of the
process, 2)  increased  reliability  in  meeting plant
discharge requirements, and  3)  reduced process
costs. If properly applied,  aeration  control  offers the
potential for significant cost savings.

6.2.1  Process Implications
Maintenance  of an inadequate  residual mixed liquor
DO concentration can  inhibit biological  activity and
contribute to problems  such as sludge bulking  (1,2)
and inhibition or  loss of nitrification (3).  Conversely,
excessive aeration can shear the activated sludge floe
and  result in settleability  problems (4,5).  Improved
process reliability, organic nitrogen removal  efficiency,
sludge settleability, and effluent quality  have been
attributed to automated DO control (6,7).

6.2.2  Economic Considerations
Aeration energy consumption usually represents 50-90
percent of the total energy demand (8) and up to  30
percent of the total operating cost (9) for activated
sludge  plants.  The  incorporation  of effective
automated control of the aeration process can result
in  considerable energy savings  (10-13).  However, as
the complexity  of a  control  system increases,  labor
costs can increase. The design goal is to select the
optimal  control  system,  i.e.,  one that achieves
satisfactory treatment at minimal total cost.

6.2.2.1 Manual Control
As indicated in  previous  chapters,  wastewater
treatment plants rarely, if ever, operate  under steady-
state  conditions. Temporal and  spatial  variations  in
oxygen demand  make  it  very  difficult,  if  not
impossible, for an operator  to manually  manipulate
airflow rates  and  air distribution to maintain  desired
mixed liquor DO  concentrations throughout  a
sustained operating period. This is true even for plants
with  well-designed,  flexible  aeration  systems.
Therefore,  manually-adjusted aeration  systems are
usually operated at a fixed airflow rate and distribution.
Changes are initiated once or twice daily,  at best, and
often  on a weekly or seasonal  basis only. Airflow  is
manually fixed at a  rate high  enough  to satisfy the
oxygen  demand  anticipated  during  peak  loading
periods. This  practice results in  unnecessary  and
costly excess aeration  during extended periods  of
reduced loading.

6.2.2.2 Automated Control
Automated DO control is the  only practical way a well-
designed  aeration  system  can  be  effectively
manipulated  to  satisfy biornass  oxygen  demand,
minimize  operational  problems  associated  with
inadequate  or excessive aeration, and minimize
aeration energy consumption. Generally, the  potential
aeration energy savings  achievable by automatic
aeration or DO control is  25-40  percent  (10,11), but
can  be as  high  as 50  percent  (12,13). Potential
savings are plant specific and depend on plant loading
characteristics,  plant configuration and  process
hardware  design, and the existing level  of  manual
control.

An example of the results achievable with automated
control vs. manual operation is given in  Figures 6-1
and  6-2  (14). The  DO  profiles  of  two  parallel,
"completely  mixed"  basins-in-series aeration trains
are shown in Figure  6-1. For the manually-controlled
                                                  149

-------
 Figure 6-1.  Comparative DO profiles for automated control vs. manual operation.
          10
     o
     8
                                                               	  Manual Train, Tank 1
                                                               —•	•—  Manual Train, Tank 2
                                                               	  Automated Train, Tank 1
                                                               	Automated Train, Tank 2,
                                                10      12

                                                  Time, hr
                                                               14
                                                                       16
                                                                              18
                                                                                      20
                                                                                             22
aeration train, the airflow rate was fixed and the DO
profiles are,  therefore, a  reflection of plant  loading
dynamics. Automated  control  of DO concentration  in
the  other  aeration  train  was   achieved through
continuous manipulation of: 1) aeration capacity using
speed regulation of rotary lobe positive displacement
blowers with a variable-frequency a.c. drive, and 2) air
distribution using automated control valves. Figure 6-2
shows  the  energy  savings achieved through
continuous control of the airflow rate in the automated
train compared with the fixed aeration rate operation
of tho manual train.
As indicated in Chapters  2 through 4, as  the airflow
rate  applied to a  fine pore diffuser increases, oxygen
transfer  efficiency  (OTE)  usually decreases  and
aeration  efficiency drops because  of  increased
pressure losses.  Therefore, continuously  minimizing
the airflow rate while  satisfying the dynamic oxygen
demand through set-point DO control maximizes OTE
and aeration efficiency and minimizes aeration energy
consumption.  The  DO  set-point chosen  for  the
controller  also has  an impact on  aeration  energy
expended.  In respiring systems,  the  dynamics of
reaeration  are well  known  and  identified  by  the
simplified model:
where,

   C
   t
   a
   KLa
           dC/dt = aF(KLa) (C'oo - C) - r
                                     (6-1)
= process water DO concentration, mg/L.-
= time, hr
= (process water K[_a of a, new  ditfuser) -r
  (clean water K|_a of a new diffuser)
= (process water KLa of a  diffuser  after a
  given time  in service) •=• (K[_a of  a new
  diffuser in the same process water)
= apparent  volumetric  mass  transfer
  coefficient in clean water at temperature T,
  1/hr
= steady-state DO  saturation concentration
  attained  at  infinite, time  at  water
  temperature   T and  field atmospheric
  pressure Ft,, mg/L
= volumetric respiration rate, mg O2/Uhr
As outlined in  Sections 3.1 and 4.2.3.1d, the residual
DO concentration  affects the  OTE under process
                                                  150

-------
Figure 6-2.   Comparative energy comsumption for automated control vs. manual operation.


       60
                                        Energy Saved - 38 Percent
                                                                                            22
conditions  (OTEf)  by changing  the oxygen  transfer
driving force (C"m - C). The maximum OTEf is realized
when the driving force is greatest,  i.e., at a residual
mixed liquor DO control set-point of  zero. The positive
DO residual set-point usually maintained for  process
performance reduces the driving  force, lowers OTEf,
and requires a greater amount of airflow to meet the
oxygen  demand. This can represent  a significant
increase  in  aeration  energy consumption.  It  is
important, therefore, to maintain  the  DO residual at
the lowest level possible without adversely  affecting
treatment.  A  detailed  example is  given  in  Section
4.2.3.1d.


6.3 Control Strategy Development
6.3.7 Degree  of Control
How much aeration control is required or desired and
can be  achieved at  an activated  sludge plant is site
specific. The degree of aeration control implemented
can  range from the extremes  of very infrequent
manual  manipulation based on manual measurements
(virtually no  control)  to  comprehensive  automated
control  of  the aeration delivery and distribution
hardware  to  maintain  on-line  measured  DO
concentration at desired set-points  throughout the
aeration trains under dynamic loading conditions.

6.3.1.1 New Treatment Facility
For a new treatment plant, the decision to incorporate
aeration control is straightforward. The capital cost  of
implementing  even  a high  degree  of  automated
control as an incremental cost above that required  to
provide  open  loop,  on-line monitoring  is  a  small
percentage of the total initial cost of plant, generally  1 -
5 percent,  depending on the size of  the facility. As
outlined in  Chapter  4,  the decision is  based on the
desired  process configuration and operational control
goals. Successful  implementation  of  automated
control of airflow and  DO in a new facility requires that
the plant be designed from the  outset with the  intent
of incorporating efficient automated control.

Careful attention to process and hardware flexibility  is
necessary  to realize the maximum benefits  from  a
well-designed aeration control system  throughout the
                                                  151

-------
plant  design life.  The  various  components  of  the
aeration system  must  be designed to  allow  for
operational  flexibility to  meet the  dynamic oxygen
requirements of the system (see Sections 5.3.1.3  and
5.4.3). Of greatest importance in this respect  are: 1)
operational constraints and turndown capacity of the
blowers, 2) air distribution control  hardware,  and 3)
diffuser allowable airflow range.

6.3.1.2 Retrofit Facility
For an existing, manually-controlled plant, the decision
to retrofit  for automated aeration  control  must be
based on one or both of the following objectives: 1)
provision  of more effective  control of  the aeration
process  to minimize  operational  problems, or 2)
optimization of the aeration process to achieve energy
consumption savings.

In either  case,  the  level  or  degree of control
implemented should be based on an incremental cost-
benefit analysis.  The  economic  analysis  should
consider the cost  of retrofitting  the plant for  various
levels of control, training requirements and personnel
changes, the potential for energy savings, O&M costs,
and the intangible  benefits of improved operation  and
plant  reliability. Automated control may also result in
more efficient use  of existing tankage and optimization
of plant capacity. This might permit postponement of
expected plant  expansions and the associated capital
financing expenditure. While this possibility is a  less
obvious  benefit of improved  aeration  and process
control, it  may outweigh the cost  savings that  are
achievable through operating cost reduction alone.

The economics for adding automated DO control  also
depend on whether the  control  system  is being
incorporated as part of an overall aeration  system
upgrade or on a stand-alone basis. The economics of
upgrading the aeration system of a hypothetical  219-
Lts (S-mgd) plant with  higher efficiency aeration
equipment  and automatic DO control has been
evaluated  (15). It was  estimated  that the  simple
payback period for the entire retrofit project would be
5-7 years  depending on  the economic  assumptions
made. It was also determined  that the capital  cost
associated with adding  automatic  control  would be
small  compared  with  the cost  of the  retrofitted
aeration equipment. Yet, the  anticipated savings
resulting from this addition would account for nearly
70  percent of  the expected energy savings  of  the
entire retrofitted aeration system. Additional economic
analysis considerations and procedures are presented
in Chapter 7.

6.3.2 Control Systems
The principles  of  control theory used in the  control
systems examined in  this section  are  available in
standard control  textbooks (16-18).  In  addition, an
EPA  report (10) describes in  detail several  control
strategies and basic techniques used for DO control in
the activated sludge process.
The  automated  control  of  mixed  liquor  DO
concentration to  a set-point value  does  not  itself
require  any in-depth  knowledge  of  biological
metabolism or  activity.  Simply  stated, the  DO
concentration  is  measured  using  on-line
instrumentation and the airflow rate  is controlled in the
zone  of the basin where  the measurement is being
made. The  manipulatable variable in this case  is
adjusted, if necessary, by a  controller directing  final
control elements accordingly.

6.3.2.1 Control Theory
The most  common DO control  strategies  involve
conventional  feedback  or  feedforward-feedback
controllers. A feedback  control loop  is  shown
schematically in Figure 6-3. The DO  sensor signal is
fed to the comparator where  it  is  compared  to a
desired set-point value, and the error is determined.
The controller calculates a control signal based on this
error  using an established algorithm or  control
function. It then sends  an output  signal to adjust a
final  control element,  such  as  a  blower or  air
distribution  valve, to  achieve a change  in  the
manipulatable variable.

If feedforward action  is added  to  the controller, as
illustrated  in  Figure  6-4,  an incoming  process
disturbance (a change  in  influent  flow  rate,  for
example) is  detected  directly through  additional
instrumentation. The controller initiates control of the
manipulatable variable immediately  rather than waiting
until the disturbance is reflected in a change in the
controlled variable. Usually, the two  controllers are
coupled  so that  the  feedforward  portion  of  the
controller paces the action and the feedback portion
serves to "fine tune" or "trim" the feedforward control
action. The process DO response to changing input is
normally slow and can  be  adequately handled  by
control systems using feedback control only.

6.3.2.2 Control Functions
The control signal is  derived  from  the  measured
variable error according to a control law or algorithm.
This can be as simple  as on/off control of the  final
control element to adjust  the manipulatable variable.
For DO control, this could involve  bringing on-line or
taking off-line  one or more  blowers in  a multiple
blower system to  change the aeration rate whenever
the  DO  was  below  or  above  the  set-point,
respectively. In cases where the resulting response of
the process  DO  concentration to this type of  air
delivery  control  is  fast, excessive  starting  and
stopping of blowers can  result. This is undesirable
from  the  standpoint  of both  hardware  O&M  and
energy demand and consumption.

These problems can be minimized by modifying the
controller  to  take  no  control action  within a user-
selectable band about  the  set-point. For example,
where the desired  DO  concentration  is 1.5 mg/L, the
controller could  be set to start up a  blower at 1.0 mg/L
                                                  152

-------
    Figure 6-3.  Block diagram for feedback control loop.
                                                                             O2 Demand

Comparator-
Error
"r
1
Command O2 1
Controller
DO Measured







r
_ __ j

                                                                                                 DO
    Figure 6-4.   Block diagram for feedforward -feedback control loop:
                                                                      input
                                                                   Measured
                                                                             O2 Dt mand


CfJUf

                                                Command
                                                 • Signal
                          02
                          Input
                           DO Measured
                         DO
and to shut off the blower at 2.0 mg/L Thus, the DO
concentration would be  allowed  to fluctuate within  a
"band"  about the set-point. This type of controller is
called a "dead band" controller and results in a longer
cycle time for blower starting and stopping. There is,
however,  a corresponding decrease in the degree of
control.

Usually, however, the control  signal is derived  from a
continuous  controller such as a  proportional-integral
controller.  For this  controller, the  control signal is
made up  of two  components: one proportional to the
error and  one proportional to the integral of the error.
Proportional  control alone can result  in  an  offset or
steady-state error.  The addition of the integral action
term  (reset)  compensates  for  this  error.  The
mathematical function describing this control signal is:
      Change in Control Signal =

                     1 ft
           K 1)0,..+ —   DO....At
            P   d>f  K J o   M

where,

   DO(jjj  = DC/sei-point " DOmeasured
   t      - time
   Kp    = proportional gain
   KI    = integral time constant
(6-2)
In some  control applications  where  very  rapid
response  is  required, a  third parameter,  the  time
derivative  of the error,  Ko(dDOdjf/dt) (where  KD  =
derivative  time), is added  to the above equation. This
is then termed a  proportional-integral-derivative
controller. Usually  proportional-integral control  is
adequate for aeration control schemes.

The  "tuning" of the controller involves  selecting the
controller  constants  of proportional gain and integral
time to obtain  the  desired corrective  action  of the
controller.  Mathematical  models and  tuning  rules as
outlined in general control theory references (16-18)
can  be used to obtain  an  initial  set  of constants.
However,  once the plant is operational, step testing of
the system is usually used to refine the  estimates for
the process  gain  and reset  time.  Sometimes, more
thorough  process identification  experimentation  may
be  required  to  develop  a process and disturbance
model  that can  be used  to design and optimize the
control algorithm.

Generally,  for  a given set of  operating conditions,
there is an  optimum  set of controller parameters.
However,  for DO concentration control, where  the
dynamics  are nonlinear and  time  varying, the tuning
parameters  that are optimum for  maintaining  contra!
under  peak diurnal  loading periods may not sustain
                                                   153

-------
that level of control over an extended operating range.
A compromise  between  ideal control  and  the
maintenance of controller stability often results in the
controller  being "detuned" to a  degree to  achieve
adequate  control  over a  wide  range of  operating
conditions. The controller is then periodically retuned,
as necessary, to accommodate gross changes in DO
dynamics that can  occur on a seasonal basis.


6.3.3 DO Control  Strategies
6.3.3.1 Conventional Control
In aeration basins designed to be completely mixed,
the oxygen demand is relatively  uniform. Therefore,
automated  control  of DO concentration in the  basin is
based only on feedback from the DO sensor. In many
other cases,  however, multiple  control  loops are
required to control  DO concentration  effectively.

As indicated in Section 5.3.1.3b, the spatially-varying
oxygen demand along a plug flow reactor requires a
nonuniform rate of oxygen  transfer to  accomplish
uniform DO control. For a steady-state condition, this
can  be achieved  by tapering the diffuser density  in
grids down the length of the basin. Automated  air
distribution control valves can be installed to regulate
the airflow  rate to, and maintain the  DO concentration
in,  each grid at  a desired  set-point.  If this is not
practical, the air distribution profile can be established
with manually-adjusted  air distribution valves and the
total airflow  to the basin automatically  regulated  to
maintain the desired DO profile down the length of the
basin (19).

Figure 6-5 shows  a typical DO control system for a
tapered diffused air delivery,  compartmentalized, plug
flow aeration  basin. A proportional-integral  controller
determines the required change  in airflow needed  to
restore the DO set-point and  "cascades" a set-point
for airflow rate to a  separate airflow control  loop.
Because the  control loops for  each basin  operate
independently, an  increase in  airflow to Basin 1 will
result in decreased airflow  to and decreased DO
concentration in Basin  2. The  system will attempt  to
compensate  for this at the next control interval. This
can result  in the controllers continually "hunting" and
perhaps even cause instability in the  control system.

Tha addition  of a pressure  control loop minimizes
"hunting."  This   loop  ensures  the existence   of
adequate airflow for both basins and  system operating
pressure set-point is minimized to achieve  maximum
energy  savings by always maintaining one  of the
valves in  its  "most  open"  position.  The  control
algorithm  usually  incorporates a  minimum  allowable
airflow  set-point  to   maintain  minimum  mixing
requirements in the reactor.

For modern digital  controllers, the timing of the control
loop action (control Interval)  is  critical.  It must  be
matched to the time constants of the variables being
controlled to  accommodate process  dynamics,
process dead time, if any, and disturbance or noise in
the process  or  primary  measurement signal that
should be ignored or rejected. For the case above, the
DO controller is the slowest loop in  the system. This
is because the aeration  process dynamics  are
relatively slow,  and it can take up to 30 minutes for
the DO  to  reach a  new  equilibrium  value after  a
change in airflow has  been  initiated, even  under
steady-state conditions. There  is little point, therefore,
to run the DO control loop more  often than once
every few minutes.

The  individual  air distribution control  loops have  a
much faster response time, and the control interval
may be on the order  of several  seconds. A pressure
optimization control loop will adjust  the final control
elements on the blowers to regulate the total airflow to
the system and maintain the desired  system  pressure.
This controller  must  account for the  disturbances
generated by the control  valve. The control interval
will, therefore, be slower (about 1-2 minutes) than the
air distribution control loops, but  not as slow as the
DO control loop. The proper tuning  of  these several
control loops is necessary to  obtain desired system
response.  This task  can  become quite  tedious
because  of the various  process  and controller
interactions involved.
6.3.3.2 Advanced Control
Programmable  digital controllers have much more
flexibility than traditional analog controllers and  are
widely used today (18). These controllers  can readily
compensate  for  the considerable  dead  time that
usually delays the response of the control variable to
disturbance or control action. They  also facilitate the
implementation of more advanced controllers that can
extend the empirical  engineering solutions  exemplified
by proportional-integral control.  Digital filtering of raw
signals  from  process  instruments  can eliminate  or
minimize controller disruption due to signal  noise.

The computing power of the digital  processor  can  be
used  to obtain important additional  information about
the process from on-line measurements. For example,
from  the DO  mass balance  identified  above  as
Equation 6-1, if the relationship between aF (K|_a) and
airflow rate (see Section 3.4.2.2) was known  and the
DO controlled, the oxygen uptake  rate (r) could  be
estimated directly  from the airflow  rate.  This  would
yield valuable process information to the operator on a
continuous basis.  The  accuracy of this  estimation
depends on the assumption that the parameters in the
established aF (Ki_a)-airflow  rate  relationship have
remained  constant over time. In reality, however,  the
aF (Ki.a)-airflow rate function may  change with time
due to changes in a and F, especially for fine pore
systems. Therefore, the change in airflow rate may  be
reflecting  a change in aF (K|_a) rather than a change
in r.
                                                   154

-------
    Figure 6-5.  Two-stage DO control system for a compartmentalized plug flow aeration train.
                            Pressure Optimization
                                Set-Point
                                                                                     Influent
                                                                                     Effluent
                                               Notes:  PI   -   proportional-integral controller

                                                     PT   =   pressure transmitter

                                                          =   airflow measurement transmitter


                                                          =   control valve
As  a result,  the  aF(K[_a)-airflow relationship  must be
checked and updated, if necessary,  on a  continual
basis.  This  information would also be very  useful in
determining  the  need for diffuser  or air  delivery
system maintenance.  These  determinations  can be
made off-line by  monitoring the dynamic response in
DO to  a change in airflow rate under conditions where
r is assumed constant and independent of  DO
concentration. This can also be conducted on-line by
temporarily  suspending  DO control,  "bumping"  the
airflow rate,  monitoring the DO  response dynamics,
and  calculating o.F(K|_a)  directly. Again,  this requires
that  the  respiration  rate  be constant.   The
determinations should therefore be made  during  off-
peak loading periods when  respiration is assumed to
be largely endogenous and relatively constant.

These  assumptions,  however,  may  not be  entirely
correct. A means to  distinguish  and simultaneously
estimate both aF(K|_a) and r is necessary to  minimize
the  errors  and  maximize  the potential   of  this
technique. This has been successfully achieved  using
advanced control techniques described below.

Advances in  industrial  process control have led to the
development and commercial application of  adaptive
controllers otherwise  known  as  self-tuning  or  auto-
tuning  controllers.  With  these controllers,«the
controller tuning  parameters can  be  calculated from
measurements of process response to control action
and  updated recursively as  process conditions and
DO dynamics change. The two interacting loops of a
self-tuning  controller  are shown  schematically  in
Figure 6-6.  The  updated tuning parameters of the
conventional controller are obtained either directly  or
determined  from process parameters that have been
estimated recursively from the process response (18).
This  minimizes  the  effects  of  time-varying
disturbances such as seasonal temperature changes,
changes in  the oxygen transfer coefficient  due  to
diffuser  clogging  (F  effects),   or  wastewater
composition variation (a effects) on the controller.

Figure 6-6.  Block diagram of a self-tuning regulator.
      Design Calculations
      Regulator
      Parameters
 Command
                   	._-J
                     Control "
                     Control
                     Signal
Process
               Output
The adaptive  controller  can  be  implemented  for
feedforward-feedback situations and  also where  the
sequential changes in the manipulated  variable must
                                                   155

-------
be constrained. This serves to minimize wear due to
excessive control action on mechanical components
such  as motor drives and valves. Self-tuning  or
adaptive control of  DO concentration  has  been
successfully  implemented at the Kappala  sewage
works in Stockholm (20).

Through special handling of an adaptive controller, the
need  identified above has  been  addressed  and a
means  developed  to  estimate simultaneously  the
process parameters aF(Ki_a) and r while  maintaining
effective DO control (21). Estimation of time-varying
parameters requires a high degree of excitation of the
system. This contradicts the regulation purpose of the
controller, and a compromise must  be  reached to
ensure that the oscillations of the controller are large
enough to  provide the  estimator  with  parameter
information, but small enough to allow for effective DO
regulation.  The  airflow rate is disturbed in every
sampling interval, which causes the DO concentration
to deviate  from its set-point, but  yields  enough
information for the  estimator to identify  both  aF(Kta)
and r.

This successful adaptation has important implications
for advanced aeration control. It provides not only an
on-line determination of the  oxygen uptake rate  from
measurements of DO concentration and airflow  rate,
which is useful for process control, but also serves as
an on-line  monitor of aeration  system performance
(aF(Ki_a)].  Changes  in the value of aF(Ki_a),
determined  on-line, can  indicate both  wastewater
composition variations (a effects) and diffuser clogging
(F effects).  These  indications could potentially  be
used  to identify plant  O&M needs such  as  diffuser
cleaning.
6.4 Control System Components

6.4,7 Instrumentation
The successful  operation  of  installed  on-line
instrumentation is critical to the successful application
of  automated aeration  control. Incorrectly applied,
installed,  or  maintained  instruments  can render an
otherwise  well-designed process control  system
useless. Not only will the objectives  of the  control
system not be met, but serious and costly operational
problems  can  result.  As such,  sensors  are often
considered the weakest link in the control system.

The control  strategies  discussed  in  Section 6.3.3
above  require DO and,  in some  cases, airflow rate
instrumentation.  Other  process instrumentation
requirements for this application generally include air
pressure,  air temperature, and air distribution valve
actuator position monitors, instrumentation required to
monitor and control blower operation depends on the
specific requirements of  the blower control  system.
Functions monitored may include speed, blower inlet
vane or valve position,  current or power  draw,  and
suction and discharge temperature and pressure.

The  above  instruments are  all,  with the possible
exception of DO monitoring, considered standards  in
the chemical and industrial process control  industry. A
large body  of  literature exists concerning  the
application, installation, calibration, and  maintenance
of this instrumentation. Some of these  references (22-
26) are  considered  "standards" in the industry, and
this information should be reviewed  in  detail  before
specifying or procuring any of this instrumentation. It
is essential  that  the  definitions  used  to describe
instrument characteristics  and performance  — such
as range, span,  accuracy,  hysteresis,  repeatability,
sensitivity, dead band, resolution,  response time, and
drift  —  be understood  to  appropriately specify and
apply these instruments.  Considerations  crucial  to
successful  application  of the  major groups  of
instrumentation are reviewed briefly below.

6.4.1.1 Dissolved Oxygen
Without  doubt, DO monitoring equipment is the most
notorious of all on-line  sensors used in wastewater
treatment. It is most  often cited as  the main reason for
DO control system  failure.  While  system failure  may
be due  to faulty  equipment in some  cases,  it is  at
least as likely to be  due to improper application, poor
installation, lack of attention and maintenance by plant
personnel, or a combination thereof. These problems
usually have  at their root  a lack of understanding  of
the instrumentation, insufficient attention  to  details
during the design phase,  or  lack  of  commitment  to
keeping instrumentation operating properly.

a. Principles of Operation
Virtually  all  DO  probe? available  today  are
electrochemical cells that contact the fluid  through an
oxygen-permeable  membrane.   The oxygen
concentration  in  the  cell  electrolyte reaches
equilibrium with that in the  bulk fluid, and  a chemical
reaction induces  a  change in voltage across the
electrodes. The subsequent current flow  across the
electrodes produces an  electrical  signal  in proportion
to the oxygen content of the fluid.  One device has no
membrane, and the probe  uses the bulk  fluid as  its
electrolyte.

fa. Selection and Installation
Instrument selection should only  be undertaken after
application  constraints  such  as  environmental
conditions, operating ranges, and design  requirements
have been identified. Existing comparative  instrument
test  data and the experience of  other  users under
both bench and field conditions should be taken into
account (26,27),  The  user-based, nonprofit North
American Water and Wastewater  Instrument  Testing
Association  (ITA)  conducts structured evaluations  in
accordance with strict peer-reviewed test protocols.  In
1988, ITA completed a comparative test  of seven DO
                                                  156

-------
measurement systems. Results  are  available  to
members directly from ITA (28).

A field protocol for selecting, locating, and maintaining
DO sensors has been developed and,  based on a 60-
day  test of equipment from several  manufacturers,
specific recommendations  regarding  probe location,
calibration,  and maintenance have been offered  (27).
The manufacturer  usually  stipulates  installation
conditions;  however, in general, DO  sensors  should
be installed in the aeration basin in a "dip" mode, i.e.,
directly in  contact with the mixed liquor  and easily
removed and  serviced by operations personnel.  This
minimizes  time delays and  the  maintenance  of
ancillary equipment associated  with flow-through cell
configurations. The  mounting  hardware should be
accessible  and of a quick-release type to  allow for
maintenance (27).  It is also  important to  provide
flexibility in mounting hardware and signal  cabling to
permit relocation of the sensor in the basin (10).

c.  Field Verification, Calibration,  and Maintenance
Each  manufacturer has  a specific  recommended
calibration procedure. Usually,  it is a  simple one- or
two-point calibration. As part of the field  verification
procedure for the installed instrument,  accurate output
should be verified  over the entire expected operating
range. Other performance checks, such as  response
time, hysteresis, and  repeatability should also be
made.

In  the environment of the bioreactor, the tendency for
fouling of the membrane is of substantial concern as
foulants affect membrane permeability  and, hence, the
accuracy of the DO measurement. Biofouling in the
form of a slimy biofilm is usually the most  significant
problem. It can  usually be effectively  removed  from
the membrane surface  by  careful wiping with a wet
tissue or soaking in a-10-percent HCI solution, thereby
restoring probe  performance. Mineral deposits or oil
can change the permeability of the membrane and are
not easily removed by cleaning.

Frequency  of cleaning will vary  depending on process
loading characteristics and operating configuration. To
minimize maintenance requirements, it is important to
clean the probes only when necessary.  Checking the
process  probe for conformance  with  a  reference
probe  can  indicate when probe servicing is required.
The reference probe must be accurately  calibrated
and  have  a time response  similar to  the process
probe, and the linearity  of  both  meters   must be
known.

The frequency  for  carrying  out the  conformance
procedure  can  be  optimized  with  experience.
Generally, the process  probe should  not be touched
until  successive conformance  checks  show  a
significant deviation - often taken to be  0.4-0.5 mg/L.
The criteria  ultimately  selected  is  site   specific.
Cleaning will  normally restore probe performance. If it
does not,  intensive recalibration or even replacement
of the electrochemical cell may be necessary.

Total  annual labor requirements for  conformance
checks, cleanings, and calibrations for four on-line DO
probes are 80-120 labor-hr at one site (29).
6.4.1.2 Airflow Measurement
Airflow  monitoring  equipment  is  an  important
component of  most  aeration control  schemes.
Effective distribution of  airflow to several points in the
aeration basin requires  accurate airflow measurement.
Determination of  mass airflow  rate is necessary  to
ensure  the correct quantity  of  oxygen  is  being
delivered.  The  importance of this  measurement
demands that the required instrumentation be carefully
selected and installed to maximize the probability that
performance expectations will be met.

a. Principles of Operation
The  various airflow  metering  systems  widely  used
today are based  either  on differential pressure across
a control element or mass  flow. Differential  pressure
meters all use the same basic relationship to measure
flow rate:

      Flow Rate =  (Velocity)  (Area) (Factor)   (6-3)


The  differences  between differential flow meters  is
largely  a function   of  how  the   velocity  term  is
determined. Plate orifice and  Venturi meters  use a
constriction to produce  a measurable differential head
or pressure.  This  differential pressure  is  then
converted to velocity using fundamentals of mass and
energy  conservation. To obtain a  measurement  of
mass flow, temperature and  pressure  corrections
must be applied.

Mass flow meters generally operate on the principle of
a hot wire  anemometer. A wire is placed  in the flow
stream with an electric  current applied to maintain the
wire  at  a preset temperature. The  rate of cooling  of
the wire, based on the  current required to maintain it
at the preset temperature, is proportional to the  mass
flow rate of air.

b. Selection and  Installation
A large number  of  airflow instruments are available,
each  with  its  own  advantages and disadvantages.
These  are  discussed  in  detail in  the  references
previously  cited  (22-25).  Care must be taken  in
selecting the airflow  meter to not unnecessarily
constrain the air  delivery system in  terms of  headloss
or turndown requirements.  It is also  very important
that the meter  be  sized to accommodate  only the
immediate future  expected  ranges   in  airflow.
Oversizing  and attendant poor  flow measurement will
occur if the range used to size the meter is based too
far into the future.
                                                  157

-------
Careful installation is needed to avoid jeopardizing the
performance  of an  appropriately  applied and  sized
meter.  This  includes  closely  following  the
manufacturer's recommendations  for approach and
downstream conditions to ensure the accuracy of the
velocity determination.

c. Field Verification, Calibration, and Maintenance
The fundamental principles outlined for commissioning
and maintaining  DO sensors  also apply to airflow
monitoring equipment.  The  meter usually comes
calibrated to the user's specification from the factory.
Once  the meter  is  installed  in  the system,  a
conformance check should be made, if possible. Even
an  approximation  of  conformance using blower
performance  curves is worthwhile and  can  likely be
done for  the entire range  of the meter (30).

Since the sensors are not in contact with wastewater,
much  less  maintenance  is  required, once  their
performance is verified, than for DO probes. Periodic
conformance checks are  recommended,  however.

6.4.1.3 Pressure  and Temperature
Pressure and temperature measurements are used in
the aeration control system to monitor blower suction
and discharge  conditions.  They also  provide on-line
information for converting volumetric field flow rates to
Standard  flow  rates. These  instruments  are  standard
throughout the process  control industry and are  not
discussed in  detail  here.  The  same  fundamentals
described for DO  and  airflow  rate  measurements
concerning  selection,  verification, calibration,  and
maintenance  apply to  these instruments. It is  most
important to incorporate  instruments that are suitable
for the environment encountered at a wastewater
treatment plant and able  to meet the objectives of the
application  in  terms  of operating   range  and
performance.

6.4.2 Final Control Elements
The final control elements in any  control system  are
required to carry out the  desired control action.  For  a
DO control system, this  means adjusting the delivery
and distribution of air through manipulation of blowers
and air control valves.

6.4.2.1 Air Delivery Blowers
The two major classifications of blowers normally used
in aeration control systems are  rotary  lobe positive
displacement  (PD) blowers and  centrifugal  blowers.
The operating  and  performance  characteristics  of
each have been identified in Section 5.5.3. Additional
details are available elsewhere (10,19,31,32).

The effective  and efficient  control of the blower  or
blowers in an aeration control system is  almost totally
dependent on  good design. The blowers  must be
sized such that an operating map of the air delivery
system not only   meets  the expected  variations  in
process air requirements, but also maximizes blower
efficiency throughout the delivery range (see Sections
5.5.3, 5.5.5.3, and 6.5). The control  system must be
designed  to  not  only   minimize unnecessary
disturbances  in airflow rate  and DO concentration, but
also to  maximize the energy savings extracted  from
variable air delivery. The control strategies for the two
different classes of blowers are  different because of
their principles of operation.

a. Positive Displacement Blowers
As indicated in Chapter 5, the PD blower is essentially
a  constant-volume, variable-pressure machine.  The
pressure the blower runs  at is  dictated  by system
requirements. Historically,  changing  the number  of
blowers in  service and  controlling blowers  with
multiple-speed  motors  were  the only  options for
controlling  PD  blowers. Today,  use of  a  variable-
frequency a.c. drive also allows the PD blower to run
as  a variable-volume,  variable-pressure  machine.
Controlling and saving airflow, however, must translate
into  energy  savings.  Variable-frequency  drives  can
consume up  to  10  percent of the energy applied as
heat. This premium must be considered carefully as,
sometimes,  it  can  quickly  effect  the  economic
incentive for automated DO control.

From an operations viewpoint, the variable-frequency
drive should  not be permitted to lower the  blower
speed  below the manufacturers  recommended  limit.
Less heat  dissipation  at  lower-than-recommended
speed could  result  in overheating of and  damage to
the blower. At the other end of the speed scale, most
a.c.  variable-frequency drives can   operate  at  110
percent of normal frequency, thereby  increasing motor
speed  and,  hence, blower  speed  by  10 percent
overall.  While this places  a greater load on the motor
and  consumes additional  energy,  it is  one  way  to
increase the  capacity and operational flexibility of the
PD blower.

It  is very  important  that  the  blower  and drive
manufacturers understand,  in  detail,  the application,
control  strategy, and anticipated  operating conditions
of the  air delivery system.  This   ensures proper
integration of appropriately  sized motors, drives,  and
monitoring instrumentation.

b. Centrifugal Blowers
Usually, the volume output from centrifugal blowers is
manipulated by adjusting  speed or inlet  guide  vanes,
or throttling either the inlet or outlet valves.  These
approaches  vary  in  terms  of difficulty,  energy
efficiency, reliability, and effect on stability (see Table
5-3). The control of these blowers is  also complicated
by the need to stay between the  low  output operating
limit or surge point and the maximum  blower hydraulic
limit. Often,  large,  complex blower  packages come
with built-in  controls  to  prevent surge conditions.
Detailed explanations of various  types of centrifugal
blower  control  schemes  are available  elsewhere
(10,19,31,33).
                                                  158

-------
The  same degree  of involvement of the manufacturer
with  the designer  and instrumentation  and  control
specialist  as for a PD  blower  application is also
essential for centrifugal blowers.

6.4.2.2 Air Distribution
Control  valves  are used to distribute  air to aeration
basins,  headers,   and grids. Control  valves  are
designed to  maintain  a  relationship between airflow
rate  through  the valve and valve  travel as it is varied
from 0  to 100  percent of its "most  open"  position.
Valves  are  thus characterized to  provide  relatively
uniform  control loop stability  over  the  expected
operating range.

A  valve with a linear flow  characteristic  yields  an
airflow rate that is  directly proportional  to valve travel.
For  an  equal-percentage  valve, equal increments of
valve travel produce equal percentage  changes in the
existing flow. Equal-percentage  valves are  typically
used on pressure  control applications where  a large
percentage of the pressure drop is normally absorbed
by  the  system itself, with  only  a  small  amount
available at the  valve.

The  sizing  of  valves, not  unlike airflow meters, is
critical  for  good air  distribution  control.  The valve
should  be  sized  to  control  over the  immediately
foreseeable  operating  range.  In some cases, valves
with  replaceable "trim" are  available, which  provides
the flexibility to  handle different ranges of flows.

A valve and actuator may have a  degree of hysteresis
associated  with their  response  to a  control signal.
Hysteresis is a  measure of the difference in the valve
response for a  particular input signal, depending  on
whether the  new   position  was  approached from  a
more open  or more closed position.  If  such  a
condition exists, it must be  accommodated in  the
control algorithm to obtain good control.

It  is  also important to note  that,  while  the valves
should always be  maintained  as  open as possible to
minimize overall system pressure and aeration energy
expenditure,  it  is necessary  to sacrifice some head-
loss to  have control.  This trade-off requires constant
balancing. Energy  losses due to increased headless
can  be minimized with  properly sized  and  applied
valves.  The air distribution designer should work with
the  valve supplier  applications engineer in selecting
and  specifying these valves and their  operating
characteristics.

6.4.3 Controller
Control  hardware changes so rapidly  it is difficult to
stay current  with available equipment  options.  In the
past, real-time  minicomputers were the only realistic
option for on-line process control  of a treatment plant.
Now, desktop  microcomputers  are capable  of the
same if not a better level of control when coupled with
industrial  programmable  logic  controllers in  a
distributed  network. Industrial programmable logic
controllers  are much more sophisticated in terms of
digital control than they were just 2 or 3 years ago. As
a result, the distributed  system offers a high degree of
reliability and control capability.

An in-depth discussion and evolution of control hard-
ware is beyond the scope of this  manual.  Manual of
Practice  SM-5  (34) is an  excellent  reference for
details on  the types of control system options now
available.
6.4.4 Software
A detailed discussion of the software required for an
effective automated aeration control system is beyond
the scope of this document. Manual of Practice SM-5
(34) is a good primer on software  categories and the
important characteristics of each. Generally, software
is  classified as "system"  software and "application"
software. System software controls and directs opera-
tion  of  the  computer and  supports operation of the
application  software.  It  includes the  computer
operating system software, programming  languages,
and utility programs.

Application  software  performs  specific tasks required
by  the  user.  This  includes process  control  tasks
ranging from  simple pump  scheduling to  more
complex tasks such  as on-line,  sensor-based  DO
control,  solids  retention time (SRT) or solids inventory
control,  and energy  optimization. Generic control
packages,  off-the-shelf  packages  for  data  base
management,  spreadsheets, word  processors,  and
preventive maintenance packages  are also types of
application software.

The desired control  strategy  cannot be  effectively
implemented if the application software designed to
achieve real-time  control of process operations such
as aeration is faulty.  Careful attention to the following
general  considerations will ensure successful imple-
mentation and effective use of the control system by
the plant operating  staff.

Requirements  for the application software should be
clearly stated and understood by the designer, control
and  instrumentation specialists, O&M personnel,  and
computer specialists and programmers. Employing a
team approach to  define and  design process control
goals and strategies,  operator interface with the data
acquisition and control system, and reporting systems
is  recommended. The team should  also  select  and
specify  appropriate  computer/controller  hardware,
system  software,  and on-line  instrumentation.   The
software must be designed to be  flexible  and "user-
friendly." Wherever possible, the software  should also
be designed to be modular, menu-  or graphics-driven,
and  easily  modified. This requires the continual
involvement  of  the  users  in  the  design   and
implementation of software.
                                                  159

-------
Software documentation must be accurate, complete,
and readily understood. The "Operations Manual" and
"Computer  Control  System  Manual"  should  be
compatible and  include,  for  each  application,
"readable" process  and  control narratives; control
loop  block  diagrams  that  identify the  specific
instrumentation and  control  hardware  to  be  used;
technical guidelines for data acquisition, manipulation,
and storage; interlink maps and  diagrams  of various
computers,  controllers,  and system software; and
examples of  the graphics  interfaces  and  report
templates.  Finally, extensive  user/operator-interface
documentation  is  required  to  illustrate  how  to:  1)
change set-points, scales or ranges of instrumentation
inputs,  and control  loop  and  controller parameters,
and 2) use the control system in general.

It is important that the control system be designed to
permit override or suspension of automated control
and  allow  manual operation. Further,  the control
system should should  default to a  safe  operating
condition that maintains  process integrity in the  event
of control system failure. For example, air distribution
valve  positioners should default to open the air valves
if the  system fails, with the valves then being operated
manually until the problem is corrected.


6.5 Aeration Control Example
The example design presented in Chapter 5 for the air
diffusion and supply system  of a 'plug flow activated
sludge plant will be expanded to illustrate the design
considerations  necessary  for  the   successful
incorporation of an effective aeration control system.

The general layout of the process  air piping for this
example is presented in Figure 5-14. For this design,
the diffusers in each plug flow reactor are arranged in
three  grids or zones  to allow for adjustment of airflow
down the length of the reactor to more closely match
the anticipated spatial  oxygen demands.  As  stated
previously, the degree of aeration control implemented
can  generally  range  from  the  extremes of very
infrequent manual manipulation based  on manual
measurements  to comprehensive, automated, set-
point  DO concentration control.  These options also
exist for this example.

Ideally,  a  set-point  DO  concentration would  be
maintained under dynamic loading conditions in each
of the  12 aeration zones  shown in  Figure 5-14 by
automatically  configuring  the  blower system  to
efficiently meet total air requirements and manipulating
the air distribution control  valves to vary the air
delivery rate to each aeration zone. This would  result
in a rather complex control system. The least complex
control  system for this  example would automatically
manipulate the  blower configuration and  total air
output to the four aeration trains based on maintaining
a desired DO set-point at  one point in one  of the
reactors. The desired DO concentration  profile  along
the length  of each reactor would  be maintained by
periodically manually  adjusting  air delivery  to  each
zone.

For this example design,  several control  options lie
between  these  extremes,  each  representing  a
compromise between  the  level  of control  achievable
and the complexity and cost of the  control system.
These options are outlined in further detail below.

6.5.7 Air Delivery Control
For this example, the  basic  air delivery control
hardware  requirements  are similar  for virtually all
anticipated  control options. The four 1,320-L7s (2,800-
scfm) blowers selected  in  Chapter 5 were  sized to
accommodate the diurnal, seasonal,  and yearly
variations in total air requirements expected when the
plant  reaches  its  design  loading. The  air  delivery
control  strategy  for  any  option  considered  would
continually  optimize the number of blowers in service
and their operating points to efficiently deliver the  total
air required at any time in the design life of the facility.
This strategy is essential  to  achieve  the  goal of
maximized  energy savings through effective airflow
control. Often,  a  well-designed control  system
maintains DO concentration set-points as desired, but,
because of inefficient  operation  of the blowers,  does
not realize its full potential aeration energy savings.

6.5.1.1 Blower Operating Map
Manufacturer blower curves  should  be  used to
develop a detailed operating map for configuring the
number of  blowers in service and their  individual or
collective   operating   points to  achieve  the  most
efficient air delivery possible at  all times. This  map
should  incorporate the  effects  of  environmental
conditions such  as temperature  and  humidity. Once
the blowers are installed, it is possible to fine tune this
operating   map  and  determine any  site-specific
operating  limitations through  measurements of
temperature, humidity, pressure rise,  power  draw,
speed, and overall airflow.

Mathematical and empirical relationships  incorporating
these parameters can  be used to optimize the on-line
configuring of  the air delivery  system.  The blower
manufacturer  application  engineer  should  be
requested   to confirm the  most  efficient operating
configurations to meet any  operating set-point.  Any
operating  limits or  restrictions  identified  by  the
application engineer should be strictly observed.

6.5.1.2. Start/Stop Control
For any aeration system, careful attention must be
given  to blower start/stop control. A well-designed
automated control system is able to use the maximum
range of a blower, or  combination  of  blowers, to
minimize the  necessity of bringing on-line  or  taking
off-line additional  blowers.  Implementation of
automatic  start/stop of blowers  is straightforward. In
some control systems, however, operator approval is
                                                  160

-------
a required additional judgement step prior to start-up
or nonemergency shutdown of any blower to minimize
starting  and stopping  of blowers.  In other cases,  the
control  system only  flags  the  need  for  bringing
additional blowers on-line or taking some off-line and
the actual operation is carried out manually.

To  minimize operational problems, a blower (whether
automatically or manually initiated and controlled) is
usually  started off-line and  brought  up to operating
conditions by  delivering air  through  either a recycle
loop or vent  valve. Only then  are the appropriate
isolation valves  manipulated  to  allow the  blower to
discharge into the common header and be integrated
into  the control loop. Under certain operating  and
start-up conditions,  the  main  header operating
pressure may have to be reduced to permit the blower
to open its check valve and come on-line. In addition,
to minimize  energy  demand  charges,  it  is  often
necessary to reduce air delivery, and thereby power
consumption, of the running blowers prior to  bringing
a  new  blower on-line.  Finally, the  control  strategy
must respect  any  restrictions on minimum  run time
and time between starts.

The above  considerations are machine and facility
specific. For the design example, specific operating
requirements  for  the selected  blowers would  be
incorporated  in  the initial  control software  design.
They would be fine tuned following system  start-up
and  after a modest level of operating experience  had
been obtained.

6.5.1.3 Variable Capacity Control
For this example, the  variable delivery capacity of the
operating blowers will  be  controlled  by  adjusting  the
blower  inlet guide  vanes.  The  selected  aeration
system  is designed to operate at relatively constant
system pressure,  which  results  in  a  nonlinear
relationship  between guide vane  position and blower
throughput  However,  blower power draw or motor
amperage vs. throughput is nearly a linear relationship
and  will be  used to control  the inlet guide vanes as
shown generically  in  Figure  6-7 for one of the four
blowers.

Where  multiple air  valves are  manipulated  by
independent  automatic  DO  controllers,  generally in
more complex  control  systems,  a  main  header
pressure controller   is  added  to  minimize  the
disturbance  of air distribution control on the blower
controller.  The pressure  controller cascades  the
amperage  or  power  draw  set-point to  the amp
controller as shown in Figure 6-8 for one of  the four
blowers. The addition of a variable pressure set-point
optimization routine  to further  maximize  aeration
energy  savings by  always  minimizing  the header
operating  pressure may be justified  in  this case.
However, because  of the relatively small size of the
blowers in  this example, this refinement would most
likely be justifiable  only after the  design  capacity of
Figure 6-7.   Blower inlet guide vane control schematic.

                ^             From Process

         •	^     Amp
               Controller
                                      To Process
    Motor
    Amp
    Draw
 Vane
Position
Guide
Vanes
 Set-
 Point
                        Blower
        Air Inlet
the plant  was  being  approached  and operation  of
three blowers  was  required  to  meet  the  normal
oxygen demand.

6.5.1.4 Surge Protection
When a centrifugal blower  cannot develop  enough
pressure  to  overcome  the downstream  process
pressure, the blower can begin to surge - a condition
in  which momentary  reversible  pulsing of air occurs
inside the compressor.  Since  this condition  can be
particularly destructive to centrifugal  blowers, surge
protection is  essential.  Most blower  manufacturers
provide independent surge protection control systems;
however, additional levels of surge protection can, and
should,  be incorporated in the overall blower control
algorithm. Surge can occur at certain high-pressure or
low-throughput conditions. Therefore, careful  attention
to  the interaction  between  air  distribution  and air
delivery controllers  is necessary  to  minimize  line
pressure.

A  minimum  blower amperage  limitation   that
corresponds to  the above critical operating conditions
is  often  used  for surge protection. However, when
inlet  guide vanes  are  manipulated  to  control
centrifugal  blower delivery,  as  in  this example,  the
surge point varies  with  each  guide vane  setting.
Typically, when  guide  vanes   are manipulated  to
reduce blower throughput, the motor amperage draw
corresponding  to the surge point is  also reduced.
Care must be taken, therefore, not to set  the absolute
minimum  amperage  too high,  thereby  reducing
effective blower turndown  under  normal operating
conditions. The  relationship  between  guide  vane
position and  surge point amperage  draw  is often
nonlinear  over  the  range of guide vane operation.
Once established,  it can  be  used with  on-line
measurements   of blower  differential pressure  or
                                                  161

-------
    Flguro 6-8.   Blower control schematic.
                                                                    Multiple Air Distribution Control Loops
                                                                      r
            r— »
      Motor
      Amp
      Draw
 Vane
Position
                                                  Header
                                                   Line
                                                  Pressure
Guide
Vane
Sel-
Poinl
             Intel
                           Blower
                                                                                           NX
                                                                                           S\
                                                                                           \/
                                                                                           /\
                                                                Airflow Measurement
individual blower  discharge  airflow to  predict  and
appropriately update safe operating limits.

6.5.1.5 Blower Performance Evaluation
Provisions necessary  to periodically  evaluate the
performance of the blower control system should be
incorporated in  the initial design.  This  is particularly
important in this case where multiple blowers operate
in parallel. At various periods in the plant design life,
the operaling objective may be to either balance the
load among  blowers or maximize  air delivery overall
efficiency in  accordance with the  operating map.  To
confirm  that  these objectives  are being  achieved,
provisions for measuring  individual blower  operating
parameters must be made. This requires inclusion of
appropriately located standard pipe taps  in the suction
and discharge piping of individual blowers to facilitate
temporary installation of instrumentation for measuring
such  variables as temperature, pressure, and airflow
rate.  These  data  provide a  means for  assessing
Individual machine as  well  as overall  system
performance.

6.5.2  Mr Distribution Control
Control valves are required for effective  distribution of
air  to the individual zones in  each aeration train and,
                                        depending  on the  control system,  to the  aeration
                                        trains themselves. Whether automatically or  manually
                                        adjusted, the valves must be properly  sized  and of  a
                                        type suitable  to  minimize  pressure  headless,  yet
                                        maintain controllability over their anticipated ranges of
                                        airflow. The manufacturer applications engineer should
                                        be  consulted  to  ensure  that  type, size,  and  flow
                                        control characteristics are compatible with control
                                        objectives over the anticipated design life of the  plant.
                                        For automated control, electric or  pneumatic  valve
                                        actuators,  complete with  positioners,  where neces-
                                        sary, are used in place of manual hand  operators.

                                        To  illustrate  how  the design  of  the  air  distribution
                                        control system can  be incorporated into  the overall
                                        aeration  system  design  (see  Figure 5-14 for air
                                        delivery  system  layout),  two  control options  are
                                        outlined below for the design example.  The  first
                                        represents  a low-complexity option, and the second  a
                                        moderate-complexity control  option. A third control
                                        scheme,  not recommended for this size plant, is also
                                        presented to  show  a high-complexity control strategy
                                        that  could  be considered for  large  plants. These
                                        options are presented to illustrate the various degrees
                                        to  which aeration  control  can  be  implemented. The
                                        specific features of any option  could be integrated to
                                                   162

-------
generate  a modified,  i.e., a  more- or less-complex
control scheme, as appropriate.

6.5.2.1 Low-Complexity Control Strategy
The least-complex control strategy  for this  example
involves  manipulating  the blower configuration  and
total aeration output to the four reactors to maintain a
desired set-point DO  concentration at one location in
one aeration basin. Initially, Zone 2 is selected for the
DO probe location for this control measurement. Zone
3 is not selected because load changes and  changes
in  system  oxygen demand may  not  always  be
detected in this zone  or  will be  detected too late to
respond  to  changes in DO concentration  in  Zones 1
and 2. Zone 1 is not  selected because DO  changes
typically  occur more  rapidly  here than further down
the tank  and may result in unnecessary or erroneous
corrections  to  airflow in Zones  2  and  3.  This is
discussed further in Section 6.5.3.

A proportional-integral  DO  controller  cascades  the
airflow set-point to the air demand  controller,  which
regulates the blower output through the  manipulation
of inlet guide vanes to maintain an amperage  set-point
as  outlined in  Section 6.5.1.3.  The operator  would
strive to  maintain  an acceptable DO concentration
profile along the length of the reactors by periodically
manipulating  the  aeration grid  distribution valves
manually to adjust air delivery to other aeration zones.
In this option, careful operator attention is required to
maintain  the distribution valves  in their collective,
"most open"  positions  to  minimize  air  header
pressure.

The  control loops  for this  option  are shown
"schematically in Figure 6-9. The blowers are  operated
by  the control  system to respond to variable oxygen
demands. The  number of on-iine blowers depends on
the  load  to the plant.  Bringing  blowers  on-line or
taking them off-line is carried out automatically upon
receiving  an on/off  signal  from the air  demand
controller.

The blower system operating  map would  be used to
determine  control of  the on-line  blowers to achieve
optimum energy efficiency. This may be accomplished
most effectively by controlling all on-line blowers  with
the same signal from the air demand controller. This
strategy  controls  all  on-line  blowers at  the  same
operating  point  while  matching the  variable  total
airflow demand.

Alternatively, one blower could  be  operated by the
control  system  to  respond  to variable  oxygen
demands and  one or more  of  the other  blowers
operated at constant output to provide a "base
supply"  of  air. Periodic  substitution  of  a  different
blower to serve as the variable delivery blower allows
for load  balancing and  accommodates maintenance
requirements. The blower manufacturer  application
engineer  should be  consulted  to  ensure  that  the
envisioned  control strategy achieves the  control
objectives.

It may be  desirable to initially override part  of  the
automated  control of the  blowers  and allow  the
operating staff to manually control certain aspects of
the  blower  system  operation.  This  could  be
accomplished by providing the  controller  output
information to the operator who would  in turn make
the required  adjustments to guide  varies  and  the
bringing on-line or taking  off-line  of  blowers.  As  the
level of comfort  with the control  system increases,
fully automated control  could  be implemented on  a
staged basis.

This control strategy assumes  the four parallel
aeration  basins   are  operated to  achieve uniform
spatial  and temporal oxygen demand  profiles and that
by similarly adjusting the air distribution valves in  the
parallel  zones,  an acceptable DO  profile can  be
obtained in  each aeration basin.  However, even
though the aeration basins are operated in parallel,  it
is unlikely they will perform identically.

An  important assumption  inherent in this  strategy is
that the  system  hydraulic design  results  in equal
wastewater flow  distribution  and  that air  system
headlosses  are  very  close, if  not  equal, for each
aeration basin. Since the diffusers may have different
operating and cleaning schedules, DWPs  could vary
from basin to basin.  Different levels  of fouling of  the
diffusers  in each zone would also result  in different
OTEs and affect  DO dynamics  between basins. This
may make  it difficult for the operator to continuously
maintain  a uniform DO  concentration profile in all
aeration basins  and  emphasizes  the  importance  of
monitoring DO in all basins.

Ideally, an  on-line DO  sensor would be provided for
each aeration zone (a total of 12 sensors)  to optimize
monitoring  of system  DO concentration  dynamics.
However, for reasons outlined in Section 6.4.1, it may
be  appropriate  to reduce, at least  initially,  on-line
probes to a number with  which O&M personnel  are
comfortable.

DO monitoring can be accomplished by placing  DO
probes with a continuous readout in the  other three
basins  at the same location as the DO probe in  the
control basin. This design  permits  any of  the  four
readout probes and their respective basins  to serve as
the DO control system depending on  which basins  are
out of service  for cleaning. This arrangement  also
allows  any  of the basins to be removed from service
during low-loading periods.

A calibrated portable DO probe can  be used  by  the
operator to monitor and maintain desired DO levels in
those sections of the system  not served by  on-line
meters. As acceptance of the on-line instrumentation
and  the  comfort  level of the plant  staff increases,
                                                  163

-------
                                                                                                                                                                                  2
                                                                                                                                                                                 to
                                                                                                                                                                                  c
                                                                                                                                                                                  5
                                                                                                                                                                                  en
                                                                                                                                                                                  to
                                                                                                                                                            Recorder
                              Amp Set
                                Point
              Amp
              Draw
CT3
                                                                                                                                     DO Probe
                                                                                                                                      (typical)
                                                                                                                           Basin 1
Header
Pressure
Recorder

Airflow
Recorder


                                                                                  Airflow
                                                                               Measurement
                                                                                                                           Basin 2
                                                                                                                           Basin 3
                                                                                                       Aeration Grid
                                                                                                        Distribution
                                                                                                      Valve (typical)
o

8
                                                                                                                                                                                 o
                                                                                                                                                                                 o
o
a-
(B
s»
5'
                                                                                                                          Basin 4         Q"

-------
additional  on-line  probes  could  be added.  It is
important that hardware and software provisions  for
easily accommodating  expanded monitoring capability
be incorporated in the control system design.

The  tuning of  the  rather  coarse aeration  controller
recommended for this low-complexity strategy would
be  carried out  as identified  in  Section  6.3.2.2.
Selection of the final DO feedback location and set-
point would be adjusted following system start-up and
after process response was known.
6.5.2.2 Moderate-Complexity Control Strategy
A  moderate-complexity control strategy  is described
here as  an alternative to  the  low-complexity control
system  to provide more exact DO control  in  each
basin.  This system also  facilitates  more  accurate
control of airflow to each basin by using individual DO
set-points, controllers, airflow control valves,  and air
headers for each basin.

The basic controller design is shown schematically in
Figure 6-10. The major difference between this design
and the  low-complexity  control system  is that  each
aeration  basin is  provided with its own separately
controlled air distribution header. Thus, the air control
systems  for the four basins are independent of each
other,  and  the need  to  assume,  or dictate,  that
adjacent  aeration  basins are operated  identically  is
eliminated. The control of the DO concentration profile
in   each   aeration  basin would  be  similar  to  that
described in the low-complexity example.

One measurement  of  DO concentration  initially  in
Zone 2 of each aeration basin would provide feedback
to  the airflow controller for  that  basin. Again,  as
indicated  in the low-complexity example, it  would be
ideal to monitor the DO concentration in each of the
other zones with  on-line  probes. However, the DO
concentration  could, initially at least,  be  monitored
with a portable DO probe and  meter  in the  other
zones.   Periodic  manual  adjustment of  the  air
distribution  valves  in  the  individual  aeration  zones
maintains the desired  DO concentration profile. This
would be easier to achieve than in the previous case
outlined  because of the greater independence of the
four aeration control systems. As before, the operator
would  strive  to keep the  air distribution  valves as
collectively  open  as  possible  to minimize  header
pressure.

Automated valves  located in the  four  individual
headers  distribute the total blower output to the four
aeration  basins. At least one of these valves is always
maintained in  its "most open" position to minimize the
main air  header  pressure.  A  pressure  controller
located  in the main header regulates blower output by
manipulating  the  inlet  guide vanes  as  described
above.  The  same  considerations  regarding blower
control options  for achieving  optimum  system
efficiency apply to this control strategy as well.

For this strategy, implementation of fully-automated air
distribution control  could also be staged to build
operator confidence. Initially,  the control  strategy
would be simplified. The operator would  designate the
basin with the highest oxygen demand as the "control
basin"  and use the output from the DO controller in
that  basin to manually set its air  header distribution
valve in its "most open"  position. The DO  controllers
for the  other three  basins would automatically adjust
their respective air  distribution  header control valves
to  maintain the  desired  DO set-point in  their
respective basins.

Implementation of the four airflow controllers shown in
Figure  6-10  could  be  postponed and  the  DO
controllers  used  to  manipulate the  air distribution
control valves directly. As before, the manual aeration
grid distribution  valves  in  each  basin   would  be
adjusted by  the  operator to  achieve the desired
profile.  The output from the DO  controller  with the
manually-controlled  header valve would be  used to
control  the blowers by providing the set-point for an
air  demand  controller,  as  shown in  Figure  6-9.
Installation of the pressure controller shown in Figure
6-10 would  also  be postponed. Once  experience is
gained  with, and  the operating staff has  accepted
automated  control  of,   the  air header distribution
control valves, full  automated control of the  air
distribution system can be implemented,  if desired.

