&EPA
            United States
            Environmental Protection
            Agency
            Municipal Environmental Research
            Laboratory
            Cincinnati OH 45268
EPA-600/2-79-134
August 1979
            Research and Development
Disinfection/
Treatment of
Combined Sewer
Overflows

Syracuse, New York

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was  consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental Health  Effects Research
      2.  Environmental Protection Technology
      3.  Ecological Research
      4.  Environmental Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports

This report has  been assigned  to the  ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate  instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution-sources to meet environmental quality standards.
 This document is available to the public through the National Technical Informa-
 tion Service, Springfield, Virginia 22161.

-------
                                             EPA-600/2-79-134
                                             August 1979
   DISINFECTION/TREATMENT OF COMBINED SEWER OVERFLOWS

                   Syracuse, New York
                           by

                    Frank J.  Drehwing
                    Arthur J. Oliver
                   Dwight A.  MacArthur
                     Peter E. Moffa
             O'Brien & Gere Engineers, Inc.
                Syracuse, New York  13201
              Grant No. S802400 (11020HFR)
                     Project Officer

                      Richard Field
            Storm and Combined Sewer Section
              Wastewater Research Division
Municipal .Environmental Research Laboratory (Cincinnati)
                Edison, New Jersey  08817
       MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
           OFFICE OF RESEARCH AND DEVELOPMENT
          U.S. ENVIRONMENTAL PROTECTION AGENCY
                 CINCINNATI, OHIO  45268

-------
                              DISCLAIMER

     This report has been reviewed by the Municipal Environmental Research
Laboratory, U.S. Environmental Protection Agency, and approved for
publication.  Approval does not signify that the contents necessarily
reflect the views and policies of the U.S. Environmental Protection
Agency, nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.

-------
                               FOREWORD
     The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the
health  and welfare of the American people.  Noxious air, foul water, and
spoiled land are tragic testimony to the deterioration of our natural
environment.  The complexity of that environment and the interplay between
its components require a concentrated and integrated attack on the problem.

     Research and development is that necessary first step in problem
solution and it involves defining the problem, measuring its impact, and
searching for solutions.  The Municipal Environmental Research Laboratory
develops new and improved technology  and systems for the prevention,
treatment and management of wastewater and solid and hazardous waste
pollutant discharges from municipal and community sources, for the pre-
servation and treatment of public drinking water supplies and to minimize
the adverse economic, social, health, and aesthetic effects of pollution.
This publication is one of the products of that research; a most vital
communications link between the researcher and the user community.

     The deleterious effects of storm sewer discharges and combined
sewer overflows upon the nation's waterways have become of increasing
concern in recent times.  Efforts to alleviate the problem depend upon
characterization of these flows in both a quantity and quality sense.
This report describes the results of a full-scale facilities demonstration
study of a number of treatment devices for controlling the quality of
combined sewer overflow discharges.


                              Francis T. Mayo
                              Director
                              Municipal Environmental Research Laboratory

-------
                                  ABSTRACT

     The Syracuse demonstration program was designed to evaluate treatment
of combined sewer overflows (CSO) for the Onondaga County Department of
Drainage and Sanitation.

     The demonstration study covered field evaluations of high-rate
screening and disinfection by the following unit processes:  fine-mesh
screening by three separate microscreening devices, swirl regulator/
concentrator, and disinfection utilizing chlorine (Cl2) and chlorine dioxide
(C102).  Applied flowrates to the microscreening devices ranged from 210 to
1660 gpm (13.2 to 104.6 I/sec), and applied flowrates to the swirl regulator/
concentrator ranged from 140 to 5280 gpm (8.8 to 332.6 I/sec).

     The screening facilities operation covered a total of 16 overflow
events during the period of March 1975 through October 1976.  A total of
11 overflow events were studied using the swirl unit during the period
of May 1974 to September 1975.  All studies evaluated the effects of
hydraulic and pollutant loadings and influent quality on system performance.

     The three microscreening units were a 5-ft (1.5m) diameter Sweco
Wastewater Concentrator utilizing 105y screen apertures, a 6ft x 6ft
(1.8m x 1.8m) Zurn Micromatic utilizing 71u screen apertures,  and a
7.5ft x 5ft (2.3m x 1.5m) Crane Microstrainer utilizing 23ja screen apertures.
Storm average hydraulic loading rates employed in the evaluations ranged
from 11.2 to 66.4 gpm/ft^ (27.3 to 162.0 m/hr) for the Sweco unit, 3.0 to
12.3 gpm/ft2 (7.3 to 30.0 m/hr) for the Zurn unit, and 1.7 to 7.7 gpm/ft2
(4.1 to 18.8 m/hr) for the Crane unit.  Using multiple regression analysis
techniques, mathematical performance models were .developed for each of the
microscreening units relating SS removal rates to influent hydraulic loading
and to influent SS concentrations.  Average SS removal rates in terms of
concentration were approximately 32 percent for the Sweco unit, 45 percent
for the Zurn unit, and 58 percent for the Crane unit.  In terms of mass
removal, the Sweco unit averaged 48 percent - a significantly higher
removal rate than exhibited in terms of SS concentration removal efficiency.
The increased level of solids mass removal is due to the large fraction of
influent flow returned to  the intercepting sewer.  Comparisons of the
performance results with results published in the literature indicated similar
trends of SS removal.

     A 12.33 ft diameter swirl regulator/concentrator was evaluated for SS
removal efficiency at flowrates ranging from 0.2 to 7.6 M6D (0.5 to 20 cum/
min) or hydraulic loading rates ranging from 1.2 to 44.4 gpm/ft^
(2.9 to 108.3 m/hr).  Mathematical models were developed relating SS removal
efficiencies to influent hydraulic loading rates and influent SS concentra-
tions.  Results indicated SS concentration removal efficiencies ranging from

                                      iv

-------
18 to 55 percent and SS mass removal efficiencies ranging from 33 to 82 percent.
A settling velocity distribution analysis of solids particles was conducted
to determine the predicted SS removal under a given set of operating conditions
for comparison with actual measured SS removal efficiencies.  The results
tended to confirm the predicted performance curve determined in previous
model studies.

     Multiple regression modeling of the Cl2 and C102 disinfection data
yielded statistically significant performance equations for the high-rate
disinfection systems.  The results of the study at Maltbie Street indicate
that application of high-rate disinfection processes can result in signifi-
cant reduction of bacterial populations in CSO.  C10£ dosages in the order
of 6 to 12 mg/1 applied in the initial stages of overflows reduced FC levels
to 200 counts/100 ml.  Applied dosages of 4 mg/1 after first-flush loadings
had passed through the treatment system, maintained the 200 counts FC/100 ml
level in the majority of the samples collected.  Application of Cl2 at
dosages of from 12 to 24 mg/1 during the initial stages of overflow also were
able to achieve 3 to 4 log reductions of FC, while lower dosages (12 mg/1)
produced similar reductions after the first 30 to 45 min.  Sequential
addition of disinfectants  (2 mg/1 C102 followed by 8 mg/1 Cl2 after 15 sec)
at a total contact time of 1 min produced 3 to 4 log reductions of FC.  The
limited data obtained in the sequential addition tests precludes a comparison
of this method of disinfection with the application of Cl2 or C102 separately.

     Regression analyses of the disinfection data collected indicated that
removal of SS would improve the reduction of bacterial populations by the
disinfection processes.  The effects of solids removal  are slightly more
pronounced for C102 than for Cl2 disinfection with both exhibiting improved
FC kills of about one-half to one order of magnitude.

     Capital and operating cost estimates indicated that disinfection
facilities would be less expensive utilizing Cl2 as the disinfectant rather
than C102-

     Preliminary evaluations were conducted on the impact of transmitting
CSO treatment residuals to the Metropolitan Syracuse Sewage Treatment
Plant (Metro).  The analysis included the effects of transmission of CSO
sludges as well as transmission of dilute residuals from on-site sludge  •
dewatering processes on both the primary and secondary treatment facilities
at Metro.  The effects of transmitting CSO sludges directly to the Metro
sludge handling facilities were also examined.  Results of the analysis
indicated that return of CSO treatment residuals would result in a solids
overload of the primary facilities at Metro and that the secondary treatment
facilities would suffer an organic overload if the bleedback of CSO residuals
occurred at flow conditions higher than average dry-weather conditions.

     Transmission of CSO treatment residuals directly to the dry-weather
plant sludge handling facilities would result in hydraulically overloading
the gravity thickeners, thereby reducing the solids retention time in the
thickeners from a design retention time of 7 hours to 4 hours or less for a
1 year - 2 hour storm.  Hydraulic overload would also be evident in the
sludge digesters and would result in a reduced solids retention time.  A

-------
return of incomplete digestion products to  the head of the treatment plant
would be expected.  The impact of CSO residual bleedback was also evaluated
from a solids loading viewpoint.  Results indicated that the bleedback would
cause the solids loading to be approximately equal to the peak allowable
solids loading, and the sludge handling facilities would be limited in their
capacity to handle solids loadings from back-to-back storm events.

     Capital and operating costs are presented for point-source treatment
of all overflows in Syracuse based on a 15 MGD (56,775 cu m/day) capacity
satellite CSO treatment facility serving a 115 acre (46.5 ha) drainage area.
Costs were then extrapolated for the entire Syracuse CSO drainage area. These
costs are compared to previous estimates which considered centralized
treatment adjacent to Metro and point-source treatment at individual
overflow sites utilizing conventional disinfection design parameters.  The
study projected annual costs for CSO treatment facilities in Syracuse as
follows:

     Treatment Device         Cl? Disinfection         ClOg Disinfection

          Sweco               $ 12,294,500             $ 12,665,700
          Zurn                   7,669,900                8,076,700
          Crane                  5,888,900                6,307,500
          Swirl                  3,264,300                3,641,900

     Special studies were conducted during the demonstration program and are
also presented in this report.  The special studies included virus inactiva-
tion, bench-scale ultraviolet light disinfection, on-line monitoring of SS,
continuous monitoring of TOC, and a brief discussion of chlorinated
hydrocarbon analyses.

     This report was submitted in fulfillment of Grant No. S802400 (formerly
11020HFR) by O'Brien & Gere Engineers, Inc. under the partial sponsorship
of the U.S. Environmental Protection Agency.   This report covers the period
from 1972 to 1976, and work was completed as of January 1978.
                                     VI

-------
                                  CONTENTS

Foreword 	    i i i
Abstract	    iv
Fi gures	    vi i i
Tab! es	    xi
Abbreviations and Symbols	    xiv
Acknowl edgement	    xvi i

      1.  Introduction	       1
      2.  Conclusions	       7
      3.  Recommendations—	      14
      4.  Experimental Plan	      16
               Plan Objectives	      16
               Solids Removal Considerations	      16
               Disinfection Considerations	      19
               Virus Considerations	      21
               Development of Experimental  Plan	      21
      5.  Pre-Construct!"on Studies	      27
               Preliminary Monitoring	      27
               Bench Seal e Studies	      37
      6.  Facilities Description and Operation	      43
               Maltbie Street Facility	      43
               West Newel 1 Street Facil ity	      56
      7.  CSO Loadings At Demonstration Sites	      65
               Rain Data Analysis	      65
               Field Program	      66
               Simplified SWMM Analysis	      67
      8.  Microscreening - Results and  Discussion	      70
               General	      70
               Suspended Sol ids Removal	      70
               Organic Solids Removal	      78
               Heavy Metal s Removal	      78
      9.  Swirl Regulator/Concentrator  -  Results and  Discussion	      82
               General	      82
               Settling Velocity Tests	      82
               Suspended Sol ids Removal	      88
               Organics Removal	      93
               Heavy Metal s Removal	      95
               Coarse Fl oatabl es Removal	:.      95
     10.  Solids Handling Considerations  -  Results  and  Discussion...      97
               General	      97
               Full Return of CSO Treatment Residuals to  Metro	      98
               Return of Dilute CSO Treatment Residuals to  Metro	     103
               Direct Transmission of CSO Treatment Residuals  to
                 Metro Sludge Handling  Facilities	     106

                                    vii

-------
     11.  Disinfection - Results and Discussion	    109
               General	    109
               C102 Disinfection - 1975 Operations	    109
               Cl2 and C102 Disinfection - 1976 Operations	    114
               Multiple Regression Analysis	    120
               Discussion of Disinfection Results	    130
     12.  Capital  and Operating Cost Estimates	    133
               General	    133
               Actual Capital Costs	    133
               Actual Operati ng Costs	    135
               Projected Costs for Ful 1 CSO Treatment	    136
     13.  Virus Studies - Maltbie Street Facility	    141
               General	    141
               Phase I Program	    141
               Phase 11 Program.	    147
               Summary of Virus Studies	    154
     14.  Special Considerations	    156
               General	    156
               Special Analyses	    156
               Special Investigations	1	    163
               Special Instrumentation	    187

References	,	    213

Appendices

     A.   Typical Preliminary Monitoring Data	    221

     B.   Chlorine Dioxide Analytical Data	    227

     C.   Chlorine Analytical Data	    230,

     D.   Analytical Data for Sequential  Addition of
            Disinfectants	    232

     E.   Analytical Data for Swirl  Prototype  SS Removal	    234

     F.   Swirl Prototype Storm-Averaged  Data	    236

     G.   Analytical Data for Adenosine Triphosphate
               Investigations	    238
                                    vm

-------
                                  FIGURES
Number                                                                Page
     1    Onondaga Lake	  2
     2    Typical Overflow Event Being Monitored at Maltbie Street	 28
     3    Maltbie Street Site Plan	46
     4    Maltbie Street Site Location	46
     5    Maltbie Street Process Orientation	47
     6    Crane Microstrainer	50
     7    Zurn Micromatic	50
     8    Sweco Wastewater Concentrator	51
     9    Schematic of Horizontal Shaft Drum Screen	52
    10    Schematic of Vertical Shaft Drum Screen	52
    11    Maltbie Street - Chlorine Disinfection Equipment	55 .
    12    Maltbie Street - Chlorine Dioxide Generators	55
    13    West Newell Street Site Plan	58 •
    14    West Newell Street Site Swirl  Regulator/Concentrator	58.
    15    Isometric View of Swirl Regulator/Concentrator	61
    16    Schematic Profile - West Newell Street CSO Facilities	62
    17    Swirl Regulator/Concentrator- Dry-Weather Operation	63
    18    Swirl Regulator/Concentrator -Wet-Weather Operation	63
    19    West Newell Street - Chlorine Dioxide Generators	63
    20    Syracuse Area-Rainfall Intensity vs. Duration	65
    21    Mass Emission of BODs vs Total Rainfall	67
    22    Mass Emission of SS vs Total Rainfall	67
    23    Mass Emission of VSS vs Total  Rainfall	68
    24    Mass Emission of TKN vs Total  Rainfall	68
    25    Mass Emission of NHaN vs Total Rainfall	68
    26    Mass Emission of TIP vs Total  Rainfall	68
    27    Microscreening-SS Concentration Removal vs Hydraulic
            Loading Rate....'.	75
    28    Microscreening SS Mass Removal vs Hydraulic Loading Rate	75
    29    Microscreening-SS Removal vs Screen Aperture Size 	77
    30    Microscreening-SS Removal vs Influent SS Concentrations	78
    31    Settling Velocity Distribution Curve -West Newell Street
            CSO	83
    32    Predicted Performance of Prototype Swirl vs Flowrate	86
    33    Prototype Swirl Regulator/Concentrator SS Removal	90
    34    Observed vs Predicted SS Removal-Swirl Prototype	91
    35    Percent Foul Fraction vs SS Concentration Removal -
            Swirl Prototype	91
    36    VSS/SS Ratio vs VSS Removal -Swirl Prototype	94
    37    Coarse Floatables Removal - Swirl Prototype	95
    38    C102 Disinfection Following Sweco Storm 2, 1975	  112


                                    ix

-------
Number
Page
    39    C102 Disinfection Following Sweco Storm 3, 1975	112
    40    C102 Disinfection Following Sweco storm 5, 1975	112
    41    C102 Disinfection Following Sweco Storm 6, 1975	112
    42    C102 Disinfection Following Sweco storm 8, 1975	113
    43    C102 Disinfection Following Sweco storm 9, 1975	113
    44    C102 Disinfection Following Sweco storm 10, 1975	113
    45    C102 Disinfection Following Sweco Storm 11, 1975	113
    46    C102 Disinfection Following Zurn Storm 3, 1975	115
    47    C102 Disinfection Following Zurn Storm 5, 1975	115
    48    C102 Disinfection Following Zurn Storm 6, 1975	115
    49    C102 Disinfection Following Zurn Storm 8, 1975	115
    50    Cl2 Disinfection  Unscreened CSO - Storm 1, 1976	117
    51    Clp Disinfection Unscreened CSO - Storm 2, 1976	117
    52    C102 Disinfection Unscreened CSO - Storm 3, 1976
    53    Cl2 Disinfection Screened CSO (23y) - Storm 3, 1976	117
    54    Cl£ Disinfection Unscreened CSO - Storm 4, 1976	118
    55    C102 Disinfection Screened CSO (23u) - Storm 4, 1976	118
    56    C102 & Cig Sequential Disinfection Screened CSO (23y) -
            Storm 5, 1976	118
    57    C102 & Cl2 Sequential Disinfection Unscreened CSO -
            Storm 5, 1976	118
    58    C102 Disinfection Unscreened CSO - Storm 6, 1976	120
    59    Regression Model Results-GT vs FC Reduction	123
    60    Effect of Cl2 Dosage - GT vs FC Reduction	126
    61    Effect of C102 Dosage - GT vs FC Reduction	126
    62    Effect of SS on Cl2 & C102 - Dosage vs FC Reduction	127
    63    Effect of FC on Cl2 & C102 - Dosage vs FC Reduction	127
    64    Effect of pH on Cl2 & C102 - Dosage vs FC Reduction	129
    65    Cl2 Disinfection - Observed vs Predicted Kill	129
    66    C102 Disinfection - Observed vs Predicted Kill	129
    67    Sequential Addition of C102 & C12 - FC Count vs FC Reduction 131
    68    Sequential Addition of C102 & Cl2 - GT vs FC Reduction	131
    69    Iso-Kill Curves - Syracuse and Rochester Studies	132
    70    Inactivation of f2 Phage - Storm 1	149
    71    Inactivation of f2 Phage - Storm 4	149
    72    Inactivation of 0X174 Phage - Storm 1	149
    73    Inactivation of 0X174 - Storm 4	149
    74    Schematic of Bench-Scale UV Disinfection Testing Apparatus..165
    75    Number of Surviving TC vs yW-sec/cm2	166
    76    Percent Inactivation of TC vs uW-sec/cm2		168
    77    Effect of Ultraviolet Light Intensity on Disinfection	169
    78    Effect of UV Exposure on Cl2 Disinfection	170
    79    Effect of Cl2 Dose on Disinfection Following Ultraviolet
            Light Exposure	170
    80    Surviving TC vs Reaction Time for Simultaneous Disinfection
            by UV and Cl2	172
    81    Single-Stage Disinfection of Poliovirus Sabin Type I with
            Cl 02	185
    82    TOC Monitor Data Obtained During Calibration Procedures	188
    83    Comparison of TOC Monitor Results with Laboratory Results...139

                                    x

-------
Number                                                                Page
    84    Schematic of Suspended Sol ids Monitor	•	192
    85    Electronics Package for Suspended Solids Meter	197
    86    Depolarization Ratio vs Ungraded Kaolin Clay Concentration -
            No Depolarization-Polarization Gain Adjustment	199
    87    Depolarization Ratio vs Ungraded Kaolin Clay Concentration -
            Unbalanced Depolarization-Polarization Gain Adjusted	200
    88    Depolarization Ratio vs Ungraded Kaolin Clay Concentration -
            Adjusted for Reradiation  Effects	200
    89    Depolarization Ratio vs Graded Kaolin Concentration	203
    90    Depolarization Ratio vs Kaolin, Diatomaceous Earth,
            Rottenstone, and Pumice Concentrations	204
    91    Depolarization Ratio vs Biological Solids Concentration	205
    92    Depolarization Ratio vs Nigrisine Black Concentration	206
    93    Depolarization Ratio vs Nigrisine Black and Kaolin
            Concentrations	207
    94    Depolarization Ratio vs Treatment Process SS Concentrations. 208
    95    SS Monitor Percent of Scale Readings vs SS Concentrations in
            Syracuse Field Tests	210
                                     XI

-------
                                   TABLES

Number                                                                Page
     1    West Newell Street Disinfection Evaluation Schedule	  26
     2    Preliminary .Monitoring-Water Quality Parameters	  28
     3    Combined Sewer Overflow Quality -Maltbie Street Overflow	  30
     4    Combined Sewer Overflow Quality - West Newell Street
            Overflow	  31
     5    Combined Sewer Overflow Quality - Rowland Street Site	  32
     6    Wet-Weather Onondaga Creek Quality Upstream of CSO's At
            Dorwin Avenue	  34
     7    Wet-Weather Onondaga Creek Quality- Downstream of CSO's
            At Spencer Street	  35
     8    Quantities of CSO Pollutants Discharged	  36
     8a   Summary of Quality Data for Selected Parameters on a
            Site-By-Site Basis in Onondaga County	  38
     8b   Illustration of "First-Flush" Effects from Onondaga
            County CSO.	  39
     9    Maltbie Street Overflow Characteristics	  43
    10    Microscreen Operating Characteristics	  49
    11    West Newell'Street Overflow Characteristics	  56
    12    3wirl Regulator/Concentrator Design Dimensions	  61
    13    Maltbie Street Operation Schedule - Microscreening	  71
    14    Sweco Centrifugal Wastewater Concentrator Suspended Solids
            Removal	  72
    15    Zurn Micromatic Suspended Solids Removal	  74
    16    Crane Microstrainer Suspended Solids Removal	  74
    17    Sweco CWC-Organic Solids Removal	  79
    18    Zurn Micromati c Organic Sol ids Removal	  79
    19    Crane Microstrainer Organic Solids Removal	  79
    20    Sweco CWC-Heavy Metals Removal	  80
    21    Zurn Mi cromati c Heavy Metal s Removal	  80
    22    Particle Size Distribution (Percent) and Solids Ranges in
            Sanitary Sewage, Stormwater Runoff and Combined Sewer
            Overflows	'	  84
    23    Predicted vs Actual SS Removal - Swirl Regulator/
            Concentrator	  87
    24    Swirl Regulator/Concentrator Suspended Solids Removal	  89
    25    Multiple Regression Analysis for Swirl Prototype SS
            Removal	.'	  92
    26    Swirl Prototype BOD5 Removal	  93
    27    Swirl Prototype TOC and VSS Removals	  94
    28    Swirl Prototype Heavy Metal s Removal	  96
    29    Effect of CSO Treatment Sludges on Hydraulic Loading
            at Metro STP	  99
    30    Effect of CSO Treatment Sludges on Solids
            Loading at Metro STP	<	  101
                                    xii

-------
Number                                                                Page
    31    Organic Characteristics (BOD) of CSO Treatment
            Sludges	 103
    32    Solids Loading to Secondary Facilities at Metro
            STP	 104
    33    Effect of Dilute Effluent From On-Site Dewatering
            of CSO Sludges to Metro STP	 105
    34    Volatile Solids Content of Sludges from Various CSO
            Treatment Devi ces	 108
    35    Maltbie Street Operation Schedule -Disinfection	 HO
    36    Multiple Regression Analysis Results for C102	 122
    37    Multiple Regression Analysis Results for Cl2	 123
    38    Actual Costs for Maltbie Street Screening
            Facil ities	 134
    39    Actual Costs for West Newell Street Swirl
            Regul ator/Concentrator	 135
    40    Design Parameters - CSO Treatment Facilities	 137
    41    Summary of Capital Costs - Syracuse CSO Treatment
            Facilities	 138
    42    Summary of Annual O&M Costs -Syracuse CSO Treatment
            Facil ities	 138
    43    Projected Capital Costs of Pumping and Site Work -
            CSO Treatment Facilities	 139
    44    Projected Annual Costs -Syracuse CSO Treatment
            Facilities	 140
    45    Comparative Inactivation of Viruses and Enterobacteria	 150
    46    Disinfection of CSO Seeded With PVl-Storm 3	•	 152
    47    Disinfection of CSO Seeded With PVl-Storm 4	 152
    48    Disinfection of CSO Seeded With PVl-Storm 5	 153
    49    Disinfection of CSO Seeded With PVl-Storm 6	 153
    50    Chlorinated Hydrocarbon/Pesticide Scan During Disinfection
            Tests	;	 156
    51    Volatile Chlorinated Organic Concentrations Produced From
            Various Disinfection Schemes Applied to Simulated CSO	 160
    52    Ultraviolet System Costs - 15 M6D Capacity	 174
    53    Combined Cl2 and UV System Costs - 15 MGD Capacity	 175
    54    Chlorine System Costs - 15 MGD Capacity	 176
    55    Combined Cl2 and ClOg System Costs - 15 MGD Capcity	 176
    56    Summary of Costs for Disinfection Systems - 15 MGD
            Capacity..'	 177
    57    ATP Content of Bacteria	 180
    58    Effect of Cl2 on ATP Assay Mixture	 182
    59    Effect of CIO? on ATP Assay Mixture	 182
    60    Effect of Sodium Thiosulfate on ATP Assay Mixture	 183
    61    Linear Regression Correlation Coefficients of ATP vs Test
            Bacteria	 183
    62    Disinfection of CSO in Full-Scale Facility	 186
    63    Deviation of Measured TOC From Known TOC Standard	 189
    64    Ratios Involving the Perpendicular and Parallel Components
            of Depolarized Light	 195
                                    xi ii

-------
                          ABBREVIATIONS  AND SYMBOLS

ABBREVIATIONS

APWA      —American Public Works Association
ATP       --adenosine triphosphate
BOD5      --5 day biochemical  oxygen  demand at  20°C
C         --centigrade
cm        --centimeters
COD       --chemical oxygen demand
CSO       --combined sewer overflow
CSS       —Onondaga County Comprehensive  Sewerage Study
cu m      —cubic meters
DC        --direct current
dia       —diameter
DWF       —dry-weather flow
ENR       --Engineering News Record
ft        —feet
ft2       --square feet
ft3       —cubic feet
EDTA      --ethylenedi ami netetraacetate
EPA       --U.S. Environmental  Protection  Agency
FC        —fecal col iform
fps       --feet per second
FS        —fecal streptococcus (or - cocci)
G         --velocity gradient (sec"1)
gal       --gallons
gpm       —gallons per minute
ha        --hectares
hr        --hour
in        --inches
kg        —kilograms
KWH       —kilowatt hours
1         --liters
Ib        —pounds
m         —meters
m2        —square meters
m3        —cubic meters
ma        —mi Hi amperes
Metro     --Syracuse Metropolitan Sewage Treatment Plant
MG        --million gallons
mg/1      --milligrams per liter
MGD       --million gallons per day
MIS       —Main Intercepting Sewer
ml        — mill il Her
                                    xiv

-------
ABBREVIATIONS (CONT'D)
MLSS      —mixed liquor suspended solids
mm        —millimeter
min       —minute
yw/cm2    --microwatts per square centimeter
nm        --nanometers
NYSDEC    --New York State Department of Environmental  Conservation
0&6       --oil and grease
O&M       --operation and maintenance
OrgN      --organic nitrogen
PFU       --plaque forming units
rpm       --revolutions per minute
SCSO      --simulated combined sewer overflow
sec       --second
Sett-S    —settleable solids
sq km     --square kilometers
sq mi     --square miles
SS        —suspended solids
SWMM      —Stormwater Management Model
TC        --total coliform
IDS       --total dissolved solids
TIP       --total inorganic phosphorus
TKN       --total kj el da hi nitrogen
TOC       —total organic carbon
TS        —total solids
TSS       —total suspended solids
T-Alk     —total alkalinity
UV        —ultraviolet                                         *
VAC       —volts of alternating current
VS        --volatile solids
VSS       --volatile suspended solids
yr        —year
SYMBOLS

a
Al
Aids
Ca
Cd
ci-
C12
C102
C102-
C02
HC1
HOC!
K
-angstroms
•aluminum
-aluminum chloride
-calcium
-cadmium
-chloride ion
•chlorine
-chlorine dioxide
•chlorite ion
-carbon dioxide
•hydrochloric acid
-hypochlorous Acid
-potassium
                                    xv

-------
SYMBOLS (Cont'd)

Mg        --magnesium
NaOH      --sodium hydroxide
NHsN      --ammonia nitrogen
Ni        --nickel
N02       --nitrites
NOs       --nitrates
OC1-      --hypochlorite ion
P         --phosphorus
Pb        -lead
Zn        --zinc
                                     xvi

-------
                               ACKNOWLEDGMENTS

     O'Brien & Gere Engineers, Inc.  gratefully acknowledges the cooperation
of the Onondaga County Department of Drainage and Sanitation.  Appreciation
is expressed to Mr. John J. Hennigan, Jr., Commissioner, Mr. John M. Karanik,
Deputy Commissioner, and Mr. Randy Ott, Project Engineer for their cooperation
and assistance.
     The support of this effort by the Storm and Combined Sewer Section,
Edison, New Jersey of the USEPA Municipal Environmental Research Laboratory,
Cincinnati, Ohio, and especially of Richard Field, Chief, Storm and Combined
Sewer Section, USEPA, for their guidance, suggestions, and contributions
is acknowledged with gratitude.
     Valuable assistance to the program was furnished by Dr. James E. Smith
of Syracuse University  and his staff through the design, conduct, and
analysis of the virus studies at the Maltbie Street facility.
                                     xvn

-------
                                   SECTION 1

                                 INTRODUCTION
GENERAL

     Discharge of untreated domestic and industrial waste and of primary
sewage treatment plant effluent has long been recognized as a major
source of pollution in the nation's waterways.  More recently, within
the past two or three decades, the importance of storm and combined
sewer overflow (CSO) as a major source of contamination has also been
recognized. The significance of CSO has increased as a result of the
addition of wastewater from new residential, commercial, and industrial
developments to the finite capacity of existing combined sewer systems.

     Although many advances have been made in waste treatment technology
in recent years, receiving water quality cannot be consistently maintained
without controlling and/or treating the high volume discharges from
combined and storm sewers.  It was the purpose of the study described in
this report to demonstrate the feasibility of treating a CSO at the
overflow point rather than at a centralized treatment facility through
high-rate treatment techniques.  These techniques included microscreening
followed by high-rate disinfection, and high-rate quality and quantity
control through application of the swirl regulator/ concentrator.


BACKGROUND

Description of Study Area

     Onondaga County, located in the central region of New York State
has a total area of 792 sq mi (2,050 sq km) and a population (estimated
1976) of approximately 473,000.  Most of the county is located in the
Oswego River Basin.  Onondaga Lake, which drains to the Oswego River, is
included in the lowland region which extends across the northern part of
the county.  The lake is shown in Figure 1.

     Soils in the county are predominantly glacial till, which is composed
of silt, sand, and gravel.  The soil contains numerous cobbles and
boulders, deposited by receding glaciers.  Precipitation is generally
distributed evenly across the county.  The average annual precipitation
is about 36.9 in./yr (94 cm/yr).

-------
     The City of Syracuse is located approximately at the center of the
county.  The city has a land area of about 16,000 acres (6,480 ha) and a
                     FIGURE  1.   Onondaga  Lake
population (estimated 1976) of approximately 173,000.  Sewage from the
central urbanized area of the county, consisting of the city and adjacent
sections of the suburban towns of Dewitt, Salina, and Geddes, is conveyed
either to the Metropolitan Sewage Treatment Plant (Metro) or to the Ley
Creek Sewage Treatment Plant.  Primary treatment is presently provided
at the Ley Creek plant.  All flow from the Ley Creek plant, including
treated and untreated excess flows, is pumped to Metro.

     All sewage is conveyed in one of three intercepting systems.  The
Main Intercepting Sewer (MIS), which follows Onondaga Creek, and the
Harbor Brook Intercepting Sewer, which follows Harbor Brook, are both
tributary to Metro.  The Ley Creek intercepting system is tributary to
the Ley Creek Sewage Treatment Plant.

-------
     The MIS runs from Metro at Hiawatha Boulevard south along Onondaga
Creek to Pacific Avenue, a distance of about 27,000 ft (8,200 m).  It
serves a total area of about 11,750 acres (4,760 ha) in the City of
Syracuse and the Town of Dewitt.

     The Harbor Brook Intercepting Sewer runs from Metro southwesterly
to Velasko Road, a distance of approximately 15,700 ft (4,800 m).  It
serves a total area of about 2,150 acres (870 ha).

     The Ley Creek trunk sewer serves an area of about 8,750 acres
(3,500 ha) in and adjacent to the northeastern section of the city.  The
trunk sewer is approximately 30,000 ft (9,200 m) long.

     An area of approximately 9,000 acres (3,650 ha) in the ^central
urbanized area is served by combined sewers.  This combined system is
within the Main and Harbor Brook Interceptor Systems.  The collection
system tributary to the Ley Creek trunk sewer is a separated system,
collecting and conveying only sanitary and industrial wastes.

     The MIS and the Harbor Brook Intercepting Sewer together have a
maximum capacity of 150 M6D (6.6 m3/sec), which is about twice the
anticipated dry weather flow from the served areas.  Diversion devices
are located in manholes at points where combined sewers intersect the
intercepting sewers.  Wastewater in excess of the capacity of the intercept-
ing sewers is discharged directly into Onondaga Creek or Harbor Brook at
these diversion manholes.  The diversion devices include leaping weirs,
side overflow weirs, dam and orifice devices, and drop manholes, and are
in varying states of structural condition.

Previous Studies

     The Federal Water Pollution Control Act Amendments of 1972 were
enacted on October 18, 1972.  The objective of this legislation was:
"to restore and maintain the chemical, physical, and biological integrity
of the nation's water".  Included in the goals and policies were the
following declarations: "It is the national goal that the discharge of
pollutants into the navigable waters be eliminated by 1985.", and "It is
the national policy that a major research and demonstration effort be
made to develop the technology necessary to eliminate the discharge of
pollutants into the navigable waters	".

     In 1974 the Environmental Protection Agency -(EPA) published a
report entitled, "Urban Storm Water Management and Technology:  An
Assessment" (1).  This report stated that "During the next decade, it is
expected that billions of dollars will be spent in the United States to
combat the degradation of streams and other water bodies by pollutants
released through storm discharges and combined sewer overflows."  The
statements above emphasize the Federal government's concern about the
quality of the nation's waters and demonstrate its determination to find
a solution to CSO discharges through control and/or treatment.

-------
     The County of Onondaga has shared this concern with the Federal
government.  It is the overall goal of the county to reduce or totally
eliminate pollutional discharges to the waters of Onondaga Lake and by
so doing, to upgrade the water quality of the lake and protect downstream
waters, which include the Seneca River, Oswego River and Lake Ontario.
To achieve this goal the Onondaga County Department of Drainage and
Sanitation in conjunction with EPA has undertaken a number of studies to
define and ultimately abate pollution to the lake.

     The studies completed to date are as follows:

     1.   The Onondaga County Comprehensive Sewerage Study (2)
     2.   An Industrial Waste Study (3)
     3.   An Onondaga Lake Study (4)
     4.   The Onondaga Lake Monitoring Program (5)

     Two of the above studies, the Comprehensive  Sewerage Study and the
Onondaga Lake Study, have direct bearing on the demonstration studies
described in this report.

     The Comprehensive Sewerage Study  (CSS) was. published in November,
1968. Its objective was to set forth a broad master plan of sewage
requirements for Onondaga County which could be implemented to assure
acceptable quality for the receiving waters in the county.  The master
plan formed the basis upon which more specific facilities planning could
be conducted.

     The CSS included the assembly and evaluation of all pertinent
existing information, inspection and investigation of existing private
and municipal sewage facilities (including overflow points and combined
sewers) and evaluation of major municipal and industrial wastewater
discharge outlets. From the resulting data, alternate solutions for
collection systems were prepared.  Proposals for treatment of wastewater
within the study area were developed and recommendations for the most
feasible solutions to the problem of wastewater discharge were presented.
Within the central urban area of the county, a total of 87 CSO locations
to Onondaga Lake or its three tributaries, Ley Creek, Onondaga Creek,
and Harbor Brook were identified. Numerous discharges were found to
occur at these overflows, primarily due to inadequate or reduced wet-
weather capacity of conveyance systems.

     The CSS indicated that capacities for design peak dry-weather flow
rates are commonly from two to four times the average dry-weather rates,
estimated to be realized in the design year.  The average rate of sewage
flow is estimated to be equivalent to the runoff rate of about 0.01 in/hr
(0.025 cm/hr) from the area served.  Intercepting sewers, thus, are
usually designed to have capacities not greater than for equivalent storm-
water runoff rates of about 0.02 to 0.04 in/hr (0.05 to 0.10 cm/hr).  The
capacities of Syracuse intercepting sewers fall within this range.  The
combined sewer system, not including the intercepting sewers, is reported
to have a capacity to carry storm flows equivalent to a runoff rate of
about 0.50 in/hr (1.3 cm/hr), in addition to dry-weather flow.

-------
     Even at times of generation of sanitary sewage at average or less
than average rates, there is little capacity in intercepting sewers for
mixed sanitary sewage and stormwater.  Thus, it is true, that even for
storms of low rainfall intensity, the greater proportion of mixed sewage
and stormwater cannot be accepted by the intercepting sewer system from
the combined sewers and must discharge through outlets to the streams or
to Onondaga Lake.

     After evaluating various methods of treating CSO, the Comprehensive
Study recommended that the CSO be conveyed to a centralized treatment
facility adjacent to Metro.  The estimated construction cost for centralized
treatment totaled $170,000,000 at an ENR Index of 1475 in 1971.  After
application of the September 1978 ENR Index of 2861, the estimated cost
in terms of 1978 dollars was $330,000,000.  The CSS estimated the cost
of point-source treatment at $250,000,000 in 1971.   This estimated cost
is $485,000,000 in terms of 1978 dollars.

     The high projected costs for point-source treatment were based on
construction and operation of chlorine contact tanks sized for a minimum
of 10 min detention at design flow.  Due to the excessive costs and land
requirements associated with conventional chlorination, it was decided
that high-rate disinfection would be investigated.  If successful, high-
rate disinfection (detention times of approximately 1 to 2 min) would
greatly reduce the size of the contact tanks and significantly reduce
the cost of treatment facilities.

     The Onondaga Lake Study was conducted to determine the trophic
status of the lake in terms of physical, chemical, and biological water
quality parameters.  Engineering evaluations were conducted to determine
the effects future pollution abatement facilities would have on the
lake. In addition, a monitoring program was established to provide
continuous updating of the lake water quality data.

     One conclusion of the lake study was that although lake bacterial
concentrations during dry-weather periods usually fall below contact
recreation limits established by the New York State Department of
Environmental Conservation (NYSDEC), the bacterial input of CSO discharges
precludes any guarantee of public safety.

PURPOSE AND SCOPE

     As a result of the previous studies, three major areas of concern
became evident to the responsible government agencies:

     1.   That low dissolved oxygen in the lake would continue
          because of the high organic concentrations discharged to the
          lake, primarily in the effluent from Metro, but also from the
          CSO.

     2.   That bacterial and viral contamination of the lake would
          continue during and for some period after storm events as long
          as the CSO remain untreated.

-------
     3.   That significant addition of nutrients to the lake could
          result from CSO.

     The total organic loading to the lake will be significantly reduced
by the upgrading of Metro to tertiary treatment.  The upgrading was
mandated by the 'New York State Health Department in mid-1969.  Construction
of the required additional treatment facilities was started in February,
1975, under an EPA construction grant.  The upgrading and expansion will
provide for 85 percent removal of biochemical oxygen demand (BODs) and
suspended solids (SS), and for tertiary treatment specifically for
reduction of phosphorous concentrations to less than 1 mg/1 in the plant
effluent (6).  It is expected that upgrading will reduce the average
rate of BODs discharged to the lake at Metro from 8,000 Ib/day (3,600
kg/day) to 3,600 Ib/day (1,600 kg/day). Discharge of SS is expected to
be reduced from 38,000  to 21,400 Ib/day (17,300 to 9,700 kg/day).
These reductions will tend to emphasize the significance of untreated
CSO.

     In 1971, EPA awarded Demonstration Grant No. 11020 HFR (now
S802400) to Onondaga County, in order to provide partial funding for a
demonstration study of various aspects of point-source treatment of CSO.
The broad objectives of the study were as follows:

     1.   To investigate the feasibility of high-rate disinfection
          techniques;

     2.   To investigate removal of nutrients, particularly nitrogen and
          phosphorus;

     3.   To investigate techniques for removal of suspended solids,
          and the impact of solids removal on high-rate disinfection.

     Treatability of CSO, in terms of bacteria, viruses, nutrients, and SS,
was to be demonstrated using commercially available treatment equipment.
Emphasis was placed on efficiency of removal, automatic response and
control during storm events, and definition of operation and maintenance
requirements.

     The activities and results of the studies conducted under 11020
HFR (S802400) are summarized in this report.  In addition, two separate
reports have been prepared under the same grant, as follows:

     1.   Bench-Scale High-Rate Disinfection of Combined Sewer Overflows
          with Chlorine and Chlorine Dioxide, (EPA-670/2-75-021) (7).

     2.   High-Rate Nutrient Removal for Combined Sewer Overflows, (EPA-600/
          2-78-056) (8).

-------
GENERAL
     3.
     4.
                               SECTION 2

                              CONCLUSIONS
High-rate treatment processes for solids reduction and
disinfection are viable and economical alternatives
in CSO abatement and management.  These treatment processes
are adaptable and well suited for point-source applications.

Initial field studies provide important physical and quality
data on CSO flotos and loadings for the selection of treatment
facilities.

Analysis of long term rainfall data provides a good basis of
correlation between projected and monitored overflows.

The Simplified Stormwater Management Model is an excellent
tool for analysis of projected loadings and preliminary
abatement planning.
MICROSCREENING

     1.   Operational problems were encountered during start-up and
          initial performance of prototype screening devices, particu-
          larly as related to the Crane Microstrainer.

     2.   The Sweco unit achieved an average SS mass removal efficiency
          of 48 percent and a concentration removal efficiency of 32
          percent, operating in a range of 4.9 to 61.7 gpm/ft^
          (12 to 150 m/hr).

     3.   The Zurn unit provided an average SS mass removal efficiency
          of 45 percent and a concentration removal efficiency also of
          45 percent, operating in a range of 3.3 to 13.7 gpm/ft2
          (8 to 33 m/hr).

     4.   The Crane unit achieved an average SS mass removal efficiency
          of 58 percent and a concentration removal efficiency of 58
          percent also, operating at loading rates of 1.7 and 7.7 gpm/ft2
          (4.1 and 18.8 m/hr).

     5.   Relationships of SS removal efficiency to hydraulic loading rate
          were developed for each of the units and generally indicated that
          efficiency decreased as hydraulic loading rate increased.

-------
     6.   The Sweco and Zurn units removed proportionately greater amounts
          of organic solids than total SS while the Crane unit removed
          proportionately lesser amounts.

     7.   Significant and/or consistent removal of heavy metals was not
          achieved by any of the screening units.

     8.   The performance data of the screening units in this study cor-
          responded closely with  the reported results of other micro-
          screening studies and noted that better SS removals were
          achieved at smaller screen apertures.  However, sustained
          hydraulic loading rates could not be achieved for the Zurn
          Micromatic and only for one short time interval (5 min) was the
          design hydraulic loading rate of 30 gpm/ft^ (75 m/hr) achieved.
          The Sweco unit appeared capable of achieving its design
          loading rate of 60 gpm/ft2 (150 m/hr) although most evaluations
          were at lower hydraulic loading rates.

SWIRL REGULATOR/CONCENTRATOR

     1.   The swirl regulator/concentrator was found to function satis-
          factorily as a concentrator of CSO solids when operating within
          the design flow range.

     2.   Operational problems with peripheral equipment installed at the
          site (pumps, instrumentation, etc.) prevented a more detailed
          evaluation of the swirl regulator/ concentrator.

     3.   Analysis of the settling velocities for a range of particles found
          at the selected overflow tended to confirm the predicted perfor-
          mance curve determined from previous model studies.

     4.   The SS concentration removal efficiency of the swirl regulator/
          concentrator ranged from 18 to 55 percent and from 33 to 82
          percent in terms of mass removal.

     5.   Regression analysis of SS removal data indicated a slight increase
          in SS concentration removal efficiency with increased foul
          fraction of. the swirl unit.
     6.   The BODs concentration removal efficiency of the swirl regulator/
          concentrator ranged from 29 to 79 percent and from 51 to 82
          percent in terms of mass removal.

     7.   The overall average TOC and VSS concentration removal efficiencies
          were 33 and 34 percent, respectively.

     8.   Significant and/or consistent removal of heavy metals was not
          achieved by the swirl regulator/concentrator.

-------
SOLIDS HANDLING CONSIDERATIONS

     1.   Transmission of CSO treatment facilities concentrated underflows
          (residuals) from the entire Syracuse CSO drainage area to the
          primary treatment facilities of Metro would not result in a
          hydraulic overload for any of the four solids removal devices
          investigated.

     2.   A solids overload at Metro would result from transmission of CSO
          treatment residuals for a 1 year-2 hour storm for all of the
          treatment devices considered.

     3.   The organic loading to the Metro secondary clarifiers would not
          exceed the acceptable limit of 1.5 times the average dry weather
          flow (DWF) loading for the two storms considered upon trans-
          mission of CSO treatment residuals to Metro.

     4.   No hydraulic overload would result from transmission of dilute
          residuals to Metro.

     5.   No solids overload would result for the Zurn, Crane and swirl
          units used as CSO treatment devices when CSO dilute residuals
          are transmitted to Metro.  However, if the Sweco unit
          were utilized, a solids overload would result from a 1 year-2
          hour storm.

     6.   Transmission of CSO treatment residuals directly to the Metro
          sludge handling facilities would result in drastically overloading
          the gravity thickeners hydraulically.

     7.   The total solids loadings from CSO treatment residuals transmitted
          directly to the existing Metro sludge handling facilities
          would be close to the peak recommended solids loading rates
          under average DWF conditions.

DISINFECTION

     1.   Application of high rate disinfection processes can result in
          significant reduction of bacterial populations in CSO.

     2.   Chlorine (C12) dosages of 12 to 24 mg/1 during initial stages of
          overflow were able to achieve 3 to 4 log reductions of fecal
          coliform (FC).

     3.   Cl2 dosages of 12 mg/1 achieved similar log reductions of FC
          after the first 30 to 45 min of the overflow event.

     4.   Chlorine dioxide (C102) dosages of 6 to 12 mg/1 in the initial
          stages of overflow reduced FC levels to 200 counts/100 ml.

     5.   C102 dosages of 4 mg/1 following first flush loadings of overflow
          events maintained 200 counts FC/100 ml level.

-------
     6.    There were limited data to evaluate disinfection by sequential
          addition of Cl2 and C102.

     7.    Regression analyses of disinfection data indicated that removal
          of SS improved reduction of bacterial populations.

     8.    High disinfectant residuals of C102 and Cl2 could be expected in
          the effluent after a contact time of 1 min.

     9.    Chlorination equipment functioned safely and reliably under
          intermittent operations.

    10.    Prototype C102 generation equipment exhibited continuous
          mechanical problems and required close attention during
          operations, but produced results to warrant consideration for
          full development.

CAPITAL  AND OPERATING COSTS

     1.    The capital and operating costs of actual facilities provided
          the basis of cost comparisons between treatment facilities.
          The capital and operating costs developed were:

                    Capital Costs	O&M Costs
Treatment
Device
Sweco
Zurn
Crane
Swirl
Note:
2.
Treatment
Device
Sweco
Zurn
Crane
Swirl
Using
$/Acre
11,120
5,860
3,700
1,950
Acres
MGD x
Actual
annual
jected

C12 Using C102 Using Cl2 Using C102
$/MGD $/Acre $/MGD $/Acre $/MGD $/Acre $/MGD
82,250 11,420 87,500 290 2,220 310 2,350
44,960 6,150 47,210 240 1,830 260 1,950
28,350 4,000 30,610 230 1,780 250 1,910
14,950 2,240 17,200 140 1,090 160 1,210
x 0.4047 = hectares
3875 = cu m/day
capital and operating costs were the basis of projected
costs for treatment of all Syracuse overflows. The pro-
annual costs based on 20-year amortization at 7% were:
Cl2 Disinfection C102 Disinfection
$13,179,000 $13,577,000
8,222,000 8,658,000
6,313,000 6,761,000 .
3,500,000 3,900,000
Note:     Acres x 0.4047 = hectares
          MGD x 3875     = cu m/day

     3.   The swirl regulator/concentrator was found to be a more cost-
          effective alternative then microscreening.

                                      10

-------
     4.   Disinfection by Cl2 was projected to be less costly than C102 as
          indicated in  the data presented above.

     5.   Treatment of CSO by application of point-source high-rate
          application processes would be significantly less expensive than
          would be centralized CSO treatment or point-source treatment
          utilizing conventional application processes.

VIRUS STUDIES

     1.   The Aquella virus concentrator and its system of selective
          adsorption of viruses by filter media are reasonably well suited
          for study of CSO.  However, difficulties were experienced with
          continuous operation,  and it was found that the batch concentra-
          tion capability of the equipment is very limited.  Prospects for
          use of the Aquella concentrator for frequent or continuous
          monitoring of viruses in CSO seem poor.

     2.   The population of wild viruses in CSO is generally at a low level,
          with high sample variation.  It appears questionable whether a
          meaningful measure of disinfection effectiveness can be made on
          the basis of observed reductions in wild viruses found in CSO.

     3.   Seeding of CSO with three indicator organisms (coliphages f2
          and 0X174, and poliovirus Sabin K-l) was found to be a satis-
          factory method for increasing viral populations in CSO to
          reliably measurable levels;  However, phage f2 proved not to be
          a good simulant of enteroviruses in CSO.

     4.   Use of massive doses of Cl2 (24 mg/1 with a Cl2 residual of
          1.0-1.4 mg/1) in a high-rate application resulted in
          virtually complete kill of seeded organisms.

     5.   Screening had no influence on virus inactivation in CSO.
                                                           f
     6.   Simultaneous reductions in bacterial and viral titers were
          common but there was no direct proportional relationship between
          them.

     7.   Sequential addition of disinfectants did not result in increased
          viral kills.

SPECIAL ANALYSIS

     1.   Analyses for total chlorinated hydrocarbon species in disinfected
          and undisinfected CSO indicated values of 1 to 50 yg/1 with respect
          to aldrin.  A variety of compounds  (10 to 20 in number) contributed
          to the total value.

     2.   Containment time in the delivery system of C102 generated on-site
          was determined to significantly affect the strength of the C102
          at the point of injection to the microscreened CSO.

                                      11

-------
     3;   Bench-scale analyses of the interaction of Cl2 and C102 with
          organic species in CSO indicated a lack of correlation of
          chlorinated organics formation versus dosage upon addition of
          either Cl2 or C102 or upon sequential addition of C102 and Cl2-
          However, 5 out of 8 samples indicated significantly increased
          levels of chlorinated organics after addition of Cl2 and/or C102-
          The change in chlorinated organics levels ranged from -2.2 to
          +200 percent.

     4.   Bench-scale testing of the formation of volatile chlorinated
          organics upon addition of 12 mg/1  Cl2 or 8 mg/1  C102 to simulated
          CSO indicated that only low levels of such organics are produced
          when C102 is applied as a disinfectant.  Formation of volatile
          chlorinated organics upon addition of Cl2 and C12/C102 combinations
          is more significant.  Results indicated that tetrachloroethylene
          is  the most significant volatile chlorinated organic produced,  which
          increased by as much as 150 percent at a detention time of 10 min.
          Only a slight increase was observed at a detention time of 1 min.
          The background levels of both chloroform and tetrachloroethylene
          in the simulated CSO were significant (5.1 ug/1  and 8.1 ug/1 >
          respectively).

SPECIAL INVESTIGATIONS OF CSO

     1.   Disinfection by ultraviolet (UV) radiation is feasible and is
          reported to have no residual toxic effects.

     2.   Chlorination at a dosage of 10 mg/1 for one minute had approximately
          the same disinfecting power as a UV lamp radiating at an intensity
          of 5800 iaw/cm2 for one minute.

     3.   The initial rate of bacteria inactivation was greatest when UV
          radiation and chlorination were used simultaneously.

     4.   Chlorination followed by UV radiation produced higher bacteria
          kills than UV radiation followed by chlorine.

     5.   The highest bacteria kills were produced when in contact with
              for a minimum of 45 sec prior to UV radiation.
     6.   No definite advantage was demonstrated for sequential  application
          of UV radiation alone.

     7.   Bench-scale results indicated that a target level  of 2400 counts/
          100 ml of TC bacteria could be achieved by chlorinating with 8 mg/1
          for 40-45 sec and Eradicating with UV for 15-20 sec.

     8.   Cost analysis data indicates that a UV disinfection system would
          be substantially higher than Cl2 and C102 systems.

     9.   Consideration must be given to the turbidity of treated CSO to
          evaluate the effectiveness of UV disinfection.

                                      12

-------
10.    Current determinations of adenosine triphosphate (ATP),  utilizing
      the luciferin-luciferase bioluminescent assay system,  and indicator
      bacteria from CSO have indicated the feasibility of using ATP
      as a reliable and rapid indicator to control the disinfection
      process.  In the dosages required for high-rate disinfection,
      Cl2) C102 and sodium thiosulfate did not significantly interfere
      with the ATP assay.
                             i
11.    Although operational problems were experienced, the concept of an
      on-line TOC monitor for continuous monitoring of TOC in CSO is
      feasible.

12.    Laboratory testing indicated that an on-line SS Monitor for
      continuous measurement of SS concentrations in CSO is feasible.
      The prototype was shown to be insensitive to the color of
      dissolved solids or to the shape and size of SS particles within
      the particle concentration range tested of 10 to 100,000 mg/1.

13.    The size of the SS Monitor transducer probe and associated
      instrumentation allows operation in confined spaces and permits
      portable operation.

14.   Field testing of the SS monitor at the Tulsa, Oklahoma, Mohawk
      Park Sewage Treatment Plant and at the Syracuse demonstration
      facilities indicated good qualitative correspondence to measured
      SS concentrations.
                                 13

-------
                               SECTION 3

                            RECOMMENDATIONS

 1.   Design and installation of demonstration facilities should be preceded
     by intensive field studies.

 2.   Future demonstration programs should include the use of the Simplified
     Stormwater Management Model  for projection of flows and loadings to
     CSO facilities.

 3.   Future development work should be continued by manufacturers of
     microscreening equipment to eliminate or minimize operational difficulties
     such as blinding,  drive malfunction, and screen breakage.

 4.   Consideration should be given to further research on the feasibility
     of microscreening  in the range of solids removal above that attainable
     with the swirl regulator/concentrator.

 5.   Full-scale swirl  units, covering a wide range of CSO hydraulic and solids
     loading conditions, should be evaluated.

 6.   Future swirl units should include provision for hydraulic regulation
     of the foul fraction.

 7.   Although the actual SS removal efficiency of the West Newell Street
     swirl unit appeared to confirm the ,SS removal efficiency as predicted
     from model studies, pilot and prototype scale swirl units should be
     tested side-by-side under actual field conditions to verify the scaleup
     procedures from model to full-scale swirl units.

 8.   In-depth research  studies are required on residual CSO solids and sludges
     to promote solutions for handling and disposal.

 9.   Bleedback of CSO sludges to sewerage systems should be re-evaluated
     for specific abatement alternatives and programs.

10.   Possible deleterious by-products of disinfection by Cl2 and C102
     should be more comprehensively identified in subsequent demonstration
     programs.

11.   Further research should be performed to optimize high-rate dosages
     of Cl2 and C102 for CSO applications.
                                    14

-------
12.   More detailed evaluation of sequential  addition of disinfectants in
     a subsequent demonstration program should be considered.

13.   Consideration should be given to the development of fully operational
     equipment for field generation and feeding of C102-

14.   Site specific factors should be considered in the design  of CSO
     facilities, particularly in regard to pumping requirements.

15.   Viruses found in CSO discharges require further studies on identification
     and quantification.

16.   Further independent research into the mechanisms governing the existence
     of viral  organisms in CSO, and their response  to disinfectants, should
     be conducted.

17.   The formation of chlorinated organics,  especially volatile chlorinated
     organics, and other refractory residuals in high-rate disinfection
     systems using Clg and C102 should be more thoroughly evaluated.

18.   Further research is necessary to develop positive control of injecting
     the generated C10£ into the wastewater.

19.   Full scale studies should be performed on ultraviolet (UV) disinfection.

20.   The development of an instrument to monitor ATP in unattended operation
     is a realistic venture and should be actively pursued, with particular
     attention directed to evaluation of instrument stability  and the cost
     of the enzyme reagents.  ATP monitoring could be applicable for automated
     disinfection.

21.   Further evaluations of on-line TOC monitoring are required to eliminate
     problems resulting from intermittent operation of CSO treatment
     facilities.

22.   Encouragement should be given to further technical development of a
     SS monitor for CSO and other monitoring and control applications.

23.   Future design of swirl regulators should consider the criteria of using
     a design treatment flowrate (Q^) for frequent storm occurrences, e.g.
     6 to 25 per year.  A peripheral relief weir should then be included for
     flows in the range of 1.5 to 2.0 times Qj.
                                      15

-------
                                 SECTION 4

                           EXPERIMENTAL PLAN
PLAN OBJECTIVES
     The specified objectives of the Syracuse CSO demonstration study
are summarized as follows:  (1) to demonstrate the feasibility of mechan-
ical screening for solids removal, (2) to demonstrate the feasibility of
swirl flow regulation/concentration for overflow regulation and solids
removal, (3) to demonstrate the feasibility of high-rate disinfection
for reduction of microbial and viral populations with and without
preliminary screening, and (4) to develop design parameters and cost esti-
mating information for high-rate CSO treatment facilities.

     In addition, certain special considerations were added to the
program objectives during the course of the study.  These peripheral
investigations included either an analysis of related research and
demonstration work which would enhance this and possible future work
on CSO or field testing of specific applications.  The areas covered
under special considerations included analyses, investigations and
instrumentation and are presented in Section 14.

     An experimental plan was devised to provide for a specific, rational
program for the accomplishment of the overall objectives of the demonstra-
tion study.  Considerations involved in this planning, and the major
elements of the plan itself, are outlined in the following subsections.


SOLIDS REMOVAL CONSIDERATIONS

     CSO's frequently contain high concentrations of SS, which when dis-
charged to receiving waters may settle to the bottom or accumulate in
shoals, creating noxious conditions, excess oxygen demands on overlying
waters, and other physiochemical and aesthetically displeasing effects.
The receiving water's capacity to assimilate downstream wastes is reduced,
magnifying other pollution problems downstream from CSO.  It has also been
hypothesized that solids removal may enhance disinfection (7, 9, 10).

     The available techniques for solids removal were considered to be the
following:

          1.   Sedimentation
          2.   Mechanical fine mesh screening (microscreening)
          3.   Swirl concentration

                                     16

-------
     The feasibility of using each of these techniques was considered.

Feasibility of Sedimentation

     Normal sedimentation techniques for solids removal were not consi-
dered practical for the Syracuse CSO demonstration study.  The space
available in the vicinity of the Syracuse CSO outfalls was found to be
inadequate for sedimentation tanks sized to handle anticipated peak
discharge rates.

Feasibility of Mechanical Fine Mesh Screening (Microscreening)

     In general, a microscreen consists of a rotating drum, covered with
finely woven fabrics of stainless steel with a range of aperture'sizes
of approximately 20 to 120 microns.  It is constructed to rotate around
either a horizontal or a vertical axis.  Water enters the inside of the
drum, flows radially through the drum screen into the outlet chamber,
and deposits suspended solids on the inside of the drum screen.  This
mat of deposited solids on the screen increases the resistance to flow
through the unit.  As a result the head differential across the unit
increases.  The screening unit must have some provision for the removal
of the solids mat.  Generally, backwashing (either continuous or inter-
mittent) has been used.

     Glover and Herbert (9) concluded that high-rate fine mesh, mechanical
screening, using screens with apertures in the range of 23 to 35 microns,
followed by high-rate disinfection, is a practical method for CSO treatment.
The process demonstrated in Philadelphia was termed Microstraining,
which is a copyrighted name under Crane Company.  For purposes of this
study, mechanical screening of this general type has been designated
microscreening.

     Maher (10), expanding on previous work by Glover and Herbert, observed
that when a microstrainer having openings of 23 microns was operated at
an influx rate of approximately 26 gpm/ft2 (63.4 m/hr), suspended solids
in CSO were reduced from 50-300 mg/1 to 40-60 mg/1.  Maher concluded,
based on the findings of his study, that the microstrainer could be
operated at differentials of about 24 in. (60 cm) of water, and at an
influx range of 15 to 25 gpm/ft2 (36.0 to 60.0 m/hr), resulting in
effluent suspended solids of 40 mg/1.  Organic matter, in the form of
volatile suspended solids was reduced by about 70 percent.

