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
-------
-------
-------
-------
-------
-------
|