6.5.2.3 High-Complexity Control Strategy
A more complex controller design option  for  an
aeration system configuration similar to  the one used
in this  example provides  independent set-point control
of DO  concentration in each  zone or grid  of each of
the four parallel aeration  basins. This control strategy
would  normally be considered only for a much larger
plant (>890 Us [20 mgd]) as it is probably  not cost
effective  for  the plant  size  used  in  this design
example.  This system  uses:  1)  12 cascaded  DO
concentration/airflow control loops to control the air to
each  zone,  and  2) a  main  air  header  pressure
controller  to regulate  blower  output through
manipulation of inlet guide  vanes to  maintain  an
amperage set-point  (as  outlined in  Section  6.5.1.3),
and  the number  of on-line blowers.  The • control
system is shown schematically in Figure  6-11.

The complexity of this control  scheme requires  that
additional  instrumentation and  final control elements
be  provided.  Besides the 12  DO  sensors, airflow
measurement (standard  conditions) is  required  for
each zone.  An automated actuator/positioner is also
required for each  aeration grid  distribution control
valve.  These valves must be appropriately selected,
sized,  and located  to achieve  the  necessary control
action.
                                                   165

-------
                                                                                                                                                                                   s
                                                                                                                                                                                   I
05

O3
                                                                                                       Airflow

                                                                                                    Measurement
                                                                                                                             Basin 1
                                                                                                                                             DO Probe

                                                                                                                                             (typical)
                                                                                                                                                                       1 Of 4

                                                                                                                                                                       (typical)
3
o
a.
ft

3

9
A
o


"2.
(9
X
o
o
3
                                                                                                                                                                                   o
                                                                                                                                                                                   3-
                                                                                                                                                                                   Q


                                                                                                                                                                                   tu
                                                                                                                             Basin 2
                                                                                                                                                                  Aeration Grid

                                                                                                                                                                   Distribution

                                                                                                                                                                 Valve (typical)
                                                                                                                              Basin 3
                                                                                                                               Basin 4

-------
                                                                                                                                                                               •n
                                                                                                                                                                               &
                                                                                                                                                                               I
                                                                                                                                                                               01
                              Amp Set
                                Point
                                                                                                   Basin 1
                                                                                                            O
                                          DO Probe
                                            (typical)
                                    o      o—
              Amp
              Draw
CD
                                                                 4	
       Airflow
    Measurement!
       typical)
                                                                    Pressure
                                                                   Transmitter
 Airflow
Recorder
                                                                                   Airflow
                                                                                Measurement
                                                                                  1 Of 12
                                                                                  (typical)
                                                                                                                                                                               1O
                                                                                              3-
                                                                                              !D

                                                                                              fl)

-------
In each of the control strategies described  above,
tuning of  control parameters and  selection of  control
loop intervals are critical. The control intervals will be
on  the order of those  identified in  Section  6.3.3.1
(e.g., 20 minutes for the DO control loop, 30 seconds
for  the airflow control  loop, and 3  minutes for  the
pressure  control  loop).  Fine tuning of overall  control
performance will be required once the process is  on-
line  and operating  at desired DO concentration set-
points.

6.5.3 DO Probe Location and DO Set-Point
The selection  of DO concentration control points and
set-points is a major consideration in the successful
operation  of  any  DO/airflow  control  system.  As
outlined previously  in this chapter and in  Chapters  3
and  4,  selection of the  operating DO concentration
affects process performance and aeration efficiency.

In two of  the cases outlined above, DO concentration
is set and controlled at only one point in the plug flow
reactor and airflow  distribution is manually  adjusted to
maintain an acceptable DO concentration profile. The
DO  profile  is  dynamic, changing  in response  to
influent concentration  and  hydraulic load  variations
throughout the day.

The limitations of having only one DO concentration
reading in a plug flow reactor are illustrated in Rgure
6-12  (35). In this example, three  different  shapes of
the profile could exist if DO is measured only at Point
C.  If Profile 2  is  the  desired  profile for  optimum
process performance, Profile 1 suggests the biological
reactions  have  been  completed  too soon,  with  a
corresponding wastage of  air.  Conversely,  Profile  3
suggests  the  reactions have not  been completed in
the  desired time. The  use of  three  on-line  DO
measurements (i.e., Point  C plus Points  A and  B)
provides  valuable  additional information,  particularly
near the area of the inflection point in  the DO profile.

Figure 6-12.  Significance of additional DO measuring probes
            on interpolation of DO profile.
                Distance Along Aeration Tank
For Profile 1, the early completion of the reactions and
corresponding wastage of energy are  confirmed by
the high value of DO at Point A. Point A on Profile 2
suggests partial completion of the biological reactions
with neither  underaeration  or  overaeration  at  this
location in the basin, while Points B and  C indicate the
reactions have been substantially completed for a  long
enough period to achieve both good process removals
and sludge settleability. On the  other hand, the DO
readings on Profile 3  would seem to indicate either;
excessive process  loading or underaeration,  or both,
in the critical middle segment of the reactor.

This analysis (35)  illustrates the  considerations  that
must be taken into account when trying to maintain an
acceptable residual DO profile in a plug flow reactor.
Further, the  profiles in  Figure  6-12 show  that ,an
estimate of the profile  slope, based on two or three
DO probes placed near the outlet end of the aeration
basin,  would  be  a better feedback  variable  than  a
measurement from  a single  DO  probe. This could be
easily  implemented  using modern  microprocessor-
based  controllers  and  the  reliable  DO monitoring
instrumentation available today.

For this example,  if 12 probes  are  provided  initially,
the three probes in each basin should be used in the
early  phase of operation to adequately  define  the
location and profile of the DO inflection  curve.  These
data can then be used  to  adjust process operating
parameters as required. The DO set-point  should be
high enough to ensure that  variations in DO in other
parts of the basin (primarily in  Zone  1), which result
from this relatively  high variance control strategy, do
not become rate limiting for extended periods of time.
Once  the  process has  been   fine tuned  and  is
performing  satisfactorily  on a  consistent  basis,  the
optimum location of the probes  can be determined.
Subsequent significant changes  in process operating
conditions may require reestablishment of optimum
probe locations.

In the more complex control strategy, the controller is
designed to maintain a DO concentration set-point  in
each zone. The DO set-point values are selected  to
maintain the  lowest  possible  DO concentration
residuals  without adversely   affecting  treatment
performance.  To achieve nitrification, higher set-points
may be required  than  would  be necessary to meet
nonnitrifying requirements. In this case, the  reduced
variance of the controller permits tighter  DO set-points
to  be.  maintained without adversely  affecting
performance.  The traditional assumption that  DO  set-
points  should always  remain constant may  not be
valid  (20,36).  Site-specific  operating experience  is
necessary to  determine the most appropriate DO  set-
point values.
                                                   168

-------
6.6 Experiences with Automated Aeration
Control
Incorporation of automated aeration control systems in
wastewater treatment plants  has been reported for  a
number of cases  (6,7,10,12-14,19,29,33,37).  The DO
concentration  control schemes  employed at 12
wastewater  treatment facilities,  in  sizes of 44-7,000
L/s (1-160 mgd),  have been described and reviewed
in  detail (10).  The  results  achieved at  these 12
facilities  have  been  further summarized (38). in  this
summary, and in  almost  all  other cases cited where
the aeration control  systems  were  well-designed,
properly implemented, and well maintained, significant
aeration energy  savings  and  improved  process
efficiency resulted.

Where  automated control   systems  were
inappropriately designed,  improperly implemented, or
poorly maintained, problems occurred that often led to
the abandonment of the system. These problems
included inability to control DO  because of low  plant
loadings, limitations  in  blower  turndown  capacity,
oversized air distribution  control  valves accompanied
by  ineffective  air  distribution  control, inadequate
mixing,  operator  indifference,  system complexity,
instrumentation  failure,  and high instrumentation
maintenance.

Many of these experiences occurred 10-15 years ago
when microprocessor-based control technology was in
its  infancy  and  on-line  instrumentation  was  less
reliable. The more recent successful applications have
resulted  from   advances  in  control  hardware
technology  and  control  strategy  development,
improved on-line  instrumentation reliability, a greater
understanding  of the  control process,  and greater
interaction between design and control specialists and
end  users.  Several  recent  examples  of  successful
implementation of automated airflow/DO control are
described below.

6.6.7 Piscataway, MD
A  computer-based  process control  system at  the
Washington  Suburban Sanitary  Commission  (WSSC)
activated sludge  plant in Piscataway,  MD, has been
operating successfully since  1981 (39). The plant has
a  design capacity  of 1,315 Us  (30 mgd)  and an
average  flow of approximately 50  percent of design.
The process includes primary clarification followed by
first-stage  activated   sludge  treatment  for
carbonaceous  BOD5 removal  and  second-stage
activated sludge  treatment  for nitrification  and
phosphorus  removal.  The computer control system is
capable  of  continuous and  sequential control of ail
major unit processes and equipment  in the second-
stage (nitrification) treatment train.

Operating experience has shown that computer-based
control  of  the  nitrification  aeration  system is
economical. The system functions  without  major
problems. Nitrification aeration  energy  costs have
been  reduced  by 34 percent  as compared  with
manual control.  No adverse effects on the treatment
process have been reported.

The nitrification  aeration system  consists of two four-
pass,  plug flow  (or step feed) aeration basins. Air is
supplied by four 10-m3/s (21,190-scfm), 500-kW (670-
hp) centrifugal  blowers discharging into  a  common
header from where it is distributed into  the  aeration
basins through  coarse bubble diffusers.  The control
system and strategy are very similar to those outlined
for the design  example in  Section 6.5.  The blower
control strategy  is virtually identical to that described
in Section 6.5.1. Inlet throttling valves are employed to
control blower  capacity using the control   strategy
illustrated in  Figure  6-8. The  computerized  blower
controller,  which provides  surge  protection  and
automatic start-up, has resulted  in continuous,  sate
aeration system  operation.

The  basic  design  of the air distribution controller
utilized to maintain  DO set-points is similar  to that
shown in Figure 6-10. However, in this case Basins 1
and 2 (Figure 6-10) are actually equivalent to the first
two and  last two passes, respectively, of one 4-pass
aeration basin. Each pair of passes is supplied with air
through a separate air distribution header. DO probes
are located in the second and fourth  passes in each
aeration basin. The controller incorporates a minimum
airflow set-point to  maintain  adequate   mixing.  The
controller successfully maintains DO at the  desired
set-points (±0.1 mg/L).

The  system was  initially  commissioned  with  manual
control and full computer  monitoring.  During this
period, plant staff  became familiar with the on-line
instrumentation  and  process  operation. This  was
followed  by full  computer control during  the day shift
of the normal  work week  and  finally with 24-hr
automated control. The plant is  run by 33 operators,
13 electromechanical maintenance personnel, and 2
instrument  technicians.  An  instrument   service
company provides assistance,  as  required, under
contract to  WSSC.  The  computer   system  is
maintained  by two data  processing  personnel. The
instrumentation  technicians  and  data  processing
personnel are considered essential to the successful
operation of the automated control system.

During the  period  of manual  control,  airflows  to
Passes 1 and 2 and Passes 3 and 4 were periodically
manually adjusted and the resulting DO concentration
profiles  monitored.  High DO  concentrations  are
evident  in both Passes 2 and 4 during  the  evening
and morning hours (Figure 6-13).

The on-line instrumentation for measuring airflow took
longer to commission than expected.  The airflow
sensors  were downsized to achieve more accurate
measurement at low airflows.  This emphasizes  the
                                                 169

-------
Figure 6-13. Manual control of DO at Piseataway.
6PM   7    8
                                              MID  1AM

                                               Time, hr
careful attention necessary in selecting and specifying
instrumentation to ensure proper operation during all
stages of the anticipated plant design life. Following
extensive  staff  training on the  control  system,
automated operation of  the  nitrification train was
phased in. Initially, the plant was run in the automated "
mode  during th©  day shift,  Monday to  Friday only.
This staged approach resulted in a confident transition
by the operators from manual operation to continuous,
fully automated  control.  Twenty-four hour automated
operation was initiated  once the  O&M staff was
confident and comfortable with the system.

The performance  of the system  has been excellent.
The overaeralion problem encountered during  manual
control has been nearly eliminated. DO concentration
profiles under fully automated operation are plotted in
Figure 6-14. It is  evident that, except for periods of
minimum airflow limitations in the fourth pass, DO
concentration was effectively controlled  at the  set-
points  of 0.8 mg/L and 1.0 mg/L in the second and
fourth passes, respectively.  Adjustments  to set-points
were  made to  optimize  process and  aeration
efficiencies  as  experience with the  system was
gained. The DO  set-points  were gradually  lowered
from 4.5  mg/L to 0.6-1.0  mg/L with no adverse effects
on the process. In addition, the minimum airflow set-
                                      point for each  pair of aeration passes was  lowered
                                      from  0,71 to 0.47 ms/s  (1,500 to 1,000 scfm), also
                                      without adverse process effects.  The  individual  air
                                      diffuser valves  were manually adjusted to  obtain  an
                                      even mixing pattern throughout the basins.

                                      Automated control to  lower  DO  concentrations  has
                                      resulted  in  a corresponding  reduction in average
                                      blower power per unit loading of  34 percent in the
                                      nitrification system.  It is expected these aeration
                                      energy savings  will  further  increase as  process
                                      optimization continues.

                                      The nitrification  aeration  system has been operated
                                      with automated control more than  90 percent of the
                                      time. The main interruptions have  been because of
                                      routine sensor maintenance. The plant staff clean and
                                      calibrate  the   DO   probes  weekly. The  other
                                      instrumentation associated with the  aeration system
                                      (airflow, pressure, and temperature measurements) is
                                      checked every 3 months.

                                      Early involvement of the  operations staff at the
                                      Piscataway plant, and their continuing commitment to
                                      maintaining and optimizing the system, was reported
                                      to be critical in achieving successful operation of the
                                      plant's automated aeration control system (39).
                                                  170

-------
 Figure 6-14, Automated control of DO at Piscataway.
                40-
                30-
                           DO         P2 -
                           DO         P4 -
                           Airflow   P1  + P2 -
                           Airflow   P3 + P4 -
                20-
                10-
              !S. 10-
             m

              •l  5-
              o
              E  2;
                  MIDI  2345678  9 10 11NOON1 2  34  56789 10 11 MID
nnnonnnnnnnnflflflnnnnnnnn
6.6.2 Madison, Wl
A distributed control system was incorporated as part
of  a recent plant expansion and  aeration  system
upgrade at the  Madison, Wl Metropolitan Sewerage
District  Nine  Springs  Wastewater Pollution  Control
Plant. The expansion and upgrade were undertaken to
achieve nitrification in a  single-stage activated sludge
process. Fine pore diffusers were incorporated in  all
aeration basins. The control system was installed to
save energy, estimated to  be about 10 percent, and
minimize the need for additional staff  to operate the
expanded  plant.  The  automated system controls  a
variety  of  processes and unit  operations. These
include  influent flow splitting, primary effluent flow
splitting, waste  activated  sludge  pumping,  recycle
sludge  pumping,  UV disinfection, effluent  pumping,
digester sequencing, and DO and blower control.

Details regarding  the background and  history of the
upgrade and  the  control system design have been
presented  elsewhere  (40-42) and  are  further
summarized in Chapter 8.  Experience  with  operating
the automated aeration and DO control system is of
interest here.

The treatment plant has a design capacity of 2,190
Us (50 mgd) and  now operates at an average flow of
1,665 Us (38 mgd). As a result of various expansions,
there are two plants on site designated as the East
(Old) plant  and the West (New) plant.  The East plant
consists of six three-pass  aeration basins,  while the
West plant is comprised of  four three-pass aeration
                              basins. The East plant is further split into Plants 1 and
                              2 and the West plant into Plants 3 and 4.  There are,
                              therefore, effectively four treatment plants on site that
                              can be operated independently from one another. An
                              overall flow schematic is presented in Figure 6-15,

                              The  aeration  systems for the East and West plants
                              are completely separate. For each of the  three-pass
                              aeration  basins,  a DO probe located at the end of
                              each  pass  is used  to  manipulate an air  distribution
                              control  valve for that pass. The DO  controller
                              cascades the airflow  set-point  to  the air  control
                              algorithm  for   that  pass.  A  minimum  airflow
                              requirement limits the  total airflow turndown to the
                              pass. Air is further distributed to each of the  three
                              aeration grids in  each pass by manual adjustment of
                              butterfly valves.

                              The  inlet guide  vanes on the centrifugal blowers are
                              adjusted  to vary air delivery to maintain  a  constant
                              header pressure, plus or minus a deadband - much as
                              previously described in Section 6,5,2.3.  For the  West
                              plant, it was anticipated that only one of three single-
                              stage centrifugal blowers would be needed to satisfy
                              air requirements. The  East plant has one gas engine
                              driven blower, two  2-stage centrifugal  blowers, and
                              two PD blowers available. It was initially assumed that
                              one  PD  blower  plus one  of the 2-stage  centrifugal
                              blowers would be used on a routine basis.

                              On system start-up, problems were encountered in
                              operating the West plant  blowers. They  had  been
                                                171

-------
 Figure 6-15.  Plant schematic for Nine Springs wastewater treatment plant - Madison, Wl.

                                              Aeration Tanks
           West
           Plant
                           Primary
                           Clarifiers
4
            East
            Plant
                                                  30
                                                  29
                                                  28
                                                  26
                                                  25
4
                                                  24
                                                  23
                                                  22
                                                  21
                                                  20
                                   Secondary Clarifiers
                                                                            Secondary Clarifiers
                                                                                 00
supplied with a  rated  capacity  of  24 m.3/s  (50,000
scfm) each,  even though they had  been specified to
have an operating range of 6-12 m3/s (12,500-25,000
scfm).  Motors capable  of  driving the blowers to the
12-m3/s  (25,000-cfm) delivery point had been supplied
with hardwired lockouts to the guide vane controllers
to prevent the motors  from being  overloaded  if the
guide vanes opened too  far. The  lockouts were not
tied to the control system initially.  As a result,  when
air  demand  subsequently dropped and  the airflow
distribution valves began to close, the blower(s)  would
go  into a surge  condition before the operator  could
react.

Individual watt transducers  were  installed  on  the
blowers  and interfaced  to the control  system. The
control  system could then control  the blower  guide
vanes   between  motor operating  limits, and  the
hardware lockouts  were  subsequently  removed.  In
              addition,  changes  were made  to  the  blower control
              algorithm  to  facilitate  blower  rotation.  The  operator
              manually  overrides the blower  control  system  and
              throttles  the  inlet guide  vanes  to cause the  air
              distribution control valves  to  open  and  reduce the
              main  header  pressure, thereby enabling  a second
              blower to be  brought  on line. The valves are then
              "frozen"  for 10 minutes while the  first blower is shut
              down. This prevents pressure  rise in the system due
              to closing  of  distribution  valves and  eliminates the
              possibility  of  blower  surge.  The  solution to  this
              problem is one example of effective use of process
              control computer communication to minimize adverse
              impacts on plant  equipment  and prevent  process
              upsets. The blowers for the East plant are handled
              similarly.

              Reduced  loading  to the Nine Springs  plant initially
              caused some problems  with  tuning the  air delivery
                                                   172

-------
controller  to  work effectively at low airflows.  At the
lowest  guide vane settings,  a  small  percentage
change in guide vane  position resulted  in a  much
larger percentage change in airflow.  A wide pressure
set-point band was therefore incorporated  to minimize
this  problem and facilitate effective  tuning  of the
control algorithms. To minimize energy requirements,
a  "lookup table"  of pressure  set-points  vs.  total
system airflow demands  was  incorporated  in the
process control  computer to  automatically  minimize
the pressure  set-point, yet maintain effective control.

To achieve  some  degree of airflow  reduction and
energy savings while operating in  an , overall reduced
loading condition, one-half of the East  plant  was shut
down and the corresponding  flow routed to  the  West
plant.  This flow adjustment allowed the  gas  engine
driven  blower to supply total  air requirements  for the
East plant.  Even  though  this  blower  could not  be
automatically controlled, the utilization of digester gas
resulted in reduced  cost.  The  West plant had  been
unable to  realize the full benefits of DO concentration
and  airflow  control because  of limitations in  blower
turndown  capacity. Increased  OTE  and  the low
loadings resulted in overaeration much of the  time.
The increased loading  achieved by diversion of East
plant flow allowed the  West plant  to maintain typical
set-points of 1..2, 1.6, and 2.0  mg/L, respectively, in
the three  successive passes. This resulted in  more
efficient air delivery and stable process  performance.

Options  for increasing   overall  aeration  system
flexibility are currently being examined, at  Madison.
One option  being  considered is connecting  the East
and  West plant main  headers together.  This  would
provide substantially increased  flexibility in  blower
turndown  capacity and an  improved ability to configure
the  size  and number  of blowers  in operation  to
achieve maximum efficiency.

The DO probes have required the greatest amount of
control system  maintenance  at  Madison.  Probe
cleaning is required every 3-4 days in the  first  passes
when  plant  loadings  are high and  the  plant  is
operating  at  lower sludge ages. Lower loadings and
operating  at higher sludge  ages  reduce  probe
maintenance requirements. The probes  are  usually
cleaned and calibrated every  2  weeks. Operators
handle the  cleanings,  calibrations, and simple
maintenance such as  membrane  replacement.
Electronics technicians are called in  for more serious
problems.  Careful monitoring  of  maintenance
requirements and probe performance has  shown that
the probes are not  performing as well as expected.
Efforts to determine,whether this is  an equipment or
procedural problem are proceeding.

The fine  tuning of the process control  system has
involved changing several control algorithms. In many
instances, this  was caused  by  changes  in  plant
operating  conditions. In other instances, it was caused
by overdesign and  the  inability  to  control hardware
optimally.  However, it  is  to  the  credit of the Nine
Springs plant staff that  they  have made the system
work under these conditions. Although the expected
energy savings have not yet been realized, effective
utilization of the automated process control system for
sludge age and  aeration/DO control has  resulted  in
exceptionally  stable  operation  of the  treatment
process. Active efforts to optimize the control system
and plant  hardware will  result in energy savings and
more operational flexibility as plant loadings increase.


6.7 Summary
The  benefits  associated with automated  aeration
control have  been  known  for many  years. Many
systems  have been successfully  implemented  and
have  generated  the expected  benefits of  aeration
energy savings and  improved process control. Yet, in
many situations where  the potential benefits are
evident and  far  outweigh the associated  costs, the
user community is very reluctant to implement an on-
line, sensor-based automated control system.

This reluctance is  often  due to perceived problems
with the technology. In  some cases,  such  an attitude
is  justified based  on   past experiences.  In  most
instances, however, these problems could  have been
avoided with  careful attention to  system  constraints
and process requirements.

Instrumentation,  long cited  as  the  cause  of  most
problems,  can no  longer be used  as a  scapegoat.
Reliable instruments are available, but they  are not
black  boxes. They require continual maintenance that,
if properly applied, is usually not excessive.

The  amount  of  energy  used for  aeration,  and the
potential  for  saving  a  portion  of  that  energy with
effective  control, must be weighed  against the life-
cycle  cost of the control system, including hardware
and software. The aeration system designer, process
control engineer,  and operating staff should review
each aeration system design  option and determine the
associated cost of the various control options. During
this iterative process, it is important to consider the
flexibility  and expansion  potential of each option  to
accommodate all variations in loading anticipated over
the design life of the plant. This time-consuming, task
is  vital to the selection  of an  automated  control
strategy that will be compatible with the final aeration
system design and  maintain  process integrity at the
least possible cost.

The  intent of this  chapter has been to  identify the
major benefits of automated  aeration control and the
most important factors to consider in the design and
implementation of automated control  systems. Where
automated control has been  successfully incorporated
in  the operation of activated  sludge wastewater
treatment  plants, a major contributing factor has been
                                                  173

-------
comprehensive interaction  between  the  design
consultant, plant  O&M  staff,  process  control
personnel,  and  instrumentation and control hardware
manufacturers. The successful application of  an
automated control system begins at the design phase
of the aeration system. The operational success of a
well-designed  control  system depends  on  the
commitment of the plant staff.


6.8 References
When an NTIS number  is cited in  a reference,  that
reference is available from:

       National Technical Information Service
       5285 Port Royal Road
       Springfield, VA 22161
       (703) 487-4650

  LPalm,  J.C.,  D.  Jenkins  and  D.S.  Parker.
   Relationship Between Organic Loading, Dissolved
   Oxygen Concentration and Sludge Settleability in
   the Completely Mixed Activated Sludge Process.
   JWPCF 52(1G):2484-2506, 1980.

  2. Sezgin,  M.,  D,  Jenkins and  D. S.  Parker. A
   Unified  Theory of Filamentous  Activated Sludge
   Bulking, JWPCF 50(2):362-381, 1978

  3. Process Design Manual  for  Nitrogen  Control.
   EPA-625/1-77-007,  NTIS No.  PB-259149,  U.S.
   Environmental Protection Agency, Cincinnati, OH,
   1975.

  4. Parker,   D.S., W.J.  Kaufman  and  D. Jenkins.
   Physical Conditioning of Activated Sludge Floe.
   JWPCF 43(9):1817-1833, 1971.

  5. Tuntoolavest, M., E. Miller and C.P.L. Grady  Jr.
   Characterization of Treatment Plant Final Clarifier
   Performance, Technical Report No. 129, Purdue
   University  Water Resources Research  Center,
   West Lafayette, IN, June 1980.

  6. Roesler, J.F. Plant Performance Using Automatic
   Dissolved  Oxygen  Control.  J.  Env. Eng.  Div.,
   ASCE 100{EE5):1069-1076, October 1974.

  T.Weils,  C.H. Computer Control  of Fully  Nitrifying
   Activated Sludge Processes. Instrument. Technol.
   26(4);32-36, 1879.

  8. Wesner, Q.M., L.J.  Ewing,  Jr., T.S. Lineck  and
   D.J. Hinrichs. Energy Conservation in Municipal
   Wastewater Treatment.  EPA-430/9-77-011, NTIS
   No.  PB81-165391, U.S. Environmental Protection
   Agency, Washington, DC, March 1977.

  9. Energy  Conservation in the Design and Operation
   of Wastewater Treatment Facilities. Manual of
    Practice FD-2, Water Pollution Control Federation,
    Washington, DC, 1982.

10.  Flanagan,  M.J.  and  B.D.  Bracken. Design
    Procedures for Dissolved  Oxygen  Control  of
    Activated Sludge Processes. EPA-600/2-77-032,
    NTIS  No.   PB-270960,  U.S.  Environmental
    Protection Agency, Cincinnati, OH, 1977.

11.  Stephenson, J.P. Practices  in Activated Sludge
    Process   Control.   In:  Comprehensive
    Biotechnology:  The  Principles, Applications and
    Regulations  of  Biotechnology  in  Industry,
    Agriculture  and Medicine, Moo-Young, M.,  Editor,
    4:1131-1144, Pergamon Press, Oxford, England,
    1985.

12.  Robertson,  P.,  V.K. Thomas  and  B. Chambers.
    Energy  Saving - Optimisation of Fine  Bubble
    Aeration: Final Report  and  Replicators Guide.
    Water  Resources Centre,  Stevenage Laboratory,
    Stevenage, England, May 1984.

13.  Andersson,  L,G. Energy Savings at  Wastewater
    Treatment Plants. Report to the Commissioner of
    the  European  Communities and  the  Danish
    Council of  Technology, Water Quality  Institute,
    DK-2970, Horsholm,  Denmark, 1979.

14.  Speirs, G.W.  Direct  Digital Control  of the
    Tillsonburg,  Ontario  Municipal Activated Sludge
    Wastewater Treatment  Plant - A Case  Study.
    Presented at AQTE/CSCE  Specialized  Workshop
    on  Computer  Control of Wastewater  Treatment
    Plants, Montreal, Quebec, Canada, May 1986.

15.  Stenstrom,  M.K.,  H.R.  Vazirinejad  and A.S. Ng.
    Economic  Evaluation  of Upgrading Aeration
    Systems. JWPCF 56(1):20-26, 1984.

16.  Ogata, K.  Modem Control Engineering.  Prentice-
    Hall, Englewood Cliffs, NJ, 1970.

17.  Coughanowr,  D.R.  and L.B. Koppel.  Process
    Systems Analysis and Control. McGraw-Hill, New
    York, NY, 1965

18.  Astrom,  K.J.  and  B.   Wittenmark. Computer
    Controlled  Systems  - Theory  and  Design.
    Prentice-Hall, Englewood Cliffs, NJ, 1984.

19.  Aeration.  Manual of  Practice  FD-13,  Water
    Pollution Control  Federation, Washington,  DC,
    1988.

20.  Olsson, G., L.  Rundqwist, L. Eriksson and L. Hall.
    Self Tuning Control of the Dissolved Oxygen
    Concentration  in Activated  Sludge Systems.  In:
    Instrumentation and  Control  of  Water and
    Wastewater Treatment  and  Transport Systems,
                                                174

-------
    IAWPRC,  473-480,  Pergamon  Press,  Oxford,
    England, 1985.

21. Holmberg,  U. Adaptive Dissolved Oxygen Control
    and On-line  Estimation of  Oxygen Transfer and
    Respiration Rates.  Presented  at  Annual AICHE
    Conference, Miami, FL, November 1986.

22. Liptak, B.G. and K. Venczel. Instrument Engineers
    Handbook: Process Measurements. Chilton Book
    Company,  Radnor, PA, 1982.

23. Considine, D.M. Process  Instruments  and
    Controls Handbook,  3rd  Edition.  McGraw-Hill,
    New York,  NY,  1985.

24. Manross,  R.C.  Wastewater  Treatment Plant
    Instrumentation Handbook. EPA-600/8-85-026,
    NTIS  No.  PB86-108636,  U.S.  Environmental
    Protection Agency, Cincinnati, OH, 1983.

25. Process Instrumentation and  Control Systems.
    Manual of  Practice OM-6, Water Pollution Control
    Federation, Washington, DC, 1984.

26. A Study  of Recently  Developed  Continuous
    Dissolved  Oxygen Measurement  Systems Based
    on Their Field Performance. Technical Bulletin No.
    440, NCASI, New York, NY, August 1984.

27. Kulin  G.,  W.W.  Schuk  and I.J.  Kugelman.
    Evaluation  of a  Dissolved Oxygen  Field  Test
    Protocol. JWPCF 55(2): 178-186, 1983.

28. A Collection of Seven  Reports on Individual On-
    Line   DO  Meter  Performance. Water  and
    Wastewater Instrumentation Testing Association,
    1225 I Street, Suite 300, Washington, DC 20005,
    1988.

29. Speirs,  G.W.  and  M.J.  Hribljan.  Full-scale
    Evaluation   of the   Benefits  of  On-line
    Instrumentation and Automated Process Control in
    Wastewater Treatment. Presented at the  61st
    Annual Conference of the Water Pollution Control
    Federation, Dallas, TX,  October 1988.
30. Speirs, G.W.  and R.D. Hill. Field Verification of
    On-line Instrumentation at a Municipal Wastewater
    Treatment Plant. Wat. Sci. and  Tech.  19(3/4):669-
    680, 1987.
31. Nisenfeld, A.E.  Centrifugal  Compressors:
    Principles  of Operation and  Control.  ISA,
    Research Triangle Park, NC, 1982.
32. Prime  Movers.  Manual of Practice SM-5, Water
    Pollution Control  Federation, Washington,  DC,
    1984!

33. Lutman,  C.G. and  R.G.  Skrentner.  Controlling
    Low Pressure Centrifugal Blowers - A  Tutorial.
    The Communicator 7(2), EMA, Inc., St. Paul, MN,
    1987.

34. A  Primer  for   Computerized  Wastewater
    Applications.  Manual  of  Practice  SM-5, Water
    Pollution Control  Federation, Washington,  DC,
    1986.                    ,      _

35. Olsson, G. and J.F. Andrews. Dissolved Oxygen
    Control in the Activated Sludge Process. Wat. Sci.
    and Tech. 13(10):341-347, 1981.

36. Hermanowicz,  S.W. Dynamic  Changes  in
    Populations of the Activated Sludge Community:
    Effects of Dissolved Oxygen Variations, Wat. Sci.
    and Tech. 19(5/6):889-895, 1987.

37. Grinker, J.R. and R.F.  Meagher. Five. Years of Full
    Scale  Experience  With a Computer Controlled
    Wastewater  Treatment Plant. Presented  at  the
    56th Annual  Conference  of  the  Water Pollution
    Control Federation,  Atlanta,  Georgia, October
    1983.

38. Genthe, -W.K., Roesler, J.F. and B.D,  Bracken.
    Case Histories of Automatic Control of Dissolved
    Oxygen. JWPCF 51(10):2257-2275, 1978.

39. Johnson,  F.B.,  Fertik, H.A.  and  C.G. Lutman.
    Operating Experience with  Computer  Control of
    Air Nitrification. JWPCF 56(12):1223-1230, 1984.

40. Reusser, S.R; and R.R. Riesling. Air Supply and
    UV Disinfection  - Design  and  Control  at  the
    Madison Metropolitan Sewerage  District Nine
    Springs  Wastewater Treatment Plant.  Presented
    at  the Central  States Water  Pollution  Control
    Association 59th Annual Meeting, Milwaukee,  Wl,
    May 1986.

41. Winden,  R.A.  and   S.R.  Reusser.  Active
    Participation in  Design and Installation Leads to
    Operating Success. In: Proceedings  of  the  ISA
    1987 International  Conference,  Anaheim*  CA,
    October 1987.                    >   '

42. Reusser,  S.R.  Operational  Experience  at a
    Computer Controlled  Single-Stage Nitrification
    Facility with  Fine  Bubble  Diffusers. Presented at
    the Central  States  Water  Pollution  Control
    Association  61st Annual Meeting,  Arlington
    Heights, IL, May 1988.
                                                175

-------

-------
                                             Chapter 7
                                        Economic Analysis
7.1 Introduction
The principal reason for installing fine pore aeration
systems is the savings  in aeration energy costs made
possible  through  the higher oxygen  transfer
efficiencies (OTEs) of fine pore devices. Any decision
to employ fine pore diffusers should be justified by an
economic analysis that confirms these savings are
large  enough  to offset  any additional equipment,
installation, and  maintenance expenditures  that may
be  required  compared  to  more energy-intensive
aeration  alternatives.  This  chapter  presents  a
summary of those factors that need to  be considered
in performing such an analysis.
This  chapter  is organized  as  follows.  First,  the
significant  components of aeration  costs  are
summarized.  Next,  a technique  for  performing  an
economic analysis  is  discussed. This method  is
applied to the design example introduced in Chapter 5
to illustrate the  economic impact of  different fouling
rates  for a fine pore aeration system  compared to a
traditional, nonfouling coarse  bubble aeration system.
Examples  of  desktop  and   computer-based
spreadsheet  implementation of this  method  are
provided.  Finally, a compendium  of actual component
cost data obtained from various fine pore installations
is presented.


7.2 Cost Components
Fine  pore aeration  systems may be retrofitted to
existing aeration basins or installed in  new aeration
basins.  The  following cost  items  need  to  be
considered when developing cost estimates for these
aeration systems:

•  Aeration basin dewatering, cleaning, and degritting

•  Old equipment removal, disposal, and salvage

•  Installation of new diffuser system  and  associated
    below-water  air piping

•  Installation or  rehabilitation  of  above-water  air
    piping, including valves

•  Installation of a new or upgrading  the existing air
    filtration system
•   Installation  of  airflow  and   pressure
    monitoring/control equipment

•   Installation  of  a  dissolved   oxygen  (DO)
    monitoring/control system

•   Upgrading or replacing the existing blower system

•   Upgrading or replacing the existing blower prime
    movers

•   Installation of variable frequency drives, inlet guide
    vanes (centrifugal), or alternative  equipment for
    blower control

•   Installation of acid (liquid or gas) cleaning system

•   Installation  of  other  cleaning  systems  or
    components, such as high-pressure pumps

Initial  costs include  all  construction  activities,
equipment purchases and  installation.  Items such  as
mobilization, bonding,  insurance, engineering, and
legal and administrative fees represent additional initial
cost items depending  on  the  particular  project  in
question.

Ongoing  operations and maintenance  (O&M)  costs
over the evaluation period  must also be  taken into
consideration. These  include all energy, labor, and
materials costs.  Particular  attention  must be paid  to
the cost of cleaning  or  replacing  diffusers  on  a
periodic basis.

Replacement  costs  are  expenditures  made  to
purchase and install equipment whose useful  life is
less than the period of analysis. Diffusers,  blowers,  air
filters,  and monitoring devices are examples of items
that may have to be replaced on a periodic basis.


7.3 Economic Analysis Procedure
This section presents  a method for estimating and
evaluating the cost of diffused aeration systems. This
method is applicable to both coarse bubble and fine
pore diffuser systems for either existing, new, -or
retrofit projects. It computes total system costs as the
present  worth  of  initial expenditures and  future
                                                 177

-------
equipment replacement  costs, monthly  energy and
equipment maintenance  costs, and  periodic diffuser
cleaning costs.

The  method can  be  used  to judge  the  cost
effectiveness of alternative  aeration  system designs.
Typical issues that can be addressed include:

*  choosing between coarse  bubble and  fine  pore
   aeration,

•  finding  the optimal cleaning frequency  for fine pore
   diffusers,

•  determining the least-cost choice of  such  design
   variables as number of diffusers and blowers, and

•  examining  the sensitivity of results to  uncertainties
   in  oxygen demands,  OTEs,  fouling  rates,  power
   pricing, etc.

Because different aeration system costs  are incurred
at different points in  time, a  consistent method is
needed to express the value  of  future dollar  outlays
against  current ones.  Present  worth  analyses
recognize that future expenditures do not  have the
same value  to society as  an   equal  level  of
expenditures  made today. Interest factors are used to
place the  two types of costs on an equal footing. The
well-known compound interest formula can be used to
relate a present worth to a future worth in terms of a
number of time Intervals, and the  decimal interest rate
per lime interval:
                 Fw = P
(7-1)
where,
   Pw  = present worth cost, $
   Fw  = future expenditure, $
   i    = periodic discount rate, decimal
   n   = total number of time periods

The ratio of Pw to  Fw,  termed  the  single payment
present worth factor, PWF, is given by:
          PWF'=
Consider, for example, a $1,000 expenditure made 5
years into the  future with the time value of money
taken as 8 percent/yr (i =  0.08):

          PWF = 1 -r{1 + 0.08)5 =  0.6806

The present worth cost is 0.6806(1,000)  = $680.60.
This means that it is cost effective  to spend $680.60
or less  now to avoid the $1,000 expenditure 5 years
hence.

Likewise, a uniform series of end-of-period payments,
each of magnitude A and made  at  the end of each
one of n periods, may be related  to a present worth,
Pw, using the capital recovery factor, CRF:

     CRF = (A/PW) = i(1  + i)n*((1  + i)n - 1]  (7-3)

or its  reciprocal, the uniform series present worth
factor,  SPWF:

  SPWF =  (PW/A) =  [(1 +  i)n - 1]*[i(1  + i)n]   (7-4)

The  present worth  of  an  annual savings in  power
costs may  be  calculated  by multiplying the  annual
savings by the uniform  series  present  worth  factor.
Consider, for example, annual savings in power costs
over a  10-yr life of an aeration system when i = 0.08:

  SPWF  = [{1 + 0.08)10- 1]*[0.08(1 + 0.08)10]= 6.71

This means that $1 of annual  savings in each of  10
years is equivalent to  $6.71 of  present  worth. Put
another way, it is cost  effective  in  this example  to
spend  up to $6,710  in  initial  investment  to save
$1,000/yr in power costs.

A fourth factor, useful for calculating the present worth
of diffuser cleaning expenses, is  the periodic series
present worth factor, PSPWF:

PSPWF = (PW/ZC)                              (7-5)
        = [PWF(im,nm)] [SPWF
-------
SPWF(im,n.m) = SPWF(0.0067,5)
            = [(1 + 0.0067)5-1]* [0.0067(1 +0.0067)5]
         .   = '4.90

PSPWF = [PWF(im,nJ][SPWF(im,nmM)] + [SPWF(im.nm)]
    -••'" =  (0.967)(82.28) *4.90 = 16,23             .

Thus, the present worth of the diffuser cleaning costs
is  16.23(1,000)  = $16,230.  This means that, for this
example, it is cost effective to spend up to $16,230 in
initial costs  to  lower cleaning  costs by $1,000 per
cleaning event.

The discount rate i used in  these factors reflects the
degree to which society is willing to postpone current
expenses and pay  more  at  a later  point in  time. It is
only partially reflected by the current rate of return on
investment obtainable in financial markets. The choice
of a proper  discount rate  to use  for present  worth
analyses,  especially  for public  sector  projects,  is an
arguable  point and  no  specific  rate  will  be
recommended here.

Another issue  that  arises, in considering  future
expenditures  is inflation.  Two  approaches are
available. The first inflates future expenditures as they
occur  and then discounts  them back to time  zero
using  the  above  equations.  The  second  uses  a
discount rate that has been adjusted for inflation and
keeps   all  future costs  inflation free.  If  the  same
inflation rate applies to all expenditures,  the two
methods provide identical results.

The material that follows assumes  that an inflation-
adjusted discount rate i will be used and that no future
cost increases due to inflation need be made. Also, it
is  convenient to use a month as the basic time period.
Therefore,  if  the discount i is stated as an annual rate,
im should be used in  the above equations when n and
m are  given in months.

The total  present worth of the project is the sum  of
initial  costs  plus the present worth  of  ail future
expenditures discounted  using  one  of the  above
equations.  A desktop approach for calculating the
various present worth  elements  of  a project  rs
presented  in  Sections  7.3.1   through  7.3.6.  The
information  required  to perform the calculations that
follow  is identified in Table 7-1.  A sample desktop
analysis is  presented in Section  7.4  An alternative,
computer-based spreadsheet approach is presented in
Section 7.5.

7.3.1 Calculate Initial Costs
Initial  costs include expenditures for  equipment,
construction,  and installation  of the items  listed  in
Section 7.2. Empirical cost data are  presented for
general guidance in  Section  7.6.  By definition, the
actual   cost  of  these items is  equal to  the  present
worth.
Table 7-1. Information Needed to Perform Desktop Economic
        Analysis of Diffused Aeration Systems

     Component Cost Data

     Initial costs
     Equipment replacement intervals and costs
     Unit cost of power
     Monthly routine maintenance cost '
     Cost per diffuser for major cleaning
     Time interval between majot uiimser cleanings
     Length of analysis period
     Discount Rate

     Aeration Zone Data

     Number of diffusers
     Average oxygen demand"
     Average ralio of field to standard OTE"
     Rate of loss of OTE due to fouling
     Maximum percentage loss of OTE possible
     Pressure drop across clean and fouled diffuser
     Orifice pressure drop at 1 sefm of airflow
     Airflow required for mixing
     Minimum and maximum airflows per diffuser
     Relation between SOTE and diffuser airflow

     Air Delivery System Data

     Barometric pressure
     Headless due to depth of submergence and line losses
     Overall blower efficiency

"Annual average values; monthly values are used  for computer-
 based spreadsheet method.
7.3.2 Calculate Energy Costs
Energy  costs embody  the  blower  motor' horsepower
required to deliver  sufficient air to  the diffusers  to
meet process oxygen demands. These costs will vary
from  day to  day over the analysis period as oxygen
demands change and  OTE  decreases  because  of
diffuser fouling.  When a major diffuser cleaning effort
occurs, OTE will be substantially restored and a new
time  pattern  of energy consumption will begin.  For
desktop analysis, it  is not practical to track these
effects over each discrete time  period (hour, week  or
month)  of the  analysis  period.  Instead,  energy
consumption is computed using  annual  average
oxygen demands and the average degree of fouling
that exists between  major diffuser cleanings. A major
cleaning is one that restores a diffuser's OTE to  its
original  unfouled condition.

As was done for the design example  in Chapter 5, the
cost analysis divides the aeration  basin into one  or
more aeration zones,  where the  oxygen  demands,
OTEs, and fouling rates vary by zone. The following
simplifying assumptions are made for  each zone:

1.  Oxygen  demand  equals  the annual  average
    demand (Ib/d).

2.  The ratio of the field QTR (OTR()  to standard OTR
    (SOTR)  for a  clean  diffuser equals  the annual
    average value of that ratio.
                                                    179

-------
3.  The  fouling factor, F, decreases  from 1.0 at  a
    linear rate with time downto some lower limit.

4,  The  pressure drop  across a fine pore diffuser
    increases in  direct  proportlonto the degree of
    fouling,

5.  A major cleaning restores F to 1.0.

With these  assumptions, total present worth  energy
costs can be estimated using the following five-step
procedure:

Step 1.  Assume a diffuser  cleaning interval and
         perform  the following for each aeration
         zone:

la. Find   the  average F, Fa>  that  exists over  this
    cleaning interval.

1b. Determine the airflow per diffuser needed to meet
    the oxygen demand in the  aeration zone:

  q » 0.04
-------
estimated  from  the  theoretical  adiabatic  work
expression ((Equation 4-11):

The expression for the universal gas law is:

                 PbVs = my 144           (7-12)

where,

  Pb = field atmospheric pressure, psia
  Vs = specific volume, cu ft/Ib
  R  = ideal gas  constant = 53.3 ft-!b/ib-°R
  Ta = blower inlet air temperature, °R

Further, mass rate  of  afr  is related to volumetric
airflow rate and specific volume by:
                   w = 60
                             (7-13)
where,
  w  =  mass rate of air, Ib/hr
  qs  =  field  standardized volumetric airflow  rate,
          scfm

By substituting the above terms in Equation 4-11 and
using K = 0,283, the following working equation can
be developed:

      WP = (O.Ot15qsPbfe){[(Pb+Pd)/Pbp283..1j (714)

where,

  WP =  wire power consumption, kW
  &  -  combined blower/motor efficiency, fraction
          (using  a  constant  blower  efficiency
          simplifies  this  calculation;  however,
          efficiency  will  change under  varying
          operating conditions)
  Pb =  fjey atmospheric pressure, psia
  PCJ =  blower discharge pressure, psig
  and 0.0115 is in units of kW-sq in-min/(ft-lb)-sq ft.

The average monthly energy consumption  (WE^) in
kWh can  be calculated by multiplying  Equation 7-14
by 720 hr/month to yield:
   WEM = (8.268
Step 5. Compute  the present  worth cost  of the
        monthly average  energy  consumption
        over the entire analysis period:
Ze =
                        (Ep) (SPWF)
(7-16)
where,
   Ze = present worth power cost, $
   Ep = unit cost of power, $/kWh
7.3.3 Calculate Maintenance Costs
Maintenance costs  are  expenditures made  on  a
regular basis for labor and materials needed to keep
the aeration system functioning  properly. They  cover
normal maintenance functions for such equipment as
blowers and instrumentation and might also include
the continuing costs of using a noninterruptive method
of diffuser  cleaning, such as gas cleaning.  The
present worth of maintenance costs is calculated  by
multiplying  the monthly maintenance cost  by SPWF
(Equation 7-4).

7.3.4 Calculate Diffuser Cleaning Costs
The analysis  assumes  that process interruptive
diffuser cleaning occurs  on a  regularly  scheduled
basis every nm months. The total number of cleanings
over the analysis period is  M.  The  present worth of
cleaning costs is calculated  by multiplying the cost of
an individual cleaning by PSPWF (Equation 7-5).

7.3.5 Calculate Replacement Costs
Calculation of  equipment costs  is similar to diffuser
cleaning costs. Assume a specific piece of equipment
must be replaced every nm  months and that the total
number of replacements over the analysis period is M.
Then the present worth replacement cost of this item
is its installed cost times the periodic series present
worth factor calculated by  Equation 7-5.  Repeating
this calculation for each replacement item and adding
the results together produces the total present worth
replacement costs.

7.3.B Calculate Jotaf Present Worth Cost
Add the initial, energy, maintenance, cleaning, and
replacement present worth  costs to obtain total
present worth cost for the cleaning interval selected.

7.3.7 Determine Lowest Total Present Worth Cost
K the lowest present worth cost  is desired for various
alternatives, repeat the calculations in Sections 7.3.1
through 7.3.5 using different conditions (e.g.,  cleaning
intervals) to determine the lowest total present worth.


7.4 Sample Desktop Economic Analysis
A  desktop economic analysis is performed below for
the fine pore  aeration system  design presented  in
Example 5-10. The  cost of  this system is  compared
with that of a coarse  bubble .aeration system. The
analysis demonstrates the sensitivity of aeration costs
to different  assumptions concerning fouling rates,
cleaning frequencies and prices,  and  power prices.

As; in  Example 5-10, each of the four aeration basins
is divided  into  three aeration  zones. The  average
nitrifying month and average  nonnitrifying month
OTRjs used in that example are assumed to be based
on the following  distribution  of monthly average
OTRfs:
                                                 181

-------
                           Total System OTR, Ib/d
                                                         Target DO, mg/L
                                  OTFVSOfR'
Month
Jan.
Fob.
March
April
May
Juno
July
Aug.
Sept-
Oct.
Nov.
Doc,
Average month, nttnlymg
Average month, nonnitrifying
Ovorall Averauo
Zonei
2,108
2,497
3,047
3,410
3,722
4,290
4,730
5,445
4,400
4,202
3,047
2,497
4,613
2,904
3.619
Zone 2
1,192
1,551
1,903
2,097
2,438
3,300
3,586
4,134
3,322
3,223
1,903
1,551
3,513
1,804
2,517
Zone 3
275
605
737
821
1,155
1,408
1,562
2,009
1,434
1.375
737
605
1,559
704
1,060
The nitrification period for this example plant is June-
October,

Oxygen transfer rates, even for clean diffusers,  vary
with  lime of year  due  to  changes  in  wastewater
temperature and target  DO levels in the  aeration
basin. The relationship of OTRf to SOTR for a new or
clean diffuser (F = 1.0) is:
where,
   KLa


   C
   T
   BC.T
       OTR(/SOTR = a BCj

                       H
                                            (7-17)
= (process   water  Ki_a  of   a  new
  diffuser) •? (clean  water  Ki_a of  a  new
  diffuser)
= apparent  volumetric  mass  transfer
  coefficient in  clean water at temperature T,
  1/hr
= process water DO concentration, mg/L
= process water temperature, °C
* OT20 {OiB°trM20 - C)-irC^o, as described
  in Equation 5-5.
Fof fixed DO concentrations, BCJ varies only slightly
within  the  temperature range of 10-25°C, having
values of about 0.86 and 0.77 at DOs of 1.0 and 2.0,
respectively.  Since  target  DOs in  the  example
problem were assumed to  vary by month, the
corresponding ratios of OTRf to SOTR are shown  at
the top of the next column.

Nole that Zone 2 DO in August is allowed to drop to 1
mg/L  in response to  the highest oxygen  demand  of
the year. Otherwise,  Zone 2 DO is  maintained  at 2
tngn.  during the nitrifying months and  1  mg/L during
the nonnitrifying months.
Months
Jan.-May
June-July
Aug.
Sept. -Oct.
Nov.-Dec.
Average
Zone 1
1
1
1
1
1

Zone 2
1
2
1
2
1

Zone 3
2
2
2
2
2

Zone 1
0,86o
0.86a
0.86a
0.86a
0.86a
0.86a
Zone 2
0.86a .
0.77a
" 0.86a
• 0.77a
0.86a,
0.83a
Zones
0.77a
Q.77q
Q.77,a
0.77a
0.77a
0.77a
Baseline conditions for the economic analysis assume
that  this  is  new  construction  using  an  annual  net
discount rate of 8 percent, a 20-yr analysis period, a
power  price  of $0,05/kWh, and a diffuser cleaning
cost of $1 .QQ/diffuser.  No equipment replacement
costs are considered.  Blowers are assumed  to  be
wired to an automated DO control system and capable
of having their output adjusted  to keep pace with the
airflows and pressures demanded.

7,4.7 Fine Pore System Design
Table 7-2 summarizes the information used to perform
a  cost analysis  of  the example  fine pore aeration
system. Initial costs of  the system  are $927,000 and
are itemized  in Table  7-3. In  this example,  annual
maintenance  costs  are taken  as 2  percent  of  the
installed blower costs. The assumed average a values
of 0.4, 0.6,  and  0.9 in Zones 1, 2,  and 3 result in
average OTRf/SOTR values of  0.344,  0.498, and
0.693, respectively (for new diffusers, F =  1.0).

The  number  of  diffusers  in each  zone  for all, four
aeration basins (from Example 5-10, Step 4) and their
respective densities are:

The SOTE values used in  this  analysis for the  above
diffuser densities were estimated by linear regression
of curves  presented in  Figure 5-9.  This was done to
simplify the analysis. Therefore, the calculated values
used in this chapter are slightly different than those
used in Example 5-10.

For this example, it is  assumed the  OTRf  of a fully
fouled fine pore diffuser is 40 percent  lower than for a
clean diffuser. Thus, the lowest possible F is 0.6. The
rate at  which  fine pore diffusers will foul is one  of the
most difficult operating  parameters to predict. In this
example,  four different  fouling  rates are  considered,
as shown  in Table 7-2.  They are based on the range
of values observed  at  actual  installations   as
summarized in Chapter  3. For all cases, Zone  2 fouls
1.4 times faster than Zone  3, while Zone 1 fouls twice
Zone
1
2
3
Number of
Diffusers
1,920
1,152
576
Diffusr Density,
Number/1 00 sq fl
48.2
28.9
14.5
                                                 182

-------
Table 7-2. Design Information for Example Fine Pore Aeration
         System
 Iniiiai Costs, $ (from Table 7-3)
 Replacement Cosls, $
 Monthly Maintenance Costs, $
 Unit Cosl of Energy, $/kWh
 Cosl of Diffuser Cleaning, $/diffuser
 Cleaning Interval, months
 Discount Rate, percent
 Analysis Period, months
 Barometric Pressure, psia
 Headloss from Submergence and
    Line Losses, psig
 Overall Blower Efficiency
927,000
      0
    400
      0.05
      1.00
(variable—see text)
      8
    240
      14.3
      6.2

      0.7
                           Zone 1    Zone 2
        Zones

          576
        1,060
            0.9
            0.693
 Number of Diffusers           1,920     1,152
 Average OTR,, Ib/d           3,617     2,517
 a                             0.4       0.6
 Average OTR,/SOTR"            0.344 '   0.498
 Rate of Fouling, percent loss of
    OTE/month (i.e., F x 100):
      "Case 1                 , 0.5       0.35
      "Case 2.                  1.0       0.7
      "Case 3                  3.0       2.1
      "Case 4                  5.0       3.5
 Maximum Percent Loss of        40       40
    OTE Possible
 Pressure Drop Across Diffuser, in:
    Clean                       5        5
    Fully Fouled                 24       24
 Orifice Pressure Drop at          2.67     2.67
    1 scfm, in
 Mixing Requirement, scfm       400      400
 Minimum Airflow/Diffuser,         0.5       0.5
    scfm
 Maximum Airflow/Diffuser,         2.5       2.5
    scfm
 SOTE at Minimum Airflow,        36.5      30.3
    percent
 SOTE at Maximum Airflow,       26.4      24.9
    percent
            0.25
            0.5
            1.5
            2.5
           40
            5
           24
            2.67

          400
            0.5

            2.5

           28.4

           23.2
Table 7-3. Initial  Costs of Design  Example  Fine Pore and
         Coarse Bubble Aeration Systems
Category
Piping
Diffusers
Blowers
Instrumentation & Control
Contingency
Mobilization, Bonding, Insurance
Fine Pore
System, $
51 ,000
131,000
240,000
150,000
143,000
57,000
Coarse
Bubble
System, $
42,000
24,000
320,000
150,000
134,000
54,000
 Total Estimated Construction Cost       772,000     724,000

 Engineering @ 15 percent              116,000     109,000
 Legal & Administrative @ 5 percent       39,000      36,000

 Total Estimated Initial Cosl              927,000     869,000
as  fast as Zone 3.  Overall,  the four sets  of fouling
rates increase in relative proportions of 1:2:6:10.

7.4.2 Coarse Bubble System Design
Table 7-4 lists  the information needed  to perform  a
cost  analysis of a new, equally-sized coarse bubble
aeration system to provide the same OTRfS  and DO
levels as  in  the fine  pore system  design. The  initial
cost of $869,000 is broken down as shown in Table  7-
3. Higher  capacity  blowers are needed for the coarse
bubble  system  than  for  the fine pore  system,
increasing the blower initial cost by  33 percent. A total
of  600 diffusers is  used.  The assumed average   a
values  in  Zones 1,  2, and 3 of  0.6, 0.7,  and 0.9,
respectively,  result in  average OTRf/SOTR  values  of
0.516,  0.581, and  0.693. SOTE values  are  assumed
to be 10 percent at  all airflows in  all aeration zones.
Since coarse bubble diffusers do not foul, there  is no
need to prescribe  a fouling  rate, a cleaning  cost,  or
minimum and maximum diffuser pressure drops.
Table 7-4. Design Information  for  Example Coarse  Bubble
         Aeration System

 Initial Costs, $ (from Table 7-3)           869,000
 Replacement Costs, $                         0
 Monthly Maintenance Costs, $                535
 Unit Cost of Energy, $'/kWh  "                  0.05
 Cost of Diffuser Cleaning, $/diffuser              0
 Cleaning Interval, months                  None
 Discount Rate, percent                        8
 Analysis Period, months                     240
 Barometric Pressure, psia                      14.3
 Headloss from Submergence and                6.2
   Line Losses, psig
 Overall Blower Efficiency                       0.7
                                                                                       Zone 1    Zone 2
                                                                    Zone 3
Number of Diffusers
Average OTRt, Ib/d
a
Average OTR,/SOTR
Rate of Fouling, percent loss of
OTE/month
Maximum Percent Loss of
OTE Possible
Pressure Drop Across Diffuser,
Clean
Fully Fouled
Orifice Pressure Drop at
1 scfm, in
Mixing Requirement, scfm
Minimum Airflow/Diffuser,
scfm
Maximum Airflow/Diffuser,
scfm
SOTE at Minimum Airflow,
percent
SOTE at Maximum Airflow,
percent
300
3,617
0.6
0.516

0 .
0

in:
0
0
0.1

400
2

15

10

10

200
2,517
0.7
0.581

0
o •


0
0
0.1

400
2

15

10

10

100
1,060
0.9
0.693

0
0


0
0
0.031'

400
2

15

10

10

                      " Umited mixing condition in Zone 3 permits the use of a larger orifice
                       with corresponding lower pressure drop.
                                                        183

-------
7.4.3 Comparison of Present Worth Costs
The desktop  method was  used in conjunction  with
Table 7-2 to determine the  present worth cost of this
system under each  fouling  rate  case  for  several
different diffuser cleaning frequencies.   Tables 7-5
and 7-6 include the  calculations  made for a cleaning
interval of  18 months and Case 3 fouling to illustrate
the steps involved. These calculations were repeated
for several other cleaning intervals to produce Table
7-7 and Figure 7-1, which plots total present worth  of
the  energy,  cleaning, and  maintenance  costs as a
function of cleaning  interval. The  optimum cleaning
interval for Case 3 fouling is seen to be 9  months with
a  present  worth  energy, cleaning, and  maintenance
cost of $394,000. Adding the initial cost of $927,000
yields  a total present worth  cost  of $1,320,000 for
Case 3 fouling.