     An economic study by Keilbaugh, et al. (11), on the cost of various
methods for treating CSO concluded that microstraining followed by high-
rate .disinfection was the least expensive and most compact of several
methods investigated.  Combinations of 12 basic treatment processes were
compared.  Only a surface impounding basin located at the CSO outfall
with pumpage to a sewage treatment plant was determined to be less
expensive for CSO treatment.  However sufficient land area for this
method is often not available in urban areas.  Consideration of the
swirl regulator/concentrator was not included in their economic study.
                                    17

-------
     Bench-scale demonstrations were conducted under the Syracuse study
to test the feasibility of CSO treatment by microscreening followed by
disinfection.  Based on the results of the bench-scale work it was
decided that microscreening would be tested in a full-scale demonstration
study.  Consequently, microscreening was used for solids removal in one
of the two Syracuse full-scale demonstration facilities in order to
determine solids removal efficiencies and subsequent effects on disinfection
processes.

Feasibility of Swirl Regulator/Concentrator
     The hydraulic concept leading to the development of the swirl
concentrator as an overflow regulating device was first presented in the
United States in a report published by EPA in 1970 (12).  This report
presented the results of a study conducted by the American Public Works
Association (APWA) Research Foundation in 1968-1970 on combined sewer
overflow regulator and control facilities found in the United States,
Canada and selected foreign countries.

     The discussion pertinent to the development of the swirl concentrator
was on a device known as the circular "vortex", which had been developed
in the City of Bristol, England.  The "vortex" device was designed as a
regulator, but was also found to effectively separate and concentrate
solids found in CSO.  The report subsequently recommended additional
investigation.

     An opportunity to confirm and supplement the work carried out at
Bristol and to to develop a basis for a similar regulator device suitable for
the different CSO conditions found in the United States was provided by
the plan of the City of Lancaster, Pennsylvania to construct a CSO
regulator facility upstream of a proposed CSO control/treatment facility.

     The APWA was retained by the City of Lancaster to conduct an intensive
study aimed at achieving these goals.  This study was accomplished
through use of laboratory hydraulic modeling and development of a
mathematical model calibrated to provide the best possible match with
experimental model results.  The study was sponsored, in part, by the
Office of Research and Development, EPA, under Demonstration Grant No.
802219 (formerly 11023GSC).  The final study reports were published in
1972, 1973 and 1974 in the EPA Technology Series (13, 14, 15).

     A study finding indicated that an unimpeded free vortex must be
avoided with the large flows and minimum sized chambers associated with
CSO in the United States.  A different hydraulic phenomenon, the swirl,
was determined to effectively separate solids from CSO.  The report
concluded that:

     1.   A practical, simple facility has been developed which offers a
          high degree of performance in reducing the amount of settleable
          solids contained in CSO, as well as enabling the quantity of
          flow to the interceptor to be controlled, all with a minimum
          of moving equipment.


                                     18

-------
      2.   The design of the swirl  concentrator has been developed for
          rapid calculation of its different elements enabling ready
          transferability to the regulation of various quantities of
          flow.

      3.   The swirl  concentrator is very efficient in separating both
          grit and settleable solids in their middle and larger grain
          size range (>0.2mm).   By weight,  these fractions represent
          about two-thirds of the respective materials in the defined
          combined sewage.  Separation of the smaller grain sizes was
          less efficient, although still appreciable.

It recommended that:

      1.   A demonstration facility should be constructed of sufficient
          size to be totally effective for flows of 103 cfs (2.92 m3/sec).
          The facility should be monitored to verify the hydraulic and
          mathematical modeling which was accomplished in the study.

      2.   Additional hydraulic and mathematical modeling should be
          accomplished to determine the effectiveness of the swirl
          concentrator concept in the. various phases of primary sewage
          treatment.  Such research should also have application in many
          industrial waste situations.

      3.   Further investigation should be made to determine if better
          efficiency could be obtained with two or more concentrators
          operated in parallel  or in series.

     The Lancaster hydraulic study demonstrated the potential of the
swirl concentrator as a CSO quality and quantity regulating device.
Under the Syracuse study, it was decided to construct and operate a
prototype demonstration facility based on the data obtained during the
Lancaster study, in order to prove the effectiveness of the swirl for
solids removal during full-scale operation.

DISINFECTION CONSIDERATIONS

     Investigation of disinfection in this demonstration study was based
on the application of chlorine and chlorine dioxide and limited bench-
scale analysis of ultraviolet radiation.

Feasibility of High-Rate Disinfection

     Chlorine  (Cl2) is used extensively for disinfection of sewage.  Chlorin-
ation of sewage can be thought of as a two-stage process:

     1.   Addition of sufficient Cl2 for satisfaction of Cl2 demand.
          In this stage, free Cl2 combines with other material in the
          waste. The major combined forms of Cl2 are chloramines
          NH2C1, mono-; NHC12. di-j and NCI3, tri-), which are formed
          by the reactions of ammonia with free chlorine.  Minor reactions

                                    19

-------
          occur with organic material, particularly with reducing agents.

     2.   After satisfaction of the Cl2 demand, further addition of
          Cl2 results in maintenance of free chlorine, in the form
          of Cl2»  HOC! (hypochlorous acid) or OC1" (hypochlorite ion).

     Free Cl2 is a fast-acting disinfectant.  Cl2 in combined
forms disinfects much more slowly. In standard disinfection of sewage
effluent, addition of sufficient Cl2 to satisfy the Cl2 demand is not
attempted, and it is assumed that only combined Cl2 is present.    To
provide reasonable assurance that disinfection is achieved detention times
of at least 15 minutes have been required by state and federal regulatory
agencies.

     In general, provision of 15 minutes or more of detention time
following addition of Cl2 to CSO was considered infeasible because
of the high flows encountered.  Therefore, chlorination planning for
this study was based on a concept of adding enough Cl2 to overcome
the side reactions quickly and provide free Cl2 for rapid disinfection.
It was hypothesized that Clg added in sufficient amounts to overflows
could reduce bacteria and viruses to acceptable levels with contact
times of two minutes or less.  Bench-scale tests indicated that target
levels of bacteria and viruses could be accomplished within two minutes
in simulated CSO with a Cl2 dosage of 25 mg/1 (7).  The following
target levels were achieved at this dosage rate:

     1.   Total coliform bacteria (TC) - 1,000 colonies per 100 ml.
     2.   Fecal coliform bacteria (FC) -   200 colonies per 100 ml.
     3.   Fecal streptococci bacteria (FS) - 200 colonies per 100 ml.
     4.   Poliovirus, Sabin K-l (Poliovirus 1) - Five log reduction
                                                  in population.
     5.   Coliphage (0X174 and f2) - Five log reduction in population.

     No federal or state regulation specifies target levels of viruses,
but it was assumed that a five log reduction in population would reduce
the highest anticipated viral counts in CSO to essentially zero.  Work
by Crane Company (10) and by Glover and Herbert (9) concluded that it
is practical to use Cl2 for high-rate disinfection of CSO.  The Syracuse
demonstration study, therefore, included facilities for C12 addition and
high-rate disinfection.  The effects of screening, contact times, mixing,
pH, and temperature on disinfection by Cl2 were to be investigated.

     Use of chlorine dioxide (C102) for disinfection is a concept of
relatively recent origin.  It appears that C102 may have certain advan-
vantages over Cl2-  C102  is a good disinfectant at high
pH values, does not combine with ammonia to form chloramines, and
destroys phenolic compounds that combine with other types of Cl2 compounds
to produce chlorinated phenols which yield undesirable taste and odors in
drinking water.

     The lack of a specific test to determine C102 concentra-
tion in sewage has prevented detailed studies of the behavior of C102

                                     20

-------
during the disinfection process.  A technique known as electron
spin resonance was recently developed that may be useful for this purpose
in the future.  Bench-scale studies indicated that C102 on a weight
basis, is about twice as effective as Clz in reducing bacterial and
viral counts to target levels (7).  It is not understood exactly how
Cl2 and C102 act on bacteria and viruses to accomplish disinfection.
However, during the bench-scale work, it appeared that lipid (organic fat)
content in the cell membrane may be a factor.  Cells with relatively
low lipid content in the membrane appeared to be relatively more resistant
to disinfection.  There is some basis to assume that lipids may be more
sensitive to C102 than to Cl2s which might explain the superior disinfection
ability of C102.  The Syracuse demonstration study was to include C102
addition to compare its disinfectant capability with C12 under similar
conditions of screening, contact time, mixing, pH, and temperature.


VIRUS CONSIDERATIONS

     The hazards to public health posed by the presence of water-borne
viruses have been well documented (16, 17, 18).  The principal sources
of these viruses are the alimentary tracts of humans and domestic animals.
A broad spectrum of viruses is contributed whose members have only one
general feature in common -- they are naked, nonenveloped virions for
the most part: picornaviruses, adenoviruses, reoviruses, parvoviruses and
perhaps papovaviruses.  The viruses have relatively long thermal decay
times and are not readily inactivated by osmotic shock resulting from
high dilutions in water, by pH variations normally associated with water
supplies, by adsorption to natural colloids in water or by the action of
other types of microorganisms.  On this basis alone it is not surprising
that animal viruses are transmitted virtually unchanged when domestic
sewage and storm runoff are combined.

     Although there is no evidence to date that any virion is completely
resistant to disinfection by Cl2 and Cl2 compounds, there are
ample data to demonstrate that the method of their application and the
presence of interfering organic substances control the efficiency of this
treatment (19, 20, 21).  Previous research (22) had shown that C102 as
well as Cl£ (as HOC!) reduced the titers of several enteroviruses in
simulated and real CSO.  These attempts to disinfect CSO and to evaluate
the relative merits of both Cl2 and C102 had been limited largely to
bench-scale studies.  A follow-up was required to determine if the same
principles  could be applied to disinfection of CSO on a full scale.


DEVELOPMENT OF EXPERIMENTAL PLAN

     The experimental plan which was developed to satisfy the overall
objectives of the demonstration study was based on use of facilities at
the Maltbie Street overflow to demonstrate screening techniques and use
of facilities at the West Newell Street overflow to demonstrate the swirl
concentrator.  High-rate disinfection was planned at both sites.  The
procedures for selection of sites for demonstration purposes are outlined
in Section 5.

                                   21

-------
Maltbie Street Facilities

Microscreening--

     It was decided that the facilities at Maltbie Street would include
three screening devices, operated in parallel.   The screens would be
commercially available units, of comparable design.  This would allow
evaluation of the relative merits of each under roughly identical loading
and operating conditions.  An economic analysis was made to determine
the feasibility of designing the screening and disinfection facilities
for the peak anticipated overflow of 30 MGD (78.9 cu m/min).  A
reduced design flow of 15 MGD (39.4 cu m/min) was selected due to
physical and financial constraints.   Flow was to be split evenly between
the three units, with each receiving 0 to 5 MGD (0 to 13.2 cu m/min), depend-
ing on overflow quantity.

     The three screening devices to be tested in this program consisted
of two microscreens each rotating on a horizontal axis (a Crane Micro-
strainer and a Zurn Micromatic) and a third microscreen (a Sweco Wastewater
Concentrator) rotating on a vertical axis.  Details of the design of
each device are contained in Section 6.  The screening devices were equipped
with different size screen apertures to determine what effect the removal
of different levels of suspended solids would have on disinfection processes.
The screen aperture sizes were, specifically, 23, 71 and 105 micron stainless
steel mesh.  It was expected that these different screen sizes would
produce different suspended solids levels in the effluents to the disinfec-
tion contact chambers.  All other parameters being equal, an evaluation
of varying solids concentrations on the disinfection process could be
made.

     Evaluations of the screen loading rates were to be investigated to
determine the optimum loading rates in terms of operation, maintenance,
and efficiency of solids removal for each unit, in order to minimize the
capital and operating costs of treatment facilities.

     In addition to varying the screen loading and flowrates, two overflow
events were to be evaluated for reduction of bacterial levels with no
screening prior to disinfection.  This evaluation would further indicate
the relative importance of solids levels on disinfection.

Disinfection - Bacteria --

     The basic objective of the disinfection experiments at the Maltbie
Street facility was to optimize the disinfectant dosages required to
reduce microbial populations to acceptable levels.

     Microscreens were believed to have a significant advantage regarding
disinfection in that microorganisms which would otherwise be protected
from disinfection by larger grease and solids clumps would be more
vulnerable to disinfection in the relatively small clumps passing through
                                   22

-------
the screen.  In addition, it was believed that microscreening would have
the following downstream advantages:

     1    As a result of reducing the particle sizes, the time for
          natural assimilation of oxygen-demanding substances in the
          receiving stream would be shortened.

     2.   Some removal of oxygen-demanding substances would be accom-
          plished, in that some fraction of the organic matter in the
          waste stream would be included in the screenings.

     For the purpose of the demonstration phase of this study, target
levels for the receiving stream were established as follows:  total
col i forms of less than 1,000 counts/100 ml and fecal coliforms of less
than 200 counts/100 ml.  This indicated that acceptable levels in the
discharge could be as follows:

     TCd =     5s x 1000/100 ml
               Qd

     FCd =     fis x 200/100  ml
               Qd

     where:  Qs is stream flow, after CSO discharge
             Qd is CSO discharge
            TCd is total col i form count in Qd
                is fecal col i form count in Qd
     In practice, it was impossible to pace disinfectant dosage using
the Qs/Qd ratio since its value was not known.  In order to be certain
of achieving the target levels in the stream, it was decided that the
acceptable levels in the discharge would be defined as being exactly
equal to the target levels in the stream (i.e. Qs/Qd =1).

     The conclusions of the bench-scale study (7) had indicated that a
Cl2 dosage of 25 mg/1 , a C102 dosage of 12 mg/1 , or a two-stage (sequential)
addition of 2 mg/1 of C102 followed by 8 mg/1 of Cl2 in 15 to 30 sec, would
reduce bacteria counts to the defined acceptable levels in a two-minute
exposure period (7).  These dosages were to be verified or optimized in the
full-scale facility.             .

     It was hypothesized that two-stage disinfection by sequential addition
might enhance disinfection beyond the additive effects of the respective
doses.  This might occur in two ways:  (1)  The first disinfectant
would precondition the waste so that the second disinfectant could work
more efficiently; and (2) interactions between two disinfectants could
lead to improved efficiency of one or both.  Four combinations are
theoretically possible for two-stage chlorination with Cl2'and C102:
Cl2 followed by Cl2» Cl2 followed by C102, C102 followed by C102, and
C102 followed by Cl2-  Bench scale tests showed that disinfection was
enhanced by addition of C102 followed by addition of Cl2 in 15 to 30
seconds.  It was speculated that C102, which is a stronger disinfectant,

                                     23

-------
may be regenerated through interaction of the chlorite ion (C102-) and
Cl2-  Sequential addition of the same disinfectant did not enhance
disinfection beyond the expected additive effects of the chemicals.  Two-
stage disinfection was demonstrated at the Maltbie Street facility only.

     It was also felt that rapid mixing of the disinfectant with the CSO
could result in the achievement of acceptable levels of bacteria at reduced
disinfectant dosages.  Thus, provisions were made in the experimental plan
to investigate various mechanical mixing techniques, including single flash
mixing at the point of injection, sequential flash mixing, no mixing, and a
previously reported technique (9) utilizing corrugated baffles throughout
the length of the disinfection tanks.  Evaluations of the bacterial level
reduction achieved by these mixing techniques versus cost were to be made.

Disinfection - Viruses —

     Although it seems patently obvious that combined storm and domestic
sewage overflows should yield a spectrum of enteric pathogens, both the
qualitative and quantitative nature of these pathogens are largely unknown.
This is particularly true regarding our knowledge of virus pathogens from
humans and animals since the viruses generally are diluted greatly and the
technology for quantitative recovery of viruses from large volumes of water
is relatively new.

     A wide variety of methods for quantitative recovery of viruses have
been studied including precipitation, ion exchange, electroosmosis,
two-phase separations, high speed continuous centrifugation and gel
filtration (22, 23, 24).  However, only high speed filtration methods
combined with selective adsorption appear to have the capacity to remove
viruses selectively from tens or even hundreds of gallons of water (22).
Wall is, et^al-  (25, 26) have developed the most successful system for
this purpose and a commercial version of their instrument has been marketed
under the trademark Aquella (Carborundum Co., Buffalo, N.Y.).  Numerous
published accounts of the development and performance of this instrument
have appeared recently (21, 22, 25, 26, 27, 28, 29, 30).

     The Aquella virus concentrator has been applied successfully to tap
water (25, 27, 28), sewage (22), estuarine waters (30, 31) and sea water
(30, 31).  Both wild viruses and virus seed have been recovered with
efficiencies approaching 77 to 100 percent when infective units were less
than one per gallon (27, 28).

     Thus the Aquella concentrator appeared to be the best available
choice for quantitating viruses in CSO.  A recent paper (32) revealed
that BGM, a continuous line of African green monkey kidney, could detect
enteroviruses in storm sewage (and incidentally, more efficiently than
primary African green MKC or rhesus MKC).  Since a study was underway
to determine the effectiveness of C102 as a viricide for CSO it was
decided to attempt the use of the Aquella to measure the number of wild
viruses which survive C102 treatment of storm overflows. An effort was
also made to use phage f2 as a virus indicator which would mimic enter-


                                    24

-------
oviruses arid could serve as an internal control for the concentration
of viruses, measuring the recovery efficiency.

     The primary objectives in the first phase of the virus investigations
were: (1) to determine the feasibility of using the Aquella virus con-
centrator to detect virus pathogens in CSO; (2) to determine levels of
virus pathogens in CSO at the Onondaga County demonstration facilities
on Maltbie Street and West Newell Street, Syracuse, N.Y.; (3) to
investigate the use of bacteriophage indicators as standards for Aquella
performance and efficiency; and (4) to systematize the survey methods
for animal viruses in CSO.

     A second phase of the project was conducted at the Maltbie Street
treatment facility during the summer and fall of 1976 with the following
objectives: (1) to compare Cl2 and C102 for their relative efficiency in
disinfecting CSO seeded with indicator viruses; (2) to determine if a
combined application of Cl2 and C102 provided enhanced activity; (3) to
determine if the treatment reduced the wild type enteroviruses naturally
occurring in CSO; and (4) to study the correlation between reduced seed
virus titers and reduced enterobacterial counts.

     Results of these virus investigations are presented in Section 13.

West Newell Street Facilities

     At the West Newell Street site, experiments on treatment of CSO
were to be investigated with the use of the swirl regulator/
solids concentrator with pre-and post-disinfection by Cl2 and C102-

Swirl Regulator/Solids Concentrator —

     The swirl regulator/concentrator operates with relatively little
mechanical equipment, thus providing a marked contrast in operating and
maintenance expenses when compared to other primary treatment devices.
During an overflow conditions swirling action produced by the momentum
of flow into the swirl chamber and the geometry of the tank causes solid
particles suspended in the flow to move toward the outside of the swirl
chamber and toward the bottom, where they are removed through a foul
sewer line.  Approximately 20 overflow events were to be investigated
for removal of SS and BOD5 by swirl concentration.

     A series of reports from 1967 through 1972 (12,  13, 33, 34, 35,  36)
outlined considerations and developed parameters for design of the swirl
regulator/concentrator.  In general, these reports advanced two design
approaches.  The first approach was developed from the results of
hydraulic model studies conducted by the LaSalle Hydraulic Laboratory
Ltd. (Laboratoire d'Hydraulique), LaSalle, Quebec, Canada.  The second
design approach was the product of mathematical modeling calibrated with
experimental data from the LaSalle model study.  The mathematical model
was developed by the Re-Entry and Environmental Systems Division of
General Electric Company.
                                     25

-------
     The physical model design approach used anticipated flow to the
swirl chamber as the basis for deriving design dimensions.  The mathema-
tical model used solids concentrations, specific gravity of solids, and
particle size distribution to determine design dimensions.

     Although some disparity existed between the design dimensions
obtained from the two approaches, it was considered to be outside the
scope of this study to evaluate their relative merits, particularly
since neither approach had been previously proven on the prototype
scale.  The LaSalle model approach (13) was chosen to provide the design
dimensions for the swirl chamber at the West Newell Street facility.

Disinfection —

     The basic objective of the disinfection experiments at the West
Newell Street facility was to develop data on required dosages after
treatment by the swirl regulator/concentrator.  Provisions were made for
injection of disinfectant to that portion of flow spilling over the
weir (post-disinfection) as well as to the total flow entering the swirl
chamber (prerdisinfection), to take advantage of the mixing action inherent
in the swirl.  During the initial operation of the facility, an Englehard
Chloropac sodium hypochlorite generating system (Englehard Industries
Division, Englehard Minerals and Chemicals Corporation, East Newark,
N.J.) was to be installed.  The system was to utilize an existing brine
supply as its influent, and through the use of Englehard1s electrolytic
process, sodium hypochlorite was to be generated, stored on site, and
fed to the swirl concentrator effluent during overflow conditions.  An
evaluation of the use of sodium hypochlorite in the two-stage disinfection
process was to be made.

     C102 was to be generated on-site by means of a Nitrosyl Chloride
 generation system (U.S. Patent 3754079, Chemical Generators, Inc.,
 Rochester, N.Y.).

     Various combinations of pre- and post-addition of disinfectants
were to be investigated.  Table 1 indicates the various combinations
that were proposed for field application.

	TABLE 1.  WEST NEWELL STREET DISINFECTION EVALUATION SCHEDULE	


                    C102     (mg/1)          Cl2       (mg/1)
     Trial	Pre	Post	Pre	Post	

       1                        12
       2            15
       3            15           2
       4                                                25
       5                                      8         20
       6                         48
                                    26

-------
                               SECTION 5

                       PRE-CONSTRUCTION STUDIES
PRELIMINARY MONITORING

General

     The activities involved in evaluations of high-rate treatment of
CSO in Syracuse have been divided into pre-construction and post-construction
studies.  The pre-construction studies included a preliminary monitoring
program and bench-scale investigations.

     In order to establish a rational basis for designing full-scale
demonstration facilities, a one-year sampling and monitoring program was
conducted at three selected overflow sites.  Selection was based on size
of outfall and associated drainage area, accessibility for sampling, and
availability of sufficient land to accommodate the required monitoring
structures.  The sites selected were the overflows at Maltbie Street,
Rowland Street and West Newell Street.

     A monitoring program was established at these sites to collect
background data on water quality characteristics, flowrates and volumes,
and rainfall intensities.  Figure 2 shows a typical overflow event being
monitored at Maltbie Street.  The water quality parameters that were
determined during the preliminary monitoring program are presented in
Table 2.
     The geometric mean has been used in the statistical data analysis
presented in this section in preference to the arithmetic mean.  It is
felt that for most design purposes, it is more appropriate to design on
the basis of modal conditions, since by definition these conditions will
be most frequently encountered in actual operation.  Most natural phenomena
are zero-limited and have a right-skewed distribution.  For this type of
distribution, the geometric mean is closer to the'modal value than the
arithmetic mean.  In addition, 'outliers' which occur in the measurement
of natural phenomena tend to occur to the right of the median value,
and therefore further emphasize the right-skewness of the distribution.
This effect is reduced by use of the geometric mean, since the geometric
mean is always to the left of the arithmetic mean, and is closer to the
mode than the arithmetic mean.
                                     27

-------
           TABLE 2.  PRELIMINARY MONITORING - WATER QUALITY PARAMETERS

                                   Parameters
Flow
Rain Intensity
PH
Total Coliform (TC)
Fecal Coliform (FC)
Fecal Streptococcus (FS)
Adenosine Triphosphate (ATP)
Alkalinity (T-Alk)
Total Inorganic Phosphorus (TIP)
Chemical Oxygen Demand (COD)
Biochemical Oxygen Demand (BODs)
Total Organic Carbon (TOC)
Oil and Grease (O&G)
Organic Nitrogen (Org N)
 Ammonia Nitrogen
 Nitrates (N03)
 Nitrites (N02J
 Settleable Solids (Sett-S)
 Total Solids  (TS)
 Volatile Solids (VS)
 Suspended Solids (TSS)
 Volatile Suspended Solids (VSS)
 Dissolved Solids (TDS)
 Chlorides (Cl)
 Calcium (Ca)
 Magnesium (Mg)
 Sodium (Na)
. Potassium (K)
           FIGURE  2.  Typical  Overflow  Event  Being  Monitored at
                     Maltbie  Street
                                       28

-------
     The geometric mean is defined as the nth root of the product of n
values.  The spread factor is defined as the antilog of the standard
deviation of the logs of the n values.  Expressed mathematically, these
definitions are as follows:

          Geometric mean (G) =      (Xi...Xn)

                                           n p               f] h
          Spread factor = antilog          Z [(log x-j - log G) J

                                                 fPi

These parameters, for a right-skewed distribution, are comparable to the
arithmetic mean and standard deviation  of a normal distribution.

     Subsequently, Maltbie Street was selected as the site for demonstration
of rotary screening for solids removal and high-rate disinfection of bacteria
and viruses, and West Newell Street was chosen for demonstration of a
swirl regulator/concentrator for solids removal and high-rate disinfection
of bacteria.

Summary of Results

     During the preliminary phase of the project, the quality and quantity
of overflows at Maltbie Street, West Newell Street and Rowland Street
were monitored.  Tables 3 to 5 give the results of a statistical analysis
performed on the data gathered at each of the sites.  The drainage area
tributary to the Maltbie Street overflow consists of approximately 115
acres (47 ha).   Predominant land use is by commercial and light industrial
establishments.  The estimated population of the area is 1,350, based
on the 1970 census.  In addition, dry weather flow is contributed by
industrial establishments with a total estimated population equivalent
of 4,500.  The combined sewer in West Newell Street drains about 54
acres (22 ha) of medium-density residential area.  The major sources of
stormwater are runoff from roof drains and overland runoff.  The estimated
population of the tributary area is 1,200.  In addition, dry-weather flow
is contributed by commercial establishments with a total estimated
population equivalent of 300.  The drainage area tributary to the Rowland
Street overflow consists of 125 acres (50 ha), primarily medium-density
residential.  The population of the area is approximately 2,800.  Statistical
analysis of data collected from Onondaga Creek at Dorwin Avenue (upstream
of the overflows) and at Spencer Street (downstream of most overflows)
is shown in Tables 6 and 7.

     As can be seen in Tables 3 to 5 the numbers of total coliforms
discharged during periods of wet weather were much higher from the
residential land use areas of West Newell and Rowland Streets than from
the commercial/industrial area of Maltbie Street, by a factor of two to
five times.  Fecal coliform (FC) counts from the residential areas were
three to three and one-half times as high as from the commercial/industrial
area.  Fecal streptococcus (FS) counts were inconclusive, with values
                                     29

-------
       TABLE 3.   COMBINED SEWER OVERFLOW QUALITY  -  MALTBIE  STREET OVERFLOW
Geometric Spread Upper Confidence
Parameter No
Overflow Volume,
MG
Total Col i form
count/100 ml
Fecal Col i form
count/100 ml
Fecal Strep
count/ 100 ml
BOD5, mg/1
TOC, mg/1
SS, mg/1
VSS, mg/1
TKN, mg/1
NHaN, mg/1
OrgN, mg/1
N02N, mg/1
N03NS mg/1
T-IP, mg/1
COD, mg/1
TDS, mg/1
Cl, mg/1
T-Alk, mg/1
. Points Mean
9
189
166
181
151
224
173
113
226
228
261
263
259
262
258
203
260
258
0.26x106
6. 38x1 05
1.25xl05
3. 76x1 O4
27
26
159
40
3.09
0.72
2.03
0.07
0.40
0.56
34
289
42
100
Factor
7.28
7.36
10.98
3.60
3.60
2.45
2.40
3.09
1.88
2.16
2.12
2.21
3.22
3.02
2.45
1.95
2.14
2.06
95.0 Percent
6.81x106
0.1 7x1 08
0.64x10?
3. 09x1 O5
226
112
672
255
8.70
2.58
6.98
/
0.27
2.77
3.46
'150
868
145
328
Level
99.5 Percent
43.58x106
l.lOxlO8
6. 05x1 O7
10. 26x1 O5
747
259
1526
733
15.67
5.30
14.08
0.57
8.27
9.72
347
1622
296
645

Samples obtained during 15 storm periods.
                                      30

-------
   TABLE 4.   COMBINED SEHER  OVERFLOW  QUALITY - WEST  NEHELL STREET OVERFLOW
Parameter No.
Overflow Volume
Total Coliform
count/100 ml
Fecal Coliform
count/100 ml
Fecal Strep
count/100 ml
BODs, mg/1
TOC, mg/1
SS, mg/1
VSS, mg/1
TKN, mg/1
NHsN, mg/1
OrgN, mg/1
N02N, mg/1
NOsN, mg/1
T-IP, mg/1
COD, mg/1
IDS, mg/1
Cl , mg/1
T-Alk, mg/1
Points
10
117
94
101
129
172
154
152
171
171
189
188
161
177
190
134
190
190
Geometric
Mean
2.00xl04
13. 39x1 05
4. 67x1 O5
1.44xl04
59
69
87
48
• 10.24
4.91
3.55
0.06
0.20
1.51
85
708
81
197
Spread
Factor
6.45
54.89
27.60
15.71
1.87
1.65
2.88
2.85
1.78
1.71
3.07
2.68
3.63
2.48
2.86
2.53
4.28
2.86
Upper Confidence
95.0 Percent 99.
4.29xl06 2
9. 74x1 O8 411
10. 96x1 O7 242
13.35xl05 173
166
157
498
269
26.56
11.90
22.53
0.31
1.66
6.74
498
3261
883
1108
Level
5 Percent
.45x106
.92x108
.76xl07
.37xl05
297
251
1342
718
45.65
19.68
64.36
0.78
5.56
15.76
1278
7770
3443
2959

Samples obtained during 10 storm periods.
                                      31

-------
       TABLE 5.   COMBINED SEUER OVERFLOW QUALITY  -  ROWLAND STREET SITE
Parameter No.
Overflow Volume,
M6
Total Col 1 form
count/100 ml
Fecal Coli form
count/100 ml
Fecal Strep
count/100 ml
8005, mg/1
TOC, mg/1
SS, mg/1
VSS, mg/1
TKN, mg/1
NH3N, mg/1
OrgN, mg/1
NO£N, mg/1
NOsN, mg/1
T-IP, mg/1
COD, mg/1
TDS, mg/1
Cl, mg/1
T-Alk, mg/1
Points
16
264
249
243
242
319
274
265
312
312
348
352
338
350
350
253
346
345
Geometric
Mean
248,000
34.02x105
4. 46x1 O5
8. 23x1 O4
29
28
80
38
4.50
2.33
1.53
0.07
0,35
0.79
32
326
59
117
Spread
Factor
3.22
10.57
13.74
4.06
2.05
2.12
2.79
2.78
1.97
2.31
2.71
2.35
2.52
3.09
2.17
2.00
2.90
1.92
Upper Confidence Level
95.0 Percent
1.70xl06
1.65xl08
3. 32x1 O7
8. 27x1 O5
96
95
431
202
13.73
9.20
7.88
0.27
1.58
5.04
114
1018
341
342
99.5 Percent
5.07xl06
14. 94x1 O8
38.49x107
30.69x105
188
191
1126
526
25.87
20,11
20.04
0.61
3.73
14.44
235
1944
921
628

Samples obtained during 18 storm periods.
                                     32

-------
for residential areas both less than and greater than the recorded
values for the commercial/industrial area.

     The West Newell Street overflow contained the highest concentrations
of 8005, TOC, TKN and NHsN as indicated in Table 4.   The geometric mean
BOD5 a"t West Newell Street was approximately twice that for either
Rowland Street or Maltbie Street.  Similar comparisons of TOC values
show West Newell Street to be twice as high in TOC as either Maltbie or.
Rowland Streets.  These results tend to indicate that the CSO at West
Newell Street is much higher in organic matter than either of the other
two sites monitored.  This conclusion is further supported by examination
of the TKN and NHsN data where, again, the geometric mean is significantly
higher at West Newell Street than at either Maltbie Street or Rowland
Street.  The relatively high values found for the West Newell Street
overflow may be due to a greater rate of dry-weather organic deposition
in the West Newell Street tributary area, which has relatively flat
sewers.

     The volatile suspended solids (VSS) at the three overflows were roughly
equivalent.  However, the level of fixed suspended solids (SS-VSS) at
Maltbie Street was approximately three times that at either West Newell Street
or Rowland Street, indicating the presence of relatively large amounts
of grit in the Maltbie Street overflow.

     In general, grit particles can be expected to settle out in trans-
porting conduits more readily than lighter organic material.  In a
commerical/industrial section such as the Maltbie Street area, relatively
greater amounts of grit would be expected to enter the sewer system and
settle out.  When a storm occurs, the first flush picks up these inert
materials and transports them to the outfall, resulting in a relatively
higher solids loading.  In contrast, in residential sections, such as
the West Newell Street and Rowland Street areas, lesser amounts of
gritty material are deposited in the transporting conduits during dry
weather.  This effect is seen in a comparison of the ratio of fixed suspended
solids to total suspended solids for Maltbie Street (0.75), vs. West
Newell Street (0.45) or Rowland Street (0.53).

     Tables 6 and 7 demonstrate that there is considerable deterioration
of the quality of Onondaga Creek as it passes through the City of Syracuse.
Substantial increases are seen in BODs, TOC, SS, VSS, TKN, Org-N, and
chlorides.  Relatively insignificant changes are seen in nitrites, nitrates,
TIP, COD, and alkalinity.  CSO are unquestionably major sources of
contamination to the creek and lake, although the total contribution
cannot be known without comprehensive monitoring of the overflows.

     Table 8 lists the quantities of pollutants discharged from the
CSO at Maltbie Street, West Newell Street and Rowland Street, in terms
of pounds per acre-inch (kg/ha-cm) of runoff.  Quantitatively, Maltbie
Street discharged the greater quantity, 268,000 gal (1,020 m3) per overflow
event.  Rowland Street discharged a mean of 248,000 gal (940 m3) per
event and West Newell Street discharged a mean of 20,000 gal (75 m3) per
event.  These overflow quantities are consistent with the rainfall data

                                      33

-------
TABLE 6.  WET-HEATHER ONONDAGA CREEK QUALITY UPSTREAM OF CSO's AT DORWIN AVENUE
Parameter
Total Col i form
Count/100 ml
Fecal Col i form
count/ 100 ml
6005, mg/1
TOC, mg/1
SS, mg/1
VSS, mg/1
TKN, mg/1
NHsN, mg/1
OrgN, mg/1
N02N, mg/1
NOsN, mg/1
T-IP, mg/1
COD, mg/1
Cl, mg/1
T-Alk, mg/1
No. Points
10
10
2
111
88
84
89
89
79
112
112
98
110
112
112
Geometric
Mean
1.9xl05
2.5xl03
24
11
82
15
6.74
0.27
0.46
0.03
0.50
0.18
7.8
47
205
Spread
Factor
15.16
13.02
-
1.54
1.80
2.04
1.63
1.69
1.99
1.49
1.89
3.25
2.28
1.54
'l.!8
Upper Confidence Level
95.0 Percent
1.64xl07
1.70xl05
-
22.3
217
47
1.65
0.65
1.42
0.05
1.41
1.28
30
95
269
99.5 Percent
2.11xl08
1.88xl06
-
33.4
377
93
2.61
1.06
2.70
0.07
2.56
3.85
65
142
314

Samples obtained during 5 storm periods.
                                      34

-------
        TABLE 7.   WET-WEATHER ONONDAGA CREEK QUALITY  DOWNSTREAM OF  CSO'S
                                AT SPENCER STREET

Parameter
Total Col i form
Count/100 ml
Fecal Col i form
count/ 100 ml
BOD5, mg/1
TOC, mg/1
SS, mg/1
VSS, mg/1
TKN, mg/1
NH3N, mg/1
OrgN, mg/1
N02N, mg/1
N03N, mg/1
T-IP, mg/1 -
COD, mg/1
Cl , mg/1
T-Alk, mg/1
No. Points
11
2
2
147
121
116
125
124
116
147
147
140
147
147
147
Geometric
Mean
4.97xl05
3.4x104
128
16.4
144
21
1.08
0.40
0.64
0.04
0.36
0.20
9.6
101
189
Spread
Factor
2.74
127.95
-
1.58
2.23
2.56
1.48
1.72
1.94
1.89
1.45
2.20
2.41
1.68
1.24
jjpper Confidence Level
95.0 Percent
2.61xl06
9.94xl07
-
34.7
538
100
2.07
0.97
1.90
0.10
0.66
0.74
41
237
269
99.5 Percent
6.70x106
9.28xl09
-
53.2
1138
242
2.99
1.60
3.53
0.19
0.94
1.55
93
384
328

Samples obtained during 4 storm periods.
                                      35

-------
                                 TABLE  8.   QUANTITIES  OF CSO POLLUTANTS DISCHARGED
oo
CTl

Parameter
BOD5
TOC
SS
VSS
TKN
NHsN
Maltbie
Geometric*
Mean
6.22
5.79
36.00
9.06
0.70
0.16
Street
Spread
Factor
3.60
2.45
2.40
3.09
1.88
2.16
West Newell
Geometric*
Mean
- 13.36
15.63
19.71
10.87
2.32
1.11
Street
Spread
Factor
1.87
1.65
2.88
2.85
1.78
1.71
Row! and
Geometric*
Mean
6.57
6.34
18.12
8.61
1.02
0.53
Street
Spread
Factor
2.05
2.12
2.79
2.78
1.97
2.31

         *Expressed  in  Ibs/ac-in.

         Conversion:  1 lb/ac~in.= 0.441  kg/ha-cm

-------
obtained for the storm events at each site.  The average rainfall intensity
was approximately 0.20 in./hr (0.51 cm/hr) with an average peak intensity
of 1.00 in./hr. (2.5 cm/hr).  At West Newell Street the average rainfall
intensity for monitored storms was approximately 0.07 in./hr. (0.18 cm/hr)
with an average peak intensity of 0.29 in./hr. (0.74 cm/hr).  At Rowland
Street the average rainfall intensity was 0.10 in./hr (0.25 cm/hr) with an
average peak intensity for all monitored storms of 0.56 in./hr (1.42 cm/hr).
The difference in rainfall intensities between sites considered together
with the effective runoff areas (total rainoff area x runoff coefficient) of
63 acres (25.5 ha) at Maltbie Street, 53 acres (21.4 ha) at Rowland Street,
and 18 acres (7.3 ha) at West Newell Street account for the Maltbie Street
overflow discharging the largest volume of overflow event and West Newell
Street discharging the least volume of overflow/event.

     The West Newell Street overflow had the higher contaminant levels in
terms of pounds per acre-inch as indicated in Table 8.  As noted earlier,
the higher contaminant levels are believed to be due to a greater rate of
dry-weather organic deposition in the relatively flat trunk sewer serving
the West Newell Street tributary area.

     In the analysis of selecting sites for full scale demonstration of
the swirl regulator/concentrator, the Rowland Street site was not
considered feasible based on operational expenses, cost estimates and
collected data.  Therefore the West Newell Street site was chosen for demon-
stration of the swirl unit.

     Subsequent to this demonstration study, a comprehensive CSO monitoring
program was conducted in Onondaga County to determine the variability in
quality and quantity of the CSO from individual drainage areas in the
combined sewer system.  For informational purposes, only, Tables 8a and 8b
present summary data for selected parameters as obtained in the monitoring
program, details of which are available in the referenced report.

BENCH SCALE STUDIES

     A detailed literature search revealed that very little data was
available to establish a basis for design and operation of prototype
high-rate treatment facilities.  A series of bench-scale studies were
conducted prior to preliminary design in order to obtain this information.
The major conclusions and recommendations of the bench-scale studies are
summarized in this section.  Detailed information about the bench-scale
studies is presented in a separate report (7).

     The bench-scale study resulted in several tentative conclusions,
contingent on verification in the full-scale demonstration of prototype
treatment facilities at Syracuse.  The degrees of mixing that occur in
full-scale facilities could not be simulated on a bench scale.  Complete
mixing was therefore used in the bench-scale study, and the conclusions
of that study are based on complete mixing.  The following conclusions
were made.
                                     37

-------
TABLE 8a.  SUMMARY OF QUALITY DATA FOR SELECTED PARAMETERS ON A SITE-BY-SITE
           BASIS IN ONONDA6A COUNTY +


SITE NO.
003
004
014
015
020
021
022
025
026
027
028
029
030
031
033
034
035
036
037
039
040
042
043
044
046
051
052
058
059
060
063
073
074
076
077
080

SS
573
283
262
148
125
224
564
13
179
392
4
177
367
664
544
103
134
563
53
139
47
189
443
333
216
472
442
316
-
197
633
198
188
1,272
704
651
Geometr'
VSS
207
67
93
40
62
82
133 .
3
103
247
1
59
152
67
110
61
84
283
47
76
'23
99
153 -
141
92
119
156
66
-
65
105
86
40
178
202
58
ic Mean c f Selected Parameters
BOD 5
104
43
53
10
97
52
45
110
44
116
-
53
91
219
44
23
92
130
18
60
50
74
87
51
118
38
55
12
-
56
48
35
23
18
69
45
TKN
2.74
1.20
2.28
2.27
3.21
1.67
0.35
21.70
1.08
3.45
-
1.18
5.17
2.16
5.97
1.62
-
5.95
1.38
2.74
9.11
5.86
5.63
2.27
3.30
2.20
3.38
0.11
-
2.06
3.46
0.70
1.22
0.44
2.62
0.84
TIP
0.32
0.14
0.27
0.17
0.44
0.16
0.07
1.60
0.18
0.56
-
0.15
0.64
0.15
0.29
0.23
0.31
0.44
0.50
0.36
1.07
0.46
0.40
0.23
0.28
0.26
0.66
0.07
-
0.29
0.21
0.16
o.n
0.07
0.31
0.13
FC
1,548,760
3,108,610
1,583,510
1,950,460
2,226,620
1,078,920
77,549
32,011
594,155
-
10,000
626,230
1,284,940
-
-
4,157,600
999,987
1,201,530
56,322
1,271,180
999,976
4,795,000
1,845,580
2,632,920
1,325,280
1,228,130
1,437,630
585
-
622,433
80,738
222,199
6,830,100
92,507
1,009,560
116,436

+  Adapted from:  Progress Report, Combined Sewer Overflow Abatement Program,
   Department of Drainage and Sanitation, Onondaga County, New York.  O'Brien
   and Gere Engineers. 1978.
                                      38

-------
TABLE  8b. ILLUSTRATION OF "FIRST-FLUSH" EFFECTS FROM ONONDAGA COUNTY CSO+

Site
003
004
014
020
021
022
026
027
029
030
031
034
036
039
042
043
044
052
058
060
063
073
074
076
080

0-30*
819
630
580
160
343
520
163
527
223
498
515
307
546
170
261
566
369
795
316
500
919
228
313
1767
599
SS
30-60*
523
210
333
68
146
1370
133
389
207
420
-
259
484
186
473
295
447
534
-
116
450
357
148
-
933

60-90*
214
99
179
-
52
-
220
215
83
413
-
53
456
180
192
318
276
330
_
59
391
207
104
-
814

0-30
195
102
142
_
68
38
56
171
76
135
219
129
128
75
74
109
134
151
12
91
63
39
26
23
47
BOD5
30-60
97
25
62
-
13
31
36
105 '
54
125
.
54
181
63
162
67
42
46
_
42
27
33
28
4
39

60-90
50
19
43
_
17
-
37
71
21
90
-
15
88
57
80
49
33
40
_
29
27
20
17
-
55

0-30
4.83
1.27
4.00
2.95
1.84
0.32
1.08
4.49
1.40
6.88
2.16
2.62
5.35
3.59
6.19
5.02
3.40
5.85
0.11
3.42
5.17
0.81
1.02
0.79
0.81
TKH
30-60
1.58
0.76
1.88
1.74
0.60
0.28
1.02
2.78
1.16
5.89
-
1.40
6.39
3.74
6.80
7.63
2.36
2.24
.
1.68
3.22
0.82
1.34
0.10
0.89

60-90
1.47
0.67
2.03
-
1.51
-
1.27
2.90 •
0.98
3.70
-
1.41
4.65
2.97
5.11
5.03
2.37
2.57
-
0.98
1.67
0.65
1.46
-
0.97

0-30
1,322,022
3,346,813
3,474,542
-
1,424,458
67 ,-954
592,682
-
717,863
1,326,850
26,671
8,825,250
936,215
970,900
5,118,182
2,003,141
3,576,146
4,686,115
585
951,358
79,015
228,876
8,843,971
302,571
90,938
FC
30-60
2,773,920
-
963,126
-
80,000
2,106,644
590,591
.
939,065
2,217,945
-
2,800,000
7,955,378
1,381,639
4,411,049
1,620,059
2,135,924
3,500,000
-
656,175
119,642
644,000
5,948,316
10,000
238,831

60-90
2,699,414
2,778,559
1,958,384
_
189,228
-
590,055
-
491,224
1,285,137
-
2,234,643
4,802,279
1,063,662
3,692,996
924,979
4,097,124
3,600,000
-
277,346
55,807
303,926
4,370,282
-
50,043
* 0-30  = First 30 minutes of overflow
 30-60  = Second 30 minutes of overflow
 60-90  = Third 30 minutes of overflow
Adapted from:  Progress Report, Combined Sewer Overflow
Abatement Program, Department of Drainage and Sanitation,
Onondaga County, New York.  O'Brien & Gere Engineers. 1978.

-------
1.   Total coHform (TC) bacteria in simulated combined sewer
     overflows (SCSO) were reduced to the target levels of 1,000
     colonies per 100 ml in two minutes by the following disinfectant
     dosages:

     a.   25 mg/1 Cl2 (only 50 percent of all trials)

     b.   12 mg/1 C102

     These same conditions reduced FC and FS bacteria to 200 colonies
     per 100 ml in two minutes.

2.   These same conditions also achieved five log reductions in
     poliovirus-1 and 0X174 coliphage.  Although target levels of
     viruses are not specified as part of federal or state water
     quality effluent criteria, five log reductions in virus populations
     would reduce the highest anticipated viral counts in actual
     overflows essentially to zero.

3.   High-rate treatment by microscreening, followed by disin-
     fection,  is a feasible method of reducing microbial con-
     tamination of CSO to an acceptable level.

4.   The enhanced disinfection by using two-stage (sequential)
     addition of C102 followed by Cl2  in 15 to 30 sec may be
     due to the regeneration of C102 through the interaction of
     chlorite ion (CIO;?) and Cl2.

5.   There is no enhancement of disinfection beyond the expected
     additive effects when sequential addition of the same disin-
     fectant is practiced.

6.   On a weight basis, C102 is approximately twice as effective as
     Cl2 in reducing bacterial and viral populations to target
     levels.

7.   In the disinfection of contaminated waters with Cl2j the
     initial rapid disinfection is accomplished by free Cl^ which
     is converted to the less potent combined Cl2 species in one to
     two minutes.  C102 is converted to the less potent ClOjJ in
     the same time period.

8.   Microscreening had no measurable effect on high-rate disin-
     fection with Cl2 and only a slight positive effect with C102.
     A possible explanation is that the increased rate of reaction
     between disinfectant and demands that resulted from the shredding
     of parti culates upon screening offset the increased numbers of
     exposed bacteria to yield no net increase in disinfection.

9.   Microscreening alone decreased SS, but in some cases increased
           and bacteria counts.
                                4.0

-------
10.   An advantage of microscreening is that only the smaller SS par-
      ticles will  pass through fine mesh screens thereby increasing
      the disinfectant penetration potential.

11.   Cl2 and C102 demands can be attributed to different substances
      in wastewaters.
12.   Within the dosages required for acceptable disinfection,
      and C10£ do not measurably change pH, TOC, BODs, COD, TKN or
      NH3N.

13.   The temperature variations associated with the northeastern
      climate had only a slight positive effect on disinfection of
      wastewaters with Cl2 and C102-  This deviation from the large,
      positive temperature effects observed in no-demand waters is
      most likely due to a wide variety of competing chemical reactions
      that occur in wastewaters.

14.   Microorganism aftergrowth was not observed to be a significant
      factor in this study.  However, the results may be more a
      reflection of the difficulties in simulating the conditions
      for aftergrowth than aftergrowth itself.

15.   Significant decreases in adenosine triphosphate (ATP) concentration
      that parallel bacterial reductions have been observed during
      the disinfection process.  The results of ATP measurements
      point to the potential of using this indicator parameter as an
      effective means of measuring bacterial concentration after
      disinfection or controlling disinfectant dosages.

16.   The effects of screening and temperature upon disinfection
      were difficult to observe because they were of similar magnitude
      as the variations in duplicate trials.

17.   The order of resistance of bacteria to disinfection with
      and/or C102 is FS >  TC >  FC.

 As a result of the bench- scale studies, the following recommendations
 were made.
  1.  The results of the operation of the full-scale demonstration
      units be correlated with the bench-scale results to determine
      the validity of using these bench-scale studies for certain
      design parameters.

  2.  The role of SS in disinfection of CSO be evaluated
      on a full-scale to determine the effect of screening on the
      disinfection process.
                                  41

-------
      3.  The comparison of the effects of disinfection between the
          swirl regulator/concentrator and the various microscreens receive
          particular attention in the operation of the full-scale facilities
          in view of the conclusions about microscreening previously
          mentioned.

      4.  The full-scale facilities be operated in different seasons to
          evaluate the effects of temperature on disinfection.

      5.  The effect of chlorite (CIO?) in receiving waters be investigated
          before the widespread use of C102 is implemented.

      6.  C102 be considered as a disinfectant pursuant to the previous
          recommendation.

      7.  The effects of mixing on high-rate disinfection be thoroughly
          investigated on a full-scale study.

      8.  Two-stage disinfection with Cl2 and C102 be investigated to
          determine the mechanism through and conditions under which
          enhanced disinfection occurs.

      9.  For bench-scale comparisons of the factors that affect the
          disinfection of CSO, a SCSO may be satisfactory, if SCSO is
          prepared properly.


     10.  The procedural difficulties in running bench-scale studies
          such as these must be recognized.  The greatest care must be
          taken to preserve the intended experimental conditions and to
          maintain sample integrity.

     11.  Blending be adapted as a preliminary step in the bacteriological
          examination of waters that contain significant amounts of
          particulate matter.  Because of the differences in individual
          blenders, a study of bacterial counts vs blending time should
          be performed to determine the optimum time for each model.

     In general, the bench-scale studies indicated that high-rate treatment
of CSO, consisting of solids removal followed by high-rate disinfection,
is feasible.  Although additional conclusions and recommendations were
presented as a result of the bench-scale studies, those concerned only
with the full-scale studies have been presented above.  Bench-scale
results were valuable in developing the full-scale prototype treatment
facilities which are described in detail in Section 6.

-------
                                 SECTION 6

                    FACILITIES DESCRIPTION AND OPERATION
MALTBIE STREET FACILITY

Description of Drainage Area

     The drainage area tributary to the Maltbie Street overflow consists
of approximately 115 acres (46.5 ha) located west of Onondaga Creek.
The principal land use is for commercial and light-industrial purposes.
The tributary area which has an estimated population (based on the 1970
census) of 1,350, is served by approximately 15,500 ft (4,725m) of trunk
and lateral sewers.  The sewers convey sanitary and combined sewage to the
main intercepting sewer via an 8 in-(20.3 cm) siphon under Onondaga
Creek.  Upstream from the siphon there is a diversion device, consisting
of a side overflow weir, located in a manhole at the intersection of
Leavenworth Avenue and Evans Street.  The overflow from the diversion
structure is a 30 in.(76.2 cm) diameter concrete pipe, which originally
ran directly to the west bank of the creek.  This overflow provides the
combined sewage treated at the Maltbie Street facility.

     An average time of concentration (including inlet time and transport
time to the point of overflow) was determined for the Maltbie Street
trunk sewer based,on an inlet time estimated at 15 min.  The actual
time of concentration as observed from start of rainfall to start of over-
flow in this study varied between 20 to 45 min and averaged about 26.5 min.
Each individually measured time of concentration varied with storm intensity.
The basic physical characteristics of this site are listed in Table 9.

           TABLE 9.  MALTBIE STREET OVERFLOW CHARACTERISTICS
     Drainage Area Characteristics

Size- 115 acres (46.5 ha)

Runoff Coefficient - 0.55

Population-Tributary - 1,350

Industrial Population
Equivalent             4,500

Total Population
Equivalent             5,850
Overflow Outfall Characteristics

Length - 3*571 ft (1088 m)

Diameter - 30 in (76.2 cm)

Slope - 0.0043

Inlet Time - 15 min
Transport Time - 5 to 30 min
Time of Concentration - 20 to 45 min
                                     43

-------
Objectives and Implementation of Prototype Facilities

     The primary objective of the facilities designed for installation
at the Maltbie Street overflow was to demonstrate under field conditions
the feasibility of high-rate, fine-mesh screening for solids removal and
disinfection enhancement followed by high-rate disinfection.  Secondary
objectives were to:

     1.   Evaluate the relative performance of commercial screening
          units under similar hydraulic and solids loading conditions.

     2.   Investigate different means of high-rate disinfection.

     3.   Investigate the effect of solids removal on high-rate disinfection.

     4.   Investigate the effects of various mixing techniques on high-
          rate disinfection.

     5.   Investigate the impact of additional solids volumes from CSO
          treatment on existing treatment facilities.

     These objectives, together with the physical characteristics of the
site and the specific constraints discussed in Section 5, dictated the
following major elements for implementation of the experimental plan:

     1.   A means of pumping controlled amounts of CSO to wet-weather
          treatment units.

     2.   Screening units, with parallel flow pattern.

     3.   Instrumentation to monitor flows and to vary flows to specific
          units.

     4.   Parallel disinfection units, with the flexibility to apply
          Cl2> C102 or both.

     5.   Piping to convey treated overflow to the creek.

     6.   Monitoring and sampling equipment to record data needed for
          evaluation of equipment performance.

     7.   Fresh water and electric power for system operation.

Facilities Installed

     The prototype facilities installed at the Maltbie Street overflow
for implementation of the experimental plan included the following major
components:

          -pumping station
          - screens
               Zurn Micromatic

                                     44

-------
               Sweco Centrifugal Wastewater Concentrator
               Crane Microstrainer
          - disinfection basins
          - control instrumentation
          - electrical service
          - fresh water piping
          - chlorination equipment
          - chlorine dioxide generating and dosing equipment
          - influent and treated effluent piping

     The overall configuration of the Maltbie Street facilities is shown
in Figure 3, and Figure 4 illustrates the site location with the pumping
station, screening building and effluent discharge depicted.

Pumping Station—
     A pumping station was constructed at the Maltbie Street facility to
convey overflow to the screening units.  The pumping station structure
was divided into three compartments:  (1) an influent chamber; (2) a
metering chamber; and (3) a wet well, and pump chamber as shown on
Figure 5.

     The influent chamber was equipped with a bar screen to remove
coarse solids and an emergency bypass to allow bypassing of the total
overflow in the event of pump failure or flows in excess of pumping
capacity.

     The metering chamber contained a 30 in. (76.2 cm) magnetic flowmeter
(Brooks Instrument Division, Model 7100 Series, Emerson Electric Company,
Hatfield, Pennsylvania) to measure the total flow entering the treatment
facilities from the overflow regulator.  A signal from the .flowmeter
activated a circuit to start the treatment units and sampling equipment.

     The third chamber acted as a wet well for the pumping systems
located above the chamber.  The overall pumping scheme had three parallel
pumping systems, each consisting of a 2.5 MGD (6.6 cu m/min) constant
speed drive and a 2.5 MGD (6.6 cu m/min) variable speed drive.

     The pumping systems provided either constant or variable flow to
each of the three screening units.  The variable speed pumps were activated
by a level sensor in the wet well.  Under constant flow conditions the
variable speed pump was manually inactivated, and the constant speed
pump was automatically started when a pre-determined wet well level was
sensed.

     Under variable flow operations, each pump combination operated in
sequence.  The variable speed pump was activated first, and as this pump
approached maximum capacity, a relay was energized in the pump controller
to activate the constant speed pump.  Once the latter pump reached full
speed, the variable speed pump equalized the pumping rate with incoming
flowrate.
                                    45

-------
             ^	OVERFLOW
             \ SCREENING
               \ BUILDING
- OVERFLOW REGULATOR
\ (LEAPING WEIR TYPE)


  \
             18 INTERCEPTOR CONNECTOR
             TO MAIN INTERCEPTOR
                                                             8" PUMP DISCHARGE
12" SOLIDS  CONCENTRATE LINE
                                                  SIPHON TO MAIN
                                                  INTERCEPTOR •
            FIGURE  3.   Malthie  Street Site Plan
           FIGURE 4.   Maltbie  Street Site  Location.   Pumping
                        Station  in Foreground,  Overflow Screening
                        Building  in  Background.
                                   46

-------
Control System—
     The control  system at the Maltbie Street  Facility provided automatic
activation of  pumps,  screens, disinfection  equipment and sampling equipment.

     At a predetermined level of flow, a  signal  from the 30 in. (76.2 cm)
inlet flowmeter activated a current trip  relay,  which completed a 120 V  AC
power supply circuit.   Elements of this circuit  included the following:

     1.   Screening unit drives
     2.   120  V AC electrical outlets for sampler operation
     3.   A time-delay relay to activate  a  4-20  ma DC pump control
          circuit

          A flowmeter on each of the pump discharges provided  process control
for the following equipment:

     1.   The  chlorinator associated with each individual pump discharge
     2.   The  chlorine dioxide feed pumps associated with each individual
          pump discharge.
                                              DISINFECTION
                                              CONTACT TANKS
                                                        24" DISINFECTED
                                                        EFFLUENT LINE
          SCREENING BUILDING AND DISINFECTION TANKS
                                        OVERFLOW^''
                                        9YPASS    PUMPING STATION
                                                            X~BAR
V-OVERFLOW
 BYPASS
               FIGURE 5.  Maltbie  Street Process Orientation
      Signals  from the inlet flowmeter and  the  individual flowmeters on
 the pump discharges were telemetered to  the  offices of O'Brien & Gere
 Engineers,  Inc.   Flows were also recorded  on strip charts at the Maltbie
 Street facility.
                                     47

-------
Microscreens--
     The three screening devices selected for demonstration at the
Maltbie Street facility were:

     1.   Crane Microstrainer, manufactured by Crane Company - Cochrane
          Environmental Systems, King of Prussia, Pennsylvania.

     2.   Zurn Micromatic, manufactured by Zurn Industries, Inc., Erie,
          Pennsylvania.

     3.   Sweco Centrifugal Wastewater Concentrator, manufactured by
          Southwestern Engineering Company, Massilon, Ohio.

Operating characterics of the three units are listed in Table 10 and photo-
graphs of each presented in Figures 6 through 8, respectively.

     The Crane Microstrainer was designed to provide a maximum of 5 MGD (13.1
cu m/min) at a hydraulic loading rate of approximately 40 gpm/ft2 (98 m/hr)
of submerged screen area.  In 1973 a report (9) indicated that effective
removal of suspended matter could be achieved by the microscreening
process at hydraulic loading rates of 35 to 45 gpm/ft2 (85 to 110 m/hr).
Further study in 1974 (10) indicated that the latter flux were not achieved
by the Crane unit but were limited to a maximum of 23 gpm/ft2 (56 m/hr) for
treatment of the CSO.  However, after coagulent aids were added to the
overflow throughput was increased to 39 gpm/ft2 (95 m/hr).  Since construc-
tion of the Syracuse facilities preceded the final 1974 study referenced
above, the design flux of the Crane unit in Syracuse had been selected as
40 gpm/ft2 (98 m/hr), or 1500 gpm.  The Zurn unit design provided a maximum
flux of 30 gpm/ft2 (75 m/hr), or 2160 gpm (490 cu m/hr), utilizing 71 micron
screens.

     The Sweco unit was designed for hydraulic loadings up to 60 gpm/ft2
(150 m/hr), or 3470 gpm (788 cu m/hr), based on Sweco's experience with
similar units in the past.  A 1974 study by the Ontario Ministry of the
Environment (37) on a comparably-sized Sweco unit handling a treatment
plant bypass during storm events resulted in 22 percent removal of total
SS using 105 micron screens at a hydraulic loading rate of 66 gpm/ft2
(160 m/hr).

     The Crane and Zurn units both utilize drum screens rotating on a
horizontal axis at a variable speed of 4.5 to 6.5 rpm, dependent
on the head differential between influent and effluent at the screen (See
Figure 9).   These units are equipped with backwash jets which wash
deposited solids off the inside of the screens, into washwater troughs
and then to the sewer.  Washwater consumption is normally 0.5 to 3
percent of the throughput.flow.