The results of the calculations for all four fouling rates
are  summarized  in  Table 7-8. Figure  7-2 plots the
optimal cleaning interval, the percentage  of operating
costs devoted to  cleaning, and the present worth cost
per  diffuser for cleaning  against the fouling rate (as
measured  relative to Case  1,  the lowest rate used).
The optimal cleaning intervals range from 7 to 27
months. Even under the most  severe  fouling rate,
cleaning is less than  15 percent of the total operating
costs.

As indicated in Table 7-7, cleaning costs represent a
small fraction  of either Initial or energy costs for Case
3  fouling.  Further, the  trade-off between cleaning
cosls  and energy costs as  a function  of cleaning
frequency is relatively constant. Therefore,  for similar
situations where  operators  are faced with  a  Zone 1
fouling rate of approximately 3 percent/month, a wide
choice  of  cleaning  frequencies  is  available and  a
decision to  clean on  an  annual  basis would  be
consistent with the analysis shown.

Tables  7-9 and  7-10 summarize  the  desktop cost
computations  for the coarse bubble aeration system.
Total present worth  costs are $1,513,270, of which
$580,500 (38 percent)  are  for energy. These costs
are compared with fine  pore aeration system costs  in
Table 7-8. The fine pore system is more cost effective
than  the coarse  bubble  system under  all assumed
fouling rates.
7.4.4 Sensitivity Analysis
Additional computations were made to determine the
sensitivity of these economic results to changes in the
discount rate and the unit costs of energy and diffuser
cleaning. Using a 4 percent discount rate instead of
an 8 percent rale resulted in no change in the optimal
cleaning intervals.  Operating costs increased by  a
factor  of  1.38.  This is  exactly the  increase in the
uniform series present worth factor between discount
rales of 8 and 4 percent.
Table 7-11 illustrates the effect on total present worth
operating costs  (energy, cleaning, and  maintenance)
as the unit costs of power and diffuser  cleaning are
changed. As  the power price  increases and  the
cleaning  price decreases, it pays to clean more often.
When both prices rise in the same proportion, there is
no  change  in  the  optimal  cleaning interval. Total
operating costs are many times  more  sensitive to
changes  in  power  pricing  than to equal  relative
changes  in diffuser cleaning costs.

7.4,5 Retrofit Comparison
The  previous analysis compared fine pore  vs. coarse
bubble aeration  for  a  new  treatment plant.  This
section includes  a  similar  analysis assuming that a
coarse bubble aeration system already exists and  that
replacement with a fine pore aeration system  is being
considered. Operating  costs for the two  systems are
the same as calculated  before. No capital costs are
incurred  for the existing coarse bubble  system.  The
new fine pore system can use the existing blowers,
instrumentation and  control equipment,  and  above-
water air piping. Capital costs  for diffusers,  below-
water air piping,  and  basin cleanup/modification  are
assumed to be $65/diffuser. For 3,648 diffusers,  this
totals $237,120.

The  costs of  the fine pore aeration retrofit for the  four
fouling rates used previously are compared to the cost
of the existing coarse bubble aeration system in Table
7-12. The retrofit is  uneconomic only at the  highest
fouling rate using an 8 percent discount rate.


7.5  Lotus Spreadsheet
 A Lotus  1-2-3  (Version 2)  spreadsheet  calculation
approach  is presented in Appendix D. The user  can
input varying diffuser fouling rates, diffuser cleaning
intervals, and  monthly variations in  OTRf.  The
spreadsheet outputs  present worth costs for  the  fine
pore aeration system and provides graphical display of
costs and airflow requirements.

Tables 7-13  through 7-16 display the  spreadsheet
data input forms for the fine pore aeration  Case  3
fouling  rate problem  analyzed earlier in  this chapter.
Table 7-17 lists the total present worth costs of  this
system calculated based on the input design data  and
using the optimum  cleaning  interval of 9 months.
Figures 7-3 through 7-5 display the plots generated by
the spreadsheet. (The monthly cost and airflow graphs
were generated with  an analysis period of  5 years to
maintain sufficient detail in the plots.)

Table 7-8 also displays the  present  worth costs of
both the  fine  pore  and  coarse  bubble aeration
systems  calculated   with  the  computer-based
spreadsheet method. This method is identical to  the
desktop method except that OTRf, OTR,/SOTR, and F
values can vary by month. Because of this,  the results
will  be  somewhat sensitive to  the  month of the year
                                                  184

-------
Table 7-5. Sample  Desktop Calculations:  Case  3 Fouling  Rate for Fine Pore Aeration  Design  Example (18-month Cleaning
          Interval)


 A, Present Worth Initial Costs
    From Table 7-3, Pw (initial costs) =  $927,000

 B. Present Worth Energy Costs

    ,1.  Average airflows and pressure drops within each aeration  zone (see Table 7-6 for detailed calculations in each zone);

                                                                                        Zone 1      Zone 2      Zone 3   •

       Average airflow/diffuser, scfm                                                           0.882        0.729      0.694
       Average pressure drop, psig                                                            0.69         0.57        0.46
       Number of diffusers                                                               1,920       1,152       576

    2.  Total system airflow (qs):
       From Equation 7-10, qs = 0.882(1,920) + 0.729(1,152) •+  0.694(576) = 2,934 scfm

    3.  Total system pressure drop (P,j):
       Submergence  + line loss = 6.2 psig
       Maximum drop in aeration zones = 0.69 psig
       From Equation 7-11, Pd  = 6.2 + 0.69 = 6.89 psig

    4.  Monthly  energy consumption (WEM):
       Barometric pressure = 14.3 psia
       Blower efficiency = 0.7
       Compressibility factor =  [(14.3 + 6.89)/i4.3]°-283 -1  = O.T18
       From Equation 7-15, WEM = 8.268(2,934)(14.3)(0.118)/(0.7) = 58,475 kWh

    5.  Present worth cost of energy:
       Unit cost of power  = $0.05/kWh
       Discount rale = (8 percenl/yr)/100/12 = 0,0067/momh
       Analysis period = 240 months
       Monthly power cost = 0.05(58,475) = $2,924
       From Equation 7-4, SPWF = [(1 + 0.0067)240 . 1J/[Q.0067(1  •+• 0.0067)240] = 119.2
       Pw (energy costs) = $2,924(119.2) =  $348,540

 C. Present Worth Cost of Maintenance
       Discount rale = 0.0067/month
       Analysis period = 240 months
       Monthly maintenance cost = $400
       Pw (maintenance costs)  = $400(119.2) = $47,680

 D. Present Worth Cleaning Costs
       Cost to clean a diffuser = $1.00
       Total number of diffusers = 3,648
       Interval between cleanings  = 18 months
       Total number of cleanings = 240/18 = 13
       Cost per cleaning  = $3,648

       From Equation 7-2,
       PWF(0.0067,18) =  1/(1  + 0.0067)<«  = 0.887

       From Equation 7-4,
       SPWF(0.0067,234) = [(1 + 0.0067)234 - ij/[Q.0067(1 + 0.0067)234] = 118

       and

       SPWF(0.0067,18)  = [(1 +  0.0067)« - 1]/[0.0067(1 + 0.0067)18] = 16.9

       From Equation 7-5,
       PSPWF = 0.887<118)/16,9 = 6.19

       Pw (cleaning costs) = 3,648(6.19) = $22,580

 E, Total Present Worth Cost
       Pw (total cost)  = $927,000  + S348.540 + $47,680 + $22,580 = $1,345,800
                                                              185

-------
Tablo 7-6a. Sample Desktop Calculations (Case 3 Fouling Rate): Zone
          Cleaning Interval)
      1 of Fine Pore Aeration Design Example (18-month
           a Find average F:
                Cleaning Interval - 18 months
                Rate ol OTR loss = 3 percent/month
                Maximum loss of OTR = 40 percent
                Time to reach 40 percent loss = (40)/(3) = 13 months
                Average  F for months 1 to 13 = (1  + 0.6)/2 = 0.8
                Average  F for months 14 to 18 = 1  - (4Q)/{100) = 0.6
                Overall average F = (0.8)(13)/(18) + (0.6)(5)/(18) = 0.744

           b  Determine airflow per diffuser (q):
                Avoraae  OTR, - 3,617 Ib/d
                Average  OTR|/SOTR = 0.344
                Average  F » 0.744
                Number of diffusers = 1,920
                From Equation 7-6, q = 0.04(3,617)/[(0.344)(0.744)(l,920)(SOTE)] = 0.294/SOTE
                From Table 7-2 (assuming linear relation between q and SOTE):
                SOTE = 0.348 far q = 0.5, and SOTE = 0.264 for q =  2.5, giving
                SOTE » 0.369 - 0.042(q)

                From Equations 7-18 and 7-19, qd = 0.882 scfm/diffuser

           c. Chock constraints on q:
                Minimum airflow = 0.5 scfm/diffuser
                Mixing requirement = (400)/(1,920)  »  0.208 scfm/diffuser
                qd - MAX {0.882,0.5,0.208} » 0.882 scfm/diffuser
                Maximum airflow * 2.5 scfm/diffuser
                qd * WIN {0.882,2.5} = 0.882 scfm/diffuser

           d. Find pressure drop across diffuser and orifice:
                Pressure drop across clean diffuser = 5 in
                Pressure drop across fouled diffuser  =  24 in
                Average  degree of fouling = (1 -F)/(40/100) = 0.63
                Average  diffuser pressure drop = 5  +•  0.63(24 - 5} » 16.97 in
                Average  orifice pressure drop = (2.67)(0.882)2 = 2.08 in
                Total average pressure drop = (16.97  + 2.08)(0.036 psig/in) = 0.69 psig
                                       (7-18}



                                       (7-19)
chosen to begin the analysis.  For this example,  June
was taken as the start-up month.  A comparison of the
results In Table 7-8 shows that the optimal cleaning
Intervals arrived at by  the  two methods  are virtually
identical. The  energy costs resulting from use of the
spreadsheet method average 6.4 percent higher than
from the desktop  method,  while  the  total  present
worth costs vary by no more than 2 percent.


7.6 Compendium of Empirical Cost Data
The actual  cost of any fine pore aeration system will
be  highly dependent on site-specific factors. Although
empirical cost  data have  been obtained from many
sources,  the  application  of  these  data to  the
evaluation  of  any specific  installation  must  be
performed with caution. Where cost information was
obtained from  Canadian  installations,  a constant
conversion  rate of $0.80 U.S.  =  $1.00  Canadian was
used.

7.6.1 Basin Cleaning and Preparation Costs
Costs for basin cleaning  and  old  equipment  removal
are  too site  specific  for  general  estimates.  At the
Terminal Island treatment plant in Los Angeles, CA
(1),  only 40 labor-hr were required  for a 91.4 m x 9.1
m x 4.6 m sidewater depth (SWD) (300 ft x 30 ft x 15
ft) basin. At Monroe, Wl (2), 20 labor-hr were required
for each 31.1  m x 15.2 m x 4.6 m SWD (102 ft x 50 ft
x 15 ft) basin  for initial cleaning and degritting plus an
additional 12 labor-hr for removal of air lift knee joints.
At Green Bay, Wl (3), it was estimated that 120 labor-
hr were required  to dewater  and  clean each set of
contact (74.4  m x 11.1 m x 6.8 m SWD [244 ft x 36.3
ft x 22,3 ft]) and reaeration (74.4 m x 22.3 m x 6.2 m
SWD [244 ft x 73.3 ft x 20.5 ft]) basins.

Labor  for dewatering and  cleaning and degritting at
these three locations can be summarized as follows:
 Location
 Terminal Island
 Monroe
 Green Bay
Labor-hr/1,000 sq ft floor area
             4.4
             3.9
             4.5
7.6.2 Diffuser Costs
The variety  of fine pore  diffuser types,  materials,
sizes, and throughput capacities described  in Chapter
2 is matched by  an equally extensive variety of unit
costs for  installation  of these  systems.  Initial cost
information reported by various  participants in the
EPA/ASCE Fine Pore Aeration  Project and others is
summarized  in Table 7-18.  As far as practical, these
data  represent just the cost  of  the  diffusers, the
below-water air piping, and their installation. However,
reference to the comments in the  table is essential for
                                                      186

-------
Table 7-6b. Sample Desktop  Calculations (Case 3 Fouling Rate): Zone  2  of Fine Pore Aeration Design Example (18-month
          Cleaning Interval)
           a.  Find average F:
              Cleaning Interval = 18 months
              Rate of OTR loss = 2.1 percent/month
              Maximum loss of OTR = 40 percent
              Time to reach 40 percent loss = (40)/(2.1) = 19 months
              Average F for months 1 to 18  = 1 - [0.05(18)(2.1)/100] = 0.811
              Overall average F = 0..8

           b.  Determine airflow per diffuser (q):
              Average OTR, = 2,517 Ib/d
              Average OTR,/SOTR = 0.498
              Average F = 0.811
              Number of diffusers = 1,152
              From Equation 7-6, q = 0.04(2,517)/[(0.498)(0.8H)(l.l52)(SOTE)] = 0.216/SOTE
              From Table 7-2 (assuming linear relation between q and SOTE):
              SOTE =  0.303 for q = 0.5, and SOTE  =  0.249 for q = 2.5, giving
              SOTE =  0.316 - 0.027(q)

              From Equations 7-20 and 7-21, qd = 0.729 scfm/diffuser

           c.  Check constraints on q:
              Minimum airflow = 0.5 scfm/diffuser
              Mixing requirement = (400)/(1,152> = 0.347 scfm/diffuser
              qd = MAX {0.729,0.5,0.347} = 0.729 scfm/diffuser   .               ,
              Maximum airflow = 2.5 scfm/diffuser
              qd = MIN {0.729,2.5} = 0.729 scfm/diffuser

           d.  Find pressure drop across diffuser and orifice:
              Pressure drop across clean diffuser = 5 in
              Pressure drop across fouled diffuser = 24 in
              Average degree of fouling  = (1 - F)/(40/100) =  0.473
              Average diffuser pressure drop = 5 + 0.473(24 - 5)  = 13.98 in
              Average orifice pressure drop  = (2,67)(Q.729)2 = 1.42 in
              Total average pressure drop = (13.98 •*- 1.42)(0.036 psig/in) = 0.57 psig
                                        (7-20)


                                        (7-21)
a proper interpretation of the costs. As  indicated, it
was not always possible to isolate costs to this level.
Also, as noted in the comments, some costs include a
gas cleaning system as  part of a  ceramic  fine pore
aeration system package.

Oxygen transfer testing is an  additional  capital cost
item. Normally, the cost of a field  test is  not closely
related to the number of diffusers in an aeration basin.
Such testing can  add  several  dollars per diffuser to
the total  cost  of  the  system,  depending  on
specification  requirements. Shop  tests  also add to
initial costs, but not to  the extent of field tests.

7.6.3 Air Filtration Costs
As  discussed in Section 5.5.2,  a new or upgraded air
filtration system may be needed. At Madison, Wl (10),
the $31,000 cost of the new air filtration system (static
filters  plus  rotating   fiberglass  filters  for  three
centrifugal blowers each with  a capacity of 11,800  Us
[25,000  scfm])  was   equivalent  to  $1.99/installed
diffuser.  This  system   was estimated to  require  10
labor-hr/yr for  maintenance and $200/yr for materials.
Green  Bay, Wl  (3) selected  an  in-line,  12,980-L/s
(27,500-scfm)  capacity canister filter for their ceramic
disc  quadrant.   The  $30,000  installed   cost
($3.62/diffuser)  achieves  90  percent  removal  of
particles /1 micron,  and an estimated  18 labor-hr/yr
are required for maintenance.          ,

The air filtration system at Monroe, Wl (2) - air supply
capacity of 3,825 Us (8,100  scfm) - is  an American
Airfilter UF-H  unit consisting  of  a disposable glass
fiber prefilter and a Biocel DH final filter. Maintenance
is estimated to require  30  labor-hr/yr and $200/yr for
materials.  The 525-L/s  (12-mgd) Whittier  Narrows
treatment  plant in  Los  Angeles  County,   CA   (i)
installed an air filtration system in 1980 at a cost of
$13,000 for three centrifugal  blowers with a  nominal
total output of  13,925 Us (29,500  scfm).  They replace
the glass fiber prefilters four  times per year  and the
high-efficiency  pleated  paper  filters  once per year.
Estimated annual filter replacement costs are $2,000.

San Mateo, CA (21)  has a double dry coarse and fine
air filtration system for an installed blower capacity of
8,310  Us (17,600 cfm)  that requires  96 labor-hr/yr for
maintenance  and $2,000/yr  for the  replacement
elements.  Durango,  CO  (22) uses a 3-stage  air
filtration system for  a 2,125-L/s (4,500-cfm)  total
installed  blower  capacity  to achieve  95  percent
removal of 0.5-micron 'particles.  Filters  are  changed
annually  (2 labor-hr/yr),  and  materials  costs  are
estimated at $600/yr.
                                                      187

-------
Tobto 7-6c. Sample  Desktop Calculations (Case 3  Fouling Rate): Zone  3 of  Fine Pore Aeration Design Example (18-month
           Cleaning Interval)
            a Find average F:
                 Cleaning interval = 18 months
                 Rate of OTR toss = 1.5 percent/month
                 Maximum toss of OTR  = 40 percent
                 Time to reach 40 percent loss =  (40)/(1.5) = 27 months
                 Average F for months 1 to 18 -  i - [0.5(1 8){1.5}/100J = 0.865
                 Overall average F = 0.865

            b. Determine airflow per diffuser (q):
                 Average OTR, =  1,060 Ib/d
                 Average OTR,/SOTR - 0.693
                 Average F = 0365
                 Number of ditfusers =  576
                 From Equation 7-6, q = 0.04(1 ,060)/[(0.693)(0.a65)l576)(SOTE)l = 0.123/SOTE
                 From Table 7-2 (assuming linear relalion between q and SOTE):
                 SOTE  = 0.284 for q = 0.5, and SOTE = 0.232 for q = 2.5, giving
                 SOTE  =  0.297 - 0.026(q)

                 From Equations 7-22 and 7-23, qd = 0.430 scfm

            c. Cl«ck constraints on q;
                 Minimum airflow = 0.5 sefm/diffuser
                 Mixing  requirement = (400J/(576) = 0.694 scfm/diffuser
                 %  « MAX {0.430,0.5,0.694}  =  0.634 scfm/diffuser
                 Maximum airflow = 2.5 scfm/diffuser
                 qd  = MIN {0.694,2.5} = 0.694 scfm/diffuser

            d, FKK! pressure drop across diffuser and orifice:           -                 ,
                 Pressure drop across ctean diffuser =  5 in
                 Pressure drop across footed diffuser = 24 in
                 Averafle degree of fouling - (1 - F)/(4Q/10Q) = 0.338
                 Average diffuser pressure drop = 5 + 0.338(24 - 5)  = ti.42 in
                 Averafle orifice pressure drop - (2.67)(0.694)2 = 1 .29 in
                 Tote! average pressure drop = (11.42 + 1.29)(0.036 psig/in) = 0.46 psig
                                         (7-22)


                                         (7-23)
Table 7-7. Present Worth Costs as a Function of Cleaning Interval for Case 3 Fouling Rate
Ctaarwno, Interval,
months
3
4
6
8
| 0
10
11
12
13
15
18
24

Maintenance
48
48
48
48
48
48
48
48
48
48
48
48

Cleaning
144
108
72
53
47
42
38
35
32
28
23
17
• Present Worth Costs, $1,000
Energy
270
274' • .
284
294
299
304
310
316
322
334
349
371

initial
927
927
927
S27
927
927 . •
927
927
927
927
927
927

Total
1,389
1,357
1,331
1,322
1,320
1,321
1,323
1,326
1,329
t,337
1,346
t,363
7.6.4 Blower Costs
Figure 7-6 summarizes blower costs updated to 1988
(Marshall &  Swift ECI  = 847} obtained from a variety
of sources.  The data from  EPA's  CAPDET cost
estimating program (24) are for a 1,415-L/s  (3,000-
scfm) rotary positive displacement blower; a vertically
split, multistage 5,665-L/s  02,000-scfm) centrifugal
btower;  and a  pedestal-type, single-stage 23t600-Us
(SQ.QOQ-scfm) centrifugal blower. Also shown in Rgure
7-6 are uninstalled blower costs for several Canadian
installations  (20).  These  costs are for  individual
blowers  of the  indicated capacity. Increasing  these
costs by 40-60 percent may be a reasonable estimate
of installation  costs  for retrofits  where  a blower
buMing  already  exists   (20).   For  totally  new
construction,  purchased  blower  costs  can  be
increased  by  200-300  percent to  estimate  final1
installed costs (20).
                                                       188

-------
          Figure 7-1.   Present worth operating costs of example fine pore aeration system for Case 3 fouling rate.
           Present Worth Cost, $1,000
                  600  ._
                  500
                  400
                  300
                  200
                  100
                                                                 Energy
                                                                 Cleaning
                                                    10             15
                                                , Cleaning Interval, months
                                                             20
                                                                                               25
Table 7-8. Economic Comparison of Newly Constructed Fine Pore and Coarse Bubble Aeration Systems
                                                             Lowest Present Worth Cpsis, $1,000
                   Optimal Cleaning 	
    Fouling Rale
Optimal Cleaning
 Interval, months
Maintenance
Cleaning
Energy
                                                     Initial
                                                                      Total
    Results of Desk-Top Computations
    Case 1
    Case 2
    Case3
    Case 4"
    Coarse Bubble
      27
      18
       g
       7
    48
    48
    48
    48
    64
    Results of Spreadsheet Compulations
   14
   23
   47
   61
    0
  277
  284
  299
  314
  580
927
927
927
927
869
1,265
1,281
1,320
1,349
1,513
Case 1
Case 2
Case 3
Case 4
Coarse Bubble
24
19
9
T

48
48
48
48
64
17
21
47
61
0 -
293
305
320
337
610
927
927
927
927
• ' 869
1,284
1,301
1,341
1,372
1,543
                                                          189

-------
            Figure 7-2.   Optimal cleaning intervals and costs for example fine pore aeration system for
                        alternative fouling rates.
             Dollars or Months

                   30  _
                  25
                  20
                   15
                   10
                                                               $ of Operating Costs
                                                                         10
                                                                                   12
                                             Fouling Rate Relative to Case 1
Figure 7-6 also shows uninstalled  blower and piping
costs lo the aeration basin for five  facilities plotted as
a function of total installed blower capacity rather than
as a function of individual blower sizes (22). For these
five  facilities,  total  blower costs (including buildings,
blowers, piping to  basins, electrical work,  and all
related  construction costs) averaged 300 percent of
the uninstalled blower and piping costs.

Many small facilities use rotary, positive displacement
blowers. For small  to intermediate  size (<220 Us [5
ifiOdl)  plants, the  capital  cost  of  the blowers
{uninstalled)  is usually about 70-80 percent  of the
capital cost of the aeration equipment itself (19). This
cost estimate  applies to blowers <75 kW (100 hp).

In 1985, Madison, Wl (10) installed  three  new 900-kW
(1,200-hp) centrifugal blowers, each with a capacity of
5,900-11,800  Us  (12,500-25,000  cfm).  Installed
blower  costs  excluding  the blower building were
$663,000 ($18.73/L/s [$8.84/scfm]  capacity)  with an
additional $254,000 for an automated control system.
The installed  cost  per  individual blower  (adjusted to
1988) is shown in  Figure  7-6. For 1986, DO  sensor
maintenance  associated  with  the  control  system
required 180  labor-hr and  $90  for  materials.  In
comparison, the new blowers required 40 labor-hr and
$200 for materials in 1986.
The blower system at the Terminal Island  treatment
plant in Los Angeles,  CA (1) was. not upgraded when
fine pore diffusers were added. Three existing 1,120-
kW (1,500-hp) centrifugal Roots blowers are capable
of delivering 18,400 L/s (39,000 acfm) each. The plant
aeration  system  is automated,  requiring 500  labor-
hr/yr  for  DO sensor maintenance.  The  blowers,
originally installed in 1976,  required an estimated 500
labor-hr for maintenance  and an estimated  $12,000/yr
for materials in 1987,
Monroe,  Wl  (2)  installed four new 956-L/s  (2,025-
scfm) positive displacement blowers in their existing
blower building in 1985.  Installed  costs, including the
air f 1,120-kW (1,500-hp)  centrifugal  Roots blowers
are capable  of delivering 18,400 L/s  (39,000 acfm)
each.  The  plant  aeration  system  is  automated,
requiring 500 labor-hr/yr  for DO sensor maintenance.
The  blowers, originally installed in 1976, required an
estimated  500  labor-hr for  maintenance  and  an
estimated $12,000/yr for materials in 1987.
                                                   190

-------
Table 7-9. Sample Desktop Calculations for Coarse Bubble Aeration Design Example
 A. Present Worth Initial Costs
    From Table 7-3, Pw (initial costs) = $869,000

 B. Proseril Worth Energy Cosls

    1. Average airflows and pressure drops within each aeralion zone (see Table 7-10 for detailed calculations for Zone 1):

                                                                                        Zone 1       Zone 2      Zone 3

       Average airflow/diffuser, scfm                                                           9.35         8.66        6.12
       Average pressure drop, psig                                                            0.315        0.270       0.042
       Number of diffusers                                                                 300         200        100

    2. Total system airflow (qs):
       From Equation 7-10, qs = 9.35(300) + 8.66(200) + 6.12(100) = 5,149 scfm

    3. Total system pressure drop (Pd):
       Submergence  + line loss = 6.2 psig
       Maximum  drop in aeration zones = 0.315 psig
       From Equation 7-11, Pd  = 6.2  + 0.315 = 6.52 psig

    4. Monthly energy consumption (WEM):
       Barometric pressure = 14.3 psia
       Blower efficiency = 0.7
       Compressibility factor =  [14.3 + 6.52)/i4.3]0.283 -1 » 0.112
       From Equation 7-15, WEM = 8.268(5,149)(14.3}(0.112)7(0.7) = 97,400 kWh

    5. Present worth cost of energy:
       Unit cost of power  = $0.05/kWh
       Discount rale =  (8 percent/yr)/iOO/i2 = 0.0067/month
       Analysis period = 240 months
       Monthly power cost = $0.05(97,400)  = $4,870
       From Equation 7-4, SPWF = [(1  + 0.0067)24<> - 1J/[O.OQ67(1  + 0.0067)210j = 119.2
       Pw  (energy costs) = $4,870(119.2) = $580,500

 C. Present Worth Cost of Maintenance
       Discount rate =  0.0067/morilh
       Analysis period « 240 months
       Monthly maintenance cost =  $535
       From Equation 7-4, SPWF = 119.2
       Pw  (maintenance costs)  = $535 (119.2) = $63,770

 D. Present Worth Cleaning Costs
       Diffuser cleaning costs = 0

 E. Total Present Worth Cost
       Pw (total cost)  = $869,000 + $580,500 + $63,770  + $0 = $1,513,270


Table 7-10. Sample Desktop Calculations for Zone 1 of Coarse Bubble Aeration Design Example

             a.  Average F = 1.0

             b.  Determine airflow per diffuser, q:
                  Average oxygen demand  = 3,617 ib/d
                  Average OTR,/SOTR  = 0.516
                  Average F = 1.0
                  SOTE = 10/100  = 0.1
                  Number of diffusers = 300
                       From Equation 7-6, q = 0.04(3.617)*[(0.516)(1.0)(300)(0.1)] =  9.35 scfm/diffuser

             c.  Check constraints on q:
                  Minimum airflow  = 2 scfm/diffuser
                  Mixing requirement =  (400)/(300) = 1.33 scfm/d iff user
                  qd = MAX (9.35, 2, 1.33} =  9.35  scfm/diffuser
                  Maximum airflow =15 scfm/diffuser
                  qd = MIN {9.35, 15} = 9.35 scfm/diffuser

             d.  Find  pressure drop :
                  Average orifice pressure drop  = (0.1)(9.35)2 = 8.74 in
                  Total average pressure drop = (8.74)(0.036 psig/in)  = 0.315 psig
                                                             191

-------
 Table 7-11. Sensitivity of Fine Pore Aeration Costs to Changes in Price of Power and Diffuser Cleaning
Power Price,
S/kWh
Fouling Rale
0.05
0,05
0.10
0.10
Fouling Rate
0.05
0.05
0.10
0.10
Cleaning Price,
$/diffuser

1.00
0.50
1.00
0.50

1.00
0.50
1.00
0.50
Tablo 7-12. Economic Comparison
Optimal Cli


Interval, months Mainl.

27
19
19
13

9
7
7
5










of Retrofit Fine Pore and
Existing Coarse Bubble Aeration
Lowest Present Worth Costs,
FoulMifl Rate

8'perceftl Discount

Case 1
Case 2
Case 3
Case 4
Coarse Bubble
4-pofccml Discount
Case 1
CasoS
Case 3
Case 4
Coarse Bubble
Operating

Rate

333
354
397
422
647
Rale
468
489
543
584
893
Initial



237
237
237
237
0

237
237
237
237
0
Systems
$1,000
Total



576
591
634
659
647

705
726
780
821
893

48
48
48
48

48
48
48
48
Present Worth C
Clean.

14
11
21
16

47
30
61
43
lost, $1,000
Energy

277
271
541
532

299
288
577
558
Table 7-13. Input Data Form for Oxygen

Tot. Operating „

339
330 ;,
610 ,
596

394
366
686 '
649
Requirements for
Economic Analysis Spreadsheet - Case 3 Fouling


Month

Jan.
Feb.
Mar.
April
May
June
July
Aug.
Sep.
Oct.
Nov.
Dec.

Rate
Total System OTR

i, Ib/d
Zone 1 Zone 2 Zone 3 Zone 1

2,108 1,192
2,497 1,551
3,047 1 ,903
3,410 2,097
3,722 2,438
4,290 3,300
4,730 3,586
5,445 4,134
4,400 3,322
4,202 3,223
3,047 1,903
2,497 1,551


275 0.344
605 0.344
737 0.344
821 0.344
1,155 0.344
1,408 0.344
1 ,562 0.344
2,009 0.344
1 ,434 0.344
1,375 0.344
737 0.344
605 0.344


OTR,/SOTR
Zone 2 Zone 3

0.516 0.693
0.516 0.693
0.516 0.693
0.516 0.693
0.516, 0.693
0.462 0.693
0.462 0.693
0.516 0.693
0.462 0.693
0.462 0.693
0.516 0.693
0.516 0.693

                                                     Press < All-M > to return to Main Menu
                                                     Press < All-D > to return to Data Menu
Monroe,  Wl  (2)  installed four new 956-L/s (2,025-
sclm) positive displacement blowers in their existing
blower building in 1985.  Installed costs, including the
air foperation/hr  of  maintenance.  DO  sensor
maintenance at this plant requires 35 labor-hr/yr and
5100 for materials.

San Mateo, CA  (21) operates two 112-kW (150-hp)
and five  93-kW  (125-hp) blowers that  require  about
200 labor-hr/yr for routine repair and maintenance and
an additional $20,000/yr in materials costs.

The capital cost  of two new  1,220-L/s (2,590-icfm)
centrifugal  blowers purchased by Frankenmuth,  Ml
(12) was estimated in 1985 as $22,000, or $9.01/L/S
($4.25/scfm) capacity.  The  uninstalled  cost  per
blower, adjusted to 1988, is also given in Figure 7-6.
In the  absence of specific blower maintenance  cost
information,  estimates  of  annual maintenance
requirements of  3  percent  of the  uninstalled
mechanical equipment cost for centrifugal blowers and
5 percent for positive displacement blowers (20), and
5 percent for either type  of  blower  (25) have been
recommended.

7.6.5 Ceramic Diffuser Gas Cleaning Costs
Sanitaire markets a proprietary  gas  cleaning process
(U.S.  Patent  No.  4,382,867)  that normally  uses HCI
gas. Over 60 systems  had  been  installed  as  of
November 1988. The 1988 installed cost for the gas
delivery hardware and license  fee was reported  (26) to
average  $10/diffuser with a typical range  of $8-
12/diffuser, depending  on  system size.  Of  this total,
the royalty or license fee is $6/diffuser.  For large
                                                  192

-------
Table 7-14. Input Data Form for Diffuser Characteristics for
          Economic Analysis Spreadsheet - Case 3 Fouling
          Rate

                             Zone 1   Zone 2   Zone 3
                 Table 7-17. Output  Display from Economic Analysis
                           Spreadsheet - Case  3 Fouling  Rate (9-month
                           Cleaning Interval)
Minimum Airflow, scfm/diffuser
Maximum Airflow, scfm/diffuser
SOTE at Min. Airflow, percent
SOTE at Max. Airflow, percent
OTE loss rate, percent/month
Maximum percent OTE Loss
Minimum (Clean) DWP, in
Maximum (Fouled) DWP, in
Orifice P-drop at 1 scfm, in
Mixing Requirement, scfm
Number of Diffusers
0.5
2.5
34.8
26.4
3
40
5
24
' 2.67
400
1,920
0.5
2.5
30.3
24.9
2.1
40
5
24
2.67
400
1,152
0.5
2.5
28.4
23.2
1.5
40
5
24
2.67
400
576
Press < Alt-M > to return to Main Menu
Press < Alt-D > to return to Data Menu
Table 7-15. Input Data Form for Economic Factors  for
          Economic Analysis Spreadsheet - Case 3 Fouling
          Rate (9-month Cleaning Interval)
              Factor
 Value
     Discount Rate, percent
     Diffuser Installation Cost, $/unii
     Other Initial Costs, $
     Electricity Rate, $/kWh
     Diffuser Cleaning Cost, $/unit
     Cleaning Interval, months
     Routine Maintenance, $/yr
     Analysis Period, months
     Start-up Month (Jan = 1, etc.)
      0

927,000 ,

      0.05
      1

      9
  4,800

    240

      6
Press < Alt-M > to return to Main Menu
Press < Alt-D > to return to Data Menu
Table 7-16. Input  Data  Form for  Blower Characteristics for
          Economic Analysis Spreadsheet - Case 3 Fouling
          Rate
               Item
                                        Value
Barometric Pressure, psia.
Pressure Head w/o Diffusers, psig
Overall Blower Efficiency, percent
14.3
6.2
70
Press < Alt-M > to return to Main Menu
Press < Ait-D > to return to Data Menu
sanitary authorities with  many  plants,  it  may  be
possible to obtain a "blanket license" for as low as
$2-3/diffuser. The gas dosage per cleaning is usually
90-100 g (0.02-0.2 Ib)  HCI/diffuser (2.7). Gas cleaning
costs are summarized in Table 7-19 for nine treatment
plants.

The  estimated  installed  cost of  a  Sanitaire  gas
cleaning system  at  Frankenmuth, Wl  (12),  including
                                                             Operating Period, months
                                                             Present Worth Costs, $1,000:
                                                               initial
                                                               Energy
                                                               Cleaning
                                                               Maintenace
                                                               Total
                                                          240


                                                          927.0

                                                          319.5

                                                          46.7

                                                         , 47.8

                                                        1,341.0
fees,  was  $25,000, or $i0.42/diffus.er. At this plant,
dynamic wet pressure (DWP) is  monitored on  a
weekly schedule (3 diffusers} requiring about 50 labor-
hr/yr.  Cleaning is normally  performed at 38-46 cm  (15-
18  in) DWP. The average HCI  cost for 1987  was
$3.42/kg ($1.55/ib). From  April 1986 through  August
1987, an average  32 kg (71  Ib) HCI/month were used,
with the average dose per diffuser varying during the
period, ignoring  the very high dosages during the  first
2 months of  start-up, the average monthly dose was 9
g  (0.02   lb)/instal!ed diffuser.  Average  labor
requirements for  gas  cleaning were  0.1  labor-hr/kg
(0.045 labor-hr/lb)  HCI delivered.

The total  installed cost of a gas  cleaning system at
Green  Bay,  Wl  (3), consisting  of the  building
($59,100),  royalty, and feed system, was $128,334, or
$15.50/diffuser in  1986. The normal gas  dose  was 45
g  (0.1  Ib) HCI/diffuser at an  average  HCI  cost of
$1.48/kg ($0.67/lb). Thus far,  5  of the  10  installed
grids  have been cleaned  with a labor expenditure of
14  labor-hr/cleaning,  which  is equivalent to  0.075
labor-hr/kg (0.034  labor-hr/lb) HCI or 4.6 labor-hr/1,000
diffusers,  for changing cylinders  and  monitoring  the
gas cleaning  system during the procedure.

A  gas  cleaning  system  at  the   Whittier  Narrows
treatment plant  in  Los Angeles County,  CA  (1)  was
installed at a nominal cost for research purposes.  HCI
costs (1987) were  $2.58/kg ($1.17/ib)'using  272-kg
(600-lb)  cylinders, plus $500/cylinder for shipping.
Additional  costs  are   incurred  for demurrage  and
loading and  unloading cylinders.   Gas dosages have
averaged  23 g  (0.05  lb)/diffuser.  The  estimated time
required for a cleaning is:

      Setup  time                  2.0 hr
      HCI cleaning time:
         1,000 units per grid       2.5 hr
         500 units per grid        1.5 hr
      Cleanup time               1.5 hr

Setup and cleanup times include moving cylinders,
attaching  hoses, and  loading and unloading  needed
equipment.
                                                     193

-------
     Figure 7-3.   Present worth costs generated by economic analysis spreadsheet - example fine pore aeration system:
                Case 3 fouling rate.
               Presenl
               Worth
               Cost,
               $1,000

                 1,4

                 1.3


                 1.2


                 1.1


                   1


                 0.9


                 0.8


                 0.7


                 0.6


                 0.5


                 0.4


                 0.3


                 0.2


                 0,1
927.0
             319.5
                                                     46.7
                                                                  47.8
                          Capital
             Energy        Cleaning      Maintenance      Total
7.6.6 Other Diffuser Cleaning Methods and O&M
Costs
A  variety of O&M activities  are  associated  with  the
use of fine pore aeration systems, including:
•  monitoring for diffuser fouling,
•  basin drainage for visual inspection,
•  cleaning diffusers by high-pressure hosing,
•  cleaning diffusers by brushing,
•  cleaning  diffusers  by  chemical  treatment/hosing
   (Milwaukee method),
                             •  other less  frequently  used diffuser  cleaning
                                methods,
                             •  diffuser membrane replacement, and
                             •  repair of cracks, leaks, etc.


                             The cost of  ceramic dome diffuser cleaning  was
                             significantly  influenced  by  the  work  rules  and
                             procedures  employed at a  particular  plant (28).  This
                             analysis  of the unit  operations associated  with  basin
                             and diffuser cleaning led  to  the  generalized   time
                                                    194

-------
    Figure 7-4.   Monthly operating costs generated by economic analysis spreadsheet during first 5 years of operation -
                example fine pore aeration system: Case 3 fouling rate.
                  7 -
              "96-
            o  3  :
            Q  o  •
                  5 -
                  4 -
                 3 -
                 2  -
                                    Total
                           Power
                                               20
                                                                         40
                                                      Month
estimates for various  unit operations summarized in
Table 7-20.

At Madison, Wt (10), a combination of off-gas testing
and  DWP monitoring  requires 40  labor-hr/yr.  Annual
inspections  require just 4 labor-hr  for basin drainage
and  gross cleaning of each  10,200-m3 (2.7-mil gal)
basin. High-pressure hosing and acid application were
each estimated to require 0.03 labor-hr/diffuser with a
chemical cost for acid  application of $0.014/diffuser.

By  the  end  of 1986,  Ridgewood,  NJ (6)  had 3-3/4
years of experience with  their ceramic dome diffuser
system. Diffusers were cleaned with a high-pressure
fire hose or acid cleaned with 20 percent HCI diluted
1:1  with tap water  and  scrubbed  into each  diffuser.
Hose cleaning  required  0,012 labor-hr/diffuser and
acid cleaning  0.025 labor-hr/diffuser. A summary of
diffuser cleaning and  repair costs for Ridgewood is
given in Table 7-21.

Repairs included tightening bolts, sealing cracks, and
adjusting and replacing domes and bolts. Ridgewood
now uses only fire hose  cleaning in  early Spring and
late Fall for each basin (8).

The modified Milwaukee diffuser cleaning method was
applied  at the  Whittier Narrows treatment plant (29).
 The method consists of: 1) low-pressure hosing  for
 approximately 7.5 seconds/diffuser, 2) application of
 50-percent muriatic acid solution to the media surface
 of each diffuser, 3) air/acid  agitation for approximately
 10 minutes, and  4)  another 7.5 seconds/diffuser of
 low-pressure  hosing.  Cleaning 2,026 ceramic disc
 diffusers required  42 labor-hr, or 0.021   labor-
 hr/diffuser. Of this total, 12 labor-hr were for the top '
 man connected with a safety line to  the  bottom men
.because  the  basins were  considered  as  confined
 spaces.  Cleaning costs for  acid,  air,  water and
 capitalized  equipment  costs  (spray  equipment,
 protective clothing, etc.) amounted to $145/cleaning,
 or $0.071/diffuser. The $20/hr labor cost increased the
 total unit cleaning  cost to $0.49/diffuser.

 Ceramic  dome  diffusers  at  the San  Mateo,  CA
 treatment plant (21)  are cleaned every 6  months
 because  of the development of coarse  bubbling and
 increased blower discharge  pressures. Dewatering the
 aeration basins  (3,785 m3  [1.0 mil gal] capacity)  by
 pumping and gross cleaning require about 24 hours or
 0.008  labor-hr/diffuser.  Several  cleaning methods,
 including  acid cleaning, have been evaluated but the
 combination  of  high-pressure  hosing  and  brushing
 yields satisfactory results and has become the method
 of choice. Brushing  and high-pressure hosing require
                                                   195

-------
     Figure 7-5.  Monthly airflow generated by economic analysis spreadsheet during first 5 years of operation - example
                fine pore aeration system: Case 3 fouling rate.
                                                                                                    6(
                                                                 Zone 3
                               A  Total
 a total of 0.016 labor-hr/diffuser. Checking bolt tension
 and leak  checking the  system  after cleaning adds
 another 0.014 labor-hr/diffuser,  resulting in a total cost
 of $0.57/d!ffuser  ($15.00/hr labor)  for a  complete
 cleaning  cycle. The diffuser failure  rate  is  only  1-2
 units/1,000 diffusers/cycle.  This rate  has been  steady
'over the 4 years since system installation.

 Hartford, CT  (15,16)  cleans  according to the
 Milwaukee method as follows:  dewater (air on), high-
 pressure hose (air on),  apply 50 mL acid/diffuser (air
 off), brush diffusers (air off), and rehose (air  on). The
 acid solution  is a  commercial  product  (ZEP)
 containing 22 percent HCI  and  surfactants. Estimated
 costs for the Milwaukee method as applied at Hartford
 are summarized in Table 7-22.

 With a  labor  rate  of $20/hr, the cost per cleaning at
 Hartford, including  leak repair,  was  $1.35/diffuser.
 Excluding the costs for  spare  parts and  leak  repair,
 the  average  cleaning cost was  $0.70-$0.75/diffuser.
 These cleaning costs do not include grit removal.

 Monroe, Wl  (2)   has   three  2-pass  basins,  each
 containing 900 ceramic disc diffusers. Aeration basin
 dewatering requires  pumping  and  consumes   0.009
 labor-hr/diffuser. Cleaning is by high-pressure hosing,
acid spray/soak, brushing, and rehosing.  Both hosings
require  a total of  0.006  labor-hr/diffuser,  acid
application 0.005 labor-hr/diffuser, and brushing 0.007
labor-hr/diffuser. An additional  0.004 labor-hr/diffuser
is  spent  checking  bolt  tension  and  for  leaks.  No
failures were observed in the first year of operation. At
Monroe's $12/hr labor rate, cleaning costs  for labor
total $0.37/diffuser,  of which $0.11/diffuser is  for basin
dewatering.

The cost of a  kiln, electrical hookup,  and accessories
installed  at Bozeman, MT (30,31)  in  1982  was about
$3,000. The ceramic disc  diffusers  in a given basin
(682/basin) are  changed every 1-2  years  and  the
removed  diffuser  stones fired  50  at a  time  at
temperatures  of  1,540-1,760C  (2,800-3,200F).  The
diffuser stones are  placed  in the blower room to dry
for about  1 month before firing since firing wet stones
led  to  problems with  breakage.  Careful  placement of
stones in the  kiln  is required, representing  about  1
hr/load (0.02  hr/diffuser). Removal time is about 30
minutes/load. In  1978, labor for  removing the diffuser
stones from the basins  (remove screw  caps and 0-
rings,  take to kiln) was reported as  0.035 labor-
hr/stone and reinstallation was also reported  as 0.035
labor-hr/stone. Stone removal time has been reduced
                                                    196

-------
Table 7-18.   Initial Cost Information for Selected Fine Pore Aeration Systems
Plant
Green Bay, WI
Green Bay, Wl
Ridgewood, NJ
Madison, Wl
Frankenmuth, Ml
Valencia, CA
Terminal Island, CA
Terminal Island, CA
Trinity River Aiith., TX
Newark, .OH
Hartford, CT
Glastonbury, CT
Whittier Narrows, CA
Monroe, Wl
Richardson, TX
Amsterdam, NY
Sycamore Creek, OH
Piqua, OH
Guelph No. 1, Canada
Guelph Nos. 2 and 3,
Canada
Mid Hallon, Canada
Georgetown, Canada
Oakville, Canada
Celdonia, Canada
San Mateo, CA
Broomfield, CO
Yakima, WA
Serra, CA
Litllelon, CO
Edmonds, WA
Dale Installed
May 1986
Jan. 1986
April 1983
1985
Deo. 1985
SepL 1986
June 1987
April 1987 •
Oct. 1987
Spring 1985
Fall 1982
Summer 1984
1980
1985
Nov. 1987
Sept. 1986
Under Consl.
March 1988
1986
1988
1988
1987
1987
1988
1985-1986
1987
1988
1989
1985
Planned
Diffuser Type
Sanitaire Ceramic
Discs
Parkson Pert.
Membrane Tubes
Gray Ceramic Domes
Sanitaire Ceramic
Discs
Sanitaire Ceramic
Discs
Nokia Rigid Porous
Plastic Discs
AERTEC Nonrigid
Porous Plastic Discs
Parkson Pert.
Membrane Tubes
Sanitaire Ceramic
Discs
Parkson Pert.
Membrane Tubes
Norton Ceramic Discs
FMC Rigid Porous
Plastic Tubes
Sanitaire Ceramic
Discs
Norton Ceramic
Domes
Sanitaire Ceramic
Discs
EDI Ceramic Plates
EDI Ceramic Plates
EDI Ceramic Plates
EDI Ceramic Plates
Sanitaire Ceramic
Discs
Sanitaire Ceramic
Discs
Sanitaire Ceramic
Discs
Sanitaire Ceramic
Discs
Sanrlaire Ceramic
Discs
Envirex Pert.
Membrane Discs
Norton Ceramic
Domes
Gray' Ceramic Domes
Sanitaire Ceramic
Discs
Statiflo Ceramic
Domes
Gray Ceramic Domes
Sanitaire Ceramic
Discs
Diffusers Material,
Installed $/diffuser
8,276 33.83
6,018 55.33
1,080
15,576
2,400 47.08
5,490 24.42
770 32,47
1,000 60.00
40,250
4,000
13,316 -25
320 -38
2,026
5,064
2,700
112 288.40
160 237.50 "
427 278.69
170. 296.18
4,000 44.27
6,704 47.36
1,428
6,082 50.64
3,391
232
1 1 ,800
3,792
7,400
9,170
4,800
5,000
Labor, Total,
hr/diffuser $/diffuser
44.87
68.18
201.85
28.44
0.33 53.75
59.20
0.125 36.21
0.96 ' 89.00
34.75
75.00
-45
0.75 ~53
62.98
31.85
0.5 ~303
0.5 -253
0^5 -294
0.5 -311


61.92

73.60
92.80
50.00
35.60
46,22
46.67
58.33
60,00
Notes
C,R,a
C.R.a
C,R,b
C.N.c
P.R.d
C.R.e
P,R,f
P.R.9
R,h
C,R,i
C,R,j
P.R.k
C.R.I
C,R + N,m
P,R,n
n
n
n
o
P
q
r


C,R,s
C,n,t
C.R.U
C,R,v
w
X
Ref.
3-5
3-5
6-8
9,10
11,12
1,13
1,13
1,13
14
14
15,16
16,17
1,13
2,18
19
19
19
1.9
20
20
20
20
20
'20
. 21
22,23
22,23
22,23
22,23
22,23
                                                         197

-------
                                               Tola! cost includes  bond, mobilization, concrete  ($18,400),  demolition ($10,000),
TaWa 708 Notes:
 C Installation by contractor.
 P Installation by plant personnel.
 N New installation.
 R Retrofit installation.
 a Basin dea/illing/cleaning by plant  personnel.
    equipment, labor, and miscellaneous,
 b Payment lor Installation made annually based on energy savings for fiat year. Instated cost is sum of all payments to be made until a
    tola) of $201.BSftfiffuser is reached.
 0 Cost is  17.1 percent of tola! cost of blowers, blower building, automated controls, above-water  air  piping, air filtration system, and
    diffuser installation.
 d Does not include gas cleaning system cost. Includes 6 new airflow isolation/control valves. Total  cost based on plant superintendent
    estimate of average labor cost.
 o Cost includes basin cleaning, old equipment removal, and new air filtration system.
 (  Tube diKuscfs retrofitted directly lo existing swing arms; labor does not include basin cleaning and old equipment removal.
 0 Does not include labor for basin cleaning and old equipment removal; capital cost includes dilfusers at'$30.00 each plus piping and pipe
    supports.
 I) Not including basin cloaning costs.
 I  Alt now air piping from outside blower building to diffusers; cost includes basin preparation.
 j  Includes removal of coarse bubble system, installation of new drop pipes arid diffuser system, and installation of'"biocell" air filters in
    inlot Mower plenums. Air filtration cost was about $25,000. Component costs  estimated from lump sum bid.....
 k Retrofit  to  existiny reconditioned  swing arm assemblies; $30.00/diffuser unassembled; $7.80/diffuser miscellaneous  pipe
    fittinfls/hardware; $l5,00/difluser installation.
 I  Total cost including air filtration estimated at $446,500; cost of $62.98/diffuser is average  for both ceramic diftuser types. Total price
    includes 350 spare Sanitaire diftusers, 876 spare Norton diffusers, and installation of plugged baseplates on the laterals lo accept the
    spare diffusers, il needed.
 in Does not include costs lor basin cleaning, old equipment removal, and installation of stainless steel leveling rods tor new system.
 n 4 sq It coramie plate diffuser; capital cost is bid price; installed cost estimated by vendor and includes concrete in diffuser base and
    betoW'Waler piping.
 0 Materials cost tof dtllusers, piping to headers, supports, etc., delivered lo site was $30.95;  gas cleaning system cost including license
    loo was $l3,32/dtfluser.                    .                                                      .     ;
 p Materials cost includes diffusers, piping to headers, supports, etc., delivered to site plus gas cleaning system including license fee.
 q Installed cosl for dilfusers, piping and supports was $39.20/diffuser; $22.72/diffuser was for gas cleaning system.
 r  Materials cosl lor diffusers, piping lo headers, supports, etc., delivered to site was $34,2/diffuser;  gas cleaning system cost including
    teense foe was Stt3.44/diffuser.
 s Cost represents two separate contracts, one for $550,000 in 1985 ami another for $40,000 in 1986.            :
 I  Cosl basod on October 1985 bid data. Addition of new stainless steel header piping increased installed cosl to $87.02/diffuser.
 u Cost based on January 1988 bid data.
 v Cost basod on  contractor bid. Addition  of new stainless  steel header piping plus 500 spare  domes increased "installed cosl  to
    $86«6S/diffuser.
 w Cosl basod on contractor breakout in March 1985.
 x Cosl basod on contractor 1988 bid data, installation planned in 1990.
lo 5 hr/bastn (0.007 hr/diffuser) as the operators have
gained more experience with this  operation. Power for
the kiln firing averages $0.02/stone.

7.6.7 Power Charges
Electric  energy charges are highly  specific  to the
utility  providing the  service.  The following  types  of
monthly charges may be contained in a  utilities rate
structure:

*  Flat customer charge

*  Demand  charge  ($ per  maximum  kW demand  in
   past)

*  Energy charge ($ per kWh used in current month)

»  Power factor clause

These charges may also vary with time of day, day of
week,   or  season  of  the  year.  An  example  rate
schedule for the Wisconsin  Public Service Corporation
is  shown in Table 7-23.
                                                            Rate  schedules  of this  type  make  it  difficult  to
                                                            estimate aeration  energy costs  accurately.  Use of a
                                                            flat average $/kWh  energy price in the  analysis may
                                                            produce misleading  results unless  it  represents  a
                                                            composite number that somehow reflects  the relative
                                                            contributions of all demand charges and  peak/off-peak
                                                            pricing.

                                                            There  can  be a significant cost disadvantage to the
                                                            use of large blowers with  substantial excess capacity
                                                            to  satisfy  an  occasional  peak  demand.  Consider a
                                                            hypothetical situation using the rate'schedule in Table
                                                            7-23 in  which  a  750-kW (1,000-hp)  blower  motor
                                                            drawing  70 percent of its  rated load  is  turned on  to
                                                            satisfy peak load requirements during a period of peak
                                                            plant demand.  The incremental cost to operate this
                                                            blower for various  time periods  is  shown  in  Table  7-
                                                            24.  The  cost  of  an  occasional  blower  start-up  is
                                                            evident.  Even though the  on-peak  energy charge  is
                                                            only  $0.0369/kWh,  the  actual  incremental  cost  to
                                                            operate  this blower can be several dollars per kWh
                                                            once the demand  charges are  included.  Use of the
                                                            appropriate rate  schedule  is  an essential part of a
                                                            cost-effectiveness analysis.
                                                          198

-------
Figure 7-6.   Blower costs as a function of capacity.








03
03
O) _

5
"o
cd
13 ~
ts
8













in6










10s










•in4
5






















/
r
DO






















'
i






















/























.
























/



K

















/
/°
/


o <

}3













y
/^
y'
/



,
O

y













/







<
3
^
^












x







©

.
X













/








J(
^













/









/













^









/














^
f








'













,1

|







/





1





/
jL
/m
l





/
X. f
/
/
'





D4


X^
f
f


D


M

/
r












/




O""' 	




/













/






4


/





















(

/
















€






/
























^
































































f
105
                                                             Capacity, scfm

               X  Uriinstalled Individual Blower Costs (24)
               (•)  Uriinstalled individual Blower Costs (20)
               [•]  Unmstalled Total Capacity plus Piping (22)
               M.  Installed Individual Blower Cost - Madison, Wl (8)
               F  Uninstaller Individual Blower Cost - Frankenmulh, Ml (10)
               ^f  installed Individual Biower plus Filter Cost - Monroe, Wl  (18)
                                                               199

-------
Table 7-19. Costs Associated with Gas Cleaning
Plant
Albitqitorqiia, NM
SoytiKXir, Wl
Plymouth, WI-*
RipoiJ, Wt»(
Sluryoon Bay, Wl"
MiHur/Fiilton, NY
Frantonmuth, WI
Gicon Bay, WI
Wliiliior Narrows. CA
WlHlliut Narrows, CA
No. Did.
25,000
1,240
1,350
1,092
1,374
11,250
2,400
8,276
1,000
500
" Cluotmig on contract basis using 60-lb HCL cylinders
b $0,10 if cylinder shipping is included.
HCL Usage Per Diffuser Per
Cleaning Labor Per Cleaning
Ib
0.24
0.10
0.13
0.22
0.22
0.05
0.02
0.10
0.05
0.05
al $140 each.
Table 7*20. Estimated Time Requirements for Basin and
Dfffusor Cleaning
Time Required
Operation iabor-hr/lank
Basin Dcwatcnng
Basin Cleaning
Low-Prcssuro Hosing
High-Pressure Hosing
Acid Washing
Dome Removal and Replacement

Tablo 7*21. Diff user Cleaning and
RIdgewood, NJ
No.
Typo of Cleanings
Year Cleaning per Year
4/83-12/83 None
1S84 Hose 2
198S Hoso 8
AciU 3
1986 Husu 3
Acid 1
4 lo6

Repair Costs •
Cleaning"
Cost,
$/yr
0
350
1,400
1,125
525
375
labor-
hr/diffuser
0.02
0.02
0.025
0.06
0.085


Repair Cost,
$?yr
0
700"
350

0

$ hr/1,000 diff. hr/lb
0.24 6.3
' 0.21 3.2
0.31 2,2
0.51 9.2
0,51 5.8
0.05 3.6
0.03 1.0
0.07 4.6
0.06^ 6.0
0.06b 10.0

HCL del. Ref.
0.026 26
0.033 26
0.017 26
0.042 26 ,
0.027 26
0.067 26 "
0.045 12 f
0.034 3
0.12 1
0.20 " 1

Table 7-22. Milwaukee Method Diffuser Cleaning Costs -
Hartford, CT
Labor for hosing, acid cleaning, and
leak repair"
Cleaning chemical
Spare parts (gaskets, bolls, supports,
domes)*"
Equipment and protctive clothing
0.03 labor-hr/diffuser
$0.075/diffuser
$0.45/diffuser
$0.22/diffuser
"Leak repair required 0.01 labor-hr/diffuser.
*" 5-10 percent of the diffusers required leak repair in the first two
cleanings. At the 5-yr cleaning point {October 1988), repair of
bolts and gaskets was required for 35-40 percent of the diffusers.














Average labor cost is $21.80/hr.
Includes installation ol 220 new diffusers on existing blanks.
                                                             200

-------
Table 7-23. Wisconsin  Public  Service  Corporation  Rate
           Schedule
 AVAILABILITY

 The schedule is applicable lo customers whose monthly demand is
 > 1,000 kW.

 MONTHLY RATE

 Customer Charge: $300.00/month

 Demand Charge

 1.  Customer Demand:  $1.00/kW per kW o( maximum demand
                      during the current or preceding 11
                      months, plus:

 2.  System Demand:
    Peak Load         $6.32/kW
    Intermediate Load   $4.55/kW
    Base Load         $0.00
 3.  Energy Charge:
    On Peak .
    Off Peak

 MINIMUM CHARGE
$0.0369/kWh
$0.0261/kWh
 The minimum monthly charge is the customer charge plus ihe
 demand charges. The sum of the system demand charges for the
 annual period, including the current and the succeeding 11
 months, shall not be less than $4i/kW of maximum peak load
 demand during the current month.
 Table 7-24. Cost of Operating a 750-kW (1000-hp) Motor at 70 percent of Rated Load for Various Periods Over a Year (Demand
            = 522 kW)

                                                     Annual Cost, $
Period of on- Peak
Use
1 hr/yi*
1 hr/monttf
6 hr/monttf
100 hr/montrr
300 hr/month™
rower
Consumed,
kWh/yr .
522
6,264
6,264
626,400
1,879,200
Demand
Customer
6,264
6,264
6,264
6,264
6,264
Charge
System
3,299
39,588
39,588
39,588
39,588

Energy Charge
19
231
1,387
23,114
69,342

$/yr
9,582
46,083
47,239
68,966
115,194
Total Cost
$/kWh
18.36
7.36
1.26
0.11
0.06
    On-peak use occurs at peak plant demand.
    On-peak use occurs at peak plant demand at least once per month.
                                                         201

-------
7.7 References

 1. Personal communication from M.K.  Stenstrom,
   University  of  California at Los  Angeles, Los
   Angeles,  CA,  to  J.A.  Heidman,  U.S.
   Environmental  Protection Agency,  Cincinnati, OH,
   December 17,  1987.

 2. Personal communication from  D.T. Redmon,
   Ewing  Engineering Co.,  Milwaukee, Wl, to J.A.
   Heidman, U.S. Environmental Protection  Agency,
   Cincinnati, OH, June 12, 1987.