     The Sweco unit contains a series of screens attached to a cage
which revolves around a vertical axis at 55 rpm.  Flow enters the inside
of the screen cage at the bottom and flows upward to a deflection plate
at the top of the unit.as schematically depected in Figure 10. Deflected


                                    48

-------
             TABLE 10.  MICROSCREEN OPERATING CHARACTERISTICS

Parameter
Drum Size
Screen Aperture, microns
Screen Mesh
Rotating Speed, rpm
Design Flow, gpm**
(cu m/hr)
Total Screening Area ft2
(m2)
Effective Screening Area, ft2*
(m2)
Design Hydraulic Loading Rate
gpm/ft2
(m/hr)
Crane
7.5 ft x 5 ft
23
230
4.5 to 6.5
3470
790
94
8.7
90
8.4
40
100
Zurn
6 ft x 6 ft
71
100
4.5 to 6.5
2160
490
108
10
70
6.5
30
75
Sweco
5 ft dia
105
150
55
1500
340
25
2.3
25
2.3
60
150
Backwash Volume, percent of
 inflow                          0.5 to 3           0.5 to 3       0.5 to 3

Backwash Pressure Ib/sq in.             40                 30             80
                  (kg/sq cm)           2.8                2.1            5.6
* Effective Screening Area is the actual area of screen used to remove solids.
  In the case of the Zurn and Crane units, it is equivalent to the submerged
  area of screen.

**These design flowrates are based on the specified screen aperture and screen
  panel support material as obtained from the equipment manufacturers.
                                      49

-------
FIGURE 6.   Crane Microstrainer
  FIGURE 7.   Zurn Micromatic
              50

-------
flow passes through the screens and is collected outside the cage.
Solids which are entrapped on the screens, together with a fraction of
the throughput, are withdrawn at the bottom of the cage and conveyed to
the interceptor.

     The Sweco Concentrator method of operation utilizes a flow "split"
operation such that up to 25 percent of incoming flow is directed back
to the dry-weather interceptor with the solids screened from the CSO,
The Zurn and Crane units utilize only backwash water of up to 3.0 percent
of the total throughput to the unit to carry the screened solids back to
the interceptor.
            -FIGURE  8.   Sweco  Wastewater  Concentrator
     By returning up to 25 percent of the incoming flow to the interceptor,
the Sweco unit mass loading calculations will indicate removals of a
portion of the solids even if the screens were not effective in filtering
the solids, since the mass balance equation for this unit is:
     where
          Ql
          Ci
          Q2

          G£
          Q3

          C3
= Q2C2 + Q3C3

influent flow to the unit, MGD (cu m/min)
SS concentration of the influent flow, mg/1
effluent flow from the unit to the disinfection process,
MGD (cu m/min)
SS concentration of the effluent flow, mg/1
effluent flow from the unit returned to the interceptor,
MGD (cu m/min)
SS concentration of flow returned to interceptor, mg/1
If Cl = C2 = Cs, the overall efficiency of the unit in removing SS would
be equal to Q3/Q1 x 100, even though the concentration levels in each
                                     51

-------
  Drum  Drive
                                       Backwash
                                       Hood
FIGURE 9.  Schematic of Horizontal Shaft Drum Screen
     Screen
     Panels
                                            Backwash
                                            Nozzles
                                             Influent Flow
                                             Automatic Valve
             ^Screened Effluent
 FIGURE 10.  Schematic of Vertical Shaft  Drum Screen
                             52

-------
wastewater stream do not change.  Thus, when comparing the SS removal
efficiencies of the two units, the concentration removal efficiency
should be considered.

     However, when attempting to predict the effects of the concentrated
SS being diverted to the dry-weather treatment facility, the mass loading
removal efficiencies should be considered since the total poundage of SS
returned to the interceptor could cause upsets in the operation of the
dry-weather treatment facility during wet-weather periods.  In addition,
the evaluation must consider that many intercepting systems do not have
the capacity to accept 25 percent of the total overflow from CSO treatment
facilities.  For example, during wet-weather events at Maltbie Street,
the intercepting system has essentially zero capacity to handle the
runoff from that drainage area.

     The Sweco unit was followed by an air flotation cell to achieve
further reduction of SS by taking advantage of dissolved air entrained
in the wastewater during the screening process.  The flotation cell was
a 8 x 38 x 2 ft (2.4 x 11.6 x 0.6 m) concrete basin with a manually-
operated scum collector for skimming off floating solids.  Dimensions
were as specified by the Sweco manufacturer and resulted in a surface
loading rate of approximately 5 gpm/ft^ (12.2 m/hr) at the design flow
of 2.2 MGD (5.8 cu m/min).

     The screens were operated during storm events in 1974, 1975 and
1976.  Various operating and sampling problems effectively invalidated
the 1974 screening data.  These problems were resolved, and the analysis
of results found in Section 8 is based on the 1975 and 1976 data.

     During screen operation, the influent and effluent from each unit
were sampled, in order to provide data to evaluate the relative performance
of the microscreens and the effects of different solids levels on dis-
infection.  The previous bench-scale studies did not result in any
conclusive finding regarding the effect of solids on disinfection.
Periodic analyses of BOD and of metals (Pb,Cr,Fe,Zn,Cu,Cd,Ni,Hg) were
also performed.

     The operation and maintenance requirements of each screening unit
were determined.  Data was accumulated on the ability of the units to
restart after both long and short periods of operation, on the effectiveness
of their backwash cycles, and on the durability of the screening material.
Other screen operating parameters examined included power requirements,
head losses and rotational speeds.

Disinfection Equipment—

     Equipment was provided at the Maltbie Street facility for the
addition of chlorine and chlorine dioxide to the effluent from each
screening unit.

     Chlorine gas was stored at the site in 150 Ib (68  kg) cylinders and
delivered to the wastewater flow by Fisher-Porter solution-feed, vacuum
type gas chlorinators (Model 70C1751).  The chlorination  equipment is shown

                                     53

-------
in Figure 11. It should be noted that the Cl2 used in the bench-scale
studies was obtained as a 5 percent solution of sodium hypochlorite
(94.25 percent available Cl2).  It has been assumed that this difference
in source of Cl2 between the bench-scale and full-scale studies had no
bearing on the results, since results in all cases were related to
dosage of free Cl2» regardless of source.

     Chlorine dioxide was generated at the site by means of a Nitrosyl
Chloride generation system (U.S. Patent 3754079) furnished by Chemical
Generators, Inc., Rochester, New York.  The process used in this system
consisted of separately pumping equal amounts of sodium chlorate-sodium
nitrite (NaC103-NaN02) slurry and nitric acid (HMOs) into a lucite
reaction chamber, in which they were mixed.  The resulting reaction was
expected to produce a 12 percent solution of C102 which could be fed
directly to a disinfection contact tank.  The equipment used is shown
in Figure 12.  It should be noted that the Nitrosyl Chloride system
involved a new and relatively unproven process.   The state-of-the-art in
C102 technology prior to this project was that field generation of C102
at the time of disinfection is necessary (since C102 deteriorates rapidly
in storage), and that large-scale, low unit cost methods for field
generation are generally lacking.  The C102 used in the bench-scale
studies was generated in small amounts according to the laboratory
procedure then described in the 13th Edition of Standard Methods.

     During operation, the rate of delivery of Cl2 and/or C102 to each
disinfection basin was controlled by a 4-20 ma DC signal from a flowmeter
on each of the screening unit pump discharges.  Disinfectant strengths
of Cl2 and C102 solutions were monitored in the field.  Bacteriological
and viral samples were taken before and after disinfection for analysis.

     Particular attention was given to determining and controlling the
strength of C102 solutions produced, since the performance of the
generation equipment was found to be erratic.

     The disinfection contact facilities consisted of three parallel
tanks with an approximate flowpath distance of 30 ft (9.1 m).  The tanks
were designed for a one-minute contact time at all flows which was
provided for by use of a proportional weir on the downstream end of the
disinfection tanks.  This constant contact period facilitated comparisons
of disinfection techniques.

     In order to demonstrate the effectiveness of different mixing
applications, experiments were attempted during the 1976 facility
operations in which a common header from the Zurn unit to all three
contact tanks was installed.  This modification was to provide influent
of virtually identical quality to each of the tanks, thus maximizing the
validity of any comparison of mixing procedures.  However, the amount of
flow which could be passed through the Zurn unit screens was not sufficient
to operate the three disinfection basins in parallel at design levels.
These experiments were then performed by removing the screen panels of
the Zurn unit and using unscreened CSO.
                                     54

-------
FIGURE 11.   Maltbie Street - Chlorine Disinfection Equipment
  FIGURE 12.  Maltbie Street •< Chlorine Dioxide Generators
                              55

-------
     One of the three contact tanks was equipped with two flash mixers,
located at the point of disinfectant injection (tank entry) and at the
downstream end (approximately 10 ft (3.05 m) of the first longitudinal
baffle.  Each of the other two tanks was equipped with a single flash
mixer at the point of disinfectant injection.  This allowed for comparison
of single-flash mixing with sequential-flash mixing. Positive results
had previously been observed by Kruse during a full-scale sewage treatment
plant study of sequential flash mixing (38).

     Previous pilot work by Glover (39) indicated that an increase in
turbulence throughout the length of a contact tank increases the efficiency
of the disinfection process.  Glover's design objective in the high-rate
pilot contact chamber was to achieve a GT value of 10,000 (GT is a
unit]ess measure of mixing intensity where G is the velocity gradient in
sec"' and T is the contact time in sec).   In order to maintain GT at a
level of 10,000 without using long contact time, or a large tank size, G
was increased by inserting corrugated, closely-spaced baffles in the
tank parallel to the flow.  In Glover's work a flash mixer was also used
at the head of the contact tank.

WEST NEWELL STREET FACILITY  -

Description of Drainage Area

     The drainage area tributary to the West Newell Street overflow
consists of approximately 54 acres (21.9 ha) located within the City of
Syracuse,.east of Onondaga Creek.  The area is primarily residential
with only one major commerical establishment, a laundromat.  The population
of the area at the 1970 census was 1,200.  The tributary area is served
by approximately 6,400 ft (1,950 m) of combined sewers.  All dry weather
and combined sewage from the area is collected in a 24 in. (61 cm) dia
trunk sewer and dry-weather flow is conveyed to the main intercepting
sewer on the west bank of Onondaga Creek via an 8 in. (20.3 cm) siphon
under the creek.  Upstream from the siphon a diversion device directed
excess storm flows into an overflow pipe.  Table 11 lists the overflow
characteristics of the West Newell Street site.

	TABLE 11.   WEST NEWELL STREET OVERFLOW CHARACTERISTICS	
Drainage Area Characteristics           Overflow Outfall Characteristics

Size- 54 acres (21.9 ha)                Length - 2,202 ft (671 m)

Runoff Coefficient - 0.34               Diameter - 24 in (61 cm)

Population-Tributary - 1,200            Slope - 0.003

Industrial Population                   Inlet Time - 15 min
Equivalent               300

Total Population       .                 Transport Time - 5 to 30 min
Equivalent             1,500            .                                 .
	Time of Concentration-20 to 45 mm

                                    56

-------
Objectives and Implementation of Prototype Facilities

     The primary objectives of the facilities designed for installation at
the West Newell Street overflow were:

     1.   To determine under field conditions the reliability and operating
          parameters of the swirl regulator/concentrator as a device for
          reduction of suspended solids in CSO.

     2.   To demonstrate the feasibility of a concept combining solids
          reduction by the swirl regulator/concentrator with disinfection
          to accomplish various CSO pollution abatement levels.

     These objectives required the following major construction elements
at the West Newell Street site:

     1.   Gravity diversion of all flow in the West Newell Street trunk
          sewer through the treatment facility.

     2.   Construction of a swirl regulator/concentrator.

     3.   Disinfection equipment.

     4.   A pump downstream of the swirl regulator/concentrator
          to provide pump-out of dry-weather flow.

     5.   Monitoring and sampling equipment to record data
          needed for evaluation of equipment performance.

     6.   Fresh water and electrical power.

Facilities Installed

     The facilities installed at the West Newell Street overflow for
implementation of the experimental prototype included the following
major components:

          - flow diversion
          - swirl regulator/concentrator
          - control instrumentation
          - electrical service
          - fresh water piping
          - chlorine dioxide generating equipment
          - effluent pumping

     The site plan of the West Newell Street facilities is shown in Figure
13» and a photograph of the constructed facilities is shown in Figure 14.

Flow Diversion--
     In order to divert combined sewage to the swirl regulator/concentrator,
the upstream trunk sewer was intercepted and a 24 in. (61 cm) dia concrete
diversion pipe was installed.  An inflatable plug was placed in the


                                    57

-------
                                       24 INCH EXISTING COMBINED
                                       TRUNK SEWER -
         FIGURE 13.   West Newell  Street Site Plan
                                                b.   Swirl Unit
a.  Site  Location

 FIGURE 14. West Newell Street  Swirl Regulator/Concentrator
                              58

-------
trunk sewer downstream from the new diversion manhole, to force flow
into the diversion pipe.

     The original overflow pipe was sealed off with sandbags.  A new 24
in. (61 cm) dia overflow line was run from the downshaft of the swirl
regulator/concentrator to Onondaga Creek.  A 12 in. (30.5 cm) dia foul
sewer (concentrate line) was installed between the swirl regulator/con-
centrator and the original trunk sewer in order to return solids removed
from the CSO discharge to the sanitary intercepting sewer.

     New manholes were constructed as follows:

     1.   A flow monitoring manhole was constructed in the new overflow
          line.

     2.   A rise manhole was constructed in the new overflow line so as
          to maintain full pipe conditions upstream, in order to provide
          for proper operation of the overflow flowmeter.

     3.   A flow monitoring manhole was constructed in the foul sewer.

     4.   A so-called "grit" manhole was constructed in the foul sewer,
          to a depth necessary to establish gravity flow through the
          swirl concentrator.  A submersible pump was installed in the
          "grit" manhole to remove grit and dry-weather flow.

Swirl Regulator/Concentrator--
     Design dimensions for the swirl regulator/concentrator were obtained
by applying the design approach developed by the LaSalle Hydraulic
Laboratory, LaSalle, P.Q., Canada (13).    The design approach was based
on anticipated flows to the swirl unit as the foundation from which all
design dimensions were obtained.  Supplemented design approaches have
since been developed (14)  which provide design parameters to give
the greatest solids removal efficiencies under specific hydraulic conditions.
The flows pertinent to the West Newell Street swirl regulator/concentrator
were:  (1) a maximum inlet flow of 8.9 MGD (23.4 cu m/min) corresponding
to a flood flow, (2) a design flow of 6.8 MGD (17.9 cu m/min), (3) a
dry-weather flow range of 0.50 to 0.75 MGD (1.3 to 2.0 cu m/min), and
(4) an inverted siphon flow to the sanitary interceptor of 1.2 MGD
(3.2 cu m/min).  The major dimensions are presented in Table 12.

     The foul sewer (concentrate line) was sized to convey maximum dry-
weather flow and prevent  line blockage by large solids.  A diameter of
12 in. (30.5 cm) was determined to be the minimum size necessary to fulfill
these conditions.

     The swirl regulator/concentrator was constructed with a gently-sloped
concrete floor, with floor gutters in place and vertical steel sides.
The various appurtenances associated with the swirl chamber, e.g. inlet
ramp, flow deflector, scum ring, weirs, spoilers, floatables trap, and
                                    59

-------
floor gutters were installed as recommended in the physical model design
approach.  Figure 15 presents an isometric view of the swirl regulator/
concentrator.

Control System—
     The control system was based on two independent measurement systems,
described as follows:

     1.   An ultrasonic flow monitoring system (Badger Meter, Inc.,
          Precision Products Division, Tulsa, Oklahoma) on the foul
          sewer to control operation of the submersible pump in the
          grit manhole.

     2.   A magnetic flow monitoring system (Brooks Instrument Division,
          Model 7100 Series, Emerson Electric Co., Hatfield, Pennsylvania)
          on the overflow discharge pipe, to control operation of the
          disinfection system and the automatic samplers.

     The basic control strategy at the West Newell Street facility was
such that dry-weather flow diverted from the trunk sewer flowed through
the swirl regulator/concentrator and the foul sewer to the grit manhole.
(Refer to Figure  16.) During periods of low (dry-weather) flow, the
submersible pump in the grit manhole was automatically activated at a
level set to prevent surcharge in the swirl regulator/concentretor, and
dry-weather flow was pumped back to the sanitary interceptor via the
West Newell Street trunk sewer, at a point downstream from the diversion
manhole.  During periods of high (wet-weather) flow, a signal from the
flowmeter in the foul sewer deactivated the submersible pump, and
the concentrate (underflow) was forced through the foul sewer system by the
raised liquid level in the swirl which eventually caused overflow.
As a result of pump deactivation, the grit manhole continued to surcharge
until the level of sewage reached a 12 in. (30.5 cm) dia gravity relief
line located 7.79 ft (2.38 m) above the invert of the manhole.  This relief
line allowed gravity return to the trunk sewer leading to the siphon and
interceptor.

     During periods of high flow, a gravity flow regime was established,
with diverted flow from the trunk sewer entering the swirl regulator/con-
concentrator, and exiting either via the gravity relief line from the
grit manhole or via the overflow to the creek directly from the swirl unit.

     When high flow subsided, a signal from the foul sewer flowmeter
reactivated the submersible pump, drawing down the system and preventing
wastewater from standing in the swirl chamber.

     The swirl regulator/concentrator was operated during eleven storm
events in 1974 and 1975.  As discussed in greater detail in subsequent
sections of this report, a constant operational difficulty was the lack
of sufficient storm flow to simulate design conditions.  Figures 17 and
18 illustrate the swirl under dry- and wet-weather conditions, respectively.
                                    60

-------
       TABLE  12.    SWIRL REGULATOR/CONCENTRATOR  DESIGN DIMENSIONS
Basic Design  Dimensions
                                                                        Size
                     ft
m
Inside Chamber Diameter
Inlet Diameter
Scum Ring Diameter
Weir Diameter
Overflow Outlet Diameter
Radius of Inlet gutter
0° - 90°
90° - 180°
Radius of Secondary gutter
90° - 270°
go _ 9go
270° - 360°
Offset Distance for Determining Gutter Radii
Floor to top of Circular Weir
Inlet Pipe Invert to Chamber Bottom
Depth of Circular Weir Skirt
Depth of Scum Ring
12.33
2.00
8.00
6.67
2.00

4.67
1.00

1.25
2.25
7.33
0.33
3.00
1.67
1.00
0.67
3.76
0.61
2.44
,2.03
0.61

1.42
0.30

0.38
6.78
2.23
0.10
0.91
0.51
0.30
0.20

               OVERFLOW
                                  LEGEND'
                                    a.  Inlet
                                    b.  Flow Deflector
                                    c.  Floor Gutters
                                    d.  Scum Baffle
                                    e.  Overflow Weir
f. Weir Plate
g. Spoilers
h. Downshaft
i. Floatables Collector
j. Foul Sewer Outlet
                                                                   INFLOW
         FIGURE 15.   Isometric View of  Swirl  Regulator/Concentrator
                                            61

-------
        West Newell St
          Elev. 45.10 -
                         .—Grit
                         \ Manhole
_Flow
 Metering
 Manhole
Grade
Elev. 41.0 -7
   2.0' Combined
   Sewer Inlet.
= 12.33'
          r-Existing 2£>
           Trunk
           Sewer to
           Interceptor
                                                                 /-2O1 Magnetic
                                                                   Flowmeter
                                                     - Sampling Points
        FIGURE 16.   Schematic Profile - West Newell  Street  CSO' Facilities

     Samplers were  installed to collect samples from the swirl  influent,
the swirl effluent  prior  to  disinfection, and the swirl effluent  after
disinfection.

Disinfection  Equipment--
     The high-rate  disinfection system installed at the West  Newell
Street facility  (Figure 19)  was a Nitrosyl Chloride generation  system
identical with that previously described in this report for the Maltbie
Street facility.  Provision  was made to introduce the disinfectant
either at the inlet to the  swirl  concentrator (pre-disinfection)  or  at
the bottom of the concentrator downshaft (post-disinfection).   C102
was to be injected  at a rate proportional to the quantity  of  flow
measured in the  discharge pipe.

     No valid disinfection  data was obtained during the 1974  and  1975
storms, due to rapid deterioration of the disinfection equipment  and a
continuous lack  of  cooperation on the part of the supplier to make his
equipment operable.

Corrosivity —
     Corrosivity problems at this site were especially severe since  the
chemical storage tanks were  housed in the same area as the instrumentation
controls and  C102 generating equipment.  Copper water pipes were  severely
corroded and  one terminal strip in the instrumentation panel  had  to  be
replaced.  Also, leaks in the C102 piping system resulted  in  additional
corrosion problems.  Storage of the chemicals should be accomplished in
a separate storage  area that is well -ventilated and protected from sunlight.

Flow Sensing  Devices--
     A magnetic  flowmeter was installed in the swirl effluent line discharg-
ing to Onondaga  Creek.  When calibration procedures were necessary,  a
technician was required to  descend into a damp, crowded manhole to calibrate
the flowhead.  Then the signal converter mounted in the equipment building
had to be calibrated.  Oils, grease and solids attached to the  probes  •
presented difficulties in calibrating the unit.  In contrast, an  ultrasonic
                                      62

-------
     FIGURE 17.  Swirl Regulator/Concentrator - Dry-Weather Operation
CO
                                                                         FIGURE 19.  West Newell Street  -
                                                                            Chlorine Dioxide Generators
     FIGURE IB.  Swirl Regulator/Concentrator - Wet-Weather Operation

-------
flowmeter installed in the foul sewer (with probes attached to the outside
of the pipe) was easily calibrated from the equipment building.  Constant
calibration of the magnetic flowmeter was required while only occasional
calibration of the ultrasonic flowmeter was required.  A non-contact flowmeter
is preferred over a magnetic flowmeter.

Insulation—
     The West Newell Street equipment building was insulated at the beginning
of the project in anticipation of winter operation.  A portable, electric
heater was provided to maintain above-freezing temperatures and although
this precaution was taken, water pipes burst on one occasion during the
winter when a power failure occurred in the area.  CSO facilities should  be
adequately insulated to minimize freezing problems.

Pumping Control —
     Where a pump is required for dry-weather operation of a swirl regulator/
concentrator, the pumping should be controlled by the clear effluent flow-
meter rather than the foul sewer flowmeter.  Such an arrangement would allow
a pump to continue to operate up to a predetermined flow in the overflow
pipe, at which point the flowmeter would activate a relay to remove the pump
from service and allow the swirl unit to operate under gravity conditions.

     At the West Newell Street facility the capacity of the pump installed
exceeded the desired predetermined overflow rate.  Thus, when the pump was
operating during dry-weather conditions, the pump itself occasionally created
a flow through the ultrasonic flowmeter which deactivated the pump entirely
resulting in standing water depth of three feet in the swirl chamber.
(See Section 5 for a description of the hydraulic limitations of this site.)
A pump operating mode controlled by a flowmeter in the overflow line would
avoid this problem.

Washdown Facilities--
     After each overflow event, manual washdown of the swirl chamber was
required.  This requirement could be eliminated by installation of an auto-
matic washdown system.  References are available in the literature (13,15)
for an effective means of incorporating such a system.
                                      64

-------
                                 SECTION 7

                    CSO  LOADINGS AT DEMONSTRATION SITES
RAIN DATA ANALYSIS
     Precipitation  records (40) from the U.S. Weather Bureau's  station
at Hancock Airport  for the period 1948 to 1973 were statistically analyzed
to develop rainfall  frequency-duration-intensity curves for  the Syracuse
area.  Technical  Paper No. 40 published by the U.S. Department  of Commerce
was also used  in  the development of these curves (41).  Examination of
the existing records indicated that precipitation patterns in the Syracuse
area are highly variable.   High intensity-short duration  events are
usually associated  with thunderstorms; less intense, longer  duration
rainfall events are usually caused by cyclonic activity.  The computed
relationships  of  intensity versus duration for different  storm  return
periods is illustrated in  Figure 20.
   STORM

	  lyr,
	gyr,
	Syr.
	 lOyr,
                                 4945 093 II 15
                                 4588 0.86 rae
                                 8085 093 1270
                                 IOI 16 0.94 1257
                                      10

                                   DURATION,T (MIN.)
                                                   H	1-
                                                    IOO  ZOO
         FIGURE 20.  Syracuse  Area-Rainfall Intensity vs. Duration
                                      65

-------
FIELD PROGRAM

     During 1974, 19 overflow events were monitored at the Maltbie
Street demonstration facility during a period from May 1974 through
November 1974.  The durations of the individual rainfall events
ranged from a minimum of 1.0 hr to a maximum of 11.0 hr with a mean
duration for all storms of 4.8 hr.  Average intensitites of individual
storms ranged from 0.05 in./hr (0.1 cm/hr) to 0.60 in./hr (1.5 cm/hr) with
a mean intensity for all storms of 0.17 in./hr (0.4 cm/hr).  The average
intensities of individual storms were calculated by dividing the total
rainfall by the duration of the rainfall for each event.  The mean
intensity of all monitored storms was calculated by averaging the individual
storm intensitites on a duration-weighted basis.  To attempt to define
the severity of individual storms, a maximum hourly intensity was calculated
by determining the maximum intensity during any 10-min period.  The
maximum hourly intensitites for individual storms were determined to
range from 0.02 in./hr  (0.05 cm/hr) to 1.30 in./hr (3.3 cm/hr).  All of
these storms had a return frequency of less than one year, as can be
seen by entering Figure 20 with values of 1.30 in./hr (3.3 cm/hr) and a
20 min duration.

     Between March and November 1975, 10 overflow events were monitored
 at the Maltbie Street site.  The rainfall durations of individual
events ranged from 1.0 hr to 12.0 hr with a mean duration of 6.5 hr for
all storms.  Average intensitites of individual storms ranged from 0.03
to 0.26 in./hr (0.08 to 0.66 cm/hr) with a mean intensity for the 10
storms of 0.10 in./hr (0.25 cm/hr).  Maximum hourly intensities were
calculated to estimate the severity of individual storms based on the
maximum rainfall during any 30 min period.  The results indicated a
range of maximum hourly intensitites of 0.12 to 1.50 in./hr (0.30 to 3.8
cm/hr) for the 10 storms monitored.  As in 1974, all of these rainfall
events fall into a category of less than one-year storms in terms of
return frequency (see Figure 20).

     From March 1974 to September 1975, a total of 11 storms were
 monitored at the West Newell Street demonstration site.  The durations
of individual rainfall events ranged from a minimum of 2.25 hr to 37.0
hr.  The average intensities of individual storms ranged from 0.04 to
0.40 in./hr (0.10 to 1.0 cm/hr) with a time-weighted average intensity
for all storms of 0.16 in./hr.  An effort to determine the severity of
each rainfall was attempted by determining the maximum intensity occurring
in any 20 min period during the rainfall.  Maximum rainfall intensities
ranged from 0.08 to 1.95 in./hr (0.20 to 5.0 cm/hr).  All of these storms
had return frequencies of less than 1 year.

     Runoff coefficients for the two demonstration sites were determined
from rainfall and runoff measurements.  The ratio of total runoff measured
to the total rainfall for each demonstration site resulted in an average
runoff coefficient of 0.55 for the Maltbie Street drainage area and 0.34
for the West Newell Street drainage area.  From the data -gather, no
                                     66

-------
correlation of  runoff coefficient with preceding dry-weather  intervals
could be developed.
SIMPLIFIED SWMM ANALYSIS

     Using techniques  of Simplified Stormwater Management  Model  (Simplified
SWMM) analysis (42), relationships between mass emissions  of  various
pollutants from the Maltbie and West Newell Street overflows  and the
total rainfall for given storm events have been established.   Figures
21 through 26 illustrate the mass emission of BODs, TSS, VSS,  TKN,
and TIP, respectively,  for both the Maltbie and West Newell Street
overflows for various  storm events.   The Simplified SWMM  approach
provides an accounting  of the runoff from rainfall events  by  mass balance
calculations considering such factors as drainage area, runoff coefficient,
and total rainfall (or  average storm intensity times storm duration).
The mass emissions are  determined by superimposing the runoff quality on
the volume of discharge.   No adjustment has been provided  for the variability
of the runoff coefficient within given storm events.  Actual  data measure-
ments obtained in the  demonstration study are plotted on the  graphs also.

     As indicated in Figures 21 through 26  Maltbie Street overflows
result in higher pollutant loadings for all parameters, except NHsN,
than does the West Newell  Street overflow.  This is a direct  result of
the larger drainage area and higher runoff coefficient for the Maltbie
Street location.  West  Newell Street overflows result in higher quantities
of NHsN being discharged than at Maltbie Street, largely as a result of
the larger NhhN concentrations at West Newell Street, e.g.  4.4mg/l NHsN
at West Newell Street  as compared to 0.8 mg/1 NHsN at Maltbie Street.

      LEGEND

       • WEST NEWELL STREET

       • MALTBIE STREET
                                        IS, 10 =
LEGEND

 • WEST NEWELL STREET

 • MALTBIE STREET
                                                      MALTBIE STREET<
                                                                WEST NEWELL STREET
                 TOTAL RAINFALL,
-------
u3=
as
s|
      LEGEND

        • WEST NEWELL STREET

        • MALTBIE STREET
                                 •WEST NEWELL STREET
                                                      I02
                 MW.TBIE STREET-
                                                                                 •WEST NEWELL STREET
                                                                                 LEGEND

                                                                                  • WEST NEWELL STREET

                                                                                  • MALTBIE STREET
                    TOTAL RAINFALL, in Inn- 2.54cm)
                                                                       TOTAL RAINFLALL,in(fn«2.Mcin>
  FIGURE 23.    Mass Emission of VSS
                  vs  Total  Rainfall
FIGURE 24.   Mass  Emission  of TKN
                vs  Total  Rainfall
      LEGEND

        • WEST NEWELL STREET

        • MALTBIE STREET
                     TOTAL RAINFALL.In llin" 2.54cm)
       WEST NEWELL STREET

       MALTBIE STREET
                                                                      TOTAL RAINFALL, in, (l«.'2.54cm)
  FIGURE 25.    Mass  Emission of  NH3N       FIGURE  26.    Mass  Emission of TIP
                  vs  Total  Rainfall                             vs Total  Rainfall
                                                 68

-------
     The pollutant loadings expressed in Ib/acre-in (kg/ha-cm) as measured
during the demonstration program were:
                    Maltbie St. Site
Parameter    mg/1   Ibs/acre-in  Kg/ha-cm
  SS
  VSS
  TKN
  NH3N
  TIP
325
163
2.56
0.82
1.23
73.46
36.84
0.58
0.19
0.28
32.45
16.27
0.26
0.08
0.12
West Newell St. Site
mg/1  lb/acre-in  Kg/ha-cm
276
132
8.46
4.39
0.85
62.38
29.83
1.91
0.99
0.19
27.55
13.18
0.84
0.44
0.08
                                      69

-------
                                 SECTION 8

                    MICROSCREENING - RESULTS AND DISCUSSION
GENERAL
     During the 1974 Maltbie Street facilities operation, many operational
problems were encountered relating to equipment start-up and performance.
Evaluation of sampler performance indicated that intake velocities of
0.1 ft/sec (3.0 cm/sec) and less were much too low for obtaining representa-
tive samples of the process wastewaters, although the latter also flowed at
velocities generally less than 0.1 ft/sec (3.0 cm/sec).   In all
cases, samples were lifted distances of at least 5 ft (1.5 m) from the
process streams and settling of solids was observed to occur in the
sample tubing.  In late 1974, the samplers were modified to provide
intake velocities up to 3.0 ft/sec (0.9 m/sec), thus yielding more
representative samples.  Other operational problems such as replacement
of punctured microscreen panels and electrical malfunctions of disinfection
equipment delayed the start of meaningful data collection until the
spring of 1975.

SUSPENDED SOLIDS REMOVALS

     During 1975, ten overflow events were monitored at the Maltbie
Street demonstration site, during which time the Sweco and Zurn units
were evaluated for solids removal capabilities.  The Crane microscreen
unit was functional during the 1975 monitoring period for only Storm
16.  Table 13 presents the screen loading rates for the units during
this period.  It should be noted that the Sweco unit was operated at
much higher hydraulic loading rates than the Zurn unit.  The Zurn unit
was operated at the maximum pressure differential possible between
influent chamber.and overflow weir level that would not result in influent
bypass via the fixed elevation emergency weir.  Nevertheless, flux for the
Zurn unit was limited to less than 15 gpm/ft^ (36.6 m/hr).

     Solids removal capability of the Crane Microstrainer was investigated
for two additional overflow events in 1976, as indicated in Table 13.
However, since the 1976 evaluations focused on disinfection investigations,
variable loadings to the Crane were not attempted; instead, flowrates
were held constant.

     The results of the solids removal evaluations for the three microscreens
are discussed in the following paragraphs.
                                     70

-------
       TABLE 13.  MALTBIE STREEET OPERATION SCHEDULE - MICROSCREENING
Overf1ow
 Avg. Screen Loading
   Rate (gpm/ft2)
            Screen Aperture
             Size (microns)
                 Sweco
         Crane
Zurn
Sweco
Crane   Zurn
  2-75
  3-75
  4-75
  5-75
  6-75
  7-75
  8-75
 10-75
 11-75
 16-75
40
36
32
28
43
33
66

50
11
 12
 11
  7
  9
 10
  8
  4
 12
  9
  3
 105
 105
 105
 105
 105
 105
 105

 105
 105
 23
71
71
71
71
71
71
71
71
71
71
1-76
2-76
3-76
4-76
5-76
6-76
-
_
8
8
8
— •"
no screen
no screen
no screen
no screen
no screen
no screen
-
-
-
23
23
••
no screen
no screen
no screen
no screen
no screen
no screen

Note:gpm/ft2 x 2.44 = m/hr

Sweco Centrifugal Wastewater Concentrator Performance

     Results of the performance data for the Sweco unit are presented in
Table 14.  Average influent SS concentrations for individual storm
overflows ranged from 106 to 529 mg/1 with an overall time-weighted SS
concentration of 284 mg/1.  Effluent SS concentrations ranged from 65 to
396 mg/1 with an overall time-weighted average of 196 mg/1.  SS concentra-
tion removal efficiencies ranged from 7 to 65 percent with an overall time-
weighted average of 32 percent.   The operation of'the Sweco unit utilizes
a "flow-splitting" technique whereby up to 25 percent of the influent
volume is returned to the sewer system in the form  of a concentrated slurry.
This method of operation effectively results in physical removal of a
portion of the raw wastewater from the overflow discharge. When this
removed volume is accounted for in the mass removal  efficiency calculations,
the mass removal efficiency of SS achieved by the Sweco unit ranged from
30 to 74 percent for individual storms with an overall time-weighted
average of 48 percent.  The 25 percent of total flow entering the Sweco unit
which is returned to the interceptor may be too large a volume to be accepted
                                     71

-------
                                                         TABLE 14.   SWECO CENTRIFUGAL HASTEMATER CONCENTRATOR   SUSPENDED SOLIDS REMOVAL
ro
Storm
No.
2-75
3-75
4-75
5-75
6-75
7-75
8-75
11-75
16-75
Total
Time
Min
360
225
50
120
120
30
75
45
165
Time-Weighted
Total
Volume
qal
360,000
204,750
40,500
82,800
129,600
24,900
124,500
56,250
46.200
Average
Suspended Solids
Raw Effluent
mq/1 mq/1
324
240
181
217
529
277
499
124
106
284
242
157
163
76
396
212
314
115
65
196
Concentration
Removal
Efficiency
Percent
25.3
34.6
9.9
65.0
25.1
23.5
37.1
7.3
38.7
32.3
Average
Flow Rate
gpm
1,000
910 •
810
690
1,080
830
1,660
1,250
280
900
Loading Rates
Hydraulic Solids
qpm/ft2 Ib/dav/ft2
40.0 156
36.4 105
32.4 70.4
27.6 71.9
43.2 274
33.2 110
66.4 398
50.0 74.5
11.2 14.3
35.9 137.5
Solids
Influent
Ib/min
2.70
1.82
1.22
1.25
4.76
1.91
6.91
1.29
0.25
2.39
Average
Flow Rate
Effluent
gpm
750
680
610
520
810
620
1,250
940
210
670
Solids
Effluent
Ib/min
1.51
0.89
0.83
0.33
2.67
1.10
3.27
0.90
0.11
1.25
Mass Removal
Efficiency
Percent
44.1
51.1
32.0
73.6
43.9
42.4
52.7
30.2
56.0
47.9
              Conversions:   gal  x 3.785 x 10'3 =  cu  m
                            gpm  x 3.785 x 10" 3 =  cu  m/min
                            gpm/ft2 x 2.44 = m/hr
                            lb/day/ft2 x 4.89  = kg/day/m2
                            Ib/min x 0.454 = kg/min

-------
by the dry-weather treatment facilities and/or interceptor systems in many
municipalities.  This feature should be evaluated when considering use of
Sweco units.

Zurn Micromatic Performance

     Average influent SS concentrations to the Zurn unit for individual
overflow events ranged from 120 to 748 mg/1 with an overall time-weighted
average of 308 mg/1, as presented in Table 15.  Effluent SS concentrations
ranged from 46 to 340 mg/1 with an overall time-weighted average of 172
mg/1.  SS concentration removal efficiencies ranged from 25 to 62 percent
with an overall time-weighted removal efficiency of 45 percent.  In
terms of mass removal efficiency, the Zurn Micromatic SS removal efficiencies
were the same for individual storms as were the concentration removal
efficiencies since the Zurn unit did not utilize a flow-splitting method
of removing screened solids.  Rather, backwash water from the city water
supply system in the order of 0.5 to 3.0 percent of influent flowrate was
used to return screened solids to the sewer system.  For the storms
monitored in this study, 3.0 percent of the average flowrate would amount
to approximately 30 gpm or 43,200 gpd for one Zurn unit in operation.
Although most interceptors would have the capacity to accept this volume of
flow, the backwash flow volumes generated should be evaluated on an individual
basis.  Provision can be made to utilize screened effluent for backwash water.

Crane Microstrainer Performance

     The Crane Microstrainer was operated for a total of three storms
during the demonstration study, as indicated in Table 16.  Influent SS
concentrations ranged from 118 to 971 mg/1 for individual storm events
with an overall time-weighted concentration of 619 mg/1.  Effluent SS
concentrations ranged from 42 to 588 mg/1 with an overall time-weighted
concentration of 290 mg/1.  SS concentration removal efficiencies ranged
from 39 to 79 percent with an overall time-weighted average of 58 percent.
As in the case of the Zurn Micromatic, the mass removal efficiencies
were the same as the process removal efficiencies, since no volume of
wastewater is removed from the process flow.  Backwash water from the
city water supply was utilized to return screened solids to the sewerage
system.  However, provisions can be made to utilize screened effluent for
backwash water, which would reduce operating cost by eliminating city water
use for the purpose of backwash'ing.

Comparison of Screening Unit Performance Data

     Regression analyses were performed on the process efficiency data
for each of the microscreens relating process efficiency to the hydraulic
loading rate.'  Although the correlation was not as high as might be
desired between these two parameters (r = 0.3 to 0.5), the equations
developed are considered to represent the performance trends for the
three microscreens.  The correlation results are presented graphically
in Figures 27 and 28.
                                     73

-------
                                                 TABLE 15.   ZURH HICROHATIC SUSPENDED SOLIDS REMOVAL



Total Total
Storm
No.
2-75
3-75
4-75
5-75
7-75
8-75
10-75
11-75
16-75
Time
min
260
240
50
120
. 100
75
120
45
150
Time-Weighted
Volume
gal
223,600
182,400
24,000
76,800
53,000
21,000
98,400
27,900
52,500
Average

Suspended

Solids
Raw Effluent
mg/1 mq/1
380
291
186
277
277
748
321
124
120
308
230
143
138
132
175
340
240
80
46
172
Concentration
Removal
Efficiency
Percent
39.5
50.8
25.8
52.3
36.8
54.5
25.2
35.5
61.7
44.5

Averaqe
Flow Rate
gpm
860
760
480
640
' 530
. 280
820
620
210
616

Load ins
HydrauJic
qpm/ft
12.3
10.9
6.9
9.1
7.6
4.0
11.7
8.9
3.0
9.1

Rates
Solids ,
Ib/day/ft^
56.1
38.1
15.4
30.3
25.3
35.9
45.1
13.3
4.3
34.5

Solids
Influent
Ib/min
2.73
1.84
0.74
1.50
1.22
1.75
2.20
0.64
0.21
1.68

Solids
Effluent
Ib/min
1.65
0.91
0.55
0.70
0.77
0.79
1.64
0.41
0.08
0.97

Treatment
Efficiency
Percent
39.5
50.8
25.8
52.3
36.8
54.5
25.2
35.5
61.7
44.5
Conversions:  gal x 3.785 x 10
 -3
n-3
                     cu m
gpm x 3.785 x 10"" = cu tn/min
gpm/ft2 x 2.44 = m/hr
lb/day/ftZ x 4.89 = kg/day/m2
Ib/min x 0.454 = kg/rain
                                              TABLE 16.   CRANE MICROSTRAINER SUSPENDED SOLIDS REMOVAL


Storm
No.
16-75
4-76
5-76

Total
Time
min
75
100
75

Total
Volume
qal
11,475
69,400
52.050
Time-Weighted Average
Conversions: gal x
gpm x
3.785 x
3.785 x


Suspended Solids
Raw Effluent
mq/1 mq/1
118
971
652
619
10 = cu m
10 = cu m/min
42-
588
140
290

Concentration
Removal
Efficiency
Percent
64.4
39.4
78.5
58.6


Average
Flow Rate
gpm
153
E94
694
532


Loadinq
Hydraulic
qpm/ft2
1.7
7.7
7.7
5.9, '


Rates
Sol ids
Ib/day/ft?
2.4
89.8
60.3
54.7 '


Solids
Influent
Ib/min
0.15
5.62
3.77
3.42


Solids
Effluent
Ib/m1n
i
0.053
3.40.
0.81
1.62


Treatment
Efficiency
Percent
64.4
39.4
78.5
58.6

             gpm/ft2  x  2.44  =  m/hr
             lb/day/ft2 x  4.89 =  kg/day/m2
             Ib/min x 0.454  =  kg/m1n

-------
   IOC

   , 90
  UJ
  o 80
  y
  t 70
    60
    50
    40
    30
    20
         \
                           CRANE UNIT
                               STORM AVERAGE
                               • ACTUAL DATA-SWECO UNIT
                               A ACTUAL DATA-ZURN UNIT
                               • ACTUAL DATA-CRANE UNIT
                                     ZURN UNIT
                                                              SWECO UNIT
                10
                          20         3O         40         50
                      HYDRAULIC  LOADING RATE, gpm/ft2 (Igpm/ft 2=2.44m/hr)
                                                                    60
FIGURE 27.
Microscreening  - SS Concentration  Removal  vs
Hydraulic  Loading  Rate
                                              STORM AVERAGE
                                              • ACTUAL DATA-SWECO UNIT
                                              A ACTUAL DATA-ZURN UNIT
                                              • ACTUAL DATA-CRANE UMT
                                                              SWECO UNIT
V)
(0   10
                10
           20         30         40         30
       HYDRAULIC LOAOMO RATE ,  gpm/M2 (Upm/tt2 « 2.44 m/hr)
                                                                     60
 FIGURE 28.   Microscreening-  SS Mass  Removal  vs  Hydraulic
                Loading  Rate
                                    75

-------
     In Figure 27, the SS concentration removal efficiencies for each of the
microscreens is seen to decrease as the hydraulic loading rate is increased.

     The highest storm-average hydraulic loading rate at which the Sweco
unit was operated approached 66.4 gpm/ft2 (150 m/hr).  On the basis of the
regression equation results, the SS concentration removal efficiency at
this loading rate should be about 18 percent.  At the lower hydraulic loading
rate near 4.9 gpm/ft2 (12 m/hr), SS concentration removal efficiencies should
exceed 45 percent.  Corresponding mass removal efficiencies for the Sweco
unit for the above hydraulic loading rate would be 40 to 55 percent,
respectively as indicated in Figure 28.

     The range of storm-average hydraulic loading rates at which the
Zurn unit operated was 3.3 to 13.7 gpm/ft2 ( 8 to 33 m/hr).  The concentration
removal efficiencies  (and mass  removal efficiencies) as  represented
by the regression equations in Figures 27 and 28 ranged from 56 to 35
percent.  The highest instantaneous hydraulic loading rate attained by
the Zurn unit during the study was 30 gpm/ft2 (73 m/hr) which resulted
in a concentration removal efficiency of 19 percent.  Attempts to apply
loading rates in excess of 30 gpm/ft2 (73 m/hr) resulted in the creation
of excessive differential levels greater than 6 in. (15.2 cm) between
the unit's influent and effluent chambers and subsequent bypass of a
portion of the influent flows to the receiving water.  The hydraulic
loading rate appeared to be limited by the relatively fine mesh size (71
microns) and inherent support material of the microscreen, and to some
degree to the blinding of the screens by oil and grease, since the
backwash system did not incorporate any detergent or other oil and
grease removers.  The backwash spray system operating at 36 gpm (136
1/min) and 40 psig (2.8 kg/cm2) was not sufficient to thoroughly remove
the oil and grease particles.  Increased drum rotational speed, although
automatically increased as head loss across the screen increased, did
not significantly increase the flow-through capacity of the Zurn Micromatic.

     The Crane Microstrainer was operated at only two hydraulic loading
rates: 1.7 and 7.7 gpm/ft2 (4.1 and 18.8 m/hr) during the demonstration
study.  These hydraulic loading rates are significantly lower than hydraulic
loading rates investigated on a Crane-Glenfield microstrainer in Philadelphia,
Pa., which operated at an average flowrate of 16 gpm/ft2 and achieved an
average SS concentration removal efficiency of 70 percent.(10)  The data
obtained at these loading rates was not considered adequate enough to define
the performance capabilities of the unit; however, the regression analysis
did indicate that higher SS removals might be achieved by this unit than by
the Zurn or Sweco units.  The SS removal trend is consistent with the removals
expected by a finer mesh screen of 23 microns as compared to the screen
sizes of 105 microns on the Sweco unit and 71 microns on the Zurn unit.
The testing of the Crane unit was not sufficient to define possible problems
with screen blinding.

     SS concentrations removal efficiencies as determined for the Syracuse
microscreens indicated that better SS removals were achieved at smaller
screen sizes.  At a hydraulic loading rate of 15 gpm/ft2-(36.6 m/hr) the
Sweco unit would produce a SS removal efficiency of 29 percent, the Zurn unit


                                      76

-------
32 percent, and the  Crane  unit 51  percent.   However, consideration must
be given to the fact that  the Zurn unit was limited to a maximum hydraulic
loading rate of 30 gpm/ft2 (73.2 m/hr) and  the Crane unit to 7.7 gpm/ft2
(18.8 m/hr), whereas the Sweco unit treated up to 60 gpm/ft2 (146.4 m/hr),
although at reduced  concentration  removal  efficiency.

     In terms of the SS mass  removal  efficiency as depicted in Figure 28.
the Sweco unit performed much better than  the Zurn unit and nearly as
well as the Crane unit.  As an example, at  the 15 gpm/ft2 (36.6 m/hr)
hydraulic loading rate, the Sweco  unit achieved about 47 percent removal
of SS, the Zurn unit achieved 32 percent removal and the Crane unit
achieved 51 percent  removal.   However, the  increased mass removal per-
formance of the Sweco  unit (as compared to  the concentration removal
performance) was due to the physical  removal of up to 25 percent of the total
influent volume, as  a  result  of the hydraulic split.  The impact on the
existing dry-weather treatment plant and collection system of returning
the latter volume to the existing  sewer system must be considered.
           90
           80
          £60
          [£
          o
          111
          I 30
          w

           20
        • DATA FROM LITERATURE (9) (IO)(43:(44)(45K46)(47)(67)
        A DATA FROM SYRACUSE STUDY
            CRANE"
                       ZURN
                        A
SWECO
 A
                    50
                             100       ISO
                            SCREEN SIZE, MICRONS
                                             200
                                                      250
                                                              3OO
         FIGURE 29.   Microscreening-SS Removal  vs  Screen  Aperture Size

     When the performance  data of the Maltbie Street microscreening
units are compared with-the  performance  results of other studies, the
data are seen to correspond  closely,  as  illustrated in Figure 29.
Suspended solids removals  were slightly  greater for the Sweco and Zurn
units than was  reported elsewhere, and the Crane unit produced removals
about the same  as others have reported (9, 10, 43, 44, 45, 46, 47, 48).

     The relationship  between the SS  removal  efficiencies versus influent SS
concentrations  is illustrated in  Figure  30.   The data gathered in the Syracuse
study lie within the general  range found in  other projects and summarized in
Reference 48.   The data from this project are quite scattered, however, and
do not show the increase in  removal with increase in influent SS indicated
by the curves from that reference shown  in Figure 30.
                                      77

-------
ORGANIC SOLIDS  REMOVAL

     The organic  solids removals for individual  storms  for each of the
screening units are presented in Table 17 through  19.   A summary of
average values  follows:
                          Time-Weighted Concentration  Removal
                     Sweco               Zurn
Parameter
SS,mg/l

VSS, mg/1

TOC, mg/1
In   Out  Removed    In    Out  Removed
284  196

143   79
128   57
32

45
55
308  172

172   74
112   54
45

57
52
Crane
In
619
212
158
Out
290
127
100
Removed
55
40
37
     The removals  of VSS achieved by the Sweco and  Zurn  units were greater
than the removals  for total  SS.   The percent VSS/SS for  the Sweco, Zurn
and Crane units were 50, 56  and 34 percent, respectively.   These values
are lower than the 60 to 85  percent VSS/SS ratios typical  of sewage treat-
ment influents and are somewhat indicative of the contribution of inorganics
in combined sewer  overflows  due to storm runoff.  Removals of TOC paralled
the removal rates  of VSS.  These two parameter removal rates indicate that
the organic matter contained in the combined sewer  overflow is more related
to the suspended matter than to dissolved material,  and  that the inorganic
portions of the total SS measurement consisted of considerable quantities of
fine materials.
                      Legend
                      	  DATA AS  REPORTED IN REFERENCE 48
                      •  CRANE MICROSTRAINER  (23 JJL )                  '
                      A  ZURN MICROMATIC  (71 ^)
                       0  SWECO CONCENTRATOR (I05/O
                  100
                1
                  80
                 g 60


                 o: 40


                 W 20
                                               23 ,u
                    0   100 200  300  400  500 600  700  800 900  1000  .
                            Influent  SS  Concentrations, mg/1
    FIGURE 30.   M'Icroscreem'ng-SS Removal vs  Influent  SS Concentrations

HEAVY METALS  REMOVAL

     Heavy metal  concentrations in the combined sewer  overflow at Maltbie
Street were in  the range of concentrations reported elsewhere (48, 49, 50)
with the exception of nickel  which was up to  five times  higher than other
values reported.   Since the area tributary to this overflow  is a mixture
                                      78

-------
                           TABLE  17.  SVIECO CUC   ORGANIC SOLIDS REMOVAL
Storm No. Duration min
2-75
3-75
4-75
5-75
6-75
7-75
8-75
11-75
16-75
Time-Weighted

360
225
50
120
120
30
75
45
165
Average


In
324
240
181
217
529
277
499
124
106
284
TABLE
TSS,
Out
242
157
163
76
396
212
314
115
65
196
rng/1
Percent
Removed
25
35
10
65
25
23
37
7
39
32

In
200
112
42
75
212
163
290
51
44
143
VSS,
Out
106
52
25
54
115
99
157
30
38
79
mq/1
Percent
Removed
47
53
40
28
46
39
46
42
14
45

In
_
63
54
96
339
-
150
27
-
128
TOC,
Out
_
35
43
38
no
-
99
26
-
57
mq/1
Percent
Removed
_
44
20
60
68
-
34
4
-
55
18. ZURN HICROHATIC ORGANIC SOLIDS REMOVAL

Storm No. Duration rain
2-75
3-75
4-75
5-75
7-75
8-75
10-75
11-75
16-75
Time-Weighted
26.0
240
50
120
100
75
120
45
150
Average

In
380
291
186
277
277
748
321
124
120
308
TABLE 19
Storm No. Duration, min
16-75
4-76
5-76
75
100
75

In
118
971
652
TSS,
Out
230
143
138
132
175
350
240
80
46
172
mg/1
Percent
Removed
39
51
26
52
37
55
25
35
62
45

In
317
166
45
94
163
305
86
51
81
172
. CRANE MICROSTRAINER
TSS,
Out
42
588
140
mg/1
Percent
Removed
65
39-
79

In
86
378
118
VSS,
Out
106
50
18
81
106
159
67
14
29
74
mg/1
Percent
Removed
66
70
60
14
35
48
22
73
64
57
ORGANIC SOLIDS
VSS,
Out
37
224
89
mq/1
Percent
Removed
57
41
25

In
_
154
56
101
-
183
50
28
-
112
REMOVAL

In
_
239
50
TOC,
Out
_
62
32
57
-
106
25
20
-
54

TOC,
Out
_
141
45
mg/1
Percent
Removed
_
60
43
44
-
42
50
29
-
52

mg/1
Percent
Removed
_
41
10
Time-Weighted Average    619     290
55
212   127
                                                                40
158   100
                                               37
                                                79

-------
of commercial, light industrial and residential, higher concentrations of
metals than is normally associated with residential areas were expected.

     Significant and consistent removal of heavy metals by the microscreening
units was not achieved during this study.  The influent and effluent
concentrations for all storms are presented in Table 20 and 21, as well as
the range of influent values encountered during the study.
             TABLE 20.  SWECO CONCENTRATOR   HEAVY METALS REMOVAL

Heavy
Metal
Fe
Cr
Cu
Pb
Zn
Cd
Ni
Mean Concentration, mg/1
Influent
1.93
0.22
0.09
0.20
0.19
0.00
0.70
Effluent
2.14
0.31
0.08
0.17
0.24
0.00
0.69
Range of Influent Values
Low
0.00
0.00
0.00
0.00
0.02
0.00
0.00
Hiqh
17.20
1.60
1.60
0.90
2.40
0.02
19.00

             TABLE 21.  ZURN MICROMATIC    HEAVY  METALS REMOVAL

Heavy
Metal
Fe
Cr
Cu
Pb
Zn
Cd
Ni
Mean Concentration, mg/1
Influent
1.65
0.18
0.08
0.19
0.24
0.00
0.52
Effluent
1.13
0.25
0.07
0.09
0.26
0.00
0.60
Range of Influent Values
Low
0.00
0.00
0.00
0.00
0.02
0.00
0.00
Hiqh
17.20
1.60
1.60
0.90
2.40
0.02
19.00

      In  general,  it was concluded that screens could remove metals only via
 particulate removal either through direct removal of heavy metal compound
 precipitate or as  a result of a dissolved metal being sorbed on a particle.
 The  only metal that showed removals at measurable and significant concen-
 trations was  Pb.   All other metals showed highly variable removals and/or
 relatively low concentrations.  The removal of Pb is most probably due to the
 fact that Pb  compounds are insoluble or only slightly soluble in water and
 would be removed  relatively easily with other suspended matter.
                                     80

-------
     Data collected during this study are considered insufficient to draw
conclusions about heavy metals removal capabilities of the microscreening
units.  Whatever removals that were achieved are considered incidental to
the microscreening process.  Since reduction of heavy metals in terms of
concentration was so erratic, no attempt is made here to determine the mass
removal rates of the individual units.
                                     81

-------
                                 SECTION 9

            SWIRL REGULATOR/CONCENTRATOR - RESULTS AND DISCUSSION
GENERAL
     Studies of the hydraulic model at LaSalle Hydraulic Laboratories
determined that particle removal effectiveness was a function of the
particle effective diameter and specific gravity, or in other words  the
particle settling velocity (13).  In the model studies, removals of
grit with specific gravity of 2.65 and greater than 0.33 mm in diameter
were greater than 90 percent.  This removal percentage for grit decreased
to less than 40 percent for 0.1 mm particles.  For settleable solids of
specific gravity 1.20, efficiency ranged from 80 to 100 percent for
particles larger than 1.0 mm, 30 percent for 0.5 mm particles, and no less
than 20 percent for 0.3 mm sizes.

     In all of the model studies, the hydraulic loading rates for individual
tests were held constant.  Also, since simulated solids were injected at
constant rates, the SS loading rates were also constant for individual
tests.  In the Syracuse prototype swirl regulator/concentrator at the
West Newell Street site, these uniform loading rates were not possible to
obtain since the swirl was installed directly on a combined sewer.  This
situation makes it difficult to compare results obtained in the field to
results obtained in a highly controlled laboratory model analysis.  However,
a series of grab samples totaling 30 gal (113 1) of inflow to the Syracuse
swirl prototype was collected during one overflow event for particle settling
velocity analysis, and comparison of the observed particle removal
efficiency to the predicted model results was made.  The following para-
graphs describe the settling velocity tests and the overall performance
of the swirl unit in terms of SS, VSS, TOC and heavy metals removal.

SETTLING VELOCITY TESTS

     Settling velocity relationships of sanitary sewage and stormwater runoff
have been presented previously (51).  In tests of the settling velocity
distributions in sanitary sewage, the median settling velocity observed
was 0.00018 fps (0.054 cm/sec) with 31 percent of the particles having
settling velocities of less than 0.00033 fps (0.01 cm/sec).  Similar tests
of urban stormwater showed that 78 percent of the particles have settling
velocities less than 0.00033 fps (0.01 cm/sec) with a mediam velocity of
less than 0.000033 fps (0.001 cm/sec).  In this particular case, it is not
known under what storm intensity conditions the urban stormwater sample was
collected.  If the storm intensity was low, the runoff may not have nad
sufficient velocity to carry larger-si zed particles to the storm conduits.

                                      82

-------
This possibility could account for the  lower  settling  velocities  obtained in
the urban stormwater sample than was found  in the  sanitary  sewage sample.
A second test conducted on a portion of the same urban stormwater sample
after storage of six days at 4°C showed a median settling velocity of 0.00023
fps (0.007 cm/sec) with only 56 percent of  the solids  having  settling
velocities less than 0.00033 pfs (0.01  cm/sec), indicating  improved settling
characteristics after storage, apparently due to agglomeration  of small
particles during storage.
                         i   i I I I   I
                         5   10  15 20  3(
                                 30 40  50 60 70 8085 90  95  98


                             PERCENT LESS THEN OR EQUAL TO
  FIGURE 31.  Settling Velocity Distribution Curve - West Newell Street  CSO
     A similar  settling  velocity test was conducted on the West Newell
•Street overflow  during one  overflow event.   As  seen in Figure 31, only
17 percent  of the  solids were  found to have a settling velocity less
than 0.00033 fps (0.01 cm/sec).   A mean settling velocity of 0.0021 fps
(0.064 cm/sec)  was determined.   These data indicate that the mean
settling velocity  of  the solids  at West Newell  Street was similar to the
sanitary sewage sample,  although the range of settling velocities was smaller.
The results are  compared against  those  obtained  for the  stored stormwater
sample by Dalrymple,  et  al_  (51).   In  the  stored  stormwater sample,  56 percent
of the solids had  settling  velocities less  than  0.00033  fps (0.01 cm/sec)
compared to 17  percent for  the West Newell  Street sample.   Dalrymple, et al_,
determined that  nearly 40 percent of the  particles  had a settling of less
than 0.000033 fps  (0.001 cm/sec)  while the West  Newell Street sample showed

                                      83

-------
that only 13 percent had a settling velocity of 0.000033 fps (0.001 cm/sec)
or less.  Also, 52 percent of all particles in the West Newell Street CSO
had settling velocities between 0.00033 fps (0.01 cm/sec) and 0.0033 fps
(0.10 cm/sec).  The similar sample tested by Dalrymple, et^ a]_, indicated
that only 40 percent of the particles in urban stormwater had settling
velocities between 0.00033 fps (0.01 cm/sec) and 0.0033 fps (0.10 cm/sec),
while 56 percent were less than 0.00033 fps (0.01 cm/sec).  Unfortunately,
at the time of the Dalrymple report, a suitable sample of CSO was not avail-
able for settling velocity analysis  and thus no comparison of the West
Newell Street CSO sample with another CSO sample can be made here.

     In summary, the CSO at West Newell Street for the one storm event
where settling velocity characteristics were determined, contained particles
with settling velocities somewhat higher than particle settling velocities
of sanitary sewage and stormwater runoff.  The mean settling velocity of
CSO should lie within a range determined by the mean settling velocity of
sanitary sewage as the upper range limit, and the mean settling velocity of
stormwater runoff as the lower range limit, since CSO is considered to
consist of some mixture of sanitary sewage and stormwater runoff.  The
mean settling velocity of CSO at West Newell Street is only slightly greater
than for sanitary sewage; however, further testing to define the settling
characteristics of CSO is warranted.

     No particle size distribution data was gathered at the West Newell
Street site.  Table 22 presents a condensed form of the size distribution
relationships presented by Dalrymple, et al_,in the previously cited report.