 3, Personal communication from M.J. Pierner, Green
   Bay Metropolitan  Sewerage District, Green  Bay,
   Wl,  to J.A.  Heidman,  U.S.  Environmental
   Protection Agency, Cincinnati, OH, April 15, 1987.

 4. Donohue  & Assoc.,  Inc.  Fine  Pore  Diffuser
   System Evaluation for the Green Bay Metropolitan
   Sewerage  District.  Study  conducted under
   Cooperative  Agreement CR812167,  Risk
   Reduction  Engineering  Laboratory,  U.S.
   Environmental  Protection Agency, Cincinnati, OH
   (to be published).

 5. Personal communication from J.J.  Marx, Donohue
   &  Assoc., Sheboygan, Wl, to J.A. Heidman, U.S.
   Environmental  Protection Agency,  Cincinnati, OH,
   January 1988.

 6. Mueller, J.A. Case History of Fine Pore Diffuser
   Retrofit at Ridgewood, NJ. Study conducted  under
   Cooperative  Agreement CR812167,  Risk
   Reduction  Engineering  Laboratory,  U.S.
   Environmental  Protection Agency, Cincinnati, OH
   (to be published).

 7. Personal  communication from  P.O.  Saurer,
   Manhattan College, Bronx,  NY, to J.A. Heidman,
   U.S. Environmental Protection Agency, Cincinnati,
   OH, May 1987.

 8. Personal communications from  J.A.   Mueller,
   Manhattan College, Bronx,  NY, to J.A. Heidman,
   U.S. Environmental Protection Agency, Cincinnati,
   OH, September 1987 and December 1988.

 9. Boyle, W.C.  Oxygen Transfer  Studies at the
   Madison Metropolitan Sewerage District Facilities.
   Study  conducted  under  Cooperative Agreement
   CR812167,   Risk   Reduction  Engineering
   Laboratory, U.S. Environmental  Protection
   Agency, Cincinnati, OH (to be published).

10, Personal communication from  S.R. Reusser,
 •  Madison Metropolitan  Sewerage District, Madison,
   Wl,  to J.A.  Heidman,  U.S.  Environmental
   Protection Agency, Cincinnati, OH, May 1987.
11. McNamee, Porter & Seeley Engineers/Architects.
   Fine Pore Diffuser Case History for Frankenmuth,
   Ml.  Study  conducted  under  Cooperative
   Agreement  CR812167,  Risk  Reduction
   Engineering  Laboratory, U.S.  Environmental
   Protection  Agency,  Cincinnati,  OH  (to  be
   published).                                 ,

12. Personal communication  from T.A.  Allbaugh,
   McNamee, Porter & Seeley, Ann Arbor, Ml, to J.A.
   Heidman, U.S. Environmental Protection Agency,
   Cincinnati, OH, October 21, 1987.               ;

13. Stenstrom, M.K. Fine  Pore Diffuser Fouling: The
   Los  Angeles  Studies. Study  conducted under
   Cooperative  Agreement  CR812167,  Risk
   Reduction  Engineering Laboratory,  U.S.
   Environmental Protection  Agency,  Cincinnati, OH
   (to be published).

14. Personal communication  from E.L.  Barnhart,
   Hydroscience,  Inc., Harbor Island, SC,  to -J.A.
   Heidman, U.S. Environmental Protection Agency,
   Cincinnati, OH, November 17, 1988.

15. Aeration  Technologies,  Inc.  Off-Gas Analysis
   Results and  Fine Pore Retrofit Case History for
   Hartford, CT MDC Facility.  Study conducted under
   Cooperative  Agreement  CR812167,  Risk
   Reduction  Engineering Laboratory,  U.S.
   Environmental Protection  Agency,  Cincinnati, OH
   (to be published).                    ,

16. Personal communication from   R.G. Gilbert,
   Aeration Technologies, Inc., North Andover,  MA,
   to J.A. Heidman, U.S. Environmental  Protection
   Agency, Cincinnati, OH, December  27, 1988

17. Aeration  Technologies,  Inc.  Off-Gas Analysis
   Results  and  Fine  Pore Retrofit Information  for
   Glastonbury,  CT Facility,  Aeration Tank  No. 2.
   Study conducted under Cooperative  Agreement
   CR812167,  Risk  Reduction   Engineering
   Laboratory,  U.S.  Environmental  Protection
   Agency, Cincinnati, OH (to be published).

18. Ewing Engineering Co. The Effect of Permeability
   On  Oxygen  Transfer  Capabilities,  Fouling
   Tendencies, and Cleaning  Amenability at Monroe,
   Wl.  Study  conducted  under  Cooperative
   Agreement  CR812167,  Risk  Reduction
   Engineering  Laboratory, U.S.  Environmental
   Protection  Agency,  Cincinnati,  OH  (to  be
   published).

19. Personal communication from  C.E.  Tharp,
   Environmental Dynamics, Inc.,  Columbia,  MO, to
   J.A.  Heidman,  U.S.  Environmental  Protection
   Agency, Cincinnati, OH, January 27, 1989.
                                              202

-------
20. Personal communications from G.G. Powell, Gore
    & Storrie  Ltd., Toronto, Ontario,  Canada,  to J.A.
    Heidman,  U.S.  Environmental Protection Agency,
 -,  Cincinnati, OH, February 6, 1989 and  March 1,
    1989.

21. Personal communication from R. Hall, City of San
    Mateo,  CA, to J.A.  Heidman, U.S. Environmental
    Protection  Agency, Cincinnati,  OH,  March  3,
    1989.

22. Personal  communication from H.H.  Benjes,  Jr.,
    HDR Engineering, Inc., Dallas, TX, to W.C. Boyle,
 ,   University of Wisconsin, Madison, Wl, October 2,
 •   1988.

23. Personal  communications from H.H. Benjes,  Jr.,
    HDR Engineering,  Inc.,  Dallas,  TX, to  J.A.
    Heidman,  U.S.  Environmental Protection Agency,
    Cincinnati, OH, January-April 1989.

24. Harris,  R.W., M.J.  Cullinane, Jr. and  P.T. Sun.
    Process Design and Cost  Estimating Algorithms
    for the  Computer Assisted  Procedure for  Design
    and Evaluation of Wastewater Treatment Systems
    (CAPDET). NTIS  No.  PB-190455,  U.S.
    Environmental Protection  Agency, Washington,
    DC, January 1982.

25. Personal  communication  from  F.K.  Marotte,
    CH2M-Hill,  Denver,  CO, to R.C. Brenner,  U.S.
    Environmental Protection Agency, Cincinnati, OH,
    April 13, 1989.
26. Personal communication  from  J.D.  Wren,
    Sanitaire -  Water  Pollution  Control  Corp.,
    Milwaukee,  Wl, to W.C. Boyle, University  of
    Wisconsin, Madison, Wl, November 2, 1988.

27. Personal communication  from  J.D.  Wren,
    Sanitaire -  Water  Pollution  Control  Corp.,
    Milwaukee,  Wl,   to  J.A.  Heidman,  U.S.
    Environmental Protection Agency, Cincinnati, OH,
    January 20,  1989.

28. Barnhart E.L.  and  M.  Collins. The Measurement
    and Control  of  Fouling  in  Fine  Pore  Diffuser
    Systems.  Study conducted  under , Cooperative
    Agreement  CR812167,  Risk  Reduction
    Engineering Laboratory, U.S.  Environmental
    Protection  Agency,  Cincinnati,  OH  (to  be
    published).

29. Yunt, F.W. Some  Cleaning Techniques for Fine
    Bubble Dome and Disc Aeration Systems. Internal
    report, Los  Angeles County Sanitation Districts,
    Whittier, CA, October 1984.

30. Personal communication from D. Noyes,  City  of
    Bozeman, MT, to  J.D. Wren, Sanitaire  -  Water
    Pollution Control Corp., Milwaukee, Wl, May  19,
    1987.

31. Personal communication from D. Noyes,  City  of
    Bozeman,   MT,  to  J.A.  Heidman,  U.S.
    Environmental Protection Agency, Cincinnati, OH,
    January 27,  1989.
                                                203

-------

-------
                                              Chapters
                                            Case Histories
8.1  Introduction

Case histories provide essential information about the
progression of a  new technology that  can not be
totally incorporated in  the categorical approach used
in this manual. Case histories illustrate how engineers
and  wastewater treatment plant authorities apply such
technologies to solve specific problems.  Although
these  problems arise from site-specific situations,
valuable general guidance can often be gained from  a
study  of  such installations  by others contemplating
use  of the same technology. It is with this objective in
mind  that  the  following summaries  have  been
prepared.
Case histories have been developed for several fine
pore diffuser systems  evaluated  by  contractors
retained on the EPA/ASCE Rne Pore Aeration Project.
Substantial  in-process oxygen transfer  testing  was
conducted  in these  field  studies,  and  the data
generated are  summarized in the  respective case
histories  reported  herein. In  addition, to provide  a
broader perspective of fine pore diffuser applications,
advertisements  were placed in the  professional
newsletters of ASCE and WPCF to solicit information
on other potential  case  histories. Several  case
histories have been developed from these  sources
and  are also included here. For the most  part, no field
oxygen transfer data are available for these additional
case histories. Performance of these systems must be
judged  on overall  power consumption, operation and
maintenance  (0&M)  requirements, and  effluent
quality.

In general, each case  histpry is organized to provide
information on: 1)  the  reason why fine pore diffusers
were selected, 2)  the type  of plant and wastewater
involved,  3) the fine pore diffusion  system selected
and  some detail on its installation, 4} how the system
performed during the 12 months or more of operation
following rts installation, and 5) the benefits derived as
a  result of employing the new system. Art effort has
been  made to provide summaries that  are as
complete  as possible to direct potential users of this
technology to situations that most closely resemble
their own.

As might be expected,  many of the case histories
involved retrofitting plants previously equipped  with
coarse  bubble  diffused aeration systems. These
retrofit examples generally reflect the economic power
savings  possible with  fine pore  aeration.  In  a few
cases,  however,  predicted  economics  were not
realized in actual operation. In several other cases,
problems were encountered that were not envisioned
beforehand. Conversely, diffuser fouling  problems
anticipated at several plants never materialized.

When judging the success or failure of a particular
retrofit, readers are cautioned against making a strict
economic judgement for the following reasons.  First,
many of the  fine pore  systems described in this
chapter are just beginning their useful life and full life-
cycle costs are not available. Second,  it is not always
equitable to  directly compare operating  data  from
before a retrofit to after because of a variety of other
changes that may have occurred  during either  period
for which  documentation  is  either  unavailable or
incomplete. Obviously,  such changes could bias the
comparison. Third,  plants  retrofitted with fine pore
aeration  often achieve nitrification  as  an added
benefit  Finally,  retrofitted systems  are  frequently
operated at different dissolved oxygen (DO)  levels
than prior  to the  retrofit- Where  possible,  these
different before and  after conditions have  been
reported but have not  necessarily been factored into
cost differentials.

Case histories, as a whole, permit the formation of
generic  conclusions  regarding the claims for a new
technology and the likelihood that  a real  advancement
has  been achieved.  Two such conclusions may be
drawn  from  the case histories presented herein
concerning the increasing trend toward employing fine
pore diffusers in treating municipal wastewater.  First,
the projected greater oxygen transfer rates for simifar
energy inputs  have generally been realized,  although
complete life-cycle costs are not yet available and, in
some cases,  cost  savings are  marginal.  Second,
although diffuser fouling has not always  accompanied
the changeover to fine pore diffusers, many plants are
periodically cleaning their diffusers by  a variety of
methods in the  belief  that such  efforts will prevent
fouling from developing.

The case  histories that follow offer the  reader  a
number of diverse situations. They should be useful to
                                                  205

-------
Table 8-1.
 Facility
Summary of Case Histories
                                     Design Flow, tngd
Fine Pore Aeration System
 Performance Evaluation by Off-Gas Testing
  Frankeronuth, MI
  Glaslontouiy. CT
  Groon Bay, Wl
  Hartford, CT
  Jones Island, Wl
  Madison, W!
  Ridflfiwood, NJ
  Whitta Narrows, CA
 Performance Evaluation by Means Other Than
 Off-Gas Testing
  Cleveland, Wl
  Plymouth, Wl
  Ronton, WA
  Ripon, Wl
  Saukvillo, Wl
                                           1,8
                                           3.6
                                          52.5
                                          60.0
                                         200.0
                                          38.0
                                           3.0
                                          15,0
                                           0.2
                                           1.3
                                          72.0
                                           2~0
                                           0.5
Ceramic Discs
Rigid Porous Plastic Tubes
Ceramic Discs and Perforated Membrane Tubes
Ceramic Domes
Ceramic Plates
Ceramic Discs and Ceramic Domes
Ceramic Domes
Ceramic Discs and Ceramic Domes
Rigid Porous Plastic Plates
Ceramic Discs
Perforated Membrane Tubes
Ceramic Discs
Ceramic Discs
those  considering  fine pore difiusers  for  new  or
existing  plants.  Table  8-1  lists  the plants that  are
described in  this chapter.  The plants are divided into
two groups depending on how the plant's performance
was evaluated. Section  8.2 includes plants that were
evaluated by off-gas testing data; Section 8.3 includes
plants that had to be evaluated  by means  other than
off-gas  testing,  including  treatment  performance,
energy used, and quantity of air supplied per unit of
BOOg removed. Section 8.4 summarizes the various
sources of information used to assemble these case
histories.
                                                      206

-------
                       8.2 Performance Evaluation by Off-Gas Testing
   8.2.1 Frankenmuth Wastewater Treatment Plant

   LOCATION: Frankenmuth, Michigan

   OPERATING AGENCY: City of Frankenmuth

   DESIGN FLOW: 79 L/s {1.8 mgd)

   WASTEWATER: Domestic Plus Brewery

   ORIGINAL AERATION SYSTEM: Broad Band Coarse Bubble Tube Diffusers

   FINE PORE AERATION SYSTEM:  Sanitaire Ceramic Disc Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION: 1986

   BASIS OF PERFORMANCE EVALUATION: Off-Gas Testing
                                           Overall Process Performance

   CLEANING METHOD: In-Situ HCI Gas Cleaning
I. INTRODUCTION
Frankenmuth is  a small  community  of  about  4,000
people in  central Michigan. It is  also the home of a
brewery that produces beer and  other products. The
brewery wastewater accounts for  about 25-30 percent
of the flow and 50-70 percent of  the BOD load  to the
wastewater treatment facility.

In January  1986, conversion from a  stainless  steel
broad band coarse bubble diffuser system to fine pore
aeration was  completed  in  all  of the  six  existing
aeration tanks. Two smaller centrifugal  blowers with
inlet filters capable of removing particles  >2Q microns
in size were installed as well  as an  in-situ HCI gas
cleaning system. Plant energy consumption per unit of
BOD  removed was  measured and recorded before
and after  the  fine pore aeration  conversion. Oxygen
transfer efficiencies  (OTEs) were  measured by off-gas
testing on  13  different days between April 1987 and
May 1988. Some of the  off-gas  tests were done  on
successive days before and after gas cleaning  of the
diffusers to observe the effect of cleaning on OTE.

In addition, an  attempt  was made  to  monitor the
condition of fine  pore aeration diffusers by measuring
dynamic  wet  pressure (DWP)  and  pressure  drop
across air distribution  orifices  in test diffuser
assemblies. Four diffusers were  placed in one  of the
six  activated sludge aeration cells, and measurements
were  generally obtained  at 1-2  week intervals after
equipment installation.

Summary  details are presented below on the original
aeration system,  the basis  for changing to fine pore
aeration, and  the design,  performance, and evaluation
of the new system.
II. ORIGINAL TREATMENT FACILITY
The process flow scheme for the plant consists of  a
manual  bar screen,  raw wastewater pumping,  a
square aerated grit tank,  two  rectangular  primary
clarifiers, aeration tanks, two 15-m (50-ft)  diameter
final  clarifiers,  and  a  chlorine  disinfection system.
Primary  and waste  activated sludge  are  combined
prior to  anaerobic digestion.  The digested sludge is
dewatered  with  vacuum filtration  and hauled  for
disposal off site. In-plant recycle  streams are returned
to the raw wastewater pumping station.

The  aeration  tank  layout,  shown in Figure  8-1,
consists of six individual aeration  cells. Each cell is
13.4 m by 6.7 m (44 ft by 22 ft) with a 4.6-m (15-ft)
sidewater depth  (SWD), resulting  in a total aeration
volume  of  2,465 m3  (651,000  gal). The hydraulic
design for  the aeration cells  allows  a  variety  of
process flow  regimes  ranging from  modified contact
stabilization to conventional plug  flow activated sludge
by  operating  the  six  cells in   series.  Contact
stabilization can be varied by  using 2-4  of the six cells
in series for return  sludge aeration followed  by  the
remainder of the cells (2-4)  in series for  wastewater
aeration,  with primary  effluent added to the first  cell
after  return  sludge  aeration.  One  modified  contact
stabilization operating mode has  been to operate with
Cells  1-3 for return sludge aeration, followed by Ceils
4-6 in series  for wastewater aeration with  primary
effluent added to  Cell  4. Normal operation,  however,
is to  use Cells 1 and  2 (and possibly 3) for return
sludge aeration and Cells 4 and 3, and 5 and 6 as two
parallel, two-tanks in series activated sludge trains.

The  air diffusion equipment initially  installed  was
stainless steel broad  band  coarse bubble diffusers
                                                 207

-------
Figure 8-1.  Aeration tank arrangement - Frankentnuth, Ml.
 aiM
 L
1

I-




U-
-J-ixl-
4
l-|xj_



. 	 tx}_
5

•»
Ifc



•H
•»





C
Aeration
[ No. i
I
[ No. 3
r
1
I'

' No. 5
Tank
No. 2

NO. 4 ;
t
1
*
No. 6
i


T
I


1
                    PE - Primary Effluent
                    RS - Return Sludge
                    SN - Supernatant
                    ML - Mixed Liquor
mounted  on galvanized steel headers.  Four multi-
stage blowers  were  provided, each with a  nominal
capacity of 3,070 Us (6,500 scfm).

The plant's NPDES  permit  requires  that  30-day
average effluent  BODs and  SS  concentrations each
be  less than 30  mg/L.  There is  no  effluent ammonia
nitrogen discharge requirement.

III.  FINE  PORE AERATION MODIFICATION
EVALUATION

Basis for Change
The retrofit to  fine pore aeration was not predicated
solely  on  the desire  to  reduce  operating costs
because of higher OTEs  associated  with  fine pore
aeration. Rather,  it was  motivated  by  the  need  to
replace a  significant  number  of coarse  bubble
diffusers that had broken off due  to corrosion,  and the
inability of the  system  to always supply the required
oxygen  demand.

A pre-design analysis indicated that  the initial  cost for
the retrofit  to  fine pore aeration was approximately
$160,000, the same as replacing the existing system
with new  coarse bubble aeration  equipment.  Initial
cost estimates for the fine  pore  aeration  system
included new diffusers and diffuser grid piping, new air
drop pipes in each cell, in-situ gas cleaning, two new
small blowers, and new air inlet filters. Initial costs  for
the coarse bubble system included new diffusers and
in-tank piping and new stainless steel drop pipes. The
estimates were based on  using 2,400  ceramic disc
diffusers and 2,000  stainless steel broad band coarse
bubble units.

Before initiating the  final design, a test header with
four ceramic disc  diffusers was installed  near the inlet
end of Cell 5 to monitor the potential for plugging and
fouling with this  relatively high-strength wastewater
application. The DWP and orifice pressure drop were
monitored daily for  about 10 weeks to  ascertain the
possible  level of plugging and fouling. At the end of
that period, the test header diffusers were cleaned  by
injecting  HCI gas into the air supply to the header.

Initial DWP readings were approximately 15 cm (6  in)
w.g. Within the first  week,  the DWP increased to  38
cm (15 in) w.g. During the fifth  week, the air supply
was increased in  an  attempt  to  "air bump"  the
diffusers to reduce fouling.  The DWP decreased from
about 43 cm (17 in) w.g. to  30 cm (12 in) w.g. the
following week, but this was within the range of other
DWP fluctuations  so no conclusions could be made
with regard to the benefits of air bumping. After gas
cleaning, the DWP decreased from about 35 cm (14
                                                  208

-------
in) w.g. to 24 cm (9.5 in) w.g., the lowest observed
level  since the first week of the  test,  so it  was
concluded that  fouling  and  plugging  of fine pure
diffusers could be controlled in this application by the
use of gas cleaning. Thus, the final design of the fine
pore aeration system was continued with plans to use
gas cleaning.

Existing System Evaluation
The anticipated range of airflow rates for the  fine pore
aeration system was 378-1,888 L/s  (800-4,000) scfm
to satisfy  minimum mixing requirements and  peak
oxygen demands.  The estimated average airflow rate
was  approximately 1,038 L/s  (2,200 scfm). This was
near the minimum operating level or surge point of the
existing blowers.  An evaluation  of  blower operating
efficiencies at  this  airflow  rate from  the blower
performance curve found that blower efficiency would
be very low, especially compared to what is possible
using smaller size  blowers.

To  ensure  that anticipated energy savings  for fine
pore aeration could be achieved and to  operate above
the blower surge point, new blowers  were selected for
the retrofit system. Two multi-stage 149-kW (200-hp)
centrifugal blowers with  a nominal capacity  of 1,038
L/s (2,200 scfm) each were selected. Only the
blowers were replaced. The  existing   blower  bases,
motor  starters, .valves, flexible connectors, and piping
were  used with the  new  equipment. Two of  the
existing 187-kW (250-hp) blowers were  left in place to
provide standby capacity.

The inside lining of the  existing air  piping  was
determined to be in excellent  condition; the only new
piping  required was for the air header drop pipes and
the diffuser grid  system.  Due to the age of the existing
piping  system, the design engineers recommended in-
line air filters be placed  immediately upstream of the
air header drop pipes to each cell.  Specifications for
these filters required 97 percent removal of particles
>0.3 micron in  size to protect the diffuser equipment
from air side fouling. After finding the  existing piping
was in good condition, the  City elected not  to install
the in-line filters.  New  elements  for existing  blower
inlet air  filters  capable  of  removing  particles >20
microns in size were installed  instead.

Cost-Effectiveness Evaluation
Table  8-2  summarizes  the  assumptions  and
calculations for  the  pre-design aeration  energy
comparison between fine pore  and coarse bubble
aeration alternatives. The estimated O&M cost for the
fine  pore  system was  based on periodic  HCI gas
cleaning   of the  fine  pore diffusers.  The  average
annual aeration operating cost for fine pore aeration
was estimated  to  be about  one-half   of the coarse
bubble aeration cost. Based on this evaluation, even  if
the coarse bubble  system  didn't  need replacing,
replacement could  be  considered  with  the annual
energy savings providing a simple  replacement
payback  within  6 years  based  on  an  initial  cost
estimate of $160,000.

Table 8-2.    Comparison  of Fine Pore  and Coarse Bubble
           Aeration  Energy  Requirements:  Pre-design
           Estimates - Frankenmuth, Ml

                                          Stainless
                                           Steel
                                Ceramic    Coarse
 Parameter                        Fine Pore    Bubble
Average Primary Effluent Flow, mgd1
Average Primary Effluent BODg1, mg/L
lb,d .•
Required Oxygen Transfer Rate
(AOR)2, Ib/d
Assumed Actual OTE3, percent
Average Airflow Required3, scfm
Average Horsepower, hp
Estimated Annual Energy Cost4, $ .
Estimated Annual O&M Cost4, $
Total Estimated Annual Cost, $
• Estimated Annual Savings, $
1.4
512
5,977 .
6,575
11
2,214
88
26,140
3,000
29,140
29,675
1.4
512
5,977
6,575
7.2
3,641
180
58,815

58,815

 1 Flows and loads based on averages for November 1983-October
  1984.
 2 Assumes 1.1 Ib O2 required/lb BOD applied.
 3 At average mixed liquor DO of 2.0 mg/L.
 4 Assumes power cost of $0.045/kWh.
IV.  FINE PORE AERATION  RETROFIT  DESIGN
DESCRIPTION
The  new aeration  system consisted of ceramic disc
diffusers installed in  a full-floor, coverage grid pattern
and in-situ HCI gas cleaning. Installation was begun in
December  1985  by the  treatment  plant  staff with
technical  assistance  provided by  the  equipment
manufacturer. The new equipment was fully in service
by January  1986.  The  total  equipment  cost was
$160,000, and the plant staff invested  approximately
800  labor-nr in  installation  and start-up.  The total
project  cost, including  equipment,  installation, and
engineering, was approximately $190,000.

A  total  of  2,400 diffusers were installed with 400
diffusers evenly  spaced  in each of the six  aeration
cells. The diffuser density was thus  one diffuser/0.22
m2 (2.42  sq ft)  of tank floor  area. The anticipated
maximum airflow rate at peak load was  approximately
0.8 L/s  (1.7 scfm)/diffuser. The minimum  airflow rate
was estimated to be 0.3;L/s (0.7 scfm)/diffuser. Airflow
rates to each cell are controlled by manually, operated
butterfly valves.

V.  OPERATIONAL  PERFORM ANCE  AN D
EVALUATION

Treatment Performance             7 .
The  wastewater treatment facility consistently met its
NPDES permit  for effluent BOD after the retrofit  in
spite of the  high  primary  effluent BOD  concentration
                                                   209

-------
Tabto 8-3.   1986 Performance Summary - Frankcnmuth, Ml.

                                Primary Efllueni
                               Final Effluent
Oaio
January 1986
February
March
April
May
Juno
July
August
September
October
Novombor
December
Flow, _
mfld
1.46
1,20
1.77
1.47
1.35
1.52
1.39
1.27
1.90
1.72
1.20
1.39
BODS
mQ/L
539
729
587
838
866
621
769
723
474
490
873
520
Ib/d
6,720
7.545
8,129
10,605
10,172
7,947
9,111
7,959
6,724
6,830
7,185
6,655
SS
mg/L
283
430
288
378
479
231
422
248
220
232
289
222
Ib/d
3,553
4,459
4,267
4,788
5,812
2,836
5,034
2,658
3,523
3,474
2,931
2,642
Volumetric Load,
lbBOD5/d/l,OOOcu ft
83.3
79.9
86.1
112.3
107.8
84.2
96.5
84.3
71.2
72.4
76.1
71.6
BOD6
mg/L
15 ,
22
29
51
29
24
22
17
16
9
14
29
Ib/d
183
233
452
658
339
316
262
174
220
125
137
349
SS
mg/L
23
21
23
32
24
18
12
20
27
22
32
24
Ib/d
293
220
354
395
276
239
1,411
207
405
314
310
278
due to the brewery wastewater contribution. Table 8-3
summarizes monthly  average  plant  loadings and
treatment performance for  1986. The mean  solids
retention  time  (SRT)  was typically  5-10 days. The
hydraulic  retention time (HRT) was 10-12 hours, and
typical aeration tank  mixed liquor suspended  solids
(MLSS) levels were in the range of 3500-7000 mg/L.

Aeration System Oxygen Transfer Efficiency
Off-gas testing was performed on selected  aeration
cells  on  13 different  days  from April 1987  to May
1988. The purpose was to measure OTEs of the fine
pore  aeration system  in  the  various operating cells
and to observe the effect of diffuser gas cleaning  on
diffuser performance.  The  standard  off-gas testing
procedure was followed using  a 61-cm by 305-em  (2-
ft  by 10-ft)  fiberglass  off-gas collection  hood. The
hood  was typically located at a total of four locations
near the center of the aeration cell. There were two
locations  on each side of the transverse center line.
Those two  locations  were end-to-end  to  provide
almost complete coverage of the aeration  cell width.

During testing  in  May  1988,  two additional  pairs of
hood  locations  were used to sample near the  aeration
cell inlet and outlet. The measured OTEs many times
varied by  20-30 percent  for  the different  hood
locations in  a given aeration cell. This may have been
due to  cell  mixing characteristics. The  OTE values
were  averaged to allow a comparison  of aeration
performance between cells and for different test  days.

Two  operating modes  were  used  during  off-gas
testing. The most common mode was the use of Cells
1-3 for return sludge aeration  followed by  primary
effluent addition to Cell 4 and operation of Cells  4-6 in
series. During  off-gas  testing  in April  1987 and May
1988,  Cells  1  and 2 were used for return  sludge
aeration, followed by  two parallel  trains with primary
effluent addition to Cells 4 and 5. Mixed liquor then
flowed from Cell 4 to Cell 3 and from Cell 5 to Cell 6.

Table  8-4 summarizes  the  average off-gas  testing
results from June  1987 through March  1988 for the
series  operating  mode.  Cell  3  was  used for return
sludge aeration as  well as Cells 1 and 2, and  primary
effluent was added to  Cell 4. The airflow rate  in each
cell averaged  about 0.9  Us (2.0 scfm)/diffuser.  The
results illustrate the effect of the degree of wastewater
stabilization  on the average aF(SOTE).  aF(SOTE)
values were  in the  7-8 percent range in Cells 4 and  5,
but following  substantial soluble  BOD  removal,
aF(SOTE) values averaged 12.1 and 13.2 percent  in
Cells  6 and 3, respectively.  The weighted average
aF(SOTE) was 10.0  percent  for  all four cells at an
average  airflow,  rate  of   about  0.9  L/s  (2.0
scfm)/diffuser.  Using  a  reasonable assumption  that
aF(SOTE) values  for  Cells  1  and 2  were  between
those  for Cells 3 and 6,  the  weighted average
aF(SOTE) for the system was estimated to be about
10.9 percent.  This compares to the value of 11.9
percent used  in  the  pre-design  analysis to evaluate
potential energy savings of fine pore aeration.

Table 8-4.   Off-Gas Test Results - Frankenmuth, Ml
                           Aeration Cell No.

Airflow, scfm/diffuser
Avg. aF{SOTE), %
aF(SOTE) Range, %
No. Test Days
3
1.9
13.2
10.0-16.3
5
4
2.0
7.1
6.8-7.4
2
5
1.9
7.8
7.0-9.6
8
6
2.1
12.1
10.3-16.2
9
Effect of Cleaning on Performance
HCI gas cleaning was initiated to control  DWP below
41-46 cm (16-18 in)  w.g.  This  occurred at various
                                                  210

-------
Table 8-5.   Estimated Energy Cost Savings - Frankenmuth, Ml
                                                                Year

Average BODH treated, Ib/mo
Average kWh.used/lb BODS removed
Average monthly power cost1, $
Average morithy savings, $
Annual savings, $
1 985 - Coarse Bubble
295,939 -
0.83
12,281 '
-
-
1986 - Fine Pore
295,939
0.59
8,730
3,551
42,612
1 987 - Fine Pore
295,939
0.68
10,062
2,219
26,628
1988 - Fine Pore
295,939
•0.88
13,021
-740
-8,880
 ' @'$0.05/kWh.
intervals  that  were typically  about  once per  month
during the first 2 years of operation. After September
1987, the amount of HCI gas  used during  cleaning
was  reduced and, in 1988,  cleaning was performed
only  3 out of the first 7 months. The actual gas used
averaged about 150  g (0.33 Ib  )/diffuser/yr compared
to an estimated  consumption  of  450  g (1.0
lb)/diffuser/yr in the pre-design  evaluation stage. The
cost of gas  cleaning, including labor and gas, ranged
from  approximately $2,400  during  the first year of
operation  to approximately  $4QQ/yr when  cleaning
frequency and gas dosage were decreased.

Off-gas testing was  performed  on days  immediately
before and  after gas  cleaning  on five  different
occasions between June  1987 and May 1988 in Cells
5 and 6. No increase in  OTE was found on the day
following gas  cleaning. This  may have been due to
variations in wastewater characteristics, poor mixing of
primary effluent with  recycle sludge in Cells  4  and 5,
and possible variations in  the test results.

Gas  cleaning was found to effectively  control the
D.WP of the diffusers.  The  DWP level of 14-15 cm
(5.5-6.0 in)  w.g. typically measured on new  diffusers
was  not always achieved,  but DWPs were usually
decreased  to 15-25  cm (6-10  in) after  cleaning.
Although no direct OTE  benefits could be  seen for
cleaning, long-term control of DWP may be helpful in
minimizing  operating  pressures  at  the blower
discharge, and energy requirements.
VI. ECONOMIC CONSIDERATIONS
Table 8-5 summarizes total energy consumed per unit
of BOD5 removed for operating periods before  1985
and after the retrofit.  Average monthly savings  were
calculated by comparing the average observed kWh/lb
BODs  removed ratio  to that  before  the  retrofit and
multiplying that value  by the  average monthly BODs
removal rate.  The average  monthly BODs  removal
rate for 1985  was used for the subsequent  year in
Table 8-5 to provide an equal  basis for comparing the
energy demands between fine pore and coarse bubble
aeration.  The average  annual  energy savings
calculated for  1986 and  1987  is $34,620, which is
above  the value  of $29,140  used  in the  pre-design
fine pore aeration economic  evaluation. The  energy
consumption ratio was much higher in 1988 because
plant personnel maintained high aeration rates  even
though the BOD loading to the  plant had  significantly
decreased. Thus,  the potential energy savings benefit
of fine  pore aeration  was not realized  during  this
period.                            •
                                                 211

-------
   8.2,2 Glastonbury Water Pollution Control Plant

   LOCATION: Glastonbury, Connecticut

   OPERATING AGENCY: Town of Glastonbury

   DESIGN FLOW: 158 Us (3.6 mgd)

   WASTEWATER: Domestic

   ORIGINAL AERATION SYSTEM: Delrin Coarse Bubble Diffusers

   FINE PORE AERATION SYSTEM: FMC Pearlcomb Rigid Porous Plastic Tube Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION: 1984

   BASIS OF PERFORMANCE EVALUATION: Off-Gas Testing

   CLEANING METHOD: Detergent Washing	
I. INTRODUCTION
A  secondary wastewater  treatment facility with an
average design flow of 158 Us (3.6 mgd} was placed
into operation at Glastonbury, CT in 1972. The original
aeralfon system consisted of coarse bubble spargers
on  swing-arm assemblies. As electricity  costs
increased from the mid- to Iate-I970s, Glastonbury
became interested in  converting the aeration system
to fine pore diffusers to reduce plant operating costs.

Alternative  fine  pore  diffuser systems were
investigated  by obtaining manufacturers' product
information and by making site visits to other facilities
that had been  retrofitted  from coarse  bubble  to fine
pore aeration.  Continued use of the existing  swing-
arm assemblies was desired because they were still in
goad condition and the need to drain an aeration  tank
during cleaning of  full-floor  coverage  fine pore
diffusers was not an  attractive operational feature for
this small plant. Bids were received for new fine  pore
diffusers in  October  1983, and the Pearlcomb  rigid
porous  plastic  tube  diffuser  was  selected for the
retrofit. Air filtration equipment was also purchased.

The  new aeration equipment was installed by the
Town public works employees and placed  in service
during the Fall  1S84. A smaller,  more efficient 75-kW
(100-hp) blower was installed in  1985 to better match
the airflow rate requirements of the new system.

An extensive aeration system performance  evaluation
involving six separate sfte visits was undertaken
during the  period November 1985-September 1988.
Over  160 individual  off-gas tests were performed
across the aeration  tank  to  determine  OTEs as a
function of diffuser location and age.

Summary details are  presented herein for the original
system  and  retrofit  designs, the  performance
evaluation of Ihe fine pore aeration system, and the
estimated  economic  benefits  of  the  aeration
modification.

II.  ORIGINAL WASTEWATER  TREATMENT
FACILITY
The original  wastewater  treatment facility  was
designed for a 1990 average daily flow of 158 Us (3.6
mgd} and  a  peak flow of  355  Us  (8.1  mgd).
Population growth and industrial flow projections  used
for design did not develop as anticipated. During the
aeration system  evaluation period, the average and
maximum daily flows were about 66 Us (1.5  mgd) and
88  Us  (2.0 mgd), respectively. The wastewater
treatment facility consists of preliminary treatment, two
primary clarifiers, activated sludge aeration tanks, two
19.8-m  (65-ft) circular secondary  clarifiers,   and
chlorine disinfection. Thickened waste sludge from the
plant is trucked to  the  Hartford  treatment facility for
final treatment and disposal.

The secondary treatment  system  contains  two
identical aeration tanks, but only one  has been  used
since actual flows have been less than design.  Each
aeration tank is 50,6 m  (166  ft) long and 12.2 m (40
ft) wide with an SWD of  4.6-4.7 m (15.0-15.5 ft) and is
divided into four equal-size passes. Each pass is 25.3
m (83  ft) long and 6.1 m  (20  ft)  wide. Total tank
volume is 2,890 m3  (764,000 gal). Figure 8-2 is  a
schematic of the aeration tanks. Return  sludge is
directed to the head of Pass 1  at an average rate
equal to about 34 percent of the ptant flow. Primary
effluent is normally  introduced at the head of Pass 2,
However, during the aeration evaluation site visits in
1988, primary effluent was fed to the head of Pass 1.

The original  aeration equipment installed in  1972
consisted of Walker Process Jacknife Sparger-Header
Diffuser  assemblies with   Defrin coarse bubble
spargers  installed on  the air manifolds at the end of
each jacknife. The coarse bubble spargers  contained
                                                212

-------
Figure 8-2   Aeration tank schematic - Glastonbury, CT,


	 ,.,.
_^RAS
-

r:

^

Pa
n






Pass 4

SS 1

^ ° fc
20' -8",
typ






1
, n 11 , 	 ,llpllulllim,l,.n
0 J 0
L^ PE
1— -Ik

_
0 I 0


166'
Pas


Swing-Arm
Assembly




Pass 3

s 2


20', typ
i







'

<








\
I

V

^

                                            PE - Primary Effluent
                                            RAS - Return Activated Sludge
                                            ML - Mixed Liquor
four 6-mm (1/4-in) diameter air orifices at 90 degrees
to one another. The total airflow capacity was 6.2 Us
(13.1 scfm)/diffuser at 28 cm (11 in) w.g. headless. A
total of 320 spargers was installed in each tank, with
20  spargers installed on  each  swing-arm assembly.
There are four swing-arm assemblies in each of the
four  aeration  tank  passes.  The  spargers  were
uniformly  spaced along  the air manifolds at 61-cm (2-
ft) intervals, and no aeration tapering was possible.

The centerline of the horizontal air manifold of each
swing-arm assembly was positioned 76-91 cm (2.5-3.0
ft) above  the aeration tank floor and 76 cm  (2.5 ft)
from  the  tank sidewall. Sparger  submergence  was
3.7-3.8 m (12.0-12.5 ft),.  The spargers  were  located
between the drop-leg of the swing-arm  and the  tank
sidewall. A spiral roll aeration and mixing pattern was
created by this diffuser placement.  aF(SOTE)  of  this
coarse bubble diffuser system was estimated to be 4-
5 percent

Air for the original aeration system was supplied by
three 149-kW (200-hp) Hoffman multi-stage centrifugal
blowers, each  with  an  airflow capacity  of 945-2,125
L/s  (2,000-4,500  scfm)  and an average discharge
pressure of 150 kPa (7.0 psig). Operation of only one
blower was required at any given time, and the airflow
rate was adjusted by throttling the blower inlet valve.

Mixed liquor DO and air supply control  were carried
out  with  manual  measurement of  DO  and  manual
adjustment of airflow to  Passes 1 and 2 and Passes 3
and 4 using in-line valves. In addition, each swing-arm
assembly  contained an isolation valve that could be
used to throttle airflow to that assembly.

III.   FINE  PORE   AERATION   SYSTEM
MODIFICATION
Retrofitting the plant to fine pore aeration was  based
on  the desire to  reduce  electrical  power costs,
although  several other considerations were identified
as part of the retrofit evaluation. These were;

1. Achieve a short initial cost payback  period  based
  on power savfngs, hopefully about 24 months.

2. Minimize any  additional  O&M  costs over  the
  existing aeration equipment.

3. Use the existing swing-arm assemblies so that tank
  draining  for  diffuser  cleaning would  not  be
  necessary.

4. Provide the same operational flexibility  as  afforded
  by the existing equipment.

5. Maintain the same  process treatment performance
  as before the retrofit.

6. Avoid the need to use multiple blower units to keep
  electrical demand charges acceptably low.

Prior to the retrofit feasibility study, one of the original
blowers was modified  to  provide lower airflow rates
                                                  213

-------
and  power  draw.  The  modification allowed airflow
turndown  from the original surge  point of 895  Us
(1,900 scfm) to under 755 L/s (1,600 scfm).

Glastonbury solicited bids  from various  fine  pore
aeration suppliers. Several were from fine  pore tube
manufacturers, including perforated membranes and
various rigid  media  types. At  least  one  bid was
received from  a fine pore dome diffuser manufacturer.
Evaluation parameters used by the Town to compare
the various bids included:

1. energy savings over the current aeration process,

2, expected equipment life,

3. operating enrgy costs over the system life,

4. O&M costs  over the system life,

5. initial cost,

6. equipment  flexibility  within the  system life  to
   accommodate varying flows  and loadings, and

7. location  of  systems currently using the proposed
   new aeration equipment.

For  the evaluation,  the  system life was set at  10
years, or the actual proposed life of the equipment, if
less  than 10 years.

The  successful  bid  called  for 320  Pearlcomb tube
diffusers, Model SP-35, to be installed in one of  the
aeration tanks. Each tube diffuser consisted of:

-  one  tube  adapter manufactured  from  an  ABS
   polymer  provided  with  a  stainless  steel insert
   suitable  for connection  to  a  19-mm (3/4-in) NPT
   thread and control orifice,

-  one diffuser tube,

-  one end cap manufactured from an ABS polymer,

-  one stainless steel rod threaded  at both ends with
   PVC nuts, and

-  one set  of Neoprene  gaskets  and  polyethylene
   washers.

They were shipped unassembled, requiring assembly
at the job site  during installation.

Similar to  the previous  system, 20  diffusers  were
installed on  each of the existing swing-arm assemblies
at 61-cm (2-ft) intervals. Although two  coarse bubble
diffusers had  been positioned between  the sidewall
and  air manifold  at  each interval location, only one
tubo was  located between  the  sidewall and  air
manifold, and the  other  was located on the inboard
side of  the  manifold  due to space  limitations. The
airflow rate  recommended  by the manufacturer  for
adequate  tank  mixing  was  1.9-2.4  L/s  (4-5
scfm)/diffuser,  compared to an airflow rate of  3.3  L/s
(7 scfm)/diffuser used for  the  coarse bubble  system
prior to  the retrofit. Thus, the anticipated airflow was
605-755 L/s (1,280-1,600 scfm)/tank.

One of  the three existing 149-kW (200-hpj blowers
was replaced with  a  75-kW  (100-hp)  Hoffman,
multistage centrifugal  blower.  This  blower has  an
airflow rate operating  range of 470-1,415 L/s  (1,000-
3,000 scfm} at a discharge pressure of 149 kPa (6.85
psig),

Mixed liquor  DO concentrations  are controlled  by
manual  DO measurements in the basin and throttling
the blower inlet valve.  DO concentrations were 1.5-4.5
mg/L in  the  winter and 0.5-1.2 mg/L during summer
operation.
IV.  OPERATIONAL  PERFORMANCE  AND
EVALUATION
After  a few  weeks of  operation,  several  diffusers
appeared to  be  clogged or  plugged and the airflow
patterns did not appear to be uniform. Investigation of
the  problem revealed that  mixed  liquor  was
backflowing   into the  blow-down  holes in the  ajr
manifolds on  each swing-arm assembly and plugging
the inside of the diffusers. Steps were taken to correct
this problem by plugging the holes,  and  the diffusers
were thoroughly cleaned.

During the aeration system evaluation period, average
plant influent  BOD5 was approximately 240 mg/L and
average primary  effluent BOD5 approximately  120
mg/L. Operating MLSS levels were  1,500-3,000 mg/L,
Average HRT and SRT were about  11 hours and 12
days,  respectively.  Effluent  BODg  and  SS
concentrations were usually below 10 mg/L.       .

For the aeration  performance evaluation, six  off-gas
tests were performed at the following times:

   November  21 and 22, 1985
   March 26 and 28, 1986
   September 17, 1987
   January 11, 1988
   May 17, 1988
   September 7, 1988

The sampling plan for  off-gas tests in November 1985
and  March 1986 consisted of two  sets of replicate
tests  in the  influent and effluent quadrants  of  each
aeration pass. Each  replicate test  consisted  of two
individual off-gas tests,  one taken with  the  hood
positioned across that half of the aeration pass  width
nearest the swing-arm assembly and one taken with
the hood  placed at  the half-width  position furthest
from the swing-arm assembly. The off-gas airflow was
                                                 214

-------
much higher in the first hood  position than in the
second position.

For the remaining four off-gas test visits, the sampling
plan was modified to use  a larger  hood  and one
replicate  per run  at  each sample  point.  Three
consecutive test  runs were conducted during  each
site  visit with this  plan. The larger collection  hood
extended  across  the full width of  the  aeration pass.
This sample plan reduced the time necessary to test
the total aeration tank on a once-through basis and
was more  representative  of  overall  performance
conditions in the aeration tank. Another advantage for
using the full-width hood was the  elimination of the
need to  calculate  airflow weighted  average OTEs at
the sampling locations.

The lowest off-gas OTEs were  observed during the
first series  of  tests in  November 1985. aF(SOTE)
vales were about 3 percent compared to 7 percent for
later  tests. Also,   at that  time,  significant  coarse
bubbling was observed at the liquid surface  directly
above the  diffusers. This was found to be due to air
leaking  around the gaskets at the ends of the tube
media for several diffusers. Tightening of the retaining
nuts on the diffusers eliminated the leaking problem.
There was no obvious trend  in changes in overall OTE
as a function of the length of service for the fine pore
tube diffusers. The  overall efficiencies were similar for
the  March  1986-September 1988  testing period.  It
should be  noted that the diffusers  were not cleaned
during this period. The only cleaning occurred before
the November  1985 testing, after observation of air-
side fouling due to backflowing mixed liquor.

Table 8-6 summarizes the average aF(SOTE) values
for the entire tank and for each pass. These data are
based on the  five  off-gas  testing   events  after
November 1985. All the averages  shown are airflow
weighted averages  using the  individual  sample
location  measured airflows  and off-gas  results.
aF(SOTE) was consistently lower in the first pass than
the other three passes  by  10-15  percent. This may
have been due to leakage around the diffuser gaskets,
indicated by more coarse bubbling in Pass 1.
Tattle 8-6.
 Basis"
Summary of  aF(SOTE)  Determinations
GlastonBury, CT
                Average
                               Average
                            — Estimated
             Minimum  Maximum    aF
                     aF(SOTE), percent
Entire Tank Avg.
Pass 1 Avg.
Pass 2 Avg.
Pass 3 Avg.
Pass 4 Avg.
7.2
6.7
7.3
7.8
7.5
6.6
6.1
4.7
7.3
7.3
7.6
7.3
8.4
8.3
7.7
0.61
0.56
0.61
0.66
0.63
   All efficiencies are airflow weighted averages. Does not include
   November 1985 test.
aF values shown in Table 8-6 were determined using
clean water oxygen  transfer data  obtained from  the
manufacturer and  during an  EPA-sponsored  oxygen
transfer test project conducted by Los Angeles County
Sanitation Districts. The estimated  average SOTE for
the Glastonbury fine pore tubes is 11.9 percent. aF
values ranged from 0.56 to 0.66 with an  average of
0.61 for the entire aeration tank.

Power Consumption and Airflow Rate
Although  no long-term  baseline information exists for
pre-retrofit  blower  electrical power  usage,  measured
power reduction,  on a spot-check basis, after  the
retrofit and installation  of  the  new 75-kW (100-hp)
blower was approximately 50 kW (67 hp), a reduction
of about 40 percent.  Although most  of  the  power
reduction is due to in-tank equipment improvements,
there  was  a significant  improvement  in  air
compression efficiency due to  the  installation  of  the
new  blower.  This  power  reduction  compares
reasonably well to the observed reduction in airflow to
the retrofitted aeration  tank. Before the  retrofit  the
airflow rate ranged from 1,040 Us (2,200) to 1,180 Us
(2,500 scfm), and  after the retrofit it  ranged from
below 470  Us  (1,000 scfm) to  about 895 Us  (1,900
scfm).

Diffuser Cleaning
The fine  pore tube diffusers have  been cleaned only
once, which was  a  few weeks after  start-up  when
internal fouling of  the  diffusers was discovered.  No
routine cleaning program has been performed since.
During the cleaning,  the diffusers were disassembled
and  a detergent  wash  used.  The diffusers  were
disassembled and the diffuser tubes soaked for about
30  minutes  in  a  solution of  warm  water  and
commercial grade detergent. A  firm bristle brush was
then used  to  wash the inner  and  outer surfaces to
loosen deposits. Finally, a thorough clean  water rinse
was applied before  reassembling  and  placing  the
diffusers  back into service.

V. ECONOMIC CONSIDERATIONS
aF(SOTE) averaged  6.5-7.0 percent for the fine pore
tube  diffusers., compared  to  an  estimated  4.0-4.5
percent for the original coarse  bubble  diffusers. The
post-retrofit aF(SOTE)  is thus 150-160  percent of the
original aF(SQTE).

Based on the observed blower  reduction of about 50
kWh, the estimated  electricity cost savings is  about
$50/d, or $15,000-$18,QOO/yr,  depending on  blower
usage and  maintenance performed.

The approximate total initial cost for the retrofit project
was $28,000. The price of the diffusers  was  about
$30  each,  unassembled, for  a total cost of  $9,600.
The time required for Town employees to remove the
old diffusers,  recondition the swing-arm assemblies,
and  install the new  diffusers was  about  0.75  labor-
hr/diffuser.  At a rate of  $20/labor-hr, the installation
                                                  215

-------
cost was approximately $6,500. Miscellaneous pipe
fittings  and  hardware cost about $2,000.  The
estimated total cost of the new blower was $10,000,
including air filtration equipment and installation.
The actual payback  period is  affected by the diffuser
cleaning method used and frequency of cleaning. The
effort required for the  detergent cleaning  operation
was much greater than routine in-situ cleaning  of the
exterior surface of  diffusers.  The time required to
disassemble  and clean  each  diffuser  was  20-30
minutes  with  a cost of  $5.00-$7.50/diffuser  (or  an
actual cost of $1,600-$2,4QO).

Cleaning  of  diffuser  exterior  surfaces  was  not
performed at  Glastonbury, but  hosing  and brushing
should take no more than  2-3  minutes/diffuser. Annual
maintenance should  include  at least  one exterior
cleaning  of the   diffusers  with  an  inspection  and
tightening of retaining  nuts and gaskets  to  prevent
coarse bubbling due to air  leaks. The estimated cost
for annual diffuser exterior  surface cleaning is  about
$500 for the Glastonbury system.

Based on the estimated total initial cost and electrical
power savings, the simple payback  period is  18-21
months. Simple payback periods of less than  2  years
may thus be  possible  for  replacement of spiral roll
coarse bubble aeration equipment with spiral  roll fine
pore  tube diffuser  aeration equipment.  The actual
payback  period will be a  function of the total  initial
cost and the  ability to reduce blower power usage
through  turndown  and/or  shutdown  of  blower  units
after  retrofitting  with  new  equipment. For the
Glastonbury type  of  retrofit,  in-tahk   equipment
modifications are generally inexpensive if air piping
and  swing-arm  assemblies  are in  good working
condition. The cost of  blower  modifications can vary
depending on the site.
                                                  216

-------
   8:2.3 Green Bay Wastewater Treatment Plant

   LOCATION: Green Bay, Wisconsin

   OPERATING AGENCY: Green Bay  Metropolitan Sewerage District

   DESIGN FLOW: 2,300  L/s (52.5 mgd)

   WASTEWATER: Domestic Plus  Pulp and Paper

   ORIGINAL AERATION SYSTEM: Sparged Turbines

   FINE PORE AERATION SYSTEM:  Sanitaire  Ceramic Disc and Parkson Perforated  (Flexible)  Membrane
                                    Tube Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION: 1986

   BASIS OF PERFORMANCE EVALUATION: Off-Gas Testing

   CLEANING METHOD:   Ceramic Discs; Hosing-ln-Situ HCI Gas Cleaning-Hosing
   	                  Perforated Membrane Tubes; Hosing-Brushing-Hosing   	_^ 	  	
I. INTRODUCTION
Based  on pilot-plant studies conducted  in 1983, the
Green  Bay Metropolitan Sewerage District (GBMSD)
decided to investigate the use of fine pore aeration as
an alternative to existing sparged turbine aerators to
reduce plant operating  costs. Both ceramic disc and
perforated membrane tube  diffusers were considered
viable  alternatives,  but  long-term  testing  was
undertaken to determine the possibility and effects of
diffuser fouling  on  OTE  under  actual wastewater
treatment conditions.

After installation of the two types of fine pore aeration
systems in individual  activated sludge  treatment
quadrants,  an 18-month field  testing program was
initiated  in May  1986 to  evaluate their  long-term
performance.  Initially,  the  two fine  pore  aeration
quadrants treated the entire plant flow until a decrease
in OTE required  directing some of the influent flow to
one  of the  two remaining sparged turbine  aeration
quadrants. The  diffuser  performance evaluation
included  an extensive field  off-gas testing program to
observe  changes in  OTE vs. operating time  and the
effects of  diffuser  cleaning  on oxygen  transfer
performance. A  pilot program was  also carried out
concurrently to  evaluate characteristics of  diffuser
fouling, changes in diffuser performance after fouling,
changes  in diffuser operating characteristics  such as
DWP and  bubble release  vacuum  (BRV),  and the
effects of  different cleaning methods  on restroing
diffuser performance.

Summary details are presented herein on  the Green
Bay retrofit design, the  performance and evaluation of
the two types of fine pore diffusers, the results of the
diffuser cleaning evaluation program, and the results
of an economic analysis on  the  conversion  to fine
pore aeration.

II. HISTORICAL BACKGROUND
GBMSD uses  a  modified  contact  stabilization
activated sludge facility for  secondary treatment.
Sparged  turbine aeration was used  in  the original
facility, and aeration energy costs  typically accounted
for about one-half of the total plant's electrical energy
costs.

With an objective of reducing  plant operating costs,
recommendations were made in 1983 (based on field
test  work in  Summer  1983)  to consider  fine  pore
ceramic diffusion as  a possible replacement for the
existing  mechanical  aeration  system.  However,  a
major concern with fine pore  diffusers  was potential
media fouling and the effects that such fouling could
have  on overall system economics  by increasing
operating pressure, decreasing OTE, and increasing
maintenance  costs  for diffuser  cleaning. It  was
recommended,  therefore,  that fine pore  diffused
aeration be  tested initially  by installing a  full-scale
ceramic diffuser system in only one of four quadrants
of the activated sludge system complex.

Since the\1983 field  tests showed that a substantial
portion of the foulant materials was soluble in acid, the
recommended  ceramic  installation  was to include an
in-situ  HCI gas injection  system.   Preliminary
calculations indicated that,  to operate the fine  pore
system in a mode equivalent to design year loadings,
approximately 40 percent of the influent flow would
have to be fed  to the  fine pore system and 60 percent
to one  or  more of the  existing sparged turbine
systems.
                                                 217

-------
In the fall of 1984,  GBMSD  considered  perforated
membrane tube diffusers as  an additional alternate
fine  pore system.  The  manufacturer  claimed that
these diffusers normally do not foul and,  if fouling
occurrs,  it can  be  controlled  by flexing  the  units.
Based on the data  available at the time, the ceramic
diffusers were expected  to  have a higher OTE than
the membrane diffusers when  the  systems were
clean. However,  since  fouling was  expected  to
adversely affect OTE of the ceramic  diffusers,  the
perforated  membrane units could possibly produce
better OTEs in the  long run if,  in fact, the membrane
diffusers did not foul.

Because the cost effectiveness of ceramic disc and
perforated membrane tube  diffusers  was  believed to
be  similar,  and   fouling characteristics  and
maintenance procedures were  potentially  different,  it
was decided to install and test both types of fine pore
diffusers. Testing two fine pore diffuser systems also
provided the  opportunity to try  to  treat  all  of  the
wastewater  in the  two  test quadrants and achieve
substantial energy savings during the test period. The
membrane diffuser  system was installed  first and
placed in service in January 1986. By April 1986 the
ceramic  diffuser  system installation  was  completed.
To provide a clean  start  for both systems, the basins
containing the membrane diffusers were  drained and a
cleaning   procedure   recommended  by  the
manufacturer followed.  This  involved  hosing  the
membranes, scrubbing them with a stiff bristle brush,
and rehosing.  Both  systems were placed in service in
May 1986.
III.  FINE PORE  AERATION  RETROFIT  DESIGN
DESCRIPTION
The GBMSD activated sludge facility treats a mixture
of metropolitan (Metro) and pulp and paper mill (mill)
wastewaters. The Metro wastewaters  are  comprised
of municipal and industrial wastes including a seasonal
contribution  from  vegetable  canning  industries. The
Metro  wastewaters receive  preliminary  and primary
treatment before entering the activated sludge system.
The mill wastewaters receive primary treatment at the
mills before  they enter  the plant through separate
interceptor sewers and  are  pumped directly to the
activated sludge system.  Plant recycle streams, which
include decant  liquor  from  a Zimpro  sludge  heat
treatment process, are returned to  the primary effluent
channels.

The GBMSD activated  sludge  facility  is  a contact
stabilization  system  consisting  of four  quadrants
(Figure 8-3).  Each quadrant includes a  22.3-m x  74.4-
m x 6.2-m (73.3-ft x 244-ft  x 20.5-ft)  SWD contact
basin (10,370-m3 [2.74-mil gal] volume) and a 11.1-m
X 74.4-m x  6.9-m (36.3-ft  x 244-ft x  22.5-ft)  SWD
reaeration basin (5,680 m3 [1.5-mil gal] volume). Inlet
flumes were added to the contact  basins to distribute
the incoming wastewater across  the  entire width  of
the basin. Each contact basin was originally equipped
with 12 93-kW (125-hp) sparged turbine aerators. The
reaeration basins  had  six  56-kW  (75-hp)  sparged
turbine aerators. The  process  air is supplied by four
1,864-kW (2,500-hp) centrifugal blowers, each having
a capacity of 21,000 Us (44,500 scfm) at a discharge
pressure of 184 kPa (12-psig). The centrifugal blowers
did not need modification  because  the 184-kPa (12-
psig) discharge pressure was more than adequate for
the 5.8-m (19.1-ft) submergence of the diffusers.

Quadrant 2 was retrofitted  with Sanitaire ceramic disc
diffusers,  while   Quadrant  4  received  Parkson
perforated membrane  tube  diffusers.  The design
criteria for selecting the number and distribution  of
fine pore  diffusers  in the  aeration basins are
presented in Table 8-7. The design criteria  were
based on the results of the field studies conducted in
1983, operating experience  with the sparged turbines,
and information supplied by the manufacturers of the
fine  pore  aeration systems. The  a values used for
design,  0.68 in the contact basins and 0.90 in the
reaeration basins, were  selected based on  the results
of the 1983 field work and  with consideration given  to
other design constraints such as maximizing power
savings during the test period by treating the  entire
plant  flow  in  the  two  fine pore diffuser  equipped
quadrants.  However,  this  design strategy  was only
viable if the actual a values were equal to or greater
than the values used  for design.  In the event that a
values were lower or plant  loadings  higher than
expected, additional aeration capacity could be added
by  placing one  or both of the  remaining  sparged
turbine quadrants in service.