TABLE 22. PARTICLE SIZE DISTRIBUTION (PERCENT) AND SOLIDS RANGES  IN
	SANITARY SEWAGE, STORMWATER RUNOFF, AND COMBINED SEWER OVERFLOWS.*(5l)

               Sanitary       Stormwater     Combined Sewer      W. Newell
   Parameter    Sewage	Runoff	Overflows	CSO	

Sett-S  (<100y)    45             90               27

SupraColloidal    35              9               50
        (1 to  lOOy)

Colloidal        J20            __]_              23
      (1 my to lu)

     Total       100            100              100


VSS Ranges     70-85          10-40            25-86               24-87

Sett-S  Ranges  37-65          70-90            37-87                 19
*Expressed as a  percentage  of  SS


                                     84

-------
     From Table 22 it is seen that the VSS fraction of the West Newell Street
CSO compares with the VSS fraction as determined in other studies of CSO,
having a very wide range.  The settleable solids value as determined on one
sample used in the settling velocity tests is lower than the range yielded
by other test data, 19 percent for West Newell Street as compared to a range
of 37 to 87 percent in other studies.
     Based on the settling velocity data gathered during the storm event
of July 19, 1975, a comparison of predicted SS removals and actual SS
removals was made.

     Adjustments of the performance curve as predicted by the model studies
were necessary to arrive at the corresponding performance curve predicted
for a 12.3 ft (3.7 m) unit since the results of the model studies were scaled
from a 3 ft (0.9 m) diameter unit (13).  The performance curves for swirl units
of various sizes are related by the Froude number since the latter is the
parameter upon which the flows and settling velocities of the simulated
solids used in the model are based.   From Froude number relationships, the
equations relating the performance curves of various size swirl units are
described below:
      '12.3  =
                     1/2
where
     V12 3 = settling velocity of particles in a 12.3 ft (3.7 m) diameter unit

     V3 = settling velocity of particles in a 3.Oft (0.9 m) diameter unit

     d!2.3 = diameter of 12.3 ft (3.7 m) unit

     d3    = diameter of 3.0 ft (0.9 m) unit
'12.3  =  "3
                  12.3
                      1/2
                           = 2.025
     The settling velocity curve referenced above was therefore adjusted
by multiplying all values by 2.025.  Flowrate, Q was adjusted by a similar
value derived from the Froude number relationship of
      ^12.3  =
                      5/2
where
            =  design flowrate of the West Newell Street unit

            =  design flowrate of a 3.0 ft (0.9 m) diameter unit

                                     85

-------
      d!2.3  =   diameter of a 12.3 ft (3.7 m) unit

      d3     =   diameter of a 3.0 ft (0.9 m)  unit
      n         n        5/2         n
      y!2.3  =   ^3  (4.1)     = 34.04 U3
                      100

                       90

                    B  80
                    z
                    o  .

                    1
                    HI  en
                    n.  60
                    3
                    £  50
                    in
                    d  40

                    a  30

                       20

                       10

                       0
                                5   10       50  100
                            FLOWRATE, cfs (cfs x 1.7 = cu. m/min.)

       FIGURE  32.   Predicted Performance of Prototype Swirl vs Flowrate


     Figure 32 represents the resulting  curve of  predicted performance of
the West Newell Street prototype swirl  concentrator.   The  family  of curves
presented in Figure 32 are  representative  of  the  performance  of the
prototype swirl at a foul sewer flow  of  22 percent of  the  total inflow.
(the foul sewer fraction of the July  19,  1975 storm).   Table  23 illustrates
the predicted performance and the actual  performance for the  July 19,  1975
event.  The predicted performance is  arrived  at by entering Figure 32  at
the desired flowrate and determining  the  percent  removal of particles  of
the various settling velocity ranges.   The percent removal  of particles
for each settling velocity  range is then multiplied by the fraction of
total particles represented by the particles  of each given settling
velocity range.  The sum of the individual particle removals  is the pre-
dicted removal for the swirl regulator/concentrator.   The  procedure is
outlined in Table 23.

     As Table 23 indicates, the mass  removal  of total  SS for  the  July
19, 1975 storm should have  been approximately 32  percent,  while the
SS mass removal measured on the day of  actual overflow was determined  to
be 33 percent.  The predicted removal is  slightly lower than  the  actual
removal despite the fact that the West  Newell Street settling velocity sample
was stored for five days as described earlier. Dalrymple,  et  al_,  reported
that the settling characteristics of  solids appear to  be improved with
storage (51).  Since the settling characteristics of West  Newell  Street
                                     •86

-------
      	TABLE  23.   PREDICTED  VERSUS  ACTUAL  SS  REMOVAL  -.SWIRL  REGULATOR/CONCENTRATOR


      Predicted  Removal  - Analysis  for  Stored  Sample
       Particle  Settling
       Velocity, ft/sec
Percent of Total
Particles With
Given Settling Velocity
Percent of Particles
With Given Settling
Velocity That Are Removed
Percent of Particles
Removed as Percent
of Total Particles
00
0.00003
0.00028
0.00085
0.00138
' 0.00174
0.00236
0.00298
0.00387
0.00574
0.01082
0.03279
13
7
10
10
10
10
10
10
10
5
5
22 '
22
22
24
25
26
28
34
42
56
86
2.9
1.5
2.2
2.4 •' .
2.5
2.6
2.8
3.4
4.2 ,
2.8
4.3
      Total Percent SS Removed
                                                                    31.6
      Actual Removal of SS

           From  Table 24,  Storm 12-75:
           Total  Mass in = 551  IDS  (250 kg)
           Total  Mass out  = 370 Ibs (168 kg)
           Percent  Removal  = 33%
      Conversions:  1 ft/sec = 30.5 cm/sec

-------
overflow were determined after storage and the actual removal determined on
unstored sample, these results tend to indicate that the actual performance
of the swirl unit is better than predicted performance.

     In summary, the one settling velocity distribution performed on the
West Newell Street swirl influent tends to confirm the predicted per-
formance curve determined from the model studies (13).

SUSPENDED SOLIDS REMOVALS

     Relatively good SS removal efficiencies were determined over the
entire storm-flow range for this prototype (Table 24).  Total mass loading
and concentration removal efficiencies ranged from 33 to 82 percent and 18
to 55 percent, respectively, as flowrates ranged from relatively minor
flow of 0.2 MGD (0.5 cu m/min) to a high of 7.6 MGD  (20 cu m/min).

     Under dry-weather flow conditions, most regulators are designed to
direct the entire flow and associated solids to the  intercepting sewer.
When flow conditions exceed the maximum capacity of  the regulator to direct
flow to the the interceptor, overflows result whereby flows in excess of
the regulator capacity are then discharged from the  sewer system.  The
swirl has the added advantage of concentrating solids as well as convention-
ally diverting flow during overflow events.  This concentrating effect is
evidenced by removal efficiencies for the swirl in terms of SS concentrations
varying from 18 to 55 percent (Table 24), as previously stated; whereas
conventional regulators are assumed not to concentrate solids and result
in zero percent removal.  (See Table 24, Footnote 3)

     If a hypothetical conventional regulator was developed that did not con-
centrate solids, the net mass loading reductions (attributable to the SS
conventionally going to the intercepted underflow) would have ranged
from 17 to 64 percent as compared to a more effective range of 33 to 82
percent for the actual swirl unit.  This may be a better way to compare the
effectiveness of the swirl to conventional CSO regulators.

     For low-flow storms, approaching the maximum dry-weather capacity
of the interceptor, the advantages of swirl concentration are reduced as
would be expected based on the physical principle involved.  In other
words, as the ratios of inflow to foul outlet underflow (or weir overflow
to foul outlet underflow) decrease, the SS removal advantage from swirl
concentrating also decreases, since the intercepted  hydraulic loading to
underflow becomes more significant in the net mass loading calculation
of the hypothetical or conventional regulator, according to the equation:

  Hypothetical Regulator   _  n.r  n r       „  1nn     Q.-Q     inn  0-   ,__
  Percent SS Mass Removal  ~  frfl:9°Cff-     x  10°  =  £J!o  x 100 =^f x 100
                                UT L»"l                    U •            U •
                                  i                       1            ^^

    Where Q.. = influent flow

          Q0 = overflow discharged to receiving stream

          Qf = foul sewer flow

                                     88

-------
                        TABLE 24.   SVJIRL REGULATOR/CONCENTRATOR   .SUSPENDED SOLIDS REMOVAL
                                                                                           1
00
10
Swirl Concentrator
Average SS




per
storm, mg/1

Mass

Loading
Ib


Avg. j
Storm
No.
2-74
3-74
7-74
10-74
14-74
1-75
2-75
6-75
12-75
14-75
15-75
Mean

1.
Flow
M6D
2
1
2
2
2
0
1
1
2
2
3

1
•
•
•
•
*
*
*
•
*
•
•

•
2
0
8
9
2
5
5
4
8
4
3

9
SS removals
influent
2.
3.
and
Inf
535
182
no
230
159
374
342
342
291
163
115

276
were
Eff.
345
141
90
164
123
167
202
259
232
119
55

176
calculated
?
Rem.
36
23
18
29
23
55

Inf.
824
152
205
564
218
227
41 1020
24
20
27
52

®
from actual
247
551
367
258

496

Eff.
394
75
134
295
126
53
368
137
370
216
46

221
VL
Rem
52
51
34
48
42
77
64
45
33
41
82

&t
/SK
/ U \J
( 	 s
measured data sets
Conv.
Mass



s
. Inf.
824
152
205
564
218
227
1020
247
551
367
258

) 496
of
Regulator
Loading
Ib
	 	 	 	 • — ' 	 1

„ W^
Underflow Rem. ^
222 27 2-ST°»
73 48 3 0/;
44 22 /3C<
108 19 3?v
57 26 /& %
145 64 /J%j
374  3**4
68 TtT '%%
106 / 19 //%
80 / 22 1 /?x£
159 / 61 \ A/%
1 \
164 1 33
..
effluent samples. " Xr2fbC>%>
Data reflecting negative SS
For the
conventional regulat
removals at
.or removal
tail
end of
calculation,
storms
it is
not included
assumed that
•
the ^ ^ '
            SS concentration of the foul underflow equals the SS concentration of the
            inflow.
       4.    Conversion:   1  Ib x 0.454 = kg

-------
     C0 = Cf = C-j = SS concentration  in  influent
This phenomenon can be illustrated by comparing  the  removals  shown in
Table 24 of Storm 10-74 in which the average  flow  was  relatively
high (2.91 MGD) (7.7 cu m/min) to Storm  1-75  in  which  the  average flow
was relatively low (0.50 MGD)  (13 cu m/min).   Comparison of the removal
efficiency of the swirl versus a conventional  regulator shows that for
Storm 10-74, the SS removal by the swirl was  48  percent compared to
19 percent for the conventional regulator,  a  difference of 29 percent.
Storm 1-75, the difference in SS removal was  only  13 percent.
                                                     For
                 CE

                 V>
70


60


50


40


30


20


10
                                    I
                                        I
                                           l
                                                 Influent SS
                                                 Concentration
                                                   500 mg/1
                                                   100
                                                   50
                                               I
                                                  j_
                                    6      9      12
                                  FLOWRATE, cfs (I cfs = 1.7 cu m/min)
         FIGURE 33.   Prototype Swirl Regulator/Concentrator SS Removal
     Despite the  apparent  decreased advantage of swirl  concentration at low
flows it  is important  to consider that many overflow outfalls are designed
to pass 20, 100 or  even 1,000 times average dry-weather flow, as opposed
to West Newell Street  which,, at best,  passes only 10 times average dry-
weather flow.  For  the former cases, the swirl concentrating effect will
exhibit distinct  advantages  over conventional regulators for SS removal.

     Figure 33 illustrates that at increased flowrates, the percent concen-
tration removal of  SS  is decreased.  In addition, increased influent SS
concentrations tended  to result in increased removals of SS.  It is believed
that the  SS concentrations in the CSO fluctuate in response to scouring
velocities in the sewer line.   Thus, wastewaters during the first flush
tend to have a greater proportion of solids of larger size and specific
gravity which are more easily removed by the swirl regulator/concentrator.
                                      90

-------
      In order to  account  for  the  influences  of both flowrate and. influent
SS concentrations,  the  performance  data of the swirl  was statistically
fitted using multiple regression  analysis to an equation of the form:

      Percent SS Removal   =  K-,QK2SSI<3
where  KI,  K2,  l<3  are  constants,  Q is  hydraulic flow or f-lux to the unit (MGD
or gpm/ft2)(cu m/min  or m/hr) and SS  is the influent SS concentration (mg/1).
The results of the  regression analysis  are shown in Table Z5.   The signs
associated with the regression coefficients indicate that SS concentration
removals generally  increased with an  increase in influent SS concentrations
and decreased  with  increasing flowrate.   Values of "t" associated with Q
and SS indicated  degrees  of confidence  of greater than 99 and  95 percent,
respectively,  and an  "F"  value for the  overall expression of greater than
99 percent.
.90
 80
£70
O 60
 40
 30-
O
£20
u

1,0
a.
      Note:
             Data from Appendix E
Note:  Data from Appendix F
                                         80
                                         70
                                        560
                                        O
                                        !50
                                        z
                                        940
                                         20
                                        V)
                                        at 10
                                 I
      10 20 30 40 50 60  70  80  90  100
           OBSERVED SS REMOVAL, %

  FIGURE  34.   Observed vs Predicted SS
              Removal-Swirl Prototype
                                          0     10     20     30     40     50
                                                        FOUL FRACTION, %
                                         FIGURE 35.  Percent Foul Fraction vs
                                                     SS Concentration Removal
                                                     Swirl Prototype
     The  final  regression  equation thus obtained was

     Percent  SS Concentration  Removal  =32.4 Q^'^SS0'07

     The  trends indicated  by the regression equation are presented in
 Figure 33 where the  higher flowrates  indicate .lower SS removals for given
 influent  SS concentrations.  A plot of the observed SS removal versus
 predicted SS  removals  is presented in  Figure 34.

     A regression  analysis of  SS removal  rates  versus the foul fraction
 flow as a percentage of influent flow  indicated a slight increase in SS
 removal efficiency with increased foul fraction.  The results are pre-
 sented in Figure 35  along  with measured data of 11 overflow events at the
 West Newell Street facility, consisting of an average of seven data points
 per storm, or a total  of 79 data points.     A  possible explanation for this
 effect is that  at  increased foul fractions, the SS deposited around the floor
 gutters and outlet are removed more effectively, and the chances of

-------
                     TABLE 25.  MULTIPLE  REGRESSION  ANALYSIS  FOR SWIRL  PROTOTYPE SS REMOVAL
ro

Independent Standard
Variable Mean Deviation
Log Q, MGD 0.25 0.33
Log SS , mg/1 2.31 0.36
Dependent Variable
Log (Percent 1.65 0.12
SS Removal)
Intercept 1.51
Multiple Correlation 0.40
Std. Error of Estimate 0.12
Source of Variation
Attributable to Regression
Deviation from Regression
Correlation
X vs Y
-0.34
0.27

Regression Std. Error of
Coefficient Regr. Coef.
-0.11 0.04
0.07 0.04

Analysis of Variance for the Regression
Degrees of Sum of Mean
Freedom Squares Squares
2
84
0.21 0.11
1.11 0.01
Computed
t-Value
-2.97
2.04
.
V Value
7.94
      Total
86
1.33

-------
resuspension and loss to overflow are lessened.

ORGANICS REMOVAL

     Since nonbiodegradable synthesized solids were used, no evaluation
of BODs removal was made in the laboratory swirl hydraulic model.  Pro-
totype analyses indicated greater than 50 percent BODs removals in terms
of mass loading and concentration (Table 26).  Total mass loading removals
and treatment efficiencies in terms of concentration ranged from 50 to
82 percent and 29 to 79 percent, respectively.

	TABLE 26.   SHIRL PROTOTYPE BODs REMOVAL	
Mean
                              Average BOD5
                             per storm, mg/1
                                      Mass Loading, 1b
Storm
No.
7-74
1-75
2-75
Avg. Flow
MGD
2.8
0.5
1.5
Percent Percent
Influent Effluent Removal Influent Effluent Removal
314
165
99
65
112
70
79
32
29
610
214
385
106
66
189
82
69
51
1.6
193
82
58
403
120
70
Conversion:
 1 lbxO.454 = 1
 1 MGDX2.63 = 1
 kg
 cu m/min
     Twelve overflow events were monitored for TOC at West Newell Street
during this project.  Data collected showed removals of TOC by the swirl
unit as indicated in Table 27.  The TOC removals achieved in individual
storms ranged from 5 to 53 percent in terms of concentration.  A certain
portion of the CSO influent flow is diverted back to the interceptor via
the foul sewer outlet.  This portion of flow diverted back to the interceptor
varies greatly depending upon the severity of the storm and in this study
ranged from an average of 14 to 59 percent for individual storms and an
overall average of 31 percent.  At future CSO treatment sites, the foul sewer
fraction may not be as variable depending on the hydraulics at those sites.
An overall time-weighted average calculated for all storm data indicates
33 percent removal of TOC in terms of concentration.
                                      93

-------
          TABLE  27.    SWIRL PROTOTYPE   TOC AND VSS REMOVALS
               TOC, mg/1
                          VSS. mq/1
Storm
No.
2-74
3-74
7-74
10-74
14-74
1-75
2-75
6-75
10-75
12-75
13-75
14-75
Influent
166
56
139
83
48
128
44
62
202
39
30
25
Effluent
99
35
65
66
35
84
38
44
118
37
27
19
Percent
Removal
40
38
53
20
27
34
14
29
42
5
10
24
Influent
304
135
68
65
96
94
173
130
289
132
-
46
Effluent
203
73
38
60
64
80
121
65
170
77
-
21
Percent
Removal
33
46
44
8
33
15
30
50
4i
42
-
. 54
VSS/SS
57
74
62
28
60
25 '
51
38
41
45
-
29
Time-
Weighted
Average   80
54
33
132
87
34
46
Removals of VSS  for. individual  storms are also presented in Table  27.   The
overall time-weighted  average removal of VSS at the West Newell  Street swirl
was 34 percent,  with a range for individual storms of 8 to 54  percent.   Gen-
erally speaking,  the removal of VSS paralleled TOC removal rates although
there were exceptions.   A general tendency of increased VSS removals  at the
higher VSS/SS  ratios was observed as depicted in Figure  36.  The data  had an
R value of 0.54.
                        90


                        80


                        70


                        60


                        50


                        40


                        30


                        20


                        10
                            O.I 0.2 0.3 0.4  0,5 0.6 0.7 0.8  0.9  1.0

                                     VSS/ SS RATIO
        FIGURE 36.   VSS/SS  Ratio vs VSS Removal - Swirl Prototype
                                      94

-------
HEAVY METALS REMOVAL

     Heavy metals concentrations at this site were generally low except for
Fe and Zn.  Pb and Cu were measurable and Cr, Cd, and Ni were barely
detectable. Table 28 presents a summary of the heavy metals concentrations
for all storms. All heavy metals were generally less than 1 mg/1 in
concentration except for occasional spikes, particularly with regard to Fe.
Removal of metals by swirl concentration was erratic as indicated in
Table 28.  Evaluations for each of these parameters did not yield consistent
removals throughout any of the storms.  It is interesting to note that the
standard deviation of effluent heavy metals concentrations was always less
than the standard deviation of influent concentrations.  Apparently heavy
metal influent spikes are dampened out by passage through the swirl unit.
However, removal of heavy metals by swirl concentration would have to be
considered incidental to the units more effective role in removal of
suspended solids.

COARSE FLOATABLES REMOVAL

     The coarse floatables/scum removal mechanism worked satisfactorily.
Visual observations during overflow events revealed floatables to be
effectively contained by the scum ring in the outer ring of the chamber
and forced into the floatables trap (under the weir plate) by the swirl
action for subsequent drawn-down and removal to the foul sewer during
dry weather. Figure 37 illustrates flbatables entrapment during wet-
weather operation.
          FIGURE 37.  Coarse Floatables Removal - Swirl Prototype
                                     95

-------
TABLE 28  SWIRL PROTOTYPE HEAVY METALS REMOVAL

Concentration, mg/1
Heavy
Metal
Fe
Cr
Cu
Pb
Zn
Cd
Ni
No.
Points
46
52
46
42
50
42
26
Influent
Mean
2.26
0.00
0.05
0.01
0.55
0.00
0.00
Std Dev.
4.28
0.00
0.09
0.04
1.60
0.01
0.00
Range
Effluent Influent
Mean
2.29
0.00
0.05
0.01
0.13
0.00
0.00
Std. Dev.
3.74
0.00
0.05
0.03
0.07
0.00
0.00
Low
0.10
0.00
0.00
0.00
0.02
0.00
0.00
of
Values, mg/1
High
19.10
0.01
0.14
0.20
0.34
0.02
0.00
                        96

-------
                                   SECTION 10

             SOLIDS HANDLING CONSIDERATIONS - RESULTS AND DISCUSSION
GENERAL
     A major factor to be considered in design of CSO treatment facilities
is disposal of solids.  Therefore, an attempt is made in this report to
define the impact additional solids loadings to  the Syracuse Metropolitan
Sewage Treatment Plant (Metro) would have on existing solids handling
capabilities and treatment efficiency.  From this specific example,
inferences may be drawn on the general quantities and impacts of solids
in other situations.

     In performing these analyses, a number of simplifying assumptions
are made concerning the sewage collection and treatment system.   First,
it is assumed that the main intercepting system has the hydraulic capacity
to accept CSO treatment residual wastewater from satellite CSO treatment
facilities.  In reality, the Syracuse intercepting system has little
wet-weather hydraulic capacity, which in itself is a major cause of the
overflows.  Second, it is assumed that once the CSO residual wastewaters
enter the intercepting system, no settling of the heavier particles
occurs in the collection system, i.e., all residual wastes entering the
collection system arrive at Metro.  Third, it is assumed that the flow
of CSO residual wastes occurs over a 24 hr period. This assumption implies
that a solids concentrate holding tank has been constructed at each of the
satellite CSO treatment facilities.   In the actual Syracuse
CSO system, overflows for the storms selected for examination occured
over a time frame of four to six hours.  Last, it is assumed that the
flow of CSO residuals occurs under average DWF conditions.

     Any changes in the assumptions outlined above would require new evalua-
tions as to impact of CSO treatment residuals.  These changes should be
fully re-examined upon selection of a final CSO residual waste handling
alternative.

     Three cases have been considered in this section, as follows:

     1.   Full return of CSO treatment residuals to Metro.

     2.   Return of dilute CSO treatment residuals to Metro.

     3.   Direct transmission of CSO treatment residuals to Metro sludge
          handling facilities.
                                    97

-------
FULL RETURN OF CSO TREATMENT RESIDUALS TO METRO

     Since the characteristics of sludge generated at the Syracuse
demonstration sites were not determined in this study, considerable
reference to published data is made in the analysis of the impact of CSO
sludges on existing treatment facilities (52).

     The total sludge loading from the Syracuse CSO system has been projected
under four hypothetical conditions, assuming complete treatment by each of
the four devices investigated in this program (Zurn, Crane, Sweco, swirl
regulator/concentrator).

     For comparative purposes, the concentration of SS contained in the
effluent from each of the three screening units was estimated by assuming
the SS removal efficiency of each unit to be 40 percent at an influent SS
concentration of 325 mg/1, the time-weighted average influent SS concentration
to the screening units experienced in the demonstration study.  Table 29
presents a preliminary estimate of the quantity of sludge solids that would
be produced in Syracuse assuming full treatment of all significant CSO.  The
estimate does not include provisions for increased solids in the sludge
resulting from chemical addition to. the process wastewaters.

Hydraulic Loading Considerations

     Since the volume of sludge which was returned to the interceptor was
estimated at 3 percent for the Zurn and Crane microscreens and 25 percent
for the Sweco unit during the demonstration study, these figures were used
in developing the sludge quantities and volumes produced by the microscreens.
The average foul sewer fraction returned to the interceptor by the swirl
unit was approximately 30 percent.  However, this percentage is considered
to be higher than is necessary for proper operation of the unit.  For this
reason, the foul sewer fraction used in estimating sludge quantities was
taken as 3 percent for the swirl regulator/concentrator.  The 3 percent
value corresponds to the foul fraction investigated in the'hydraulic model
studies in the LaSalle Hydraulic Laboratory.

     Table 29 illustrates the effect of discharges of CSO treatment sludges
on the hydraulic loading at Metro for each of the hypothetical treatment
systems.  Volumes of overflow for the two storm events considered in this
analysis are equal to 16.7 MG (63,200 cu m) from an average storm of 0.22 in.
(0.56 cm) total rainfall and 83 MG (314,000 cu m) from a 1 year-2 hour storm
which has 1.11 in. (2.82 cm) of total rain.

     Data presented for the average storm in Table 29 indicates that the
hydraulic loading to Metro is increased by about 5 percent when the
Sweco unit is used as the CSO treatment device, and by less than 1
percent when the Zurn, Crane, and swirl units are individually considered
to be the treatment units.  For the CSO treatment conditions presented
for the average storm, no hydraulic overload results from CSO treatment
residual wastes, since Metro was designed to operate as a secondary
                                     98

-------
                                   TABLE 29.  EFFECT OF CSO TREATMENT SLUDGES ON HYDRAULIC LOADING AT METRO STP
CSO Treatment
  Process
                         Average Storm = 0.22 1n. Total Rainfall
                                                                                 1 Year-2 Hour Storm = 1.11 in. Total Rainfall
          Sludge Volume
       Percent of CSO
          to Metro      MGD*
Sludge Volume Plus
    Average DWF     Hydraulic
       MGD          Overload
   Sludge^ Volume
Percent of CSO
   to Metro      MGD*
Sludge Volume Plus
    Average DWF     Hydraulic
       MGD          Overload
Sweco Unit
Zurn Unit
Crane Unit
Swirl Unit
25
3
3
3
4.2
0.5
0.5
0.5
84.2
80.5
80.5
80.5
No
No
No
No
25
3
3
3
20.8
2.5
2.5
2.5
100.8
82.5
82.5
82.5
No
No
No
No
*Based on total CSO concentrate flow volume returned to Metro STP over 24 hr period.
Notes:
     1.
     2.

     3.

     4.
DUF = Dry Weather Flow, average of 80 MGD
CSO Volume Treate'd = 16.7 MG (average storm overflow)
                   = 83 MG (1 year-2 hour storm overflow)
Hydraulic Overload determinations made by comparing Sludge Volume Plus
 Average DWF with maximum design solids loading of 1.5 x DWF solids loading.
Conversions:  MGD x 3.785 = cu ra/day
              in. x 2.54 = cm

-------
treatment facility up to a hydraulic loading rate of 1.5 times the
average DWF hydraulic loading.

     Data for the 1 year - 2 hour storm indicate that the hydraulic
loading would increase by about 26 percent for the Sweco CSO treatment
system and by about 3 percent for the other units considered individually
as the CSO treatment devices.  No hydraulic overload results from CSO
residual wastes for the 1 year - 2 hour storm.

Solids Loading Considerations

     A second factor that must be considered when determining the effects
of CSO treatment residuals is the impact of increased solids loadings on
Metro.  For Metro the design solids loadings is 120,000 Ib/day (54,500
kg/day) to the primary treatment facilities.  For purposes of this analysis,
it is assumed that solids loadings up to 1.5 times the design loading could
be tolerated for short periods of time without drastically upsetting
treatment processes.  It is further assumed that only a small percentage of
the.total SS entering the treatment plant is removed in the aerated grit
facilities.  Since the CSO bleedback solids concentrations are highly vari-
able in nature, the above assumption is considered sufficient for illustra-
tive purposes.

     A solids overload would not result from transmission of CSO treatment
residual wastes from the average storm, but a solids overload would
result at Metro for the 1 year - 2 hour storm.  The unit solids loading to
the primary clarifiers at Metro would increase by around 100% during the
1 year -2 hour storm, to an unacceptably high level greater than 180,000
Ib/day (81,720 kg/day).   The  result  of the  excessive  solids  loading  (greater
than  1.5 times the  design  loading) could  be lowered primary  effluent  quality
and overall  treatment  efficiency.  Adverse  effects of the  solids  overload
would probably be carried  over  to  the  aeration  tanks  and secondary clarifiers.

     Table 30 indicates the effects on Metro of CSO residual solids from
treatment by the various treatment units for the average storm and
1 year - 2 hour  storm.

Organic Loading Considerations

     One of the criteria used in evaluating organic overload is associated
with the activated sludge portion of the sewage treatment process.  In
calculating the BODs characteristics presented in Table 31 for the
specific treatment units studied in this demonstration project (Sweco,
Zurn, Crane and swirl regulator/concentrator,  an initial concentration
of BOD5 in the CSO of 90 mg/1 was used.

      It is assumed,  based on  data  collected during this  demonstration study,.
that  20 percent  of  the BOD5  in  CSO is  removed by the  microscreening  units,
and 40 percent of the  BOD5 is removed  by  the swirl regulator/concentrator.
Use of these estimates results  in  BODs concentration  in  the
                                    . 100

-------
                                                  TABLE 30.   EFFECT OF  CSO  TREATMENT SLUDGES  ON SOLIDS  LOADING AT METRO STP
O

Average Storm » 0.22 in. Total
CSO Sludge
Treatment to
Metro
Process MGD
Sweco Unit
Zurn Unit
Crane Unit
Swirl Unit
Notes:
1. DWF
4.2
0.5
0.5
0.5'
Sol Ids
Percent Dry Solids Removal
Solids Ib/day Percent
0. 075 26, 300 40
0.45 18,800 40
0.45 18,800 40
0.45 18,800 40
= Dry Weather Flow of 80 MGD (average)
2. Average DWF
3. CSO
Treated
solids loading at Metro of 120
= 16.7 MG (average storm flow)
Rainfall



1 Year-2
Hour Storm
Sludge
CSO + DWF Solids
Ib/day
146,300
138,800
' 138,800
138,800
, 120 MGD (maximum
Solids
Overload
No
No
No
No
design)
to Metro
MGD
20.
2.
2.
2.


8
5
5
5

Percent
Solids
0.075
0.45
0.45
0.45

Dry Solids
Ib/day
130,000
93,800
93,800
93,800

=1.11 in. Total Rainfall
Solids
Removal
Percent
40
40
40
40


CSO + DWF Solids
•Ib/day
250,000
213,800
213,800
213,800


Solids
Overload
Yes
Yes
Yes
Yes

,000 Ib/day (average)









                                   = 83.0 MG (1  year-2  hour  storm flow)
                  4.    Solids Overload determination  made  by comparing CSO + DWF Solids  with
                        the design solids loading of  1.5 x design solids  loading =  180,000 Ib/day
                  5.    Conversions:  MGD x 3.785 = cu m/day
                                     Ib/day x 0.454 = kg/day
                                     in.  x 2.54  = cm

-------
                                    TABLE 31.  ORGANIC CHARACTERISTICS (BOD) OF CSO TREATMENT SLUDGES
CSO Treatment
Process
Sweco Unit
Zurn Unit
Crane Unit
Swirl UnU
BOD
mq/1
150
675
675
1250
Notes:
"1. Assumes 20
Average
Vol ume
to Metro
HGD
4.2
0.5
0.5
0.5
percent BOD rj
Storm = 0.22 in. Total Rainfall
BOD
to Metro
1 b/day
5250
2820
2820
5210
=moval by
BOD Removed By
Primary Treatment
1 b/day
1050
560
560
1040
primary treatment at Metro
BOD to
Activated
Sludge
1 b/day
4200
2260
2260
4170
at hydraulic
1 Year-2 Hour Storm = 1.11 in. Total Rainfall
Volume
to Metro
MGD
20.8
2.5
2.5
2.5
loading
BOD
to Metro
1 b/day
26,000
14,100
14,100
26,000

BOD Removed By
Primary Treatment
1 b/day
5200
2820
2820
5200

BOD to
Activated
Sludge
1 b/day
20,800
11,280
11,280
20,800

      rates up to 1050 gpd/ft'
2.   Conversions:  MGD x 3.785 = cu m/day
                   1b/day x 0.454 = kg/day
                   in. x 2.54 = cm

-------
Sweco unit sludge of 150 mg/1, in the Zurn and Crane unit sludges of
675 rng/1, and in the swirl unit sludge of 1250 mg/1.  The calculations for
Table 31 are based on full treatment of all  Syracuse CSO by each of the CSO
treatment processes.

     In general, the analysis of organic loadings to Metro of CSO treatment
residuals indicates that return of residuals to the sewer system during
storm events would be acceptable only when the storm occurs during
average or less than average DWF periods.

     The solids loadings imposed on the secondary treatment units also affects
the overall operation of secondary treatment facilities.  The secondary
clarifiers at Metro are designed for close to the maximum recommended
solids loading of 30 lb/day/ft2 (146 kg/day/m2) (52).  However, sufficient
reserve capacity is available so that, as indicated in Table 32, a
solids overload to the secondary clarifiers would not result from return
of CSO treatment sludge over a 24 hr period for either the average storm
or the 1 year - 2 hr storm.

     Significant impact on the sludge handling facilities is not anticipated
for the situation where CSO residual bleedback is directed to the head of the
treatment plant since the rate of drawoff of the primary and secondary
sludges can be limited to prevent hydraulic overloading of the gravity
thickeners.  The SS loading rate to the gravity thickeners would be
increased by as much as 35 percent for the 1 year-2 hour storm and result in
decreased efficiency in the thickeners, and higher loadings to the digesters.

RETURN OF DILUTE CSO TREATMENT RESIDUALS TO METRO

     An alternative to full return of raw CSO treatment sludges is to
return dilute residuals resulting from on-site dewatering of CSO sludges.
Dewatered sludges in this case would be transported from the CSO treatment
sites to ultimate disposal elsewhere.  Although it is probably uneconomical
to dewater CSO treatment sludges at individual overflow points, calculations
are presented in Table 33 to illustrate the probable impact of on-site
dewatering to reduce the quantity of solids sent to Metro from CSO
treatment facilities.  The data presented in Table 33 are based on the
sludge from the Sweco unit at one percent solids and for the Zurn, Crane
and Swirl units at 10 percent solids.  The concentration of solids in
the Sweco unit sludge prior to dewatering is approximately 750 mg/1 and
it is unlikely that the solids content could be increased to much greater
than one  percent even with thickening and vacuum filtration.

     Table 33 indicates that for the 1 year - 2 hour storm, solids
overload at Metro can be prevented when using the Zurn, Crane or swirl
unit as the CSO treatment-device.  However, the upper limit of permissible
solids loading of 180,000 Ib/day (81,720 kg/day) as defined earlier would
be reached if the Sweco unit were used as the CSO treatment device,
largely as a result of the large volume of dilute residuals, 20 M6D (75.7
cu m/day), transmitted to Metro from the Sweco unit.
                                   103

-------
                                                         TABLE 32.   SOLIDS LOADING TO SECONDARY FACILITIES AT METRO STP
o
Average Storm =
0.22 in. Total Rainfall
Bleedback Solids Removed By Solids to
CSO Treatment Solids to Metro Primary Treatment Activated Sludge
Process Ib/day
Sweco
Unit 26,300
Zurn Unit 18,800
Crane
Swirl
Notes:
1
2

3

4
Unit 18,800
Unit 18,800

Ib/day
15,780
12,400
12,400
12,400

Ib/day
10,520
6,400
6,400
6,400

Solids
1 Year-2 Hour Storm = 1.11 in. Total Rainfall
Solids Removed By
Solids to Metro Primary Treatment
Overload Ib/day Ib/day
No
No
No
No

130,000 78,000
93,000 61,900
93,800 61,900
93,800 16,900

Solids to
Activated Sludge Solids
Ib/day Overload
52,000 No
31 ,900 No
31 ,900 No
31 ,900 No

Bleedback solids obtained from Table 30.
Assumed 66 percent SS removed
Assumed 60 percent SS removed
by primary
by' primary
Solids overload established when the CSO
30 Ib/day/ ft2
Conversions: Ib/day x 0.454 =

kg/day
treatment at overflow
treatment at overflow
rates up
rates up
solids to the activated sludge




to 750 gpd/ft2
to 1050 gpd/ft2
system exceeded







                                     in.  x 2.54 = cm

-------
                     TABLE 33.  EFFECT OF DILUTE EFFLUENT FROM ON-SITE DEWATERING OF CSO SLUDGES TO METRO STP
1 Year-2 Hour Storm - 1.11 in. Total Rainfall
CSO Treatment
Process
Sweco
Zurn
Q Crane
en
Swirl
Notes:
1. Dilute
Dewatered
HGD
20.8
• 2.5
2.5
2.5
Effluent Solids
Dilute Effluent
Flow
HGD
20.0
2.4
2.4
2.4
Concentration based on 1
SS
mg/1
380
521
521
521
percent
Solids
Ib/day
64,300
10,400
10,400
10,400
sludge solid content
CSO Effluent
+• DWF Flow
HGD
100.0
82.4
82.4
82.4
for
Solids
Ib/day
184,300
130,400
130,400
130,400

Solids
Overload
Slight
No
No
No

      Sweco unit and 10 percent sludge solids content for the other three units.
2.   Dilute Effluent flow volume based on 96.0 percent of influent flow to CSO
      treatment facilities returned to Hetro STP for all units.
3.   Calculation of SS concentration in dilute residual bleedback:
          Sweco unit: (20.8 HGD) (750 mg/l)= (0.8 HGD) (10,000 mg/1) +• (20.0 HGD)(c).-. c = 380 mg/1
          Zurn, Crane, Swirl units:  (2.5 HGD)(4500 mg/1) = (0.1 HGD) (100,000 mg/1) +•  (2.4 MGD)(c)  .-.c = 521 mg/1
4.   Conversions:  HGD x 3.785 = cu m/day
                   Ib/day x 0.454 = kg/day
                   In. x 2.54 = cm

-------
DIRECT TRANSMISSION OF CSO TREATMENT RESIDUALS TO METRO SLUDGE HANDLING
FACILITIES

     Although the volume of dry weather residual sludges obtained at a sewage
treatment plant is relatively small, usually 2 to 3 percent of the wastewater
volume treated, sludge handling and disposal is complex, troublesome, and
can represent up to-25' to 50 percent of the capital and operating costs of
a typical sewage treatment plant (52).

     At Metro, sludge handling for ultimate disposal consists of a series
of dewatering steps in which the volume of sludge is progressively reduced
by removal of water associated with  the sludge solids.  The major portion
of water removed is accomplished by gravity thickening.  Further treatment
and sludge volume reduction is obtained by anaerobic digestion in primary and
secondary digesters.   Final discharge of the digested sludge is to sludge
drying beds.

     The sludges discharging to the gravity thickeners consist of combined
sludges from primary treatment and contact stabilization activated sludge
processes.  Detention time in the thickeners is about 7 hours.  The normal
thickened sludge concentration from the gravity thickeners is projected
to be 6 percent solids.

     Further reduction of sludge volume is achieved by passage of the sludge
through high-rate primary digesters and a secondary digester in series.
The solids retention time in the primary digesters is designed at 15 days.
The primary digester underflow concentration is designed to be 4.4 percent
solids (dry basis).  The design retention time in the secondary digester is
4.5 days.  This is expected to produce an underflow solids concentration of
digested sludge of 8 percent (dry basis).

Hydraulic Loading Considerations

     The daily design volume of sewage sludge to the sludge handling facili-
ties at Metro is 2.9 MGD (10,970 m3/day).  Shown in Table 29 are projections
of CSO treatment sludge volumes.  Table 29 indicates that for the average
storm, and with the Sweco unit used as the CSO treatment device, the volume
of CSO sludge is much higher than the design daily dry-weather sludge
anticipated.  For the other treatment devices investigated (Zurn, Crane and
swirl units), the CSO treatment sludge represents a comparatively minor
fraction of the design daily dry-weather sludge.  If the CSO sludge were
transmitted directly to the Metro sludge handling facilities, solids
retention time in the gravity thickeners would be decreased to about 3 hr
for the Sweco unit, and to about 6 hr for the other units.  For the 1 year -
2 hr storm, the retention time would be decreased to less than 1 hr for the
Sweco unit, and to less than 4 hr for the other units.

     The effect of transmitting Sweco unit CSO sludge to the Metro
sludge handling facilities would be an unacceptable hydraulic overload.
Direct transmission of sludge from the other units would result in
severe impacts, which might be tolerated for short periods of time.  The
detention time in the primary and secondary digesters would also be


                                 .    106

-------
shortened as the result of adding CSO treatment sludge.  The hydraulic
overload could be expected to result in greater return of incomplete
digestion products to the head of the treatment plant, adversely affecting
overall plant performance by increasing the total solids loadings entering
the primary and secondary treatment units.

Solids Loading Considerations

     The daily design loading of dry sewage solids to the Metro sludge
handling facilities is 190,000 Ib/day (86,300 kg/day).  Projected CSO
treatment solids (dry weight basis) are presented in Table 33.  For the
average storm, the solids generated by any of the four treatment units
considered would be only a small fraction of the design sewage solids at
Metro.  However, for the 1 year - 2 hr storm, the CSO treatment solids
would be around 50-60 percent of design sewage solids.  The total solids
loading at the Metro sludge handling facilities for the 1 year - 2 hr
storm would be approximately equal to the peak allowable loading.  Excess
loading rates might be experienced if back-to-back overflow events occurred.

Organic and Inert Solids  Considerations

     The organic fraction of Metro sludge has been estimated to be 55 percent
on a dry solids basis.  The design volatile solids loading to the sludge
handling facilities is 104,500 Ib/day (47,400 kg/day).  Projected CSO treat-
ment volatile solids (dry weight basis) are presented in Table 34.  The
results of analysis indicate that for both the average storm and the 1 year -
2 hr storm, the additional volatile solids loading to the Metro primary
digesters resulting from CSO treatment would not cause excessive loading.
                                    107

-------
         TABLE 34.  VOLATILE SOLIDS CONTENT OF SLUDGES FROM VARIOUS CSO TREATMENT DEVICES
Percent
CSO Treatment Volatile
Process Solids

i— »
0
00

Sweco
Zurn
Crane
Swirl
Notes:
1.
56
61
61
45
Conversions:
Averai
Total Solii
Ib/day
26,300
18,800
18,800
18,800
Ib/day x 0.454 =
ae'Storm = 0.22 in. Total
is Volatile Solids
Ib/day
14,730
11,470
11,470
8,460
kg/day
Rainfall
Inert Solids
Ib/day
11,570
7,330
7,330
10,340

1
Total Solids
Ib/day
130,000
93,800
93,800
93,800

Year-2 Hour Storm =1.11
Volatile Solids
Ib/day
72,800
57,200
57,200
42,200

in. Total Rainfall
Inert Solids
Ib/day
57,200
36,600
36,600
51,600

in.  x 2.54 = cm

-------
                                 SECTION 11

                  DISINFECTION-RESULTS AND DISCUSSION
GENERAL
     One of the major objectives of this project was to determine the
feasibility of high-rate disinfection of CSO following some level of
solids removal.  Owing to the short distances and limited spaces for
construction between overflow structures and the points of discharge,
and the high-volume rapid flows associated with the majority of over-
flows in the City of Syracuse, contact times of one minute were investi-
gated.

     Studies had previously been conducted on a bench-scale level to
optimize dosages required for adequate disinfection of CSO and to aid in
determining the various design parameters for full-scale prototype
facilities (7).  The results of the bench-scale work are summarized in
Section 5 of this report, and a description of the facilities constructed
appears in Section 6.

     The facilities at both the Maltbie Street and West Newell Street
sites included instrumentation for automatic feed of disinfectant, such
that a constant dosage of disinfectant would be applied to the treated
CSO even under varying flowrates.  However, a number of problems arose
through both the limitation of storm events (overflow would not occur
unless there was a severe storm) and malfunctions of disinfection
equipment, which prevented significant disinfection investigations at
the West Newell Street site.  'Notation of these problems is given in
Section 6.

     All the findings and results presented in this section relate to
disinfection performance of Cl2 and C102 at the Maltbie Street site
following screening.  The operation schedule for disinfection at Maltbie
Street is given in Table 35.

C102 DISINFECTION - 1975 OPERATIONS

     Eight storms were evaluated for reduction of fecal coliform (FC)
levels by C102 disinfection on the Swe'co treatment system and four
storms were evaluated on the Zurn system. C10_2 generating equipment
problems on the Zurn process train resulted in four of the eight storms
not being evaluated for that unit.  Various back pressure valves and pump
diaphrams failed, limiting use of this C102 generator.  Figures 38 through
                                   109

-------
i
                                                                        TABLE 35.  HALTBIE STREET OPERATION SCHEDULE - DISINFECTION
Avg. Screen Loading
Overflow Rate (qpm/ft?)

2-75
3-75
4-75
5-75
6-75
7-75
8-75
10-75
11-75
16-75
1-76
2-76
3-76
4-76
5-76
6-76
Sweco
40
36
32
28
43
33
66
-
50
11

-
-
_
-

Crane Zurn
12
11
7
9
10
8
4
12
9
2 3
- no screen
- no screen
8 no screen
8 no screen
8 no screen
no screen
Screen Aperture
Size (microns)
Sweco
105
105
105
105
105
105
105
-
105
105

-
-
_
-

Crane Zurn
71
71
71
71
71
71
71
71
71
23 71
no screen
no screen
no screen
23 no screen
23 no screen
no screen
Cl2
Tank
#1
0
0
0
0
0
0
0
0
0
0

-
-
.
-

Dosage (mg/1)
Tank
#2
0
0
0
0
0
0
0
0
0
0

-
12
0
8

Tank
#3
0
0
0
0
0
0
0
0
0
0
9-24
12
0
12
8
7.3
C102
Tank
#1
1-6
0-6
0
7-8
5-7
7-8
3-4
4-9
5-6
0

-
-
-
-

Dosage (mq/1)
Tank
K
0
0
0
0
0
0
0
0
0
0
.
-
0
3.4
2

Tank
#3
0
0-4
0
3-5
5-7
0
3-10
0
0
0
0
0
11.0
0
2
0
Mixing
Tank
#1
F
F
-
F
F
F
F
F
F
-

-
-
-
-

Tank
K
.
-
-
-
.
-
-
-
-
-
.
-
F
F
F

Tank
#3
.
SF
-
SF
SF
-
SF
-
-
-
SF
SF
SF
SF
SF
SF
                           F - Single flash mixing
                          SF - Sequential flash mixing

-------
45 relate to the Sweco disinfection system results and Figures 46 through
49 relate to the Zurn disinfection system results.  The disinfection results
were plotted on log-normal paper to facilitate evaluation of the bacterial
kills for various C102 dosages and one minute detention time.

     Figure 38 indicates FC kills of 1 to 3 logs at C102 dosages of 3 to
7 mg/1.  The results presented in Figure 39 indicate that at dosages of
less than 1 mg/1, no reduction of FC populations were achieved; however,
as the dosage increased from less than 1 mg/1 to 4 to 6 mg/1, the reduction
increased to 1 to 2 logs.  Figure 40 indicates 4 log reduction of FC at
C102 dosages of 7 to 8 mg/1.  During this storm, the actual FC populations
were reduced to less than 10 counts/100 ml, well below the desired level
of 200 counts/100 ml as discussed in reference (7). Figure 41  exhibits
2 to 5 log reductions at dosages of 5  to 7 mg/1 C102-  The lower reductions
for this storm (in the order of 2 logs) were observed during the first hour,
indicating that an increased C102 demand in the first flush adversely
affected disinfection.  Figure 42 indicates 2 to 3 log reductions of FC
achieved at C102 dosages of 3 to 4 mg/1 with the higher levels of reduction
achieved during the later stages of the overflow.  Figure 43 illustrates
2 to 3 log reductions achieved at C102 dosages of 3.5 to 6.0 mg/1 C102»
with the standard of 200 counts FC/100 ml being achieved during the
latter stages of the overflow.

     Figure 44 shows 3 to 4 log reductions of FC being achieved throughout
the overflow after the first 30 min.  Initial dosages of 7.7 to 8.7
mg/1 of C102 reduced FC populations to less, than 10 counts/100 ml.'  This
level of reduction was maintained through the remainder of the storm at
reduced C102 dosages of 4.4 to 5.0 mg/1.  Figure 45 exhibits 2 to 3 log
reductions of FC at dosages of 5.4 to 5.9 mg/1.  Again, the better
reductions were achieved during the latter stages of the storm overflow
and were sufficient to reach the desired level of 200 counts/100 ml.

     Data collected on the disinfection system associated with the Sweco
unit indicate that the desired level of 200 counts FC/100 ml in the
effluent was reached in all storms when greater than 4 mg/1 C102 was
injected to the system after the first 30 to 45 min of the overflow.
During the first 30 to 45 min, dosages of 7 to 8 mg/1 were required to
reach the target level of 200 counts/100 ml as indicated in Figure  40.
The disinfectant demand is apparently higher at the beginning of the
overflow as a result of the first-flush phenomenon, thus requiring higher
dosages to achieve a given level of reduction.  Also, an initial period
of up to 15 min was required for the C102 pumping system to stabilize in
its delivery of C102 to the disinfection tank.  In part, this may have
been due to the time necessary for the C102 piping system to fill upon
activation of the C102 generator.  In addition, variable back pressure
created by the rise and fall of the liquid level in the disinfection
tanks at variable CSO fTowrates may have resulted in somewhat erratic
delivery of the C102-  C102 delivery rates would be more consistent
under constant CSO flow conditions, resulting in more consistent FC
kills.  This condition is partially indicated by comparing log reductions
of Storms 2 and 3 (Figures 38 and 39 where CSO flowrates are variable)  .
to the remainder of the storms where more constant CSO flowrates were

                                    111

-------
                                                     _  0
                   1200     1300
                     TIME
  FIGURE  38.   CIO? Disinfection
  Following  Sweco Storn  2,  1975
iJ  '«

§

O rf
                           Jlnfluwir FC Livil
                           FC Uvtl Alter Oiimrtction
                          D.T. = I Ulnult
                          Flash Hiiinij
                          Scrt«n Optrarur* • 105 ft
                    '   0230    OSOO   053O
   FIGURE 40.    C102  Disinfection
   Following  Sweco Storm 5,  1975
                                                                                Influent FC L*v«l
                                                                                  Ltv«  After OiWitKtion
                                                                              O.T. = I Minute
                                                                              Flatn J/innq
                                                                              Scrtin Apvaturt * 105 it
                                                         430    ISOO     I93O    1600    1630     rTOO    1730
FIQURE  39.   C102 Disinfection
Following Sweco  Storm  3, 1975
                           Inliutnt FC L*vd
                             vtl Atwr Ownfcction
                         DT 
    A,—I	1	1	1	'  •*>!  '	1	L
    17    IM7    MIT     1447    BIT
                                                                                  BTT     B4T
FIGURE 41.   C102  Disinfection
Following  Sweco Storm 6,  1975
                                               112

-------
!
                          ^Influent PC Level
                          AFC Level AMir Disinfection
                          OT = I Mmule
                          Floih Muing
                          Scr*«n Apwaiur* • |OS p
    0900    09»
                 0930    0945
                      TIME
                              •000    K3I5     KSO
  FIGURE  42.   ClOg  Disinfection
  Following  Sweco Storm 8,  1975
                          A FC Lnal Aft>
 C
                                                                         55      57
0 Influent FCLneli
A PC Uvtl AM(* Citmfietion
O.T : I Minul*
FlOih Minng
Strun Ap«ra1urt * 103 }i
        21 29    21 39    21-49    21 39    22'C9   22*19
                       TIWC .

FIGURE 45.    C102  Disinfection
Following  Sweco Storm  11,  1975
                                                 113

-------
applied.  FC kills varied throughout Storms 2 and 3 from 0 to 3 logs
when flowrates varied from 0.3 to 2.1 MGD (0.8 to 5.5 cu m/min) within
individual storms.  For the remainder of the storms, flowrates were held
relatively constant throughout individual storms.  FC kills are seen to
be less variable within individual storms as seen in Figures 40 through
45.

     The results of the 1975 storms utilizing the disinfection system
associated with the Zurn unit are presented in Figures 46 to 49.

     Figure 46 indicates no significant reduction of FC populations
at C102 dosages of less than 1 mg/1.  Reductions of about 1 log were
achieved at a dosage of 2.4 mg/1.  Figure 47 depicts rather erratic
results.  During the first hour of the storm flow, over 4 log reductions
were achieved at C102 dosages of 3.7 mg/1 while the last one and one-
half hours shows only 1 to 2 log reductions at a C102 dosage of from
3.7 to 5 mg/1.  Since parallel data for the Sweco disinfection system
(Figure 39) illustrated lower levels of reduction during initial  stages
of the storm, the data presented in Figure 47 indicates that contact
with the disinfectant may have continued after the samples were drawn.

     Figure 48 indicates 4 log reductions of FC achieved at C102 dosages
of 6 to 7 mg/1 after the first 30 min of the overflow.  During the first
30 min dosages near 5.5 mg/1 resulted in only 1 to 2 log reductions.
This data indicates the time required for C102 delivery rates to stabilize
and/or the effects of first-flush effects on disinfection.

     Figure 49 illustrates FC kills ranging from 1 to 5 logs at C102
dosages varying from 3 to 10 mg/1; the higher kills being achieved at
the higher dosages.  Lower C102 dosages of 3 to 6 mg/1 applied during
the last half of the storm achieved nearly the same log reductions as
the higher dosages of 6 to 10 mg/1 achieved during the first half of the
storm.  Again higher pollutant loadings occurring during the initial
periods of the storm could account for the higher dosage requirements
during initial storm periods.

     Overall, data collected during 1975 on the Zurn disinfection system
indicate much the same trends as did the Sweco disinfection system.
Target levels of 200 counts FC/100 ml were achieved at dosages of 7 to 8
mg/1 C102 at the beginning of the overflow period, while lower dosages
at later stages'of the overflows achieved the same results.

     At no time during 1975 was the Crane system investigated with
respect to disinfection.  This was due to the continuing failure of the
Crane Microstrainer to operate properly.

Cl2 AND C102 DISINFECTION - 1976 OPERATIONS

Description

     Six additional overflow events were monitored in 1976 at Maltbie
Street to further examine high-rate disinfection of CSO.  Since consider-

                                     ,114

-------
   0-
CT  5-
1  ,0.
S
s
s *>>•
                           4 taffutflt FC L«v«U

                           A FC Level Aflir Disinfectio
                           O.T. = I M.nute
                           Seqgennal FloihMixin?

                           Scntn -Apvaturt «
   I4-3O    1500
  FIGURE  46.   C102  Disinfection
  Following  Zurn Storm  3,  1975

!"
s

                      5lnfluini FC Level
                      FC Level After Otimfection
                     D.T, = | Minute
                     Sequential Ftaih Mliinq
                     Sa**n  Aptranr* • 71 p
                       14.47 ~ 73'IT
 FIGURE 48.    C102  Disinfection
 Following  Zurn Storm 6,  1975
                 I5=3O    16-00    16-30   17^00    \T 30
                      TIME
                                           I&IT
JInfluent FC Uvd
                                                                              FC Level After
                                                                             D.T. = | Minult
                                                                             Sequvttd Floih M.iinq
                                                                                ApwoTur* » 71 u
                                                           Ol'QO    01-30    02-00   C2 3O    OToQ0373O   D*-00
                                                      FIGURE 47.   C102 Disinfection
                                                      Following Zurn  Storm  5, 1975
                                                          09-00   09--I5    09-30    09<45   10 CO    10 6    10=30
                                                      FIGURE  49.   C102  Disinfection
                                                      Following Zurn  Storm  8,  1975
                                               115

-------
able data had been collected relative to the solids removal capabilities
of the screening units at this site and since budget constraints limited
the number of laboratory analyses to be performed, no effort was directed
to assessing the Sweco (105y) and Zurn (71y) microscreens during these six
events, although solids removals by the Crane (23y) unit were evaluated
for two storms in 1976, to supplement data gathered for one storm in 1975.

     Since the main effort of the 1976 program was to further evaluate
the feasibility of high-rate disinfection, it was desired to remove as
many operational variables as possible to produce a more controlled
process train.  The most important of these variables was that of
flowrate through the treatment processes.  Therefore, in all six events
monitored in 1976, the flowrates were fixed.  Excess CSO above the fixed
rate was allowed to bypass directly to Onondaga Creek.

     Processes involving unscreened CSO were performed by pumping CSO
from the wet well through the Zurn Micromatic, from which all screen
panels had been removed.  The screened processes were accomplished by
passing CSO through the Crane Microstrainer.

     During Storms 1,2,3,4 and 6 the disinfectant was injected at the
upstream end of each disinfection tank, thus providing a one minute
detention time through the tank.  During Storm 5, C102 was injected at
the upstream end of the disinfection tank and Cl2 was subsequently
injected at the end of the first baffle in the tank.  This sequence of
injection thus provided a theoretical one minute detention time for C102
and a 40-45 sec detention time for Cl2.  Immediately after injection of
C102» the C102 begins to deteriorate into the C102" ion which is far less
potent as a disinfectant than is the C102 molecule.  It has been theorized
that addition of Cl2 at some point subsequent to the addition of C102
converts ClOjj back to the more potent C102 molecule.  By sequential
addition of C102 and Cl2 it was anticipated that the disinfection capa-
bility would be enhanced. (7).

     Flash mixing was provided at the point of injection of each disin-
fectant on the unscreened CSO, and flash mixing of only C102 at the
point of injection was provided on the screened CSO.  Analyses for the  •
reduction of bacterial populations were limited to analysis for FC for
the entire duration of each overflow event.  However, viral studies were
conducted during which- grab samples were collected at 2 min time
intervals.  These grab samples were analyzed for TC, FC and FS.   Other
parameters analyzed included TOC, SS, TKN, NHsN and pH. Chlorine demand
was determined from laboratory analyses of composite samples.

Results of 1976 Disinfection Tests

     Figures 50 through 58 are plots of the results of disinfection for
the six storms monitored in 1976.

     During Storm 1, as shown in Figure 50, Cl2 dosages ranged from 0
to 24 mg/1.   Sequential flash mixing of the unscreened CSO was provided
at the upstream end of the tank and at the end of the first longitudinal
baffle.   Reductions of FC ranged from 1 to 6 logs.

                                     116

-------
                                   I9'30   20 00
                      TIME
  FIGURE 50.   Cl2 Disinfection
  Unscreened  CSO -  Storm  1,  1976
u  10 3
8  '^
« o5
                         9 IMIutm FC Ltvtl
                         A FC L«vcl Atttr Oiiinf.elio
                         OT = I M«JI«
                         Stcwtnliol FlOfh Mixing
                      I6>30

                      TIME
  FIGURE 52.   C102  Disinfection
  Unscreened CSO -  Storm 3,  1975
                                                       I02
                                                         .
                                                          ! 1 ,
                                                         .'1 i i
                                                         . . 1 \
                                                         -' ' i 1
                                                         1 ' • ^    .
                                                         '
                         • InllLitnt FC uxl
                         A Fc I-***1 Af '«r Otifl^
                         D.T. i 1 Ujniii
                         Stqutntiol Floth Mixing
                         uncmnM SonxKn
                                                                  ;/\ A;
                                                                  i V  '
                                                                      \
                                                                       1    /  /  1

                                                                      1  U^'  i4
   1300   13-30    i4-00    14.30   15-00   I5-3O    I6OO
                       TIME
FIGURE  51.   C12  Disinfection
Unscreened  CSO -  Storm 2;  1976
                                                      o3
                         0 In fluent FC Uv«r


                         OT: IMmutt
                                                                             Scrttn Aoiratur* • 23 p
                      I&30
                    TIME
FIGURE  53.   Cl2  Disinfection
Screened CSO (23u) -  Storm 3,  1976
                                              117

-------
                           Influent FC Lcvil
                           FC Level Afrer Disinfecta
                           ' s I Minute
                          Sequential Flash Mrxfng
                          Uraa'tentd Samptes
 I5-3O   1600    1630   1700
                   TIME
                           7*30    18 00    18 30
FIGURE  54.   Cl2 Disinfection
Unscreened  CSO  -  Storm 4,  1976
                         lr.1ii.enl FC Levels
                        A FC Level* Aim Oi
                        DT : 1 Mrnule
                          Mixing
                        Scr**n Apgrorure * 23
 0915    03^30    OS'45    10 00    1015    10 50    10-45
   15=30    I&OO    i6«3O    1700    (7=30    18*00   18 30-
                       T1ME
FIGURE  55.  'C102 Disinfection
Screened CSO  (23u)  - Storm 4,  1976
                                                      10'-
                                                             CiOz D05E = Zmg/l DT -60 tec
                                                               DOSE>emg/l OJ-46«cc
                         0 Inllutnt FC Livili
                         A FC Level JUitf Oomfcelio
                         DT t IMinuie
                         &«quentrol Ftoth Unwig
                                                                          10 00
                                                                        TIME
 FIGURE 56.    C102  & Cl2 Sequential           FIGURE 57.   C102  & Clz Seouential
 Disinfection  Screened CSO (23vO  -          Disinfection  Unscreened CSO -
 Storm 5,  1976                             110  Storm  5,  1976
                                              118

-------
     Figure 51 illustrates the log reductions of FC during Storm 2.  The
Cl2 dose during Storm 2 was set at 12 mg/1 with resulting Cl2 residuals
ranging from 5.0 to 9.8 mg/1 for the one minute of contact time available.
Figure 51 indicates a general tendency of increased log reduction of FC
with time.  A 3 to 4 log reduction of FC for Storm 2 was observed, and the
target level of 200 counts/100 ml was attained only during the latter stages
of the storm.