The contact basins were divided into three zones and
subdivided  into ten  grids each.  The number  of
diffusers per grid  was  the highest in the  tank inlet
zone and  lowest  in  the  tank  outlet  zone.  The
reaeration basins were divided into six grids  each.
Since oxygen  demand was expected  to be  relatively
uniform in these basins, the number of diffusers per
grid was constant.

The number of diffusers installed in each grid and the
design airflow rates per diffuser  are  summarized  in
Table 8-8. The design average airflow rates used for
the ceramic diffusers,  1 Us (2.1  scfm)/diffuser  in the
contact basin  and 0.9  Us  (1.9 scfm)/diffuser in the
reaeration basin, were relatively high compared  to the
normal practice of designing  for  about 0.6 Us (1.25
scfm)/diffuser.  This was due to the high  organic
loading  applied to  the activated  sludge system. The
airflow rates in  the contact  basin are a result  of
providing the maximum number of diffusers that  would
physically fit in the basin inlet grids and then using the
same diffuser  airflow  rate  for all  grids in the  basin.
This relatively high average airflow rate  per diffuser
was possible because the  mill waste  discharge
resulted in minimal diurnal load variations.
                                                  218

-------
Figure 8-3.   Treatment plant schematic - Green Bay, Wl.

            Secondary Clarifers
  Secondary
   Effluent
                        RAS
Reaeration
Quadrant 2
Contact
• "J

l
•••
r*.
' 1
> i
t
•*!
Reaeration
Quadrant 1
Conlact
y :-__
                                                                                    Primary Clarifiers
                                      ML
                                                                            ML
                                                                   Decant
                                                                            ML
1 f
Quadrant 4
Contact
Reaeration
*•
i
MB—
i
l i
t j
*
*|
[
l

r
Quadrant 3
Contact
Reaeration
                                                                                                   Metro WW
                                                                                                    Mill WW
      RAS - Return Activated Sludge
      ML - Mixed Liquor
      Decant - Decant Liquor from Sludge •
             Heat Treatment Process
                                            RAS
                             Quadrants 1 and 3 - Sparged Turbine Aerators
                             Quadrant 2 - Ceramic Disc Diffusers
                             Quadrant 4 - Perforated Membrane Tube Diffusers
Table 8-7.    Fine  Pore  Aeration System Design Criteria
            Green Bay, Wl

 Parameter	.	Criteria	_^

 BODgi, Ib/d
  Average day (50th percentile)
  Maximum month
  Maximum day
 O;> Requirement2
 Contact Basins
  Size
  Number
  Volume, ga! per quadrant
  O2 demand
  O2 demand profile, %: Zone 1
                    Zone 2
                    Zone 3
         161,000
         202,000
         239,000

 1.0 Ib 02/lb BODS applied
  Diffuser submergence, ft
 Rearation Basins
  Size
  Number
  Volume, gal per quadrant
  O2 demand
  O2 demand profile  •
  Diffuser submergence, ft
73.3 ft x 244 ft x'20.5 ft deep
          4
       2,740,000.    ;  -
   75 % of quadrant total
          55
          30
          15
         0.68
         0.95
         19.1


36.3 ft X 244 ft X 22.5 ft .deep
          4
       1,540,000
   25 % of quadrant total
   Uniformly Distributed
         0.90
         0.95
         19.1
  1 Calculated from data reported for January 1982-April 1983." A
   total of 478 values were used.
  2 Based on 1983 off-gas results.
The membrane system was designed for  an airflow
rate of 1.4 Us (2.9 scfm)/diffuser in the contact basin
and 1.2 Us (2.6 scfm)/diffuser in the reaeration basin.
The manufacturer's recommended  design  range  for
these diffusers was between 0.9 and 2.4  Us (2.0 and
5.0 scfm)/diffuser.

Before  the aeration  equipment was  approved  for
installation,  manufacturer shop tests were  conducted
to  determine  clean  water  OTE.   The  tests were
conducted  in  accordance  with the  ASCE Standard
procedure. The  Sanitaire  ceramic  discs 'met  or
exceeded the specified SOTE values for each  contact
basin zone and for the sludge reaeration  basin with
the  respective diffuser densities and diffuser airflow
rates shown in Table 8-8.  These  SOTEs were 31.0,
35.7, 38,3,  and 36.2 percent for contact  basin Zones
1, 2, and 3  and the reaeration basin, respectively.  The
Parkson membrane  diffusers achieved a  much higher
SOTE than the specified value of  27.5  percent.  The
SOTEs were 31.3,  33.8, 31.7, and- 34.9 for  contact
basin Zones  1,  2 and 3  and the reaeration basin,
respectively. Since the number of membrane diffusers
was not reduced to account for the  higher SOTEs,  the
projected design airflow rates per membrane  diffuser
were reduced  to about 0.9 Us (2.0 scfm)/diffuser in
the reaeration  basin and  1.2 Us (2.5 scfm)/diffuser in
the contact basin.

Five removable pilot headers were provided in  each
contact  basin, and  two  were provided in each
                                                     219

-------
Tabto 8*8.   Fine Pore Diffuser Configuration Summary - Green Bay, wi


Contact Basin 2 - Ceramic Disc Oiffusers



Rcaoratm Basin 2
Contact Basin 4 •



RcaofotKJH Basin 4




- Ceramic Disc Diflusers
Perforated Membrane Tube
Dilfusors


Perforated Membrane Tube
Dilfuscfs
Zone
1
2
3
Total
1
1
2
3
Total
1

No.
Grids
4
4
2

6
4
4
2

. 6

Diffuseis/Grid
805
490
474

358
616
350
378

233

Diffusers/Zone
3,220
1,960
948
6,128
2,148
2,464
1,400
756
4,620
1,398

Diffuser Density,
sq ft floor
area/diffuser
1.85
3.03
6.25

4.17
2.44
4.35
7.69

6.25

Design
Airflow,
sclm/diffuser
2.1' '
2.1
2.1

1.9
2.9
2.9
2.9

2.6

reasratton  basin to  obtain diffusers for  visual
observalion  and testing without having to  drain  the
basins.  Each  removable header  was  equipped  with
four diffusers  at a submergence of about 5.2 m  (17
ft).  The headers were placed at the inlet, middle, and
outlet of the contact basins and at the inlet and outlet
of  the  reaeration  basins.  The diffusers  on   the
removable headers and two diffusers  per  fixed  grid
were provided with pressure taps for monitoring DWP.

A two-stage, in-line air filter was provided on the main
header to the ceramic disc quadrant. The air filter was
originally designed to remove more than 99.9 percent
of all particles  2:0.3 micron  in size.  After start-up,
problems  were encountered with  excessive  headloss
across the filter so the second-stage elements were
changed to a coarser medium. The filtration efficiency
of the modified filter was 98  percent of all particles
a 1.0 micron in size.

IV.  OPERATIONAL  PERFORMANCE AND
EVALUATION

Operation and Testing Procedures
The main goals of the evaluation program were to
operate each  fine  pore diffuser  quadrant   at equal
organic  loadings,  follow  recommended  diffuser
cleaning methods, and  monitor OTE performance of
the  diffusers  with  the standard  off-gas testing
procedure.  Table  8-9  summarizes  the  monthly
average operating conditions for the  two systems for
the duration of the test program. A low  operating SRT
was used to  prevent nitrification.  The  average
operating  DO  concentrations  and   SRTs were
essentially equal for the two systems. The membrane
diffuser system had a slightly higher average organic
loading, which was due to the inability  to  divert an
exactly equal distribution of  influent to each quadrant.

For  the first 6 months of  operation before a major
diffuser cleaning effort  was  carried  out, ,half of  the
ceramic grids were treated  with HCI gas at 1-3 month
intervals and half of the membrane grids were flexed
every 3 weeks. Normally, the HCI gas  dose is added
in  sufficient  quantities during cleaning until  the DWP
decreases to a predetermined  level.  Since  DWP did
not increase appreciably during plant operation,  a
fixed  HCI gas dose of approximately  45  g  (0.1
Ib)/diffuser was used.

The membrane flexing procedure was done according
to  the membrane diffuser manufacturer's instructions.
One grid was  flexed at  a  time ,by closing  the
downcomer air valve, bleeding off the air in the header
system so the membranes would collapse completely
onto the frame, increasing the airflow rate to  about 3.8
Us (8 scfm)/diffuser for 2-5 minutes, and returning the
airflow rate to the previous operating level.

The first  of,the nine off-gas tests was conducted  on
May 13, 1986 (just after the start of the test program),
and five  more tests  were performed before a major
diffuser cleaning in November  1986.  An off-gas test
was conducted just after cleaning in December 1986,
and the other three off-gas tests were conducted in
June,  August, and October 1987.  The August 1987
test followed another major diffuser cleaning  in July
1987.

A typical off-gas test was conducted over 2 days. One
day was  used to survey 24 positions in each of the
contact basins. The second day was used to survey
12  positions in each of  the  reaeration  basins.
Corresponding  positions  in each  basin  were
alternately sampled and analyzed at  about the same
time during the  sampling  day. The  results of the off-
gas testing were summarized by calculating a weighed
aF(SOTE) value for  each contact  and reaeration
basin.  The  weighted average is  based  on  the
measured off-gas flow and OTE.

Based on pilot tests, the rigorous diffuser cleaning
methods  used in November 1986 and July 1987 were
modified  from the  original procedures. The cleaning
procedure selected for use on  the ceramic diffusers
was hosing from the tank  top with fire hoses, partially
filling the basin with service water,  gas cleaning with
45 g  (0.1 Ib) HO/diffuser, draining  off the  service
                                                 220

-------
Table 8-9.   Operating Data - Green Bay, Wl

                        Quadrant 2 - Ceramic Disc Diffusers
             Quadrant 4 - Membrane Tube Diffusers
Month
May 1986
June
July
August
September
October
November
December
January 1987
February
March
April
May
June
July
August
September
October
Average
BOD5 Loading,
1 ,000 Ib/d
56.5
55.5
57.0
58.8
56.0
58.1
49.7
51.2
52.9
64.7
55.3
54.1
31.0
36.7
35.4
52.6
44.1 '
43.2
50.7
SRT, days
3.00
2.36
3.23
2.65
2.52
2.66
3.14
3.21
2.90
2.89
2.82
2.73
4.26
2.96
4.72
3.24
3.50
3.91
3.15
DO, mg/L
2.4
1.6
1.8
1.2
1.3
1.5
2.3
2.5
2.5
1.6
1.7
2.1
2.4
2.2 ,
2.0
1.6
2.0
2.4
2.0
BOD5 Loading,
1 ,000 Ib/d
54.7
65.2
59.4
62.8
58.7
62.1
55.3
53.8
54.2
64.8 ,
58.1
59.1
38.2
39.5
38.3
46.0
45.2
44.1
55.3
SRT, days
3.02
2.36
3.03
2.86
2.86
2.87
3.37
3.64
3.42
3.26
3.09
2.92
3.67
2.74
4.91
3.55
3.98
4.11
3.31
DO, mg/L
2.3
1.6
1.8
1.4
1.6
1.6
1.8
2.5
2.4
2.1
2.0
2.0
2.2
2.0
1.8
1.8
2.0
: 2.3
2.0
water,  and  rehosing  from the tank top. All  of  the
membranes were cleaned  by hosing  from the basin
floor,  scrubbing  with  a stiff-bristled  brush,  and
rehosing.

Aeration Performance
Table  8-10  summarizes weighted aF(SOTE)  values
determined  from  off-gas testing of the ceramic and
membrane  diffuser  systems. A comparison  of
performance between  the  two types of diffusers is
shown chronologically, in  Figure 8-4. The two systems
started out at nearly equal aF(SOTE)s. aF(SOTE) then
decreased substantially tor both systems such that,
after 3 months of operation,  values were about 75
percent of their respective initial values.  In October
1986,  the fifth off-gas test was conducted and  the
aF(SOTE)s  were  still substantially below the  values
measured in May 1986. There was also a problem
with maintaining acceptable DO concentrations at the
inlet end of the contact basins, so it was decided to
drain the  basins for inspection and rigorous cleaning
of the  diffusers. Once the cleaning was complete, the
systems  were put back into  service and a follow up
off-gas test  was conducted. The increased aF(SOTE)
of the  ceramic system indicated that  the discs were
essentially restored to their original condition in  both
the contact  and reaeration  basins.  A smaller increase
in  aF(SOTE)  was  observed in  the  membrane-
equipped basins.

The next  off-gas  test was conducted  in June  1987.
Measured aF(SOTE)s were similar  to those measured
in  October 1986 before  the systems  had  been
cleaned.  Based on the off-gas test results, a decision
was made to  drain,  inspect,  and clean the two test
quadrants. After cleaning, the  systems  were put back
into service and off-gas tested. The aF(SOTE) of the
ceramic  system  showed an  increase to above its
origjnal  level.  The  membranes showed  little  it any
improvement.

The  last off-gas, test was conducted during the last
week of October 1987. The aF(SOTE) of the contact
basin ceramic  disc system decreased from nearly 25
percent to about 17 percent. A similar reduction was
experienced in the reaeration basin ceramic  disc
system. The membrane systems  remained  relatively
constant at about 12 percent in the contact basin and
13 percent in the reaeration basin.

In summary, off-gas testing before and  after rigorous
cleaning indicated the ceramic  diffusers could achieve
higher aF(SOTE)s with cleaning,  but that membrane
system  aF(SOTE)s  were  relatively unaffected by
cleaning. This is further  illustrated  in  Table  8-11,
which also shows the  aF values  calculated for the
systems  using  the earlier determined  clean water
SOTE values. The aF values approached similar
values in  the  range of 0.34-0.44 for both  diffusers
after  fouling increased. Clean  ceramic  diffusers  in
either basin yielded aF values in  the range of 0.5-
0.65.

aF  values  were  also  not greatly different for the
contact and reaeration basins. Contrary to some other
studies,  aF values were not  depressed  at the
wastewater addition points. Tank geometry may have
influenced this observation, since GBMSD tanks have
a length-to-width ratio of 4:1, which is lower than for
many plug flow type  systems that  exhibit a  more
significant reduction in  aF at the  influent introduction
zone.

Cleaning and Diffuser Evaluations
Diffusers were  tested  at  various  times during the
study to observe changes in  diffuser characteristics
                                                  221

-------
Figure 8-4,  OTE (off-gas method) vs. time - Green Bay, Wl.

  oF(SOTE), percent

   25  r-
                                                                                        Conlaot Basin
                                                                                      Reaeration Basin
                                                 Ceramic Discs
                      100
                                    200
      300

Time in Service, days
                                                                  400
                                                                                 500
                                                                                               600
Table 8-10.  aF(SOTE) Values torn Off-Gas Testing
           Bay, Wl

                                  aF(SOTE)
Green
Basin
Contact








Roaoraiiort








Test Dale
5/13/86
S/15/86
7/2/86
7/30/86
10/30/86
12/3/86
6/18/87
8/5/87
10/28/87
5/12/86
5/16/86
7/1/86
7/29/86
10/29/86
12/2/86
6/17/87
8'4/87
10/27/87
Ceramic
Diffusers
14.8
14.7
17.1
9.7
12.2
19.1
11.8
20.2
16.5
18.2
17.0
21.2
14.3
11.6
19.6
11.8
23.2
19.5
Membrane.
Diffusers
16,5
16.2
17.0
12.1
14,3
16.3
11.4
12.1
11.5
17.6
17.0
18.4
11.2
13.4
13.1
11.1
13.7
12.7
        caused by  fouling  or  aging  and to  evaluate the
        effectiveness of preventive maintenance and cleaning
        procedures.  Both types   of  diffusers  were
        characterized  by DWP  and steady-state clean water
        OTE tests.  The  ceramic diffusers  were  further
        evaluated using the BRV test. Ceramic disc thickness
        and strength and perforated  membrane size, weight,
        hardness, and modulus  of elasticity were measured to
        quantify any changes caused by aging. Steady-state
        clean water OTE tests were run in the laboratory in a
        76-cm (30-in)  diameter by 3-m (10-ft) water depth
        tank  to  characterize  diffuser OTE.  Steady-state
        conditions were established by feeding a constant rate
        of  sodium  sulfite  solution  to maintain  a  DO
        concentration  of 1-3 mg/L. OTE was  measured using
        the off-gas method. Diffusers for testing  were obtained
        from the removable pilot headers or from the in-basin
        grids when the basins were drained for inspection and
        cleaning.

        Ceramic diffusers removed from the  reaeration basin
        had different  fouling  characteristics than  those
        removed from  the contact basin. They both  had an
                                                  222

-------
Table 8-11.  uF as a Function of Time in Service - Green Bay, Wl

                                     Time in Service, morrths
 Basin
                              Total
                                                       Since Cleaning
                      Ceramic
                                   Membrane
                                                  Ceramic
                                                               Membrane
                                                                                       aF
                         Ceramic
                                                                                           Membrane
Contact
Reaeration
Contact
Reaeration
Contact
Reaeration
Contact
Reaeration
Contact
Reaeration
Contact
Reaeration
Contact
Reaeration
Contact
Reaeration
<-,
<1
2
2
3
3
6
6
7
7
13
13
15
15
18
18
4
4
6
6
7
7
10
10
11
11
17
17
19
19
22
22

-------
Tnbto 8-12.  Cost Summary Comparison for Alternative Aeration Systems - Green Bay, Wl
 «• Dabod on treating 55,000,000 lb/BQDs/yr.
 >» Based on power cost of $18.56/1,000 Ib BODS applied.
 « Based on power cost of Si 0,00/1,000 Ib BOD8 applied.
 «i Based on powor cost of $15,40/1,000 ib BODS applied.
 * Based on 20 years at 8-7/8 percent intcresl, PWF = 9.21.
                                                                         Perforated Membrane Tubes
Cost
Capital, $
Annual O&M, S/yr
iloclncal Powor«
Moclidiueal Maintenance
Oilkisor Cleaning
Tolal
O&M Present Worlli", $
Tolul Prasont Wortli, $
Sparged Turbines
0

1,020,800**
57,000
0
1,077,800
9,927,500
9,927,500
Ceramic Discs
2,221,600

550,000C
5,400
45,900
601 ,300
5,538,000
7,759,600
New
1 ,998,200

550,000«
5,400
8,600
564,000
5,194,400
7,192,600
Used
1,998,200

847,000d
8,100
8,600
863,700
7,954,700
9,952,900
The acid gas storage and feed building and acid feed
equipment were included in the ceramic  diffuser cost
estimate.  A royalty  payment of $236,800  was also
included in the ceramic  diffuser  system  initial cost
estimate for use  of the in-situ acid gas cleaning
procedure.
The costs for operating  the  two fine  pore  aeration
systems and  the sparged  turbine aeration  system
would  typically include labor  for process  monitoring
and  adjustment  of  activated  sludge  control
parameters, and  the  electrical power to  run the  air
supply blowers and turbine mixers. However,  since all
three aeration  systems would require  an equivalent
amount of effort  for monitoring the  activated sludge
process and adjusting the airflow rate to  the basins,
these  estimated  costs were not  included in  the
comparison.
Estimated aeration  energy costs were  calculated  by
first  determining the actual average energy  required
per unit of  BODS applied for  the  three systems. A
value was determined for the sparged turbine system
by averaging energy consumption and  BODs applied
data for 1983-1985. Using a historical electrical cost
of $0,04/kWh,  the  unit power cost for the  sparged
turbine system  was $40.92/1,000 kg ($18.56/1,000  Ib)
BODg applied  . The  fine  pore diffuser performance
results for the fast 4 months of the study were used to
obtain the unit  power cost of these aeration systems.
These data reflect the higher  OTE of the  ceramic
diffusor due to  its favorable response to  cleaning. The
unit power costs used for the ceramic and membrane
diffusers were  $22.04 and $33.i5/1,000 kg ($10.00
and $15.40/1,000 Ib) BOD5 applied,  respectively.
Diffuser  cleaning costs  for  the ceramic  system
assumed  draining  the basins  twice annually  for
inspection and cleaning with hosing followed by in-situ
HGI gas treatment and rehosing.  Monthly  in-situ acid
gas treatment is assumed  for the  reaeration basins to
prevent a buildup of the inner black foulant layer. A
minimal  cleaning  effort was  included  for the
membrane  diffusers, which  consists  of draining the
basins once per year for  inspection and H-B-H
cleaning. The sparged  turbine system  is  shown  in
Table 8-12  to have the highest total maintenance cost
followed by used perforated membrane tube diffusers,
then  ceramic  disc  diffusers,  and, finally,   new
perforated membrane tube diffusers.

A  comparison of electrical  power costs  and  O&M
costs between the ceramic disc diffuser and sparged
turbine aeration system (Table 8-12) shows a savings
of about $471,000/yr in  electrical power  costs and
$476,500/yr in total O&M  costs for the ceramic disc
system. The simple payback period on  the  initial
investment  would be about 4.5 years.

The  membrane  tube diffusers would have shown
similar economics had they  not lost  their initial  high
OTE.  Based on the performance of new diffusers, the
annual savings  in total O&M costs are  estimated  at
$514,000,  which would  have  provided  a  simple
payback period of about 4 years. However, the lost
transfer efficiency resulted in a substantial increase in
the electrical power requirements  and  an actual  O&M
savings of  only $214,000/yr. The  simple  payback
period under these circumstances  would be more than
9 years. Based on a 20-yr life of the  equipment and
an interest  rate of 8-7/8 percent, retrofitting with the
membrane  tube  diffusers  would  not  be  economical
because  the total present worth  of  the  retrofit  is
greater than the cost  of  continuing  to operate the
sparged turbines.
                                                  224

-------
   8.2.4 Hartford Water Pollution Control Plant

   LOCATION: Hartford, Connecticutt

   OPERATING AGENCY: Hartford Metropolitan District Commission

   DESIGN FLOW:  2,630 Us (60 mgd)

   WASTEWATER: City of Hartford Plus Six Area Towns

   ORIGINAL AERATION SYSTEM: Chicago Pump  "Defieetofuser" Coarse Bubble Diffusers

   FINE PORE AERATION SYSTEM: Ceramic Dome Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION: 1982

   BASIS OF PERFORMANCE EVALUATION: Off-Gas Testing

   CLEANING  METHOD: Hosing/Acid Application/Hosing ("Milwaukee Method")	
I. INTRODUCTION
In  November 1982,  the Hartford Water Pollution
Control Plant in South Meadows, CT went on-line with
a retrofit  from a coarse bubble to a fine pore aeration
system. Four of the six aeration tanks  were retrofitted
with diffusers, air-line  valves  and  controls  were
modified, and air filters were installed  on the inlets to
all three  existing blowers. Extensive aeration  studies
were conducted on the fine pore system in 1985-1987
to  evaluate  OTE over time and the effectiveness of
diffuser  cleaning. Information  is presented  on the
original aeration system,  the basis for changing to fine
pore diffusers, retrofit design and performance, and an
economic evaluation of the new system.
II. HISTORICAL BACKGROUND
The Hartford plant treats wastewater from six greater
Hartford area towns. The secondary activated sludge
treatment  facility processes  an average daily flow of
about  1,970 L/s  (45  mgd);  design is  based on  an
average daily flow of 2,630 L/s (60  mgd) and a peak
hydraulic flow of 4,775 L/s (109 mgd).

Faced  with  the need  for additional  aeration capacity
(greater oxygen transfer), steadily increasing electrical
rates,  and the demand  charge for placing  a second
2,240-kW (3,000-hp)  blower  on  line,  Hartford
Metropolitan  District  Commission  (MDC)  staff
engineers initiated a retrofit evaluation project in 1978
for the purpose of examining ways of reducing future
power  costs.

In addition  to the  objective of  significant electrical
savings, other objectives of the retrofit were: 1) simple
payback  of  initial  cost in approximately  3  years, 2)
adding minimum O&M costs  over the existing aeration
system,   3)  providing  operational  flexibility,  4)
increasing  OTE to avoid having to use two  large
blowers simultaneously,  and 5) maintaining  process
performance and effluent water quality.

Original Aeration System
In  November 1972, the new secondary  wastewater
treatment facility  began  operation. The activated
sludge aeration  system consisted of six  identical four-
pass aeration tanks, each 24.4 m wide x 59.2 m long
(80 ft x 194 ft) with a nominal operating liquid  depth of
4.7 m (15.5 ft) (Figure 8-5).

Primary effluent is pumped to the secondary  facilities
and  split  among  four  of  the  six  aeration  tanks
(Aeration  Tanks 5  and  6 have  never been  in
operation). Each 6,815-m3 (1.8-mil gal)  aeration tank
is divided into four passes of equal size. The normal
mode of operation  is step  feed. Return activated
sludge is fed into the head  end of  Pass  1  of each
aeration tank, and primary effluent  is  normally split
equally among Passes 2, 3, and 4 (Figure 8-5).

The air supply  for  the  activated  sludge  process is
furnished by  one (or more)  of three identical rotary
vane blowers. Each blower  is  rated at a maximum
output  of 28,320 L/s (60,000 scfm)  at  153 kPa (7.5
psig). Each blower is driven by a 2,240-kW (3,000-hp)
motor.  The airflow rate generated by the blowers is
adjusted by the  positioning of the inlet guide vanes on
the suction side of each blower.

The original aeration equipment consisted of  Chicago
Pump  "Defieetofuser" coarse bubble  diffusers with
large 1-cm (3/8-in) diameter orifices  on  the periphery
of  the  diffuser.  Approximately 250 of these diffusers
were installed on each  of seven  drop  pipe/manifold
assemblies per  pass. A spiral roll aeration and mixing
pattern was established by this design geometry. The
SOTE of the  "Defieetofuser" system at Hartford was
estimated to be  6-7 percent.
                                                 225

-------
Flguro 8-S.   Secondary treatment process schematic - Hartford, CT.
                                                                                                  Secondary Effluent
i
t
\
\
to>
to>
i
sax — »- -
^— -*— -^s
g^-to. N°- O
c


-«*
j ,
i ,
^t " 	 jf V
^*— Sluice Gate (Manual) — ^ ^''•^

»•
»»
1 1

** ~->t3
&>, 	 if ^
V.NO-4^^^
J

•^
j
•9
t
i '
1
»»•
t
fr-
gs^. 	 .,»,. —
V. -* 	 ,03
_ NO. B y
&$, — to- -^
c ^

-*
j
^
J
1

>Mam Influent Channels
f
*>
i
•a*-

c
&N — IP- -s.
No, 1^,^42
By* ^ —
i
1
\
. ^- Motor Operated Sluice Gate (typ.)
to-
^ -.
.. ^ — Xs
^ 	 *J*>- 3 -^
^ 	 ^g
\
•«
(
•*
1 1
1 ]
#-
!
r
g^— *- -^
— .No. 5^^__x4a
gy~- »- —
•«•
!
-«
I
T


P
1 E



Aeration Tanks ^^"^
	 _ _ ^ 	 __ 	 r
                                                                                                                Primary
                                                                                                                Effluent
                                                                                                    Secondary Efflueril
                                                           226

-------
The  air supply control system is designed with  flow
control  and flow indicators for  each pass  of  each
aeration  tank.  A  DQ  probe senses  the  DO
concentration  in the  pass. The signal is fed to an
analyzer and then to an electronic controller where the
DO  is  manually  set. The output  signal  from the
controller modulates the 30-cm (12-in) motor-operated
butterfly valves that meter the correct amount of  air to
match the set-point DO that has been manually set.

From start-up in 1972, one blower always operated at
full capacity. Yet DO demand in  the aeration system,
particularly at the inlet points, was not met during the
summer months. The only operational solution was to
turn  on a second 2,240-kW  (3,000-hp)  blower.  This
was  a costly proposition for providing a small  amount
of additional air.

Basis for Changing to Fine Pore Aeration
Total plant power  costs  had  increased steadily from
about $30Q,OQO/yr in 1973 to over  $900,000/yr in
1979. Between 1979  and 1982, a large  decrease in
energy usage was realized by the upgrading of sludge
handling and  treatment equipment  from  coil filters to
belt  filter  presses  and the  initiation  of  a   new
incinerator operating  mode. Despite these electrical
cost reduction improvements,  total  plant  electrical
costs continued to rise at an alarming  rate.  Energy
efficient aeration equipment needed to be considered
on ice this equipment accounted  for about 60  percent
of the energy used at the plant.

Over a 5-yr period beginning in 1978, MDC engineers,
with  the assistance of equipment manufacturers and
consultants, developed a system that was compatible
with  existing facilities  and was believed cost effective
and  efficient.  Ceramic domes and discs were chosen
for pilot testing because they appeared  to have the
greatest  OTE and satisfied the compatibility
requirements  of tank geometry and  the air supply
system.

The  pilot  tests  indicated the possibility  of  a 50-60
percent reduction  in total air  supply requirements for
aeration. Total estimated  air supply  requirements were
set at  8,495-10,385  L/s  (18,000-22,000 scfm), which
included air to suspend solids in the influent channel.

With one  blower  normally operating at maximum
capacity of 28,320 Us (60,000 scfm)  for the  existing
aeration system, the next concern  was to investigate
the  turndown capability  of the  existing  blower
equipment. It was  determined that each  blower could
be turned  down  to  below 4,720 L/s (10,000 scfm)
without  surging; however,  operating  efficiency would
diminish as airflow was reduced.

The  original air piping system at  the Hartford  plant
was  constructed  of spiral  welded  steel  and wrought
iron  pipe.  Inspection  revealed:  1)  the  bituminous
epoxy coating in the air  mains and suction lines was
in  excellent condition  and would be  suitable for the
fine pore system without  repair, 2) the coating  in the
30-cm (12-in) air pipes in  Aeration Tanks 1, 2, 3, and
4 was in good condition (paradoxically, the coating in
the same air pipes in  Aeration Tanks 5 and 6 (never
operated) was damaged and  contained rust  and
scale),  and  3) the  15-cm  (6-in) drop  pipes  and
manifolds  to  which  the  "Deflectofusers"  were
connected were rusted, and the drop pipes could not
be used with the fine pore aeration system.

Fine pore ceramic domes  would require removal of 95
percent  of alt particles >0.3 micron  in size in the air
supply.  The existing  automatic oil bath filters  were
capable  of  removing  only  25-30  percent of  the
particles and, therefore, were not usable alone for the
fine pore system. "Biocell" filters,  which could  be
installed inside  the  existing  inlet  plenums,  were
selected.

Oxygen  transfer projections and energy requirements
for coarse and fine pore diffuser systems over a 20-yr
period were  examined. Minimum air requirements for
the proposed fine pore  dome system would  be
dictated  by  the  air needed to keep the MLSS in
suspension. Peak air demand  for the domes could be
supplied by  one  existing blower  throughout  the
planning period,  while peak air demands  for the
existing coarse  bubble system  would  require
simultaneous operation of two blowers during  summer
months. The total system head on the blowers  with
the new system was expected  to be well within the
capacity of the existing equipment, and blower surging
would not be expected to occur for the  expected air
demand.

In  1981, the estimated initial cost of all recommended
improvements  to  retrofit  to fine pore  domes was
between  $1,115,000 and  $1,830,000  with  an
estimated simple payback  period of 3-6 years.
HI  FINE  PORE  AERATION  RETROFIT  DESIGN
DESCRIPTION
The  cost-effectiveness  study for Hartford included a
complete  design  review  for  sizing  the  fine  pore
aeration system to meet process needs through the
year 200 1 . The projected food-to-microorganism (F/M)
loading of the  retrofit was  0.2-0.3 cH, which resulted
in a  ratio of  oxygen required:BOD5 removed of about
1.0 kg 02/kg BOD5 based on design criteria in WPCF
MOP No. 8.

Average daily oxygen requirements at field conditions
(OTRj) were  projected to be 33,115 kg (73,000 lb)/d in
the year 2001 during  the  period  May-October,  and
21 ,230 kg (46,800 lb)/d the rest of the year. For  1 982
conditions, the expectation was  for OTRf values of
22,635 and  13,700  kg  (49,900 and 30,200 lb)/d  for
May-October  and  the  remainder of  the  year,
respectively. The  standard oxygen  transfer  rate
(SOTR) was calculated using:  1)  a = 0.75 for  the
                                                 227

-------
ceramic  fine  pore domes  (0.85  was  used for the
existing  coarse  bubble  diffusers),  2) 6 = 0.95, 3)
operating DO  of  2.0  mg/L, and  4) average  mixed
liquor  temperature of  15°C (59°F).  a was  selected
from  literature information. Average  SOTEs  were
selected as 29 percent for the fine pore discs and 6
percent for the existing coarse bubble diffusers.

The resulting  projected air  requirements  for  1982
conditions  including nitrification  were 32,145 Us
(68,100 scfm)  using the existing system and 12,270
Us (26,000 scfm) using the fine pore dome system.
Airflows  without  nitrification  were estimated  to  be
21,665 Us (45,900 scfm) for the existing system and
9,675  Us  (20,500 scfm)  for  the fine pore  dome
system.  An additional  5,665 Us  (12,000 scfm)  was
also allocated for mixing in the influent channel.
Four of six aeration tanks originally placed in service
in 1972 would continue  to be  used after the retrofit,
No modifications to the tanks or to the process liquid
piping  would  be necessary. Only in-tank air piping
would be changed. The mode of operation both before
and after retrofit is step feed. Pass 1 of each aeration
tank is used  for reaeration of the return activated
sludge. Normally, one-third  of  the primary effluent  is
fed into the head end of each  of the remaining three
passes.

The layout and design of the fine  pore dome diffuser
system was planned so  that most of the existing air
piping could be utilized without modification. The  new
submerged  air distribution  piping and diffuser  grid
piping  was designed to  facilitate  an  easy  and
economical installation.

The acceptable fine pore diffuser system had  to have
a guaranteed  SOTE of  28  percent  at 0.24  Us  (0.5
scfm)/diffuser, 26 percent  SOTE at  0.47 Us  (1.0
scfm)/diffuser, and 23 percent  SOTE at 0.94 Us  (2.0
scfmj/diffuser  when operated at  the average design
diffuser density  in 4.7 m (15.5 ft) of clean water.  In
addition,  a  minimum  airflow  of  0.24 L/s  (0.5
scfm)/diffuser  was  established  to  prevent water-side
fouling. Further,  a  minimum mixing  requirement of
0.61 L/s/m2  (0.12  scfm/sq ft) of tank  bottom  was
established together with a minimum diffuser spacing
of 61 cm (2 ft) on  center. Certified oxygen transfer
and mixing test results were required  as part of the
bid documents of each manufacturer submitting a  bid.
Diffuser performance data are presented  as  part of
Figure 8-6.
It  was  planned that Aeration Tanks  1  and  2  be
retrofitted  first and then  placed on-line  before
completion of the retrofits in  Aeration Tanks 3 and 4.
Tanks  1  and  2 would  then be  field tested and
additional diffuser  density modifications incorporated
into Tanks 3 and 4,  as needed.
DO measurements were taken in Aer,ation Tanks  1
and 2 soon after being placed  in operation. The DO
profiles indicated that more diffusers were  needed at
the influent end of the passes. Redistribution of the
diffusers was easily accomplished  in Aeration Tanks 3
and 4 by use of the spare saddles specified in the
design. Final diffuser density and  quantity information
by grid and pass are contained in Figure 8-7 for one
aeration tank.

No  blower air  handling capacity  modifications  were
planned as part of the aeration system  retrofit. The
existing 28,320-L/s (60,000-scfm)  rotary vane blowers
were  found to be suitable for  use with the new fine
pore diffuser system. Although preretrofit airflow from
a single blower  had been at the  28,320-Us (60,000-
scfm) level, the  new fine pore  aeration system  would
require less than 14,160 Us  (30,000 scfm).  By inlet
guide vane adjustment, turndown  of the blowers was
possible to a surge point of approximately 3,775 Us
(8,000 scfm). The only cost of turndown  was in the
reduced efficiency  of  the blower  at lower airflow
output rates.

Norton  dome diffuser  equipment  was  installed  in
Aeration Tanks 1-4. Each diffuser is an 18-cm (7-in)
diameter porous ceramic dome  secured  to a  PVC
plastic pipe saddle by a dome orifice bolt. Air emerges
from  the 10-cm  (4-in)  grid piping  network at  each
saddle location  up through a hollow plastic dome boit
that contains a 5-mm (13/64-in) diameter orifice  in the
bolt sidewall.

IV.  OPERATIONAL  PERFORMANCE   AND
EVALUATION
By November 1982, installation of the entire project
was complete. With all  four aeration tanks on-line with
the new fine pore aeration  equipment,  air  usage
dropped to 1,020,000 m3  (36,000,000 cu  ft)/d,  down
from  1,810,000  m3 (64,000,000 cu ft)/d  prior to the
retrofit, and power fell  to a level of 22,000 kWh/d,
down  from  31,000 kWh/d.  For the remainder of  1982,
the system was operated at this level.

The  automatic  control  system did not perform as
planned. Blower discharge pressure was too large to
deliver  air  to all passes,  and air  supply to  some
passes was reduced to levels lower than the minimum
recommended values  required for solids suspension
and estimated to be necessary to prevent water-side
fouling of the ceramic diffusers.

The blower discharge pressure  was 151-153 kPa (7.1-
7.4 psig) in the automatic air control  mode.  Airflow
control was switched to manual operation  to alleviate
the high discharge  pressure.   The  blower discharge
pressure was adjusted to 149 kPa  (6.8 psig).

Foaming of the  aeration tanks worsened after start-up
of the fine  pore system. It was determined that high
SRTs (>3-5 days) caused foaming to increase above
                                                  228

-------
Figure 8-6.   OTE characteristics - Hartford, CT.
  OTE, percent

     35  r-
     30
    •25
     20
     15   —
     10
                       LACSD Data
                                                        AERTEC Data
                [Dome diffuser SOTEs based on 15.5-fl
                  SWD and 14.5-ft air release depth]
                                                                                *  Specified SOTE for
                                                                                   dome/disc diffusers
                                                    Specified uF(SOTE)
                                                       fuF = 0.75]
                         Expected aF{SOTE)
                             [aF = 0.55]
                                   Dome Diffusers
            Dome diffuser average ..•
            aF(SOTE) results from  \
                off-gas testing
                                            11/85
4/87
                                       "6/87
                                                                i 2/87
                                                        3/86
                                                                                   8/87 i
                                                                          SOTE  = 6.25
                                                                                               • 7/86
                          Onyinal Equipment   •;;;	

             ("Deileciofusers" at 13.0-tl air release deplh
               and 15.5-fl SWD, spiral roll configuration]
            uF(SOTE) = 4.4
              faF =  0.70]
                                                      Airflow Rate, sofm/diffuser
                                                            229

-------
Flgura 8-7.  Typical diffuser layout for one aeration tank - Hartford, CT.
                194 ft
          Pass 	
          Domcs/Pass
RAS -  Rolum Activated Sludge
PE   -  Primary Effluonl (ono-third/feed point)
EFF  -  Ellluont from Aeration Tank
* Domtos Number of OiHusors per Grid
t Donates Oilftisor Density, diffusers/100 sq ft
EFF
1
71*
(12.8)f
76
{13.7}

8?
(15.6)
t
87
(15.6)

98
(17.5)
134
(22.2)
136
(24.4)
^
t
PE
4
679 H
PE
1
r
147
(26.3)
140
(25.0)

115
(20.8)

103
(18.5)
\.
97
(17.5)
95
(17.2)
96
(17.2)
J


3
793
20ft
(typ.)
83
(14.9)
95
(17.2)

97
(17.5)
t
103
(18.5)

115
(20.8)
140
(25.0)
160
(28.6)
^.
f
PE
2
793
RAS
1
209
(37.0)
219
(40.0)
\
186
(33.3)

160
(28.6)

100
(17.9)
95
(17.2)
95
(17.2)
J


1
h 1,064
                                                                                Grid
                                                                               3,329
an  acceptable level. Short-term  remedial  action  to
reduce  foam consisted of reducing the mixed  liquor
DO concentration to near zero or zero. Foaming had
been noticed in the 1979 pilot tests  of the fine pore
diffusers.

MLSS concentration, with a design design range  of
2,000-2,500  mg/L,  varied  widely  after  retrofit
implementation through the period of  off-gas  testing.
Concentrations  as  high  as  6,000  mg/L  were
measured,  and the average  MLSS  level was much
greater  than  the  upper limit design  value  of 2,500
mg/L. The elevated  mixed  liquor solids inventory
resulted  in increased  oxygen  demand  and  energy
consumption. The foaming problem brought on by the
higher MLSS values and  resulting higher SRTs also
appeared to adversely affect OTE.

Following the  change  to manual airflow control, the
established mode  of operation  was  to leave  all
                                                   230

-------
aeration  pass control  values  fully open,  allowing all
diffusers to  operate at approximately  equal rates
throughout  the system,  if  DO  concentrations
increased in some aeration passes or tanks, air supply
throttling was initiated  at the individual aeration pass
or  passes  having  the  greater DO  concentration.
Throttling was not continued to a point  below which
airflow to a  pass would be less  than the  minimum
airflow  to  achieve  mixing  and  solids  suspension
(approximately  0.16  Us/m3 [1.3 scfm/1,000 gal]} or
less than  0.24  Us  (0.5  scfrn)/dome. Plant effluent
quality   remained  consistently  high  after
implementation  of fine  pore aeration.  Effluent  BOD5
and TSS  concentrations  nearly always remained
below 10 mg/L.

Aeration Performance Evaluation
Table 8-13 summarizes aeration performance data by
site visit for Aeration Tank  2. From 36  to 48 individual
off-gas  tests were conducted  during  each  site visit.
The whole tank average values are based on  seven
site visits from  November 1985 to August 1987 and
represent the summary of  over 340 individual tests.
During each  site visit,  each aeration pass was  tested
at the influent, middle, and  effluent grids.

These data are plotted in  Figure 8-8 for the average
whole  tank  test results by site  visit. The  average
expected SOTE for the  Hartford design was 27.5
percent, and the  average  whole  tank aF(SOTE), as
measured by off-gas testing, was 10.0 percent.

Average uF for  the whole tank was 0.37 with a range
of  0.29-0.45.  Usually, the very low values  were
measured  at the influent feed  point  sample locations
in any of the aeration passes.

The SOTE of the original coarse bubble spiral  roll
aeration  system was estimated to be 6.25  percent.
The efficiency of this aeration equipment is reduced to
4.4 percent  [aF(SOTE)]   when  an  aF of 0.7  is
assumed.  The resulting OTE in mixed liquor with a
DO concentration of 2.0 mg/L is  3.2  percent for  the
coarse bubble system.

The average  aF(SOTE)  for  all  whole-tank tests
conducted over the 2-yr study period on the fine pore
aeration  system in  Aeration  Tank No.  2 was 9.97
percent,  with  a  range of 8.2-12.6  percent. This
represents 2.25 times  the transfer  efficiency  of  the
original   equipment.  The ratio  of  2.25 is in general
agreement with  the  ratio of airflows before  and after
retrofit.
'Effect of Cleaning on Performance
 Prior to the beginning  of  off-gas -testing in Aeration
 Tank 2  in the fall  of 1985, the  tank was dewatered
 and the diffusers  cleaned. The  other three aeration
 tanks were also cleaned in the fall of 1985 for the first
 time. Only  Aeration Tank  2  was cleaned  a second
time - in May 1987. There was no performance basis
used to initiate cleaning in May 1987.

The cleaning method used both times is known as the
"Milwaukee  Method"  (see Table  4-1), which uses
hosing and acid application. A high pressure water jet
is applied to the diffuser surface  (air on) followed by
acid spraying  (air  off)  with  ZEP,  a commercially
available cleaning compound with  an  HCI content of
22 percent, and brush scrubbing after letting the acid
soak  for 30  minutes.  A  second hosing is then
performed (air on) to remove any solubilized  foulant
and residua! acid.

Immediately after cleaning the diffusers, air distribution
and leak testing was  conducted  prior  to placing  the
aeration  tank  back  on-line. Plant  effluent  was
introduced to the aeration tank until the diffuser  grid
was submerged by 51-76 mm (2-3) inches of liquid.
Airflow  was adjusted  to approximately 0.24 Us  (0.5
scfm)/diffuser,  and observations were made for proper
air distribution and  leaks. All gasket, dome bolt,  and
other air leaks were repaired throughout the tank,  and
any  leveling  of,  or  repairs to, pipe  supports  was
accomplished at this time.

Off-gas test results in  Aeration Tank 2 before  and
after the May 1987 cleaning are summarized in Table
8-14. Although aF(SOTE) values were  higher in Pass
2  after  cleaning  than  before,  lower OTEs were
observed in Passes 1, 3, and 4 after  cleaning. As  a
result,  total tank average performance  results  for the
after-cleaning  tests were  actually lower than those
recorded before cleaning. The before-cleaning results
were the highest aF(SOTE) values measured during
the entire test period since the initial tests  conducted
in November 1985.

Although  diffuser  cleaning  may have  little effect on
OTE at Hartford, the cleaning of diffusers on a routine
basis is beneficial. It removes built-up deposits  of both
inorganic and organic materials that cause increased
backpressure.  Additionally, diffuser cleaning facilitates
an  inspection  of  the aeration equipment and  the
undertaking  of  air  distribution  and leak tests.  Any
necessary repairs to limit gasket and other leaks  and
the performance of any other repairs constitutes good
maintenance practice at the time of tank cleaning.

The average  downtime  required  to clean each
aeration tank was 1 week, including tank draining  and
filling time. No adverse effects on  plant effluent quality
occurred  during  cleaning.  However, if peak  loading
had  occurred  during  cleaning, effluent quality could
have been reduced due to reduced retention  time in
the aeration process and possible diffuser air capacity
limitations.

V. ECONOMIC CONSIDERATIONS
The estimated initial cost of the retrofit based on the
consultant's estimate was  between $1,115,000  and
                                                  231

-------
Figure 8-8.  Oxygon transfer performance for Aeration Tank 2 - Hartford, CT.
        35-
        30-
        25-
        20-
        15-
        10-
         5-
                                                          Average SOTE
                                                                                  n
                Dillusofs
                Cleaned
                10/15/85
                                   Average oF(SOTE)
Diffusers
Cleaned
 5/1/8?
                   I
| 1386
1 1 1
3 4
I
8
1987
1 1 1 I 1 1 1 1 1
12 16 20 24
1
28
           t
8,'1/85
                                               Elapsed Time, months
$1,830,000. The actual total initial cost was less than
$600,000, completely installed.

The installed cost was less than $50/diffuser, including
modifications to the instrumentation and additional air
filtration. In-tank diffuser equipment and piping costs,
alone, probably represented  about half  of the  total
cost of the project.

Annual operating savings were over $200,000 for the
first year of operation, yielding  a simple payback
period of less than 3 years. A daily power reduction of
about 12,000 kWh was realized, and the electrical rate
in 1933 was about $0.05/kWh. Similar  savings in the
cost of electricity have been observed for succeeding
years.

Annual maintenance  costs for the  fine pore  system
have  not increased  significantly  over maintenance
costs for the  coarse bubble  system. The ceramic
dome diffusers in all four aeration  tanks were cleaned
In late 1985. Up  until  that time, no cleaning or other
maintenance had  been  performed  on  the  fine pore
system. The cost to clean each aeration tank in 1985
                                        was approximately  $4,500,  or $18,000 for  all four
                                        tanks.  The  breakdown  for cleaning the diffusers  in
                                        each tank is presented in Table 8-15.

                                        The cost of replacing domes,  gaskets, and bolts with
                                        new equipment would be approximately $35,000 per
                                        aeration tank. In addition,  removal  of grit, debris,  and
                                        sludge could cost  an additional $15,000  per tank.
                                        Total  rehabilitation  of the fine pore diffuser  system
                                        should not be required more often than every 5 to 8
                                        years. However, gaskets and plastic bolts may require
                                        replacement as often as  every  3 years.

                                        Reductions in blower power consumption achieved by
                                        replacing spiral  roll  coarse bubble aeration equipment
                                        with full floor coverage  fine pore dome/disc  aeration
                                        equipment, such as at Hartford, are predicated on the
                                        ability to turn  down and/or shut down  blower units
                                        after retrofitting  with  the  new equipment. Historical
                                        blower power consumption for the  ceramic  dome
                                        diffuser system  in Aeration Tank 2  is plotted in Figure
                                        8-9 for over 6 years  of operation, along with airflow
                                        and BODs removed (BODR), all normalized for plant
                                        flow.
                                                   232

-------
Table 8-13.  Tank Average Off-Gas Results (Aertation Tank 2 ) - Hartford, CT
Dale
11/12/85
3/24/86
7/14/86
2/4/87
4/22/87
6/18/87
8/13/87
Test No.1
1A
2A
3B
46
5B
6B
7B
IVIJA*JU
Liquor
Temp., "C
18.9
13.9
22.5
13.1
14.7
21.6
24.7
uF(SOTE), ,
percent
12.60
8.18
9.40
9.00
11.36
9.35
9.88
Airfloti
per dome
0.96
1.28
2.40
1.49
1.41
0.85
2.20
v, sefm
total
3,195
4,261
7,994
4,953
4,674
2,819
7,309
New SOTE,
percent
28.2
28.2
25.4
27.3
27.5
28.6
25.6
aF
0.45
0.29
0-37
0.33
0.41
0.33
0.39
aP(SOTR),
Ib O2/hr
417
361
778
462
550
273
748
New SOTR,
Ib O2/hr
932
1,248
2,107
1,401
1,332
836
1,942
  i "A" lets designate 4 replicate tests per sample location and 1 pass through the aeration lank.
   "B" lets designate 1 replicate tests per sample location and 3 passes through the aeration tank.
Table 8-14.  Average Off-Gas Test Results Before and After
             May 1987 Cleaning (Aeration Tank 2) - Hartford,
             CT

Primary effluent BODS, mg/L
Final effluent BOD6, mg/L
Plant flow, ntgd
MLSS, mg/L
SRT, days
Entire Tank, Average uF(SOTE),
percent
Entire Tank, Average aF
Pass I, Average uF(SOTE), percent
Pass I, Average aF
Pass 2, Average uF(SOTE), percent
Pass 2, Average uF '
Pass 3, Average uF(SOTE), percent
Pass 3, Average uF
Pass 4, Average uF(SOTE), percent
Pass 4, Average aF
Before
Cleaning
(4/22/87)
49
4
62.3
3,200
7.9
11.36
0.41
11.43
0.40
9.80
0.36
11.63
0.43
10.87
0.41
After
Cleaning
(6/18/87)
112
9
47.2
4,500
9.7
9.35
0.33
8.66
0.30
11.65
0.40
10.40
0.36
7.55
0.27
Table 8-1 S.
     Item
Diffuser Cleaning Costs - Hartford, CT
                                      Cost, $
                                                                 Labor (100 man-hr, incl. overtime)                      2,000
                                                                 Chemicals (ZEP cleaner)                                250
                                                                 Cleaning equipment and protective clothing                750
                                                                 Spare parts (domes, bolls, gaskels, pipe hangars, elc.)    1,500
                                                                     .Total Tank                                     4,500
                                                            233

-------
Figure 8-9,   Power consumption profile for Aeration Tank 2 - Hartford, CT,

          1,3
                                                               Power, kwh/a/mgd

                                                             BODR,lb/d/mtjci
                 1982
I I I Mill I I | I I I I I  I I I I
   1984            1985
         Time, months
                                                                                          I I I I 1 I 1 I  I I 11 I I I I I I 1
                                                                                              1987
                                                                                                            1988
                                                        234

-------
   8.2.5 Jones Island Wastewater Treatment Plant

   .LOCATION: Milwaukee, Wisconsin

   OPERATING AGENCY: Milwaukee Metropolitan Sewerage District

   DESIGN FLOW: 8,765 Us (200 mgd)

   WASTEWATER: Domestic Plus Industrial

   ORIGINAL AERATION SYSTEM: Ceramic Plate Diffusers

   FINE PORE AERATION SYSTEM: Ceramic Plate Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION: 1925

   BASIS OF PERFORMANCE EVALUATION: Air Usage; Off-Gas Testing

   CLEANING METHOD: Various Methods, Including Water Hosing, Sand Blasting, and Acid Washing
I. HISTORICAL BACKGROUND
The Milwaukee Jones  Island wastewaler treatment
plant  has a long history of using fine  pore  ceramic
plates in its activated sludge facilities. The West plant,
with an initial design capacity  of 3,725 Us (85 mgd),
was placed in operation in 1925 and consisted of 24
aeration tanks 71.9 m long by 13.4 m wide (236 ft x
44  ft) with a 4.6-m (15-ft) SWD. The aeration tanks
are operated as a two-pass  system.  Ceramic  plate
diffusers  in a  ridge-and-furrow  configuration  were
selected for the original design. Containers that hold
nine  30-cm  (1-ft) square porous  ceramic  diffuser
plates were placed across the width of the  aeration
tanks between  the  ridge  separators.  One  row of
containers also ran the  length of each tank and held
seven diffusers per section. The ratio of floor area to
diffuser area  was 4:1, yielding a 25-percent coverage
of the floor surface with plate diffusers.

The original  ceramic  plates used in  the West  plant
were  Filtros fused silica 38-mm (1.5-in) thick plates.
Norton  Alundum 25-mm (1-in) thick  plates  replaced
the Filtros plates in several West plant aeration tanks
(North battery) shortly after the plant  was plaped in
operation.  Eventually, these Alundum plates  were all
replaced with Filtros fused silica plates. The original
Filtros fused silica plates installed in 1925 in the entire
South battery of the West plant were still in service as
of February 1989.

The  East  plant was constructed  in  stages as  an
addition  to the  West plant, with the first  addition in
1935. Final additions in 1952 brought the total nominal
treatment capacity for the Jones Island plant to 8,765
Us (200 mgd). The first 12 of 20 East plant aeration
basins  installed in 1935  contained a  plate  diffuser
layout along one side of the aeration tank to provide a
spiral roll  mixing  pattern. Each aeration tank  was
112.8 m long by 13.4 m wide (370 ft x  44 ft) with a
4.6-m (15-ft)  SWD.  These basins were also  operated
initially in a two-pass flow mode so that each channel
was 6.7 m  (22 ft) wide. The diffuser design consisted
of two rows of Norton Carborundum square alumina
plates 46 cm (18 in) apart  with 1,296 plates in each
aeration pass.  In 1942 and  1943, an additional  row of
plates was  added to each of the 24 aeration passes,
increasing the  number of diffusers per pass to 1,944.
The permeability rating of these plates was 32-36.
An  additional  expansion  completed  in  1952  added
eight aeration tanks to the East plant with four rows of
plate diffusers  along one side of the tank. These were
all Filtros  fused silica plates  with  permeabilities  of
19.5-24. At this time, the total  flowsheet  consisted of
bar screening,  grit removal,  and fine screening prior to
activated sludge treatment.  The West plant contained
15 secondary  clarifiers with  4.6-m (15-ft) SWDs. The
diameter was 29.9 m (98 ft) for 11 of the clarifiers and
13.0 m (42.5 ft)  for the other four clarifiers. The East
plant contained a total of 10 secondary clarifiers 25.6
m wide by 49.2 m long (84 ft x 161.5 ft) with a 4.3-m
(14-ft) SWD.

In 1957, the East plant aeration system  was modified
by replacing the spiral roll, ceramic plate diffuser
system with a  tapered aeration, ceramic .tube diffuser
system. Prior to this, in 1956, saran-wrapped diffusers
on fixed  headers were installed on an  experimental
basis:  Due  to a greater clogging problem  with  the
saran-wrapped diffusers, ceramic tubes were selected
for the tapered aeration   modification. Forty-two
headers at 5.3-m (17.5-ft)  intervals  were installed in
the two-pass aeration tanks. The number of tubes  per
header decreased incrementally from 33 to 12 in  the
first 21 headers. The next eight headers  had 12 tubes
each, the  next had 11  tubes each, and the final  two
headers contained 15 tubes each, for a total  of 711
tubes per tank.  Approximately  equal  quantities of  air
were  applied to the two plants,  averaging  10.2  m3
air/m3 wastewater flow (1.36 cu ft/gal).
                                                  235

-------
At the time of the 1957 diffuser system  modification,
East plant return sludge was separated  from that of
the West  plant.  Both  plants were operated  with
independent return sludge systems after July 1958,
However, East plant waste sludge was directed to the
West plant return  sludge line  for ultimate  sludge
wasting via the West plant.

As  the  conversion to the tapered aeration  ceramic
tube diffuser system progressed  in 1958-1960,  East
plant performance decreased. Annual average effluent
BODs increased from 19.6 mg/L in 1958 to 30.7 mg/L
in 1960 to 59.8 mg/L in  1963.

A plant-scale research project was initiated in 1961 to
study the effect  of  various plate  diffuser  placement
patterns. This ultimately led  to conversion of all  20
East plant aeration  tanks to a five-row  longitudinal
diffuser  placement  pattern  utilizing  nine
dilfusers/container with a total of 10 containers, or 90
diffuser plates on each downcomer, A total of 3,150
plates was installed in each aeration tank.  The tanks
continued to operate in the  double-pass  mode. The
ratio of theoretical tank surface to  plate  surface was
5.17:1  compared to 4:1 provided in the original ridge-
and-furrow design of the West plant. By 1965, all East
plant  aeration basins had been  converted  to  this
ceramic  plate  design.  The   effluent   BODs
concentration decreased to  10-20 mg/L  as  the
conversion progressed.

In 1983, the East plant basins were rehabilitated with
new diffuser plates and  air  piping.  The  fine pore
longitudinal  diffuser  arrangement  was  retained. A
trench was also provided along one side of the tank to
collect flushed solids during cleanup operations.

The new diflusers were Norton ceramic plates, 30 cm
x 30 cm x 25 mm thick (12 in x  12 in x  1 in). The
permeabilities of these plates is 17-23. The plates are
grouped by permeabilities in  ranges of 17-19, 20-21,
and 22-23.  Each downcomer is fitted with plates of
only one range.  The plates are grouted into concrete
containers placed flush with  the bottom  of the tank.
Each  container has  nine  plates and is connected at
ona end to  a 25-mm  (l-in)~"diameter air  pipe. The
containers are placed end-to-end  in the direction of
the tank length,  32 containers in a row and five rows
across the basin width, yielding 1,450 plates per pass
and 2,900 per two-pass basin.

Air  is supplied from  downcomers to three separate
zones in each pass. The total number of plates is 558
in Zone 1 at the  pass inlet, 450 in Zone 2, and 446 in
Zone 3. The ratio of tank  floor area to diffuser plate
surface area is   5.62:1, i.e.,  the  plates cover  17.8
percent of the tank floor area. Airflow to each zone is
set manually by a butterfly  valve, but,  within each
zone,  there  are no  orifices  for  control  of  air
distribution.  Each downcomer is equipped with  an
orifice  meter, and  the  airflow corresponding  to  the
pressure drop is read from a portable flow indicator, A
permanently-mounted orifice meter and flow indicator
for each pair of  passes  (six zones)  provides a check
on the  sum  of  the readings taken  at the  separate
zones.

Although the  number of diffuser plates in  each of the
three zones  is  different, diffuser density  is  uniform
down the length of each pass because the length  of
each zone  is different.  Airflow can be varied to each
zone to achieve a tapered  aeration pattern.  Both
tapered and  uniform  aeration  patterns  have  been
utilized.

In June 1985, the East plant basins were  converted
from two-pass  to  single-pass operation.  An off-gas
testing program  was carried  out from June 1985  to
June 1988  in the North and South passes of Basin  6,
each operated  as a  single-stage  system.  Prior  to
initiating the  testing  program,  the diffusers  were
hosed.

II. PLANT PERFORMANCE
From 1961  through 1964, oxygen transfer tests were
conducted  for a variety of diffuser configurations  in
the East and West plants.  The test method was an
early version  of the off-gas  testing procedure. A hood
was  used to collect gas  leaving the liquid surface, the
oxygen content of the gas was measured by the Orsat
method, and gas flow was measured as the rate that it
displaced water in a gas accumulation tank.