     Figures 52 and 53 illustrate the results of disinfection tests by
C102 on unscreened CSO and Cl2 on screened (23u) CSO, respectively, during
Storm 3.  The CIO;? test shown in Figure 52 shows an increase in the log
reduction of FC with time ranging from 1 log reduction near the beginning
of the storm to 6 log reduction at the end of the storm.

     Figure 53 illustrates the reduction of FC by application of a 12 mg/1
dose of Cl2 to the screened (23y) CSO.  The log reduction tended to increase
slightly with time from 1 to 2 logs at  the beginning of the storm to 2 to
3 logs near the end of the storm.

     The results of Storm 4 are depicted in Figures 54 and 55.  A 12 mg/1
dose of Cl2 to unscreened CSO produced slightly erratic reductions ranging
from 2 to 4 logs as shown in Figure 54.  The majority of points indicating
numbers of FC in the effluent fall below the target level of 200 counts/100
ml.  Cl2 residuals during this storm were measured at 5 to 8 mg/1.

     C102 applied at a dose of 3.4 mg/1 to the screened (23y) CSO in Storm 4
(Figure 55) resulted in significant log reductions of FC.  All values of FC
in the disinfected CSO were measured to be less than 1000 counts/100 ml.
The Cl2 demand for the CSO was determined to be approximately 13 mg/1.

     For Storm 5, raw CSO FC populations were determined to be 380,000
counts/100 ml while populations after screening (23y) were 312,000 counts/
100 ml.  Since this reduction is within the variability of the analysis
procedure (estimated to be 20 percent), no conclusion can be reached regarding
bacterial reduction as the result of microscreening.

     Storm 5 was conducted employing sequential addition of C102 and Cl2-
Two mg/1 of C102 was applied at the head of the disinfection tank, and
8 mg/1 of Cl2 applied at the end of the first baffle in the tank.  In the
case of both screened and unscreened CSO, the level of reduction of FC was
in the order of 3 logs as indicated in Figures 56 and 57.  The magnitude
of the Cl2 residual measured in each test was 10 to 13 mg/1 and 5 to 9 mg/1,
respectively.


     Figure 58 illustrates the results of Storm 6 where a C102 dose of 7.3
mg/1 was injected into unscreened CSO.  During the middle portion of the
storm, the FC populations were reduced to less than 100 counts/100 ml with a
Cl2 residual of 4.2 to 4.5 mg/1.  Reductions of FC were in the order of 3
logs.
                                      119

-------
 MULTIPLE REGRESSION ANALYSIS

      To evaluate the effects of variable wastewater  quality on the disin-
 fection processes investigated during the demonstration  study, mathematical
 models were developed relating specific water quality  parameters to the
 reduction of FC levels achieved under various disinfectant dosage applica-
 tions.  Multiple regression  analysis of the treatment  data was performed to
 statistically fit equations  to the experimental data.  The final equations
 fitted to the results take into account the varying  levels of FC, SS, pH, etc.
 that resulted from the microscreening processes.
                     I
                                           Influent FC Level
                                           FC Level Alter CismlKtifln
                                          O.T s I Minulf
                                          Sequential FIQth Mixing
                                          UlYKrttrMd Scrnc4e»
                       03'25   09'35   09"»5  09'55   1005   10-15   iO-25
                                      TIME

                      FIGURE 58.   C102  Unscreened CSO
                      Storm  6,  1976


     The final  form of"the regression equation  determined to be most repre-
sentative of the  observed data for the disinfection  investigations was:
                   K2 Ko  K4  K5  K6 PH
     logkill =  KjC  GT   SS  FC  10

     where      C  = concentration of disinfectant, mg/1

              SS  = concentration of SS, mg/1

              FC  = influent level of fecal coliform,  counts/100 ml

              pH  = pH
                                      120

-------
               GT = mixing intensity x detention 'time in zone of influence

               Ki,K2sl<3,K4,K5,K6 = constants from the regression analysis

     The velocity gradient G for the flash mixers was calculated from the
following equation:

               G = (550P/Vu)^

     where     P= water horsepower of the mixers, HP
               V= volume of the zone of influence of the mixer,
               U= kinematic viscosity of water at 50°F, 2.73 x 105 lb-sec/ft2

     For the disinfection tank mixing influence, the G value was calculated
from the following equation:

               G= 1730(u)"% (VS)%

     where     u= viscosity of water at 50°F, = 1.3097cp
               V= velocity of flow, fps
               S= tank slope, ft/ft

     Previous research ( 53) has indicated that disinfectant concentration
is exponentially related to reduction of bacteria.  This relationship
resulted in the inclusion of the factor Ck2 in the regression model.

     To develop the mathematical relationship between kill and dosage, the
SS, FC and pH parameters were included since they indicated statistically
significant effects with the disinfection system performance data.
Disinfection contact time was held constant at one minute throughout the
demonstration study.  Therefore, although it is one of the major factors
affecting disinfection processes, contact time was not included in the
development of the regression equation.

     Tables 36 and 37 present the results obtained from the regression
analyses for the C102 and Cl2 disinfection systems, respectively.  The
regression coefficient values correspond to the exponential K values in
the regression equation.  The value of KI is equal to Id"" where i is the
regression intercept value.

     The magnitude of the regression coefficient gives some indication
of the relative importance of the term in the regression expression; for
example, positive coefficients associated with C102 dosage application,
mixing intensity (GT), pH and influent FC levels signify that as these
values increase, the log kill of FC also increases.  The negative coef-
ficient associated with SS indicates that as this value increases, the
log kill of FC decreases.

     The statistically derived 't1 test of significance designates the
degree of confidence with which the corresponding regression coefficients
can be assumed to be correct.  In the C102 regression results, the 't1
value for the C102 dose coefficient corresponds to a degree of confidence

                                    121

-------
                            TABLE 36.  MULTIPLE REGRESSION ANALYSIS RESULTS  FOR  C102
PO

Independent
Variable
Log C
Log FC
Log Gt
Log SS
PH
Mean
0.72
5.61
3.61
2.26
6.63
Standard
Deviation
0.23
0.89
0.25
0.41
0.50
Correlation
X vs Y
0.65
0.16
0.38
-0.05
-0.01
Regression
Coefficient
0.68
0.06
0.09
-0.07
0.02
Std. Error of
Regr. Coef.
0.08
0.03
0.08
0.05
0.04
Computed
t- Value
8.99
2.02
1.18
- 1.36
0.57
Dependent Variable
      Log (Log kill)    0.39     0.27
Intercept -0.71
Multiple Correlation 0.67
Std. Error of Estimate 0.20
Source of Variation
Attributable to Regression
Deviation from Regression
Analysis of Variance
-Degrees of
Freedom
5
157
for the Regression
Sum of
Squares
5.21 1.
6.30 o.
Mean
Squares
04
04
F Value
25.98
      Total
162
11.52

-------
ro
                             TABLE 37.  MULTIPLE REGRESSION ANALYSIS RESULTS FOR C12

Independent
Variable
Log C
Log FC
Log Gt
Log SS
PH
-
Mean
1.03
5.88
3.53
2.26
7.20
Standard
Deviation
0.61
0.66
0.12
0.39
0.73
Correlation
X vs Y
0.86
0.24
0.05
0.06
-0.25
Regression
Coefficient
0.36
0.02 .
0.42
-0.07
-0.03
Std. Error of
Regr. Coef.
0.02
0.02
0.12
0.03
0.02
Computed
t-Value
17.29
0.99
3.45
-2.05
-1.47
Dependent Variable ,
      Log (Log Kill)  0,51
0.25
      Intercept                -1.09
      Multiple Correlation      0.90
      Std.  Error of Estimate    0.11.
                                     Analysis of Variance for-the Regression
Source of Variation
                                              Degrees of
                                               Freedom
                             Sum of
                             Squares
                Mean
               Squares
F Value
Attributable to Regression
Deviation from Regression
                                                  5
                                                 75
                             3.99
                             0.90
               0.80
               0.01
66.43
      Total
                 80
4.89

-------
greater than 99 percent,  while the 't1 value  for the FC coefficient  was
greater than 95 percent,  6T greater than 75 percent, SS coefficient  greater
than  80 percent and that  for pH greater than  40  percent.  The magnitude
of  the standard error of  the regression coefficient and the  't1 value
associated with the pH  indicate that the effect  of pH was fairly  insignificant.

      In the Cl2 regression  results, 't1 values for the various para-
meters indicated a degree of confidence greater  than 99 percent for  Cl2
dose  and GT coefficients, greater than 95 percent for SS, greater than
85  percent for pH, and  greater than 70 percent for FC.

      The 'F'  value in the multiple regression analysis gives an indication
of  the validity of the  entire regression equation.   The statistically derived
'F1 test for equality of  data variances conducted for each of the C102  and
Cl2 equations represented a degree of confidence greater than 99  percent.
The final  regression equations obtained were  as  follows:

                     0.68  0.90  -0.07  0.06   -0.02pH
C102:  Logkill=0.19 C    GT     SS     FC    10

                    0.36   0.42  -0.07  0.02  -O.OSpH
    =  Logkill=0.08 C     GT    SS     FC    10
Ref, 27 ) Dt.= IMin	~)
iFFLES (Re(39) O.T. = 4Min	 lCla
DISINFECTION (RH39; O.T.= 30Min. 3
                    • BEAKER TESTS (Rl
                    • CORRUGATED BftFFLES (Ri

                    A CONVENTIONAL DISH
                  	REGRESSION MODSL ROCHESTER CSO(.1efS4) O.T. • I Mm.
                  	REGRESSION MODEL SYRACUSE CSO (Thit SludylDT. = I Mm.
                 Racidiiol s Sme/l
                                                          8mg/l CI2
                                 100
                                        1,000
                                                 10,000
                                                          100,000
                                                                  1,000,000
                                            GT
                   FIGURE 59.
Regression  Model  Results-GT
vs FC Reduction
Illustrative Trends of the Regression Models

      The  separate effects of the  independent variables  on the disinfection
unit performance were examined  by  use of the regression  models.  Varia-
tions in performance with respect  to mixing intensity  as supplied by the

                                      124

-------
flash mixers and detention time were plotted for both Cl2 and C102-  An
effort to compare the Syracuse model results with the results Glover obtained
(39) using a Cl2 residual of 5 mg/1 was attempted by applying a Cl2 dosage
of 8 mg/1 and a C102 dosage of  4 mg/1 in the  regression models.  Results
were also compared to the results obtained in a similar study in Rochester,
(54) using equivalent dosage applications.  Values for the parameters
contained in the regression models were the averages experienced in the
Syracuse study.  The effects of flash mixing of disinfectants with the waste-
water is reflected in the magnitude of  the mixing intensity, 6T.

     Plots of Glover's results and the regression model results are
presented in Figure 59.  A comparison of the curves shows similar
trends.  The slope of the curves indicates that disinfection with Cl2
is greatly enhanced with an increase in mixing intensity, while the
curves for C102 indicate that increased mixing intensities do not result
in as pronounced an increase in bacterial reductions as observed for
Cl2-  Figure 59 also suggests that at low mixing intensitites and short
contact times, C102 is more effective than Cl2 in reducing bacterial
populations.

     Figure 60 is a plot of performance versus GT for different dosages
of Cl2 using the average parameter values from the Syracuse data.  The
effects of mixing intensity on Cl2 disinfection effectiveness are apparent.
The slope of "the curves indicate that the effect on performance is more
pronounced when dosages are varied at higher mixing intensities.

     A plot of performance versus GT for various C102 dosages is presented
in Figure 61.  The slopes of the curves indicate that mixing intensity
does not affect bacterial reduction when C102 is used as significantly as
mixing affects reduction of bacteria when using Cl2-  However, as in
the case of Cl2j the effects of mixing on performance are more pronounced
when the C102 dosage is varied at the higher mixing intensities.  Comparison
of Figures 60 and 61 shows C102 to be a better disinfectant than Cl2
at the lower mixing intensitites.

     The effect of varying SS levels was evaluated using the regression
models.  Figure 62 presents plots of performance versus disinfectant
dosage for both Cl2 and C102-  Comparison of the plots show that variations
in the SS levels of the applied wastewater produce relatively minor
improvement in the disinfection effectiveness of both Cl2 and C102-

     A similar sensitivity analysis was conducted for the FC levels with
results presented in Figure 63.  The set of curves indicate that the
effect of FC levels in the applied wastewater on the disinfection effec-
tiveness of Cl2 is relatively insignificant, while the effect is slightly
more pronounced for C102.   The effect on C102 is probably insignificant
from a practical standpoint since FC levels in CSO are normally In the
range of 105 to 106 counts/100 ml with only occasional FC levels in the
order of 10? counts/100 ml.

     Sensitivity analysis of the effects of pH as performed with the
regression models and illustrated in Figure 64, indicate that pH variation
is insignificant when C102 is used, while use of Cl2 results in a

                                    125

-------
       D.T. = IMin
       SS- 200mg/l
       FC - 500,000 counts/100 ml
    Q  - Cl£ Dose, os shown, mg/l
       pH -6.8
           NOTE- All Syracuse Data Included
§   4
LU
CC
o   2
                          10
                                     100
                                    GT
10,000      100,000
FIGURE  60.   Effect  of  C12  Dosage  - GT vs FC Reduction
       DT. = 1 Mm
       SS-ZOO mg/l
       FC-500,000 counls/IOOml
     _ CL02 Dose,-os shown, mg/l
                                                           CI02 DOSE
             NOTE' All Syracuse Data Included
                                     1,000
10,000
                                                          100,000
                            GT
FIGURE  61.    Effect  of  C102  Dosage  -  GT  vs  FC  Reduction
                                  126

-------
              D T   - I mm
              FC   • 500,000 counts/100 ml
              SS   - As Shown, mg/l
              GT   - 3000
              pH   -63
                  NOTE- All Syracuse Data Included
                       D T  -  I mm
                       FC   •  500,000 coonts/lOOml
                       SS   -  As Shown, mg/l
                       6T   -  3000
                       PH   -  68
                            NOTE' All Syracuse Data Included
                     4
                     CI2
6    8    10
DOSE, mg/l
     24    6    8    10
     CI02  DOSE, mg/l
                                                                             12
FIGURE  62.    Effect of  SS  on Clg  &  C102  -Dosage  vs FC  Reduction
      £  5
      fe
         4
      z
      o
      I'
              D T  -I mm
              FC  •  As Shown, counts/100 ml
              SS  -  ZOO mg/l
              GT  -  3 000
              pH  -  68
                   NOTE' AD Syracuse Data Included
SS
GT
pM
                            •  As Shown, counts/ 100 ml
                            •  2DOitig.|
                            -  3 000
                            -  68
                            NOTE- All Syracuse Data Included
                     4    6    8   10
                     CI2  ' DOSE, mg/l
                            24    6    8    10
                            CIO 2  DOSE, mg/l
                                                                            12
FIGURE  63.   Effect  of FC on  Cl2  &  C102  -  Dosage  vs  FC Reduction
                                           127

-------
slightly-more pronounced effect on log kill with varying pH.  This
observation is partially supported by the fact that a greater proportion
of the C12 disinfectant exists as the more potent HOC1 at lower values of
pH (55).  At higher pH, more of the Cl2 exists as the less potent OC1".
With C102, the most potent form of the disinfectant is C102 with the
less potent C102 formed as the result of reactions with reducing agents,
which are much less dependent on pH.

     Figures 65 and 66 indicate the correlation between observed data
and performance predicted by the regression models.

Sequential Addition of C102 and Cl2

     During Storm 5 in 1976, investigations were conducted into the
feasibility of disinfecting CSO by a process of sequential addition of
C102 and Cl2-  For these tests, C102 was injected into the CSO at the
upstream end of each of two disinfection tanks at a dosage of 2 mg/1.
Cl2 was injected at a dosage of 8 mg/1 at the end of the first baffle in
each tank.  The wastewater was thus subjected to a C102 dosage alone for
a detention time of 15 seconds and to a combination of C102 and Cl2
for a detention time of 45 seconds.   It had been suggested that C12
added 15 to 30 seconds after injection of C102 would enhance disinfection.
C102 is oxidized by various reducing agents to C102 during the disin-
fection process.  It was suggested that addition of Cl2 would, to some
degree, reduce C102 back to C102 to prolong  the existence of the more
potent disinfectant C1025 and thus enhance disinfection beyond that
expected by the sum of the respective concentrations of Cl2 and C102-

     Effluent from the Crane Microstrainer was discharged to one of the
disinfection tanks, while unscreened CSO was discharged to the second
tank. Mixing of the screened CSO was accomplished by flash mixing at the
C102 injection point,  and mixing of the unscreened CSO was accomplished
by flash mixing at both disinfectant injection points.

     From multiple regression analysis, a mathematical expression was
developed to evaluate the effects of sequential addition of disinfectants
for the specific combination of Cl2 and C102 that was applied.  The final
form of the regression model was


     logkill=0.84FC°-10GT0-008
    -The  "t1 values associated with  the regression coefficients corresponded
to a degree of confidence for the  FC coefficient of greater than 99
percent,  and for the GT coefficient slightly over 10 percent.   The 'F'
value indicating the overall statistical significance of the regression
equation resulted in a degree of confidence greater than 99 percent.  It
would be expected that with additional testing over a range of dosage
applications, GT values and C102: Cl2 ratios, a more precise regression
equation would result.
                                     128

-------
              o
              z
              O
              UJ
              cc
   D T  - 1 mm
   FC  - 500,000 counts/100 m
   SS  - 200 mg/l
8,- GT  - 3000
   pH  - As Shown
7

6

5

4

3

2
                        NOTE' All Syracuse Data Included
                          _L
                                  J_
                                      _L
                                                    D T - ' T n
                                                    FC  • 500,000 counts/lOO ml
                                                    SS  - ?00mg/l
                                                    GT  - 2 000
                                                    pH  - As Shown
     NOTE- All Syracuse Data Included
                                                           J_
                                                               _L
                                                                   JL
                           4    6   8   10
                           CI2 DOSE, mg/l
                                           12
          4   6   8
          DOSE, mg/2
                      10  12
        FIGURE  64.   Effect  of pH on  Cl2  & C102  -  Dosage  vs  FC Reduction
    7 h-
 o  -
 UJ

 §  3
         NOTE'All  Syracuse  Data Included
               23456
                  OBSERVED  LOG  KILL
 o
 3
 o
 UJ
 I-
 o
 o
 UJ
 1C
 0.
                                                            NOTE," All Syracuse Data Included
                                                                     i
                                                                          i
                                                                               i
              2345
              OBSERVED LOG KILL
FIGURE  65.   Clz  Disinfection
 Observed  vs Predicted FC  Kill
FIGURE66.   C102  Disinfection
Observed  vs  Predicted FC  Kill
                                             129

-------
     A range of FC levels from 104 to 107 counts/100 ml could affect the
reduction of FC by as much as 2 logs, as indicated in Figure 67.
Figure 6S indicates that the mixing intensity had an insignificant
effect on the reduction of FC.  Analysis of SS and pH data indicated
these two parameters to be statistically insignificant to the regression
model and therefore were not included in the final equation.

     The log reduction of FC attained during the sequential addition
tests in the Syracuse demonstration study was compared to  the results
obtained during similar studies in Rochester (54).  Figure 69 is a
reproduction of a plot of iso-kill curves obtained in Rochester for
sequential addition of Cl2 and C102> where Clg was added first.  Superimposed
on that plot is the overall average log reduction for the  Syracuse
studies of sequential addition.  Note that in Syracuse the order of
addition was C102 first.  However, an approximate comparison of the
results from the two studies is possible since the Rochester studies
indicated only slightly higher bacterial kills were obtained when C102
was introduced prior to the addition of C12.  The Syracuse results tend
to support the Rochester findings as indicated in Figure  69.

DISCUSSION OF DISINFECTION  RESULTS

     The results of the study at Maltbie Street indicate that application
of high-rate disinfection processes can result in significant reduction
of bacterial populations in CSO.  C102 dosages in the order of 6 to 12
mg/1 applied in the initial stages of overflows reduced FC levels to 200
counts/100 ml.  Applied dosages of 4 mg/1 after first-flush loadings had
passed through  the treatment system, maintained the 200 counts FC/100 ml
level in the majority of the samples collected.  Application of Clg at
dosages of from 12 to 24 mg/1 during the initial  stages of overflow also
were able to achieve 3 to 4 log reductions of FC, while lower dosages
(12 mg/1) produced similar reductions after the first 30 to 45 min.
Sequential addition of disinfectants (2 mg/1 C102 followed by 8 mg/1 Cl2
after 15 sec) at a total contact time of 1 min produced 3 to 4 log
reductions of FC.  The limited data obtained in the sequential addition
tests precludes a comparison of this method of disinfection with the
application of Cl2 or C102 separately.

     Regression analyses of the disinfection data collected indicated
that removal of SS would improve the reduction of bacterial populations
by the disinfection processes.  The effects of solids removal are slightly
more pronounced for C102 than for Cl2 disinfection with both exhibiting
improved FC kills of about one-half to one order of magnitude.

     C102 residuals in the treated effluent were not measured during the
1975 testing period.  However, C102 and Cl2 residuals (determined as
C12) measured in the 1976 tests indicated that high disinfectant residuals
could be expected in the effluent after a contact time of only one
minute. No attempts to determine the C102 residual in the form of C102
in the effluent were made during the demonstration study.  Residuals of
disinfectants are important due to their potential impact on receiving


                                     130

-------
,- 5
o
z
o
P 4
o
s

-------
water.   This  aspect of disinfection should  be further addressed  in  sub-
sequent studies  and facilities planning.

     The demonstration study also indicated  that further research  of on-
site C102 generating equipment is necessary.   The generator used  in  this
study could  not  be operated unattended  because of continuous mechanical
malfunctions.   It is believed, however,  that the potential exists  for
development  of a reliable C102 generator.   In this study, the chlorination
equipment functioned safely and reliably even though the facilities  were
operated intermittantly.  For full-scale CSO treatment applications,
consideration  should be given to use of one  ton cylinders rather  than
150 Ib Cl2 cylinders to reduce operation and maintenance costs.   The in-
stallation of  a  weighing mechanism would also provide a more accurate
record of Cl2  usage.
                                                  ISO-Kill curves
                                                  corrugated baffles mixing
                                                  CI2 first (54).

                                                     Log kill FC from Syracuse Study
                                                     CI02 first.

                                                     Log kill FC from Rochester Study
                                                     (54 ) CI2 first
                                  4     k

FIGURE 69.  Iso-Kill  Curves - Syracuse and  Rochester Studies
                          2     i
                           CI02  (mg/l)
                                      132

-------
                                 SECTION 12

                    CAPITAL AND OPERATING COST ESTIMATES

GENERAL

     Generalized capital  and operating costs have been developed, based
on analysis of the actual cost of construction and operation of the CSO
demonstration treatment facilities at both the Maltbie and West Newell
Street sites.  The various cost elements directly attributable to con-
struction and operation of the facilities are tabulated.  The costs
incurred as a result of additional peripheral equipment required for
treatment evaluations and data collection that ordinarily would not be
included in full-scale CSO treatment is subtracted from the total project
costs and compared to cost estimates presented in a previous report (56).
For the latter comparison, costs resulting from site-specific parameters,
such as site work, excavation, pumping requirements, etc., are also
subtracted from total costs, as these factors will vary within each
region and also on a site-by-site basis.

     Capital costs include structural, mechanical, piping, housing,
labor, contingency, electrical and instrumentation expenses.  The
capital costs do not include fees associated with land and site work,
engineering, legal and administrative services, fiscal concerns, or
interest during construction.  Operating and maintenance costs include
labor, power, chemicals, miscellaneous supplies, repair and replacement
parts, administration costs, laboratory and sampling costs, and yard
maintenance.  Final cost estimates are adjusted to September, 1978 according
to the ENR Construction Cost Index of 2861.

ACTUAL CAPITAL COSTS

     Presented in Table 38 are the actual capital costs incurred in con-
struction of the Maltbie Street CSO screening facilities.  The total
cost including construction bids, change orders and contingent work was
$509,514. Adjusting the costs of the items presented from June 1974
(ENR=2000) to September 1978 (ENR 2861) .results in a total capital cost
in 1978 of $728,830, or $6340/acre ($15,670/ha) or $48,590/MGD ($12.84/cu m/
day).  The third column of Table 38 represents those costs attributable
to the screening facilities exclusive of pumping station, site work,
disinfection equipment, flow metering equipment, samplers and telemetering
facilities.  Since much of the equipment installed at the Maltbie Street
site was for demonstration purposes and might not be required for a
normal facility of this type, it has been estimated that only 50 percent
of the total costs of valves and piping, electrical and instrumentation,
and 75 percent of the miscellaneous costs are directly  attributable to

                                     133

-------
the screening facilities.   Summation of the costs of items attributable
to screening facilities listed  in Table 38 result in a total cost of .
$366,960 for CSO screening facilities exclusive of disinfection equipment
and site-specific factors.  The pumping station costs were omitted at this
time since site-specific factors will dictate the pumping requirements and
size of pumping facilities, as  well as the piping requirements.

	TABLE 38.  ACTUAL COSTS* FOR MALTBIE STREET SCREENING FACILITIES


                                                          Attributable To

      Cost Component	Actual Cost	Screening Facilities
Pumping Station
Screen Housing
Vertical Shaft Screen
Horizontal Shaft Screen (2)
Valves and Piping
Electrical Instrumentation
Miscellaneous
Site Work
Chi ori nation Equipment
Chlorine Dioxide Generators
Flow Measurement
Samplers
Telemetering
172,620
137,520
53,640
104,430
74,560
44,760
15,610
31,720
20,740
26,300
22,890
18,470
5,570

137,520
53,640
104,430
37,280
22,380
11,710






Total	728,830 ,	366,960

  *Costs  adjusted  to  ENR  2861,  September  1978
     Comparison of $366,960 for the  CSO screening facilities  in Syracuse
with estimated costs of similar facilities  (56) adjusted to   1978 costs
($326,250),  illustrates that  the  capital  costs are comparable.  The
Syracuse  project results  in a value  12 percent higher  than estimated  by
previous  cost curves (56).  Therefore, projections of  total cost  for  CSO
treatment facilities required for full treatment of  Syracuse  overflows
have been determined from previously published data  (56).

     Table 39 presents the actual  capital costs incurred in construction
of  the West  Newell Street swirl regulator/concentrator.  The  total cost
including construction bids,  change  orders, and contingent work was
$112,516  in  1974, or $160,930 when adjusted to September 1978,or  $2980/acre
($7360/ha) or $23,670/M6D ($6.25/cu  m/day).  Column  3  of Table 39 lists the
capital costs attributable to only the swirl unit where 50 percent of the
.miscellaneous items and 100 percent  of the  electrical  costs are considered
directly  attributable to  the  swirl.   The  resulting total cost of  $38,300 for
the swirl regulator/concentrator  is  comparable to the  adjusted cost of
$35,500 of a similar unit estimated  from  previous data (56).  The cost of
the Syracuse swirl prototype  is approximately 8 percent more  than reference

                                      134

-------
56 estimates.    Therefore, projections of total cost for full CSO treatment
by swirl regulation/concentrators from previous data are applicable.

ACTUAL OPERATING COSTS

     The actual operating and maintenance costs for the Syracuse demon-
stration facilities were difficult to establish since such costs included
considerable charges for utilities and manpower which ordinarily would
not be included in actual CSO treatment facilities.  Items such as yard
maintenance, cleaning of blocked and/or  silted .sewer lines  and siphons
associated with the demonstration treatment facilities were performed by
personnel from various city and county departments.  Under these circum-
stances, the costs for this work were not available but recognized as a
real expense for the proper operation of the demonstration facilities.
Therefore, the operation and maintenance costs shown in this report are
based on estimates as- previously published (56).

     The operating costs include the following  items: (1) operating and
maintenance labor, (2) power, (3) chemicals, (4) miscellaneous supplies,
(5) administrative costs, (6) laboratory and sampling costs, and (7)
yard maintenance.  The following is. a brief discussion of each of the
items:

     1.  operating and Maintenance Labor - It  is assumed that the CSO
treatment facility will operate automatically  and personnel will not be
required during operation unless equipment malfunction occurs.  Dewatering
and general clean-up of the facility after a storm event is provided for
in the cost analysis.  Labor for routine visits and maintenance is
included regardless of plant operation while a variable amount of labor
is provided for the number of times and duration of overflow events.
This labor is presented in terms of manhours per year to accommodate
varying wage scales.

  TABLE 39. ACTUAL COSTS* FOR WEST NEWELL STREET SWIRL REGULATOR/CONCENTRATOR
Cost Component
Actual Cost
Attributable
  To Swirl
Site Work
Piping
Swirl Chamber
Electrical.
Miscellaneous
C102 Generator
Samplers
Pumping
Flow Measurement
Telemetering

Total
  44,240
  28,210
  28,180
   6,910
   6,390
   8,770
   6,150
  14,630
  14,020
   3,430

 160.930
   28,180
    6,930
    3,190
*Costs adjusted to  ENR 2861, September 1978
                                      135

-------
     2.  Power - Power is presented in terms of kilowatt hours/year
(KWH/yr) for the time of operation.  The average power usage is assumed
to be based on an average flowrate of 45 percent of the rated capacity
of the plant, since most overflow rates do not reach the plant capacity.

     3.  Chemicals - The chemical requirements are basically a function  ,
of the flowrate and dosage.   The amount of chemical usage  is  based on
the average flow treated.  For estimating purposes, it is assumed that
the mixing intensity 6T for the disinfection process is between 4,000-
6,000.  A Cl2 dosage of 12 mg/1, a C102 dosage of 6 mg/1, or combination
of 2 mg/1 C102, followed by 8 mg/1 of C12 is assumed to provide a reduction
of bacterial levels sufficient to achieve 1000 counts TC/100 ml and 200
counts FC/100 ml in the plant effluent.  The maximum pumping rate at the
Maltbie Street facility is 15 MGD (56,800 cu m/day).  However, the average
flowrate used in the cost estimate is 45 percent of the rated capacity,
or 6.7 MGD (25,400 cu m/day). The duration of overflow is also assumed to
be four hours for the purpose of calculating chemical requirements.

     4.  Miscellaneous Supplies - Miscellaneous supplies include spare
parts, tools, insurance, gas, oil, contracted maintenance work allowances,
and other consumable products not specifically accounted for elsewhere.
These costs are less than those associated with a continuously operated
plant but are essentially independent of the actual hours of operation
of the plant.

     5.  Administrative Costs - The administrative costs associated with
the CSO treatment facilities are assumed to represent a total  of 5
percent of the overall administrative requirements for the agency
responsible for the dry-and wet-weather treatment facilities.   Similarly,
material and supply costs associated with administrative requirements
are included under this cost category.

     6.  Laboratory and Sampling Costs - This cost category is primarily
a function of the number of samples and the types of analysis performed
on each sample.  For the estimates presented here, it is assumed that
the number of samples collected is 4 per day per overflow,  and a total
of 60 overflows per year is assumed to occur.  The cost of laboratory
materials and supplies are included in the estimate.

     7.  Yard Maintenance - The requirements for yard maintenance are
basically independent of the flow capacity of the plant.  Guidelines
which relate yard maintenance to area of site have been presented in the
Dodge Guide (57) and reference (56), and are used as a basis for estimates.

PROJECTED .COSTS FOR FULL CSO TREATMENT

     In projecting the costs for treatment of CSO in the Syracuse area, it
is assumed that the treatment devices investigated in this demonstration
study will  be^utilized, i.e., microscreens or the swirl regulator/concentra-
tor.  Since evaluations of various storage and treatment combinations and
specific sewerage conveyance capacities would require a complete Facilities
Plan, these latter factors are not addressed in this report.  Instead, it is

                                      136

-------
assumed that the design flow  to a specific CSO treatment facility is 15
MGD (56,800 cu m/day) and a design drainage area of 115 acres  (46.5 ha)
is utilized.  Costs are developed on a cost/acre (cost/ha) basis and
extrapolated to reflect the entire CSO drainage area in Syracuse of 9000
acres (3630 ha). Table 40 presents the design parameters for the CSO
treatment facilities projected for treatment of CSO in the Syracuse
area.  A swirl regulator/ concentrator required to handle the  design
flow of 15 MGD (56,800 cu m/day), would be 18 ft (5.5 m) in diameter
(13).  Capital costs for all treatment devices were developed  from
previously published data (56).  The same cost reference was applied to
the disinfection equipment.  However, the disinfectant  feedrates were
based on dosages of 12 mg/1 Cl£ and 6 mg/1 C102 for the Cl2 and C102
disinfection systems, respectively, in order to achieve desired bacterial
levels as demonstrated in this study.  The site work and pumping costs
could be significant but will be variable for specific CSO treatment
plant locations.  An estimate of pumping and site work costs is provided
later in this section.

            TABLE 40.  DESIGN PARAMETERS* - CSO TREATMENT FACILITIES
     Design Parameter
 CSO Treatment Device
Sweco Unit  Zurn Unit  Crane Unit  Swirl Unit
Design Flowrate, MGD
Capacity/Unit, MGD
No. of Units Required
Cl2 Dosage, mg/1
C102 Dosage, mg/1
Cl2 Feedrate, Ib/day
CIOz Feedrate, Ib/day
Design Drainage Area, acres
Total Drainage Area, acres
15
2.2
7
12
6
1500
750
115
9000
15
3.1
5
12
6
1500
750
115
9000
15
5.0
3
12
6
1500
750
115
9000-
15
15
1
12
6
1500
750
115
9000

*Exclusive of Site Work and Wastewater Pumping Facilities
 Notes:  1.  Conversion:  MGD x 3785 = cu m/day
                          Ib/day x 0.454 = kg/day
                          acres x 0.4047 = hectares

     Presented in Table 41 is a summary of the capital costs for identi-
fied Syracuse CSO treatment facilities handling 15 MGD (56,800 cu
m/day). The capital costs for the entire CSO system ($/Total System) were
derived by multiplying the cost per acre by 9000 acres (3640 ha), which
represents the total acreage of the Syracuse combined sewer service
area.  The costs are presented to reflect the capital costs associated
with disinfection by Cl2 and C102, respectively.  From Table 41 it is
apparent that the capital costs are significantly higher for microscreening
than for a swirl regulator/concentrator.  Using the swirl unit capital
cost as a base, the Sweco unit would be five times higher, while the
Zurn and Crane units would be 2.7 times and 1.8 times higher in cost,
respectively, than the swirl facility.  Although site work, connection
piping and CSO pumping facility capital costs are not included in the
analysis, the additional expense would be proportionately higher for the
microscreens than for the swirl unit based on site work, interconnecting

                                      137

-------
piping requirements, and equipment housing.  Pumping facility requirements
would be anticipated to be similar.  The actual cost for site work and
pumping for the microscreening installation was $9,522/MG ($2.50/cu m) as
opposed to $5,413/MG. ($1.33/cu m) for the swirl installation.  However, the
major portion of cost differential (pumping) is related to site-specific
factors.

     Table 42 presents a summary of the annual O&M costs projected for
the CSO treatment facilities.  The swirl unit is shown to be less expen-
sive to operate and maintain than the microscreens.  When compared to
the swirl unit, O&M costs for the Sweco, Zurn and Crane units are
projected to be 1.8, 1.6, and 1.6 times higher than the swirl, respectively.
(Note:  In addition to cost comparisons between the microscreens and swirl
regulator/concentrator, consideration must be given to the solids removal
effectiveness of the treatment devices.  Higher removal requirements may
dictate that microscreens be utilized dispite the cost advantage of the
swirl regulator/concentrator.  At the 15 MGD (56,800 cu m/day) design
flowrate, the Crane would remove approximately 45 percent of the SS in
terms of concentration, and the Zurn, Sweco and swirl units would remove
approximately 30, 25 and 35 percent, respectively, as indicated in Figure 26),

     Examination of Tables 41 and 42 also indicate that capital and O&M
costs are projected to be lower for CSO treatment facilities which
utilize Cl2 as the disinfectant rather than C102-  This projection is a
direct result of the higher cost for generation of C102 at $0.54/lb
($1.18 kg) when compared to $0.11/lb ($0.24/kg) for Cl2-

TABLE 41.  SUMMARY OF'CAPITAL COSTS - SYRACUSE CSO TREATMENT FACILITIES*
Treatment
 Device     $/Acre
Cl2 Disinfection	         C102 Disinfection
 $/MGD   $/Total System   $/Acre   $/MGD   I/Total System
Sweco Unit
Zurn Unit
Crane Unit
Swirl Unit
11,120
5,860
3,700
1,950
82,250
44,960
28,350
14,950
100,080,000
52,740,000
33,300,000
17,550,000
11,420
6,150
4,000
2,240
87,500
47,210
30,610
17,200
102,780,000
55,350,000
36,000,000
20,160,000

*Exclusive of Site Work, Connection Piping, and CSO Pumping Facilities
 Notes:  acres x 0.4047 = hectares
         MGD x 3875     = cu m/day

TABLE 42.  SUMMARY OF ANNUAL O&M COSTS* - SYRACUSE CSO TREATMENT FACILITIES
Treatment
Device
Sweco Unit
Zurn Unit
Crane Unit
Swirl Unit
C12 Disinfection
$/Acre
290 •
240
230
' 140
$/MGD
2,220
1,830
1,780
,1,090
$/Total System
2,610,000
2,160,000
. 2,070,000
.1,260,000
I/Acre
310
260
250
160
CIO? Disinfection
$/MGD
2,350
1,950
1,910
1,210
$/Total System
2,790,000
2,340,000
2,250,000
1,440,000

*Exclusive of Power Requirements for CSO Pumping  Facilities
 Notes:  acres x 0.4047 = hectares
         MGD x 3785     = cu m/day
                                    138

-------
     An estimate of projected capital costs for CSO pumping facilities
and site work is presented in Table 43.   The basis of this estimate
assumes that the costs incurred for the demonstration facilities would
be applicable for all CSO treatment facilities throughout the Syracuse
combined sewer system.  It is recognized, however, that each site should
be evaluated with respect to hydraulics.  Table 43 includes an additional
estimated annual cost for pumping and site work of $1,230,000 for the CSO
treatment facilities if microscreening were employed, and an increased
annual cost of $561,000 if the swirl regulator/concentrator were employed.
When the pumping and site work costs are added to the projected annual
costs, the resulting total annual costs would be as presented in Table
44.

     Table 44 presents projected total annual costs for construction and
operation of CSO treatment facilities in Syracuse.  The analysis indicates
that utilization of the swirl regulator/concentrator with application of
Cl2 as the disinfectant would be the least expensive method for treating
CSO.  Use of microscreens would cost at least 1.7 times more in construction,
operation and maintenance than use of swirl regulator/concentrators.  However,
the solids removal requirements of the treatment devices must be considered
as well as cost when the specific treatment device is to be selected.

     In Section 1 of this report it was noted that the 1968 Comprehensive
Sewerage Study had estimated the capital cost of centralized CSO treatment
adjacent to Metro at $330,000,000 and a capital cost of point-source
treatment at individual overflow sites at $485,000,000 (1978 dollars).
The cost estimate presented in this report for solids removal and high-
rate disinfection treatment processes have projected a capital cost of
$33,000,000 to $103,000,000 for microscreening, and approximately $20,000,000
for swirl regulator/concentrators.
         TABLE 43.  PROJECTED CAPITAL COSTS OF PUMPING AND SITE WORK -
                            CSO TREATMENT FACILITIES
Treatment
Device
Sweco
Zurn
Crane
Swirl
$/Acre
1330
1330
1330
660
$/M6
10,200
10,200
10,200
5,410
$/Total System
11,970,000
11,970,000
11,970,000
5,940,000
Annual Cost
1,130,000
1,130,000
1,130,000
561,000

*Capital Costs amortized at 7 percent interest for 20 years.
 Notes:  acres x 0.4047 =• hectares
         MGD x 3875     = cu m/day
                                     139

-------
                 TABLE 44.  PROJECTED ANNUAL COSTS* -
                   SYRACUSE CSO TREATMENT FACILITIES
Treatment
Device
Sweco
Zurn
Crane
Swi rl
Cl2 Disinfection
$13,189,000
$ 8,268,000
$ 6,343,000
$ 3,478,000
CIO? Disinfection
$ 13,621,000
$ 8,694,000
$ 6,778,000
$ 3,904,000

*Capital Costs amortized at 7 percent interest for 20 years.

     The projected cost estimates developed under the CSO demonstration
program for Syracuse indicate that a substantial reduction in costs could
be attained through high-rate treatment application at point-source locations.
It has also been demonstrated that further cost savings could be accomplished
through the utilization of swirl regulator/concentrators in comparison to
mechanical screening.  Although the results of this program are based on a
site-specific application, it does provide a viable alternative in evaluating
overall abatement alternatives for the handling and treatment of CSO.

     The reader is reminded that the design criteria used for development
of costs for CSO treatment in Syracuse were extrapolated from actual costs
incurred for construction and operation of the demonstration facilities.
Therefore, consideration has been given to site-specific factors such as
drainage area size and characteristics, population density, runoff coefficient,
land use distribution, trunk sewer and intercepting sewer conveyance
capacities, runoff pollutant characteristics, and rainfall patterns.  Before
utilizing cost data presented in this report, the reader should consult a
USEPA report (107) published in 1976 which presents a methodology for assess-
ing intermittent urban point-source loads such as stormwater and CSO.
That study discusses in detail the importance of such site-specific
factors as mentioned above, as well as describes supplementary data
desired, levels of accuracy and spatial detail required in storm load
characterization, and the effect that existing conveyance and treatment
facilities may have on the storm load contribution.  Methodologies for
evaluating collection system control techniques, storage/treatment
options, and flow regulation measures are also presented.  Reference to
that report is recommended to the reader to assist him in determining
the level of effort and methodology of approach to address  his specific
CSO abatement problems.
                                     140

-------
                               •   SECTION 13',

                    VIRUS STUDIES - MALTBIE STREET FACILITY
GENERAL

     This section presents a summary description of the experimental
procedures utilized in the collection, handling and analysis of virus
organisms investigated in the Syracuse demonstration study.  .The
testing program was divided into two phases, as described in Section 4.
The results and discussion for each of the phases are presented separately
following the description of experimental procedures utilized in each
phase.                 •                    ,

PHASE I PROGRAM                           ,

     In the Phase I virus program, samples of CSO were taken before and
after treatment, to determine the efficiency of various treatment and
disinfection techniques for reduction of virus organisms-in CSO.  The
Aquella Concentrator was used to concentrate•numbers of organisms per
unit volume of sample, to facilitate more'reliable analysis.  It was
found that in general naturally occurring viral organisms in CSO are too
few in number to permit satisfactory analysis * even with high degrees of
post-sampling concentration.

Experimental Procedures. (Phase I)          .,'.•-
                                      n f     <   "    '
     The experimental procedures used in the'Phase I Program involved
three separate steps:

     1.   Demonstration of and familiarization with  the Aquella Virus
          Concentrator (Carborundum Company, Buffalo, N.Y.).

     2.   Actual sampling of flows during storm events.

     3.   Laboratory processing of samples.

Aquella Virus Concentrator—
     The virus concentrator basically consists of a centrifugal pump,
followed by three clarifying filters, a wound fiberglass adsorbing
filter and a membrane type adsorbing filter.  Accessories to these
functions include a flow meter, by-pass valves, a proportioning pump for
the addition of acid and cationic reagents, and an air compressor for
emptying the liquid from the system.   The filters have vents to release

                                     141

-------
trapped air as they fill with water.  In principle the orlon clarifying
filters serve two functions: to remove silt, bacteria and other suspended
solids and to remove membrane-coating organic materials which interfere
with the adherence of virions to the adsorption filters.  The viruses
are preferentially adsorbed by the final two filters in the system.  The
first,  a wound fiberglass filter of 1 nanometer  porosity, adsorbs 50 to
95 percent of the virions.   The second, a plate-filter holder with epoxy-
fiberglass with 5.0, 2.0 and 0.45 micron membranes in series, concentrates
the small number of viruses which pass the fiberglass filter.

     Adsorption of virus particles to these filters is greatly enhanced
by an acid pH and either aluminum or magnesium ions (22).  In the absence
of either magnesium or aluminum the adherence of poliovirus was very
irregular which was probably due to trace contaminants of various divalent and
trivalent ions.  It was found that optimum recoveries of adenovirus were
obtained at pH 4.5 rather than 3.5 when glycine buffer was used as the
eluent.   Aluminum ion at pH 4.5 and beef extract plus magnesium at pH
5.5 or 6.5 yielded equal numbers of viruses although both were less than
expected.  It has previously been stated that pH 4.5 is near the upper
limit for the most efficient adsorption of enteroviruses (21).  However,
the recovery of poliovirus appeared slightly better at pH 4.5 than at
3.5, which was originally recommended for the Aquella (22).

     Although  the use of 3 percent beef extract, pH 8 to 9, was proposed
as a nondestructive means of eluting viruses from membrane filters (58),
no appreciable advantage of beef extract over rapid elution by pH 11.5
glycine followed by nearly instantaneous neutralization of the filtrate
was found.

Phase I Sampling —
     Storm overflows were collected in 55 gal (208 1) drums, trucked to
the laboratory and refrigerated at 39.2°F (4°C).   They were usually concen-
trated the following day.  In the case of disinfected overflow samples, sodium
thiosulfate was added to a level of 300 mg/1 to neutralize disinfectant
residuals.  At 39.2°F (4°C) the titers of enteroviruses in storm overflows
remained constant for more than a week.  Repeated isolations of wild
adenoviruses from sewage also indicated they were not inactivated by storage
for several days at 39.2°F (4°C).

Phase I Laboratory Procedures —
     After prefiltration to remove inorganic solids, bacteria and membrane-
coating materials, the samples were split and concentrated on wound
fiberglass and/or Cox membrane filters either at pH 3.5 (with Al+++)-or
pH 4.5 (with Mg++). The concentrates were frozen after neutralization and
addition of nutrients.

     Enteroviruses, adsorbed at pH 3.5, were cultured.  Adenoviruses
had previously been isolated in Syracuse sewage samples in the laboratory
(but not in CSO) by a technique suggested by T.6. Metcalf  (59), i.e.,
prefliters were treated to reduce adsorption losses and the viruses were
adsorbed at pH 4.5 plus 0.05M MgCl2-  The virions were eluted with 3
percent beef extract and quantitated by plaque assays.

                                      142

-------
     The grouping of the viruses from pH 3.5 adsorption was performed by
using pooled antisera against polioviruses and coxsackie B viruses to
remove one or both groups selectively.  It was assumed that only the
echoviruses would break through neutralization by the combined antisera.
Presence of hemagglutinin for human type 0 cells and "ragged" plaque
morphology helps confirm the identification of many putative echoviruses
(l6).  Acridine orange staining of infected cell monolayers in microtiter
plates quickly differentiated the occasional adenovirus or reovirus
which survived adsorption at pH 3.5, since both groups exhibit green
fluorescent emission.  Sensitivity to low bicarbonate concentration in
the agar overlay and ability to grow at 104°F (40°C) were used to determine
whether the polio strains were vaccine or wild isolates.

     Three types of assays were employed for the animal viruses:  plaque
formation (is), 50 percent tissue culture infective dose assays (TC1D5Q)
(17) and minimum probable number (MPN) estimates (18).  The plaque
forming units (PFU) were determined in conventional fashion by neutral
red overlays in 60 mm plates.  The TC1D50 and MPN estimates were performed
in Mictrotest II plates with 1 cm diameter wells (17).  The wells were
seeded with growth medium, an operation which was conducted in a laminar
flow isolation hood.  The seeded plates were incubated 48 hr, 98.6°F (37°C)
5 percent C02 atmosphere, 85 percent relative humidity.  The plates were
not sealed individually during the incubation period.  Plates were
inspected for cytopathic effects at 48 to 96 hr for enteroviruses;
adenovirus plates were maintained 7 to 10 days with refeeding. The use
of L-15 medium (20) and a higher pH, 7.4 to 7.6, greatly improved the
maintenance'of Microtest II cultures for these extended periods of
observation.

     Polioviruses were neutralized in samples of concentrated viruses by
addition of pooled human gamma globulins or pooled hyperimmune rabbit
antisera. Adenovirus isolates were confirmed in part through the use of
pooled rabbit antisera.  Pooled rabbit antisera were used to neutralize
coxsackie B viruses and pooled human gamma globulin was found which had
anticoxsackie B activity as well as activity against the polioviruses.

Results and Discussion (Phase I)

     Two general types of filtration procedures were used to concentrate
viruses from CSO:  a two-step, discontinuous batch process and a continuous
process.  In the batch process a water sample was clarified and pumped
into a reservoir, after which A1++++ or Mg++ was added, the pH was
adjusted and the sample  was pumped through the adsorbing filters.  In
the continuous process the water was pumped through the clarifying
filters, a proportioning pump added inorganic ion and adjusted the pH to
a previously determined ratio, and the water then passed through the
adsorbing filters.  In both cases, the flowrate averaged approximately
40 gal/min (150 1/min) although it was reduced in very turbid water.

     There were marked differences in efficiency of virus recovery and
reproducibility for the two processes.  Batch type recoveries in volumes
                                      143

-------
varying from 0.5 to 20 gal (1.9 to 75.6 1) were about 55 percent of
adenovirus and 100 percent of enteroviruses, such as poliovirus 1 (PV1).
With the continuous process low recoveries were experienced and there
was no apparent correlation between the size of the inoculum, the volume
of water or the type of water.

Recovery of Wild Viruses From CSO —
     Viruses have been found in all sewage samples from both Syracuse
and Baltimore (50) suggesting that urban stormwater and CSO have a high
probability of carrying viruses.

     Direct isolation of virus pathogens was demonstrated in the August
14, 1974 Syracuse CSO samples collected at the Maltbie Street facility.
Poliovaccine virus probably constituted much of the isolation but other
viruses broke through the poliovirus antisera in half the positive
samples. Despite the obvious differences in sample sizes, Syracuse CSO
and Baltimore stormwater did not differ greatly in the detection frequency.
Whether or not this is fortuitous remains to be seen after a larger
number of samples have been analyzed.

     A limited number of CSO at Maltbie Street in the summer and fall of
1974 were analyzed in some detail.  Viruses were detected in the over-
flows with one sample yielding  quite high levels  of viruses.
Although full identification of the viruses in the "other" category was
not made, plaque morphology and staining characteristics suggested that
the viruses were not typical adenoviruses.

     No clear-cut pattern emerged from the Maltbie Street results
concerning the likelihood of detecting virus in early samples versus
late samples during the course of a storm overflow.  Viruses were observed
in both early and late samples.   Likewise there was no convincing
evidence from these data of a distinct seasonal variation or a correlation
between virus numbers and total  volume of flow.

Disinfection of Viruses in CSO —
     Storm overflows filtered by the Sweco microscreening unit were
assayed before and after treatment with 4 mg/1 C102.  Approximately a 1
min contact time was provided, after which the wastewater was pumped into
barrels containing excess sodium thiosulfate to halt residual disinfection.
Treatment with C102 at this level provided no significant reduction in
titer of these wild viruses. It must be noted that at the times of the
highest level of viruses surviving C102 disinfection sampling methods
had not been fully perfected and lapses of 1.5 hr occurred between the
chlorinated and unchlorinated samples.  Without nearly simultaneous
pretreatment controls it cannot be concluded unequivocally that no knock-
down of wild virus was achieved—although it seems unlikely that levels
of poliovirus as high as 230 PFU/gal (61 PFU/1) would be the remnant of
a much larger virus concentration in CSO.

     In general it appears that the low levels and random appearances of
wild viruses will obfuscate any attempt to evaluate viricidal effectiveness
of the treatment process directly.

                                     144

-------
Feasibility and Operating Experiences With  the Aquella Virus Concentrator--
     In wastewater samples with high organic content and/or silt content, a
problem was encountered with  the formation of a dense precipitate which
clogged the adsorption filters at pH 3.5 to 4.5 and greatly impeded
reconcentration of viruses from the 267, 125 and 47 nanometer membrane
filters. Virus filtrates from 55 gal (208 1) samples frequently could
not be recovered entirely on the 47 mm filter because of the low flowrate.
The precipitates formed in creek water, sewage, pond water, stormwater
and tap water.  Attempts to eliminate the precipitate by various pH
manipulations, addition of EDTA and substitution of other buffers for
the glycine failed.  The working solution was to concentrate the virions
on the 125 nanometer filters only and to elute the virus with 3 percent beef
extract, pH 9, overnight, 39.2°F (4°C).  This eliminated long exposures of
the viruses to pH 3.5 during reconcentrations.

Attempts to Develop f2 Phage as an Indicator Virus for Disinfection
Studies --
     The application of the Aquella virus concentrator to different
water sources requires considerable expertise and necessitates expensive
time consuming preliminary studies to determine the effect of a particular
environment on the efficiency of virus recovery.   Where water conditions
change continuously, such as in CSO, very elaborate controls for adsorption
efficiency must be instituted.

     An attempt was made to determine if conditions for the adsorption
and elution of enteroviruses from depth filters and membrane filters in
the Aquella system would also concentrate f2 phages.  This seemed feasible
since these phages resemble picornaviruses superficially.  Three concentra-
tions of viruses in CSO were studied and four points in the Aquella system
were titered to determine the amount of unadsorbed f2:  1) after the
three or!on clarifying prefilters, 2) after the addition of glycine-HCl
and AH++ and before the depth filter, 3) after passage through the
depth filter and the Cox filter, and 4) after elution with glycine-NaOH
buffer, pH 11.5 and adjustment to pH 7.0.

     It was found that both the adsorption and elution characteristics
of f2 phage in CSO are different from those of the enteroviruses.
Addition of A1+++ at pH 3.5 did not improve removal.  Attempts to recover
the f2 phage from the membrane filters by elution using several techniques
failed as well.  The efficiency of adsorption did not appear to be
related to the initial virus load.

Feasibility and Operation Experiences --
     The Aquella virus concentrator and its system of selective adsorption
of viruses by filter media are reasonably well suited for study of CSO.
The basic instrument met most of the manufacturer's claims.   The system
of prefiltration functioned well for stormwaters despite their high
turbidity and solids contents.   An overly fragile centrifugal pump and
chronically leaky hose connections were both a nuisance and a potential
hazard for the operating crew.  Although the Aquella is a movable instrument,
it is not very portable, particularly over broken ground, and is nearly
impossible to move to difficult sampling sites.  Field work with the

                                      145

-------
concentrator requires rather elaborate supportive facilities including
power generators, covered working space, refrigerated sample storage,
etc.

     The design oversight which permitted stratification of the acid
Aids solution  and hence poor pH control invalidated much of the first
three months work.  However, the recent addition of an efficient mixing
chamber seems to have corrected the problem and pH can be maintained
indefinitely in most CSO.

     The principal drawback with the Aquella was the limited numbers of
large samples which can be processed per working day.   Even with hard-
working, trained crews, a maximum of two 55 gal (208 1) samples per day
could be concentrated to 13 ml.  Flowrates rarely exceeded 40 gph (2.5
1/min) and usually fell to about 25 gph (1.6 1/min) halfway through the
primary isolation, apparently due to the aluminum precipitate and colloidal
materials which passed through the 1 micron prefilters and clogged the
257 nanometer Cox filters.  Prospects for using the Aquella concentrator for
frequent or continuous  monitoring of viruses in CSO seem poor.

     The levels of virus observed in the Phase I program by Aquella
concentration techniques underscored two important considerations:  1)
recoveries of seed viruses are very efficient so that an indicator
vaccine virus(es) introduced into CSO before disinfection treatment and
isolated after some standard retention time could be expected to reflect
the efficiency of the viricide (dilution phenomena could be estimated by
simultaneous introduction of a fluorescent dye indicator), and 2) the
population of wild viruses in CSO is at a very low level, and the sample
variation is very high.  Thus it seems questionable whether a meaningful
measure of disinfection effectiveness can be made on the basis of observed
reductions in the wild viruses in CSO.  Certainly a most important need
is to determine the statistical constraints on virus sampling necessary
to show any significant germicidal activity of the treatments.  It is
presently unknown whether virus pathogens in CSO approach anything like
the steady-state populations of coliforms which occur in sewage.  Closely
spaced fluctuations in virus populations have not been measured.  Thus
there is no rationale with which to defend grab sampling as a means for
evaluating any viral disinfection of CSO.

     The degree to which CSO and Onondaga Creek contaminate Onondaga
Lake with viruses has not yet been determined.  It is probably a fraction
of the virus input from the effluent of the Syracuse Metropolitan Sewage
Treatment Plant which, when upgraded to tertiary treatment, will still
discharge perhaps as much  as 1 to 7 virus units per 100 ml if it performs
like similar advanced treatment facilities (61).  However, even if Metro
could eliminate viruses in its effluent, the CSO would maintain Onondaga
Lake and a portion of the Seneca River at an unacceptable level of virus
content for fishing and bathing, since most virologists believe that 1 PFU
can constitute an infectious dose (23, 62, 63).  This hazard must be
judged marginal but real.  At present no virus standards for recreational
(or potable) water supplies exist.   Suggestions of virus levels at 1/1
or 5/1 have been made for drinking purposes (62) although this will probably

                                    146

-------
be revised upward now that more efficient concentration methods such as the
Aquella technique are available.

PHASE II PROGRAM

     In the Phase II virus program, CSO flows to treatment units were
deliberately seeded with relatively high levels of viral  indicator
organisms in order to insure more reliable analysis of numbers of organisms
present, and of degrees of viral reduction with various disinfectant
dosages.  It was presumed, based on previous work by other investigators (64),
that the log reduction in viral populations is largely independent of
the initial size of population.

Experimental Procedures (Phase II)

     The basic test system for the high-rate virus disinfection studies
in the Phase II program employed 3 viruses as indicators  which could be
used to monitor inactivation of viruses in CSO:  coliphages f2 and 0X174,
and poliovirus 1 (Sabin, Kl) oral vaccine strain.  Phage  levels were
measured prior to and subsequent to disinfection injection.  Eight
grab samples were taken at each sampling point over a 30  min interval.
Poliovirus 1 (PV1) was limited in most experiments to a single 55 gal (208 1)
grab sample upstream and downstream from the point of disinfection; the
variability due to dilution and uneven mixing was controlled by the
addition of sodium fluorescein to the inoculum.  Thus the base line was
the ratio of plaque forming units (PFU) of virus to fluorescence units.

Wastewater Sampling and Inoculation --
     Storm overflows collected in 55 gal (208 1) drums were trucked to
the laboratory and refrigerated at 39.2°F (4°C).  Sodium thiosulfate was
placed in the drums before the samples were collected; the final concentration
of thiosulfate was 300 mg/1, sufficient to neutralize any disinfectant
residuals.   At 39.2°F (4°C) the enterovirus titers remained constant for
more than a week; adenoviruses were not inactivated by storage for several
days at 39.2°F (4°C).

     The virus inoculum (0.93 gal (3.5 1) containing 4 g  disodium flourescein)
was added to the influent of the Zurn Micromatic fine mesh  (71y) drum screen.
This method of inoculum addition allowed complete mixing  of the inoculum
and CSO during the screening process.  Grab samples were  collected at
two points—at the effluent from the Zurn unit  just prior to discharge
to the disinfection tank, and at the outlet weir of the disinfection
tank after flash mixing and a one min contact time.

     The bacteriophages were sampled in 10 ml aliquots at intervals of
0, 2, 4, 6, 8, 12, 20 and -30 min after addition of the inoculum.  The 10
ml precalibrated tubes contained 0.5 ml sodium thiosulfate to halt the
action of residual disinfectant.  The samples were sterilized with
chloroform and divided.  One part was used for plaque assays after
storage at 39.2°F (4°C); the other was frozen at -40°F (-40°C) and
analyzed for fluorescein content later.


                                     147

-------
     The 55 gal (208 1) samples were collected 4 min after inoculum
injection on the Zurn effluent and after 6 min at the disinfection tank
outlet weir. Ten ml samples were removed from each barrel for analysis
of fluorescein tracer and bacteriophage survival; the animal viruses
were concentrated from 50 gal (190 1) to 20 ml.

Results and Discussion (Phase II)

Grab Sampling and Equilibrium —
     Two methods were used to determine the dwell time of the CSO between
the pre-disinfection and post-disinfection sampling points— disodium
florescein concentration and phage titers.  Fluorescein tracing tests
indicated that dye appeared at the disinfection tank outlet less than 2
min after it was detected in the Zurn effluent.  A peak of activity was
observed between 2 and 6 min in nearly every case.

     However, the peak activity showed large variations,' apparently
because thorough mixing of the inoculum injected in the influent to the
Zurn unit contact tank (receiving the Zurn screened effluent) required
several minutes to stabilize.

     Typical plots of influent and effluent viral indicator organism
levels are shown in Figures 70 through 73.  Runs were made in which no
disinfectant was added to  seeded CSO, and concentrations of viruses at
the two sampling points were approximately equal 8 min after addition of
the virus inoculum. Nevertheless, large fluctuations were observed
occasionally on the influent side for up to 12 min.  Because of the
propeller-type flash mixer, the effluent had much less variation.  At 20
min the untreated samples differed by less than 2 min (dwell time) and
were nearly equal.

     Originally it had been planned to use the declining ratio of phage
to fluorescein as a measure of relative activity of the disinfectant.
Unfortunately, florescein is bleached by C~\2 and C102 and is not a
suitable marker at concentrations above 5 mg/1 Cl2 and 2 mg/1 C102.

     Table 45presents the results of the 1976 disinfection investigations as
related to the inactivation of viruses and enterobacteria.  Each of the six
storm events are discussed individually in the following paragraphs.

     Storm 1.  A relatively massive dose of chlorine (24 mg/1 with a
chlorine residual 1.0-1.4 mg/1) virtually sterilized the CSO.  A clear-
cut equilibrium was achieved on the influent side, but high level,
constant dosing on the effluent side prevented any sort of steady-state
from being established between the replacement viruses and the flash-
mixing pool.  The rate of  kill exceeded the replacement rate.

     Storm 2.  Two serial runs were made on this storm without allowing
an interval between the experiments for the Zurn screening unit to purge
itself of the phage inoculum after the first serial run.   The first run
indicated more or less typical declines of a virus inoculum being killed
by 12 mg/1 Cl2 at the same time that it was being slowly diluted.  A

                                     148

-------
     2  <  6 8
                                            I  io3U-4
         ELAPSED TIME AFTER SEEDINC, min
FIGURE 70.   Inactivation  of fZ  Phage
              Storm 1
                                                     ELAPSED TIME AFTER SEEDING, min
FIGURE 71.   Inactivation  of fZ  Phage
              Storm  4
     2  •>  6  3    12
        ELAPSED TIME AFTER SEEDING, mm
 FIGURE  72.   Inactivation of (3X174
               Phage -  Storm  1
                                                       ELAPSED TIME AFTER SEEDING. n»n
FIGURE 73.   Inactivation of  0X174
              Storm 4
                                        149

-------
                                                       TABLE 45.   COMPARATIVE INACTIVATION OF VIRUSES AND ENTEROBACTERIA
cn
o
Storm
1
2

3

4

5

6

Disinfectant Dose, mg/1 .
Date Process Cl? CIO? Mixinq
6/16/76 U
6/30/76 U
U
7/8/76 U
S
7/29/76 U
S
9/10/76 U
S
10/8/76 U
U
24
12
12
11
12
12
3.4
8 2
8 2
7.3
7.3
SF
SF
SF
SF
F
SF
F
SF
F
SF
SF
f2
4.3
1.4d
1.1
0.3.
1.2C
2.5
2.1
0.6.
0.0e
0.2
0.3
Log Reduction
0X174 Polioc
4.7
0.9.
0.9d
0.1
0.2
2.7
2.4
0.4
0.2
0.3
0.2



4.4
3.1
5.3
5.2
3.5
4.8
2.1
2.3
•TC

5.
5.




1.
2.
1.
0.

9
3




0
7
4
7
FC FS
5.
4.
4.


3.
1.
0.
2.


3
4
1


0
5
9
3
2.3
2.1
Elapsed Time
After Seedinq, min
20
20
20
20
20
20
20
20
20
20
20
            •U  = unscreened samples;  S = screened (23u)  samples
             SF = sequential flash mixing; F = single flash mixing
            c Elapsed time after seeding = 8 min
              Elapsed time after seeding = 30 min
            e Elapsed time after seeding = 12 min

-------
second run made in the last few minutes of the storm overflow gave an
apparent confused pattern of mixing between 6 and 20 min.  Interpretations
of this anomaly can only be conjecture, but the reduced virus flow at
these points may reflect the low level of water in the pump well  at
Maltbie  Street as the storm abated.   The 30 min samples may be more
representative of the disinfection process in the second run.

     Storm 3.  The inoculum in Storm 3 contained not only phages f2 and
0X174 but poliovirus as well.  Run 1 indicates the effect of a relatively
heavy dose (11.0 mg/1) of C102-  Inactivation of phage at this level was
minimal, being less than 1 log reduction in titer.  The storm overflow
ended before a complete second run with Cl£ (12 mg/1) could be achieved.
Only a low degree of inactivation appeared to be underway despite this
relatively high level of Cl2-

     Both Cl2 and C102 resulted in very high levels of inactivation for
PV1-(Table 46), which is not surprising for C12 in view of the fact that
at this high dosage, the free Cl2 residuals are likely to be as large as
1-2 mg/1, a highly lethal dose of HOC1 (19, 20).  A killing efficiency of
this order of magnitude for PV1 is greater than has been observed before
for virus in wastewater and should be viewed with some skepticism until
the experiment is repeated several times.  That C102 would show no
effect would not be anticipated since other experiments have demonstrated
the capacity of C102 to be viricidal at both greater and lesser concentrations.

     Storm 4.  Treatment of the CSO was essentially a repetition of the
schedule for Storm 3 with the exception of sequential vs single-flash
mixing (Table 47). It is not clear how a reduction in  the concentration
of C102 from 11.0 mg/1 (Storm 3) to 3.4 mg/1 (Storm 4) could permit an
observed increase of 10-14 times the kill.  Bacterial kills in Storms 3
and 4 showed a dependency on  the time at which the samples were taken.
In Storm 3 the reduction of FC ranged from 1 log near the beginning of
the storm to 6 logs near its end; in Storm 4 the reductions varied from
1 to 2 logs at the storm beginning to 2 to 3 logs near the storm's
end.   However, the changes in CSO composition which affected the kill
rate of bacteria so drastically had very little influence on  the phages
and the poliovirus.

     It appears that the inactivation of viruses was nearly identical for C102
and Cl2 for the same storm and largely independent of the disinfectants'
initial concentration.  The disinfectants were probably in excess but an
unknown secondary factor controlled the total kill.

     Storm 5.  This test attempted to examine the effectiveness of the
disinfection procedure by sequential disinfection with 2 mg/1 C102 followed
by 8 mg/1 Cl2 after 15 sec.  There was no dramatic increase in viral kill
over the previous storms in which the disinfectants were added separately.
The total kill of PV1 (Table 48) remained at about the same level as observed
in Storms 3 and 4.  The second run produced killing curves more or less
identical to the first run.
                                     151

-------
 TABLE 46.  DISINFECTION OF CSO SEEDED WITH PV1 -  STORM 3
         Treatment of CSO+
PFU/ml concentrate
C12*       C102**
         Before disinfection

         After disinfection

         L'og10 reduction
23xlOl
4.44
66x10^
64xl02     83xl02
3.09
**Unscreened, CIO? 11.0 mg/1
 *Screened, Cl2 12" mg/1
 +CSO was seeded; unseeded CSO titer was less than 1 PFU/ml
 TABLE 47.  DISINFECTION OF CSO SEEDED WITH PV1 - STORM 4


Treatment of CSO+
Before seeding
After seeding
After disinfection
Log,Q reduction
PFU/ml
Cl2*
0
133x20
42x10
5.13
concentrate
C102**
0
3 ISOxlO3
1 0
5.20

**Screened, C102 3.4 mg/1
 *Unscreened, Clg 12 mg/1
 +CSO was seeded; unseeded CSO titer was less than  1 PFU/ml,
                               152

-------
     Storm  6.  Continuous addition of  C102 throughout the storm allowed a
large degree of stabilization of disinfectant dosage.   The previous disin-
fection patterns of phages f2 and 0X174 reduction were obtained on both the
first run and the second run.   Phage reductions were very low or nonexistent;
PV1 recoveries {Table 49) indicated reduced activity  from that obtained with
the quantities of ClOg in Storms 3, 4  and 5.  Little indication of significant
sampling anomalies in the inactivation curves of the phages was evident.

	TABLE 48.  DISINFECTION OF CSO  SEEDED WITH PV1.    STORM 5


                                             PFU/ml concentrate
          Treatment of CSO+                  Run 1*      Run 2**
Before disinfection
After disinfection
57 x 104 6 x 105
16 x 102 96 x 101
          Log1Q reduction                      3.45       4.78


       *Unscreened, C102 (2 mg/1) followed by C12  (8 mg/1)
          after 15 sec
      **Screened, C102 (2 mg/1) followed by C12 (8 mg/1)
           after 15 sec
       +CSO was seeded; unseeded CSO titer was less than  1 PFU/ml


       TABLE 49.  DISINFECTION OF CSO SEEDED WITH  PV1.    STORM 6


                                             PFU/ml concentrate
	Treatment of CSO+	Run 1*    Run 2*	

          Before disinfection                37 x  104  25 x 105

          After disinfection                 48 x  102  55 x 103

          Log1Q reduction                      2.11       2.34


       *Unscreened, continuous CICL dose, 7.3 mg/1.
       +CSO was seeded; unseeded CSO titer was less than  1 PFU/ml

 Effect of Cl? and C102 on Naturally Occurring, Wild Viruses in CSO

      Isolations of viruses from untreated water showed a  background  titer of
 about 40 PFU/gal (10  PFU/1).  These appeared to be mostly
 poliovirus of unknown origin since they were neutralized  by antipolio  sera.
 It  is possible  that some of the viruses originated from earlier  polio  seeding
 experiments.  Although adenoviruses occasionally appear in sewage, none
 were  isolated from any of the Syracuse CSO.  Flows from Storms 4 and 5 were
 seeded with 3780 PFU  Adenovirus 31/gal CSO and treated with 12 mg/1  Cl2 and


                                    153

-------
with 4 mg/1 C102 for 6 min each.  No virus was recovered from either treatment
when 0.52 gal  (2.0 1) were concentrated and the equivalent of 400 PFU were
inoculated into tissue culture flasks.

SUMMARY OF  VIRUS STUDIES

     The data indicate that several novel observations were made during the
course of the study and that after treatment of CSO with Cl£ and/or C102
the indicator viruses behaved somewhat differently than had been predicted
from the bench-scale study (7).  Phage 0X174, which was shown to be much more
sensitive to C102 than phage f2 (under conditions of low C102 demand),
exhibited almost identical sensitivity to C102 under the conditions of these
experiments.  Indicator viruses showed minimal response— and sometimes no
response-- to either the chemical form or the concentration of the disin-
fectant.  For example, in  three different experiments 12 mg/1 Cl2> 11 mg/1
C102 and 3.4 mg/1 C102 achieved essentially identical kills.

     The prediction that f2 phage would make a good simulant for entero-
viruses in CSO and would be useful for monitoring treatment plant effluent
did not prove to be true.  Polio virus in CSO was 2 to 3 orders of
magnitude more sensitive to Cl2 and C102 than f2 phage.  Neither the
concentration of suspended solids nor the chlorine demand of the water
proved to be a reliable predictor for the efficiency of disinfection.
Some of the most effective disinfection occurred when SS were highest.

     Screening had no influence on virus inactivation in CSO.  This confirmed
earlier observations from the bench-scale study that the enteroviruses and
phages occurred as single, free virions and were several orders of magnitude
smaller than the finest microscreen.

     The use of a fixed flowrate for the experiment improved the reproduci-
bility of the disinfection methods within a storm.  However, it had no
obvious effect on inter-storm variability.  Flash mixing provided evenly
distributed samples and supplied a more predictable microbial population
as targets for disinfection.  The even distribution of fluorescein label
measured immediately after flash mixing showed that mixing of the samples
and disinfectant was complete within a few seconds.

     Simultaneous reductions in bacterial and viral titers were common but
there was no direct proportional relationship between them.  Samples were
obtained in which (1) both bacteria and viruses were reduced; (2) bacteria
were reduced several logs and viruses only slightly; and (3) viruses were
reduced several logs and bacteria only slightly.

     Previous experiments reported elevated kills as a result of episodic
additions of small doses of two disinfectants (Cl2 and C102) rather than
a large dose of one.  The field trials at Maltbie Street failed to
confirm such enhanced kills.  The explanation is fairly direct.  The
multiple additions in the bench scale studies were made to still suspensions
of viruses followed by stirring.  Since multiple small additions resulted
in prolonged stirring, the kills were improved because of greater contact
time between the virus and the disinfectants (shorter diffusion path)

                                     154

-------
before the latter decayed. In the demonstration treatment plant maximum
contact was achieved in all schedules by flash mixing.   Thus no improvement
was seen by multiple additions to  the flash mixer.  In fact, reduced
total kills resulted probably because the flash-mixing of multiple small
doses failed to achieve free residual chlorine levels for as long as one
large flash-mixed dose.

     The use of either Cl2 or C102 can be justified on the basis of
their ability to reduce titers of enterovirus (PV1) in CSO by 2.1 to 4.7
logs.  Naturally  occurring wild enteroviruses (1600 PFU/40 gal) were
eliminated by treatment with 12 mg/1 Cl2 for 1 min and by 3.4 mg/1 C102
for 1 min.
     On  the basis of reductions of seeded PV1 samples it appears that the
minimum dose of C102 should be 8 mg/1, and the minimum dose of Cl2 should
be 12 mg/1.

     Mild acidification (pH 3.5) followed by chlorination would vastly
increase the efficiency of Cl2 or C102-  If the receiving water of the
creek could not tolerate this pH change, it might be feasible to pump
the effluent over a bed of crushed limestone to neutralize the acid.
                                    155

-------
                    SECTION 14  SPECIAL CONSIDERATIONS
GENERAL
     In addition to the basic objectives of the Syracuse CSO demonstration
study, certain peripheral investigations and studies were performed either
on related research and demonstration work which would enhance present and
future studies on CSO, or field testing of specific applications.

     The areas covered under this section include special analyses for
formation of Cl2 and C102 organic species formed during disinfection, and
C102 sensitivity; special investigations for ultraviolet disinfection, and
adenosin triphosphate (ATP) assay for disinfection control; and special
instrumentation applications of telemetering, a total organic carbon (TOC)
monitor, and a suspended solids (SS) meter.


SPECIAL ANALYSES

Interaction of Cl2 and C102 with Organic Species in CSO Wastewaters

Chlorinated Hydrocarbon/Pesticide Scan-
     In 1976, analyses of trace organic compounds by gas chromatography scans
(the concurrent determination of several related compounds by a single
procedure) were conducted on 13 CSO samples.  The samples were collected
from the Maltbie Street treatment processes prior to and subsequent to
disinfection with Cl2> C102> or sequential addition of C102 and C12.
Samples were analyzed for non-volatile chlorinated hydrocarbons, which
include PCBs and several common pesticides.  The analytical technique
employed was hexane extraction of hydrocarbons from a sample followed by gas
chromatography analysis as recommended by EPA in 40 CFR 136.  The chroma-
tographic conditions were:

     Column:        3 percent OV-1 on Chromosorb W-HP 80/100

     Carrier gas:   nitrogen at 0.013 gpm  (50 ml/min)

     Column temperature:  3920F  (20QOC)

     Detector:      electron capture

The objective of this trace organic scan was to provide a first-cut screen-
ing of CSO to determine the general magnitude of chlorinated hydrocarbons
rather than to quantify or identify specific compounds.
                                    156

-------
     The results of the gas chromatography as presented in Table 50 indicated
a lack of correlation of chlorinated organics formation upon addition of
either Cl2 or C102 or upon sequential addition of C102 and Cl2-  The change
in chlorinated hydrocarbon content ranged from -8.5 to +123.7 percent for
Cl2> from -7.5 to +108.3 percent for C102> and from -2.2 to +200.0 percent
for the sequential addition case.  Concentrations of chlorinated hydrocarbons
of untreated CSO ranged from 1.2 to 59 yg/1 and disinfected samples contained
from 2.5to 54.0 yg/1.  However, the majority of samples (5 of 8) indicated
significantly increased levels of chlorinated organics after addition of
disinfectants.

     The limited analyses conducted on untreated and treated CSO in the
Syracuse demonstration study indicated that chlorinated hydrocarbons  are
formed during disinfection processes utilizing Cl2 and/or Cl02 as disinfect-
ants.  However, significantly more data points are required to adequately
quantify the amounts of chlorinated organics formed.

Volatile Chlorinated Organics Formation - Bench Scale Studies—
     If both Cl2 and C102 are to be used effectively as wastewater disin-
fectants, the chemical processes associated with disinfection must be under-
stood.  It has been fairly well established that organic compounds in  CSO
compete with pathogens for disinfectants.   This has been established in
 studies of  natural  water  systems  containing  synthetic  organic  chemicals.
 Chloroform,  bromoform,  and  their  intermediates  have been  identified  after
 chlorination of Rhine  River water for  the  water supply of Rotterdam  (65)
 arid Mississippi  River  water for the  water  supply of New Orleans  (66).


     In order to review the potential  interaction  of Cl2 and C102 with
organic matter present in CSO, the aqueous chemistry associated with Cl2
and C102 must be understood.  In aqueous solutions of Cl2 five reversible
half-reactions are necessary to describe the rapid equilibria which_describe
the.distribution of hypochlorous acid  (HOC!), hypochlorite ion (OC1~),
chlorine monoxide (C120), trichloride  ion  (Cl3~), the hypochlorous acidium
ion (H20C1+) and of course undissociated chlorine (Cl2).  The chlorine used
in wastewater disinfection also contains traces of bromine and subsequently
small quantities of the respective bromooxy acids.

     At low pH's the chlorate ion interacts with hydrochloric acid to form
Cl2 and C102> C102 being a relatively  stable and water-soluble free radical.
Since this process is one of the fundamental methods of producing C102s it
should be noted that the reaction forming C102 results in the production
of significant quantities of Cl2 via the following mechanism:

     4H+ + 2C1" + 2C10s"-^ Cl2 + 2C102 + 2H20

Another method of producing C102 involves the reaction of HOC! with sodium
chlorite to produce C1202 and  its  subsequent breakdown  into  ClOg and Cl2-(67)

     Therefore unless extraordinary measures are undertaken to separate the
Cl2 and subsequent oxychlorine species, a pure solution of C102 is not


                                   157

-------
i
                       TABLE  50.  CHLORINATED HYDROCARBON/PESTICIDE SCAN DURING DISINFECTION TESTS
         en
         oo

Storm
No.
Sample
Date Location
Disinfection
Dose
CHP
ua/1
Change Upon
Disinfectant Addition
Absolute Volume
1

3


4


5


6

6-16-76 Raw
Unscreened
7-08-76 Raw
Crane
Unscreened
7-29-76 Raw
Crane
Unscreened
9-10-76 Raw
Crane
Unscreened
10-08-76 Raw
Unscreened
Influent
Effluent
Influent
Effluent
Effluent
Influent
Effluent
Effluent
Influent
Effluent
Effluent
Influent
Effluent

C12

C12
C102

C102
C12

CIO?
C102

CIOz

= 20

= 12
= 11

= 3.
= 12

& Cl?
& C12

= 7.

mg/1

mg/1
mg/1

4 mg/1
mg/1

*
*

3 mg/1
59.
54.

4.
3.
17.
36.
39.
4.
13.
4.
1.
2.
0
0

1
7
7
6
6
6
8
5
2
5

- 5.

+ 0.
- 0.

+18.
+21.

+ 9.
- 0.

+ 1.

0

1
3

9
9

2
1

3
Percent Change

- 8.

+ 2.


+106
+123

+200
- 2

+108

5

5


.8
.7

.0
.2

.3
               *Disinfectant  dosages  applied as  follows:   2 mg/1  C102 followed by 8 mg/1  Cl2 after 15 sec.

-------
Ol
VO
                     .TABLE 51.   VOLATILE  CHLORINATED  ORGANIC   CONCENTRATIONS*  PRODUCED
                         FROM VARIOUS  DISINFECTION  SCHEMES APPLIED  TO  SIMULATED CSO

Disinfection
Scheme
8 mg/1 C102


12 mg/1 Cl2


2 mg/1 C102
followed by
8.0 mg/1 Cl2

Control
Reagent Blank
Contact
Time
1
5
10
1
5
10
1

5
10


min
mm
min
min
min
min
min

mm
mm


CHCla
5.
9.
5.
9.
6.
6.
6.

5.
5.
5.

-------
produced.  This will have a significant bearing when evaluating the
interaction  of both Cl2 and C102 with organic substrates  characteristic
of CSO.

     Chlorine and oxychlorine species react with organics  and organo-nitrogen
compounds via a wide range of mechanisms involving electrophilic substitution,
nucleophilic substitution, oxychlorine-catalyzed hydrolysis and free radical
mechanisms.  Some information exists which indicates that  in certain oxychlor-
ine solutions where C102 is dominant, C102 will react with various organic
constituents by an "oxidative hydrolysis" reaction mechanism whereas Cl2
results in chlorination via the chlorine-free radical mechanism.

     In an effort to determine the relative formation of volatile chlorinated
organic compounds as a result of the C12 or C102 disinfection of CSO,
several bench scale tests were conducted.  CSO was  simulated by
mixing equal parts of the Syracuse Metropolitan Sewage Treatment Plant
influent with distilled water and then subjecting equal aliquots with 12  mg/1
Cl2» 8 mg/1 C102 and a sequential addition system of 2 mg/1 C102 with 8 mg/1
Cl2 added after 25 percent of the total detention time.  Samples were
analyzed after 1, 5 and 10 min detention times via gas' chromatography
utilizing a conventional purge and trap technique.  Each sample was
analyzed for the following volatile halogenated organics:
     chloroform (CHCls)
     bromodichloromethane (BC12CH)
     chlorodibromomethane (ClBr2CH)
     bromoform (CHBra)
carbon tetrachloride (CC14)
tetrachloroethylene (C2C14)
1,1,1 trichloroethane
1,1,2 trichloroethylene (ClsC2H)
     The volatile halogenated organic analytical results are presented in
Table 51 for each evaluated disinfection scheme and contact time.  The
results indicate that tetrachloroethylene is the most significant volatile
chlorinated organic produced by application of either C102 or Cl2.  The
second most significant volatile chlorinated organic produced in the course
of disinfection is chloroform.  The concentrations of 'both chloroform and
tetrachloroethylene in the CSO control are significant with approximately
5.1 ug/1 of CHCla and 8.1 yg/1 of C2C14.

     If the control measured concentrations of each of the volatile chlorina-
ted organics are subtracted from the concentrations measured on each of the
disinfection systems, it can be seen that only low levels of volatile
chlorinated organics are produced in the C102 system.  The production of
chlorinated organics in the Cl2 and C102/C12 systems is more significant
with higher concentrations measured in those samples reflecting longer
detention times.

     The results cannot be considered conclusive; however, they do tend to
indicate advantages in the use of C102 as a disinfectant relative to
reduced volatile chlorinated organic production.  It is important to note
that Cl2 and oxychlorine aqueous systems have very complex equilibria systems
which result in measurable concentrations of C102 being present in Cl2 solutions
as well  as measurable concentrations of Cl2 present in C102 solutions.
Because of this latter phenomena and the variability in precursor concentra-
                                     160

-------
tions in wastewater samples, it is difficult to conclude that the application
of C102 to CSO disinfection would be without volatile chlorinated organic
byproduct formation.

CIO? Sensitivity

     The C102 used for conducting many of the disinfection tests at both
Syracuse demonstration sites was generated on-site by means of a Nitrosyl
Chloride system (U.S. Patent 3754079, Chemical Generators, Inc., Rochester,
New York).  The process consisted of pumping two batches, sodium
chlorate-sodium nitrite slurry and nitric acid, into a specially designed
lucite chamber where the chemicals were mixed.   The resulting reaction
was expected to produce a 12 percent solution of C102 to be fed directly
to  the microscreened CSCl.

     The two solutions of chemical reagents were pumped into the specially
designed mix chamber by duplex pumps mounted on the base of each generator.
The strengths of the reagent solutions were such that volumes had to be
mixed at a 1:1 ratio to produce the 12 percent solution of C102-  The 12
mg/1 dosage feed to the microscreened CSO was to be held constant regardless
of total effluent flow by controlling the pump motor rotational velocities
with 4-20 ma DC signals from each screening system flowmeter.  The
micrometer setting on each chemical feed pump could be adjusted to provide
a smaller C102 dose to the microscreened CSO, if desired.

     At the start of the project, the duplex pumps for each of the
generating systems were calibrated to determine the exact volumes of
liquid pumped to the microscreened CSO.  The pumps were adjusted to
deliver equal volumes of reagent to the mix chamber to meet the 1:1 ratio
mix criteria.  The pump calibration curves were checked periodically
throughout the project.  Subsequent to mixing of the two reagents, the
generated C102 flowed by gravity to the disinfection tank.

     Several problems were encountered during the project requiring
constant attention.  The first major problem was failure of the duplex
pumps to function properly.  Minor malfunctions such  as the plugging of
check valves and leaks in the piping arrangements prevented operation of
the units during the first stages of the project.  In addition, major
electrical problems occurred which made pacing of the C102 generators
extremely difficult and forced abandonment of automatic pacing.  Instead,
the units were operated manually and the resulting C102 feed dosages
were determined for the varying CSO flowrates.

     Serious questions arose as to the strength of the C102 being generated
as a result of the frequent, short duration operation of the facilities
and subsequent reagent pump warrn-up time, corrosion of the lucite chambers,
and the containment time in the pumping system after formation of the C102-
Since immediate response of the facilities was necessary when an overflow
event occurred, warm-up time of the reagent pumps became a critical factor
in producing C102-  Approximately one-half hour was necessary for warm-up
                                    161

-------
during which time the 1:1 ratio was not being achieved and the strength of
C102 became less than optimal.

     In addition, a more serious question of C102 strength at the point of
disinfectant' injection to the microscreened CSO became evident.  Depending
on the CSO flowrate, the volume of C102 being fed to each system varied
from a mean low of 20 to 40 gpd (75 to 150 I/day) to a high of 450 gpd
(1700 I/day).   These flowrates resulted in containment times in the
disinfection piping system of from 30 min to 2 min, respectively.  A
study of the effects of  the stability of C102 with time had been
conducted in another study (7) and results indicated that C102 deteriorated
by, as much as 40 percent during the first 5 min after generation.  Thus,
containment times from the point of generation to point of injection
became very significant.

     Since the containment time in the initial C102 generation piping was
significant, a technique for more rapid delivery of disinfectant to the
treated CSO was considered necessary.  After consultation with Chemical
Generators, Inc., a new mixing chamber was installed on each of the four
C102 generators at the two demonstration sites.  The lucite chamber was
replaced by a 4 in. (10 cm) diameter glass tube affixed to the front of the
C102 generator panel.  The two reagents were pumped into one end of the
glass tube at the desired 1:1 ratio.  In addition, water was pumped into
that same end of the glass tube to increase the flowrate from the tube to
the point of injection.   A Jabsco pump at a rated capacity of 3 gpm
(11.3 1/min) pumped water out of a small storage box constantly refilled
from the city water supply by a mechanical float control mounted inside
the box.  The 3 gpm (11.3 1/min) pump delivered approximately 1 gpm
(3.8 1/min) to each of the three C102 units at Maltbie Street and a total
capacity of 3 gpm (11.3 1/min) at West Newell Street.  Containment times
for various flowrates were thus reduced from 2 to 30 min to 30 to 40 sec at
Maltbie Street.  At West Newell Street the containment time was reduced to
10 to 15 sec.

     In addition, for a portion of the study, the C102 was generated at a
strength of only 4.2 percent instead of  the desired 12 percent.  It is
theorized that the effects of sunlight on the reagent storage tanks may
have resulted in a loss of strength on either or both of the reagents
thus resulting in decreased production of C102-
     Two methods of assessing ClOg presence were considered in this project.
The first is the DPD technique (N-N-diethyl-p-phenylene diamine) (68)
which has the capability of determining specific concentrations of C102,
chloramines, C102, or Cl2-  However, the concentrations of such constituents
must be below 2 to 4 mg/1 to consider the analysis accurate.  Since the
C102 strength desire'd in the field tests was supposed to be a 12 percent
solution or 120,000 mg/1, the DPD technique was not considered suitable for
accurate use since a lengthy procedure of diluting the original sample to
2 to 4 mg/1 would have to be developed.  Such a procedure is difficult in
the field where conditions are rapidly changing and thus not conducive to
good accuracy.


                                   162

-------
     The second method of analyzing C102 strength in the field is the
starch-iodide titration method (68).   This technique has the disadvantage
of measuring 0102* chloramines, and Cl2 as a group.  Thus no direct
determination of C102 strength is made with this technique.   The most optimal
solution to the measurement of C102 strength as produced in  the field would
be a DPD titration to determine if the generated solution is predominately
C102 followed by the starch-iodide technique to determine the actual
strength of the solution.

     Field analyses of the C102 generated at the Maltbie Street and West
Newell  Street sites were based on the starch-iodide titration method.  Since
the solution produced was expected to be predominantly ClO^s the measurement
by starch-iodide titration was assumed to yield a fair indication of the
C102 strength.   The field analyses for C102 strength followed the procedure
for the starch-iodide method outlined in Standard Methods.  Field samples
analyzed were collected at a point just prior to injection into the CSO at
the contact tank.  Although considerable evaluation and confirmation of
C102 activity in simulated CSO had been conducted on a bench-scale level,
using electron spin resonance techniques (7), confirmation of C102 dosages
applied during the full-scale demonstration study was not attempted due to
the sophistication and expense of the electron spin resonance equipment.

SPECIAL INVESTIGATIONS

Ultraviolet Disinfection

General--
     At CSO treatment facilities, where long Cl2 detention times are not
feasible for adequate disinfection, a bactericidal method or combination
of methods requiring minimum retention time is desired.  The bactericidal
properties  of ultraviolet radiation for certain applications are well-
established, and there are several advantages of ultraviolet (UV)
disinfection over chlorination that make UV disinfection very attractive.
The major advantage of UV radiation is that disinfection occurs without
addition of any chemicals to the waste stream that may result in harmful
residual compounds.

   '  A series of experiments were performed to determine the feasibility of
disinfecting simulated CSO using UV radiation alone and in conjunction with
Cl2-  The various combinations of disinfection tried were:

     1)   UV alone
     2)   Cl2 alone
     3)   UV followed by Cl2
     4)   Cl2 followed by UV
     5)   Simultaneous UV. and C12

The parameters that were varied were:

     1)   Cl2 dosage
     2)   Cl2 contact time


                                     163

-------
     3)   UV radiation time which affects power usage
     4)   Sample volume

Bench Scale Program—
     The samples used were raw sewage samples collected each Monday
between 8:30 and 9:30 AM at Metro.  Each sample was screened through a 74
micron mesh screen and diluted with an equal volume of deionized water.
Portions of sample which were left over from  the Monday experiments were
refrigerated; before reuse they were brought back to room temperature.

     The UV source was a PCQ 9 6-1,7 in.(17.8 cm)  photochemical  immersion
lamp manufactured by Ultraviolet Products, Inc., San Gabriel, California.
The intensity was calculated to be approximately 5,800 yW/cm2.

     Chlorine solutions were prepared by diluting chlorox 1:10 to yield a
concentration of 5.5 mg/1 Cl2-

     One liter samples were irradiated in a bell jar reaction vessel which
was designed so that the walls of  the flask were equidistant (  2.25 in.  )
(5.7 cm) from  the UV lamp.  One half liter samples were irradiated in a
plastic one liter graduate cylinder which was modified to accept the UV lamp.
100 ml samples were irradiated in a 100 ml glass graduated cylinder equipped
with a ground glass joint.  Each vessel was covered with aluminum foil to
increase the intensity of the UV radiation.  Figure 74 presents  a schematic
of the UV disinfection test apparatus.

     The prepared sample was introduced into the reaction vessel  and
stirred via magnetic stirring bar for one minute prior to any treatment.   The
action of the Cl2 was arrested by adding, at the appropriate time, sodium
thiosulfate directly to the reaction vessel or by introducing the sample to
a sterile collection bottle containing the thiosulfate.   The UV lamp was
used with no prior warm-up and was controlled by a switch.

     Samples were withdrawn from the reaction vessels via sterile 10 ml
pipettes and collected in sterile bacteria bottles.  In some instances,
1 ml samples were withdrawn  and directly pipetted into bacteria dilution
bottles.  The solution was continuously stirred during the collection of
the sample.

Disinfection Results—
     UV Only— UV disinfection of wastewater is a function of the lamp
intensity, measured as microwatts per square centimeter (yW/cm2), contact
time, measured in seconds and distance from the UV source.  The  product of
the lamp intensity and contact time is expressed as yW-sec/cm2.

     The UV source used in these bench scale studies was approximately
one-sixth as powerful as the lamps used in commercial UV disinfection units.
For that reason, the percent kills of bacteria were related to yW-sec/cm2
rather than to contact time.  The experimental data was plotted  as Number
of Surviving Colonies per 100 ml vs yW-sec/cm2 and the curve was
extrapolated to determine the power required to disinfect a one  liter sample
to a target level of less than 2,400 counts TC/100 ml, (see Figure 75).

                                    164

-------
O1
O1
         Sampling  Port
                                                     Power
                                                     Supply
Liquid  Level
                                            UV Immersion Lamp
                                            -Magnetic  Stirrer
                      Bell Jar for I  Liter  Samples
                                                    100 or 250ml Graduated  Cylinder
                                                    for  Samples  Less Than  I  Liter
                  FIGURE 74.  Schematic of Bench-Scale UV Disinfection Testing Apparatus

-------
FIGURE 75.  Number of Surviving  TC vs yVl-sec/cm2
                       166

-------
A value of approximately 500,000 per liter (equivalent to 2.3X1012 yW-sec/cm2/
MG {8.7X1015 yW-sec/cm2/cu m}) was obtained.

     Although 500,000 yW-sec/cm2 per liter are required for the desired level
of disinfection, a plot of percent inactivation vs yW-sec/cm2 for 1 liter
samples (Figure 76) reveals that 50 percent inactivation is obtained with
only 58,000 yW-sec/cm2 and 75 percent inactivation is obtained with
116,000 yW-sec/cm2.  To achieve an  additional 24.9 percent inactivation
required greater than 350,000 yW-sec/cm2.   This data suggests that the
disinfection of TC bacteria with UV radiation, under static conditions,
is dependent only on the concentration of the bacteria, and is, therefore,
a first order process.  However, it was not in the scope of these bench-
scale tests to evaluate the effects of such parameters as  color,  turbidity,
and solids levels which would affect UV disinfection performance.

     The effect of contact time and lamp intensity for the experimental
apparatus is presented in Figure 77.  The analysis of surviving TC as a
function of radiation time indicates a logarithmic relationship.  Similarly,
a logarithmic relationship is observed between lamp intensity and surviving
TC.

         Only-- Disinfection with Cl£ was performed only to serve as a
comparison to UV disinfection and to combined UV and Cl2 disinfection.
Results of Cl2 disinfection are discussed in the bench-scale study report
(7).

     UV Followed by Cl^-- Irradiating the sewage mixture for 5 sec
(29,000  yW-sec/cm2)with UV light resulted in an average TC inactivation
of 46 percent. Chlorination of the UV radiated mixture with 5.5 mg/1 Cl2
for 10 sec inactivated 8 percent of those bacteria which survived the UV
irradiation, whereas a 30 sec contact time inactivated 60 percent of the
surviving bacteria.

     Doubling the UV radiation time, while keeping the Cl2 dosage and
Clg contact time constant at 5.5 mg/1 and 30 sec respectively, increased
the C]2 efficiency by only 2 percent (Figure 78).  On the other hand,
doubling the Cl2 concentration to 11.0 mg/1, while keeping the Cl2
contact time and the UV radiation time constant, increased the percent
inactivation of UV surviving bacteria from 60 to 90 percent (see Figure
79).

     This suggests that extending the UV radiation time does not significantly
enhance the efficiency of C]2 at concentration levels such as 5.5 mg/1,
and that following UV irradiation the percent inactivation is more dependent
on the subsequent Cl2 concentration rather than on the UV radiation time.

     Chlorination Followed by UV— This series of experiments determined the
effect of UV irradiation after a sample had been chlorinated.  During the
experiments, the action of the Cl2 was not neutralized by addition of
thiosulfate when the UV radiation began; therefore, the Cl2 and UV were
acting on the sample simultaneously.
                                    167

-------
100 r-
    i  i  i i  i i  i  i i  i  i i  i  i i  i  i i  i
                              pw-sec/cm



    FIGURE 76.  Percent Inactivation of TC  vs
                            168

-------
  ao
IRRADIAT
                           30       40
                          TIME,  sec
                                                    60
FIGURE 77.   Effect of Ultraviolet  Light Intensity on  Disinfection
                               169

-------
                  ro
                  O
 s
T
                            Inactivation  of Total Coliform , Percent
a<
o
a>
o
CD
o
(0
o
                                                   n - r
                                         _.  60% Kill of Surviving Bacteria
O
O
                                                                       Constants*

                                                                         CI2 Cone.

                                                                         CI  Time
                                            30 sec of 55mg/l Cl
                                                   62% Kill of Surviving Bacteria
                                                    30 sec of5.5mg/l Cl
          FIGURE  78.   Effect  of  UV  Exposure  on  Cl2  Disinfection
                             Inactivation of Total Coliform, Percent
5 O O O O
1 1 1 1

5 sec UV Radiation


5 sec UV Radiation

SO) -*] 00 t£) O
O O O O O
1 1 1 1 1 1
Constants1
UV Energy
Cl2 Time
> 60% Kill of Surviving Bacteria

30 sec Contact of55mg/ICI
90% Kill of Surviving Bacteria |

30 sec Contact of llmg/l Cl

FIGURE  79.   Effect of  C12  Dose on  Disinfection  Following  Ultraviolet
               Light  Exposure
                                          170

-------
     Experimental data showed that higher levels of inactivation were
obtained when, regardless of the Cl2 concentration, longer Cl£ contact
times were in effect prior to UV .irradiation.  It was also noted that the
percent inactivation was more dependent on the Cl2 concentration than on
the UV energy supplied; in essence, greater kills were achieved by increasing
the Cl2 dosage rather than increasing the UV radiation time.

     Simultaneous UV and Cl2—• The results of simultaneous disinfection of
the combined wastewater with Cl2 and UV was also evaluated.  Figure 80
presents a comparison of the results of the effectiveness of C12 versus
simultaneous UV and Cl2 disinfection as a function of contact time,.   It
can be seen that the effectiveness of Cl2 as a disinfectant is significantly
enhanced by the application of UV radiation.  Simultaneous disinfection was
more effective in that it resulted in a higher rate of disinfection, producing
slightly higher levels of inactivation at lower contact times than the
corresponding single disinfecting components as indicated by increases in
the slopes of the curves presented in Figure 80.

Review of UV Disinfection Equipment—
     The most recent UV disinfection modules consist of sterilizing chambers
which house the high intensity UV lamps.  Depending on the manufacturer
these lamps may be a low pressure mercury vapor lamp, or a hot cathode
tungsten filament lamp.  These lamps do not actually come in contact with
the water since they are housed in specially constructed quartz sleeves.
The purpose of the quartz sleeve is to act as a temperature buffer, so that
the lamp may operate at approximately  105°F (41°C) regardless of the water
temperature.  A second feature of the quartz sleeve is that it protects
the lamp from a buildup of solids.  Many units on the market are equipped
with a device which automatically wipes the quartz jacket clean in the event
of any debris buildup.

     The water enters the module and, by means of baffles and/or helical
discs, is made to travel around and over the quartz sleeves, thereby,
increasing the time in contact with the UV radiation.  These baffles and
discs preclude the possibility of water entering and leaving the module on
a plug flow basis.

     Most commercial UV disinfection units provide in excess of 25,000 yW-sec
of ultraviolet radiation at a wavelength of 2540ft, whereas, most bacteria
require between 6000 to 13000 yW-sec of UV light at 2540A for complete
destruction.  According to manufacturer's information, lamp life is approxi-
mately 7500 hr assuming a lamp on-time of 8 hr/day.

     More intermittent operation reduces the lamp life due to the
degradation of the lamp caused by startup and shutdown.  Based on an average
12 hr/day operation, a UV lamp would need to be replaced in approximately
24 months.   Two years is the maximum time recommended for a lamp to be
used in intermittent operation because of various deterioration and
degradation factors which take place.  After extensive-use of the UV lamp,
the glass solarizes and starts absorbing some of the 2540& light, thus
reducing the amount of light available for disinfection, and reducing its
efficiency.  Replacement Tamps are available from $50 to $200 depending on the
size.

                                    171

-------
                                     Smuttojwoiw UV
                                     a C12 OisJnftcHon -
               10
          20       30       40

            REACTION TIME, sec
                                                  50
                                                           60
FIGURE  SO.
Surviving TC vs  Reaction Time for-Simultaneous
Disinfection by  UV and
                               172

-------
     The efficiency of UV disinfection is reduced as the concentration of
UV absorbing material in the water is increased.  Suspended and colloidal
matter will interfere with disinfection by absorbing the UV light.  Also
dissolved matter, such as organic matter and iron will absorb UV radiation.
Dissolved alkali metals are known to absorb UV light but to a much lesser
degree than iron salts.

     To eliminate any of the adverse effects of extraneous matter found in
CSO, the influent to the UV disinfection system should be pretreated by
chemical coagulation and filtration or some reasonable equivalent.

Comparative Cost Analysis of Disinfection Systems—
     Based on the results of the bench scale investigations a cost estimate
was developed for UV and combined Cl2 and UV disinfection systems.  Estimates
were also prepared for Cl2 and combined Cl2 and C102 disinfection systems
based on results presented in Sections 11 and 12.  The capital and operating
costs for all systems are presented in Tables 52-55, and have been developed
for a design flow of 15 MSD in accordance with comparative cost relation-
ships established in Section 12.  A summary of costs for all disinfection
systems is presented in Table 56.

     Annual costs were not developed since these investigations represented
bench-scale applications, and the purpose of presenting capital and operating
costs is merely for basic cost comparison.   Although it appears that
disinfection by UV is not cost-effective when compared to Cl2> consideration
has not been given to the toxicity effects from Cl2 residuals which may
require dechlorination and consequently would increase the cost of Cl2
application.

Adenosine Triphosphate Assay for Disinfection Control

     The present methods of determining the efficiency of wastewater disin-
fection processes and controlling the addition of disinfectants- may not be
adequate to meet recently enacted standards of 200 FC/100 ml of sampled
water (69) without the risks involved in the presence of excess disin-
fectants (70).  The traditional approach to the assessment of disinfection
efficiency has been the enumeration of the coliform bacteria, especially
fecal coliforms, which are generally accepted as an indication of contamina-
tion from human or animal sources and, consequently, the presence of
pathogenic bacteria and viruses (71).  The principal objection to this
approach is that one of the most common and rapid methods of bacterial
enumeration, the membrane filter technique, requires a 24 hr  incubation
period, during which incomplete disinfection or the addition of excess
disinfectants may occur in the system.  A second objection is that coliforms
may not be as resistant to disinfection as some pathogenic microorganisms,
therefore an inadequate measure of disinfection efficiency (72).

     The most common method of controlling disinfectant dosages is the
monitoring of residual disinfectant to maintain a fixed level in the treated
effluent (73), a method that admittedly has its limitations (74).  Other
methods may include addition of disinfectant based on historical flow patterns


                                    173

-------
	TABLE  52.  ULTRAVIOLET SYSTEM COSTS - 15 M6D CAPACITY
Capital Cost
a)   American UV Unit*
          1.7 MGD unit @                                    $100,000
          9 units required for 15 MGD                         x    9
          Total                                             $900,000
b)   Sterilaire Unit**
          0.22 MGD unit @ $2,500 extrapolated to
          15 MGD                                            $170,000
          Compensation for solids and turbidity
          (Number of units tripled)                            x   3
          Total                                             $510,000

Operating Costs/Day
a)   American UV Unit*
          (9 units)(l,280W/unit)(24 hrs)($0.02/KWHR)        $5.55
                                   day
          Total                                             $5.55
b)   Sterilaire Unit**
          78.3 MG treated @ cost of $105.00 extrapolated
          to 15 MG                                         $20.20
          Total                                            $20.20
* American Ultraviolet Company, Chatham, New Jersey
**Ultraviolet Products,  Inc., San Gabriel, California
                                    174

-------
	TABLE .53.   COMBINED Cl2 & UV SYSTEM COSTS - 15 MGD CAPACITY


Capital Cost

UV component*:  (35 units) x ($l5600/unit)                  $  56,000

Chi orination component:                                        37,000

Total                                                       $  93,000


Operating Costs/Day

UV power:  (40 units)(280 W/unit)(24 hrs/day)($0.02/KWHR)   $   5.38

Chlorine:  (8 mg/l)(15 M6D)(8.4)($0.10/lb)                    100.80

Total                                                       $ 106.18
*Sanitron Modular UV Disinfection Unit (Atlantic Ultraviolet Corp., Bay Shore,
 N.Y.)

     Price:                   $1,600
     Capacity:                 5,000 gph
     Power Requirements          280 W
     Contact Time              13 sec      2
     Intensity                 30,000 yW/cm   2                      ?
     Disinfecting Power**(13 sec) (30, 000 yW/cm )  = 390,000 yW-sec/cm

     **Bench scale studies revealed that 87,000 yW-sec/cm  (15 sec x 5800
        of UV light alone resulted in a 95 percent kill of TC bacteria.
Since the UV would be used as a polishing step following chlorination, it
is felt that a disinfecting power of 100,000 yW-sec/cm2 is sufficient to
achieve the desired TC kills and the capacity of this unit, which produces
390,000 yW-sec/cm2, may be increased by a factor of 3.9, (available power/
required power).  Therefore, 35 such modular units would be necessary to
handle 15 MGD (39.5 cu m/min).
                                     175

-------
	TABLE  54.   CHLORINE SYSTEM COSTS - 15 MGD CAPACITY

Capital Costs:
     Chlorine Equipment                              $15,000
     Holding Tank                                      7,000
     Appurtenances                                    15,000
     Total                                           $37,000

Operating Costs/Day:
     Chlorine:  (8 mg/1) (15 MGD)  (8.4) ($0.10/lb) = $100.80
     Electricity                                   =    2.20
     Total                                           $103.00

     TABLE  55.   COMBINED Clp AND  CIO? SYSTEM COSTS - 15 MGD  CAPACITY

Capital Cost:                                        $37,000
     Chlorine Dioxide &
     Chemical Generator                                5,000
     Total                                           $42,000
Operating Cost/Day
     Chlorine (15 MGD)(8 mg/1)(8.4)($0.10/lb) =      $100.80
     Chlorine Dioxide: (15 MGD)  (2 mg/1)  (8.4)
     ($.50/lb) =                                      126.00
     Electricity                                        4.20
     Total                                           $231.00
                                    176

-------
  TABLE 56.   SUMMARY OF COSTS FOR DISINFECTION SYSTEMS - 15 MGD CAPACITY


                                                       Operating Cost*
System	Capital Cost	Per 15 MGD Flow

Chlorine                      $37,000                       $103.00

Combined Chlorine
& Chlorine Dioxide             42,000                        231.00

Ultraviolet                   900,000 High                    20.20
                              510,000 Low                      5.54

Combined Chlorine
& Ultraviolet                  93,000                        106.18


^Operating Costs Based" on":

     Chlorine 8 mg/i cone, and @ $0.10/lb
     Chlorine Dioxide 2 mg/1 cone, and @ $0.50/lb
     UV Power Costs @ $0.02/KWHR

and periodic bacterial counts.  These methods may be acceptable for domestic
wastes in which the microbial variations are somewhat predictable, but
totally unacceptable for wastes such as storm and CSO in which these
variations are large and unpredictable.  The objection is that simply
maintaining a fixed volumetric residual does not guarantee that bacterial
and viral populations have been reduced to a safe level.  A more direct
reflection of microbial activity than residual disinfectants is needed.

     The quantitative determination of adenosine triphosphate (ATP), using
the bioluminescent reaction characteristic of fireflies, offers a potentially
rapid alternative to bacterial measurements as an indicator of microbial
content of a water sample (75).

     McElroy (76) initially reported that an unreactive preparation of
two extracts from firefly lanterns, one presumably containing luciferin,
the other containing the enzyme luciferase,could be made to luminesce in the
presence of ATP.  This finding laid the foundation for the development of an
ATP assay which can be performed with relative ease and accuracy, and may
be utilized for the enumeration of microbial populations (77) or as a
measure of biomass (78).

     ATP is present as the driving force of bioenergetic reactions in all
living cells (79).  It is the primary phosphorylating agent for most
biochemical enzymatic reactions and, therefore, the primary energy source
in cellular metabolsim.  Only under rare conditions is ATP found in
nonbiological systems (80, 81).  ATP that is released by dying microorganisms
is rapidly acted on by other organisms or the surrounding environment and
converted to dissimilar phosphate forms.  Therefore, an ATP determination
can be considered a measure of only the living organisms within a system (82).

                                   177

-------
This can be seen as a considerable advantage over the assay of proteins,
nucleic acids, organic nitrogen, or nitrogen/phosphorus ratios, all  of which
are relatively unaffected by the inactivation of the organisms.

     The chemistry of luciferin-luciferase bioluminescence has been
elucidated (83, 84, 85) and divided into four stepwise reactions where:

                          E = lucif erase (enzyme)

                        LH = luciferin (substrate)

                            AMP = adenylic acid

                E-LH-AMP = enzyme bound luciferyl adenylate

                             PP= pyrophosphate

                           L = dehydroluciferin

                             CoA = coenzyme A

                   E-L-AMP = dehydroluciferyl adenylate

                    L-CoA = dehydroluciferyl coenzyme A

                    E + ATP + LHM-^i E-LH-AMP + PP
     In the above equation the initiating rate-limiting step of the entire
reaction, luciferin must first react with ATP before it can be oxidized with
light production.  The ATP does not act as an electron donor for light
emission, but performs an unknown catalytic function, in some way altering
the excited state of luciferin (86).

                 E-LH-AMP + 02 — E-L-AMP + Light + ?

     In the presence of oxygen, the enzyme-bound luciferyl  adenylate
rapidly reacts to produce the excited intermediate which subsequently emits
light.   One quantum of light is emitted for each luciferin  molecule
oxidized, while one molecule of oxygen is used in the reaction (87).  The
identity of the- additional products formed is unknown.  The E-L-AMP is a
relatively stable complex and does not spontaneously disassociate to
regenerate lucif erase.

     The enzyme may be regenerated by either or both of two reactions:

                  E-L-AMP + CoA— E + AMP + L-CoA

                     E-L-AMP + PP-»-L + E + ATP

     The luciferin-luciferase ATP assay is a useful procedure in a variety
of fields:  the determination of biomass in limnology (88), oceanography
(89), and activated sludge control (90); the direct enumeration of bacterial

                                    178

-------
contaminants (82); in laboratory determination of microbial ATP levels,
particularly in growth studies (91) since changes in metabolic activity
are frequently characterized by accompanying variations in the ATP
concentration in the organism; and in the control of wastewater treat-
ment (92).

     The hypothesis tested in this study was that ATP.determinations
yield information similar to the measurement of traditional bacterial
indicators of microbial contamination with respect to disinfection
efficiency and control.

Experimental--
     All samples were blended for a minimum of 6 sec  in a Hamilton
Beach Model 50 blender to expose bacteria which may have been harbored
within particulate matter and unavailable for subsequent recovery and
enumeration (7).  Bacterial analyses for TC, FC and FS were conducted
using the standard membrane filter technique (71).

     The natural virus levels in CSO were sufficiently low that the effects
of disinfection could not be easily observed.  Therefore, samples were
seeded with the desired test virus before disinfection to raise initial
levels.  Viral titers were determined by the plaque assay technique after
concentration on a 0.45y pore size membrane filter (93).

     The Du Pont Model 760 luminescence biometer was used for the detection
and quantitation of the light reaction (94).  All  reagents and supplies
were obtained from DuPont with the exception of morpholino-propane sulfonic
acid (MOPS buffer)(Aldrich Chemical Co.).

     The biometer was calibrated using ATP standards prepared following
detailed instructions in  the instrument operations manual (94).  ATP was
extracted from wastewater by filtering 5 ml of sample through a 0.45m pore
diameter membrane filter (Millipore Corp.), discarding the filtrate.  One
ml of a mixture of 1-butanol (80 percent) (Fisher Scientific) and MOPS
buffer (20 percent, 0.01 M) was added to the filter to lyse the cells.
This was followed by two 2.5 ml rinses of MOPS buffer to elute the
released"ATP.  After 20 samples had been extracted by this procedure and
stored on ice, the ATP was assayed.  The assay technique involves the
injection of 10 yl of the ATP, either as a standard or unknown, into a
light-protected cuvette containing MgS04 (0.01 M) buffered luciferin-
luciferase mixture (100 vl).  The magnitude of the resulting light
reaction is measured by a photomultiplier and indicated as a digital
readout.  The instrument was calibrated before and after each set of
assays to assess stability.

     Mean  values of ATP- per cell for a variety of bacterial species and
other microorganisms have been determined in prior studies (Table 57), and
an average value of 5.02 X 101U ug ATP/bacterial cell may be assumed based
on this data (82, 94, 95).
                                    179

-------
     Correlations between bacteria, viruses, and ATP were determined to
investigate the feasibility of using ATP as a substitute for bacterial
enumeration and the measurement of residual disinfectant concentrations.

                    TABLE 57.   ATP CONTENT OF BACTERIA

Organism
Aerobacter aerogenes
Aerobacter aerogenes
Bacillus cereus
Bacillus cereus
Bacillus coagulans
Bacillus subtil is
Bacillus subtil is
Clostridium sporogenes
Corynebacterium striatum
Erwina carotovora
Escherichia coli
Escherichia coli
Escherichia coli
Flavobacterium arborescens
Klebsiella pneumoniae
Micrococcus lysodeikticus
Mycobacterium phlei
Proteus vulgar is
Proteus vulgaris
Pseudomonas aeruginosa
Pseudomonas fluorescens
Pseudomonas fluorescens
Sarcina lutea
Sarcina lutea
Serratia marcescens
Staphylococcus albus
Staphylococcus aureus
Staphylococcus epidermidis
Streptococcus faecal is
Streptococcus salivarius
yg ATP X 1010/cell
2.4
0.28
6.4
1.1
1.7
24.
9.9
2.1
49.
0.44
5.8
4.1
1.0
1.5
5.0
1.3
1.9
3.0
1.8
1.0
3.9
3.1
2.2
0.37
1.0
3.1
0.64
2.2
4.9
5.7
Ref
94
95
94
95
95
82
94
82
82
95
82
94
95
95
95
95
95
82
95
95
82
95
82
95
95
94
95
82
94
82
                                    180

-------
     Samples from the bench-scale studies were collected after the appropriate
contact time, and the residual  disinfectant was immediately neutralized with
sodium thiosulfate (71).   Samples from the full-scale facilities were
collected automatically by sequential, refrigerated samplers (Sigmamotor,
Inc., Middleport, N.Y.).   Sample bottles were thoroughly cleaned, and an
excess of sodium thiosulfate was added before use.

Results and Discussion—
     For ATP to serve as  a measure of disinfection, the disinfecting agents,
in this case Cl2 and C102»must not interfere with the bioluminescent reaction.
To study this possibility, Cl2 and C102 were added to a standard ATP solution
to give disinfectant concentrations from 0.0 to 100.0 mg/1  ; a range designed
to include bactericidal doses at rapid contact times.  The ATP content of
each solution was measured after 60 and 300 sec contact time followed by the
application of sodium thiosulfate to eliminate residual disinfectant.

     The results, presented in Tables 58 and 59  indicate a definite
interaction between the reaction mixture and disinfectants, particularly C12-
At  the doses of C102 used for normal or high-rate disinfection, 25 mg/1 or
less, there would appear to be only a slight interference for 60 sec exposure
(Table59 ).  The corresponding interference caused by Cl2 is considerably
greater.  In the analysis of wastewater samples, this effect would be less
than shown because the ATP becomes exposed only to the residual  Cl2 and C102
remaining after the Cl2 or C102 demands of the wastewater are met.  However,
to eliminate up to 100 mg/1 of residual  disinfectant, sufficient sodium
thiosulfate (2500 mg/1) should be added to the MOPS buffer used in all stages
of the analysis.  This is common practice to neutralize disinfectant reactions
prior to bacterial analysis, preventing inaccurate bacterial kills by
residual disinfectant.  To test the effect of sodium thiosulfate on the
reaction mixture, various amounts were added to an ATP standard solution and
the ATP concentration assayed after 60 and 300 sec.  The results in Table
60 show that the effect of sodium thiosulfate on the ATP assay system can be
considered negligible.  This procedure was routinely adopted for ATP
measurements of disinfected waters.

     An attempt was made to correlate measured ATP with the presence of
test bacteria." This was  performed in two stages; first, on untreated
samples from CSO, and second, on disinfected samples.  Samples of untreated
CSO from different times  and locations were analyzed for ATP, TC, FC, and
FS.  The correlation coefficients (Table 61) were derived from a least-
squares fit of over 250 points assuming a linear relationship, and indicated
little correlation between ATP and any of the bacterial parameters.  The
results are not unexpected.   The lack of correlation is attributable to the
presence of ATP sources in the wastewater other than the three measured
bacteria.  The results are confirmation of the small contribution the test
bacteria make to the total ATP pools of the individual wastewater samples.
Future work in this direction should include the correlation between ATP
and total bacterial levels.

     The second stage involved the measurement of ATP and test bacteria
following disinfection.  The study was conducted on a bench scale under a
variety of conditions.  Data were derived from a single-stage disinfection

                               -   .  181

-------
                               TABLE 58.    EFFECT  OF C1?  on ATP ASSAY MIXTURE

Contact time. Sec

Cl2 dose,
mg/1
0
5
10
15
25
50
100
0

ATP,yq/l
17.1
17.1
17.1
17.1
17.1
17.1
17.1
60

% Recovery
100.0
100.0
100.0
100.0
100.0
100.0
100.0

ATP,yg/l
16.80
13.60
9.80
5.17
1.04
0.84
0.25

% Recovery
98.2
79.5
57.3
30.2
6.1
4.9
1.5
300

ATP,yg/l
17.00
10.50
9.00
3.18
0.63
0.97
0.15


% Recovery
99.4
61.4
52.6
18.6
3.7
5.7
0.9

00
ro
                               TABLE  59.    EFFECT  OF  C102  on  ATP ASSAY MIXTURE

Contact time, Sec
Clz dose,
mg/1
0
5
10
15
25
50
100
0
ATP,yg/l
10.10
10.10
10.10
10.10
10.10
10.10
10.10
60
% Recovery
100.0
100.0
100.0
100.0
100.0
100.0
100.0
ATP,yg/l
9.63
8.87
9.95
9.50
9.40
7.11
6.53
% Recovery
95.3
87.8
98.5
94.1
93.1
70.4
64.7
300
ATP,yg/l
10.60
8.86
9.25
5.74
4.82
2.36
2.54
% Recovery
105.0
87.7
91.6
56.8
47.7
23.4
25'. 1

-------
                      TABLE 60.  EFFECT OF SODIUM THIOSULFATE ON ATP ASSAY MIXTURE
                                             	Contact time. Sec
Sodium
thiosulfate
dose, tng/1
0
500
1000
2500
5000
0
ATP,yg/l
15.4
14.1
16.2
15.6
16.0
60
% Recovery
100.0
100.0
100.0
100.0
100.0
ATP,ug/l
14.8
13.9
14.6
14.4
14.9
% Recovery
96.1
98.6
90.1
92.3
93.1
300
ATPiyg/1
12.2
11.6
12.0
12.4
12.1
% Recovery
79.2
82.3
74.1
79.5
75.6

CO
co
                   TABLE 61.   LINEAR REGRESSION CORELATION COEFFICIENTS OF ATP vs  TEST BACTERIA
                              Total Coliforms
            Range of ATP,  —-
               yg/1	Raw
          Disinfected
                         Fecal Col i forms
             Raw
        Disinfected
                                       Fecal Streptococci
                 Raw
          Disinfected
            0.00-0.10
            0.00-1.50
            0.00-5.00
            1.51-5.00
0.468
0.237
0.033
0.005
0.387
0.842
0.700
0.206
0.429
0.030
0.035
0.057
0.346
0.794
0.759
0.515
0.364
0.237
0.034
0.173

-------
procedure (disinfection utilizing a single-stage disinfectant, in this case
either Cl2 or CIOz) or by a two-stage procedure (the sequential  addition of
a single disinfectant, or two disinfectants).   Discrete data is  presented
in Appendix G .

     Table 61 describes the resultant correlation after linear regression
analysis between ATP and two bacterial indicators, TC and FC , upon
disinfection.   The higher correlations found upon disinfection  (than in
untreated samples) indicate that the fraction of the total wastewater
microorganisms represented by the coliforms may increase with disinfection.
The absence of this increase in the ATP range of 0.00 to 0.10 yg/1 is due
to the small number of data pairs in that sampling range, and the results
are probably not significant at that level.  The coliforms show greater
resistance to disinfection that that'fraction of the total microorganisms
present which contributes the greater portion of the total ATP measured.
Thus, an ATP level selected as equivalent to the ATP contained in the
highest limit allowable of indicator bacteria could more closely represent
the presence of such indicators, or more resistant pathogens, than initially
believed.  While they are necessarily vague in an absolute sense due to the
highly variable range of ATP from diverse sources in wastewater, the
correlations do indicate that the contribution of ATP by micrbbial sources
of nonsanitary significance decreases measurably upon disinfection.

     The results from the operation of the prototype full-scale facilities
are given in Table 62.  The CSO was diverted through a 7lvi aperture
screening unit prior to application of 7.0 mg/1 C102 for 1 min contact
time at a flow rate of 1 MGD.

     When the FC count was below the limit of 200 colonies per 100 ml, the
ATP level was lower than 0.1 ug/1.