A summary of measured aF(SOTE) values  for Jones
Island is presented in Table 8-16. From 1961 to 1963,
the efficiency of the tube  diffusers in the East plant
decreased.  After  a  program  of cleaning  and
refurbishing, the spiral  roll  tube  diffuser  aF(SOTE)
increased to  7.6 percent  in 1964.  The  ridge-and-
furrow  plate  diffuser  system  in the West  plant
produced a 20.9-percent aF(SOTE)  in 1964 after  a
program of  cleaning was carried out. This was the first
cleaning  of these  diffusers in  10 years.  The,West
plant longitudinal  plate   configuration   produced
aF(SOTE)s within the same range as the ridge-and-
furrow system. In 1985 and 1988, after more than 60
years   of  operation   and  scheduled  to be
decommissioned, the ridge-and-furroW  West  plant
diffusers  were  performing comparably  to  their
efficiency prior to cleaning  in 1964.  The  refurbished
East plant continues  to demonstrate  the  overall
performance  of plate   diffuser  plants.  Table  8-17
summarizes East plant  average  annual performance
from 1970  to  1986. Similar  effluent  BOD5
concentrations were also observed for the West plant.

After single-pass operation began  in the  East  plant
and  after cleaning the diffusers,  periodic  off-gas
testing was conducted  in  Basin 6 over a  30:month
period from 1985 to  1988. The  gas collection  hood
used had dimensions of  61  cm (2 ft) by 5  m (16.5 ft).
Twelve  test stations were located equal distances
                                                  236

-------
Table 8-16.  Summary  of Oxygen Transfer Tests - (Jones
           Island) Milwaukee, Wl
Location
Easl Plani, Spiral Flow, Tubes
1961
1962
. 1963
1964
West Plant, Ridge and Furrow, Plates
1961
1962
1963
1964
Easl Plant, Transverse Plates
1964
West Plant, Longitudinal Plates
1964
West Planl, Ridge and Furrow, Plates
1985 and 1987
East Plant, Tapered Full Floor, Plates
. 1985 and 1988
No. Tests

6
7
7
6

4
5
6
6

20

21

21

30 , ,
Average1
uF(SOTE),
percent

7.9
5.4
' 4.8
7.6

15.8
11.7
12.9
20.9

13.8

15.8

11.9

16.4
 1 aF(SOTE) based on effective clean water DO saturation of 10,5
   mg/L.
along the tank length, yielding a gas sampling area of
5 percent of the  basin  surface area.  The North and
South pass sample locations  in  Basin  6  were
alternately  tested to  obtain  data  under  similar
conditions for each pass. The hood  was positioned
across the width  at  about the center of the basins.
The  total  test  time  to  complete  sampling  of both
passes was  about 6 hours.

Table 8-18  contains  a  summary  of  the  East plant
operating  conditions and  average aF(SOTE)  values
obtained for  both passes in Basin 6. The average SRT
reported for the aeration test days for the East plant
was  3.8 days and ranged from 0.3 to 5.3 days. The
F/M loading  averaged 0,65 d-V

III. DIFFUSER  CLEANING
A variety of  cleaning methods have been used at the
Jones Island plant,  including  high-pressure  hosing,
sand  blasting, "and acid washing.  At  ,the -Milwaukee
South Shore plant,  these cleaning  methods  were
evaluated  in more detail from 1981. to 1985  and an
additional method was tried. This  is  called  the
scarifying  method  and involves cleaning  the  diffuser
plates with  a  powered wire brush.  Muriatic acid
treatment cleaning followed by a water wash has been
shown to  be effective at the South Shore plant, and
this is known as the "Milwaukee Method" of  diffuser
cleaning.

Cleaning frequency at the East plant was variable and
ranged from one  to eight times over  a  16-yr period.
Acid  washing was done in some tanks in 1980 and
1981. For the West plant, no tanks  have been cleaned
more than five times in a 22-yr period and the average
number is 3,3 cleanings in 22 years.

The  diffusers  in  East  plant Basin  6  received the
"Milwaukee  Method" cleaning  procedure before and
after the  30-month off-gas  test program.  After the
diffuser  cleaning in June  1988,  the  measured
aF(SOTE) values were  17 and 19 percent  for the
North and 'South passes, respectively. While these are
higher than the typical aF(SOTE)s observed over the
entire course of the testing,  it  is difficult to conclude
with  just this  one set of data  that cleaning alone
caused such a pronounced increase in aF(SOTE).
                                                 237

-------
Tablo 8-17. East Plant Operating and Performance Data (Annual Averages) - (Jones Island) Milwaukee, Wl
Yoar
1970
1971
1972
1973
1974
1975
1976
W7
1978
1979
1980
1981
1982
1083
1984
1985
1986
Tablo 8-18
TMIM in
Sorvtcc,
mo
3
4
16
1?
24
24
26
30
Flow, mgd
94.2
100.0
96,0
92.2
83.4
78.0
81.6
78.0
79.8
79.0
73.0
74.0
68.0
80,0
73.5
81,7
85.4
Screened Wastewater Effluent BOD5,
BOD5> mfl/L mg/L
208
220
218
261
302
347
326
329
313
290
291
273
263
304
291
278
254
16.5
19.0
18.0
17.0
18.0
22.0
21.0
18.0
21.0
20.0
14.8
14.9
15.8
, 14.7
13.4
12.8
9.3
No. Aeration Tanks
in Service
19.2
20
20
19
19
16
16
16
17
17
16
15
13
15
13
15
14
Air Usage,
cu ft/gai
1.27
1.28
1.36
1.36
1.53
1.S7
1.51
1.55
1.46
1.57
1.58
1.51
1.60
1.36
1.51
1.49
1 .26
Volumetric Loading,
lbBQDs/i,OOOcu fi/d
38,3
41.0 '
, 39.3
46.0
50.9 ' ";
60.4
60,4
58,0
55.8
51.9
48.0
47.3
50.0
65.1
62.4
58.8
58.0
. Oxygen Transfer Performance (1985-1988) for
Tapered, Full-Floor Plate Configuration (Jones , ,
Island East Plant, Basin 6) • Milwaukke, Wl
aF(SOTE).
percent
16
15
15
16
16
17
15
14
Airflow Influent
Rale, BOD5,
scfm/ditfuser mglL
1.2 360
1.5 300
1.2 210
1.1 140
0.9 220
1.0 370
1.2 320
1,0 380
HRT, MLVSS,
hr mg/L
3.5 1,680
3.8 1,270
4.1 1,220
4.0 1,000
4.1 1,310
4.1 1,610
5.3 1,100
4.9 1,460
























238

-------
   8.2.6 Nine Springs Wastewater Treatment Plant

   LOCATION:  Madison, Wisconsin

   OPERATING AGENCY: Madison Metropolitan Sewerage District

   DESIGN FLOW: 1,665 Us (38 mgd)

   WASTEWATER: 85 percent Municipal, 15 percent Industrial {food processing)

   ORIGINAL AERATION SYSTEM: Walker Sparjer and Sanitaire D-24 Coarse Bubble Diffusers

   FINE PORE  AERATION SYSTEM: Norton Ceramic Dome and Sanitaire Ceramic Disc Diffusers

   YEAR FINE  PORE SYSTEM PLACED IN OPERATION: 1977 and 1984, respectively

   BASIS OF PERFORMANCE EVALUATION:  Off-Gas Testing

   CLEANING METHOD:   At high loading, dome plant was cleaned with high-pressure hosing or steam
                          applied routinely.  Dome system with  this cleaning pattern delivers high OTE
                          after 12 years of service. At low loading, need for cleaning has decreased - one
                          cleaning after 630 days of service with high-pressure hosing did not produce
                 	significant change  in OTE.               	            	
I. INTRODUCTION
The  Madison,  Wl  Metropolitan  Sewerage District
(MMSD)  began upgrading its  Nine  Springs  plant to
fine pore diffused aeration in 1977. Additional aeration
system  upgrading  was  implemented from  1984 to
1986. In addition  to aeration system changes,  SRT
was  increased from less than 6 days to more than 9
days, DO control  was installed, and nitrification was
accomplished in single-stage tanks. Extensive aeration
studies  were performed  on the  fine pore aeration
system  to:  1) evaluate various diffuser  cleaning
methods, and  2)  determine  the effects  of  plant
operation on fine pore OTE. When it was discovered
that  serious fouling was  no longer  occurring at the
MMSD plant, the studies on diffuser cleaning methods
were de-emphasized.

Summary details are presented on the economics of
the  system as  currently  operating,  the cost  of
installing the  new  aeration  system,  the OTEs  being
achieved as compared to those envisioned, and the
effects of plant  operating  variables, such  as F/M
loading and SRT, on oxygen transfer.

II. HISTORICAL BACKGROUND
The  Nine  Springs wastewater  treatment  facility,
located  on the south side  of  Madison,  serves the
cities of Madison, Monona,  Middleton, and Fitchburg;
six villages; and portions of several townships  within
Dane County. Approximately 15 percent of the flow to
the  plant is  industrial,  consisting   mainly  of  food
processing  wastes.  Current  average  flow  is
approximately 1,665 Us  (38 mgd); the design flow is
2,190 L/s (50 mgd.)
Raw wastewater is brought to the treatment plant
through 113 miles of interceptor  sewers and force
mains with the help of 84 pumping stations. Degritted
wastewater is  subsequently  split between East and
West plant  sections  where it is settled  prior  to
activated sludge treatment.  The  plant is currently
operated to nitrify in  a single-stage activated  sludge
process. Secondary effluent receives  UV irradiation
prior to being  pumped back into the Yahara River
watershed downstream of the  chain of lakes in  the
Madison area.

The Nine Springs  facility consists of a  number  of
activated sludge process additions undertaken since
its initial construction  in 1934. Figure 8-10 shows  the
current  configuration  of  the  East and  West plant
sections.
Norton dome diffusers were installed  in  Tanks  1-6
(East plant)  in 1977.  Original  plans  called  for
installation "of  Sanitaire coarse bubble diffusers  in
these six tanks to replace Walker Sparjers in a spiral
roll configuration. It was estimated that installation of
Norton domes could  raise  the  overall aF(SOTE)  of
these tanks from 5.75 to 11.25 percent,  providing a
net annual savings of $9,000/yr at $0,03/kWh. The
cost for retrofitting  Tanks 1-6 with 6,364 domes was
estimated at $281,000, $98,000 more than retrofitting
with coarse bubble  diffusers.
Tanks 7-15 of  the  East  plant,  which  contained
Sanitaire D-24 coarse bubble diffusers,  were
retrofitted with Sanitaire disc diffusers, and new Tanks
16-18 were constructed with Sanitaire disc diffusers in
1984  and 1985. For  an  estimated contract price  of
                                                 239

-------
Figure 8-10. Plant layout - Madison, Wl.
                                             Aeration Tanks
          Wost
          Plant
Primary
Clarifiers
                                                  30
                                                  29
              f
                                                  28
                                                 ,__.
                                  *
                                                  26
                                                  28
4
                                                  24
                                                 23
                                                 22
                                                 •BMBMH
                                                 21
                                                 20
o
                                                                          Secondary Clarifiers
                                     H*O
                                                                           Secondary Clarifiers
                                                                                O     O
            Easi
            Plant
$451,000,  11,472 Sanftaire  discs  were  installed in
these 12 tanks.

The new West plant (Tanks 19-30) was  constructed
with 15,576  Sanitaire  discs in 1984-1985  for  an
estimated  $443,000.  These  prices  were  the
mechanical contractor's contract payment prices and
included grid piping,  labor,  and  diffusers.  It was
estimated in the  facilities plan  that,  at an  overall
aF(SOTE) of 12 percent for each 3-pass tank, the fine
bubble diffuser installations would be cost effective.
Prior to 1985, the East plant was operated most of the
time as a non-nitrifying or partially nitrifying secondary
plant.  The modes  of operation included  step  feed,
contact  stabilization, and  plug flow.  Filamentous
bulking incidents were encountered  regularly prior to
1985.  The reason for  such  incidents was  likely
operation of the plant as a high-rate activated sludge
plant with poor DO control. One probe on  one  of the
                            15 tanks was used to manually change blower output
                            for the whole plant.

                            III. FINE PORE  AERATION  RETROFIT  DESIGN
                            DESCRIPTION
                            From 1983 to  1985, the plant was converted from a
                            2,190-L/s (50-mgd)  secondary plant to a  2,190-L/s
                            (50-mgd) nitrification facility.  By January  1985, the
                            plant  was  completely  nitrifying.  Table   8-19
                            summarizes  the  estimated  oxygen  requirements  for
                            the expanded plant.

                            The East plant in Madison consists  of two different
                            sets of  aeration  tanks  and  Clarifiers  referred  to  as
                            Plant 1 (Tanks 1-9) and Plant 2 (Tanks 10-18) (see
                            Figure 8-10). Tanks 1-6 have Norton ceramic domes
                            and  Tanks 7-18  have Sanitaire  ceramic  discs. The
                            West plant consists of two sets of aeration tanks and
                            Clarifiers designated as  Plants 3 and  4. All  tanks in
                            Plants 3  and  4  contain  Sanitaire discs.  The  East
                            blower building serves Plants 1 and 2, and  the  West
                                                  240

-------
Table 8-19.  Oxygen Requirements - Madison, Wl
Plant influent flow, mgd
Thickener recycle flow, mgd
Total flow, mgd
BOD5 removed, Ib/d [(202 mg/L - 19 mg/L}(8.34)(54)]
Ammonia nitrogen removed, Ib/d [{32 mg/L)(8.34)(54)j
Oxygen demand based on planning data
Ib O2/lb BOD5
Ib O2/lb NH3-N •
02 safety factor, percent
Total oxygen demand, Ib/d
BOD5 [(82,400)(1.25)]
NH3-N |(14,400)(4.6)]
Subtotal
Safety factor • • .
Total
Total air required.® 9-percent aF(SOTE), scfm
50
4
54
82,400
14,400
1.25
4.6'
33
103,000
66,200
159,200
55,800
225,000 '
100,000
Table 8-20.  Blower Capacities - Madison, Wl

 Blower No.
      Capacity,
     scfm each
 East (Plants 1 arid 2)
  1 (gas engine)

  2 and 3 (2-stage centrftugal)
  4 (positive displacement)

  5 (positive displacement)

 West (Plants 3 and 4)
  1, 2, and  3 (single-stage centrifugal)
 7,875 (600 rpm)
10,500 (800 rpm)
   5,500-12,500
     7,760 (low)
   10,850 (high)
     5,840 (low)
    9,070 (high)


  12,500-25,000
blower  building  serves Plants 3 and 4. The  actual
blower capacities are given in Table 8-20.

As  previously mentioned, the  installation of fine pore
diffusers took place  in several phases. The maximum
possible airflow rate, based on blower system design,
was 14.2 L/s/m2 (2.8 scfm/sq ft) tank area on the East
side and 16.3 Us/m2 (3.2 scfm/sq ft) tank area on the
West.  Minimum  airflow  rates, which have occurred
frequently,  are  blower-limited  rather than  diffuser-
limited.  Design  minimum  airflow rates for both the
Sanitaire discs and Norton Domes were 0.24 Us (0.5
scfm)/diffuser, but minimum airflow  rates have  been
set  somewhat higher because  of minimum blower
turndown capability.  Minimum mixing rates  that have
been used  in  operation are  approximately  0.46
L/s/m2(Q.Q9 scfm/sq  ft).

Aeration system control  was significantly  upgraded
and is  briefly discussed here (Chapter 6  contains a
more detailed  description of Nine  Spring's  control
system). Basically, a DO probe, located at the effluent
end of  each  aeration tank pass, controls an air valve
for  that pass. A cascade control loop  is used for each
pass of a  three-pass tank to  control the DO and an
airflow  set-point is output to an airflow  controller.  An
airflow tube  measures the total  airflow to each pass.
The airflow controller controls a single butterfly valve
in the air supply line for that pass. A minimum airflow
is  defined  for each  pass  to  maintain  mixing
requirements. DO set-points are typically 1.2,  1.6, and
2.0 mg/L for the three consecutive passes.

For a given total airflow to the aeration tank passes,
air  header  pressure  changes  the  butterfly  valve
settings  for  each  individual  aeration  tank  pass.
Centrifugal blowers equipped  with inlet guide  vanes
are ;used for controlling the main  air header pressure
at  a  predetermined set-point,  plus  or minus  a
deadband.

Ail DO probes are calibrated by operators against an
air-calibrated YSI probe every  week  unless problems
are noticed  before  scheduled  calibrating. Membrane
replacement  is handled  by  the operators.  Other
problems are turned over to the  electronics  staff. At
first, probes were calibrated every 2 weeks, but too
many  problems  with  calibration  were  being
encountered. Now,  first-pass probes  are  cleaned
every week prior to calibration, as growth may easily
coat the membrane within a 1-week period. Second-
and  third-pass  probes may be cleaned  only  once
every 2 or 3 weeks.

IV.  OPERATIONAL  PERFORMANCE  AND
EVALUATION
Following the 1985 conversion of the plant to single-
stage nitrification with a plug flow operating mode and
with  automatic  DO  control, plant performance  has
been consistently  good  and filamentous  bulking
incidents rare. Plant performance for  2 years between
November 1986 and September 1988 was excellent.
The raw wastewater flow during that period was 1,550
Us (35.4 mgd) with an average daily  influent BOD5 of
178 mg/L (0=8.9), TSS of 176 mg/L (0=18.2), and
TKN of 26.7 mg/L  (0=1.6).  Final  effluent  average
concentrations for the same period were: BOD5  = 2.9
mg/L (0=1.5), TSS = 4.5  mg/L (0 = 1.5),  and NH3-N
=  0.2 mg/L (0 = 0.2).

In the spring of 1985, studies  included  monitoring of
OTEs of Tanks 1-6  with  the  off-gas method,  along
with  studying the effect of operational parameters on
OTE in  parallel  aeration tanks  in the  new West plant.
Oxygen transfer studies using off-gas  procedures in
the East  plant  began in  May 1984 and  continued
through the summer of 1987.

Shown in Figure 8-11 are oxygen transfer  test results
for aeration Tanks 1-3 of the East plant. Data for two
periods  of operation  are  summarized:  May  1984-
August  1985 when the plant was run as  a high-rate
system  (SRT = 1.4-5.8 days),  and  November  1986-
July  1987 when the  plant was run as a single-stage
nitrification system (SRT= 10-16 days), .af values are
presented in Table 8-21. These estimates  of aF were
derived  from clean water tests performed  by  the Los
Angeles County Sanitation  Districts on ceramic domes
for a variety of gas flows and diffuser  densities.
                                                  241

-------
Figure 8-11. aF(SOTE) profile for East plant Tanks 1-3 - Madison, Wl.
      22 -1
      20-
      18-
   I  16-
   IH

   °    J
   uT  14-
   o
      12-
      10-
                 SRT = 1.4-5.8 days
               Power Outage
                                                                                       SRT = 10-16 days
                                                    Tanks Oul of Service
         0              200             400

               Day 0 - May 21, 1984

Figure 8-11  indicates overall system aF(SOTE) values
were higher when the  plant was run as a low-loaded,
single-stage  nitrification  facility than  when the  plant
was  run as  a  high-rate   system.  Wastewater
characteristics  were virtually identical  during  both
periods.  It could  be speculated  that a decrease  in
load, as measured by high  SRT (low  F/M  loading),
resulted in higher aF values due, in part, to biological
degradation  of  surfactants that were typically present
in the  final effluents when SRT was lower  and
nitrification occurred only sporadically. Lower loadings
also seemed to eliminate slime growth and fouling on
diffusers  further downstream from  the inlet.  This
observation  is  based  on viewing diffusers when the
tanks were  emptied as  well as examining BRV and
DWP for diffusers that  were removed and taken to the
laboratory for  testing.  Furthermore,  the high-SRT
operation appeared  to  extend  the period before
diffuser fouling  would significantly influence OTE.

In 1984,  a  power interruption and follow-up steam
cleaning of  the dome diffusers significantly  affected
transfer efficiencies as shown by the increase in  aF in
Table 8-21.  During the power outage  and subsequent
startup, clogged pores may have  been opened due  to
the  forces created  in  the diffusers  by the flow  of
mixed liquor. This benefit lasted  only 2-3 weeks,
however.  Steam cleaning, on  the other  hand, created
 600

Day
800
                                                                                    1000
                                                                                                   1200
 an increase  in aF values that lasted about 7 weeks
 according to  the  data presented in Table  8-21.
 Operations personnel believe the mixed liquor flooding
 and purging  of the diffusers  affected a small number
 of pores  but that the effect  was felt through the  full
 depth of  the diffuser. Steam cleaning, on the other
 hand, affected a greater number  of  pores, but  the
 effect was largely superficial,  i.e.,  the cleaning was
 mainly on the surface.

 Diffuser performance at the tank  inlets was greatly
 influenced  by wastewater  characteristics. Coarse
 bubbling was always observed on the liquid surface at
 inlet  zones both  during periods after cleaning and in
 1987, when the plant was  run in the  low-loaded
 nitrification mode.

 Figures 8-12 and 8-13  present the results of off-gas
 testing  of two parallel first-pass tanks in Plants 3 and
 4 of  the  West plant, respectively.  These two  plants
 were operated  in parallel throughout much of the time
 between start-up (9/25/85,  day 0) and  the end  of  the
 study (12/5/87, day 800). Over this  period of time, no
 perceptible  decrease  in diffuser  performance was
 observed in  the  first-pass  tanks based on aF(SOTE)
 measurements. The mean first-stage aF(SOTE) over
 the 800 days was 11.5 percent.  The mean weighted
 aF(SOTE) for  all three passes ranged from 12.1  to
                                                  242

-------
Table 8-21.   East Plant aF Values - Madison, Wl
Date
5/21/84 '
5/23/84
5/30/84
6/7/84
6/8/84
6/12/84
6/21/84
6/26/84
7/6/84
7/12/84
7/20/84

7/27/84
8/1/84
8/7/84
8/13/84
8/20/84
8/29/84
9/7/84
9/18/84
6/20/85
6/26/85 '
7/3/85
7/17/85
8/14/85
7/17/85
11/18/85'. .
5/8/87
5/11/87
5/27/87
5/29/87
6/3/87
6/9/87
6/17/87
6/25/87
6/29/87
7/15/87
Day
l
3.
10
POWER OUT
19
23
32
37
47
53
59
STEAM CLEANED
68
73
81
87
94
103
112
123'
398
•-••-••' 404
411
425
OUT OF SERVICE
. STEAM CLEANED
IN SERVICE
1 ,086
1 ,089
1,105.
1,107 . -
1.111
1,117
1,125
1,133
1,137
1,154
Tank 1
scfm/diffuser
0,50
0.77
0,95

0.52
0.49
0.76
0.83
0.78
0.90
0.43

1.05
0.65
0.93
0.62
0.42
0.63
1.10
0.80
0.93
0.95
0.97
0.52



0.73 I
0.73
0.72
0.59
0.62'
0.62
0.59
0.61
0.62
0.61

aF
0.26
0.25
0.24

0.44
0.38
0.33
0.34
0.33
0.31
0.32

0.43
0.38
0.30
0.30
0.33
0.34
0.31
0.29
0.28
0.31'
0.24
0.47



0.33
0.34
0.40
0.36
0.38 '
0.38
0.42
0.44
d.47
0.43
Tank 2
scfm/diffuser
0.96
1.04
1.30

0.56
0.61
1.03
1.09
1.04
0.69
0.69

1.08
0.61
0.60
0.40
0.40
0.82
1.40
1.04
1.15
" 1.15
1.28
0.63



0.90
0.86
0.86
0.51
0.41
0.81
0.68
0.84
• 0.80 .
0.85

aF
0.35
0,29
0..36

0.60
0.49
0.38
0.40
0.44
0.33
0.37

0.56
0.61
0.62
0.49
0.44
0.43
0.36
0.38
0.39
0.'48
0.38
0.57



0.64
0.63
0.65
0.54
0.54
0.55
0.67
0.78
0.74
; 0.68
Tank 3
scfm/diffuser
0.84
0.95
0.87

0.49
"0.61
0.92
1.15 '
1.12
0.86
0.64

0.84
0.46
0.70
0.67
0.54
0.62
0.81
1.00
1.13
0.92
1.16
0.72



0.79
-
0.61
0.61
- .
0.63
0.73
0.76
0.14
0.74

aF
0.43
0.38
0.42

0.56
0.51
0.40
0.36
0.51
0.40
0.41

0.64
0.71
0.76
0.68
0.56
0.54
0.59
0.45
-0.57
0.61
0.60"
0.71"



0.86
-
0.91
0.77
-
0.85
0.97 '
1.01
1.15
0.85
  3 Nocardia scum.
                                                        243

-------
Figure 8-12, aF(SOTE) profile for Tank 21 (showing Influence of high SRT and cleaning) - Madison, Wl.
     16-1
     14-
     12-
    10-
      8-
      6-
          ,9/27/85
                  100
200
300
400

Day
                                                                 500
                                              600
                                              700
                                                                                                   800
15.3 percent, based on  six  analyses  of  all three
passes.

As shown in Figures 8-12 and 8-13, during a 60-day
period  in Fall 1986, Plant 3  was operated at a higher
SRT   (approximately  11   days)  than  Plant  4
(SRT=approximately 7 days) to evaluate the effect of
process loading on aF(SOTE). Results of this trial are
described more fully  in Section 3.4.2.4.

The  diffusers  in  Tanks 19-21  (operated as a three-
pass,  series flow system  - s©e  Figure 8-10)  were
cleaned by  high-pressure  hosing within the tank 635
days  after being  placed  in  service (6/23/87).
Inspection  of  the diffusers  in those  tanks indicated
moderate fouling  in  the first pass  and little or  no
fouling in the other two. A series of off-gas tests  were
conducted on Tanks  19-24 between June 1, 1987 and
November 6, 1987. Results of these tests are in Table
8-22, (Note that  these  data are  for the two sets of
aeration basins in Plant 3  operated  under  the same
conditions. aF was determined by extrapolating clean
water test data supplied by the manufacturer to the
actual  tank dimensions and  densities at Madison.)
Review of Table 8-22 indicates that cleaning diffusers
in Tanks  19-21   did not  appreciably affect  the
performance of  that system  over the  parallel  one
                        (Tanks 22-24) that was not cleaned. The data suggest
                        that immediately  after cleaning there  may have been
                        some  improvement in Tank 21. This  was short-lived,
                        however, and  both  parallel basins produced  similar
                        aF(SOTE)s 3 weeks after cleaning. In fact, it was not
                        possible to determine whether the elevated aF(SOTE)
                        for Tank 21 on July  9 was due to diffuser cleaning or
                        the wastewater characteristics that day. No significant
                        improvement was seen in the downstream tanks.

                        The successful long-term  performance of the West
                        plant basins, as  contrasted to the East plant  basins
                        (prior to nitrification), strongly  implicates plant loading
                        as a critical factor in  fine pore ceramic diffuser fouling.
                        V. ECONOMIC CONSIDERATIONS
                        The. fine pore diffuser  installation  at Madison  is
                        proving to be very cost effective, especially since the
                        need for diffuser cleaning is very infrequent when the
                        plant is run as a low-loaded,  single-stage  nitrification
                        facility. Only one set of three tanks has been cleaned
                        since  1985, and there were no demonstrated benefits.
                        from cleaning. When tanks are cleaned, it has  been
                        estimated that the cost of steam cleaning  one set  of
                        three  tanks is about $0.61/diffuser. If only hosing  is
                        performed, this cost is reduced.
                                                  244

-------
Figure 8-13.  aF(SOTE) profile for Tank 25 (showing influence of low SRT) - Madison, Wl..
      16'
      14-
   LU

   O
   CO
      12-
       10-
       8-
              9/27/85
                    Low SRT
                     Study
                     I  '
                    100
 I  ~
200
            300
                        400

                        Day
500
 I  i
600
                       700
                                  1  I
                                   800
Table 8-22.   Off-Gas Test Results: West Plant (Plant 3) - Madison, Wf
Tanks 21 or 24
Date
6/1/87
6/10/87
6/11/87
6/23/87
7/9/87
7/10/87
8/6/87
scfm/diffuser
1.32
1,32
1.27
Steam
1.26
1.22
1.02
uF(SOTE)
10.27
11.57
10.13
Clean
13.31
8.55
11.36
aF
0.31 (24)
0.35(21 )
0.30 (24)
Diffusers
0.40 (21 )
0.25 (24)
0.33 (21)
Tanks 20 or 23
scfm/diffuser
1.32
1.29
1.53
in
1.04
1.15
1.30
aF(SOTE)
16.81
16.11
15.77
Tanks
16,89
16.14
17.76
aF
0.57 (23)
0.54 (20)
0.55 (23)
19,20,21
0.55 (20)
0.53 (23)
0.60 (20)
Tanks 19 or 22
scfm/diffuser
1.19
' 1.24
1.36

1.06
0.94
1.29
aF(SOTE)
18.93
17.49
15.70

17.12
16.78
19.47
aF .
0.83
0-59
0.53

0.56
0.54
0.66
(22)
(19)
(22)

(19)
(22)
(19)
Mean
Weighted
aF(SOTE)
13.85
14.05
13.08

14.96
12.13
15.32
 ( ) = lank dumber.                                                                                   '
 Nole: Tanks 21 and 24, 20 and 23, and 19 and 22 are parallel 1st, 2nd, and 3rd pass aeration tanks, respectively (see Figure 8-10).
Air usage since  1986 for  Plants 3 and 4 is shown in
Table 8-23  along with total plant power consumption.
Air usage in both plants  has actually dropped since
1986, as has total plant power consumption.  This is
because one  of  the  more  significant problems
associated  with  plant  start-up was  an inability  to
effectively turn down the aeration system because of
the excellent OTEs of the fine  pore  diffusers  and the
                        fact that loadings were  only 60  percent of design.
                        Minimum airflows for each aeration tank and .minimum
                        blower capacities did not enable turning down the air
                        far enough to match demand. Further, the air headers
                        for the East and  West plants were not tied  together,
                        preventing excess air on  one side of the plant from
                        being transferred to the other.
                                                    245

-------
Table 8-23.  Air Usage - Madison, Wl
                                 Plant 3
                                                                    Plant 4
Dato
November 1986
December
January 1987
February
Marcti
A pill
May
Juno
July
August
September
October
November
December
January 1988
February
March
April
May
Juno
July
AufluSJ
September
cu tt AWIb BODS
Removed
1,513
1,384
1,200
1,076
1,239
1,145
1,161
1,090
1,240
1,174
1,036
1,031
910
1,061
1,000
1,062
905
985
1,053
884
897
822
1,021
cu ft Mr/to O2
Demand Removed
803
691
639
607
692
673
668
655
759
702
574
565
514
550
532
563
512
558
595
524
506
519
606
cu tt Air/ib BOD5
Removed
1,418
1,359
1,273
1,107
1,196
1,122
1,148
1,091
1,133
1,002
1,066
1,061
959
1,151
1,023
1,067
908
914
857
799
841
790
884
cu tt Air/Ib O2
Demand Removed
755
680
678
625
670
659
659
655
693
599
588
581
555
597
544
566
513
520
484
474
473
492
525
Total Plant Power
Use, avg, kW/mo
2,538
2,517
2,504 - .
2,633
2,537
2,516
2,490
2,401
2,417
2,433
2,443
2,368
2,427
2,425
2,459 ' '
2,490
2,305
2,124
2,101
2,049
2,035
2,106
2,125
In March 1988, this problem was solved, in  part,  by
taking nine aeration tanks out of service in the East
plant, turning off all electric blowers on the East side,
and reducing the flow split to the East  plant. This
allowed operation with  a  methane gas engine blower
only In the East plant. DO control is poor since the
gas engine speed is varied manually  within a narrow
range, but no operational  problems  have occurred.
The West plant is being  loaded more heavily and is
using air  more efficiently. Experience  at Madison has
shown  that blower  and  diffuser design for  present
loadings  requires careful  consideration to take full
advantage of high fine pore diffuser efficiencies.
Economic analyses  are  not presented  for  Madison
because  of  the  step-wise  nature  of  the retrofit  and
because  of  changes in  the control  and operational
methods  used at the plant before and  after the retrofit.
One potential cost  savings not  explored  was  the
consideration of leaving coarse bubble diffusers in the
head  end  of tanks while employing  fine  bubble
diffusers  in the remainder of the tanks. Mixing would
be more  complete, and fouling might be eliminated, or
dramatically reduced, under high loading conditions.
                                                   246

-------
   8.2.5 Ridgewood Wastewater Treatment Plant

   LOCATION: Ridgewood, New Jersey

   OPERATING AGENCY: Village of Ridgewood

   DESIGN FLOW: 131 Us (3 mgd)

   WASTEWATER: Domestic

   ORIGINAL AERATION SYSTEM: Walker Sparjer Coarse Bubble Diffusers

   FINE PORE AERATION SYSTEM: Gray "Fine Air" Dome Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION: 1983

   BASIS OF PERFORMANCE EVALUATION:  in-Process  Nonsteady-State  and  Steady-State Oxygen
                                           Transfer Testing and Off-Gas Testing

   CLEANING METHOD: Hosing; Acid  Brushing	
1. INTRODUCTION
In  1983, the  Ridgewood, NJ Wastewater Treatment
Plant underwent a retrofit from a coarse bubble to a
fine pore diffused aeration system. Also, the process
was  changed  from contact  stabilization to tapered
aeration activated sludge.  The purpose  of the plant
retrofit  was to  reduce  energy consumption  and
minimize power costs.  Extensive  aeration studies
were conducted on the fine pore system in 1985 and
1986  to examine changing OTEs  with  time  and
evaluate the  effects  of  different diffuser cleaning
methods and frequencies on OTE.

Information is  presented  on  the  original  aeration
system, including oxygen transfer  test  results,  the
basis for changing to fine pore diffusers, the  retrofit
design,  and the performance and evaluation  of the
new system including cleaning methods. An economic
evaluation of the fine pore aeration  system, including
bid prices, maintenance costs,  and simple payback
period based on power savings, is also included.

II.  HISTORICAL BACKGROUND
The Ridgewood contact stabilization  activated  sludge
plant was constructed in 1959 with a capacity of 131-
L/s (3-mgd).

Original Aeration System
The  original  plant flow diagram  (coarse  bubble
aeration system) is shown in  Figure 8-14, Influent was
screened before passing through grit chambers  and
primary clarifiers.  Stabilized  return  sludge was
combined  with primary effluent in the influent channel
and flowed by gravity to Aeration Tank 1.  From this
contact  tank, flow was discharged to  both secondary
clarifiers.

Sludge  was  returned to  Aeration  Tank  2  for
stabilization prior to combining with  the primary
effluent. Aeration Tank 3 was used to stabilize sludge
supernatant. Aeration Tank 4 was not  utilized  in the
original plant.

The original aeration system (Table 8-24) consisted of
coarse bubble Walker spargers  in a wide-band, spiral
roll configuration and two blowers. Blower  discharge
temperature, pressure  and flow data were  available.
For Aeration Tanks 1 and 2, both flow tube and orifice
plate  data  were available.  During summer months,
blower  capacity was often  unable to  maintain
measurable DO  in  the  aeration tanks resulting in
periodic odors.

Basis for Changing to Fine Pore Aeration
The average aF(SOTE) of the coarse bubble diffusers
in  batch and flowing  systems  in the contact  and
stabilization tanks was  4.8  percent.  Using  an
anticipated aF value Of  0.4 and a clean water SOTE of
28 percent for dome diffusers,  an aF(SOTE) of 11.1
percent was projected  for the fine pore system. This
would allow one blower to  be used instead  of two to
satisfy the same oxygen  utilization rate as the coarse
bubble system.    :

Table 8-25 summarizes the economic advantages that
were anticipated  from the upgrade. A simple payback
period of 6.1 years was estimated for the retrofit, and
odors from the aeration  tanks  were  expected to be
eliminated. Lower sludge production was projected as
a  result  of the  new system's  ability  to supply the
sludge endogenous respiration requirement.

III. FINE PORE AERATION   RETROFIT  DESIGN
DESCRIPTION
In  1982,  a full-floor coverage fine pore diffuser (Gray
"Fine Air")  system was installed in Aeration Tanks 3
and 4. The system was  also converted from contact
stabilization to  the conventional plug flow  mode
                                                247

-------
Figure 8-14.  Plant flow diagram (original coarse bubble aeration system) - Ridgewood, NJ.
     Chtonnaton
                                               Aeration Tank 1
                                                 (Contact)  j
                                                1-4  I 1-3  I 1-2
Aeration Tank 2
 (Stabilization)
 2-1  |  2-2  j 2-3
                                                    Process
                                                    Control Btdg.

Truck Loading
%
\
                                                   Aeration Tank 3
                                                   Aeration Tank 4
                    1— Chlorine Bldg.
                      Wet Well
                      Lift Station
Tablo8-24.  Physical  Characteristics  of Original Coarse
            Bubble Aeration System (Tanks 1 and  2 )  -
            Ridgewood, NJ

                                    Aeration Tank
          Table 8-25.  Projected Energy Savings - Ridgewood, NJ

                                             Aeration System

Dillusors (Walker Sparjers)
No. compartments
Surface aroa'compartmenl, sq ft
No. dilfuscrs/compartmcni
No. dilluswstonk
Density, sq ft/diflusor
HcKjtu off tank bottom, It
Tank water dopttt, It
Blowers ISponcor Turbine -
Turbo-Compressor Model 362)
Typo
Nomibal rating, hp
Tolal number
Number in use ai any time
Typical efficiency, percent
Tank 1
(Contact)

4
678
40
160
17.0
2
14.5-15.5


Centrifugal
75
5
2
43
Tank 2
(Stabilization)

4
678
28
112
24.2
2
14.5-15.5







(Figure 8-15),  wilh  primary  effluent now  flowing  to
both aeration tanks in parallel.

At  the influent  and  effluent  ends  of  both  aeration
tanks, wooden  baffles were installed to distribute and

SOTE, percent
Temperature, °C
oF
P
C, rng/L
aF(SOT£)(, percent
Airflow, scfm
No. blowers in use
Power draw, kWh/d
Power cost (@$0,065/kWri), $/yr
Bid price for retrofit, $
Simple payback period, yr
Coarse Bubble
8.6
20
0.55
0.99
0
4.8
2,100
2
3,000
71,200
-
-
Fine 'Pore
28.0
20
0.40
0.99
0
11.1
1,100
1
1,500
35,600
218,000
6.1
          collect the flow across the total tank width.  Four grids
          were used in each tank with  a decreasing  number of
          domes  from inlet  to  effluent  end, as  summarized in
          Table 8-26.  All  domes  are  17.8-cm  (7-in) diameter
          Carborundum  (Aloxite)  diffusers  that were  initially
          connected to the saddles approximately 25  cm (10 in)
          off the  bottom using  plastic (acetal)  bolts. After  1-1/2
          years of operation, all plastic  bolts were replaced with
                                                      248

-------
Figure 8-15,  Plant flow diagram (retrofitted fine pore aeration system) - Ridgewood, NJ.
     Chlorination
                                            Aeration Tank 1
                                             (Contact)  ,
                                            1-4  i 1-3  I 1-2
Aeration Tank 2
 (Stabilization)
 2-1  j 2-2 j 2-3
                                                Process
                                                Control BIdg.
                                         —     Aeration Tank 3
                                                Aeration Tank 4
                   I— Chlorine BIdg.
                     Wet Well
                    Litt Station
brass bolts  due to numerous failures. The  fine pore
aeration system was started up in April 1983.
IV.  OPERATIONAL  PERFORMANCE  AND
EVALUATION
In March 1983,  four  clean  water oxygen  transfer
studies  were conducted on the  newly-installed dome
diffuser  system.  The  results  of  these tests  are
summarized in Figure 8-16 (see  plot labeled "Original
Equation"). Unlike the previous coarse bubble diffuser
system,  increased  airflows   resulted  in  now
significantly  decreased  SOTEs.  Therefore,  Aeration
Tanks 3  and 4  were  modified by  increasing  the
average dome density from 2.1/m2 (19.4/100 sq ft) to
2.5/m2 (23.4/100 sq ft).  The  dome configuration for
the modified density is presented in  Table 8-26. No
clean  water  studies were  conducted  after  the
modification. The equation used  to calculate SOTE as
a  function of airflow for  Ridgewood  was  modified
based on modeling concepts as follows (see Figure 8-
16):

      SOTE = 31.6 - 5.38 (Airflow)     (original)
      SOTE = 32,1 - 5.47 (Airflow)     (modified)

where airflow is in scfm/diffuser.
Often during summer months, sludge  accumulation in
the plant  was significant due  to abandonment of the
          sludge  lagoons and Aeration Tanks 1 and 2  were
          used periodically for waste sludge storage.  Digester
          sludge  supernatant quality during  this time  was
          generally  poor  with a significant quantity  of  digested
          solids probably recycling through the aeration system.
          Also, with  the retrofit,  a  significant  amount  of
          "Nocardia"  growth appeared in late spring and  early
          summer  months and  remained  until winter.  This
          resulted in a thick surface foam layer that periodically
          overflowed the tanks.

          Aeration Performance Evaluation
          Two  oxygen transfer studies were  conducted on the
          coarse  bubble  aeration system, while the remaining
          studies  were performed on the  fine pore  aeration
          system. Estimated annual averages for aF(SOTE) and
          aF are  indicated in Table  8-27 for the two  systems
          under both low  and high airflow rates.

          A nonsteady-state analysis  technique  was employed
          to evaluate the coarse bubble system. Testing was
          initiated  on October  21,  1981  with  five batch
          wastewater, two flowing wastewater, and  three clean
          water tests conducted.  In  July 1982,  18  nonsteady-
          state flowing wastewater tests were  conducted. For
          each of the flowing  wastewater tests, the  primary
          effluent and return sludge flows  were reduced  to
          provide reduced load  conditions   and positive  DO
          concentrations  for testing.  aF(SOTE) results for the
                                                  249

-------
Tablo 8-26.  Physical Characteristics of Fine Pore Retrofit System (Tanks 3 and 4) - Ridgewood, NJ
                                                                                        Grid

Dilfusor typo
Tank surface area, sq H
No. grids
Tank water depth. It
Diduscr height oil tank bollom. It
Initial Operation (4/83-9/84)
No. ditfusers
Donio density , sq ft/dome
Final Operation (9/84-oresent)
No. diffusors
Oomo density, sq It/dome
Each Tank
7-in Gray Domes
2,784
4
14.5-15.5
2

340
5.15

650
4.28
A "
696

180
3.87

234
2.97
B
696

160
4.35

208
3.35
C
696

100
6.96

104
6.60
D
696

100
. 6.96

104
6.69
Figure 8-16. SOTE vs. airflow rate (fine pore system) - Ridgewood, NJ.
               35-
               30-
               25-
               20-
               10-
                5-
       Modififld Equalion
       SOTE = 32.1 - 5.47 (Airflow)
                 Density  = 23.4 diffusers/100 sq ft
                                                                  Original Equation
                                                                  SOTE = 31.6 - 5.38 (Airflow)
                                                                  Density = 19.4 diffusers/100 sq ft
                            0,2
                                      0.4
 I
0.6
                                                         0.8
                                                                             I
                                                                             1.2
 I
1.4
                                                                                                1.6
                                                                                                          1.8
                                                         Airflow Rate, scfm/diffuser
System
Coarse BubWo
FmoPoro
Fino Poro
Test Period
1981-1983
1985
1986
No. Tests
25
21
7
20
4
Airlow Rate
High & Low
High
Low
High
Low
aF(SOTE), percent
4.8
7.5
8.9
9.6
12.6
aF
0.55
0.36
0.36
0.41
0.48
                                                            250

-------
coarse bubble system varied from 3.6 to 6.5 percent
under wastewater conditions, while clean water SOTE
was  8.6 percent.  The coarse bubble system had an
estimated  annual yearly  average aF(SOTE) of  4.8
percent and an average aF of 0.55.

From June 1985 to  September 1986, 66 flowing
wastewater tests and one batch wastewater test were
conducted  on the fine pore ceramic  dome  system.
The  off-gas  and  nonsteady-state  techniques  were
employed  during the early stages of the study, with
off-gas and steady-state test procedures emphasized
in the middle and latter stages.  Measured aF(SOTE)
values exhibited a large degree of variability during the
study with results ranging from 5.2 to 15.2 percent, in
1985, the fine pore system had an average aF(SOTE)
of 7.5 percent and an average  aF of 0.36 under high
airflows. Under low airflows, average  aF(SOTE) and
aF were 8.9 percent and 0.36, respectively.
Potential wastewater effects on aF were minimized by
conducting  low  and  high   airflow  testing at
approximately the same time of the day. For 1986, the
estimated average aF(SOTE) was 9.6 percent with an
average aF of  0.41  under high airflows. Under  low
airflows conducted in the morning hours during  low
loading periods, average aF(SOTE) and aF were 12.6
percent and 0.48, respectively. Nocardia foam was  a
problem during  the studies and  tended to affect OTE
results.

In  summary, the Ridgewood,  NJ  coarse bubble
system, tested over 2-3 years near the end of its 25-
yr life, had an average aF(SOTE) of 4.8 percent with
an average aF of  0.55. In  contrast, the  fine  pore
system, in operation for  almost 4  years,  had an
average aF(SOTE)  of about  9.5  percent  with an
average aF of 0.40 during normal  daytime  high-load
operation and with two tanks in service.

Effect of Cleaning on Performance
Two  methods of cleaning were utilized on  the  dome
diffusers, acid brushing and  water hosing. To acid
clean the domes,  a  solution made from 1/2 carboy of
20 percent HCI, diluted 1:1, was used to brush each
dome. Water hose cleaning  used  a high  pressure
stream of water from a fire hose sprayed from the top
of the aeration  tank. Typically, an  aeration tank was
out of service for less than 15 days during a cleaning.

It was difficult  to evaluate  an immediate effect of
diffuser cleaning on  OTE due to changing wastewater
characteristics and the limited amount of data for the
periods both  before and after cleaning. An immediate
increase in OTE was observed  after the July 17th and
July  28th,  1985 cleanings on  Aeration Tank 3. The
low-airflow  aF(SOTE), measured at 9.1 percent before
the cleanings, increased to 9.9 percent after the hose
cleaning (July 17th). After a  second cleaning by acid
brushing  (July  28th),  the  low-airflow  aF(SOTE)
increased  to  11.5 percent. However, primary effluent
BOD5 for the above tests decreased from 159 mg/L to
141 and  97 mg/L,  respectively.  Thus,  it is not clear
what immediate impact, if any, cleaning  had on OTE.

V. ECONOMIC CONSIDERATIONS
Preliminary  assessment  indicated  that  fine  pore
aeration retrofit would enable Ridgewood to reduce
blower power  consumption  by  50  percent. Actual
power  reduction  averaged approximately 28 percent
for 1984-1986, as  shown in Table 8-28. Table 8-29
summarizes the maintenance costs incurred with  the
fine pore system. Maintenance costs from April 1983
through 1986  averaged  $2,780/yr for  the  fine'pore
system; by  comparison,  minimal  maintenance  was
required for the  coarse  bubble  system. These fine
pore  maintenance  costs include experiments  in
diffuser cleaning and control of Nocardia foam, neither
of which may be transferabfe costs to other systems.

Figure  8-17  illustrates  the  savings  in  power
consumption obtained with the ceramic  dome aeration
system.  Two-blower operation  requires 89,700
kWh/mo, while average actual power consumption is
about  63,400  kWh/mo. The projected cost for  the
coarse bubble system is based on  continuous two-
blower operation.  Approximately  40 percent of  the
initial cost was paid off from April  1983 to December
1986. The bid  price for retrofitting the  plant with fine
pore diffusers  was  $218,000. The initial cost was to
be recovered from the power savings incurred with
the new aeration  system.  Based on  a 50-percent
power  consumption reduction and  1982 power costs,
the simple  payback period  was  projected to be  6.1
years.  Based  on  actual payments,  the  predicted
simple payback period is approximately 9.7 years. If
the increased dome maintenance co'st  is included as
computed,  the projected  simple payback  period is
11.1 years as shown in Table 8-30.

Installation of the fine pore dome system has resulted
in a significant improvement in  effluent quality with
respect to  nitrification.  Beginning in  May  1987,
Ridgewood was  required  to  provide  seasonal
nitrification.  Consequently, greater oxygen  demand
has been incurred by the fine pore aeration  system in
recent summers than experienced by  the coarse
bubble aeration  system:  Although  BOD and  TSS
removals have deteriorated  slightly, during  the last
several years, the high degree of nitrification achieved
in the summer  months  has significantly  reduced
overall oxygen  demand  on the  Ho-Ho-Kus  Brook
receiving stream. This reduced oxygen  demand is not
taken into account in the above economic analysis.

The original purchase  agreement specified  that
Ridgewood would  pay back the  initial cost of  the
retrofitted  aeration  system  incrementally based  on
actual  energy savings realized from yearly reductions
in blower power consumption.  This procedure was
followed for several years.  Because of the lowered
oxygen demand  of their treatment  plant's  effluent,
however, the Village decided to pay off the remainder
of the initial cost in  a lump sum.
                                                 251

-------
Tablo 0-28.   Average Blower Power Reduction - Ridgewood, NJ

                                              Blower
Parameter'
On-lwno, hr/mo
Power cost, S/kWh
Power usago, kWh/mo
Power cosl, $lyf
1
730
0.0746
44,840
39,950
2
313
0.0746
19,583
17,'550
Total
: -1,043

63,423
''57,550
Reduction
417

26,257
22,400
  1 Based on average values for 1984-1986. Yearly power reduction = 100(22,400)/(2)(39,950) » 28 percent


Table 8-29.   Dome Diffuser System Maintenance Costs - Ridgewood, NJ

 Year	DiHusec Cleaning, $	Repairs, $    '   Foam Chloririation, $    Foam Cleanup, $	   Total, $
4/83-12/83
1984
1985
1386
Avorago
0
250
2,525
900
1,005
0
700
350
0
280
0
0
1,825
875
720
0
0
2,275
630
775
0
, ' 1,080
8,975
2,405 „
2,780
Tablo 8-30.   Domo System Economic Summary - Ridgewood,
             NJ (1983-1986)


 Power savings from retrofit, $/yr                22,400
 increased maintenanca, $fyr                    2,780
 Not savings from feirofil, $/yr                  19,620
 Pino poro system bid price', $                218,000
 Projected simple payback period, yr
   Based on average power savings only               9.7
   Based on average not savings             ,      11.1

  «Iniual cost plus interest expenses (7 yr @ 9 percent).
                                                          252

-------
Figure 8-17,  Savings in power consumption - Ridgewood, NJ.

        100-1
         30-
         20-
         10-
                                                                      Savings
                                           Actual Blower Operation
ill  T  I  i I  I
  6      g     12
 1983
I  I  I  I  I I  I  I  I  I  I  I I  I  I  I I
6      g     12      3      6       g
1984                       198S
                                                                                    12
 I  I  I I  I
6      9
 1986
                                                                                                               12
                                                       253

-------
   8.2.8 Whittier Narrows Water Reclamation Plant

   LOCATION: El Monte, Los Angeles County, California

   OPERATING AGENCY: Los Angeles County Sanitation Districts                  '

   DESIGN FLOW: 657 Us (15 mgd)

   WASTEWATER: Domestic

   ORIGINAL AERATION SYSTEM: Coarse Bubble Diffusers

   FINE PORE AERATION SYSTEM: Sanitaire Ceramic Disc and Norton Ceramic Dome Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION:   Ceramic Discs, 1980
                                                       Ceramic Domes, 1982

   BASIS OF PERFORMANCE EVALUATION: Off-Gas Testing

   CLEANING METHOD: In-Situ HCI Gas Cleaning	      :
I. INTRODUCTION
The Whittier Narrows Water Reclamation Plant is one
of the older  secondary treatment  plants in  the Los
Angeles County Sanitation Districts (LACSD) system.
It is  considered a water reclamation facility since it
also  provides  treated  effluent  for  groundwater
recharge to  the  San  Gabriel  groundwater basins.
Industrial discharge to  the plant is closely monitored
and  limited.  The  plant is  operated with  relatively
constant flows with diurnal fluctuations  bypassed to
the downstream Joint  Water Pollution  Control  Plant
(JWPCP) in Carson, California.

In the 1970's, LACSD conducted clean water tests at
the JWPCP in an attempt  to determine the relative
performance  of various aeration systems, including
coarse bubble  diffusers, static  aerators,  jet aerators,
and fine pore diffusers  (rigid porous plastic tubes and
ceramic domes). As a result of these tests, LACSD
decided to evaluate fine pore ceramic and rigid porous
plastic  diffusers and  jet  aerators under process
conditions  in  the  aeration  tanks  at  the Whittier
Narrows  plant. Specifications were developed  and,
based on the bids received,  the three  field systems
installed  were  Sanitaire ceramic discs,- Nokia  rigid
porous  plastic tubes,  and  Cleave-Pac  jet aerators.
After  1 year  of evaluation,  the  jet  and tube systems
were  decommissioned  in  favor  of  ceramic
domes/discs. In March and May 1982,  Norton  dome
diffusers were installed  in two of the three basins. The
third basin continued to operate  with the Sanitaire disc
diffusers. The three aeration  systems have remained
in operation since, with the only  modification being the
replacement of the original dome diffusers with  new
ones in September and October 1987.

An  aeration system evaluation  program was  carried
out from April 1986 to July 1988. The objective was to
evaluate the presence and effect of long-term diffuser
fouling and the effectiveness  of in-situ HCI gas
cleaning in  preventing or retarding  the rate of diffuser
fouling.  In-situ  gas  cleaning  was  performed
periodically  in the tank  with the disc diffuser system
and in one of the tanks  containing the dome diffusers.
The second dome tank served as a control, and these
diffusers  were  not cleaned.  At  the  start  of the
evaluation program,  the diffusers were cleaned in  all
tanks by a liquid  acid cleaning procedure.  After
starting the  test program, 33 site visits were made for
off-gas testing over a 28-month period.

Summary information  is presented,  to describe the
secondary treatment and fine  pore aeration systems,
diffuser cleaning and off-gas testing programs, results
of the long-term aeration performance evaluation, and
an estimate of  the  economic benefits of using fine
pore aeration at this  site.  ,

II. DESCRIPTION OF WASTEWATER TREATMENT
FACILITY
The  Whittier Narrows  plant  is  one  of eight plants
operated  by LACSD. The plant is uniquely located  so
that  the  flow to the plan can  be  maintained at a
relatively constant  rate  by  directing  other  flows
downstream in  the  gravity flow  sewerage system  to
the JWPCP in  Carson. The  waste sludge  is also
directed downstream; as a result, the  Whittier Narrows
plant has no sludge  handling facilities and associated
recycle flows.  As with all LACSD plants, industrial
flows to the plant are monitored and limited. Thus, the
plant  has ideal. operating conditions  with regard  to
sludge recycle  streams,  flow  variations, and toxic
loads.

Whittier Narrows is  a conventional primary-secondary
treatment system consisting  of three   rectangular
primary  clarifiers, three  activated  sludge  aeration
basins, and six rectangular secondary clarifiers. These
                                                 254

-------
are followed by  sand/anthracite filtration, chlorination,
and  dechlorination  prior  to  groundwater recharge or
discharge to the  San  Gabriel River. One  of  the
secondary ciarifiers has been  converted for use as a
filter backwash storage tank. The primary ciarifiers are
91.4 m (300 ft) long by 6.1 m (20 ft) wide with a 3.7-
m (12-ft) SWD.  The secondary ciarifiers are 45.7 m
(150 ft) long by 6.1  m (20 ft) wide with a 3.0-m (10-ft)
SWD. The secondary clarifier overflow  rate, using five
units, is 41  m3/m2/d (1,000  gpd/sq  ft)  at the average
design flow  of 657 Us  (15  mgd). The three  parallel
aeration basins are  91.4 m (300 ft) long, 9.1 m (30 ft)
wide and have a nominal 4.6-m (15-ft)  SWD, resulting
in a nominal  4.8-hr  detention time at the  average
design flow. Y-walls in the aeration  basins reduce the
surface  width  to about 7.6 m (25 ft). The  aeration
basins can be operated in series with serpentine flow
or in parallel as conventional plug flow systems. In
either mode, step feeding of the influent is possible.
The conventional plug flow operating mode was used
during  most of  this ceramic diffuser field  evaluation
study, with the exception  of step  feeding from
November 1987-January  1988.

The facility is normally operated at a low enough SRT
to avoid  nitrification. The  lack of industrial  toxics  and
the  warm  wastewater temperature  (yearly average
24OQ [75 °p])  provide ideal  conditions  for nitrification.
Nitrification  reduces  the  effectiveness  of  the
chiorination  facility.  Better  chlorination is provided
when a combined residual chlorine can be maintained.
Disinfection  is  very important because the effluent
coliform  standard is the same as for drinking water
(<2.2 MPN).  Virus destruction is also of concern.
This is addressed by the State of  California through
the effluent turbidity standard, which is  <2 NTU.

During the ceramic  diffuser fine  pore aeration study
period, the average operating SRT was 2.7 days. This
was calculated using the  aeration tank solids inventory
and does not include solids inventory in the secondary
clarifier. A 3-day running average of SRT was 1.8-8.3
days during the  study. During this period, the average
F/M loading, MLSS and  MLVSS  concentrations,  and
influent  flow rate were 0.59 cM, 1,047 mg/L and  771
mg/L, and  574  Us (13.1  mgd), respectively. Primary
effluent  TSS,  BODg,  and  COD concentrations
averaged 91,   102,  and 231 mg/L,  respectively.
Treatment efficiency was generally  excellent, with an
average  effluent BOD5  concentration after  filtration
and disinfection  of 5.2 mg/L. SVI averaged 171 mL/g
and ranged from 97 to 654 mL/g. A low-SRT operation
from  August to October 1986 produced high  SVI
sludge.  On  occasion, organic polymer was added in
the  channel between the  secondary clarifier  and
aeration basin to improve solids settling in the clarifier.

III.  DESCRIPTION  OF FINE  PORE AERATION
SYSTEMS
Figure 8-18 is a schematic of the Whittier Narrows
aeration basin. Each tank  has three  aeration grids with
a downcomer to each that has one control valve and
flow measuring  tube. There are two main air headers,
with one of them  shared between Basins 2 and  3,
which contain the dome diffusers. The main air header
piping  was  kept in  service when  the  system  was
converted   to  fine  pore  aeration. The  diffuser
submergence for all basins is 3.7 m (12.3 ft), i.e., 0.8
m (2.7 ft) above the floor, which  is required due to
limited available blower pressure.

Basin 1 contains the Sanitaire 23-cm (9-in) ceramic
discs of  the old design that  have  a 25-mm  (1-in)
thickness. Basins 2 and 3 are equipped  with 18-mm
(7-in) Norton domes of the old design, which are high-
fire, tan-colored stone compared to the low-fire, gray-
colored stones  of  the new  design.  Each diffuser in
these  systems  is  equipped  with an  individual
removable inlet orifice. The Norton system is unusual
in  that fiber-reinforced  ABS bolts  and  "spongy"
gaskets were  used, as opposed  to PVC bolts and
hard  rubber gaskets. The ABS  bolts  were  used
because  PVC bolts have poor  resistance to HCI acid
application. The spongy gaskets were standard issue
at  the time. When the domes   were   replaced  in
September and  October 1987, several  domes were
equipped with hard  rubber  gaskets. The ABS bolts
failed in that application, confirming  the need to use
spongy gaskets with ABS bolts.