     The data described to this point must be evaluated carefully.  The
primary concern is that ATP can be used to differentiate between the high
bacterial counts encountered in untreated overflows and the lower levels
required by effluent limitations.  Since the coliform group, the present
indicator of sanitary quality, represents only a small fraction of the
microorganisms found in untreated CSO, poor correlations with ATP were
expected and found.  However, if the ATP concentration of the treated
overflow is maintained at a level equivalent to the ATP concentration found
at the maximum permissible indicator bacteria levels, any decisions regarding
the water quality would be at least as valid as those based on the present
indicator, the coliforms.

     To verify the inactivation of pathogenic viruses using the disinfection
procedures,'poliovirus Sabin Type 1 was treated with C102- -The results of
the study (Figure 81) show  that the levels of C102 used and the time limits
imposed by the high-rate disinfection procedure still allow a substantial
inactivation of viral pathogens.  Thus, reduction of ATP values  observed
upon disinfection at test dosages indicates not only a reduction in bacterial
levels, but in the levels of pathogenic viruses.   Again, this reduction
should not be construed as a 1:1 relationship between disinfectant dosage
and reduction of ATP levels, nor of ATP concentration and any microbial specie
                                    184

-------
  10°
  I07
  10°
  10"
u.
Q.
M~I04
 I
o
_l
o
CL
  .o3
   10
   10
                                                           KEY

                                                       SYMBOL
0

8
                                                              I
           30    60     90     120    150     180   210   240    270
                                 TIME, seconds
   300
     FIGURE 81.   Single-Stage Disinfection of  Poliovirus Sabin Type I vnth
                  C102
                                       185

-------
upon disinfection.   The data observed do describe the utility of the assay
in measuring the large differences between influent and effluent populations.

     Before the development and testing of an in-line ATP monitor can be
considered realistically, the stability of reagents, adequate sampling
technique, availability of the necessary electronics, and the automation of
the chemistry must be investigated.  The feasibility of such a system has
already been mentioned (92, 96, 97, 98).  A continuous flow monitor is
envisioned with periodic switching from overflow samples to ATP for
standardization.  The control signal  for addition of disinfectants could
be generated from an ATP monitor located at a point following disinfection.
This arrangement would require that the ATP concentration be kept below some
predetermined safe level.  Another possibility would be to measure ATP
before and after disinfection to detect sudden increases in the degree of
contamination.  The electronics necessary to continuously record the light
intensity from the bioluminescent reaction have already been developed (99).
The procedures for a continuous, automated extraction are already in use
in several of the Technicon Methodologies (100).  The expense of the reagents
is likely to be quite significant because of the high cost of the luciferin-
luciferase reaction mixture derived from the firefly lanterns and should
most definitely be studied in greater detail before attempting to develop
a monitor.

Summary of Results-
     Concurrent determinations of ATP, utilizing the luciferin-luciferase
bioluminescent assay system, and indicator bacteria from CSO have indicated
the feasibility of using ATP as a reliable and rapid indicator to control
the disinfection process.

TABLE 62.  DISINFECTION OF COMBINED SEWER OVERFLOWS IN FULL-SCALE FACILITY

                     ATP,ug/1	         FC  Counts/100 ml
Time           Influent    Effluent          Influent    Effluent
1730
1750
1810
1850
1930
1950
2.50
1.25
1.44
1.32
1.79
1.44
0.260
0.300
0.066
0.101
0.036
0.051
315 000
460 000
190 000
84 667
83 500
62 000
615
1095
20
10
10
20

     In the dosages required for high-rate disinfection", Cl 2» C102, and
sodium thiosulfate did not significantly interfere with the ATP assay.
     The development of an instrument to monitor ATP in unattended operation
is a realistic venture primarily contingent upon the stability and cost
of the enzyme reagent.
                                    186

-------
SPECIAL INSTRUMENTATION

Telemetering

     The flow data that was generated at each of the demonstration sites
was telemetered to a central location, identified, and stored on punch
tape.  The tapes were then processed and the information permanently
stored, along with coinciding analytical and rainfall data, on computer
disc files.  The purpose of the telemetry and computer system was to
minimize manhours in data handling and to establish a basis for a central
control station for the overall CSO abatement program.

TOC Monitor

General —
     During the demonstration phase of the project, an investigation was  •
made at Maltbie Street of a previously developed total organic carbon
(TOC) analyzer (101) to monitor CSO in situ and on a continuous, rapid-
response basis.   The device was developed by Raytheon for evaluation
under EPA sponsorship.  Testing of the unit was conducted for a two month
period.

     The basic method of operation of the TOC monitor required that a
constant volume of the overflow be pumped to a homogenizer prior to entry
into the TOC unit.  This homogenizer was included to provide complete
maceration of solids in the sample stream and thus minimize the possibility
of plugging problems in the TOC unit.  A finite volume of the sample stream
was delivered by a small pump in the TOC unit to a 1812°F (950°C) oven
which provided complete combustion of the sample to form C02 from all carbon
compounds in the sample stream.   The remainder of the sample stream from
the homogenizer was discharged to a drain.  The values of TOC as measured
by the unit were recorded on a panel-mounted continuous recorder for visual
observation during a storm.  A back-up recorder was also installed inside the
rear of the cabinet for use in  case of failure in the panel-mounted recorder.
The TOC unit was calibrated by a standard solution of common sugar.

     The TOC monitor was equipped with an automatic zero and span exercise,
which was activated automatically when an overflow occurred.  A supply of,'
C02-free oxygen was delivered to the analyzer to establish a baseline
zero.  Next, a gas containing a known concentration of C02 was delivered
to the analyzer and the recorder scaled to this known concentration of
C02-   The zero gas used was obtained from the oxygen supply while the
span gas was purchased to specification.

Field Program—
     During the initial startup period,5 min was  required  to  automatically
zero and span the TOC instrument.  After that time period, data was collected
continuously on CSO samples delivered to the homogenizer.

     Calibration procedures were carried out about every third day to
assure proper operation.  Generally it was observed that the unit would not
                                    187

-------
hold to zero calibration level for an extended period of inactivity
(greater than two days).  This problem made it difficult to determine actual
TOC values when an overflow event was monitored.

     Figure 82 illustrates as three examples the failure of the TOC unit
to return to the zero level after exercising.  Had a storm occurred when
the calibration was off, erroneous TOC readings would have resulted.  An
effort to determine the percent error that would have occurred was
attempted.  The TOC value above the zero point in each case (A) was
subtracted from the actual recording to determine the TOC level measured.
The latter values were compared to the TOC value that should have been
registered on the chart recorder.  For example, for data of June 20, 1975,
a sucrose standard of 276 mg/1 was injected into the TOC unit.   A
resulting TOC value of 230 mg/1 was recorded on the chart.   Once the
baseline TOC value of 25 mg/1 was subtracted from the recorded value of
230 mg/1 the resulting value of 205 mg/1 indicates the TOC measured in
the unit.  The difference between the standard of 276 mg/1 and the 205
mg/1 measured was compared to the standard (276 mg/1) to yield a percent
of error for the three illustrations presented here.   The variability of
percent error indicated in Table 63 from 9.4 to 25.7 percent indicates
the unreliability of absolute values being measured.
         June 16, 1975
                            June 18, 1975
                                                     June 20, /975
    FIGURE 82.  TOC Monitor Data Obtained During Calibration Procedures
Results--
     As a result of several mechanical failures  to  the TOC monitor
during the two-month investigation  (including  replacement of  the  1812  F
(950 C) furnace, replacement of tubing, switches and  hose lines,  failure
of the recorder clock, and required replacement  of  a  solenoid valve)
only data collected from one of the five  storm events occurring during
this period was considered to be a reliable example of the TOC monitor's
capability.  A plot of the TOC as monitored vs results from laboratory
analysis of the actual overflow is shown  in Figure  83.   As can be seen
from the plot, the Raytheon unit only roughly  followed the trend  of
TOC changes in the overflow. Through  the  method  of  linear regression,  the
                                    188

-------
degree of association between the random points as expressed by the
correlation coefficient indicated no reliable association existed
(r = 0.10).

     Data observed during one storm event indicates that the technology
exists for developing a reliable TOC monitor, although the Raytheon unit
needs further refinement to provide a unit more reliable in terms of
operational response and calibration tracking.
       TABLE 63.   DEVIATION OF MEASURED TOC FROM KNOWN TOC STANDARD
(1) (2) (3)
TOC Reading Deviation Actual
on Full- From TOC
(4)
TOC OF Percent
Date Scale Chart Zero Scale Measured Standard Error
June 16 345 95 250
June 18 255 45 210
June 20 230 20 205
276 9.4
278 24.4
276 25.7

TOC Monitoring Results
Storm 10
200—
180 -
160 —
140 -
"£, 120 —
<§ 100 -
60 -
40 —
20 -
	
16
Key
£ — ^TOC Monitor
o — oLab TOC


/ A. .A. /


cs-Sr^^S^\/->^S>-^
> 1 I 1 > 1 ' 1 >
50 1700 1720 1740 1800 18
TIME
A 390
|\
l\
I \
\
\
^^J \
£•

O— °^L ^0
/^ 0— o--0
' 1 ' 1
20 1840 1900

              FIGURE 83.   Comparison of TOC Monitor Results with
                          Laboratory Results
                                    189

-------
Suspended Solids (SS) Meter

Introduction—
     The increasing emphasis on water quality has brought with it many
demands for improved information not the least of which has been for a
more accurate method of on-line suspended solids monitoring.  Present
state-of-the-art units can be broadly classified as those which use
gravimetric methods and those which use transmission, refraction or
reflection of light.  In almost all non-gravimetric designs, the light
source is non-polarized white light, and the detector senses the intensity
of received light, some ratio of received to transmitted light or the ratio
of intensities at two or more receivers.

     All of these devices are sensitive in varying degrees to many inter-
fering parameters.  Those parameters which are most influential are the
absorption coefficient, scattering coefficient, refractive index,
particle size, particle shape, particle density and dissolved solids
concentration.  Since each of the devices is sensitive to different
parameters in differing degrees, they tend to have calibration curves for
a given suspended solid which differ (often quite significantly) from
device to device.  Even more alarming is that the calibration curve for a
given device usually varies significantly from one type of suspended solid
to another.   Thus, unless a particular unit is calibrated specifically for
a particular suspended solid, it may have very limited use as a monitoring
device.  Most, if not all  existing monitors exhibit little promise as
highly accurate monitors of suspensions where the parameters mentioned
earlier are apt to change significantly.

Background--
     The initial  feasibility study of the technique of using the depolari-
zation of backscattered polarized light to determine suspended solids levels
was undertaken by  American Standard's Research and Development Center.  The
empirical method was verified by Liskowitz (103) using standard laboratory
equipment such as beakers, cuvettes, etc.  Encouraged by the initial
results, and under a separate contract (104) a more elaborate test device
was fabricated.  The sensor element alone measured roughly 12 in. on each side.
This device included a quartz window, tungsten lamp with focusing apparatus
and an optical beam-splitter receiver.  The device received only limited
testing during the initial program and was not suited to long term unattended
service in the sewer environment.  The results of these early tests were
reported by Liskowitz and Franey (105).

     When American Standard discontinued its R&D program on the solids
monitor, Badger Meter secured all patents, design, construction and test
data from American Standard and obtained the feasibility model from the EPA.
Subsequently, Badger has redesigned the instrument to simplify its optics
and electronics, and to suit it to unattended sewer installations.

Application--
     As part of this demonstration study, Badger meter has constructed a
revised prototype model incorporating the simplified design which can be
mounted in either open channels or in pressure lines as small as

                                     190

-------
12 in. diameter.  Badger obtained laboratory verification of the prototype
operation, provided the prototype, calibrated the unit in kaolin-equivalent
mg/1 units and provided the prototype to O'Brien & Gere for field data
collection.

Theoretical Basis--
     Each particle in solution is a scattering site, slightly depolarizing
the incident  polarized light and acting as a new source of partially
polarized light.  The existence of many such particles results in multiple
scattering which more completely depolarizes the emitted light.

     The emitted light is assumed to be 100 percent plane polarized and is
referred to as the polarized (p) component.

     At any point in the medium the light exhibits two components:  That
which is still polarized - the remaining 'p1 component (with a reference
of 0 degrees) and that which is depolarized - the 'd1 component (uniformly
distributed from 0-360 degrees).

Clearly, by using the backscatter radiation, not only is the density of
particles a factor, but so is the mean effective backscatter path length.
As  the density increases, the effective path length decreases (except
for a beam of 0 degree solid angle and receptor with a 0 degree solid
angle of acceptance).

     The backscattered radiation perceived by the receptor has the two
components previously described - the 'p1 and 'd1 components.  If the
emitter could be co-located with the receptor window so as to have 0°
included angle between the emitted beam axis and the receptor's acceptance
axis, then most backscattered radiation would be singly scattered at
higher densities resulting in little depolarization.  If there is a small
included angle with axes intercepting at some distance into the medium,
then reception of multiple backscatter is enhanced at the higher densities.

     The radiation perceived by the receptor system is split into two beams
separated by 90° (Refer to Figure 84).  It is assumed that the split is
50-50 in intensity and chromatically balanced.  One beam passes through
a polarizer whose polarization axis is parallel to that of the emitter,
and the other beam passes through a polarizer whose polarization axis is
perpendicular to that of the emitter.  It is assumed that these polarizers
are capable of complete polarization.

     Light polarized at an angle other than that passed by the polarizer
can be viewed as comprised of two components:  One parallel  (0°) the other
perpendicular (90°) using the trigonometric relationships for a right
triangle.  As such, uniformly depolarized radiation would be passed at half
its intensity by any polarizer plate (assuming ideal transmission character-
istics).  Thus if the received light is uniformly depolarized, the parallel
polarizer and perpendicular polarizer will each pass equal intensities,
since the polarized intensity Ip would be zero.  As such, the equations
are:
                                      191

-------
             WINDOW
                                           WASTEWATER
                                             STREAM
             LENS AND
             POLARIZER
             CHOPPER

                           \  /             V^
                            \  /              \^>    E, DETECTOR
             LIGHT SOURCE
             FIGURE 84.   Schematic  of Suspended Solids Monitor
                     I" =  Ip  +

          where      Ii =  perpendicular polarized intensity

                     I" =  parallel  polarized intensity

                     Ip =  polarized intensity

                     1^ =  depolarized intensity

The sensor notations generally  include the subscripts ' , '  or '„' as
appropriate, and it must  be  understood that these are not synonomous
with '   ' and ' .'.
      r       ^
     If we assume an output  voltage proportional to the received intensity
at each sensor, the following equations are appropriate.
                     Ei =  ki

                     E" =  k-  (Ip  + yd)

          where      EL =  perpendicular  component

                     En =  parallel component

                     K± and  K,,  =  proportionately constants

     In the American- Standard  unit,  k,  and kn were made equal, and the
ratio of E, to E,, obtained  as  the depolarization ratio D,.
                                     192:-.-.

-------
                    D, = 2id
Note that at low densities (I . = 0), D, = 0 whereas at high densities
(Ip = 0), D, = 1.            d

     Richardson (106) proposed using a slightly different ratio, 2E,
to (E^ + E,,).  Note that if kj^ = k,,, then this ratio, call it D2 is!


               D  =  Td
     The ratio D2  also ranges from 0 to 1 but its denominator always
relates to the total received light.

     A third ratio, obtained by forming the ratio of E± to (E,,-E,)
assuming k, = k,,  yields R as

               R-'d
                   'p

It is felt that R is the true depolarization ratio.  Furthermore, it
was found that the American Standard data, presented as D^ versus P
(concentration), when converted to R versus P was linear over most of its
range.

Ambient Light Effects—
     The previous discussion assumed that the only light present was
initially emitted by the polarized source.   Should there be any ambient
light its effect must be considered.  Furthermore, it is not possible to
assume that the ambient light will be uniformly depolarized, since light
incident to the air-water interface may be slightly polarized to favor
the horizontal component under water.   With low densities, this effect
may not be completely erased by multiple scatter before being sensed.
Consequently the intensity equations become:


                    "i" = «da     'la+s = «da + «ds


                    '"a ' 'pa =«da  S* ' 'pa * !ps + «ds + «ds
          where     a = ambient light

                    s = polarized light source

Similarly, the equations for E^ and E,, must be modified.

Analysis by Poisson Statistics —
     Assuming that the depolarization of polarized light due to multiple
scattering can be approximately described by Poisson statistics, the

                                     193

-------
probability that an incident beam of polarized light would be received
fully polarized might be expressed as:

                    pn = exp(-aPx),

where P is the concentration of scattering material, x is the distance
and a is a constant.  Similarly, the probability of receiving  the light
depolarized would be,

                    PL = I - exp(-aPx).

     The ratios involving E,, and ET which have been proposed were examined,
keeping in mind that E, = hi . and E,, = I  + hi H and that  the total received
light is given by It ="I  + Td.         P     a
     This is the ratio used by American Standard which, in terms of the
received intensities, can be expressed as
and in terms of  the Poisson functions as

                    RX = {l-exp(-aPx)}/{l+exp(-aPx)>

          or        R.^ = {exp(aPx)-l}/{exp(aPx)+l>.

     This ratio varies from 0 to 1 as P varies from 0 to infinity.  Approxi-
mating exp(aPx) by the first two terms in its series expansion, i.e.,
1 + aPx (an approximation which is valid if and only if the exponential,
term (aPx) is sufficiently small that the strong inequality 1 »(aPx)ri~  ,
is satisfied for all integer (n>2) yields

                    RI = aPxX/(2 + aPx).

     In similar fashion, several additional ratios have been investigated.
The results are summarized in Table  64.  Note that the ratio R^ is the
most linear and has uniform sensitivity to the particle concentration.
This ratio has a singularity (E,, - EL) and thus for high concentrations  its
accuracy depends on an accurate balance of the parallel and perpendicular
detectors and their associated electronics.   Should the approximations
of the exponential require additional terms, they can be included more
easily in the form R4 than in any of the others.  Tests of the existing
unit would indicate that the assumption's validity is supported.

Prototype Development--

     Emitter and Sensor-- The large size of the American Standard feasibility
transducer (1.0 ft-3) (0.028 cu m) made it quite difficult to use.  Since a
large portion of the inside area was void, considerable weight was required
to submerge it.   The size and submergence problem dictated a complete

                                    194

-------
                        TABLE fff.  RATIOS INVOLVING THE PERPENDICULAR AND PARALLEL COMPONENTS OF DEPOLARIZED LIGHT

                                                          Expressed In Terms of
                    Ratio
                                      E|1 S
                                                           I  S I
                                                            P
                       Poissoh Statistics
                                                 Linearized Poisson
                                                                              {exp(aPx)-l}/{exp(aPx)+U       aPx/(2+aPx)

                                                                              {exp(oPx)-l}/exp(aPx)           aPx/U+aPx)


                                                                                  Vcxp(aPx)                   1/d+aPx)
                                                                                  exp(oPx)-!
                                                                                                                 aPx
vo
en
                    Ratio
                                      Range
                                   P=0
                                                                  FEATURES
Linearity
Sensitivity S=3R/3P
P=0         P="
                                                                                                             Remarks
                                                        linear only when         2ax/(2+oPx)2
                                                        aPx«2               ox/2            0
                                                 Poor linearity, not very
                                                 sensitive at high P.
                                   0*1
                                                        linear only when          ax/(l+aPx)
                                                        «Px«l               ox
                                                 More linear & sensitive
                                                 than Rj  the true de-
                                                 polarization ratio.
                                                       nonlinear
                                                       linear
                                                                                -ox/(l+oPx)
                                                                           —ax              o
                                                 Really a 'polarization'
                                                 ratio not useful here

                                                 Linear,  uniform sensitivity
                                                 over  the entire range.
                                                 Requires careful balance
                                                 of E,, and Ej_  to achieve.

-------
redesign of the entire transducer portion of the monitor.  The design goal
was for a transducer element which would be approximately 1 in. in diameter
and 6 in. in length.  The completed transducer unit contained both the
emitter and sensor elements as well as electronics required immediately
adjacent to the emitter-sensor elements.  Such an arrangement allowed
remote location of the major electronics package.

     The feasibility unit emitter used a single tungsten bulb which emitted
a broad spectrum white light.   This light was columnated and polarized.
Extensive tests with relatively inexpensive polarizing materials indicated
that most polarizing plates do not uniformly polarize broad spectrum light.
Such a non-uniform polarization would cause the unit's sensitivity to be
non-uniform for different particle sizes.  Ideally the unit should respond
uniformly regardless of particle size.  For this reason the spectral
content of the transducer emitter was restricted.

     The search for an emitter source of sufficiently narrow spectral content
indicated that those which had adequate intensity generally were in  the
infra-red or near infra-red spectrum.  This presented an additional problem
in that not all polarizing materials worked well in the near infra-red or
infra-red region.  However a new material being investigated by the Polaroid
Corporation proved adequate for near infra-red emitters.  The1 emitteg +     0
finally selected developed twelve milliwatts radiated power at 9400 A - 300 A
with only a 0.33 watt drive.   The polarizers used were highly effective at
this wave length and had the added benefit that they were nearly opaque to
visible light, reducing ambient light effects significantly..  Since it has
been shown that depolarization in backscatter radiation tends to fall off
dramatically when the particle size is less than 1/2 wave length in diameter,
the use of 9400 A light would dictate that the unit would be sensitive to
all particle sizes greater than 0.5y.

     The development of the new sensor paralleled that of the new emitter
system.  Initial efforts tried to duplicate in miniature the beam splitter
arrangement used in  the American Standard feasibility unit.  However, it
was found that the beam splitters which  were available at an acceptable
price did not perform satisfactorily in  the infra-red range, and added a
great deal in  the terms of complexity and loss of intensity.  Additionally,
it had to physically be displaced from the window, permitting light reflection
internally which added to depolarization, causing false indications.  For
this reason two separate sensors were used, each mounted directly behind its
polarizing plate as closely together as possible, and each located the same
distance from the emitter by placing the emitter and two sensors at the points
of an equilateral triangle.

     The objective of small size for the transducer housing required that the
emitter be placed in close proximity to the sensors.  Due to the high
intensity of  the emitter it was found that ordinary plastics were not
sufficiently opaque to totally prevent transmission from the emitter region to
the sensor.  The typical mechanism for such transmission was that as the
emitter light would pass through its polarizing plate, internal reflections
would cause sideways transmission through the plate into the receptor or
                                    196

-------
sensor polarizers where the very sensitive sensors would see the light as
returned energy.  To avoid this problem the sensor emitter housing was
fabricated from stainless steel with inset polarizing plates.

     Electronics-- Due to the extreme sensitivity of the sensor units, it
was necessary to mount the operational amplifiers and buffer amplifiers in
a sensor emitter package as close as possible to the sensors.  In the final
prototype this entire electronics package was mounted on small circuit board
approximately 0.75 in. by 3 in. and imbedded in the same plotting material
used to surround the sensors and emitter.

     In order to minimize or eliminate the ambient light effects, the emitter
was alternately turned on and off at a rate of several hundred times per
second.  The received information from the sensors was then coherently
detected and filtered, eliminating all effects of constant ambient light as
well as pulsating effects due to fluorescent lighting fixtures.   The final
electronics system was found to be totally immune to sunlight as well as
fluorescent and incandescent light.  Figure 85 presents a photograph of the
final electronic package and probe of the SS Meter.
            FIGURE 85.   Electronics Package for Suspended Solids Meter
Prototype Testing--
     The prototype testing program was accomplished in a series of
extensive tests done in the laboratory followed by some limited testing in
the field.  The field testing consisted of two phases; tests performed at
the Tulsa, Oklahoma, Mohawk Park Sewage Treatment Plant and tests performed
at selected locations at. the Syracuse Demonstration facility at Maltbie
Street.

     Laboratory T'  ts_ — For the most part laboratory tests were performed
using ungraded wlr. oe kaolin clay.  This clay was used in all color tests
using both ordinary food coloring dyes as well  as nigrosine black dye-.
                                    197

-------
Some limited tests were performed using graded white kaolin clay in size
ranges 5,10,15 and 30y.  The ungraded kaolin clay was generally sized
greater than or equal to one micron.  Additionally some limited tests
were run on pumice, rottonstone, fine sand and biological material which
was grown in the laboratory to assure purely biological nature.

     The test apparatus consisted of a 5.3 gal (20 1) cylindrical container
which was kept in constant agitation by means of an axially driven bottom
disturbing vane.  The solids meter probe was suspended in the circulating
fluid approximately 1.0 in. (2.5 cm) from the wall at a depth of approximately
4.0 in. (10.2 cm) so that it was aimed directly across the cylindrical
container.

     The  results of a typical test on ungraded kaolin clay are shown on
Figure 86.  Note how the depolarization ratio, R, of the low range of
clay concentration from 10 to 1000 mg/1 tends to increase at an increasing
rate (roll up) while it increases at a decreasing rate (roll off) in the
range from 10,000 to 100,000 mg/1.  The latter occurrence is a result of
a slight mismatch in the gains of the two channels (P for polarized
component and D for the depolarized component).  Figure 87 shows the
same data after a correction has been made to the raw data to represent
balanced gains.  Notice how the gain adjustment has straightened the
high end of the data but has little effect on the results at the low end.
The low end effect is believed to be the result of a depolarization
offset.  Fortunately, in all the tests performed, this low end phenomenon
appeared to be predictable enough so that it could  be compensated for
electronically.  In the existing prototype unit compensation was not
provided, however.

     The high end roll off problem stemming from the imbalanced gains on
the depolarized and polarized channels is a difficult one to resolve.
To date the only method which has been successful in adjusting that gain
has been to immerse the probe in a media which is at least ten times
more dense than the densest media to be observed and adjusting the gains
for as close to proper output as can be achieved.

     In Figures 86 through 88 the solid straight line represents the
linear response.   As already indicated, Figure 87 shows the corrected
response when unbalanced depolarization-polarization gains are accounted
for.  The equation

                    Rm = (1 + e)  R0 + e

          where     Rm is the measured ratio and
                    e is an empirical correction factor for probe rescatter
                    R0 is the ideal linear relationship,

represents the effect of reradiation from the probe face.  Empirically
choosing a value for e and correcting Figure 87, Figure 88 is obtained.
The low end roll up has now been straightened with little if any effect
on high end data, resulting in a nearly linear response for the unit
from 10 to 10,000 mg/1 of ungraded kaolin clay solution.

                                    198

-------
           UNGRADED   KAOLIN   CLAY ,  mg/l  (Upper  Curve)
o>
i_

o
0)
Q.
Q.
£E

 •S


O
O
N

o:
o
Q.
UJ
o
                                                                 10.0  g
                                                                       0>
                                                                       o
o:

 *fc

O



Q:

                                                                      N
                                                                      E


                                                                      3
                                                                      Q.
            UNGRADED   KAOLIN   CLAY ,  mg/l  (Lower  Curve)


 FIGURE 86.  Depolarization Ratio vs Ungraded Kaolin Clay Concentration

            No Depolarization-Polarization Gain Adjustment
                                   199

-------
           UNGRADED   KAOLIN   CLAY , mg /1 ,  (Upper Curve)
      10'
           UNGRADED     KAOLIN   CLAY, mg/1, (Lower Curve)

FIGURE 87.  Depolarization Ratio vs Ungraded Kaolin Clay Concentration
           Unbalanced  Depolarization-Polarization Gain Adjusted
                                 200

-------
            UNGRADED   KAOLIN    CLAY,  mg/l , (Upper Curve)
Q>
o



i
QL
fE


cT
O
N

cr
<


2
Ul
o
                                                                100
   0.01
                                                                      O
                                                                      £C

                                                                       *»

                                                                      O


                                                                      5
                                                                      DC
                                                                      Z

                                                                      O
                                                                      M
                                                                      2
                                                                      Ul
                                                                      o
             UNGRADED   KAOLIN    CLAY ,  mg/l , (Lower Curve)





 FIGURE 88.   Depolarization Ratio vs Ungraded Kaolin Clay Concentration -

            Adjusted for Reradiation Effects
                                  201

-------
     Limited tests using 5, 10, 15 and 30u graded kaolin samples yielded
results which were indiscernibly different from those of random kaolin
samples whose sizes are greater than or equal to one micron (see Figure 89).
This is in keeping with Liskowitz's observation that the backscatter method
is relatively insensitive to particle size provided that particle size
is greater than 1/2 wavelength.

     In limited tests, considerable difference could be discerned between
the kaolin, diatomaceous earth, pumice and rottonstone, indicating the
sensitivity of the procedure to particles of different refractive indices.
Refer to Figure 90.

     Biological solids tests show that such particles had a slightly
different response, most liekly due to their different refractive index as
well as some variation in the amount of reflected illumination from the
probe itself.  The variations show slightly different behavior at the low end,
becoming linear at the upper end but with an offset calibration variation.
Additionally the tests were performed in a beaker which further influenced
the lower concentration reading due to reflections from the beaker sides.
See Figure 91.

     Ordinary food coloring in varying concentrations and colors from dark
blue to dark red had no noticeable effect on the output of the unit.  The
nigrisine black did have an effect when the concentration of the nigrisine
black in mg/1 became equal to approximately five times the concentration
of the particulate .matter.  See Figures 92 and 93.   It is
felt that the concentration level  of this intensely black dye would be
unrealistic in most applications.   At the levels which caused change in
the readings the absorption is so high that return energy from multiple
scatters is significantly reduced making the concentration appear
artificially lower than it is.

     Field Tests-- The procedure used in the field tests consisted of
immersing the suspended solids monitor probe at each of several selected
sites and obtaining grab samples concurrently with recording  the
monitor output.  The grab samples were then returned to the lab where
gravimetric tests were performed to obtain total SS.   In
the tests performed at the local sewage treatment plant these grab
samples were analyzed .in a laboratory within several  hours, whereas the
samples obtained at the remote sites in Syracuse were obtained with
automatic sampling apparatus and these samples were not analyzed in the
laboratory for up to 48 hr after they were obtained.

     On two separate occasions the SS monitor was taken to Tulsa's Mohawk
Park Sewage Treatment Plant where several samples were obtained at various
places throughout the treatment process as indicated on Figure 94.  Samples
were obtained in the clarifier effluent, raw influent, and aeration basins.
The data shown in Figure 94, for the ratio, R' as displayed by the SS monitor,
is the raw data and has not been corrected.  The spread in the clarifier
effluent and aeration basin data is probably caused by instability in
the SS monitor.  In  general there appears to be good qualitative and
                                    202

-------
                  DEPOLARIZATION   RATIO , R
o
m
o
o
3

-------
o:

 •*

o

h-
<
cc
o
N
ce
o
Q.
UJ
O
                 10
          CONCENTRATION  of   INDICATED   MATERIAL ,  mg/l


       FIGURE 90.  Depolarization Ratio vs Kaolin, Diatoniaceous  Earth,
                  Rottenstone, and Pumice Concentrations
                                    204

-------
          BIOLOGICAL   SOLIDS,  mg/l , (Upper  Curves)
O

a>
a.
a.
DC

 »v
O
F
<
DC
Z
O

S
N
CC
O
a.
UJ
a
                                                                       0)
                                                                       O
g

P

££



Z
O

H

-------
       DEPOLARIZATION    RATIO ,  R



                            o
FIGURE  92.  Depolarization Ratio vs  Nigrisine Black Concentration
                          206

-------
   10.0
g
h-
<
N
Ct
<


2
UJ
Q
                                  10'
10'
                  NIGRISINE    BLACK ,  mg/l

       FIGURE 93.  Depolarization Ratio vs Nigrisine Black and  Kaolin
                  Concentrations
                                207

-------
   10.0
DC
N
CC
UJ
Q
          io'                         lo-

               ss  CONCENTRATIONS,  mg/l


FIGURE 94.  Depolarization Ratio vs Treatment Process  SS Concentrations
                                208

-------
quantitative correspondence in these tests with the earlier laboratory
biological tests and the tests performed using kaolin.

     In the Syracuse test program the probe was installed at several sites
where CSO was treated.  Consequently data obtained at these sites was
limited to those times during which there was a storm event.  Results of
this testing were limited for a number of reasons.  Initially the recorder
used at the site was not sufficiently fast enough in the-time scale to
allow accurate interpolation of the transient events.   Additionally,
since the SS monitor did not have an auto-arranging capability the unit had
to be left in a specified SS measuring range over which it was hoped
that the majority of the SS values measured  during the storm event
would occur.  As such the field calibration was not precisely known.  An
additional factor which limits direct comparison of SS monitor values
with the gravimetric SS determinations is that the automatic samplers
tend to attentuate peak SS values since samples are pumped into the
containers over some specified time frame, however short it may be.
Thus instantaneous SS values are not directly measurable for those
samples collected automatically.  However, there were several events for
which adequate data was obtained.  One such event is shown as Figure 95
in which the measured total SS for the SS monitor are shown versus the
time of the event.  Note the good correlation between the SS monitor
output and the total SS over the period of the event.   Unfortunately the
precise calibration of the SS monitor is not known.   The output plot
was adjusted so that the peak events of the two overlie one another.  It
would be expected that the actual peak obtained on the SS monitor would
be somewhat higher than that obtained by the automatic samplers; however,
the significant factor in this figure is the good qualitative correlation
between the two curves.

Summary of SS Monitor Results--
     The objective of this project was to fabricate a prototype SS monitor
based on the principle of depolarization of backscattered polarized light
as investigated by Liskowitz and to test this prototype both in the
laboratory and in the field.  To a large extent this was accomplished with
only the field testing in the storm overflow environment providing less
data than expected due to the field sampling difficulties discussed earlier.
The salient observations are itemized in the following.

     1.   Color Sensitivity

          The prototype was shown to be insensitive to the color of
          dissolved solids for practical concentrations.  Nigrisine black
          dye concentration of less than 10 percent of the SS concentration
          had no effect.  Similar results were obtained for red, green,
          and other color dyes.

     2.   Particle Shape and Size

          For particle sizes exceeding the radiation half wavelength, no
          size effects were noted.  While direct shape testing was not
                                    209

-------
                 Storm date:  29 August 1974

                 Sample  location: Zurn (7In)effluent at
                                 Maltbie Street  facility
      V.
      V
         UCT
               i :

¥
                          XI-
                              XJ
                                    ^
                                       efi
                                           jmg£
                                          it-
                       -K
                                                41
                                                 31
                                                retlv
                         MOM
                             s±
                                                    S3
                                                       or
                              mewi€}
W!-
                                  10
                                       Jr-
                                                            ±
                                                                    _g»
                                                                    o
                                                                    u
                                                                    0>
                                                                    (T
                                                                    O)
                              hr  of day
FIGURE 95.  SS Monitor - Percent of Scale Readings  and  SS  Concentrations

            in Syracuse Field Tests vs Time
                                   210

-------
     accomplished, no effort was employed to control shape.   There was
     no evidence of particle shape interference.

3.   Ambient Light

     Operation in ambient light conditions is limited only by sensor
     saturation.  At low concentrations, where light shielding is
     negligible, operation in direct sunlight produced no interference.
     This characteristic is vital to portable survey usage.

4.   Particle Concentration Range

     Most tests were confined to measurements in the 10 mg/1 to
     100,000 mg/1 SS range.  The unit demonstrated measurement
     capability from less than 1 mg/1 to over 1,000,000 mg/1.

5.   Probe Physical Characteristic

     The transducer probe size of 1 in. (2.5 cm) diameter and 6 in.
     (15.2 cm) length allows operation in confined spaces and
     permits portable operation.

6.   Refractive Index

     As with any optical device, sensitivity to particle refractive
     index is present.  Calibration shifts on widely divergent
     refractive indices were noted.  For refractively homogenous
     slurries, calibration stability was demonstrated.   For instance,
     in sewage sludge the composition generally has less than 20 percent
     grit.  The remainders are generally suspended and dissolved solids
     of a biological nature.  Close examination of the monitor's
     response to biological solids as opposed to sand, pumice or stone
     dust shows that the overall interference by the suspended grit
     and the determination of suspended biological solids would not
     exceed one or two percent of error.  Similarly in dredging opera-
     tions where most of the suspended matter is largely grit, proper
     calibration would permit accurate display of suspended solids.

7.   Probe Configuration

     In low concentration measurements, nonlinearity of measurement
     was traced to two sources.  First the shape of the probe face
     enhanced reradiation (by reflection) of transmitted energy which
     increased the depolarization ratio significantly.   Secondly, the
     backscatter of light, from the container sidewalls (such as a
     laboratory beaker) also enhanced the depolarization ratio.  When
     measurements of low concentration slurries is attempted, the
     container volume should be increased over that used in  the tests.
     Similarly the flat face probe design should be modified to reduce
     secondary reflections.
                               211

-------
8.   Comparative Resu1ts

     While no direct comparative tests were carried out, the depolariza-
     tion method has been shown to be insensitive to most of the
     parameters known to affect transmissive, nephelometric, and forward
     scatter measurement means.
                              21Z

-------
                                  REFERENCES

1.   Lager, John A., and William G. Smith.  Urban Stormwater Management and
     Technology:  An Assessment.  USEPA Report No. EPA-670/2-74-040.
     NTIS No. PB 240 687.  September 1974. p.23.

2.   Camp, Dresser and McKee Consulting Engineers.  Comprehensive Sewerage
     Study, Onondaga County, New York.  Project No. WPC-CS-169, Department
     of Public Works, Division of Drainage and Sanitation, County of Onondaga,
     New York, 1968.  62 pp.

3.   O'Brien and Gere Engineers, Inc. Industrial Waste Monitoring Program.
     Project No. 115.246, Department of Public Works, Division of Drainage
     and Sanitation, County of Onondaga, New York, 1975.

4.   O'Brien and Gere Engineers, Inc.  Onondaga Lake Study.  USEPA Project
     No. 11060 FAE, County of Onondaga, New York, April, 1971.

5.   O'Brien and Gere Engineers, Inc. Onondaga Lake Monitoring Program,
     Jan-Dec 1973.  Project No. 115.238, Department of Public Works, Division
     of Drainage and Sanitation, County of Onondaga, New York, 1975.

6.   O'Brien and Gere Engineers, Inc. Unpublished Data for Onondaga County
     Department of Public Works, Division of Drainage and Sanitation, 1959-1969.

7.   Moffa, P.E., et^aJL  Bench-Scale High Rate Disinfection of Combined Sewer
     Overflows with Chlorine and Chlorine Dioxide.  USEPA Report No. EPA-670/
     2-75-021.  NTIS No. PB 242 296.  April 1975.

8.   Murphy, C.B., and Orest Hrycyk. High Rate Nutrient Removal for Combined
     Sewer Overflows.  USEPA Report No. EPA-600/2-78-056.  NTIS No. PB 285 473.
     June 1978.

9.   Glover, George E., "and George R. Herbert. Microstraining and Disinfection
     of Combined Sewer Overflows - Phase II.  USEPA Report No. EPA-R2-73-124.
     NTIS No. PB 219 879.  January 1973.

10.  Maher, M.B.  Microstraining and Disinfection of Combined Sewer Overflows-
     Phase III.  USEPA Report No. EPA-670/2-74-049.  NTIS No.  PB 235 771.
     August 1974.

11.  Keilbaugh, W.A., et^ a]_.  Microstraining - with Ozonation and Chlorination-
     of Combined Sewer Overflows.  Combined Sewer Overflow Seminar Papers,
     November 1969.  USEPA Report No. 11020—03/70.  NTIS No. PB 199 361. 1970.
                                      213

-------
12.  American Public Works Association.  Combined Sewer Overflow Regulation
     and Management-A Manual of Practice.  USEPA Report No. 11022 DMU 08/70.
     NTIS No. PB 195 676. 1970.

13.  Sullivan, R.H.,et^al_.  The Swirl Concentrator as a Combined Sewer
     Overflow Regulator Facility.  USEPA Report No. EPA-R2-72-008.  NTIS No.
     PB 214 687.  September 1972.

14.  Sullivan, R.H., eit al_.  Relationship Between Diameter and Height for the
     Design of a Swirl Concentrator as a Combined Sewer Overflow Regulator.
     USEPA Report No. EPA-670/2-74-039.  NTIS No. PB 234 646. July, 1974.

15.  Field, Richard.  The Dual Functioning Swirl Combined Sewer Overflow
     Regulator/Concentrator.  USEPA Report No. EPA-670/2-73-059.  NTIS No.
     PB 227 182/3.  September, 1973.

16.  Hsiung, 6.D., and J.R. Henderson.  Diagostic Virology.  Yale University
     Press, New Haven, Connecticut, 1964.

17.  Kalter, S.S. Procedures for Routine Laboratory Diagnosis of Virus and
     Rickettsial Diseases.  Burgess Publishing Company, Minneapolis,
     Minnestoa, 1963.

18.  Meynell, 6.6., and E. Maynell.  Theory and Practice in Experimental
     Bacteriology.  Cambridge University Press, Cambridge, Massachusetts,
     1965.

19.  Rosenbaum, J.J., E.J. Sullivan, and E.A. Edwards.  Techniques for Cell
     Cultivation in Microtitration Plates and Their Application in Biological
     Assays.  In:  Animal Tissue Culture, 6.D. Wasley, ed. Williams and
     Wilkins Co., Baltimore, 1973.

20.  Leibowitz, A.  The 6rowth  and Maintenance of Tissue-Cell Cultures in
     Free 6as Exchange with  the Atmosphere.  American Journal of Hygiene,
     78:173-178, 1963.

21.  Sobsey, M.D., C. Wallis, M. Henderson, and J.L. Melnick.  Concentration
     of Enteroviruses from Large Volumes of Water.  Applied Microbiology,
     26: 529-534, 1973.

22.  Homma, A., M.D. Sobsey, C. Wall is, and J.L. Melnick.  Virus Concentration
     from Sewage.  Water Research, 7:945-950, 1973.

23.  Berg, 6. Transmission of Viruses by the Water Route.  John Wiley and Sons,
     New York, New York, 1967.

24.  Hill, W.F., Jr., E.W. Akin, and W.H. Benton.  Detection of Viruses
     in Water:  A  Review  of Methods and Applications.  Water Research,
     5:967-995, 1971.
                                       214

-------
25.  Wall is, C., A. Homma, and J.L. Melnick.  Apparatus for Concentrating
     Viruses from Large Volumes of Water.  Journal of the American Water*
     Works Association, 64:189-196, 1972.

26.  Wallis, C., A. Homma, and J.L. Melnick.  A Portable Virus Concentrator
     for Water Testing in  the Field.  Water Research, 6:1249-1256, 1972.

27.  Kruse, C.W., V. Olivieri, and K. Katvata.  The Enhancement of Viral
     Inactivation by Halogens.  Water and Sewage Works, 118:187, 1971.
28.  Wallis, C., M. Henderson, and J.L. Melnick.  Enterovirus Concentration
     on Cellulose Membranes.  Applied Microbiology, 23: 476-480, 1972.

29.  Wallis, C., and J. L. Melnick.  A Portable Virus Concentrator for
     Use in the Field.  In:  Advances in Water Pollution Research, Jerusalem,
     1972, S.H. Jenkins, ed. Pergamon, Oxford, England, 1973. pp 119-126.

30.  Metcalf, T.6., C. Wallis, and J. L. Melnick.  Concentration of Viruses
     from Seawater.  In:  Advances in Water Pollution Research, Jerusalem,
     1972.  S.H. Jenkins, ed. Pergamon, Oxford, England, 1973.  pp. 109-115.

31.  Metcalf, T.G., C. Wallis, and J.L. Melnick.  Environmental Factors
     Influencing Isolation of Enteroviruses from Polluted Waters.  Applied
     Microbiology, 27: 920-926, 1974.

32.  Dahling, D.R., G. Berg, and D. Berman.  BGM, A Continuous Cell Line
     More Sensitive Than Primary Rhesus and African Green Kidney Cells
     for the Recovery of Viruses from Water.  Health and Laboratory Sciences,
     11:275-282, 1974.

33.  American Public Works Association.  Combined Sewer Regulator Overflow
     Facilities.  USEPA Report No. 11022 DMU 07/70, July 1970. p. 139.

34.  Smisson, B. Design, Construction and Performances of Vortex Overflows.
     In:  Proceedings, Symposium on Storm Sewage Overflows, Institute of
     Civil Engineers, Great Britain, May 4, 1967. p. 99.

35.  Ackers, P., A.J..M. Harrison, and A.J. Brewer.  Laboratory Studies of
     Storm Overflows with Unsteady Flow. In: Proceedings, Symposium on Storm
     Sewage Overflows, Institute of Civil Engineers, London,  Great Britain,
     May 4, 1967. p. 37.

36.  Prus-Chacinski, T.M., and J.W. Wielgorski.  Secondary Motions Applied to
     Storm Sewage Overflows.  In: Proceedings, Symposium on Storm Sewage
     Overflows, Institute of Civil Engineers, London, Great Britain,
     May 4, 1967.

37.  Kromis, H.  Full Scale Evaluation of High Rate Screening Devices for
     Treatment of Sanitary Sewage Bypass Flow, Interim Report.  Ontario
     Ministry of the Environment, Ontario, Canada, Letter correspondence,
     June 20, 1975.
                                    215

-------
38.  Krus, C.W.  Improvement in Terminal Disinfection of  Sewage  Effluents.
     Water and Sewage Works, 120:  57-64, 1973.

39.  Glover, G.E. High-Rate Disinfection of Combined Sewer.Overflow.   In:
     Combined Sewer Overflow Seminar Papers, USEPA  Report No.  670/2-73-077.
     NTIS No.  PB 231 836.  November 1973.

40.  National Oceanic and Atmospheric Administration, Environmental  Data
     Service.  Magnetic Tape Decks 345 and 488,  1948 to 1973.  U.S.  Depart-
     ment of Commerce, Ashville, North Carolina.

41.  Rainfall Frequency Atlas of the United States, Technical  Paper  No. 40.
     U.S. Department of Commerce, Washington,  D.C.  1961.

42.  Lager, John A., Theodor Didriksson, and George B. Otte.   Development  and
     Application of a Simplified Stormwater Management Model.  USEPA Report
     No. EPA-600/2-76-218.  NTIS No. PB 258 074.  August,  1976.

43.  Fram Corporation.  Strainer/Filter Treatment of Combined  Sewer  Overflows.
     USEPA Report No. 11010 EXV07/69.  NTIS No.  PB  185 949.  July 1969.

44.  Gupta, Mahendra K., ejt al_.  Screening/Flotation Treatment of Combined
     Sewer Overflows, Volume I - Bench Scale and Pilot Plant Investigations.
     USEPA Report No. EPA-600/2-77-069a.  NTIS No.  PB 272 834.   1977.

45.  Rex Chainbelt, Inc.  Screening/Flotation  Treatment at Combined  Sewer
     Overflows.  USEPA Report No. 11020 FDC 01/72,  1972.

46.  Nebolsine, Ross, et^ a]_.  High Rate Filtration  of Combined Sewer Overflows.
     USEPA Report No. 11023 EYI 04/72.  NTIS No. PB 211 144.   April  1972.

47.  Neketin, T.H., and H.K. Dennis.  Demonstration of Rotary  Screening for
     Combined Sewer Overflows.  USEPA Report No. 11023 FDD 07/71. NTIS No.
     PB 206 814.  July 1971.

48.  Lager, John A., ejt al_.  Urban Stormwater  Management  and Technology:
     Update and User's Guide.  USEPA Report No.  EPA-600/8-77-014. NTIS No.
     PB 275 654.  September 1977.

49.  Colston, N.V., Jr.  Characterization and  Treatment of Urban Land Runoff.
     USEPA Report No. EPA-670/2-74-096.  NTIS  No. PB240 978.   December 1974.

50.  Mytelka, Alan I., et a_l_.  Combined Sewer  Overflow Study for the Hudson
     River Conference.  USEPA Report No. EPA-R2-73-152.   NTIS  No. PB 227341.
     January 1973.

51.  Dalrymple, Robert J., et aK  Physical and  Settling  Characteristics  of
     Particulates in Storm and Sanitary Wastewaters.  USEPA Report No.
     EPA-670/2-75-011.  NTIS No.  PB 242 001.  April 1975.

52.  Gupta, M..K., et^ al_.  Handling and Disposal  of  Sludges Arising From
     Combined Sewer Overflow Treatment:  Phase I Characterization.   USEPA
     Report No. EPA-600/2-77-053a.  May 1977.

                                    216

-------
53.  Rich, L.6. Unit Processes in Sanitary Engineering.  John Wiley  &  Sons,
     Inc., New York, New York, ,1963.  pp 147-164.

54;  Drehwing, Frank J., et_ al_.  Combined Sewer Overflow Abatement Program,
     Rochester, New York - Pilot Plant Evaluations.  USEPA  Project No.
     Y005141,, Final Report.  1978.

55.  Fair, Gordon M., and John C. Geyer.  Elements of Water Supply and
     Waste-Water Disposal.  John Wiley and Sons,  Inc.,  New  York,  New York,
     1965.  pp. 480-486.

56.  Benjes, Henry H. Jr.  Cost Estimating Manual - Combined Sewer Overflow
     Storage and Treatment.  USEPA Report No. EPA-600/2-76-286. NTIS No.
     PB 266 359.  December 1976.

57.  1975 Dodge Guide for Estimating Public Works Construction  Costs.   Dodge
     Building Costs Services, McGraw-Hill Information SystemsJ  bmpany,
     New York, New York, 1975.  201 pp.

58.  Rao, V.C., V. Chandorkar, N.V. Rao, P. Kumeran, and S.B. Lahke.   Unpub-
     lished Data In:  Second International Congress for Virology, Budapest,
     Hungary, 1971.

59.  Metcalf, T.G.  Personal Communication. 1974.

60.  Field, Richard, et_ al_.  Proceedings of Workshop on Microorganisms in
     Urban Stormwater.  USEPA Report No. EPA-600/2-76-244.  NTIS  No.   PB 263
     030.  November 1976.

61.,  Long, W.N., and F.A. Bell, Jr. Health Factors and  Reused Water.   Journal
     of the American Water Works Association, 64:  220, 1072.

62.  Taylor, F.B. Viruses - What Is Their Significance  in Water Supplies.
     Journal of the American Water Works Association, 66: 306-311, 1974.

63.  Potable Waters.  Hearings Before the Committee on  Commerce,  U.S.
     Senate, on Amendment 410 to S1478, Serial No. 92-57.   U.S. Government
     Post Office, Washington, D.C., 1972.

64.  Luria, S.E., et^ al_.  General Virology.  John Wiley & Sons, New  York,
     New York, 3rd Edition, 1978.  578 pp.

65.  Rook, J.J. Formation of Halogens During Chlorination of Natural Waters.
     Water Treatment Examiner, Part 2, 23:234-243, 1974.

66.  Harris, R.H.  The Implications of Cancer Causing Substances  in  Mississippi
     River Waters.  Environmental Defense Fund, Washington, D.C.  1974.

67.  Morris, J.C., I.W. Weil and R.H. Culver.  Kinetic  Studies  on the
     Breakpoint Reaction with Ammonia and Glycine.  Presented at:  International
     Congress of Pure and Applied Chemistry, New  York,  New  York,  1952.
                                       217

-------
68.  American Public Health Association, Inc.  Standard Methods for the
     Analysis of Water and Wastewater, 14th Edition.  New York
69.  Federal Register, 38:22298, August 17, 1978.

70.  Dowty, B., D. Carlisle, and J.L. Laseter.  New Orleans Drinking Water
     Sources Tested by Gas Chromatography-Mass Spectrometry.  Environmental
     Science & Technology, 9:762, 1975.

71.  American Public Health Association, Inc. Standard Methods  for the
     Examination of Water and Wastewater, 13th Edition.  New York, 1971.

72.  Chambers, C.W. Chlorination for Control of Bacteria and Viruses in
     Treatment Plant Effluents.  Journal Water Pollution Control Federation,
     43:228, 1971. -

73.  Robeck, G.G. Substitution of Residual Measurement for Distribution
     Bacteriological  Sampling.   In: Proceedings of the AWWA Water Quality
     Technology Conference, Dallas, Texas, December 2-3, 1974.  American
     Water Works Association, Denver, Colorado, 1975.

74.  Baker, R.J. Engineering Considerations in Disinfection.  In:  Proceedings
     of the National  Specialty Conference on Disinfection, July 8-10, 1970.
     American Society of Civil Engineers, New York, 1970.

75.  Reasoner, D.J.,  and E.E. Geldreich.  Rapid Bacteriological Methods.
     In:  Proceedings of the AWWA Water Quality Technology Conference, Dallas,
     Texas, December 2-3, 1974, American Water Works Association, Denver,
     Colorado, 1975.

76.  McElroy, W.D. Energy Source for Bioluminescence in an Isolated System.
     Proceedings National Academy of Sciences (USA), 33: 342-345, 1947.

77.  Levin, G.V., «rt al_.  Rapid Method for Detection of Microorganisms by
     ATP Assay:  It's Possible Application in Virus and Cancer Studies.
     Bioscience, 14:37-38, 1964.

78.  Holm-Hansen, 0.  Determination of Microbial Biomass in Ocean Profiles.
     Limnology & Oceangraphy, 14:740-747, 1969.

79.  Heunnekens, F.M., and H.R. Whitely.  Phosphoric Acid Anhydrides and
     Other Energy-rich Compounds.  Comparative Biochemistry, 1:107-180.

80.  Pohnamperuma, C., R. Mariner, and C. Sagan.  Formation of Adenosine by
     Ultra-violet Irradiation of a Solution of Ademine and Ribose.  Nature,
     198:1199, 1963.

81.  Ponnamperuma, C., C. Sagan, and R. Mariner.  Synthesis of Adenosine
     Triphosphate Under Possible Primitive Earth Conditions.  Nature, 199:
     222, 1963.
                                    218

-------
82.  Levin, 6.L., C. Chen, and G. Davis.  Development of the Firefly
     Bioluminescent Assay for the Rapid Quantitative Detection of Microbial
     Contamination of Water.  Aerospace Medical Research Laboratories Report
     No. AMRL-TR-67-71, Wright-Patterson Air Force Base, Ohio, 1967.

83.  McElroy, W.D., M. Deluca, and J. Travis.  Molecular Uniformity in
     Biological Catalyses.  Science, 157:150, 1967.

84.  Plant, P.J., E.H. White, and W.D. McElroy.  Decarboxylation of Luciferin
     in Firefly Bioluminescence.  Biochemical and Biophysical Research
     Communications, 31:98, 1968.

85.  McElroy, W.D., H.H. Seliger, and E.H. White.  Mechanism of Bioluminescence,
     Chemiluminescence and Enzyme Function in the Oxidation of Firefly
     Luciferin.  Photochemistry and Photobiology, 10:153,  1969.

86.  Rhodes, W.C., and W.D. McElroy.  Synthesis and Function of Luciferyl-
     adenylate and Oxyluciferyl-adenylate.  Journal of Biological Chemistry,
     233:1528, 1958.

87.  McElroy, W.D., and H.H. Seliger.  Chemistry of Light  Emission.  Advances
     in Enzymology, 25:119, 1963.

88.  Rudd, J.W.M., and R.D. Hamilton.  Measurement of Adenosine Triphosphate
     (ATP) in Two Precambrian Shield Lakes of Northwestern Ontario.  Journal
     of the Fisheries Research Board of Canada, 30:1537, 1973.

89.  Hamilton, R.D., 0. Holm-Hansen, and J.D.H. Strickland.  Notes on the
     Occurrence of Living Microscopic Organisms in Deep Water.  Deep-Sea
     Research, 15:651, 1968.

90.  Kao, I.e., et aj_.  ATP Pools in Pure and Mixed Cultures.  Journal Water
     Pollution Control Federation, 45:926, 1973.

91.  Hobson, P.N., and R. Summers.  ATP Pool and Growth Yield in Selenomonas
     rumi nan turn.  Journal of General Microbiology, 70:351, 1972.

92.  Levin, G.V., J.R. Schrot, and W.C. Hess.  Methodology for Application of
     Adenosine Triphosphate Determination in Wastewater Treatment.  Environ-
     mental Science & Technology, 9:961, 1975.

93.  Smith, J.E., and J.L. McVea.  Virus Inactivation by Chlorine Dioxide
     and Its Application to Storm Water Overflow.  In: Proceedings of the
     166th National Meeting, American Chemical Society, Chicago, Illinois,
     13:2, 177-185, August 1973.

94.  Instruction Manual Luminescence Biometer, Catalog No. 760018.  Instruments
     Products Division, E.I. Dupont et Nemours, Wilmington, Delaware, 1970.

95.  Chappelle, E.W., and G.V. Levin.  Use of the Firefly Bioluminescent
     Reaction for Rapid Detection and Counting of Bacteria.  Biochemical
     Medicine, 2:41, 1968.

                                   219

-------
96.  Dufresne, L., and H.J. Gitelman.  Semiautomated Procedure for Deter-
     mination of Adenosine Triphosphate.  Analytical Biochemistry 37:402, 1970.

97.  Proceedings of the First Microbiology Seminar on Standardization of
     Methods.  USEPA Report No. EPA-R4-73-022, Washington, D.C. 1973.

98.  Johnson, R., J.H. Gentile, and S. Cheer.  Automatic Sample Injector:
     Its Application in the Analysis of Adenosine Triphosphate.  Analytical
     Biochemistry, 60:115, 1974.

99.  VanDyke, K., R.Stitzel, T. McClellan, C. Szutkiewicz.  Automated
     Analysis of ATP:  Its Application to On-Line Continuous-Flow Incubations
     and Measurement of ATP-Coupled Enzyme Systems.  Advances in Automated
     Chemistry, 1:47, 1969.

100. Technicon Corporation.  Automating Manual Methods Using Technicon Auto
     Analyzer II System Techniques.  Publication No. TNI-0170-01, Tarrytown,
     N.Y., 1972.

101. Raytheon Company, Environmental Systems Center.  Stormwater Total Organic
     Carbon Analyzer.  Portsmouth, Rhode Island, 1975.  32 pp.

102. Tulumello, A. Automatic Organic Monitoring System for Storm and
     Combined Sewers.  USEPA Report No. EPA-670/2-75-067.  NTIS No. PB
     244 142.  1975.

103. Liskowitz, J.W.  An Empirical Method for Determining the Concentration
     of Solids in Suspension.  Environmental Science and Technology, 12:1206-
     1211, 1971.

104. Liskowitz, John W., et al_.  Suspended Solids Monitor.  USEPA Report No.
   .  EPA-670/2-75-002.  NTIS No.  PB 241 581. 1975.

105. Liskowitz, J.W., and G.J. Franey.  Measurement of Suspended Solids
     Concentration in Sewage by Use of a Depolarization Method.  Environmental
     Science and Technology, '1:43-47, 1972.

106. Richardson, Allyn C.  Special Assistant for Advanced Technology, USEPA
     Region I, Boston, Massachusetts.  Personal Communication, 1973.