Table 8-31  summarizes  diffuser layout and density in
each  basin  and grid.  The  design includes tapered
aeration to  provide for reduced air application  to the
third grid at the end of  each aeration  basin.  The
diffuser density  is about 23-26 percent greater for the
domes than for the  discs.  The  relative density  is
based on the assumption that the media surface area
of 1 disc is approximately equivalent to  that of  1.25
domes.

The aeration system has  three  centrifugal blowers
remaining from the original  spiral  roll coarse bubble
aeration operation.  Two are 298-kW (400-hp)  units,
each rated at 5,760 Us (12,200 scfm) at  146 kPa (6.5
psig). The  other blower is  a  149-kW (200-hp)  unit,
rated at 2,830 L/s (6,000 scfm) at 146 kPa (6.5 psig).
Normally one of the large blowers is operated. Flow
meters are  available  for each blower, aeration  tank
header, and aeration grid downcomer. A  two-stage air
filtration system was installed in the blower suction air
line but  was  not used for most  of the aeration
evaluation period because  of  corrosion  problems  in
the piping.

Automatic DO control is  used for this plant. A probe is
mounted at the effluent end  of each tank, and blower
flow rate is  increased or decreased as  needed. The
air distribution between Basin 1 and Basins 2 and 3 is
manually set by the two header valves.  The relative
flows to each grid  are set by the  downcomer valves.
Adjusting the  airflow distribution usually requires
several valve adjustment iterations.
                                                  255

-------
Figure 8-18. Aeration basin schematic - Whittier Narrrows, CA.
           ~x 3 Inlluoni Gates (typical)
     \l
      M
      u
                                                   30 ft Basin Widlh
                                                   25 fl Widlh at Surface
                            -.^                          *
                               Downcomers with Manual Control Valves (typical)

                        3 Grids per Basin (typical)
                                                                         Basin 3, Ceramic Domes (Control)
                                                                         Basin 2, Ceramic Domes (Gas Cleaned)
                                                                         Basin i, Ceramic Discs (Gas Cleaned)
                                                                          Effluent Channel (typical)
                                    300 fl
       Air Headers and Manual Valves
Toblo 8-31.  Diffuser Layout Summary - Whittier Narrows, CA
                        Dillusers (submergence  « 15 fl)
Basin
i
Discs

2 and 3
Dorms

Grid
1
2
3
1
2
3
Number
792
774
460
988/836"
S68
574^728*
Density,
No/100 sq ft
26
26
15
33KB"
32
19/24°
 » Boloro and alter August 21,1987.

Over the  period of  this  study,  the DO  control
operational strategy changed several times. Generally
the DO is maintained at 0.3 mg/L or  less in Grid 1 and
is  increased to 1-2 mg/L at the  effluent end of the
tank,

IV.  OPERATIONAL  PERFORMANCE  AND
EVALUATION
Table  8-32  summarizes the  off-gas testing and
diffuser cleaning  program schedule.  In  May and June
1986,  after operating for about  18 months without
cleaning, the diffusers were rigorously cleaned using a
liquid acid surface application method.  Off-gas testing
was conducted prior to this cleaning  to determine dirty
diJJuser OTE. Off-gas testing was performed at least
once per month thereafter to follow diffuser fouling
and performance with  time  both before  and after
cleaning events.

The times between gas cleanings were selected on a
somewhat arbitrary basis. Initially, it was hoped  the
rate of increase of diffuser pressure  loss, as indicated
by an increase in DWP or a loss in OTE as measured
by off-gas testing, would dictate the  need for acid gas
cleaning.  Variations  in  OTE  due  to  day-to-day
fluctuations  in  wastewater characteristics and  poor
precision  in the  DWP measurements  made  this
strategy impractical. In addition,  during the  planning
phase  of the project,  a change  in the gas  cleaning
philosophy was introduced by Sanitaire. Cleaning was
no  longer felt  to  be  a method to  restore fouled
diffusers, but was felt to be an  effective method  to
prevent diffuser fouling when applied periodically. The
experimental design called for  cleaning of Grid  1  in
Basins 1  and 2 every  3 months.  Grid 2 was cleaned
every  6 months, and  Grid  3 was cleaned  every 9
months.

During cleaning, the HCI gas was introduced into the
downcomer  feeding each diffuser grid. One grid  at a
time was  cleaned.  The airflow  to the  grid  being
cleaned was increased to a desired level of 1,4 L/s  (3
scfm)/dtffuser,  or  as   close  to  that as possible.
Generally, the airflow achieved was about  1.2 L/s (2.5
scfm)/diffuser. The increased airflow rate ensures that
HCI  gas permeates  through the  entire  diffuser  area.
The  HCI gas rate was approximately 9.4  L/s (20
scfm)/grid. A small decrease in DWP and increased in
air flow rate was usually observed within  30  seconds
of introducing the gas. The original Sanitaire protocol
required that 45 g  (0.1  Ib) HCI/diffuser be used. This
quantity of gas per cleaning event was reduced based
on  DWP observations. Gas addition  was curtailed
when  DWP readings decreased to  a  plateau.
Consequently,  the  gas  dosage was 10-25  g
HCI/diffuser.

On two occasions,  air leakage  was  noted for the
dome diffusers  in  Basins 2 and  3 based on the air
bubble pattern on the surface. For the first instance in
September 1987, the leaks were occurring around the
                                                   256

-------
Table 8-32.  Project Chronology - Whittier Narrows, CA
 Date
 4/28/86
Event
Off-gas testing
 5/12/86       Off-gas testing
 5/13-6/19-86  Liquid acid cleaning
             of all 3 basins
Comments
Background testing
performed to determine
dirty diffuser efficiency
Background testing
Diffusers collected for
analysis; dome gasket
leakage noted
6/20/86 .
7/02/86
7/22/86


8/01/86
8/86

8/21/86
8/26-8/27/86

9/04/86
9/17/86
10/17/86
10/31/86
1 1/1 7/86
12/9/86

1/16/87
1/30/87 .
2/13/87
2/27/87
3/13/87
3/26-3/27/87


4/03/87
4/17/87
5/22/87
6/05/87
6/15-16/87

6/19/87
7/10/87
7/31/87
8/31/87
9/9/87

9/30/87

9/30/87
10/9/87
11/13/87
12/04/87
12/24/87
1/15/88

1/26/88
1/29/88
2/19/88
3/11/88
5/88; 6/88



6/16/88
8/7/88


8/1 2/88
Off-gas testing
Off-gas testing
Off-gas testing


Off-gas testing
Process operation
changed
Off-gas testing
First HCI gas
cleaning
Off-gas testing
Off-gas testing
Off-gas testing
Off-gas testing
Off-gas testing
HCI gas cleaning

Off-gas testing
Off-gas testing
Off-gas testing
Off-gas testing
Off-gas testing
HCI gas cleaning


Off-gas testing
Off-gas testing
Off-gas testing
Off-gas testing
HCI gas cleaning

Off-gas testing
Off-gas testing
Off-gas testing
Off-gas testing
Domes replaced
in Basin 3
Domes replaced
in Basin 2
HCI gas cleaning
Off-gas testing
Off-gas testing
Off-gas testing
Off-gas testing
Off-gas testing

HCI gas cleaning
Off-gas testing
Off-gas testing
Off-gas testing
Basins 2 & 3
manually cleaned
using tank top
hosing
Off-gas testing
Basin 1 manually
cleaned using
tank-top hosing
Off-gas testing


Hoods not moved in order
to determine diurnal
fluctuations in aF(SOTE)

MLSS reduced in all three
basins

Grids 1, 2, & 3 cleaned in
Basins 1 & 2





Grid 1 of Basins 1 & 2
cleaned





Grids 1,2, & 3 of Basins 1
& 2 cleaned. Simultaneous
off-gas testing performed.




Grid 1 of Basins 1 & 2
cleaned




Some gasket leakage noted
afterwards
Some gasket leakage noted
afterwards
Grid 1 of Basin 1 cleaned




Grids 1,2, & 3 of Basin 1
cleaned
Grid 1 of Basin 2 cleaned



Gas leakage noted. Broken
bolts noted







 Basin 1  - Disc Diffusers
 Basins 2 and 3 - Dome Diffusers
diffuser gaskets.  It  was determined that the diffusers
were fouled to the extent that they had to be replaced.
In May 1988, the observed  air leakage was due to
broken bolts that were repaired.

At selected  periods, (jiffusers were  collected from
each grid and sent to the University of Wisconsin for
observation and analysis. The diffusers were analyzed
for  DWP,   BRV,  Joulant  material  mass  and
composition,  and  air distribution  profiles.  Diffusers
were collected prior to cleaning at the  start of the test
program to determine  their characteristics  after 18
months of operation without cleaning by  tank  top
hosing.

The  standard off-gas testing  procedure was followed,
and the gas hoods used in the basins were in the size
range of 3 m long by 61 cm wide (10 ft by  2 ft). The
hood was  situated  at two locations  across  the basin
width and two down the length of a grid to produce  a
total of four  measuring  points per grid. The  location
along the grid  length was at about the 1/3 and  2/3
points.  At these locations,  the  hood  was placed
against one wall and then in the center of the basin.
Both the dome and disc diffuser  basins were tested
on the same day. To obtain comparative test  results
under  as  similar  wastewater  conditions as possible,
the grids  were tested alternately. When  grid testing
was finished  at one  hood location, testing would take
place in the  other  basin while the  hood in the first
basin was moved to the  next location.

Evaluation  of the removed diffuser samples indicated
that  DWP  could  be controlled by HCI gas cleaning.
However,  DWP after cleaning was usually about twice
that of a new diffuser. BRV measurements seemed to
increase with time of service, but were also reduced
after gas cleaning.  The amount  of  fouling material
found on  the sampled diffusers  was  about twice as
high in Grid 1 as in  Grid 3, with Grid 2 diffuser fouling
in between these.

Lab  diffuser  cleaning  studies  compared  the
effectiveness of cleaning by  hosing  and liquid  acid
application methods. Based  on DWP and  BRV
measurements, liquid acid cleaning  was  superior  to
hosing in reducing DWP and BRV. In many instances,
hosing decreased BRV but not DWP.

Off-Gas Testing Results
Figure  8-19 compares aF(SOTE)  over the duration of
the study for the dome diffusers  in  Basins  2 and  3.
The  only  difference between  the two basins is that
gas cleaning  was performed on the diffusers in  Basin
2 but not in Basin 3. It  is apparent from  Figure 8-19
that gas cleaning did not affect OTE for the dome
diffusers,  quite possibly because of  diffuser gasket
leakage problems.

Two significant dips in  the  aF(SOTE) curves  at
approximately the 3rd and  17th months were due  to
                                                   257

-------
Figure 8-19. uF(SOTE) vs. time (ceramic dome diffusers)
           Whittier Narrrows, CA.
                 Basin 2 (Domes, HCI Gas Cleaned)
                > Basin 3 (Domes, No Cleaning)
   14-
    4-


    2-
 Iniiial Liquid
Acid Cleaned
                         Dome Replacement
                                      Slep Feed
                                      Operation
    aF(SOTE), percent

     Lower SRT, MLSS
                                 Domes
                                  Acid
                                Cleaned
          Gas Cleanings
   1,2,3D  1  1,2.3    1   1.2B1   1,2,361


    i   y    Y    y    y     y
     •2  0  2   4  6   8 10 12 14 16 18 20 22 24  26
            Months Since Initial Liquid Acid Cleaning

plant operational changes. Around the 3rd month, the
operating SRT and  MLSS concentration decreased
from 2.0 to 1.7 and  from  about 1,000 mg/L to 700
mg/L,  respectively. The lowest aF(SOTE)  (about  3
percent)  occurred on October  17,  1986, when  the
MLVSS concentration was 409 mg/L and  the  SRT
was 1.2  days. The  second  operational  change that
resulted  in a lower overall  aF(SOTE) was when step
feeding was temporarily implemented  at  around  the
17th month of  the study.

Figure 8-20 compares disc  diffuser  performance in
basin  1   with  gas  cleaning  to  dome  diffuser
performance in Basin  3 without gas cleaning. Contrary
to the comparison between  Basins  2 and  3, gas
cleaning  the discs  resulted in  improved OTEs.

The  effects   on  OTE  of  lowering  the  MLSS
concentration  and  using  a step  feed  mode  of
operation also  noted for the disc diffusers. Figure 8-20
shows that when new domes were installed the OTEs
were very similar between the discs and domes. With
time,  the dome OTE  decreased to  below that of the
discs.  Because of the variability  in  the day-to-day
characteristics  and off-gas results, the effect of HCI
gas cleaning on OTE is not  obvious when only the
disc diffuser data  are observed. However, comparing
these results to the uncleaned dome diffuser results
shows a long-term benefit  of higher OTEs when gas
cleaning  is applied on a routine basis.

The disc diffuser system was operating  with an
aF(SOTE) of  8.5-9.0  percent before initial  cleaning
with the modified Milwaukee method.  The  domes
were operating at 6.5-7.5 percent. Since the two types
of diffusers  exhibited  similar  clean water  OTEs and
the discs had been operated under process conditions
16 months longer than the domes, it is very likely that
much of the difference in OTE may have been due to
gasket leakage with the dome  diffusers.

Overall aF(SOTE) just  after the  liquid acid cleaning,
was 10.2 percent for Basin  1 (discs) and averaged 10
percent for the two  basins  with  dome  diffusers.
Between  the 5th and 13th months  of operation, after
the low  MLSS  operating condition and  before the
dome  replacement, the  average aF(SOTE)s of the
disc and dome diffuser systems were 9.8 percent and
7.0 percent, respectively. This results in  average aF
values of 0.34 and 0.24, respectively.

V. ECONOMIC BENEFIT
The total  initial cost of the in-basin aeration equipment
installation  was  about $420,000. Additional costs  of
$13,000 for air filtration equipment and $13,500 for air
take-off piping resulted in a total retrofit expenditure of
$446,500. However, this did  not include the royalty
payment for the gas cleaning system. The average air
application rate  observed during the study was 3,923
L/s (8,312  scfm) at 574 L/s  (13.1  mgd). Based on
prior experience with coarse  bubble diffusers at this
plant, the OTE with disc  diffusers was about twice that
of coarse  bubble  aeration. Thus, the  required  air
application rate  was decreased by approximately one-
half  using  fine  pore  aeration.  Based  on  a  blower
transfer capacity of 16.5 L/s/kW (26.1  scfm/hp), this
results in a daily power savings of 5,700 kWh. With an
actual electricity cost  of $0.085/kWh,  the  daily
electrical  cost savings is $484, or $176,964/yr.

Diffuser cleaning costs must be considered in the
overall cost analysis. For these diffusers, the cleaning
cost was based  on employing HCI gas cleaning once
every 3 months and the modified Milwaukee method
on  the average of  once  every  2 years. (LACSD's
normal practice is tank top hosing every 6 months and
the modified Milwaukee method every 2 years).  Gas
cleaning costs were based  on conservative estimates
of 45 g (0.1  Ib)  HCI/diffuser and the need for 12 labor-
hr/tank at $20/hr. The  gas cost including delivery was
estimated  to  be $1.36/kg  ($3/lb).  The  modified
Milwaukee method was  estimated to require 4 labor-
days/tank.  The above  cleaning costs  amount  to
$16,680/yr.  The estimated net O&M  savings are
therefore $160,284/yr.

The energy  savings resulting  from  conversion to fine
pore  aeration  from coarse bubble aeration,  when
combined with  gas cleaning  costs, yields a simple
payback period of 2.8 years.  The relatively high  cost
of energy in Los Angeles is a significant factor in the
cost evaluation. Diffuser cleaning  has a relatively small
impact on the cost evaluation.
                                                  258

-------
Figure 8-20.  Comparison of ceramic dome and disc diffuser
             performance - Whittier Narrrows, CA.
  aF(SOTE),
    percent
    14 T
         Initial Liquid
         Acid Cleaning
    12-
    10-
       Basin 1 (Discs, HCI Gas Cleaned)
       Baatn 3 (Domes,  No Cleaning)
  LLJ
  o
     -2-
                 Dome Replacement
                              Step Feed
                               Operation
1,2,3
                         Gas Cleanings

                         1,2,3    1   1,261
 ,,                     ,        ,,

 y    4    v     v    v      v
1,2,3 B1
     B2
       -2  0  2   4  6   8  10  12 14 16  18 20  22 24 26
                Months Since Initial Liquid Acid Cleaning
                                                           259

-------
            8.3 Performance Evaluation by Means Other Than Off-Gas Testing
   8.3,? Cleveland Wastewater Treatment Plant

   LOCATION: Cleveland, Wisconsin

   OPERATING AGENCY: Village of Cleveland

   DESIGN FLOW: 7.9 Us (0.18 mgd)

   WASTEWATER: Domestic Plus Community College

   ORIGINAL AERATION SYSTEM: Coarse Bubble Diffusers

   FINE PORE AERATION SYSTEM: EDI Reef IV Rigid Porous Plastic Plate Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION: 1986

   BASIS OF PERFORMANCE EVALUATION: Overall Process Performance

   CLEANING METHOD: No Cleaning Necessary to Date	
1. HISTORICAL BACKGROUND
The  Village of  Cleveland,  Wl,  population  1,425,
operates a wastewater treatment plant that receives
wastewater from its residents and 2,000 community
college students. The  plant was first operated  in the
contact stabilization  mode. When  the mode  of
operation  was  changed to extended  aeration  to
achieve improved BOD removal and nitrification,
sufficient DO could not be maintained in the aeration
tank.

To improve oxygen transfer performance,  the pipe
mounted coarse  bubble  diffusers were removed and
replaced with porous plastic plate fine pore  diffusers.
The 20-yr old blowers  and motors were also replaced
with posilive displacement units.

The activated sludge system consists of an 11-m (36-
ft) diameter package plant with a 5.5-m (18-ft) clarifier
in the center. The volume of the aeration bays  is 345
m3 (91,000  gal), divided inlo three sections of 129,
144,  and  72 m3 (34,000, 38,000, and 19,000 gal)
each. The aerobic sludge digestion tank volume is  95
m3 (25,000  gal). Row to the plant averages 7.9 Us
(0.18  mgd). Inlluent  BOD5,  TSS, and  NH3-N
concentrations  average  173,  165,  and  30  mg/L,
respectively. Prior to  the retrofit,  effluent BODg was
30-40  mg/L. Operating  parameters  in  the  contact
stabilization  mode were an MLSS  inventory of about
1,590 kg (3,500  Ib) and a target mixed liquor  DO of
2.0  mg/L. Two  11.2-kW (15-hp) centrifugal blowers,
one a standby, could each deliver 142 Us (300 scfm)
air.

II. FINE  PORE AERATION SYSTEM DESIGN
DESCRIPTION
The new rigid porous plastic  plate diffusers, each 1.2
m long x 46 cm wide x 6.4 mm thick (48 in x 18 in x
1/4 in) housed on top of a thin plastic box to which air
is  fed by  a  flexible  hose,  were installed  in  January
1986. Piping used for  the coarse bubble aeration
system was  reused in part for the new system. Ten
new diffusers were  installed on  the  bottom inside
periphery of  the aeration tank, and two were installed
in the aerobic digester, a 4.9-m  x 4.9-m (16-ft x 16-ft)
tank butting  against the  circular package plant.  Equal
water depth  over  the diffusers  is  maintained in the
aeration and  digestion tanks. The two diffusers  in the
hopper-bottom digestion tank were placed on stilts  to
achieve  the  equitable  depth. A cross roll aeration
pattern  is achieved with  the  fine  pore diffuser
configuration. The retrofitted plant  is operated as an
extended aeration system  with  10-14 hr HRT, an
MLSS concentration  of 1,800 mg/L, and DO levels  of
1.5-2.5 mg/L in the aeration tank and 1.0-1.5 mg/L in
the digestion tank.

The installed cost of  the retrofit  system was $11,500,
including  design, equipment, and  labor.  Plant
personnel designed and  installed  the  system.
Hardware  expenditures  totaled  $8,060,  including
retrofitting  the  piping   and purchasing  the  new
diffusers, blowers, and motors.

III.  PERFORMANCE
The new  system is reported to be performing
effectively  with no maintenance problems  in over 2
years of operation.  Average effluent  characteristics
are as follows:  BOD5  =  6.4  mg/L, CBOD5  = 2.8
mg/L, TSS = 5.1 mg/L, and NHa-N = trace amounts,
Diffusers are inspected periodically, and only a slight
growth on 3-5 percent of the media surface has been
observed.  Experiments have shown that,  if desired,
the growth can be easily removed with a 10-percent
solution of sulfuric acid.
                                                260

-------
IV. COST SAVINGS
Savings in electricity are reported  to  be $2,400/yr,
which  equates to  a simple payback period  of just
under  5 years. Normally,  one blower is  operated at
130 Us (275 scfm) and 55 kPa (8 psi). According to
operating  personnel, the old  coarse bubble aeration
system, including blowers  and motors, had expended
its useful  life and was in need of replacement at the
time the fine pore system was installed.
                                                 261

-------
   8.3.2 Plymouth Wastewater Treatment Plant

   LOCATION: Plymouth, Wisconsin

   OPERATING AGENCY: Plymouth Utilities

   DESIGN FLOW: 72 Us (1.65 mgd)

   WASTEWATER: Municipal Plus Significant Cheese Manufacturing Wastes

   ORIGINAL AERATION SYSTEM: Surface Aerators

   FINE PORE AERATION SYSTEM: Sanitaire Ceramic Disc Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION: 1986                   .   •  '

   BASIS OF PERFORMANCE EVALUATION: Overall Process Performance; Power Savings

   CLEANING METHOD: In-Situ HCI Gas Cleaning System Provided	
1. HISTORICAL BACKGROUND
The  Plymouth, Wl  activated  sludge  plant operated
from 1978 through 1985 using three aeration basins,
each equipped with  one 30-kW (40-hp) mechanical
surface aerator. The plant receives wastewater from
the Cily, population 7,000,  plus industrial wastes from
three cheese  manufacturers. The  domestic/industry
organic loading split is about 1:1.
Energy use  was not the major factor in Plymouth's
decision to retrofit their  plant with fine  pore ceramic
discs. The reasons  included: a) the need  in 1980 to
provide nitrification  in a system designed only  for
carbonaceous  BOD removal, b) the  inability of the
surface aeration system to consistently maintain solids
in  suspension, which resulted in varying MLSS levels
and effluent  quality,  and c) mechanical problems and
high  repair costs associated with the surface aerators.

The  plant  consists  of grit  removal,  primary
clarification,  activated sludge treatment,  secondary
clarification,  tertiary  filtration, and disinfection. Pickle
liquor (ferrous  chloride) is added to the aeration basin
for phosphorus removal. Sludge  is anaerobically
digested, and  digester supernatant and tertiary filter
backwash are returned to  the primary  clarifiers.
Effluent standards (summer/winter) are as follows:
m*20 mg/L BOD5, 10/20 mg/L TSS, 1/9 mg/L NH3-N,
and 1/1 mg/L TP. The plant received average loadings
in  1986 of 57 Us (1.3 mgd) flow, 990  kg  (2,182 Ib)
BODg/d, 689 kg (1,518 Ib) TSS/d, 50 kg  (111 Ib) NH3-
N/d,  and 29 (64 Ib) TP/d. The system is normally
operated at an MLSS of 3,000-3,500 mg/L and an F/M
loading of 0.11-0.13  cH.

The  three aeration basins, each 12.2 m x  12.2 m x
4.6 m  (40 ft x 40 ft x 15 ft)  SWD, were designed to
operate as a contact stabilization system in either the
complete mix  or  plug flow mode. The system was
designed to  treat 1,051  kg  (2,317 Ib)  BOD5/d from
primary clarification plus  340  kg  (750 Ib) BOD5/d from
digester supernatant. Annual power costs for aeration
were estimated at $37,200 based on 45 amps current
draw/surface aerator, 191  days  with  two aerators
operating,  174 days with three   aerators operating,
and a power cost of $0.0526/kWh.

Annual  energy savings from  installing  fine  pore
diffusers was estimated  at  $28,200. These savings
were  based on anticipated  lower  BODs loads than
included in the original design plus an average NH3-N
load of 60  kg (133 lb)/d. Average mixed liquor DO for
the retrofitted system was assumed to be 2.5 mg/L,
with aF estimated at 0.6.

II. FINE  PORE  AERATION  SYSTEM  DESIGN
DESCRIPTION
The new aeration system consisted of 1,350 23-cm
(9-in)  diameter ceramic fine pore discs with a ratio of
tank  floor area to diffuser surface area of 8.6 and  a
new aeration building containing three 37-kW (50-hp)
positive displacement  blowers,  each  capable  of
delivering 400 Us (850 scfm) air. The design criteria
were: flow  =  72 Us (1.65  mgd), BOD5  =  615 kg
(1,356 lb)/d, NH3-N = 136 kg (300 lb)/d, total oxygen
requirement = 1.15 kg 02/kg BOD5 applied + 4.6 kg
02/kg  NHa-N  applied, and airflow rate  =  0.63 L/s
(1.33  scfm)/diffuser.  Each blower was expandable to
802 Us (1,700 scfm) by upgrading the motor to 56 kW
(75 hp). Each aeration  basin  received 450 ceramic
disc diffusers, equally spaced at 66 cm (26 in) o.c. In-
situ gas cleaning capability  and  pressure monitoring
systems were also provided. Figure 8-21 shows the
Plymouth fine pore aeration system.

III. SYSTEM PERFORMANCE
Following the retrofit, average effluent BOD5 and TSS
concentrations for 1986  were 3  mg/L each.  MLSS
were  consistently maintained  between  3,000 and
3,500 mg/L.  Aeration  tank  temperatures have
averaged about 5°F higher  than before  the  retrofit,
                                                262

-------
Figure 8-21.  Air header distribution system used to retrofit fine pore ceramic discs into a tank formerly equipped with surface
            aeration - Plymouth, Wl.
               "V'* t  • «'*•&/<*
                 "'  ,jt>f '"
                WW\ . f » «
               «">/M
                                                          263

-------
and mixed liquor DO has been maintained between 5
and 6 mg/L BODg and NHa-N loadings were 6.5 and
8,2 percent higher for  the 12 months prior  to the
retrofit than during the 12 months following the retrofit.

IV. COST SAVINGS
The capita! cost of the system included a 6.1-m x 11-
m (20-ft  x 36-ft) building  ($89,628), labor for
installation of blowers and diffusers ($36,702), labor
for  removing old aerators  ($2,370), diffusers
($46,720),  blowers ($33,294),  engineering,  and  a
change order credit for a total cost of $224,207.

Power savings were based on operating two blowers
5 days each week drawing 55 amps and one  blower
on weekends drawing 35 amps. This resulted  in an
annual savings of  383,240  kWh compared  with the
previous year. The annual power savings under these
conditions were  $20,160, or  about  20  percent less
than the estimated $28,200.  Based on a continued
energy savings  of $20,l60/yr, the system  has  a
simple payback period of 11.1 years.
                                                264

-------
   8.3.3 Renton Wastewater Treatment Plant

   LOCATION: Renton, Washington

   OPERATING AGENCY: Seattle Metro

   DESIGN FLOW: 3,155 Us (72 mgd)

   WASTEWATER: Domestic

   ORIGINAL AERATION SYSTEM: Coarse Bubble Diffusers

   FINE PORE AERATION SYSTEM: Wyss Perforated (Flexible) Membrane Tube Diffusers

   YEAR FINE  PORE SYSTEM PLACED IN OPERATION: 1983

   BASIS OF PERFORMANCE EVALUATION: Overall Process Performance and Air Usage

   CLEANING METHOD: Occasional Hosing	
I. HISTORICAL BACKGROUND
The  Renton, WA  Wastewater Treatment  Plant was
placed in  operation in  1965 to  serve the  urban and
suburban  areas lying to the east,  south and  north of
Lake Washington, just east of Seattle.  The plant was
originally designed  in anticipation of future incremental
expansion. The capacity of the original facility was
1,052 L/s  (24 mgd) . This was expanded to 1,578 Us
(36 mgd) during 1972-1974 and was doubled  to 3,156
L/s (72 mgd) in late 1987.

Rising  power costs led to  an evaluation of  possible
modifications of the existing  coarse bubble  diffused
aeration system in  1982. 45-50 percent of the plant's
energy  usage could be attributed  to  the  aeration
blowers.  Plant   power  costs  increased  from
approximately $10,000/mo in  1979 to $46,000/mo in
1981. Based on  projected increases  in  anticipated
flow  and  power  rates, annual  power  costs were
expected to  increase from approximately $550,000 in
1981  to  $1,150,000  by  1986.  Consequently, a
significant  cost   savings  could  be  realized  by
conversion to a more efficient aeration system.

The  Renton plant  is a  conventional activated sludge
plant designed to  meet an  effluent TSS limit of <30
mg/L. SRT  is  controlled  at 2-3  days to  prevent
nitrification and minimize energy needs. Plant influent
is  screened before passing to  preaeration, grit
removal, and primary sedimentation.

The  plant's  three  aeration tanks each contain four
passes 9 m  (30 ft) wide by  111 m (317.5 ft) long with
an SWD of 4.6 m (15 ft). HRT at average wet weather
flow  (December-April) is 4.3 hr.  The original coarse
bubble diffusers were mounted  as sidewall diffusers
on swing arms.  The aeration system is equipped with
six single-stage centrifugal blowers that discharge into
a common air header at a discharge pressure of 153
kPa  (7.5  psig).  Two of the blowers have  a  nominal
capacity of 5,665 L/s  (12,000 scfm) each, while  the
other four  have a capacity of 6,610 L/s (14,000 scfm)
each.  An  automatic DO control  system  is used  to
adjust the airflow to each aeration tank. The  blower
inlet vanes are also throttled in response to  pressure
fluctuations in the air supply system.

The plant is  equipped with  four peripheral-feed
secondary clarifiers, each with a  diameter of 30  m
(100 ft) and an SWD of 4.3  m (14.2 ft), in addition to
12  30-m  (100-ft) diameter  center-feed  secondary
clarifiers. The secondary clarifier overflow  rate at  the
3,155-L/s (72-mgd) average  wet weather flow is 23.3
m3/m2/d (572  gpd/sq  ft). Effluent  is  discharged  to
Puget Sound.

Prior to October  1987, waste primary and secondary
sludges  were discharged to a trunk line to the main
Seattle  Metro 6,570-L/s (150-mgd)  plant.  Since
October 1987, sludge  handling facilities  have been
placed in operation and consist of diffused  air flotation
thickening, anaerobic digestion,  and belt  filter press
dewatering of primary and secondary sludges.

II. FINE PORE AERATION SYSTEM RETROFIT
Wyss  perforated membrane tube diffusers were
selected for in-plant evaluation in  1982.  They were
installed in the first two passes of one aeration tank.
This tank  was operated under  identical influent feed
rates and DO levels as the first two  passes of another
aeration  tank containing coarse bubble diffusers.  The
performance of the  two systems  was compared by
measuring the airflow rates needed to control DO  to 2
mg/L  and  indirectly  determining  OTEs. OTEs were
determined by  comparing oxygen consumption rates,
calculated  from measured basin oxygen uptake rates,
to air application rates. These tests  indicated  the
perforated  membrane diffuser system requires 30-40
percent  less  air  than  the  coarse bubble diffuser
system to maintain comparable DO levels.
                                                 265

-------
Tablo 8-33. Monthly Average Performance - Renton, WA
Influent MLSS2,
Month Flow, mgd BOD51,mg/L mg/L
January 1087
February
March
A [Mi!
May
Juno
July
August
September
October
November
December
Average
61.6
59.4
57.3
46.6
45.4
45.7
43.6
43.0
41,2
45.3
43.0
51.0
48.6
145
145
145
163
150
169
170
161
147
143
160
138
158
798
674
758
730
655
751
840
831
823
824
703
795
803
Efiluenl, mg/L
SRT, days
3.1
2.5
2.6
2.4
2.2
2.1
2.4
2.4
2.4
2.3
2.2
3.8
2.5
BODS
11,6
10.4
11.6
12.6
14.2
14.0
14.3
9.0
10.9
12.3
19.0
16.2
13.3
TSS
8.7
7.7
7.6
11.0
11.4
10.7
8.9
8,7
7.4
9.8
16.6
9.8
10.0
- cu ft Air/lb
BODS'
693
631
642
747
849
761
893
842
1,018
856
844
974
813
Eslimaied
OTE3, percenl
8.3
9.1
9.0
7.7
6.8
7.6
6.4
6.8
5.7
6.7
6.8
5.9
7.2
 ' BQD5 to secondary system.
 2 From Tank 2.
 3 Assumes 0.0272 tb O-j/cu ft air and 1.0 Ib O2/Ib BODR.
Based on the actual retrofit cost  of  approximately
$380,000, a simple payback  period  of  <5 years was
expected. By early 1983, the entire  system had been
converted to sidewali mounted perforated membrane
tube diffusers with swing  arms for expedient diffuser
removal and cleaning. The existing  blowers  and  air
piping system were acceptable. Approximately 1,000
diffusers were installed  in  each pass.  The airflow
rate/diffuser at average loads  was estimated to be 0.9-
1.4 Us  (2-3 scfm), with  a peak rate  of 2.4  Us (5
scfm).

III. PERFORMANCE
Power consumption at  the  Renton  facility has been
carefully monitored. Efforts were made from 1979 to
1982 to  reduce  plant  power consumption  by
operational  changes. Total plant  power use was
decreased  from  390 to  355  kW/1,000  m3  after
installation of the fine pore aeration system. Assuming
an average annual flow of 2,130 Us (48.6 mgd) and a
power cost of $0.04/kWh results in  an  annual energy
savings of about $92,500/yr and a simple payback
period of slightly more than 4 years.

Table 8-33 summarizes plant performance for 1987.
The  annual  average SRT was about 2.5 days, and
secondary   clarifier  effluent BOD5  and  TSS
concentrations  averaged   13.3 and  10.0  mg/L,
respectively.  Based on the  reported air supplied per
unit of BOD5 removed in the secondary system, OTEs
were estimated at 5.7-9.1  percent, with an average of
7.2 percent. These  estimates  assume  an  oxygen
consumption of 1.0 kg O2/kg  BOD5 removed.
                                                266

-------
   8.3.4 Ripon Wastewater Treatment Plant

   LOCATION; Ripon, Wisconsin                 ',

   OPERATING AGENCY: City of Ripon - • '

   DESIGN FLOW: 87.7 Us (2.0 mgd)

   WASTEWATER: 80 percent Municipal, 15 percent Industrial (Food Processing)

   ORIGINAL AERATION SYSTEM: Coarse Bubble Diffusers

   FINE PORE AERATION SYSTEM: Sanitaire Ceramic Disc Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION: 1986

   BASIS OF PERFORMANCE EVALUATION: Overall Process Performance

   CLEANING METHOD: Preventive Maintenance; In-Situ HCI Gas Cleaning   ._
I. HISTORICAL BACKGROUND
Prior to 1986, the City of Ripon, Wl operated two 31-
m (101-ft) diameter Walker Process Package Plants
constructed  in  1974-76.  Ripon,  population  7,200,
treats municipal wastewater and industrial wastes from
an appliance manufacturer and  four food  processing
plants. The industrial waste represents 15 percent  of
the flow and 50  percent of  the  organic  loading
received by the plant.

In an  effort to improve energy efficiency and replace
aging equipment, the original coarse bubble diffuser
system was  replaced  with fine pore  ceramic  disc
diffusers capable of being cleaned with HCL gas. The
installation was completed in 2 weeks and placed on-
line January 1, 1986. It was estimated in 1985 that the
aeration retrofit (including new coarse bubble units in
the aerobic  digester) would  save  $30,700/yr; these
savings would result from reduced blower operation.

Prior to the  retrofit, two blowers were used 8 months
of the  year and 1-1/2 blowers were used 4 months of
the year.  With the  blower  inlet  throttle  open  100
percent, each blower consumed 140 amps or 105.6
kW  (141.6  hp).  At an  average  power cost  of
$0.04/kWh, annual costs for power for both  aeration
and aerobic digestion were estimated to be $64,000.
The  new  diffusers were  estimated to require  one
blower year round at 150 amps. Maintenance of the
new diffusers was estimated at $2,400/yr  (two gas
cleanings/yr).  The  project was estimated  to  cost
$60,000, and  the  estimated  simple payback period
was 2.1 years.  Both ceramic  disc and tube fine  pore
diffusers  were  compared for  operating cost,
maintenance, initial cost,  and  life expectancy,  with
discs being chosen.

The treatment facility contains grit removal, secondary
aeration and  denitrification,  aerobic  digestion,
phosphorus  removal  using  pickle liquor  addition,
 filtration, and  chlorination. Effluent standards  are  10
 mg/L BOD5, 10 mg/L TSS, 5 rng/L NH3-N, and 1 mg/L
 P. In 1986, the plant received loadings of 83.2 L/s (1.9
 mgd)  flow,  1,199 kg  (2,643 lb)/d  BQD5,  1,309  kg
 (2,885  tb)/d  TSS, and 30  kg  (66, lb')/d  TP. Plant.
 effluent  concentrations in  1986  averaged  6 mg/L
. BOD5, 7 mg/L TSS, and less"than '1 mg/L NH3-N.  An
 F/M loading of 0.15 d-1 and a MLSS level of 3,500-
 4,000 mg/L  were reported as  normal operating
 conditions.       '                •     .

 II.  FINE PORE  AERATION  SYSTEM  DESIGN
 DESCRIPTION
 The  new aeration system for  each package plant
 consisted of 540 23-cm (9-in) diameter ceramic  fine
 pore disc diffusers equally spaced  in six grids. The
 new  system  for the  aerobic digester  consisted  of
 stainless steel drop pipes and stainless steel coarse
 bubble diffusers. The.digester diffusers were set  30
 cm  (12  in) above the tank  floor,  and the ceramic
 diffusers in  the  aeration tank were set,23 cm (9  in)
 above the tank floor. Equivalent headlosses in the two
 systems, were achieved by  placing  an  orifice in the
 line feeding air to the coarse, bubble'system. Figure 8-
 22 shows the disc pattern during a check for leaks in
 the air distribution system.

 Design conditions for the aeration portion of the Ripon
 plant included the following: flow  =   87.7  L/s (2.0
 mgd), BOD5 = 1,515 kg (3,340 lb)/d, NH3-N = 76 kg
 (167 lb)/d, total oxygen requirements  = 1 kg Og/kg
 BOD5 applied + 4.5  kg Oa/kg NHa-N applied, and a
 =  0.55. This resulted in a design selection of 1,092
 diffusers with an AD/AT of  0.08. The  design airflow
 rate was 0.66 L/s (1.4 scfm)/diffuser.

 III. SYSTEM PERFORMANCE
 In the 17 months of operation reported,  only one 112-
 kW (150-hp)  centrifugal  blower  has been  needed.
 Mixed liquor DO levels normally vary between 4 and 6
                                                 267

-------
Figure 8-22. Chocking for leaks and uniform air distribution - Ripon, Wl.
mg/L, which are similar to  the levels maintained prior
to the  retrofit. During the hottest period, mixed liquor
DO dropped to below 2 mg/L and it was reported to
be  difficult  to  maintain aerobic  conditions  in  the
digester.  Although  treatment efficiency  remained
satisfactory, odors developed in the  digester  and
steps were taken to supply auxiliary air to the digester
to prevent odors in the future.

An analysis of plant performance for periods of 1 year
prior to the retrofit and 1  year following the retrofit
reveals an annual power  savings  of  853,800 kWh.
This is equivalent to an annual savings of $29,318.
Although close  to the projected  annual savings of
$30,700, the  cost of  energy used in the estimate
($Q.04;kWh) was lower by  about 10 percent. Ripon's
energy bill has six components based on demand time
and  reserve  arrangements. Energy costs  averaged
$0.041/kWh for the period prior to the  retrofit  and
$0.044,kWh for the period following the retrofit.  In the
year prior to the  retrofit, 2,630,400 kWh were used
comapred to 1,776,000 kWh  in the year following the
retrofit.

Gas cleaning will be employed for purely preventative
reasons at a frequency of once per year or less. DWP
measurements taken following the retrofit indicated no
significant increases in the first 17 months.
IV. COST SAVINGS
Based on actual  performance,  allowing for  one gas
cleaning yearly  ($1,200),  and  using  the  actual
equipment and installation costs ($55,352), the simple
payback period is  approximately 2 years.
                                                  268

-------
   8.3-5 Saukville Wastewater Treatment Plant

   LOCATION; Saukville, Wisconsin

   OPERATING AGENCY: Village of Saukville

   DESIGN FLOW: 21.9 L/s (0.5  mgd)

   WASTEWATER: Municipal with Significant Butter Manufacturing Waste

   ORIGINAL AERATION SYSTEM: Sanitaire Stainless Coarse Bubble Diffusers

   FINE PORE AERATION SYSTEM: Sanitaire Ceramic Disc Diffusers

   YEAR FINE PORE SYSTEM PLACED IN OPERATION: 1985

   BASIS OF PERFORMANCE EVALUATION: Overall Process Performance and DWP Monitoring of Diffusers

   CLEANING METHOD: In-Situ  HCI Gas Cleaning     ,	
I. HISTORICAL BACKGROUND
The original Saukville treatment facility, constructed in
1982, consisted of comminution, grit  removal, two
23.3-m  (76.5-ft)  diameter Sanitaire package plants
with aeration, clarification,  aerobic digestion, and
disinfection. Alum is added to the aeration basins for
phosphorus removal. The aeration compartment is 5,3
m (17.5 ft) wide with an SWD of 4.9 m (16 ft). Each
package plant was designed for an organic loading of
380 kg  (834 Ib)  BOD5/d and can be operated in the
contact stabilization  or  plug  flow  mode.  Effluent
standards  are 30 mg/L for BOD5 and TSS  and 1 mg/L
for  P.

The original aeration system  was comprosed  of
stainless steel  coarse bubble  diffusers.  Each drop
pipe was  equipped  with  four  wide-band  diffusers
causing a  cross roll pattern around the circumference
of the aeration tanks. Three 637-L/s  (1,350-scfm), 56-
kW (75-hp) centrifugal blowers were  available, one for
each plant with the third being a standby unit. One
19.7-m3 (5,200-gal)  tank  truck  of; butter waste
averaging  40,000 mg/L BOD5 was gradually  pumped
into the splitter box between the two plants each day,
generally over a 12-hr period.

Beginning  in 1984, the plant received , an additional
butter manufacturing  waste  load. While treating  the
butter waste, the plant was operated in a plug flow
mode at an  F/M loading of about,  0.25  d'1  and an
MLSS concentration'of 4,000 mg/L.

Projected energy savings were the principal reason for
retrofitting one of the plants with  fine bubble diffusers ,
in 1985. Estimates of  50  percent reduction in energy
due to the retrofit meant that 40  amps/plant could be
saved. This  was equivalent to  22,000 kWh/mo  or
about $12,500/yr/plant. Energy costs were estimated
at $0.049/kWh. Prior  to the retrofit, each plant's
blower cost about $25,000/yr to operate.

H.  FINE  PORE  AERATION  SYSTEM  DESIGN
DESCRIPTION
In 1985, one of the two package  plants was converted
to fine pore aeration.  The retrofitted  plant contained
270 23-cm (9-in) diameter ceramic disc diffusers in a
tapered  grid pattern to  provide a higher density  of
diffusers at the inlet end. Design flow to the retrofitted
plant was  21.9 L/s (0.5 mgd), and design BOD5 load
was 380 kg (834 lb)/d, although the actual ,BOD5 load
proved to  be  about 535  kg (1,180 lb)/d.  Design
oxygen requirements were based on 1  kg  Oa/kg BODs
and an aF of 0.5. AD/AT was 0.064,  and the design
airflow rate was 0.52  L/s (1.1 scfm)/diffuser, or 145
L/s  (308 scfm).

A 340-L/s  (720-scfm), 3Q-kW  (40-hp)  positive
displacement blower was installed to provide air for
the retrofitted  aeration and  aerobic digestion zones.
Air  piping  was installed to  either, isolate the  new
smaller  blower from  the centrifugal  air distribution
system or interconnect the  two  systems  so that one
centrifugal  arid one  positive displacement blower
could  supply air to both plants.

The retrofit was engineered,  furnished, and installed in
5 months  at a cost  of $43,990. In-situ HCI  gas
cleaning  and  DWP  monitoring capability  were
provided. Gleaning  costs were estimated  at $1,920/yr
for  three cleanings. The simple  payback period was
estimated  at 4.2 years. Lease purchase financing was
used at an interest rate of 10.49 percent for 5 years.

III. SYSTEM PERFORMANCE
By  retrofitting only one plant, observations could be
made about relative performance of the two systems.
                                                 269

-------
Differences were noted  in sludge volume index (SVI),
mixed liquor DO, and diffuser fouling.

During the first 5 months  operation,  SVI increased
slightly in the retrofitted plant. SVI then: 1) increased
markedly, 2)  decreased upon cleaning the fine pore
diffusers with HCL gas, and 3) increased significantly
within  2 months  of  cleaning.  Subsequent gas
cleanings reduced SVI only for relatively short periods
following cleaning. The  high SVIs were believed due,
in part, to the lower DO levels in the new system.

In May  1986,  6 months following start-up  of  the
retrofitted plant,  DO  could  not be  maintained  at  1
mg/L during the 12 hours when the butter waste was
being processed. To  gain  control, two  centrifugal
blowers  were  employed  at that  time.  With  the
increased airflow and under maximum load conditions,
the retrofitted plant  maintained  a  DO of  1-2  mg/L,
while the coarse bubble plant maintained a DO  of 2-3
mg/L.  Because the coarse bubble diffusers were
pjaced 25-76 mm (1-3  in) higher  than the fine pore
diffusers,  a separate bypass pipe was installed  to
dedicate the positive displacement  blower to  the fine
pore system and allow the centrifugal blower to feed
the coarse  bubble aeration and  aerobic digestion
2ones and the digester zone of the fine pore  system.
However, the  redistribution  of air did not completely
remedy the DO problem.

Other  changes  made  at  the  plant  included:  1)
increasing the feeding period of the butter  waste from
12 to 24 hours, and 2)  increasing airflow to the head
end  of the fine pore aeration tank  by  cracking open
one of the coarse bubble droplegs. It was believed the
increased turbulence at  the head end of the fine pore
system  would improve a  by  breaking  up fat-
encapsulated bubbles  thought to  be  the result  of
adding the butter waste. As a result of these  and the
changes mentioned previously, the  fine pore  aeration
system SVI gradually stabilized.

DWP monitoring  showed  that the  fine pore  diffuser
system was subject  to  rapid  and  heavy  fouling.
Cleaning, initially estimated as necessary 3-4 times a
year, was required four times  in the first 7 months of
operation. Gleaning was considered necessary  when:
1) DWP was high, 2) coarse bubbling was noticeable,
and 3) DO could  no longer be maintained at minimum
levels. The fine  pore diffusers in the contact and
reaeralton zones were cleaned five and  six  times,
respectively,  during  1986.  Data  on DWP  were
available before and after  four of these cleanings and
are presented in Table 8-34.
Table 8-34.  DWP Monitoring Results - Saukville, Wl
Basin
Contact
Zone


Reaeralion
Zone


'Date
1/30/86
2/27/86
7/29/86
12/15/86'
1/30/86
2/27/86
7/29/86
• 12/15/86
Cleaning Time,
minutes
38
99
60
30
26
78
105
44
DWP, in
Before
35
13
40
29
46
12
40
9.5
After
5 "
6
15
9-;,. ;_
4.5 ''
''6.5',
7.5
. 6.5- '
IV. COST SAVINGS
A comparison of power use for the 12 months prior to
and following the retrofit shows a reduction of 14,225
kWh/mo  and an annual savings of $7,300. The fact
that actual savings ($7,300) were less than predicted
savings  ($12,500)  was attributed to:  1) increased
electric rates following the retrofit,  2) initial poor air
distribution that  -required the running of two 56-kW
(75-hp) engines instead of one 30-kW (40-hp) and one
56-kW (75-hp)  unit, a  situation  that was remedied
during the first  12 months of retrofit operation,  3)
increased BOD loadings (26.7 percent additional) from
the butter waste, 4) a shift  in the  on-peak/off-peak
utilization of  energy  by  the  plant,  and  5) the
implementation  of a  plant-wide  energy  savings
program  resulting from an  energy audit by the local
utility  authority early in the 12-month period prior  to
the retrofit.             •
                                                  270

-------
8.4 Sources of Information

The following reports and  publications were used to
develop the case histories presented in this chapter:

 Aeration Technologies, Inc. Off-Gas Analysis Results
 and  Fine Pore  Retrofit Information  for Glastonbury,
 CT Facility,  Aeration Tank No. 2. Study conducted
 under  Cooperative Agreement  CR812167,  Risk
 Reduction  Engineering   Laboratory,  U.S.
 Environmental Protection Agency, Cincinnati, OH (to
 be published).

 Aeration Technologies, Inc. Off-Gas Analysis Results
 and  Fine Pore Retrofit Case History for Hartford, CT
 MDC Facility. Study conducted  under  Cooperative
 Agreement CR812167, Risk  Reuction  Engineering
 Laboratory,  U.S.  Environmental Protection Agency,
 Cincinnati, OH (to be published).

 Boyle, W.C. Oxygen Transfer Studies at the Madison
 Metropolitan Sewerage  District  Facilities.  Study
 conducted under Cooperative Agreement CR812167,
 Risk Reuction  Engineering  Laboratory,  U.S.
 Environmental Protection Agency, Cincinnati, OH (to
 be published).

 Donohue &  Assoc., Inc. Fine Pore Diffuser System
 Evaluation for the Green Bay Metropolitan Sewerage
 District.  Study  conducted  under Cooperative
 Agreement CR812167, Risk  Reduction  Engineering
 Laboratory,  U.S.  Environmental Protection Agency,
 Cincinnati, OH (to be published).

 Ernest, L.A.  Case  History Report on Milwaukee
 Ceramic Plate  Aeration Facilities. Study conducted
 under Cooperative Agreement  CR812167,  Risk
 Reduction  Engineering   Laboratory,  U.S.
 Environmental Protection Agency, Cincinnati, OH (to
 be published).

 Huibregtse,   G.L.,  T.C.   Rooney  and  D.C.
 Rasmussen.  Factors Effecting Fine Bubble Diffused
 Aeration. JWPCF 55(8): 1,057-1,064, 1983.

 Leary, R.D.,  L.A.  Ernest and W.J. Katz.  Effect of
 Oxygen-Transfer Capabilities  on  Wastewater
 Treatment Plant Performance. JWPCF  40(7): 1,298-
 1,310, 1968.

 Leary, R.D.,  L.A. Ernest  and  W.J.  Katz. Full Scale
 Oxygen Transfer Studies of Seven Diffuser Systems.
 JWPCF 41 (3):459-473,  1969.

 McNamee, Porter &  Seeley  Engineers/Architects.
 Fine  Pore  Diffuser Case  History for Frankenmuth,
 Ml.  Study conducted under Cooperative Agreement
 CR812167,  Risk  Reduction Engineering Laboratory,
 U.S.  Environmental Protection  Agency,  Cincinnati,
 OH (to be published).
 Mueller, J.A.  Case  History of Fine Pore  Diffuser
 Retrofit at Ridgewood, NJ. Study conducted  under
 Cooperative Agreement CR812167,  Risk Reduction
 Engineering  Laboratory, U.S.   Environmental
 Protection Agency, Cincinnati, OH (to be published).

 Stenstrom, MiK. Fine Pore  Diffuser Fouling: The Los
 Angeles  Studies.  Study  conducted  under
 Cooperative Agreement CR812167,  Risk Reduction
 Engineering  Laboratory, U.S.   Environmental
 Protection Agency, Cincinnati, OH (to be published).

 Vik, T.E.  Documented Energy Savings from Aeration
 System Retrofits.  Presented  at the  60th Annual
 Meeting of the Central States Water  Pollution Control
 Association, May 1987.

 Warriner, R. Oxygen Transfer Efficiency Surveys at
 the Jones Island East Plant, August  1985-June 1988.
 Study  conducted under  Cooperative  Agreement
 CR812167, Risk Reduction Engineering  Laboratory,
 U.S. Environmental  Protection Agency, Cincinnati,
 OH (to be published).

In  addition, the following individuals provided
information on  specific facilities described in  this
chapter and their  assistance  is greatly appreciated:

 Richard Finger,  Seattle Metro Wastewater Treatment
 Facility, Renton, WA.

 Bruce Neerhof, Village of Cleveland,  Wl.

 S.R.  Reusser,  Madison  Metropolitan  Sewerage
 District, Madison, Wl.
 Thomas  E.  Vik,
 Menasha, Wl.
McMahon  Associates,  Inc.,
                                                271

-------

-------
                                            Appendix A
                      Abstracts of Contractor Studies Completed Under
                             EPA/ASCE Fine Pore Aeration Project
Contractor

Michigan Technological
University
(C. Robert Baillod)
Southern Methodist
University
(Edwin L, Barnhart)
Madison MSD
(William C. Boyle)
University of Calgary
(J. William Costerton)
Lawrence A. Ernest
Milwaukee MSD
(Read Warriner)
AERTEC, Inc.
porous (R. Gary Gilbert)
Study Description

Report: Author of Quality Assurance ProgramPlan (QUAPP); also
involved  in training  project contractors  and coordinating QA
procedures for use during project (6-9 months).

Report/lnterplant Fouling Study Director: Planning, development
of protocol, liaison with contractors,  review of data, and final
report  relative  to  interplant control study  (four-lungers)  (24-30
months).

Report/(1)  North  Texas  MWD; (2) Trinity River  Authority:
(1)  Off-gas testing in the field  and evaluation of five ceramic
dome diffuser cleaning methods in laboratory.  Laboratory  testing
was performed using clean  water with and  without detergent
addition (18-24 months); (2)  Off-gas testing  of coarse  bubble
diffusers and fine pore ceramic disc diffusers (12-18 months).

Liaison/consultation with  EPA to provide economic information to
develop life-cycle cost analysis.

Report/Laboratory Study: Characterization  of various types  of
diffusers from 20-30 plants (18-24 months).

Report/Madison: Full-scale evaluation  of ceramic dome and disc
diffuser performance under various process  conditions  (18-24
months).

Report: Investigations of fine pore diffuser  biofouling. Evaluation
of  relationship between  biofilm growth  characteristics and
diffuser operating  performance at seven contractor sites  (12-15
months).

Report/Case  Histories:  Operating,  diffuser cleaning,  and
performance experiences  (over  extended  periods) of ceramic
plate diffusers  at Milwaukee Jones Island East and West Plants
and South Shore Plant.

Report/Jones Island  East and West Plants and  South  Shore
Plant.  In-situ performance evaluation of old/new  ceramic plate
diffusers using off-gas testing (12-27 months).

Report/Qlastonbury:  In-situ  performance  evaluation of  rigid
plastic tube diffusers using off-gas testing (24-30 months).
                                                  273

-------
D. H. Houck and Assoc.
(Daniel H. Houck)
McNamee, Porter and Seeley
(S. Jon Kang)
Green Bay MSD
disc (James J. Marx)
Manhattan College
(James A. Mueller)
University of Iowa
(Wayne L. Paulson)
Ewing Engineering Co.
(Lloyd Ewing and David T. Redmon)
Univ. of California at Los Angeles
(Michael K, Stenstrom)
Report/Hartford: In-situ performance evaluation of ceramic dome
diffusers using off-gas testing (24 months).

Report/European Survey:  Review/report  of plant  operating
experiences (principally in Scandinavia) using rigid porous plastic
tube and disc diffusers (12 months).

Report/Frankenmuth: Case  history of plant retrofit from coarse
bubble diffusers  to fine pore ceramic disc diffusers.  Limited in-
situ off-gas testing to evaluate  the effectiveness of gas cleaning
on ceramic disc diffuser performance (12-15 months).

ReponVGreen  Bay:  In-situ  performance  evaluation of ceramic
and  perforated membrane  tube diffuser  systems using off-gas
testing; included evaluation of cleaning methods (15-18 months).

Report/Ridgewood:  In-situ  performance  evaluation  and  case
history of  design, O&M, performance, and  cleaning of ceramic
dome diffusers used for retrofit  (18-24 months).

Report/Data Acquisition: Collection of data and coordination with
authors relative to clean water performance of fine pore diffusers
for interim summary report and  design manual.

Report/Monroe: In-situ testing  of ceramic disc diffusers of var-
ious  permeabilities using off-gas testing (24-30 months).

Report:  Summary of  in-situ performance  data on  fine  pore
diffusers  for inclusion  in interim  summary  report and design
manual.

Report:  Fine pore diffuser  fouling/cleaning data compilation  for
interim summary report and  final manual.

Report/Whittier Narrows:  In-situ comparative  off-gas  testing of
ceramic dome and disc diffusers. Also, evaluation of impact of
gas cleaning on system performance (30-36 months).

Report/Valencia:  In-situ performance evaluation of rigid porous
plastic disc diffusers using off-gas  testing (12-18 months).

Report/Terminal  Island:  In-situ  performance  evaluation  of
perforated membrane tube diffusers, nonrigid porous plastic tube
diffusers, and coarse  bubble diffusers using off-gas testing (12
months).
                                                  274

-------
                                             Appendix B
                 Selected Diffuser Characterization and Cleaning Methods
B.I  Diffuser Characterization

B.1.1 Foulant Analysis
An important aspect of the characterization of fouled
fine pore diffusers is the analysis of the nature of the
diffuser foulant(s). Foulant analysis  provides  insights
into  the  mechanism  of  fouling and can  aid  in the
selection of diffuser cleaning techniques.

Foulant analysis  consists of scraping  the  foulant  off
the surface of the diffuser medium and analyzing  for
the weight of dry solids per unit  area, volatile and
nonvolatile content, acid  solubility, and  elemental
composition  of  the  foulant  by  energy  dispersive
spectroscopy  or inductively  coupled  plasma. The
procedure for foulant analysis is:

1.   Specify a certain area on the surface of a diffuser.
2.   Scrape the  materials off  the  surface,  divide the
    materials,  and put them into two vials.
3.   Place each vial's contents in a tared  evaporation
    dish.
4.   Measure the wet weight.
5.   Dry at 105°C  (221 °F) for >1  hr  (to  constant
    weight).
6.   Cool, desiccate, and  weigh for  total solids.
7.   Put the dishes into furnace, firing them at 550°C
    (1,022°F)  for 20 minutes.
8.   Cool,  desiccate, and weigh the dishes  for fixed
    solids.
9-   Take one dish  content for metallic ion  analysis.
    Place in a vial.
10. Add approximately 10 ml_ of 14-percent HCI to the
    other dish and stir  gently until the formation of gas
    bubbles ceases.
11. Centrifuge  the  solution  at 20,000  rpm for  15
    minutes.  Decant the  upper portion, add deionized
    water into the tube, centrifuge again, and decant.
    Repeat once more for a total of three decants.
12. Repeat the steps 5, 6,  and 9 using the centrifuged
    solids.  Compare the results with those of the non-
    acidified foulant.

B.1.2 Bubble Release Vacuum (BRV)
The  BRV test provides  a means  of determining the
effective pore diameter at any point on the surface  of
a ceramic diffuser element relative to other point(s) on
its surface. This test procedure is  useful in assessing
the uniformity of pores on the surface of clean as well
as fouled diffusers.

BRV, as indicated by the name, is a measure of the
vacuum in inches of water gauge required to  emit'
bubbles from a localized  point on  the surface  of  a
thoroughly  wetted porous  diffuser  element. This  is
accomplished by applying a vacuum to a small  area
on  the  working surface of a wetted  diffuser and
measuring the differential pressure when bubbles are
released from the diffuser at the specified flux rate  at
the point in question.

A  large number of points  are sampled to  obtain  a
distribution of bubble release  pressures. Typical test
points  for a  ceramic dome  diffuser  are shown   in
Figure B-1.

The test apparatus consists of a probe, manometer,
vacuum source, and rotameter as shown in  Figure B-
2.  The  manometer is  filled  with either  water  or
mercury depending on foulanl buildup. Water   is
acceptable for  clean  and  lightly fouled diffusers.  A
switch to  mercury is required when BRV pressures
surpass the capacity of the water-filled manometer.

The probe used in this procedure has interchangeable
tips  for  testing  vertical and horizontal  surfaces.  Test
points 1, 6, 7,  and 12 require the  vertical surface  tip
while other points require the horizontal  surface tip.

The minimum airflow rate for  ceramic dome  and disc
diffusers is normally 0.24 L/s (0.5 scfm)/diffuser.  BRV
flow rates are kept below this  value so diffuser foulant
will not  be pulled  off the stone. This translates to  a
flow rate of 0.9 mLVs (0.0019  scfm) for the  1.13-cm2
(0.175-sq in) probe. A method  for calibrating  flow rate
is  illustrated in  Figure B-3.  With an in-line rotameter,
flow calibration is done just once  to  check the
rotameter calibration curve.

The recommended practice for BRV testing is listed
below:

1.   If the diffuser is new, immerse it in  tap water until
    wetted. Remove the diffuser from water  just prior
    to the test,  and let it drain  by gravity for  not  more
    than 30 minutes. Keep the diffuser  in a horizontal
    plain while draining. Do not soak fouled diffusers.
                                                  275

-------
Flguro 8-1.  BRV test points for ceramic dome.

                           7
                                Washer
  i  —
2 3
1 f __ 	

^
/ 4 5
_T-, 	 	 I 1

t
i
i
f 1 1
2 3

1 1 ^
4 5

^

                                             — 6
2.  Set BRV flow rate.

3.  Apply the probe  to the  BRV test location. The
    water surface will rise in  the probe while bubbles
    are released at the diffuser surface.  If the water
    level becomes loo high, discard excess water by
    a quick lateral and  upward movement  of the
    probe.  If  the water  level is too  low,  apply
    additional water onto the diffuser  adjacent to the
    probe. This is especially  useful when  testing
    fouled diffusers.

A.  Equilibrium  has been  reached when  the  rise  of
    water in the probe equals the rate of rise in the
    manometer (inches w.g.).  If the  time  to  reach
    equilibrium  is  excessive, it  may  be  reduced by
    operating the bypass valve momentarily. The flux
    rate increases  dramatically when the bypass valve
    is open. The large suction force will pull foulant off
    a dirty dilluser. Because the loss of  foulant may
    alfect test  results, the  bypass valve should be
    used Judiciously.

5.  Al equilibrium, record the manometer reading and
    height  of water  in  the probe.  BRV  equals the
    manometer reading  (corrected to  inches  water
    gauge if using mercury) less the height of water in
    the probe.

6.  Repeat steps 3 through 5 for all test locations.

B.1.3 Dynamic Wet Pressure (DWP}
DWP is the pressure differential across the diffusion
element  alone when  operating  in  a submerged
condition and is expressed in inches of water gauge.

In the DWP test,  most of the  pressure differential is
due to the force or pressure required to form bubbles
against the force of surface tension and only a small
fraction  of  the  total pressure gradient  is required  to
overcome frictional resistance.

In-situ DWP is measured  in the  aeration basin as
indicated  in Figure B-4. These measurements are
normally  made  two or three  times  per  week,
Comparison of the in-situ  DWP to the clean  water
DWP, as measured in the laboratory, will indicate the
degree  of  fouling.  In  making  this  comparison,
correction should be made for  possible differences in
airflow.

The  DWP and BRV  tests both measure  bubble
release  pressure.  DWP  measures  it for  the  whole
diffuser, while BRV provides a distribution of pressure.
For a new stone,  average DWP:BRV is close to 1.0.
As a  stone fouls,  the average  BRV for the 12 points
tested on the top surface becomes greater than DWP
and average DWP:BRV becomes less than  1.0.

The  equipment required  for measuring DWP  in the
laboratory includes  an air  source, rotameter,  in-line
mercury  manometer,  thermometer, diffuser plenum
with standard  orifice, water-filled manometer,  and
aquarium. The test setup looks very much like  Figure
B-4 without manometer A and the bubbler.  The water-
filled  manometer  (manometer  B  in  Figure B-4)  is
tapped into the plenum  at one  end  and open  to
atmosphere at the  other end.  The  water in the
aquarium is high enough to cover the diffuser.

Laboratory Measurement of DWP
1.  The aquarium should be filled with sufficient tap
    water  to cover the diffuser by several inches. If
    this  is done the day before testing, the water will
    warm to room temperature.

2.  New  diffusers should be wetted the same  as for
    the BRV test. Do not soak fouled diffusers.

3.  Place the diffuser securely in the plenum,

4.  Hold  the plenum over the aquarium and turn the
    air on. This allows water entrained  in the diffuser
    to drain into the tank and not on the floor.  If the
    diffuser is  fouled, do not  exceed  the operating
    airflow rate.
                                                  276

-------
    Figure 8-2.  BRV apparatus.
                                               By-Pass Valve
                      rz
                          BRV Pmbs
           See Detail   / |^ }  Probe Tip
                           OO
                                                                                  Vacuum
                                                                                  Source
                      Rotameter
                                                    Manometer
                            Diffuser
                                   Graduated
                                   Glass Tube
                                      Rubber
                                       Seal
                                                    -°-o
                                               C
           1
                                                   DETAIL
5.   Place the plenum in the aquarium. Adjust airflow
    to  the minimum suggested  rate  {0.24 Us [0.5
    sefm]/diffuser). Visually inspect the flow profile. If
    the diffuser is  not  mounted  correctly, coarse
    bubbling  will be evident. If this is the case, take
    the plenum, out of the aquarium  and reseat the
    diffuser.

6.   Adjust airflow  to the maximum allowable rate  for
    the test being performed, and let the test system
    equilibrate  for several  minutes. This allows time
    for excess water to be driven out  of stone.  When
    testing  fouled  diffusers,  do  not  exceed  the
    operating airflow rate.

7.   Perform a DWP profile. This is done by checking
    DWP at three  or more  air airflows. Typical airflows
    are 10, 20, 30,  and 40 L/s/m2  (2, 4, 6, and 8
    scfm/sq ft). A bucket catch may be performed to
    check  airflow  rate.   In  line  pressure  and
    temperature readings are taken to translate airflow
    rates to standard conditions.

8.  After the last DWP reading, turn airflow to almost
    zero.  Measure the static head over the diffuser.
    The  static head  is  subtracted  from  DWP
    manometer readings to  give true DWP readings.

9.  Correct  airflow  data  to standard  conditions.
    Regress  DWP (y)  on airflow (x). The correlation
    coefficient should be close to  1.0.

B.I.4 Airflow Profile Test
The airflow profile test  uses quantitative techniques to
evaluate  the uniformity  of air  release  across  the
surface of ceramic diffusers, while  operating, rather
                                                   277

-------
  Figure B-3.   BRV flow calibration.
       InvortotJ GratkwlucJ Cylinder
                             Probe Tip
                               To Vacuum Source
        BRV Probe
                                   Measure Risa in
                                  Water Column (t)
                                     with Time
                Soaker
than appraising uniformity  by visual means. This  is
accomplished by  testing the diffuser element  at  an
airflow rate that is approximately equal to 10 L/s/m2  (2
scfm/sq  ft), or at  the recommended design rate with
5-20 cm (2-8 in) of water over it.

The rate  of  air  release  from  selected  areas   is
measured by displacing  water  from  an inverted
container and  recording the rate of displacement  of
water with a  stopwatch.  By combining  the container
area and  the rate of  air discharge,  a flux  rate,
expressed as Us/m2 (scfm/sq ft)  or other convenient
units, can be calculated. By comparing  the flux  rates
of the  selected  area readings with one another,  a
quantitative measure or graphical representation of the
profile can be generated.

flux rate is measured by displacement by the rising
gas stream of water from a vessel inverted over the
area of diffuser to  be characterized. The vessel  must
first be filled with  water, covered, and  deftly inverted
so that the mouth of the vessel is just submerged.
Captured gas volume is  measured over a time of a
few  seconds taking care to record  the  captured
volume at atmospheric  pressure,  i.e.,  equal  water
surface levels inside and outside  the inverted vessel.
The flow rate is determined by dividing the  captured
volume by the time interval. The flux rate is defined as
flow rate divided by the area of the capture vessel. A
flow  profile  for  a  ¥ypical  diffuser  requires  flow
measurements on each of three concentric circles as
shown in Figure B-5.  Flow rates for the annular areas
are determined by difference.

These measurements are  made  by  using  three
vessels,  each with a different surface capture area.
The large,  13.5-L  (3.6-gal) vessel captures the entire
flow.  The 2-L (0.53-gal) vessel captures  all but the
periphery  flow,   whereas  the  1,000-mL (0.26-gal)
graduated  cylinder  captures  the flow  around  the
washer. By subtraction, flux rates are obtained for the
outer,  middle  and inner areas of  the  diffuser. These
flux rates are  then compared  to the average flux rate
of the diffuser. Example flux calculations are  shown in
Table B-1.

The combination of DWP, BRV, and flow profile tests
applied to new diffusers and at various stages in their
operating history  provides a  useful diagnostic  tool in
evaluating the rate, nature, and effect of fouling, either
organic or inorganic, on fine pore diffusion elements. It
is also effective,  if judiciously applied,  in appraising
the effectiveness of various cleaning procedures.

B.1.5 Specific Permeability
The manufacturers of  ceramic diffusers have used
and  are  familiar  with the permeability  test.  It has
served as  a  quality control  procedure to assure that
units sent to a job site are similar with respect to their
average  frictional  resistance  to  flow, when  dry,  to
within  some specified limits. This  was especially
important in many older plants where several ceramic
plates were  installed  in  a  single plenum  without
individual  airflow balancing  to achieve improved
uniformity  of airflow  among  the  units  when  in
operation.

The test generally consists of sealing the ceramic unit
in a test fixture substantially as it is sealed in an actual
aeration basin and then passing sufficient air through
the dry element to produce a  pressure differential of
5.1 cm (2.0 in) w.g.  The  permeability is reported as
the airflow  rate required to produce this differential. In
English units, the airflow rate  is in  scfm  (standard
cubic feet per minute) where 1 standard  cu ft of air is
considered to occupy 1  cu ft of volume at .1 aitm, 70F,
and  36 percent relative humidity. Historically, the test
was carried out on ceramic plates 30.5 cm x 30.5 cm
x 2.5 cm (12 in x  12 in x 1 in) thick.

As an  example, if  a permeability test is conducted on
such a ceramic plate, it might take 11.8 L/s (25 scfm)
to produce  a 50.8-mm (2-in) w.g. differential pressure.
In this case, the permeability rating would  be 25. A
plate of identical material,  but half as thick, would be
expected to have  a permeability of 50 instead of 25,
since the flow paths  through the ceramic would be
                                                   278

-------
         Figure B-4.  Measurement of air line and diffuser pressure..
                                  Pressure  Temperature
              Blower

               Q_
                        Flow Meier
               o
                                                              --o-o-o-
                                                        Valve
                             PLAN VIEW SCHEMATIC
           -o-o-o
                                                                        Bubbler Pipe
                                                                      Airflow Control Orifice
                               Air Header
                                               DETAIL
about half  as long and  offer correspondingly  less
frictional resistance. Had the element been 2.5 cm (1
in) thick with an area of 465 cm2 (72 sq in) instead of
930  cm2  (144 sq  in),  the  permeability  would be
approximately 12.5,  since there would have been only
about half the area of the first case.

Even though  these  elements were made in  identical
ways with identical materials, the permeabilities of the
three vary from 12.5 to 50. Thus, using permeability to
compare  ceramic  elements of  different  shapes,
thicknesses,  and materials  of  construction is  not
meaningful and has constituted a source of confusion
in the engineering community.

In an effort to employ permeability test results as a
measure of resistance characteristics of the material,
the  term specific permeability  has  been adopted.
Specific permeability is the equivalent amount of air at
standard conditions  to produce  5.1  cm  (2  in)
differential pressure across the  dry element if  the
element were 930 cm2 (1 sq ft) in  area (30.5 cm x
30.5 cm [12 in x  12 in]) and 2.5 cm (1 in) thick.
                                                 279

-------
 Figure B-5.  Airflow profile apparatus,

                    13.5-L Buckel-
         1,000-mL Graduated Cylinder
An approximate expression to convert the permeability
of any porous structure lo specific permeability is as
follows;
            SP =
                               (B-1)
whore,

    SP
    Po
    Ao
specific permeability, scfm
permeability of the element itself, scfm
area  of  element  when  made  to
hypothetieally conform to a flat surface,
sq ft
mean  weighted  thickness  of  the
element, in
B.2 DIffuser Performance
B.2.1 Steady-State Clean Water OTE Test
This test is usually conducted in a covered fiberglass
column 60-90 cm (2-3 ft) in diameter with a sidewater
depth approximately equal to that of the aeration basin
of interest. The test is conducted in a manner similar
lo the  nonsteady-state clean  water test method,
except  that dissolved sodium sulfite  is  continuously
pumped into  the column through a distribution system
to maintain  a constant  DO concentration.  Once
steady-state  conditions are  achieved,  gas-phase
oxygen transfer efficiency (OTE) is determined using
the oH«gas analysis method. The DO concentration is
monitored  at  two  depths  corresponding  to
approximately 1/4 and 3/4 of the sidewater depth.

The method is normally used to compare clean water
OTEs  of  two fine  pore diffusers of interest (e.g.,  a
new  diffusar vs.  a  fouled one).  Each has an
independent  air  source.  One  of the  diffusers is
operated at a specified airflow rate, generally 0.5-1.4
Us (1-3 scfm), and the other is idled at a very low
airllow rate (<5 percent of the other), the object being
to keep a slight positive pressure  differential  across
the ceramic element.  The airflow rates applied to the
diffusers are measured with variable-area rotameters.
The test apparatus is illustrated in Figure B-6.

Once steady-state conditions  are  achieved, the OTE
is measured by off-gas analysis. After appropriate data
are obtained on  one diffuser,  air  supply  is rapidly
switched, causing the other diffuser to operate at  the
specified airflow rate and the air to the first diffuser to
be reduced to maintain only a slight positive pressure,
as above.

Repeating the cycle several times in this manner,  the
relative oxygen transfer  performance of the diffusers
can be obtained. A comparison of clean water SOTEs
can then be made for the two diffusers of interest.

This test procedure can also be used to:

•   measure the effect of diffuser uniformity  on OTE,
•   determine the rate and consequence of  fouling on
    fine pore diffusers, and
*   evaluate fine pore diffuser cleaning methods.

B.3 Field Diffuser Cleaning Methods
B.3.1 Low-Pressure Hosing
This method is frequently used  in conjunction with
other methods and consists of hosing at a distance of
0.6-7.3 m  (2-24  ft)  using a  low-pressure  nozzle at
207-484 kPa (30-70 psi).  The  hosing  period should
continue until the readily removable foulant has been
washed away. This time is generally on the order of
10 seconds but can vary from 5 seconds to 1 minute,
depending on the resistance of the foulant, pressure,
and distance. The air should be on during  the hosing
operation with each  diffuser  operating at roughly  0.5
Us (1 scfm).

B.3.2 High-Pressure Hosing
This method consists of hosing at a distance of  about
60 cm (2 ft) using a high-pressure nozzle at 552-689
kPa  (80-100 psi). The hosing period should continue
until the readily removable foulant has been washed
away.  This  time  is generally on  the order  of  15
seconds  but can vary  from about 5  seconds to 1
minute depending on the resistance of the foulant and
pressure.  Each diffuser should  be operating  at  ah
airflow rate of approximately 0.5 L/s (1  scfm)  during
the hosing operation.

B.3.3 Milwaukee  Method (Acid Application plus
Hosing)
This method has  been used  at  the  Milwaukee,  Wl
wastewater treatment plants  for many years. A high-
pressure * water jet is applied to the diffuser surface
followed by acid spraying and rehosing. The rationale
is to first hose off the easily removable foulant so that
the applied acid can solubilize the inorganic precipitate
inside the pores of the  diffuser. A second hosing is
then performed to remove the solubilized foulant and
residual acid. The materials  needed for this method
                                                  280

-------
                 Table B-1.   Example Dif 1 user Flux Calculations
                 Apparent Flux:      AF = q/AD
                 Local Flux:         LF = q|/A|
                 Effective Flux:       NF = {E[LF(q,)j>p(qi}J
                 Effective Flux Ratio:  EFR = NF/{[£(LF))/i}

                 where,
                     Q
                     AD
                     A,-
                     i
airflow rate per diffuser, sefm
total projected media surface area of installed diffusers, sq ft
local airflow rate, scfm
local area sampled, sq ft
no. of observations
                 Example:

                 Ceramic Disc Diffuser;
                       Diameter = 8.7 in
                       Effective Area (AD) = 0.413 sq fl/diffuser
                       Airflow Rate (q) = 0.65 scfm/diffuser
                       AF = 0.65 scfm/0.413 sq ft = 1.57 sofm/sq ft

                 Local Flux:
                       Test 10 equal-area locations.
                       Use 100-mL graduate cylinder with A( = 0.0066 sq ft for sampling.
               Location
      qj, scfm
LF, scfm/sq ft
LF (q,)
1
2
3
4
5
6
7
8
• 9 •
10
Sum
Average Flux:
Effective Flux:
Effective Flux Ratio:
0.0129
0,0154
0.0147
0.0081
0.0025
0.0138
0.0115
0.0060
0.0098
0.0105
0.1052
MF = 15.93/10 = 1 .59 scfm/sq ft
NF - 0.190/0.105 = 1.81 scfm/sq ft
EFR = 1.81/(15.93/10)" = 1.14
1.96
2.33
1 2.22
1.22
0.38
2.09
1.74
0.91
1.49
1.59
15.93

0.0253
0.0359
0.0326
0.0099
0.0010
0.0288
0.0200
0.0055
0.0146
0.0167
0.1903

are: high- or  low-pressure water hosing  equipment,
acid  spray  applicator  (Hudson Acid Sprayer or
equivalent), and 50  percent by volume of 18 Baume
inhibited  muriatic  acid.  This is  equivalent  to  a  14-
percent HCI solution.

The procedure to be followed is:

1.   Clean the  diffuser by high- or low-pressure hosing
    with  the   air  on  at approximately  0.5  Us (1
    cfm)/diffuser.

2.   Apply approximately 50 ml_ (1.7  02) of 14-percent
    HCi to the surface  of the diffuser using  the spray
    applicator. No air should be applied to  the diffuser
    during the acid application period.
                             3.   Let the acid remain on the diffuser for 30 minutes.
                                 Turn the air on for 5 minutes.

                             4.   Hose the diffuser again to remove all the residual
                                 acid.

                             B.3.4 Steam Cleaning
                             This  method has  been   used  at  the  Madison,  Wl
                             treatment plant for several years. The principle of the
                             method is to use the scrubbing  and heating  power of
                             a steam jet  to  remove the material attached to the
                             diffuser's surface. The  high temperature  of the steam
                             may lessen the ability  of  the foulant to attach to the
                             diffuser  surface.  Equipment  required  for  this  test
                             includes a  steam  generator and  nozzle.  The
                             procedure to be followed is:
                                                      281

-------
 Figure B-6.   Steady-state clean water OTE test apparatus
             Pilot Tank
       Sulfito
        Tank
                             Diffusers
   Vacuum
   Source
                                                                                      Air Compressor
1.  Turn on  the  steam generator,  and let  it  run  for
    several  minutes to reach  constant temperature
    (200C [392F]).

2.  AppJy the steam jet to the diffuser surface from a
    distance  of approximately  60  cm (2 ft)  at a
    minimum pressure of 1,135 kPa (150 psi)  until all
    the foulant has been visibly removed. The  diffuser
    air should be on  during this operation.

3.  Let the diffuser cool to ambient temperature.
B.3.5 Firing (Kilning)
This method is widely used in England. The diffuser
stones are removed, placed in a kiln, fired to remove
foulant material, and gradually  cooled. A typical  British
furnace is capable of  firing  650 domes  per  24-hr
cycle. The procedure to be followed is:
1.  Load the  diffuser  stones  into the  furnace, and
    heat the furnace to 950°C (1,742°F) over a period
    of 10 hours.

2.  Hold the  temperature  at  950-1,000°C  (1,742-
    1,832°F) for 4 hours.

3.  Cool down over a period of 10 hours.

8.3.6 Gas Cleaning
Gas cleaning refers to  a method whereby a fine pore
diffusion system  is  cleaned  by  injecting a  small
percentage of  HCI  gas into the air  supply line leading
to the diffusers. The HCI gas solubilizes some foulant
deposits and aids  in cleaning the  diffusers. The gas
cleaning system evaluated  in the EPA/ASCE  Fine
Pore Aeration  Project is a  proprietary  system
marketed by Sanitaire - Water Pollution  Control  Corp.
under   U.S.  Patent  No.  4,382,867.  The   test
                                                  282

-------
installations  were  operated  under license  from
Sanitaire,  and  the  gas  cleaning operations  were
carried out following the recommendations of Sanitaire
personnel.

Gas cleaning test results to date have indicated that a
minimum  gas concentration  is required  for effective
periodic cleaning.  Above the minimum concentration,
the  amount  of  gas  required  to  clean  deposits  is
substantially constant. However, the  time required to
clean is less at higher concentrations. Mole ratios of
HCI  gas in air between  0.0000818 and  0.0309 have
been used successfully in gas cleaning.  Based on
experience, it appears that about 113 g (0.25  Ib) HCI
gas is required per diffuser.

In operation, the need to apply  gas cleaning treatment
is judged  by monitoring the DWP drop across  each of
four diffusers installed on a removable  header. The
onset of  fouling  is indicated  by an  increase in the
DWP loss across the diffuser at a  constant airflow
rate. The  allowable pressure increase before initiating
cleaning should be specified on a case-by-case basis.
If  the DWP is allowed to rise to a  level  where the
desired combined system  pressure  during cleaning
exceeds the  blower capability,  one available option  is
partial dewatering of the basin being cleaned.
During the cleaning cycle, it is  important, to achieve
uniform gas distribution both  between  diffusers  and
throughout the area of the  individual diffuser elements.
Higher  airflow rates  promote  more  uniform  gas
distribution throughout the media  pores. Because of
this, it is  recommended that cleaning be done at a
higher airflow rate  than used during normal operation,
e,.g.  about 30-40  L/s/m2  (6-8  scfm/sq ft)  diffuser
surface,  or  1.2-1.4  L/s  (2.5-3  scfm)/diffuser.  The
increased airflow  rate also increases  the  pressure
differential  across   the  diffuser  element,  thus
distributing cleaning gas to partially clogged pores.

The increased airflow rate  needs  to applied only to the
grid  being cleaned.  This  can  be  accomplished  by
operating  an extra blower  for a  short period of time,
throttling air to the rest of  the basins, or dropping the
water level in the basin  being cleaned. If the water
level  is  reduced  a  few  feet,  the normal system
pressure would be adequate to provide  the increased
airflow rate to the basin being cleaned. This may be
the most economical alternative  as no  additional
power is required.
                                                   283

-------

-------
                                     Appendix C
           Selected Physical and Chemical Tables and Graphs
Table C-1.   DO Saturation Values        •

Solubility of Oxygen (mg/L) in Water Exposed to Water-Saturated Air at Atmospheric Pressure = 101.3
kP, (14.7 psia)                                          '
Temp., "C
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
10.0
11.0
12.0
13.0
14.0
15.0
16.0
17.0
18.0
19.0
20.0

0
14.62
14.32
13.83
13.46
13.11
12.77
12.45
12.14
11.84
11.56
11.29
11.03
10.78
10.54
10.31
10.08
9.87
9.67
9.47
9.28
9.09
Chlonnity"
5.0
13.73
13.36
13,00
12.66
12.34
12.02
11.73
11.44
11.17
10.91
10.66
10.42
10.18
9.96
9.75
9.54
9.34
9.15
8.97
8.79
8.62

10.0
12.89
J2.55
12.22
11,91
11.61
11.32
11.05
10,78
1Q.53
10.29
10.06
9.84
9.62
9.41
9.22
9.03
8.84
8.67
8.50
8.33
8.17
Temp., °C .
21.0
22.0
23.0
24.0
25.0
26.0
2/.0
28.0
29.0
30.0
31.0
32.0
33.0
34-0
35.0
36.0
37.0
38.0
39.0
40.0


0
8.91
8.74
8.58
8.42
8.26
8.11
7.97
7.83
7.6P
7.56
7.43
7.31
7.18
7.07
6.95
6.84
6.73
6.62
6.52
, 6.41

Chlorinily*
5.0
8.46
8.30
8.14
7.99
7.85
7.71
7.58
7.44
7.32
7.19
7.07
6.96
6.84
6.73
6.62
6.52
6.42
6,32
6.22
6.12


10.0
8.02
7.87
7.73
7.59
7.46
7.33
7.20
7.08
6.96
6.85
6.73
6.62
6.52
6.42
6.31
6.22
6.12
6.03
5.93
5.84

* Chlorinity = Salinity/1.80655 (See p. 415: Standard Methods for the Examination of Water and Wastewater,
  16th Edition, for definition of salinity.)
Adapted from pp. 413-415: Standard Methods for the Examination of Water arid Wastewater, 16th Edition.
American Public Health Association, Washington, DC, 1985.
                                          285

-------
Table C-2.    Hydraulic Headlosses for Appurtenances
          Apfltirtonanco - Aiphabotically
Headless as
 Multiple of
  (v2/2g)
 1,   Butterfly Valves
      Fully open                                     0.3
      Angte closed, 0 » 10°                          0.46
                  0 = 20°                          1.38
                  0 - 30°                          3.6
                  0 = 40"                           10
                  0 = 50"                          31
                  0 - 60"                          94
 2-,   Check (Rellux) Valves
      BoB Typo (lulfy open)                          2.5-3.5
      Horizontal Lilt Type                             8-12
      Swing Chock                                 0.6-2,3
      Swing Clioek (fully opon)                        2.5
 3,   Contraction - Sudden
      4il  (in lornis ol velocities of small end)           0.42
      2s 1                                            0.33 -'
      4«3                                            0.19
      also sco Reducers
 «l,   Diaphragm Valve                  •            2.3
      FuMy opoti                                     2.6
      3/4 opon                                      4.3
      iffiopon                                      21.0
      i« opon
 5.   Elbow- 90"                                0.21-0.30
              - Regular                           0.18-0.20
              - Lony Radius
      Intersection of two eyiinders (welded pipe -      i .25-1.8
      not rounded)
      Scrowod - Short Radius                        0.9
      Screwed ~ Medium Radius                      0.75
      Scrowod - Long Radius                         0.60
 6,   Elbow - «•
      Flanged - Roaular                           0.20-0.30
      Ffangod - Long Radius                       0.18-0.20
      Scrowod -  Regular                         0.30-0.42
 7,   Enlargement - Sudden
      1:4  (in terms of velocities of small end)           0.92
      1:2                                            0.56
      3:4                                            0.19
      also see Increassrs
 0.   Entranco Losses
      Bell mouthed                                   0.04
      Pipe flush wilh  tank                             0.5
      Pipe projecting into lank (Borda Entrance)       0.83-1.0
      Sightly roundod                                0.23
      Strainer and foot valve                          2.50
Appurtenance  - Alphabetically
                                                                                                                  Headless as
                                                                                                                   Multiple of
                   9.    Gate Valves
                        Open           "           ,.                  .0.19
                        1/4 closed                                     1115
                        1/2 closed                                      5.6
                        3/4 closed                                     24.0
                        also see Sluice Gates
                   10.  Increasers
                        0.25 (v,2/2g - v22/2g)
                        where v( =  velocity at small end

                   11.  Miter Bends
                        Deflection angle, 0
                        5*                                         0.016-0,024
                        10°                                       0.034-0,044
                        15°                                       0,042-0.062
                        22.5°         '                -          ;   0.066-0.154
                        30°                                       0.130-0.165
                        45°                         :       '       0.236-0.320
                        60°                                       0.471-0.684
                        90°                                       1.129-1.265
                   12.  Obstructions in Pipes (in terms of pipe
                        velocities) Pipe  to Obstruction  Area Ratio
                        1.1                                           0^21
                        1.4               '                       ',   1.15
                        1.6                          ..         ,-•      2.40
                        2.0                                           5.55
                        3.0       '  ,  ,   -                       •••   . . 15.0 .'-
                        4.0                 •                '  '       ''27.3
                        5.0                                           42.0
                        6.0               '                            57.0
                        7.0             •                              72.5
                        10.0                                          121.0

                   13.  Orifice  Maters (in terms of velocities of
                        pipe) Orifice to Pipe Diameter  Ratio
                        0.25 (1:4)   .                                   4.8
                        0.33(1:3)    "•••••                        ag
                        0.50(1:2)                                      1.0
                        0.67 (2:3)                              .        0.4
                        0.75 (3:4)                                     0.24

                   14.,  Outlet Losses
                        Bell mouthed outlet                          O.l(v,2/2g -
                                                                    v22/2g)   ,
                        Sharp cornered outlet                        (v,2/2g -
                                                                    v22/2g)
                        Pipe into still water or air (free discharge)           1.0
                                                              286

-------
                                                Headless as
                                                 Multiple of
                                                  (y2/2g)
 Table C-2.     (continued)


•	Appurtenance  - Alphabetically	
  15.  P/ug Globe or Stop Valve
       Fully open                                      4.0
       3/4 open                                       4.6
       1/2 open   •                                    6.4
       1/4 open                                      780.0

  76.  Reducers
       Ordinary (in terms of velocities of small           0.25
       end)
       Bell mouthed                                   0.10
       Standard                                      0.04
       Bushing or coupling                          0.05-2.0
Appurtenance  - Alphabetically
Headless as
 Multiple of
  (v2/2g)
                                                                79.  Tees
                                                                    Standard - bifurcating
                                                                    Standard - 90* turn
                                                                    Standard - run of lee
                                                                    Reducing - run of tee
                                                                         2:1  (based on velocities of smaller
                                                                             end)
                                                                         4:1  (based on velocities of smaller,
                                                                             end)
                                                                    Venturi Meters
                                          1.5-1.8
                                           1.80
                                           0.60

                                           0.90

                                           0.75
                                                                20.
                                                                    The headloss occurs mostly in and
                                                                    downstream of throat, but losses shown are
                                                                    given in terms of velocities at inlet ends
                                                                    to assist in design.
                                                                    Long  Tube Type - Throal-to-ittlel diameter
17.






18.




Return Bend (2 nos. 90°)

Flanged - Regular
Flanged - Long Radius
Screwed


Sluice Gaies
Contraction in conduit
Same as conduit width without top
submergence
Submerged port in 12-in wall


0.38
0.25
2.2



0.5
0.2

0.8
ItiMW
0.33 (1:3)
0.50 (1:2)
0.67 (2:3)

0.75 (3:4)
Short Tube Type - Throal-to-inlet diameter
ratio
0.33 (1:3)
0.50(1:2)
0.67 (2:3)
0.75 (3:4}

1.0-1.2
0.44-0.52
0.25-0.30

0.20-0.23


2.43
0.72
0.32
0.24
Source, pp. 702-704: Amirtharajah, A. Design of Granular-Media Filter Units. In: Water Treatment Plant Design for the Practicing
   Engineer, Edited by R.L. Sanks, Ann Arbor Science, Ann Arbor, Ml, 1978.,
                              Table C-3.   Typical Air Velocities in Air Delivery Systems
                                     Pipe Diameter (in)                Velocity (fpm)
                                            1-3
                                           4-10
                                           12-24
                                           30-60
                                                                        1,200-1,800
                                                                        1,800-3,000
                                                                        2,700-4,000
                                                                        3,800-6,500
                              Source, p. 287: Stephenson, R. L. and H. E. Nixon. Centrifugal
                                 Compressor Engineering. Hoffman Air and Filtration Division,
                                 New York, NY, 1973.
                                                            287

-------
Table C-4.   Properties of Standard Atmosphere
Altitude, ft
0
1,000
2,000
3,000
4.000
5,000
10,000
1S.OOO
Source, p, 36: Daily,
Table C-S.
Tomp., "F
32
40
50
60
70
80
SO
100
110
120
130
140
ISO
160
170
180
190
200
212
Tomp., °F
59
55.44
51.87
48.31
44.74
41,18
23,36
554
J.W. and D.R.F.
Pressure, psia








Harleman.
2,116.2
2,040.9
1,967.7
1,896.7
1 ,827.7
1,760.8
1 ,455.4
1,194.3
Fluid Dynamics,
Kinematic
Density, Viscosity x 104,
slugsfcu ft sq ft/sec
0.00238
0.00231
0.00224
0.00218
0.00211
0.00205
0.00176
0.00150








Addison- Wesley, Reading,
1.56
1.60
1.64
1.68
1.72
1.77
2,00
2.28
MA, 1966.
Velocity of Sound,
fps
1,117
1,113 •
1,109
1,105
1,104
1,098
1,078
1,058

Physical Properties of Water
Specific
weight (yw),
Ib/ou ft
62.42
62.43
62.41
62.37
62.30
62.22
62.11
62.00
61.86
61.71
61.55
61.38
61.20
61.00
60,80
60,58
60.36
60.12
59.83
Density

-------
Figure C-1.   Standard atmospheric pressure for altitudes of sea level to 10,000 ft
        Altitude, ft
       10,000 ._
        8,000  -
        6,000
        4,000
        2,000
              10
                                  11
                                                      12                  13
                                                Standard Atmospheric Pressure, psia
                                                                                             14
                                                                                                                 15
Source, p. 36: Daily, J.W. and D.R.F. Harleman. Fluid Dynamics. Addison-Wesley, Reading, MA, 1966.
                                                           289

-------
Figure C-2.   Moody diagram for friction factor in pipes.
      0,1
     009
                                                         Complete Turbulence,
                                                            Rough Pipes
     0.01

    0.009

    0,008
   SOIHCO, p. 57: Motcall & Eddy, inc. Waslcwaler Engineering:
                CoKecton/TroatmenVDisposal, 1 si EcJilion.
                McGraw-Hill Book Co., New York, NY, 1972.
0.000,01
                                                                                                              10°
                                                            Reynolds Number, Ra =
                                                          290

-------
Figure C-3.   Equivalent air pressure (EAR) curves for use in blower selection.
                                                                                    i	1
                                                                               Inlet Temperature
                                                                                  Fahrenheit
Eiovalion Referenced lo
     Sea Lovel
       10,000 fl
        9,000
        8,000
        7;000
        6,000
        5,000
        4,000
      " 3,000
        2,000
      +• 1,000
        Sea Level 0
             -1,000
                Discharge Air Pressure (psig) Required to
                          Meet Site Conditions
                                                                                                 9    10   11   12   13   14   15
                                                                     Equivalent Air Pressure (psig) EAP to be Used with
                                                                       Standard Performance Curve to Select Blower
               1) Enter left-hand side of chart X-Axis with required discharge pressure at site conditions.
              2) Move up to line corresponding to elevation at site.
              3) Move right to line corresponding to inlet air temperature (note that both minimum and maximum
                 air temperature condiiions should be considered).
              4) Move down to the equivalent air pressure (EAP) to be used in selecting the blower.

              Source, p. 18: Siephenson, R.L. and H.E. Nixon, Centrifugal Compressor Engineering.
                            Hoffman Air and Filtration Division, New York, NY, 1973.
                                                              291

-------

-------
                                            Appendix D

                                 Economic Analysis Spreadsheet
Instructions for constructing the spreadsheet" are:

1.   Begin with a new 1-2-3 worksheet and enter the
    formulas listed in Table D-1.

2.   Execute the command scripts listed in Table D-2.

3.   Save the worksheet under a name of your choice
    and exit 1 -2-3.

Whenever the worksheet is loaded into 1-2-3, a Main
Menu with four choices  is presented to the user. The
menu choices are:

 Data       enters/edits aeration system data

 Calculate   calculates aeration system costs

 Graph     graphs results of cost calculation

 Quit       returns to 1 -2-3's READY mode

The Data option presents  a second menu  from which
a category of aeration system data is chosen  for data
entry/editing. The options on this menu are:

Name   names the plant or data set being analyzed

 Oxygen    input table of monthly  oxygen demands
            and ratios  of OTRfS to SOTRs for up  to
            three aeration zones

 Diffusers   input form for diffuser characteristics  in
            each aeration zone

 Blowers   input form for blower data

 Costs     input form for unit cost factors,  interest
            rate,  analysis  period,   and  diffuser
            cleaning frequency
Data is entered or changed  in  each table  or form
using the  normal 1-2-3  procedures. After a form or
table is completed, the  user can return to the Main
Menu by pressing  the ALT-M key combination or to
the Data Menu by pressing ALT-D.
The Calculate option of the Main Menu
total present worth costs of the aeration
on the current set of design  data. On
table of capital,  energy, maintenance,
total  present worth  costs is displayed
Menu reappears.
computes the
system based
completion,  a
cleaning, and
and the Main
The  Graph  option is used  to  display plots of the
results of the cost computation. Three different types
of graphs can be drawn (providing that the computer
is able to display graphics):
 Airflows     monthly total airflow and flow to each
             aeration zone
 Operating   monthly   energy and total (energy
 Costs       cleaning) costs
 Total Costs  bar chart of  present  worth  cost  by
             category
The last option on the Main Menu, Quit, returns the
user to the native 1-2-3 environment. At this point, the
worksheet can be saved under a new name for future
use by using the 1-2-3 File Save command. To restart
the economic analysis worksheet again, one only has
to press ALT-M.
                                                  293

-------
Table D-1  Cell Entries For Economic Analysis Spreadsheet
B2;  ' ECONOMIC ANALYSIS OF DIFFUSED AERATION
AS:  \=
B5:  \=
C5:  \^
D5:  \*
E5:  \*
F5:  \=
G5:  \=
B7:  'Operating Period, Months
B8:  'PRESENT WORTH COSTS, $1,000:
B9:  ' Capital
D9:  \.
E9:  I
810: ' Energy
010: \.
E10: V.
B11: ' Cleaning
D11: \.
E11: V
B12; * Maintenance
D12: V.
E12: V
B13: ' Total
D13: I
E13: V.
D30: 'OXYGEN REQUIREMENTS
A31: '-
B31: V
C31: v.
D31: V
E31: -
F31: l,-
G31: >•
H31: V
B32: ' Oxygen Demand, Ibid
F32: '  OTR(fie!d)/SOTR
A33: "Month
B33: "ZONE 1
C33: "ZONE 2
D33: "ZONES
F33: "ZONE 1
G33: "ZONE 2
H33: "ZONE 3
A34: V
B34: V
C34: V
D34: V  E34:  V
F34: V
G34: V
H34: V
A35: "JAN
A36: "FEB
A37: "MAR
A38: "APR
A39: "MAY
A40: "JUN
A41: "JUL
A42: "AUG
A43: "SEP
                                           294

-------
A44:
A45:
A46:
A47:
B47:
C47:
D47:
E47:
F47:
G47:
H47:
C48:
C49:
C50:
A51:
B51:
C51:
D51:
E51:
F51:
G51:
D52:
E52:
F52:
G52:
Pcq.
U.OO.
PCQ-
i Do.
A54:
A55:
A56:
A57:
A58:
A59:
A60:
A61:
A62:
A63:
A64:
A65:
C65:
D65:
E65:
F65:
G65:
C66:
C67:
C70:
A71:
B71:
C71:
D71:
E71:
A72:
E72:
A73:
B73:
C73:
D73:
E73:
A74:
"OCT
"NOV
"DEC
\-
V
V
V
\-
V
\-
\-
'Press  to return to Main Menu
'Press  to return to Data Menu
' DIFFUSER CHARACTERISTICS
V
V
\-
\-
V
\-
\-
I
"ZONE 1
"ZONE 2
"ZONE 3
i*
„
,,
'Minimum Airflow, cfm/diffuser
'Maximum Airflow, cfm/diffuser
'SOTE at Minimum Airflow, percent
'SOTE at Maximum Airflow, percent
'OTE loss rate, percent/mo
'Maximum percent OTE Loss
'Minimum (Clean) DWP, in
'Maximum (Fouled) DWP, in
'Orifice P-drop at 1 scfm, in
'Mixing Requirement, scfm
"Number of Diffusers
V B65: V
V
V
V-
\-
V
'Press  to return to Main Menu
'Press  to return to Data Menu
'ECONOMIC FACTORS
V
\-
V
V
V
" Factor
"Value
\-
V
V
V
V
' Discount Rate, percent
295

-------
A75:  'Dllfuser Installation Cost, $/unit
A76:  'Other Capital Costs, $
A77:  'Electricity Rale, $/kWh
A78:  "Dtffuser Cleaning Cost, $/unit
A79:  'Cleaning Interval, months
ABO:  'Routine Maintenance, $/yr
A81:  "Analysis Period, months
A82:  'Start-up Month (Jan  = 1, etc.)
A83:  V
883:  v
083:  '»-
D83:  V
E83:  V
A84:  *  Press  to  return to Main Menu
A85:  '  Press  to return to Data Menu
090:  'BLOWER DATA
A91:  V
B91:  V
091:  «-
D91:  V
E91:  V
A92:  ' item
E92:  "Value
A93:  V
B93:  V
093:  V
D93:  V
E93:  V
A94:  'Barometric Pressure, psia
A95:  "Pressure Head w/o Diffusers, psig
A96:  'Overall Blower Efficiency, percent A97: V
B97:  \-
C97:  V
D97:  V
E97:  V
A98:  "  Press  to  return to Main Menu
A99:  '  Press  to return to Data Menu
Pi:   'CALCULATIONS
Si:   "ZONE 1
T1:   "ZONE 2
U1:   "ZONE 3
P2:   \=
Q2:   v»
R2:   \=
S2:   \=
T2:   \-
U2:   \*
P3:   "Operating period
P4:   'Month of year
S4:   @1F(@MOD(S3 ••- E82-1,12)>0,@MOD(S3 + E82-1,12),12)
PS:   'Months Uncleaned
S5:   @IF(E79 < = 0»S3,@IF{@MOD(S3,E79) > 0,@MOD(S3,E79),E79))
P6:   'Oxygen Demand
S6:   @lNDEX(B35,.D46,0,S4-1)/24.87/@IF(E64>Q,E64,1)
T6:   @lNDEX(B35..D46,1,S4-1)/24.87/@IF(F64>0,F64,1)
U6:   @INDEX(B35..D46,2,S4-1 )/24,87/@lF(G64 > 0,664,1 )SOTE
P7:   'OTEf/SOTE
S7:   @1NDEX(F35..H46,0,S4-1)
T7:   @1NDEX(F35..H46,1 ,S4-1)
U7:   @1NDEX(F35..H46,2,S4-1)
P8:   'Fouling Factor
                                               296

-------
38:   1-@M!N((S5-0.5)*E58,E59)/100
T8:   1-@M]|M((S5-q.5)T58,F59)/100
U8:   1-@MIN((S5-Q.5)"G58>G59)/100
P9:   'Kote
SB:   @IF(E55 > E54,(E57-E56)/1QQ/(E55-E54),Q)
T9:   @IF(F55 > F54,(F57-F56)/1 00/(F55-F54),0)
U9:   @IF(G55>G54,(G57-G56)/100/(G55-G54),Q)
P10:  'OTEO
S10:  + E56/100-S9*E54
T10:  + F56/100-T9AF54
U10:  +G56/100-U9t354
P11:  'CFM1
S11:  + 310*810 + 4*39*36/37/38
      + U10*U10 + 4*U9*U6/U7/U8
P12:  'CFM2
312:  @IF(S1 1 > = 0,{-S1 0 + @SQRT(S1 1 ))/(2"S9),@ERR)
T12:  @IF(T1 1 > = 0,(-T1 0 + @SQRT(T1 1 ))/(2T9),@ERR)
U12:  @IF(U11> =0!(-U10 + @SQRT(U11))/(20J9),@ERR)
P13:  'CFM3
S13:  @IF(S12>0,@MAX(E63/E64,E54,S12),S12)
T13:  @IF(T12>0,@MAX(F63/F64,F54,T12),T12)
U13:  @IF(U12>0,@MAX(G63/G64,G54,U12),U12)
P14:  'CFM4
S14:  @IF(S13>E55,@ERR,S13)
T14;  @tF(T13>F55,@ERR,T13)
U14;  @IF(U13>G55,@ERR,U13)
P15:  'CFM5
S15:  @IF(S6>0, + S14*E64,0)
T1 5:  @IF(T6 > 0, + T1 4*F64,0)
U15:  @IF(U6>0, + U14*G64,0)
P16:  'CFMtot
S16:  @SUM(S15..U15)
P17:  'DWP, in
S1 7:  + E60 + (E61 -E60)*@IF(E59 > 0,(1 -S8)/E59*1 00,0)
T17:  +F60 + (F61-F6Q)*@IF(F59>Q>(1-T8)/F59*100,0)
U17:  +G60 + (G61-G60)*@IF(G59>0,(1-U8)/G59i100,0)
P18:  'TOP, in
S18;  @IF(S6>0, + S1 7 + E62*S1 4*314,0)
T18:  @IF(T6>0,+T17 + F62T14T14,0)
U18:  @IF(U6>0, + (J17 + G62*U14*U14,0)
P19:  'PDtot, psig
319:  @MAX(S18,T1 8,111 8)*O.Q36 + E95
P20:  'Energy  Used
S20:  8.39*E94*S16*(((E94 + S19)/E94)~0.283-1)*100/E96
P21:  'Energy  Cost
S21:' +E77*S20
P22:  'Cleaning Cost
322:  @IF(E79>0,@[F(@MOD(S3>E79) = 0,E78*@SUM(E64..G64),0),0)
P23:  'Maintenance Cost
S23:  +E80/12
P24:  Total Cost
324:  + S21 + S22 + S23
AA1:  '\0
AB1:  '{BRANCH \M)
AA4:  '\M
AB4:  '{HOME}
AB5:  '{MENUBRANCH Menul)
AA8:  'Menul
AB8:  'Data
ACS:  'Calculate
                                               297

-------
AD8:  'Graph
AE8:  'Quit
AB9:  'Enter/edit aeration system data
AC9:  'Calculate aeration costs
AD9:  'Graph results of cost calculation
AE9:  'Return to 123's READY mode
AB10: '{BRANCH \D}
AC10: '{BRANCH \C> Figure 7-8. Continued
AD10:'{BRANCH  G}
AE10: '{HOME}
AE11:'{QUIT}
AA13: "D  *
AB13: '{MENUBRANCH Menu2}
AA15: 'Menu2  AB15:  'Name
AC15: 'Oxygen
AD15: 'Diffusers
AE15: 'Blowers
AF15: 'Costs
AG15:'Return
AB16: 'Name the plant being analyzed
AC16: 'Specify oxygen demands & transfer efficiencies
AD16: 'Describe diffuser characteristics
AE16: 'Describe blower characteristics
AF16: 'Provide cost factors & cleaning frequency
AG16: 'Return to Main Menu
AB17: '{GETLABEL "Name of plant being analyzed? ",C4}~
AC17:'{GOTO}A30~
AD 17: '{GOTO}A50~
AE17: '{GOTO}A90~
AF17: '{GOTO}A70~
AG17:'{BRANCH \M}
AB18:'{BRANCH \D}
AC18:'{DOWN5}
AD18:'{DOWN4}                                         ,
AE18: '{DOWN 4}
AF18: '{DOWN 4}
AC19:'{RIGHT 1}
AD19:'{RIGHT4}
AE19: '{RIGHT 4}
AF19: '{RIGHT 4}
AA22: '1C
AB22: '{HOME}
AB23:'{LETS3.0}~
AB24: '{BLANK F7..F13}
AB25: '{BLANK 615}
AB26: '{BLANK P31..Y271}-
AB27: '{GOTO}F7~
AB28: '{WINDOWSOFF}
AB29: '{FOR T3,1 ,@MIN( + E81,240), 1 .MCOST}
AB30: '{WINDOWSON}
AB31: '{LET F9,(E75-@SUM(E64..G64) + E76)/1000}
AB32:'{LET F10,@NPV( + E74/1200,V31..V271)/1000}
AB33:'{LET F11,@NPV( + E74/1200,W31..W271)/1000}
AB34: '{LET F12,@NPV( + E74/1200,X31..X271)/1000}
AB35: '{LET F13,@SUM(F9..F12)}~
AB36: '{IF @ISERR(S24)}{LET B15,**" Insufficient Aeration Capacity ***}-
AB37: '{BRANCH \M}
AA40: 'MCOST
AB40: '{LET S3, + S3+ 1}
AB41:'{LETF7, + S3}~
AB42: '{RECALC S4..U24}
                                            298

-------
AB43: '{IF @ISERR(S24)}{FORBREAK)
AB44: '{PUT P31 ..Y271,0, + 83-1, + S3)
AB45: '{PUTP31..Y271,1,+S3-1,+S15}
AB46: '{PUT P31 ,.Y271,2, + S3-1, + T15}
AB47: '{PUT P31 ..Y271,3, + S3-1, + U15}
AB48: '{PUT P31..Y271.4, +S3-1, +816}
AB49: '{PUT P31 ..Y271,5, + S3-1, + 820}
AB50: '{PUT P31..Y271,6, + S3-1, + 821}
AB51: '{PUT P31 ..Y271,7, + S3-1, + 822}
AB52: '{PUT P31 ..Y271,8, + S3-1, + 823}
AB53: '{PUT P31 ..Y271,9, + 83-1, + 324}
AA55: '\G
AB55: '{MENUBRANCH Menu3}
AB56: '{BRANCH \M}
AA59: 'Menu3
AB59; 'Airflows
AC59: 'Operating-Costs
AD59: 'Total-Costs
AE59: 'Return
AB60: 'Graph monthly airflows
AC60: 'Graph monthly operating costs
AD60: 'Graph total present worth costs
AE60: 'Return to Main Menu
AB61: VgnuAIRFLOW~q
AC61: VgnuCOSTS-~q
AD61: '/gnuPWCOSTS~q
AE61: '{BRANCH \M}
AB62; '{BRANCH \G}
AC62: '{BRANCH \G>
AD62: '{BRANCH \G}
Table D-2 Command Script For Economic Analysis Spreadsheet
/rnr
/rnc\0~-AB1~-
/rne\M-~AB4~-
/rncMENUI ~AB8-~
/rne\D~ABl3~-
/rncMENU2~ABl5~
/rnc\C~AB22~
/rncMCOST~AB40~
/rne\G~AB55~
/rncMENU3~AB59~

/gnrrgtxxP31..P271 ~aV31..V271 ~bY31,.Y151 ~
olaPower-lbTotal-fgbqtfMONTHLY OPERATING COSTS-
txMonth—ty Dollars "-qncCOSTS-
rgtxxP31..P271~aQ31-.Q271 ~bR31..R271 ~eS31..S271 ~-dT31..T271 -
olaZone  1 -IbZone 2~lc2one 3-ldTotal~fgbqtfAIRFLOW REQUIREMENTS-
txMonth~tyCFM-qncAIRFLOW~rgtbxB9..Bl3~aF9..F13~ •
otfPRESENT WORTH COSTS~ty$1,000-daF9..F13-aqqncPWCOSTS~q

/rff1~F9..Fl3~                  •           •
/wgrm

NOTE: The symbol — represents a carriage return
                                            299

-------

-------
                      Appendix E
Symbols, Terms, and Acronyms Used in this Manual
ABS
AD
a
aF

aF(SOTE)

aF(SOTR)

AOR
AP
AT
b
P
BOD
BOD5
BODR
BODU|t
BRV
BRV0
C
C*s

C*a>
CBODs
COD
CRF
CVPC
de
D
DO
dP
dPSUb
DWP
e
EACC
EAP
EAPp
En
        acrylonitrile butadiene styrene
        total projected media surface area of installed diffusers
        (process water K|_a of a new diffuser)/(ciean water K|_a of a new diffuser)
        term representing combined effect of wastewater characteristics and diffuser
        fouling/aging on oxygen transfer performance
        oxygen transfer efficiency under field conditions corrected to 20°C,  1 atm, and
        a driving force of C*«,2o
        oxygen transfer rate under field conditions corrected to 20°C, 1 atm, and  a
        driving force of C*«,2o
        actual oxygen requirement
        adiabatic power consumption
        aeration basin floor area
        decay coefficient
        (process water C*oc,)/(clean water C*,,.,)
        biochemical oxygen demand
        5-day  BOD
        BODS removed
        ultimate BOD
        bubble release vacuum
        bubble release vacuum of a new diffuser
        DO concentration of clean or process water
        tabular value of DO surface saturation concentration at water temperature T,
        standard atmospheric pressure PS. and 100 percent relative humidity
        steady-state DO saturation concentration attained at infinite time at water
        temperature T and field atmospheric pressure Pjj
        steady-state DO saturation concentration attained at infinite time at 20 °C  and
        1 atm
        carbonaceous BOD5
        chemical oxygen demand
        capital recovery factor
        chlorinated polyvinyl chloride
        effective saturation depth at infinite time
        inside  diameter of pipe
        dissolved oxygen
        pressure drop
        pressure drop across a clean diffuser
        pressure drop across a fouled diffuser
        diffuser pressure drop in Zone i
        pressure drop in air piping
        orifice drop at airflow of 1 scfm/diffuser
        head of water above diffusers
        dynamic wet pressure
        combined blower/motor efficiency
        equivalent annual diffuser cleaning cost
        equivalent air pressure
        equivalent annual power cost
        unit power cost
                           301

-------
EF            efficiency factor (see Equation 4-6)
EFR           effective flux ratio                                   .  .
EPDM         ethylenepropylene dimer
f               friction factor, from the Moody diagram
fp             fouling rate (see Equation 3-6)
F              fouling factor = (process water K[_a of a diffuser after a given time in
               service)/(Kj_a of a new diffuser in the same process water)
Fa             average fouling factor
             .  minimum value for  F
               food-to-microorganism loading
               future expenditure
               fiberglass reinforced plastic
               specific weight of water at temperature T <    :
               specific weight of air in pipe
               acceleration due  to gravity
               headloss                 '          , .
               headloss per 100ft of pipe
               water depth
               relative humidity of blower discharge air
               relative humidity at standard conditions
               velocity head in pipe                   -
               high density polyethylene
               hydraulic detention time
               hydraulic retention time
               periodic discount rate
               monthly interest rate
               immediate oxygen demand                                 •
               ratio of specific heats for air
               absolute roughness factor
               coefficient for adiabatic equation = (k - 1)/k
               derivative time           !
               headloss coefficient
               integral time constant
               apparent volumetric mass transfer coefficient in clean water at temperature T
               apparent volumetric mass transfer coefficient in clean water at 20 °C
Kp            proportional gain
I               length of pipe                      '
L              aeration basin length
m             constant that reflects the effect of  airflow rate on SOTE (see Equation 2-2)
M             number of cleanings over the life of the system
MLSS         mixed liquor suspended solids
MLVSS        mixed liquor volatile suspended solids
u.              viscosity
n              total number of time periods
nm            number of months between cleanings
N             equivalent number  of basins-in-series
Nd            number of diffusers in the zone
N|             number of diffusers in Zone i              '
NHa-N         ammonia nitrogen
NH<}-N         ammonium nitrogen               '    •
NOa-N         nitrate nitrogen
NOD          nitrogenous oxygen demand
NPDES        National Pollution Discharge Elimination System
NPT           National pipe thread
£1             pressure correction for C*oo
OTE           oxygen transfer efficiency                 '
OTEj          oxygen transfer efficiency under process conditions
OTRj          oxygen transfer rate under process conditions
Pb            field atmospheric pressure
Pd            blower discharge pressure
Ps            atmospheric pressure at standard conditions
 mjn
F/M
Fw
FRP
Yw
ya
g
hf
hf/1QO
H
Hd
Hs
Hv
HOPE
HOT
HRT
i
im
IOD
k
ks
K
KD
Kj,
KI
«l_a
                                   302

-------
PD
PSPWF
PVC
Pvd
PVS
Px
PWF
q
qa
qmjn
qmix
qs
Q
Qw
r
rr
R
Re
s
SAE
SAN
SBOD
SEM
SOTE
SOTR
SPWF
SRT
S.S.
SVI
SWD
t
0
t
T
AT
Ta
Tp
Ts
TBOD
TKN
TN
TOG
TOD
TP
TSS
v
V
Vs
VS
VSS
w
W
WCARB
positive displacement              :
periodic series present worth factor
polyvinyl chloride                                       . .
vapor pressure of water in blower discharge air
vapor pressure of water at standard conditions
saturated vapor pressure of water at temperature T
present worth cost
biomass produced
present worth factor
airflow rate per diffuser
actual airflow rate
design airflow rate per diffuser                          .-..,•
design airflow rate per diffuser in Zone i
manufacturer's recommended minimum airflow rate per diffuser
minimum airflow rate required for solids suspension per diffuser
manufacturer's recommended maximum airflow rate per diffuser
field standardized volumetric airflow rate
wastewater flow rate
waste solids flow rate
volumetric respiration rate
return activated sludge recycle ratio
idea! gas constant
Reynolds No.
sample standard deviation
standard aeration efficiency
styrene-acrylonitrile
soluble 5-day BOD              ,     ,
scanning electron microscopy
standard oxygen transfer efficiency
standard oxygen transfer rate
uniform series present worth factor
solids retention time
stainless steel
sludge volume index
sidewater depth
temperature correction for C*uo
temperature correction for K^a
time
clean or process water temperature
temperature rise through blower
blower inlet air temperature
air temperature in pipe
air temperature at standard conditions
total 5-day BOD
total Kjeldahl nitrogen
total nitrogen = TKN plus oxidized nitrogen
total organic carbon
total oxygen demand
total phosphorus
total suspended solids
airflow velocity in pipe
water volume           ,
specific volume
volatile solids
volatile suspended solids
mass rate of air
aeration basin width
mass rate of carbonaceous oxygen demand removed
mass rate of nitrogenous oxygen demand removed
mass rate of oxygen demand removed
mass rate of oxygen supplied
                   303

-------
               monthly average wire energy consumption
w.g.           water gauge
WP            wire power consumption (approximately equal to brake horsepower)
WP'           average wire power consumption
               wire power consumption for an aeration system when operating with clean
               diftusers
               wire power consumption for an aeration system when operating with fouled
               diffusers just before cleaning
               power ratio, fouled/clean diffuser operation
x              average of individual sample points
X              MLVSS
Xw            waste VSS
Yfl             yield coefficient
Zc             period diffuser cleaning cost
Ze             present worth power cost
Zp             period power cost
                                 304

-------
    Appendix F
Conversion Factors
Multiply
cm
cm2
ha
kg
kg/m2
kg/m3/d
kW
kPa
Us
Us
Us/m2
Us/m3
m
m2
m3
m3
m3/d
m3/m2/d
by
0.393
0.155
2.47
0.454
0.205
62.4
1.341
0,145
0.023
2.12
0.197
8.024
3.28
10.76
35.3
264.2
264.2
24.55
To Get
1 in
sq in
ac
Ib
Ib/sqft
lb/d/1 ,000 cu ft
hp
psi
mgd
scfm
scfm/sq ft
scfm/1 ,000 gal
ft
sq ft
cuft
gal
gpd
gpd/sq ft
                «U.S. GOVERNMENT PRINTING OFFICE: 1 9 92 . SkB. 0 oo/n J 8 J 2
        305

-------

-------

-------

-------

-------

-------