107. Areawide Assessment Procedures Manual  - Volume  I, Volume  II and Volume
     III.  USEPA Report No. EPA-600/9-76-014.  Washington, D.C.  July 1976.
                                    220

-------
                                APPENDIX A

                    Typical  Preliminary Monitoring  Data




                             APPENDIX A LEGEND
STRM NO   -    Storm Number
PRIM LOG  -    Primary CSO Location
                    1-   Maltbie Street
                    2-   Newell Street
                    3-   Rowland Street
SEC LOG   -    Secondary Location
                    1-   Untreated CSO
SEQ NO    -    Sequential Sample Number
TYPE      -    Type of Sample
                    1-   Automatic Sequential Sample
DATE      -    Date Overflow Occurred
TIME      -    Time of Sample Collection
SAMPLE    -    Number Assigned to Sample for Analytical Purposes
FLOWRATE  -    Rate of Discharge, M6D
RAINACC   -    Total Accumulation of Precipitation, in.
RAININTS  -    Rain Intensity, in./hr
pH        -    Dimensionless
TCOLI     -    Total Coliform, cells/100 ml
TOC       -    Total Organic Carbon, mg/1
COD-M     -    Chemical Oxygen Demand (Wet Chemistry Analysis), mg/1
COD       -    Chemical Oxygen Demand (Automated Analysis), mg/1
TSS       -    Total Suspended Solids, mg/1
VSS       -    Volatile Suspended Solids, mg/1
TKN       -    Total Kjeldahl Nitrogen, mg/1
TIP       -    Total Inorganic Phosphorus, mg/1
Cl        -    Chloride, mg/1
FCOLI     -    Fecal Coliform, cells/100 ml
FSTREP    -    Fecal Streptococcus, cells/100 ml
TS        -    Total Solids, mg/1
VS        -    Volatile Solids, mg/1
TDS       -    Total Dissolved Solids, mg/1
VDS       -    Volatile Dissolved Solids, mg/1
8005      -    Biochemical Oxygen Demand (5-Day), mg/1
                                    221

-------
I
                                                      Syracuse Combined Sewer Overflow Study
                                                               Maltbie Street Location
                                                             Preliminary Monitoring  Data
                                »T»H M) PHI* tOC 3EC LQC 3EU HO
                                                          0«TE  TIME
                                                                       FLU""iTE
                                                                                              TCOLl
                                                                                                    TOO COO-" COO  T8S  VS9
          ro
          po
          tva







































•









1
2
3
4
s
a
7
S
q
10
11
12
IS
10
IS
10
17
18
19
20
11
22
23
24
25
26
27
28
29
10
31
32
51
3ft
35
17
38
34
DO '
HI
42
43
••
4»
•7
06/30/72 315 226 la. 4000 .35 .20 6,9 440000. la. 27. 90.0 20.0
06/30/72 900 227 11.1500 .45 .00 7.2 200000. ». 10. 60.0 14.0
06/30/72 115 22ft 6. 4200 .50 .20 7. a 020000. 6, 10. 75.0 >,0
06/30/72 «30 229 10.9000 ,55 ,20 7.5 340000. 7. 5. 110.0 ,0
06/30V72 405 230 14T9000 ,70 .60 7.6 483000. 6. a. 70,0 .0
06/30/72 500 231 14.9000
06/30/72 515 2)2 12.7200
06/30/72 530 233 10.9000
06/30/72 515 23U 8. "200
06/30/72 600 235 5,6400
06/30/72 615 236 3.9700
06/30/72 630 237 a.0500
06/30/72 645 236 2.4700
06/30/72 700 239 1.71UO
06/30/72 715 240 1.3500
06/30/72 730 241 1.2200
06/30/72 745 3U2 .9600
06/30/72 BOO 243 .StOO
06/30/72 615 204 .8100
06/30/72 830 24S .9200
06/30/72 643 2ub ,8300
06/30/72 • 900 247 .6900
06/30/72 9l5 249 .6300
06/30/72 4)0 24* .bJOO '
06/30/72 1030 280 .4600
06/30/72 1130 231 ,4$00
06/3U/72 1230 202 .0400
06/30/72 1330 28) ,3700
06/30/72 1430 264 .2500
06/30/72 IS30 2»5
06/30/72 1630 245 ,2000
06/30/72 1730 296 .1600
06/30/71! IdiO 297 .0700
06/30/72 1930 293 .0700
06/30/72 2030 299 .3600
06/30/72 2230 301
06/30/72 2330 30* .1600
07/01/7.! 30 36) .OVOil
07/01/72 130 • 304 ,0600
07/01/72 230 305 .0900
07/01/T2 Ho 306 .1360
07/01/72 «30 307 .1400
07/01/72 J30 308 .2300
07/01/72 730 310 1.4100
07/01/72 130 311 1.2100
,80 .40 7,5 393000, 4. 7, 80,0 .0
.40 .00 7.7 330000. 11. S. 75,0 .0
.00 .40 7.7 80000. 6. S. 75.0 .0
.05 .20 7.6 290000. 12. 9. 40, 0 II. 0
.10 .20 7.7 630000. 15. 9. 375.0 50.0
,12 .08 7.7 160000. 14. , 14. 170.0 2*.0
,15 .12 7.7 20000. 20. 15. 160.0 20.0
.15 ,00 6.5 330000. 16. • 13. 640.0 70,0
.15 .00 7.1 70000. 14. IS. 350,0 «0.0
.15 .00 7.4 53000. 12. 18. 340,0 40,4
.15 .00 7,5 700000. 15. 18. 310.0 30.0
.15 ,00 7,1 560000. 21. 36. 290.0 .0
,tS ,00 r.b 31000U. Ji. 32. 180,0 ,0
.IS .00 7,0 560000. IB. 23. 170,0 .0
.15 .00 7.1 600000. 21. 22. 170.0 .0
.15 ,00 7.3~ 6lo'005. 17, 25. IU1.0 ,0
.15 .00 7.5 950000. 25. 24. 150.0 ,0
.15 .00 7,» 1050000. 23. 27. 290,0 ,0
.15 .DA 7.S 330000. . UU.U ,0
.35 .00 ,3 29. 17. 44,0 .0
.35 .00 .2 130000. 18. 14. 48,0 8.0
.35' .00 .1 V000D. il, IJ, '44,0 	 4-.ff-
.35 .00 .2 50000. 1«. 12. 40.0 8.0
.35 .00 .1 1*0000. 17. 11. 44.0 12.0
.35 .00 .3 200000. 18. 10. 34,0 .0
.35 ,00 .3 230000. 17. 11. 40.0 4.0
1 I (8 | 07/01/72 «JO • Sit i.SJOO 1,35 .00 .5 170000. 17. 14. ««,0 " «;t[-

-------
ro
ro
co
STRM NO PHlM LOt SEC LOG IE8 hO IV»t D»I£ TIME S
******
















































1
a


10
11
12
13
14
IS
10
17
ie
19
20
21
22
23
24
«s
26
27
28
29
30
31
32
33
34
35
36
IT
38
39
40
41
42
4)
45
46
47
06/30/72 345
06/30/72 "00
06/30/72 415
06/30/72 430.
06/30/72 445
06/30/72 500
06/30/72 515
06/30/72 530
06/3U/72 545
00/30/72 600
06/30/72 615
06/30/72 630
00/30/72 645
06/30/72 700
06/3U/72 715
06/30/72 730
' 06/30/72 745
06/30/72 600
06/30/72 615
L06/30/72 650
06/30/72 645
06/30/72 900
66/30/72 915
06/30/7? 930
06/3i)/72 1030
06/30/72 1130
00/30/72 1230
Ob/30/72 1330
06/30/72 1430
06/30/72 ISSO "
06/30/72 '1630
06/3(1/72 1730
00/30/72 1830
06/30/72 1930
06/3U/72 2030
00/30/72 2130
06/30/72 2230
06/30/72 2330
07/01/72 30
07/01/72 130
07/01/72 230
07/01/72 330
07/01/72 430
07/01/72 530
0//01/7S 030
07/01/72 730
07/01/72 930
' 1 	 1 4» 1 07/01/72 MO
AMPLE TKh
226 .9
22T .«
228 1.9
229 .9
230 .4
231 1.2
232 .4
233 .9
234 ,7
23S .7
236 1.2
237 1.3
238 1.4
239 ' 2.2
240 2.9
201 3.2
242 2.7
243 2.7
244 4.2
24% 4.0
240 4,3
247 3.3
248 3.9
249 4.6
280 5.3
281 6.8
•-262 5.7
283 7.7
264 7,2
26S
295 5.6
296 5.
2?7 Q (
296 3,
299 2.
300 3.
301
302 4.
io3 3.
304 3.
305 3.1
300 2.9
307 2.3
306 2,5
310 3.1
311 3.1
lit J.I
NH3N ORSN
.20 .7
.25 .6
.25 1.6
.15 .7
.15 .2
.15 1.0
.30 .1
.15 .7
.35 .3
.48 .2
.35 .8
.35 .
.50 .
.60 1.
1.40 1.
1.68 1.
1.28 1.
2.30 .
1.90 2.
1.55 2.4
2.50 l.»
l.OS 2,2
.88 3.0
1,40 3,4
1,65 3.6
1.65 5,1
,30 S.4
.35 7,3
.98 6.2
.75 a.8
.75 a._s
.75 S.o
.35 2.7
.15 1.9
.15 2.S
.58 3.5
.25 3.1
.00 3.0
.55 2.5
.50 2.4
.55 2.2
.35 l.l
.25 2.8
.20. 2,9
' ,io' ' 'J.I
N02N03 NOJM N02N TALK TIP CL FCOLI PSTRBP »S VS TOS 'VOS BOOS
.01 .00 .013 40. .1* a. 4*000. (9000. 172.0 34,0 82.0 14.0
.03 .00 .026 41. .13 10. 32000. 24000. 100.0 24.0 40.0 .0
.01 .05 .020 41. .12 10. JJoOo. 22060. 162.0 . 26.6 117,0 21.1)
.01 .06 .016 41. .10 7. 35000. 23000, 224.0 50.0 114.0 56. C
.07 .06 .014 41. .09 6. 32001. 25000. 230.0 72.0 160.0 72.0
.06 .04 ,616 42. .11 1, 26600, 26006. 216.6 84.0 IJe.O 84,0
.06 .04 .021 49. .09 8, 6300(1, 17000. 214.0 90.0 !39.0 90.0
.07 .OS .021 S3. .10 10. 19000. 15000. 206.0 84.0 131.0 84.0
,06 '.04 .021 5i. ,0V 11. . i4o06. 19660. 2og.o 126.0"" 160.6 113.6
.06 .64 .030 57. .18 16. 67000. 16000. 636.0 146.0 261.0 96,0
.10 .07 .031 83. .11 20, 34000. 34000. 324.0 B8.0 153.0 .0
.21 .1* .oil 06. .32 2i, «oo6, 1606. JS2.6 06.6 192.0 ffS.O 	
,11 .07 .036 at. .36 23. 3UOCO. 79000. 802.0 104.0 162.0 34.0
.18 .14 .044 116. .50 32. 10000. 3000. 470.0 64.0 120.0 23.0
.26 .20 ,060 135. .4V 36. 36000. 44000. 656.0 150.0 116,0 110,0
,19 .15 .040 |49. .44 44. 162000. 1SOOOO. 472.0 108.0 102.0 78.0
.25 .17 .073 163. .25 46. 178000. 22000. 464.0 130.0 174.0 130,0
.25 .16 .094 194. .60 52, 22000. 4TOOg, 416.0 J3O.O 236,6 138,0
.31 .25 .060 190. 2.49 SS, 20TOOO. 77000, 400.0 J26.I) 429,0 126.0
.35 .29 .060 192. .65 SS. 150000. 119000. 406,0 112.0 236.0 112.0
.31 ,2S .u57 1*2. 1.40 53. 104000. 496660. J7o.o 112.0 27 l7o"~IT2 , 6""
.33 .27 .063 157. ,72 4ft. 35000, 62000, 574,0 166,0 124,0 166,0
.32 .24 .081 tb». 1,21 52. 42000. 99006. 816.0 204.1) 526. 0 204.0
" .3] .2" .687' 166. .70 b7. 31666, 52006. 764.0 |~6~4~.o "#74.6 ' leA'.O
.35 .28 .066 182. .47 58. 400. 50000. 574.0 148,0 36o.O 120.0
1.47 1.22 .250 197. 1.24 64. 11000. 70000. 588.0 376.0 412,0 368.0
.95 ,6a .260 185. ."»b 54. 10000, 45000. 520.0 356,0 S8»,'6"~ 324,6
.51 .to .106 IVJ, .76 00. 400. 13600. 417.0 120.0 105.0 110,0
.64 .52 .117 212. 1.81 09, TOO. 14000. 800.0 629.0 706,0 628.0
.60 .50 .097 !9». .76 77. 3000, 26000.
.90 .51 .386 203. .79 79. 4000, 21000. 702.0 S4O.O 398.0 126.0
2,71 i.lll ,»ii ^«|4, 1,06 111, loOOO. 416110. M6.0 86.0 102.0 SZ'.O 	
.16 ,13 .031 179, ,28 83, 5000, 52600, 508.0 J10.0 U24.0 178.0
.23 .16 .073 213. .24 67. 6000. 137000. 540.0 48.0 380.0 32.0
2.67 1.84 .328 155. ,1V 54.
.81 .66 .148 175. 5.55 69, 88000. 350.0 44.0 310.0 44.0
,71 .02 .6Vj 242. 4.J/ VI, 4006, 11600. 466.6 .7
-------
                                     Syracuse Combined Sewer Overflow Study
                                           West Newell Street Location
                                           Preliminary Monitoring Data
ro
ro
»TRM NO PRIM
22
22
22
22
22
22
22
22
22
22
22
22.
22
22
22
22
22
22
' 22
22
22
22
LOC sec LOC stu NO TYPE DUE . TIME
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
1 1 09/22/72 600
_2 I 04/22/72 615
a
5
7
8
9
10
1 1_
12
13
la
IS
16
16
19
20
21
22
2*
09/22/72 630
09/22/72 645
09/22/72 700
09/22/72 715
09/22/72 730
09/22/72 70S
09/22/'2 800
09/22/72 6)5
09/22/72 836
09/22/72 845
09/22/72 900
09/22/72 9lS
09/22/72 93u
09/22/72 945
_09/22/?2 1000
04/22/72 101S"
09/22/72 1636
04/22/72 1005
04/22/72' 1 lOO"
09/22/72 1115
04/22/72 1130
09/22/72 1145
»»MPLE
874
880
831
882
883
' "884
885
886
887
888
489
890
891
892
89}
890
895
— 8*5-
897
696
694
900
401
402^
FLO«R»T£ »
.1160
.0786
.0760
.0970
.1120
.0970
.0650
.0316
.1250
.1250
.1250
Jo970
.0640
.0460
.0950
.0830
.0570 '
.0920
.0920
.0610
.0290
.0100
1 ,16«6 '
UNtCC
.10
.10
.10
.16
.16
.10
.10
.10
.10
.10
.10
" .16'
.10
.10
.10
.16
.16
.10
.10
.'16
.10
.10
.10
RilNINTS PH
.02 8.7
.00 4.1
,60 9;j
.00 4.1
.00 9.4
TCOLI
24400600.
77000060.
11100000.
34000000.
.60 6, 9" llo'OOOoO';
.00 9.5 91000.
,60 9.1 132060.
.0
15
16
17
18
19
20
21
22
23
ja
09/22/72 600
09/22/72 615
09/22/72 630
09/22/72 60S
09/22/72 700
09/22/72 715
09/22/72 736
04/22/72 785
09/22/72 900
09/22/72 815
09/22/72 930
09/22/72 005
09/22/72 900
09/22/72 915
09/22/72 930
09/22/72 °4S
09/22/72 1000
09/22/72 1015
09/22/72 1930
09/22/72 100S
09/22/72 1100
09/22/72 1115
09/22/72 1130
04/22/72 1145
SAMPLE T«N NM3N ORSN N02NOJ N03N N02N T*IK TIP CL FCOLl F97MP rS 'Vs' tOS VOS, BODS
879 11,0 6.02 5.0 .02 .61 .012 490. U.26 51. 18000066. 41006. 900.6 244.0 740.0 150.0 60.0
880 13.6 2.32 11.3 .02 .01 .012 510. 2.10 55. 200000. 22606. 798.6 248.0 708.0 180.0 78.0
881 11.2 0.61 6.6 .02 .01 .010 080. 2.16 58. 4000000. 26006, 7o6.6 19 ,159 740, 2,49 153. 600, 122000, 1140.0 230.0 988,0 150.0 75.0
891 16,7 8,47 8,2 .08 .39 .086 540. 3.03 15S. 1000, »6, 914.0 182,0 810,0 106,0 91,0
842 26.1 9,50 10.6 .40 .35 .645 388. 2.86 117. 622.0 100.0 S54.0 56.0
893 4870000, MO,
84o 10.6 5,39 5.2 ,"6 .40 .08.4 616. 2.16 92. 160. 20. 968.6 192,6 664.6 126.0 78,0
895 11.7 6.10 5.6 .04 .37 .066 381. 2.17 79. 704.0 190,0 620.0 122.0 o7_,_0_
896 '13.1 6",30 6.8 .06 .39 .041 770. 3.43 41. 1338.0 472.0 1011. 0 JiZ.O 2*1,0
897 8,2 0,54 3,7 ,46 .38 ,681 710. 1.96 66. 0. 4*0, 980.0 264.0 880. 0 196.0 93.0
898 10.2 0.23 6.0 .09 .40 .091 610. 2.09 75. 0. 100. 960.0 320.0 892.6 272,0 10.UO
899 10.0 5.51 0.5 .00 .32 ,071> 367. 3.57 68. 720.0 272.0 610.0 192.0 120.0
900 8.2 5,52 2.7 .39 .36 .025 291. 3.19 98, 690.0 202.0 olu.O 142,6 75,0
901 7.9 4.79 3.1 .31 .20 .111 362. 3.00 160. 8»8.0 250.0 790_.0182.0 70.0


-------
                                         Syracuse Combined  Sewer  Overflow Study
                                                  Rowland Street Location
                                                Preliminary  Monitoring  Data
                    $T»M MO mix lot ate me aia HO TYPE  otte  TIME at*PLE fLQ«"»TE n>n*cc R«ININTS PM    TCOLI   roc coo-*  COD  its  v»»
ro
ro
en
•«••*•«















i




1



;



»
i
;

i
i






i
i
i
i
i
i
>
t
1






1
1
1
2
1
a
5
»
1
8
9
to
11
12
11
10
IS
16
17
18
I*
20
21
22
23
24
23
26
27
28
29
10
31
32
33
30
35
36
37
38
39
ao
01
42
41
00
45
a»
1 47
1 «S J
06/10/72 "15 253
Ok/30/72 430 25 
-------

t\s
rx>
cr>
»T«M NO Ml*' UOC 3EC LOC SEU NO I»Pt DATE TIME

















\
J
*
^
3
*
3







»



1
2
3
*
5
6
7
8
9
10
11
12
13
1VT577w—
.53
.48
.3**
.50
.37
.W
.73
.3«
.19
.29
.10
.15
.1*
.«6
.«l
T;BI
ct
40,
2s.
23.
27.
33.
37.
32.
36.


T8.
82.
81.
»S.
85.
85.
86.
88.
88.
88.
89.
•12;
92.
92.
v5.
92.
93.
Oft*
93.
ai.
61,
92.
43.
93.
91.
93.
92.
V2.
91.
40.
«,
reoki
17000,
4000.
llooo.
7000.
260000.
ttoooo.
23000.
9000.
12000.
16000.
8000.
4000.
22000.
12000.
13000.
3000.
9000,
3000,
2000.
10000.
JCOOO.








FSTRE* IJ V3 TOS VOS BOOS
7000, 150.0 68.0 56,0 68.0
7000. 154.0 34.0 29.0 29.0
3000.1740.0 1440,0 1624.3 1440.4
6000. 100.0 136.0 25.0 131,0
11000. 264.0 126.0 139,0 126.0
5000. 254.0 110.0 134.0 110.0
1000, 164.0 164.0 SH. 0 164.0
2500. 300.0 130.0 118.0 10.0
3100. 328.0 152.0 218,0 152.0
15000, 376.0 158,0 201.0 113.0
12000. 416.0 154. o 231.0 114.0
4000. J80.0 232.0 215.0 192.0
•00. 432.0 158. C 287.0 158.0
1*000. 450.0 56.0 310.0 21.0
7000. 482.0 80.0 347,0 55.0
1000. 534.0 70.0 419.0 40.0
4000. S24.0 82.0 414.0 47.0
10000. 566.0 220.0 461.0 185.0
. _»•«... 554.0 200.0 429.0 160.0
7000. $40.0 86.0 420.0 51.0
7000. 572.0 128,0 472.0 103.0
*000. 564.0 90.0 424.0 5.0
1000. 60>,0 148 0 455.0 9]. 6
502.0 102,0 469.0 69.0
524.0 110.0 501.0 90.0
534.0 130^0 516.0 Il6.0"~
532.0 262,0 512.0 255.0
540.0 278.0 525.0 208.0
516.0 338.0 487,0 315,0
468.0 238,0 449.0 228.0
528.0 344.0 510.0 333.0
538.0 168.0 511.0 3Sb.o
524.0 280,0 503.0 268.0
498.0 218.0 466.0 207.0
080.0 102.0 470.6 49.6
516.0 134.0 506.0 129.0
566.0 248.0 549.0 237.0
528.0 160.0 514,0 156.0
508.0 98.0 4S6.0 93.0
•81.0 106,0 473.0 94.0
564.0 118,0 547.0 108,0
562.0 230.0 549.0 227.0
546.0 150,0 532,0 144,0
544.0 196,0 534.0 193.0
578.0 218.0 560.0 213.0
3*4.0 232.0 5*1.6 . 122.6

-------
                                 APPENDIX B

                      Chlorine Dioxide Analytical Data




                             APPENDIX B LEGEND
Actual Log Kill
C102 Dose
NH3N

Mixing Intensity GT
Influent SS

PH

Influent FC

Predicted Log Kill



Residual
Log (base 10) reduction of fecal coliform
Dosage of C102 injected into CSO, mg/1
Ammonia nitrogen concentration at point of
   disinfectant injection, mg/1
Mixing intensity expressed as-a product of the
   velocity gradient (G, sec  ) and detention
   time (T, sec), dimensionless
Suspended solids concentration at point of
   disinfectant injection, mg/1
Log of the hydrogen ion concentration,
   dimensionless
Fecal  coliform concentration at point of
   disinfectant injection, cells/100 ml
Log (base 10) of the number of fecal coliform
   killed by C102 disinfection as predicted
   by the developed performance models

Difference between the predicted log kill and
   the actual log kill.
                                    227

-------
Analytical Data for C102 Disinfection Tests
SMplt No.
Actual
LogMIl FC
C102
Dose, rog/1
NHiN
ng/1
Nixing
Intensity,
6T
Influent
SS. mg/1
PH
Influent FC
Count/100 Hi
Predicted
Logklll FC
Reslduil
*«EC02 1
3HEC.02 4
SftEC02 5
SIlElOi 7
3«EX02 8
s.frnj 9
3wtl02 10
3flEC02 11
SnEl.02 IS
SllEl.02 16
3REC03 4
3KEC03 8
SKECOI 9
3«ttUJ 10
8KEC03 11
S»EC06 4
3»EC06 5
l.tLtif, b
3*E106 7
3REC06 8
s.Flfih 9
tl)EC06 IV
SutCoti 11
3REC06 13
3REC06 14
SnELOb 16
3*EC06 17
luECOK 18
3«EC06 19
3»El06 20
3nEC06 22
ShECC'6 23
s»tco« i
3KEC08 3
s.Ernn 4
SHEC08 5
JhttuS 6
anECoa 8
S»EC08 9
R.Etoa 10
3nECo« 11
».fCO« »
SRECO* S
3*ECO« *
SHES09 •
SftECOf *
3«ECO* 11
3REC09 IS
S«tC09 |6
3nEC010 5
3KE1010 6
iNlcoin 7
3MEC010 6
ShttOlO 9
IHECOIO 11
SBlCOlO 12
9*ELU10 li
3*EC010 14
(HEkOlO 15
l.tr.nici IK
3RECOIO 17
3REC010 18
««n.n,,i 19
3HEC010 20
SnEC.010 21
3R.EC010 
1.6
4.3
(1,5
3.2
3.3
.4
,4
.1
6,5
4.7
4,5...
7.6
5.5
6.2
6.5
5,7
6.0
6.3
6.7
6.7
*,7
6.7
6.7
6.7
7.0
5,6
5.7
6,1
3,3
3.6
1.7
3.3
3.3
1.6
3.7" -
3.3
3f1
3.7
5.2
4.
*.
• 1.
1*
3.
i.
0,3
5.7
5.9
7.7
S.O
8.1
8.7
7.7
8.3
0.7
7.7
8.0
8.3
4,9
4,4
4.5
4.9
4.4
4,S
4.7
4.9
S.S
5.5
5.7
1.34
.89
1.01
1-01
1.2U
1.11
1.7? -
1.68
1.66
1 ,79
1.44
1.56
1.5*
1.30
1.24
1,21
1.13
1.02
1.39 -
3.90
3.45
2.96
3.45
3.59
2.20
1.5U
1.35
.80
.70
.60
.55
I. JO
.70
.75
.51
2.29
.34
.46
.52
__..5l
.49
.54
• V
153
.26
liS
.2*
.30
.30
.30
.21
.22
.50
.65
.51 _
.08
.12
.04
.06
.10
.09
.01
.01
.01
.09
.04
.04
,07 _.._
.01
.01
.34
.06
1000
10(10
3160
3180
MAO
284Q
3180
11)00.
1000
3865
2000
2000
20on
2000
2000
2000
3375
..3620
3933
3307
1429
3410
3429
3410 -.
3484
....^Sl*
3516
15 1«
3S16
3516
3S22 	
3662
3704
3350
3324
3429
3Soi _
2170
2157
2}57
2157
21S7
- 2»S7
2157
2157
2177 ..
2204
2344
2005
19»4
1905
-us-
1»17
1117
1*51
- 2496
2581
2620
2620
1837
3484
.3.464 ._
3484
3464
. .3484
34S4
3484
30,84
3484
2170
2170
2170
2170
2170
2170
2170
3Q3J •
3013
3031
_S1H
840
(inn
283
315
232.
190
202
247
122
110
122
135
117
140
139
136
144
88
160
216
143
172
112
too
96
100
404
540
,._ 576 ,..
555
452
412
305
180
..184
252
124
116
120
268
468 ,
480
641
634
534
540
407
527
393
__ 334..
.327
" «99
S07
200
340
161
210
220
267
2*0
227
86
187
240
114
93
104
67
52
56
92
21
118
|29
157
192
216
128
106
119
162
12*
106
93
7.0
7,2
7.3
7.3
7.2-
7.2
7.2
7..1
7.1
7.0
7.1
6.6
6.8
7.0
7.0
6.8
6.5
5.6
5.6
5,6
6.2
5.9
.S.I.
5.0
5.8
6.2
6.2
	 6.5..
6.9
6.6
	 6 .A.
6.5
6.5
6.S
6.4
6.7
..6.5
6.9
6.8
6.6_
6.6
6.6
6.6
6.6
. 6.S
6.6
6.2
6.1
5.9
»!i
6.2
6.6
6.4
S.9
S.8
6.9
6.8
-6^.9
6.8
7.0
6.9
6.8
6.9
6.9
6.9
7.1
7.0
-7.0
•6.9
7.0
7.2
7.0
7.2
7.0
7.0
7,8.
105500
56000
22000
14500
41000
48000
14500
.36000
76000
164667
122000
52000
60000
117500
25000
23000
_ 25000
60000
151667
. 131500
1175000
1025000
.3100000
895000
725000
..3800000..
2045000
1790000
2990000
5500000
JJ800000.
S100000
.2200000
71425000 . _
1260000
930000
__73SOOO
515000
925000
1945000
1210000
4900000
3940000
8300000
7900000
7900000
7500000
6400000
6600000
9900000
535000
}*»0000
160000
480000
740000
S15000
2*5000
1*3667
415000
. 130000
99333
510000
J01500
32000
17000
140000
39500
19500
. 266667.
43500
63000
46000
52000
64500
93661.
106000
46000
147333
24000
74000
25000.
34000
. «OSOO
SOOOO
1*9000
11«100
190000
2.0
2.0
2.2
2.2
2.3
2.5
2.2
.9
2.1
2.2
1.5
1.6
.3
.3
.4
,3
2.5
2.1
2.0
3.2
2.5
2.9
2.9
2.9
- 2.6
2.5
2.6
2.7
2.9
3.0
3.0
3.2
3.1
. 2.,8 .
2.9
3.1
2,7
2.6
2.7
" 1.7 ""
1.9
1.9
1.8
1.8
2.0
2.0
1.9
1.9
2.1
2.1
2.0
1.7
1.7
1.9
1.8
1.6
1.6
1.8
2.3
2.5
2.4
2.2
2.7
2.8
3.4
3.1
2.9
3,5
3.2
3.3
3.3
3.4
3.1
2.0
1.9
2.0
1.9
2.0
1.8
1.9
2.0
2.3
2.*
2.*
».* _
-1.3
•1.7
-.9
-.7
-.7
-r2
.9
-.4
-.2
1.5
• .4
-1.2
-.7
.0
.0
.1
.5
-1.4
••4
-.4
-1.6
-.9
-1.1
-.6
-1.4
-.3
,4
.2
-.7
,6
1.0
1.1
1.0
.8
2.2
2.1
1.9
.8
.7
t.l
.6
-.7
-.5
-.8
-.3
.7
...2
.4
.7
.5
it
.2
.3
.4
1.6
US
1.1
.5
.8
.5
1.1
•1.5
.3
.6
.2
.2
•l.S
.2
.5
.1
.0
.5
1.8
2.0
1.5
2.0
1.2
1.7
.2
1.3
1.3
-.*
.f
.4
I.I
                     228

-------
SnECOS 1
SnELUS 2
SnECOi ~4
SntCUS 7
3»t(,0i 0
SrtELtiS 9
SnECOS 10
3*tC05 11
snECO 12
3ntt.U5 13
S*iC05 14
s.ELO IS
SntCUS 16
Sf.tC.OS 17
S«EC.U!> 18
SnECOS 19
SdfcCUi 20
SutCUi 21
ShfcCUS 22
3r.EI.C5 23
SwELOS 24
ZURN 3 10
ZuhN 3 11
ZURN S U
ZURN S IS
7lt»N S 16
ZuRN 5 1?
ZUR" 5 18
ZURN 5 21
ZURN 5 22
ZURN 5 23
ZURN S 24
ZURN 6 1
7«ftN h 2
ZuRN 6 3
ZUHN 6 4
ZURN o o
ZURN o 7
7H0A. * H
Zll*N 6 9
ZURN 0 10
ZURN 6 12
ZURN 6 13
.- ZURN 6 .. 14
ZURN e IS
ZURN o 10
fljRN * 1 7
ZUR" 6 18
ZURN 6 19
ZUHN 0 21
ZURN 6 22
/lilifl t. 2«
ZURN 6 24
ZURN B i
man B >
ZURN 8 3
ZURN 8 4
?URN 8 *
2uR* 8 6
ZUHN a 7
7H3N 8 8
ZUHN 8 9
ZURN 8 10
. 7.IJRN 8 11
ZURN 8 12
ZURN 8 13
ZURh B 14
ZURN 8 15
2.0
2.0
2.0
3.1
oil
3,8
0.3
0.2
3.3
0.0
0.1
0.5
0.3
0.1
O.I
0,2
l.S
1.6
.9
1.7
2.0
1.7
1.0
1.3
1.0
2.5
1.3
2.3
2.5
\ 9
i!s
1.7
2.9
".'
o.O
n a_
o.s
u!l
3.9
0.0
3.5
3.7
0.9
o.o
0.7
3.3
1.3
3.7
4,3
2!5
3.5
5.0
	 S..1.
1.1
2.5
•,"
1.1
7.1
7.0
sTl
7.1
7.8
8.1
7,1 -
7.0
7.8
8.1 ___
7,1
7.4
7.8
8.1
7.1
J.4 ._ .
7.8
8.1
-UL.-.
7.4
7.8
o.o
2.6
3.7
3.7
3.7
3.7
3.7
3.7
5.0
5.0
5.0
5.4
c, q
5.4
5.0
5.5
5.5
7.1
7.1
7 p
7.0
7.5
7,5
7.5
7.5
7,5
7a
7,1
6.0
6.2
b .1
6.4
5.5
6.5
7.3
10.0
8.1
8.1
7.6
7.9
4.5
5..*.
6.1
3.1
3.8

,66
1.22
. UIS
.82
1.09
..9L
.91
.30
»35
.55
.00
.76
.71
-.7.1.
,49
.08
- .W .
.55
1.22
1.20
...,13.
.53
.57
.46
.32
.08
.55
.74
1.01
.80
4.25
^,9u
6.2U
0.05
1.85
4.55
2.15
2.00
2.80
1.35
.70
	 .40 .
.45
.35
.35
.35
.311
.25
-, .10.-.
.55
.23
.20
.19
.20.
.02
1.64
- 1.74
1.58
1.34
1.32
1.3o
1.07
5100
5100
... 51l)0
SIOO
5100
	 5100
SIOO
SIOO
5100. ,
5100
5100
	 5100
5100
5100
5100
5100
5100
	 5100
5100
51CO
SIOO
SIOO
5100
.5100.
6009
8226
8220
8226
822o
8226
8226
8220
5573
1571
5573
5573
S57S_
5573
6500
7909...
8236
8322
8010
8152
8686
	 6686
6686
6686
.8666.,
7989
8322
8322
7071
7329
. . 7397.
7466
9153
. ..7466
7400
11670
_. 15163
15163
12755
..12009
12370
7712
7466
5573
._. 6689
0689
36
9
9
10
92
. 81 .
35
04
38
80
21
., 14
52
110
56
66
32
64
94
too
96
132
78
96
128
84
240
200
-. 220 ..
176
220
323
400
488
684
200
284
244
, . 84 ._
240
SOB
. _. 836.
944
1233
900
640
. ,032
328
304
-224..
236
223
396
004
112
. 304 .
728
496
056.,
488
44B
448
420
320
396 .
392
356
288
248
144
148

6.0
6.6
6.8
6.7
— 6.7
6.7
6.9
6.. 7
6.7
6.7
6.7
6.7
6.8
	 6.7,
6.7
6.7
. . 6.7
7.3
7.6
7,0
5.5
5.8
5.?
6.2
6.0
5.9
6.0
III
7.5
7,0
6.0
6.2
f.,4
6,5
5.0
6.7
6.9
7.1
0.9
7,3
. . ,7..2
7.3
7.o
6.Z
0.7
6.6
7,0
7.0
6.9
7.0
6.8
	 6.7
6.5
6.5
. 6.6
6.5
6.6
6,6
6.0
6,6
6.5
6.8
8,1
7.0
1000
1000
47000
52000
137500
100333
1000
23000
47500
143000
164500
82000
360333
332667
79500
133000
146000
95000 .
640000
495000
153333
205000
195333
307000
86500
20500
92SOUO
15400000
5400000
1475000
1545000
0000000
2000000
6000000
3900000
1290000
79000
1750000
2600000
1655000
. .2700000 .
1305000
2055000
	 11700000 ...
3300000
7600000
9100000
10200000
1820000
2135000
3000000
1805000
.. _260000
550000
485000
1315000
1320000
485000
305000 _ .
760000
4445000
2420000
8000000
1120000
955000
1330000
510000
6300000
6400000
6200000
2440000
2600000
6800000
480000
1545000
2.5
2.8
3.6
3.6 "'
3.0
3,l_
2.6
3.0
_ 3,1..
3.2
3.6
3.7
3.3
3.2
3.2
3.4
3.3
3,1
3,6
3.6
3,0
3.2
3,2
. _3,S 	
2.2
1.4
"2'.3~~
2.3
' 2.2"
2,2
S.I
2.7
2.6
2.4
2.4
2.8
2.7
2.7
3.0
2.7
2.5
3.2
3.3
3.0
3.4
3.4
-3.5 __
3.7
3.6
. 	 J.2_.
3.1
3.2
3.3
2.9
3.0
. 2.8
2.9
2.9
2,9 ,
3.3
3.3
4.2
3.7
3.5
3.8
3.9
2.6
a, s
3.0
2.0
2,3
2.3
-.5
-.8
I'.O
.9
•*6
.1
.3
.9
.5
.1
1.0
1.0
• 6
.6
.8
.9
.9
.8
1.1
.9
.9
.7
-.7
,2
-.6
-.3
-.0
-1.2
-.9
-1.1
-.2
-1.3
-.1
.1
-1.2
-1.0
,4
l.S
.8
1.1
1.0
1.1
.6
.2
.0
,4
.5
1.7
2.0
1.4
1.9
.4
-1.7
-1.5
,4
-.2
,3
-1.0
-1.4
-.4
2,8
2.3
-1,9
.5
-1.3
-1.2
229

-------
                                 APPENDIX  C

                          Chlorine  Analytical  Data




                             APPENDIX C LEGEND
Actual Log kill
    Dose
Influent SS

PH

Influent FC

GT Mixing Intensity


Predicted Log kill


Residual
Log (base 10) reduction of fecal coliform
Dosage of Cl2 injected into CSO, mg/1
Ammonia nitrogen concentration at point of
   disinfectant injection, mg/1
Suspended solids concentration at point of
   disinfectant injection, mg/1
Log of the hydrogen ion concentration,
   dimensionless
Fecal coliform concentration at point of
   disinfectant injection, cells/100 ml
Mixing intensity expressed as,a product of the
   velocity gradient (G, sec" ) and detention
   time (T, sec), dimensionless
Log (base 10) of the number of fecal coliform
   killed by C102 disinfection as predicted
   by the developed performance model
Difference between the predicted log kill and
   the actual log kill.
                                    230

-------
    Analytial  Data  for  Clg  Disinfection  Tests
 Actual       C12    NH3N   Influent
LMklll FC  Dose, nxi/1  ran    SS.na/1
Influent FC
Counts/lOOnl
 Hlxlna
Intensity  Predicted
  6T     LogkiTl FC
                                                                    Residual
SAMPLE i
RAMBLE >
SAMPLE J
SAMPLt 4
KAMpLE 5
SAMPLt 6
SAMpLd 1
SAMPLE, rt
3AMPLE9
SAMPLtlo
SAMPL 11
SAMPLtl2
SAMPLtl3
SAMBLE 1 a
SAMpLtlS
SAMPLtl6
SAMPLE 1 1
SAMpLtlS
3AMPL 19
SSMULE20
SAMPLS22
, «AMPL>?J
SAMpLt24
SAMptt25
SAMPLE?*,
SAMBLE27
SAMPLt2B
SAMPLt.29
SAMBLtiO
SAMpLt«
SAMPLt3J
. SAMptt 34
SAMPLE SS
SAMPLES*
$AMPLt37
5AMPLt39
JAMPLtUl
SAMPLt42
5AMPLt44
SAMPLt4S
SAMPLEUS
SAMPLE47
8AMPLt48
	 SAMPJ.t4S 	
SAMPLtSO
SAMPLES 1
JAMPLES?
»ANPttiJ
tAMPkb54
fAM»Lt«
»AMPLt5»
»»MPLt*7
»A*PUIS»
•AHPLtS*
lAMPLt S
»»«>Lt 7
•AMPLE 8
•AKPL6 9
SAMPLE 10
•AMPLE 11
»A*PU 12
lAMPLt 13
SAMPLE 77
|AMPLt7»
SAMPLt79
	 JAMPLESJI 	
SAHPLtSl
JAHl»tt82
•AMPLE81
*AMPLt84
•AMPLEB5
•AMPLE96
»AMPLt87
SAMPktUB
•A«Plt«4
,3
,3
5,1
6,0
6.0
b'.D
6.0
6.0
fe.,0
6.0
6.0
6,0
6.0
6.0
z.s
2.6
4.1
3.?
3.7
3.0
2.3
2.4
4.0
Jvi-
2.3
3.5
3,0
3.0
3.6
3.7
4.0
4.S
2.6
	 2.5 	
2.6
ft
2)4
2.4.
_3ll 	
2.2
tf
3l2
3*«
3.4
«U
5^7
3,0
3.8
3.2
„ 	 1*" 	
2.1
3.1
3.1
3!?
3.2
3.2
4.1
1.7
.0
,0
.0
23.3
23.3
23.3
23,3
23.3
23.3
23.3
23.3
".*
23.3
18.6
18.*
18.6
24,0
24,0
24.0
24.0
24.0
18.0
18,0
18.0
9.0
9,0
9.0
9.0
9,0
9.0
9.0
-9..0.
12.0
12.0
—L12.0
12.0
12.0
U. 0
12.0 '
12.0
12.0
12.0
— 12.0..
12.0
12.0
1?,0
12^0
12,0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12.0
12,0
12.0
12.0
	 U.JO..
12.0
12.0
12.0
U.O
19,0
12.0
12.0
12.0
12.0
12.0
1.2B
1,11
1.9B
1.83
1.58
.96
J3
' ,68
.62
.62
.62
	 .13 	
.73
1.30
1,10
1.13
.76
.45
.56
.5*
.52
.30
.34
.40
.2B
T2H
.28
.25
.40
.45
-.53 .
.32
.11
— ...u. 	
.09
.21
.15
.17
.21"
.17
_ .28 _
.51
.43
.»•
',*}
.20
.25
.35
.40
!t>5
.71 _
.59
.60
..61
.44
1.77
.20
1.12
1.97
*2<>
.28
,3t>
.11
.16
.1?
.17
.17
.39
*62
80
180
160
392
360
275
2i(L
220
167
144
116
92
	 1j>-f
176
236
228
232
180
211
167
410
25 U
87
40
68
64
112
832
40
44
7ft.
S8S
480
	 250. .
240
220
201
193
156""
132
_. -149..
168
81
60
•2
260
250
202
186
175
. 164
130
118
8«
1240
1660
_ .3130 .
1770
750
SOO
520
530
SOO
360
280
230_.
60
20
6.0 765000
7.0 2IOUOOO
6.0 1800000
6.0 2190000
S.O 11900000
5.0 9700000
5.0 14600000
S.O 10900000
5.0 10900000
6.0 11300000
*,o itonoooo
6.0 13500000
6.0 10100000
6.0 15200000
6.0 11200000
6.0 910000
6.0 585000
7.0 665000
6.0 3800000
7.0 8600000
7.0 2000000
7.0 1445000
7.0 S60000
7.0 650000
7.0 210000
6,C 330000
7.0 540000
7.0 540000
7.0 112000
7,0 135500
6.0 116333
6.0 540000
5022
5022
5022
5022
5922
5022
5022
5022
5022
5022
	 50.22
5022
5022
_._5022._
4098
4098
4098
4098
2336
2336
2336
2336
2336
2336
2336
2336
2336
2336
2336
2336
2336
.2336
7.0 525000 2336
6.0 235000 2336
A.O 1916&7 2336
7.0 4800000
7.0 2200000
7.0 2115000
7,0 3100000
7.0 1090000
7.0 1340000
7.0 610000
7.0 810000
8.0 125667
9.0 221000
8.0 209000
7.0 645000
7.0 290000
7.0 450000
7.0 465000
7.0 325000
7,0 290000
7..0 2(0000
7.0 2*0000
7.0 315000
7.0 365000
7.0 164000
7.0 630000
(.0 230000
8.0 430000
_ .7.0 . 250000
7.0 530000
7.0 800000
7.0 .1300000
7.0 100000
7.0 900000
7.0 7.400000
6.0 370000
6.0 675000
6,0 970000
' 7.0 19500
7.0 3BOOOO
7tO- 4400.10.
7.0 21500
7.0 710000
	 7*0 	 94000.0 	
7.0 530000
7.0 460000
7.0 185000
«.0 110333
7.0 1*0000
2828
2828
2828
2828 "
2828
2828
2628
2828
282B
•2828
2828
2828
2B2B
2828
2828
2*2*
2*2*
2*21
2*2*
2*2*
2828
2*2*
409*
«0*»
4098
..4098
4098
4098
4098
4048
4098
4098
4098
4098
4098
4098
4098
4.0.98
4098
4098
4098
4098
4098
4098
4098
40**
.3
.3
.3
5.1
5.6
6.0
6.1
5.8
5.8
S.B
5.9
6.0
III
4.4
4.2
4.0
3.9
3.8
3.8
3.5
3.7
3.5
3.6
2l?
2.7
2.7
2.7
2.9
2.8
2.8
2.9
2.9
3.0
3.1
2.9
3.0
3.0
3.0
2.7
2.7
2.9
	 3.0 	
3.0
3.0
3.0
3.1
3.1
1.1
3.2
3.2
3.2
3.0
3.4
3.3
3.3
.- 3.4... .
3.5
3.5
3.6
3.5
3.6
3.9
3.3
3.3
3.1
2.8
3.2
3.3
3.1
3.0
3.4
3.4
3.5
3.4
3.6
3.9
.0
.0
.0
.0
.0
-.1
.0
.2
'.1
.1
.0
-#—
1.*
-2.1
-1.6
.1
-.1
-.*
..7
-1.3
.5
U
.1
.3
K2
.1
1.2
J-1'3
1.6
.3
.5
.5
.3
.1
.6
.2
.3
.1
.4
y
s.o
.0
1.1
1.1
"*s
.1
~^o
.7
.6
.5
2.1
.5
-.1
.1
-.7
-.1
.7
.0
!s
-.2
-.3
.7
-1.9
.2
                             231

-------
                                 APPENDIX  D

          Analytical  Data  for Sequential Addition  of Disinfectants




                             APPENDIX D LEGEND
Actual Log kill
Influent FC

Influent  SS

pH
Mixing Intensity 6T -
    Dose
C102 Dose
Predicted Log kill FC-
Residual
Log (base 10) reduction of fecal coliform
Fecal coliform concentration at point of disinfectant
   injection, counts/100 ml
Suspended solids concentration at point of
   disinfectant injection, mg/1
Log of the hydrogen ion concentration, dimensionless
Mixing intensity expressed as,a product of the
   velocity gradient (6, sec" ) and detention time
   (T, sec), dimensionless
Dosage of Cl2 injected into CSO, mg/1
Dosage of ClOg injected into CSO, mg/1
Log (base 10) of the number of fecal coliform killed
   by sequential addition of disinfectants as
   predicted by the developed performance models
Difference between the predicted logkill and actual
   logkill
                                     232

-------
Analytical Data for Sequential Addition
            of CIOz and
ACUMI
Logklll
Supl* NO. FC
CR»Nt
CUIiHf.
CRANE
CttlHk
CHAM
CBA.jE
ZURN
ZURN
ZuRn
ZuRN
{JR..
ZURN
ZURN
2URf>
ZURN
ZURN
2
E
8
10
12
13
IS
2
5
*
7
9
10
11
12
u
IS
.5
.1
.1
,3
.1
.1
!3
'.t
3
.1
.1
,0
.2
.1
!«
.6
.0
ia_
Influent FC
Counts/ 100 nl
S36000
_L65000_
183000
J 1)9000
laSnnn
181000
2101100
660000
580000
370000
nunono
300000
aiooooo
SoioOO
180000
310000
670000
7AOUOO
aaoooo
430000
HOODOO
300000
4 1)0000
	 JHOOOO ...
Influent
SS,ing/l pH
too <
100
60
ISO
200
130
120
100
90
70
170
90
430
750
1230
1490
. 1000
1490
730
79u
Oil)
390
>.t
r.4.
r.o
.1
.6
,7
.8
.1
,5
.7
,9
1.6
r.4
r.o
.1
,7
,8
,t>
.1
.0
.5
.7
.!
mxing
Intensity
GT
4220
4220
4220
_.. 4220 	
4220
4220
'«220
4220
11220
4220
7600
	 .7600 	
7600
7600
7600
7600
7aOO
' 	 7600 	
7600
7600
7600
7600
7600
	 760»._.
C1? Dose
mg/1
8.0
8.0 •
8.0
8.0
_8,0 ..
8.0
8.0
8,0
8.0
8.0
8,»
8.0
8.0
8,0
8.0
8.0
8,0
8.0
8.0
	 6,0 	
8.0
8.0
8,0
8.0
8.0
. .8.0
CIO; Dose
mg/1
2.0
1.0
1 2.0
.._2.0
2.0
2.0
.. 2,0
2.0
2*0
2.0
2.0
2,0
2.0
2.0
_ 2.0
2.0
2.0
— 2,0_
2.0
2.0
_ _2.0 _
Predicted
Logklll FC Residual
3.3
3.0
3.0
3.0
3.2
3.0
3.0
3. a
3. a
3.2
3.2
3.2
1.1
2.B
3.1
3.0
2.9
3.2
3. a
3.S
3.3
3.3
3.3
2.0 3.2 •
2.0 3.3
2.0 3.2
2
J
1
1
J 	
1
1
2
t
1
7
0
2
0
0
1
1
0
2
1
t
6
3
                   233

-------
                                 APPENDIX  E

               Analytical  Data  for  Swirl Prototype  SS  Removal




                             APPENDIX E LEGEND
Actual % SS Removed      -    Percent of suspended solids removed
Flow                     -    Flowrate, MGD (1 MGD x 3785 = 1 cu m/day)
Influent SS              -    Influent suspended solids concentration,
                                 mg/1
Foul Fraction            -    Percent of influent flow which is removed
                                 via the foul sewer outlet
Predicted % SS Removal   -    Suspended solids removed (in terms of
                                concentration)  as predicted by the
                                 developed performance model
Residual                 -    Difference between the predicted SS concen-
                                 tration removal and the actual SS concen-
                                 tration removal, percent
                                    234

-------
Swirl Prototype SS Removals
Sonic No.
SNlrtt 2

SKlKt 3
SMlfiL 7
SnlfcUO

SnIfiU4
antst 1



SwtRL 2


StIKL 6



tUlHLll



aniKLia



SWlM.16


t
2
6
7
2
l>
3
It
s
;
t
2
A
a
7
2
3
4)
13
1U
lb
16
17
19
20
a
22
23
<**
' 2
: 3
. 5
7
12
1$
; t
a
it
12
13
3
a
9
l£
24
27
•ic
>7
21
22
23
2
5
12
13
14
IS
Id
19
20
•it
22
23
3
6
7
0
10
11
It
14
10
i;
it
22
£ A
I
2
4
Actual
X SS
Removed 	
S3
49
44
34
«:9
60
36
ai
37
45
3'
69
32

-------
                                APPENDIX F
                    Swirl  Prototype Storm-Averaged Data
                             APPENDIX F LEGEND
Actual % SS Removal

Average Flow

Average Influent SS
Foul Fraction


Predicted % SS Removal


Residual
Suspended solids removed (in terms of
   concentration), percent
Average flow to the swirl unit, MGD (1 MGD
   x 3785 = 3 cu m/day)
Influent SS concentration, mg/1  ,
The flow removed via the foul sewer outlet as
   expressed as a percentage of influent flow
   to the unit
Suspended solids removed (in terms of
   concentration) as predicted by the
   developed performance model, percent
Difference between the predicted SS concen-
   tration removal and the actual SS concen-
   tration removal, percent
                                    236

-------
             Swirl  Prototype  Storm-Averaged  Data
              I SS     Average   Influent     Foul         I SS
 Stem Ho. _ Removed   Row, MGD  SS. mg/1   Fraction, I    Removed _ Residual

3H1AL  2         36     a. 15       535       28          28.2          7.8
3«1RL  I _ 23 _ 1.00 _ 182 _ 59 _ 39.7 _ -16.7
aniKk  7         T8     2750      ~TTO       25          2573         -TS73
3nlNk 10         29     2.91       210       21          24.8          4.2
SnlHlU _ 22  _ 2.22 _  1S9 _ 27 _ 27.8 _ -S.»
                          "
ariixL i
Saint i
SrilSL 6
IHIRk 1«
da
11
20
ev
27
.^o
1.03
1.42
i.'V
2.7J
•— 1M
302
342
ISO
do
36
27
!«
3U7
27.8
ia.i -
20.6
• b
.3
.8
• A
.4
SHlRk  IS         52     3.2f      115       58         39. «         12.6
                                  237

-------
                                APPENDIX G


              ADENOSINE TRIPHOSPHATE (ATP) DATA MEASUREMENTS

TABLE                              TITLE

G-l            Bacterial Reductions and ATP Measurements for Single-Stage
               Disinfection Using Cl2 or C102

G-2            Bacterial Reduction and ATP Measurements for Two-Stage
               Disinfection Using C102 (Trial A)

G-3            Bacterial Reductions and ATP Measurements for Two-Stage
               Disinfection Using C102 (Trial B)

G-4            Bacterial Reductions and ATP Measurements for Two-Stage
               Disinfection Using Cl2 and C102 (Trial A)

G-5            Bacterial Reductions and ATP Measurements for Two-Stage
               Disinfection Using Cl2 and C102 (Trial B)
                                   238

-------
TABLE 6-1.  BACTERIAL REDUCTIONS AND ATP MEASUREMENTS FOR SINGLE STAGE
                    DISINFECTION USING Cl2 or ClOg

Disinfectant
C12
C12
C12
C12
C12
C12
C12
Cl2
C12
C12
C12
C12
C102
C102
CIO?
C102
C102
C102
C102
C102
C102
C102
C102
C102
Dose, mg/1
0
5
10
15
25
100
0
5
10
15
25
100
0
5
10
15
25
100
0
5
10
15
25
100
Contact Time Bacteria, Cells/100 m"
sec
30
30
30
30
30
30
300
300
300
300
300
300
30
30
30
30
30
30
300
300
300
300
300
300
TC
4,700,000
4,900,000
4,000,000
250,000
1,300
0
4,700,000
3,400,000
44,000
1,010
5
0
33,000,000
1,150,000
300,000
130,000
6,900
300
33,000,000
760,000
40,000
34,800
700
0
FS
328,000
326,000
154,000
13,300
260
0
328,000
59,000
9,700
590
5
0
322,000
39,000
21,800
8,000
730
41
322,000
17,600
5,300
576
36
0
1 ATP
yg/l
6.320
1.250
0.750
0.730
0.068
0.011
6.320
0.730
0.220
0.070
0.039
0.009
5.120
1.100
0.740
0.210
0.047
0.039
5.120
0.460
0.280
0.057
0.019
0.006

                                 239

-------
TABLE G-2.  BACTERIAL REDUCTIONS AND ATP MEASUREMENTS FOR
       TWO-STAGE DISINFECTION WITH CIO? (TRIAL A)

Disinfectant
Dosage, mg/1
ClOg CIO?
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
0
2
2
2
2
2
2
0
0
2
2
2
2
2
0
4
4
4
4
4
4
0
0
4
4
4
4
4
Contact
Time
sec
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
Bacteria , Counts/100 ml
TC
6,380,000
407,000
165,000
22,000
49,500
22,000
5,000
4,840,000
1,430,000
1,060,000
16,500
14,300
11,000
5,500
2,860,000
715,000
29,500
7,700
3,300
1,100
770
4,180,000
1,050,000
836,000
16,500
2,640
1,430
480
FC
615,000
63,500
11,000
4,620
3,300
2,200
2,300
594,000
231,000
117,000
7,240
3,520
2,200
1,540
308,000
41,800
1,850
460
2,200
660
405
322,000
97,000
52,000
1,100
600
250
60
FS
99,000
92,400
77,000
39,600
33,000
16,500
24,900
110,000
86,900
77,000
40,700
18,700
8,580
2,970
77,000
67,100
36,300
1,100
900
770
220
114,000
91,300
83,600
11,000
3,310
540
385
ATP
yg/i
5.08
2.24
1.09
0.79
0.60
0.46
0.30
5.03
2.53
3.02
0.78
0.52
0.32
0.16
1.920
0.840
0.150
0.065
0.054
0.040
0.044
1.850
1.080
0.990
0.078
0.048
0.033
0.027

                           240

-------
TABLE 6-3.  BACTERIAL REDUCTIONS AND ATP MEASUREMENTS FOR
       TWO-STAGE DISINFECTION WITH CIO? (TRIAL B)

Disinfectant
Dosage, mg/1
CIO?
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
CIO?
0
2
2
2
2
2
2
0
0
2
2
2
2
2
0
4
4
4
4
4
4
0
0
4
4
4
4
4
Contact
Time
Sec
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
Bacteria, Counts/100 ml
TC
1,870,000
340,000
10,000
8,800
9,900
2,200
999
1,430,000
80,300
18,700
4,070
2,970
2,900
1,900
803,000
28,600
880
880
580
410
275
1,120,000
126,000
19,300
1,100
460
250
120
FC
39,600
1,100
999
110
110
330
77
29,700
1,760
360
66
77
77
55
24,800
550
20
10
10
10
0
880
440
440
0
20
0
0
FS
121,000
30,800
10,000
2,530
2,640
1,430
1,100
130,000
57,200
14,900
2,860
1,590
1,100
350
131,000
14,300
1,180
820
600
300
280
124,000
28,300
7,000
1,200
500
385
120
ATP
yg/1
3.160
0.500
0.110
0.083
0.056
0.046
0.046
1.950
0.430
0.190
0.052
0.052
0.045
0.036
2.500
0.320
0.120
0.075
0.049
	
__ —
2.020
0.520
0.057
	
	
	
	

                          241

-------
TABLE 6-4.  BACTERIAL REDUCTIONS AND ATP MEASUREMENTS FOR
   THO-STAGE DISINFECTION WITH Cl2 and C102 (TRIAL A)

Disinfectant
Dosage, mg/1
Cl2 CIO?
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
0
2
2
2
2
2
2
0
0
2
2
2
2
2
0
4
4
4
4
4
4 .
0
0
4
4
4
4
4
Contact
Time
sec
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
Bacteria, Counts/100 ml
TC
3,430,000
572,000
113,000
473,000
123,000
1,100
52,800
2,220,000
924,000
506,000
187,000
62,700
132,000
75,900
1,870,000
1,870,000
407,000
176,000
187,000
80,300
59,400
2,640,000
1,430,000
1,050,000
143,000
96,800
60,500
13,200
FC
308,000
84,700
59,400
82,500
40,700
38,500
12,100
110,000
86,900
74,800
38,500
39,600
40,700
13,200
113,000
84,700
18,700
8,800
5,500
8,910
6,160
110,000
47,300
56,100
102,000
8,800
4,180
3,410
ATP
yg/l
4.52
3.53
3.20
3.13
2.68
2.44
2.67
3.87
2.94
2.77
2.73
2.33
2.00
1.98
2.60
1.82
1.03
0.68
0.68
0.61
0.53
1.33
1.68
1.02
0.43
0.20
0.29
0.23

                          242

-------
TABLE 6-5.  BACTERIAL REDUCTIONS AND ATP MEASUREMENTS FOR
   TWO-STAGE DISINFECTION WITH C1? and CIO? (TRIAL B)

Disinfectant
Dosage, mg/1
Cl2 CIO?
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
0
2
2
2
2
2
2
0
0
2
2
2
2
2
0
4
4
4
4
4
4
0
0
4
4
4
4
4
Contact
Time
sec
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
0
15
30
45
60
90
120
Bacteria, Counts/100 ml
TC
187,000
121,000
90,000
1,100
1,210
660
110
220,000
61,600
4,400
1,100
330
99
99
660,000




999




99
100,000


18,700


110


99
FC
12,100
1,430
99
99
99
99
9
9,900
2,420
2,250
320
165
9
11
30,200
	


220




9
22,500


3,630


90


9
FS
82,500
38,500
2,200
3,300
3,850
1,980
1,870
72,900
17,600
3,300
990
2,750
1,980
1,320
11,000
	


999




99
74,800


8,800


110


44
ATP
yg/i
0.720
1.030
0.280
0.290
0.120
0.070
0.055
0.960
0.610
0.140
0.150
0.074
0.073
0.065
1.010
0.710
0.089
0.049
0.027
0.039
0.027
0.820
1.020
0.520
0.052
0.035
0.025
0.019

                            243

-------
                                  TECHNICAL REPORT DATA
                           (Please read Instructions on the reverse before completing)
1. REPORT NO.

  EPA-600/2-79-134
                             2.
                                                           3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
  DISINFECTION/TREATMENT OF COMBINED SEWER OVERFLOWS
  Syracuse, New York
               5. REPORT DATE
                August 1979 (Issuing Date)
               6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
  Frank J, Drehwing, Arthur J.  Oliver,  Dwight A. MacArthu
  Peter E. Moffa
                                                           8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  O'Brien & Gere Engineers,  Inc.
  1304 Buckley Road
  Syracuse, New York   13221
               10. PROGRAM ELEMENT NO.

                1BC822, SOS #1, Task  18
               11. CONTRACT/GRANT NO.
                                                              S802400  (11020HFR)
12. SPONSORING AGENCY NAME AND ADDRESS  ,
  Municipal Environmental Research  Laboratory—Cin.,OH
  Office of Research & Development
  U.S.  Environmental Protection Agency
  Cincinnati, Ohio  45268
               13. TYPE OF REPORT AND PERIOD COVERED
                    Final 1971-1978	
               14. SPONSORING AGENCY CODE
                    EPA/600/14
15. SUPPLEMENTARY NOTES
                     Supplement  to EPA-670/2-75-021, "Bench-Scale  High-Rate Disinfection
  of Combined Sewer Overflows"   -   Project Officer:  Richard Field (201)  321-6674
 	FTS 340-6674	
16. ABSTRACT
  The Syracuse demonstration  program was designed to evaluate  high-rate disinfection/
  treatment of CSO.  The  study covered field evaluations of  high-rate treatment and
  disinfection by the following unit processes:  three separate microscreening devices,
  swirl regulator/concentrator, and disinfection utilizing chlorine  and chlorine
  dioxide.
  The three microscreening  units were evaluated employing various  hydraulic loading
  rates.  Using multiple  regression analysis techniques, mathematical  performance
  models were developed for each unit relating suspended solids removal efficiencies
  to hydraulic and solid  loading rates, and the results are  presented in the reports.
  Similarly, performance  models were developed for the treatment efficiency of the swirl
  regulator/concentrator  and  are reported.  Multiple regression modeling of the dis-
  infection data yielded  statistically significant performance equations for the high-
  rate disinfection systems.   The models provided an analysis  of sensitivity to mixing
  intensity and detention time for the two disinfectants.
  Capital and operating cost  estimates indicated that solids removal  via swirl
  regulation/concentration  followed by disinfection by chlorine was  the least expensive
  CSO abatement strategy  for  those abatement options evaluated in  this study.
17.
                               KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                             c. COSATI Field/Group
  Combined sewers, Disinfection,  Chlorine,
  Water pollution, Waste treatment,  Sewage
  treatment, Wastewater, Sewage,  Overflows,
  Sewers, Collection
   High-Rate treatment/dis-
   infection, CSO character-
   zation, Microscreening,
   Suspended solids monitor,
   Swirl  regulator/concentra
   tor, Adenosine triphospha
   'otal organic analyzer
                                                                               13B
                                                                         •e,
18. DISTRIBUTION STATEMENT


      RELEASE TO PUBLIC
  19. SECURITY CLASS (ThisReport)
      UNCLASSIFIED
21. NO. OF PAGES
      262
  20. SECURITY CLASS (Thispage)
      UNCLASSIFIED
                             22. PRICE
EPA Form 2220-1 (Rev. 4-77)
244
                                                                ft U S. GOVERNMENT PRINTING OFFICE: 1979 -657-060/5453

-------