United States
Environmental Protection
Agency
Municipal Environmental Research EPA-600/2-79-085
Laboratory August 1979
Cincinnati OH 45268
Research and Development
Combined Sewer
Overflow
Treatment by
Screening and
Terminal Ponding
Fort Wayne, Indiana
-------
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 lntforma-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-79-08?
August 1979
COMBINED SEWER OVERFLOW TREATMENT
BY SCREENING AND TERMINAL PONDING
Fort Wayne, Indiana
by
Delmar H. Prah
Howard Needles Tammen & Bergendoff
Indianapolis, Indiana 46205
and
Paul L. Brunner
City Utilities
Water Pollution Control Plant
Fort Wayne, Indiana 46802
Grant No. 11020 GYU
Project Officers
Clifford J. Risley, Jr.
U.S. Environmental Protection Agency
Region V
Chicago, Illinois 60606
and
Hugh Masters
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 publica-
tion. 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.
ii
-------
FOREWORD
The Environmental Protection Agency was created because of increasing public
and government concern about the dangers of pollution to the health and wel-
fare of the American people. Noxious air, foul water, and spoiled land are
tragic testimony to the deterioration of our natural environment. The com-
plexity 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 so-
lutions. 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 munici-
pal and community sources, for the preservation and treatment for 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 study contained herein documents the demonstration and evaluation of a
75 MGD combined sewer overflow treatment facility to obtain plant-scale data
on the effectiveness and costs of screening combined sewer overflows by three
types of fine screens. The overall results of the project indicate the im-
portance of physical processes for stormwater treatment due to their adapta-
bility to automated operation, rapid startup and shutdown characteristics,
high rate operation and very good resistance to shock loads.
Francis T. Mayo
Director
Municipal Environmental Research
Laboratory
iii
-------
ABSTRACT
The Fort Wayne Stormwater Treatment Evaluation Project was constructed to
provide plant-scale data on the effectiveness and costs of screening combined
sewer overflows and to quantify the benefits to the receiving water by pollu-
tant removal by screening, chlorination and ponding previously by-passed
combined sewer overflows. Three types of screens were evaluated including
fixed side hill vertical screens, rotary centrifugal fine screening, and
horizontal rotary drum fine screening. Design flowrate for the Facility
constructed in 1973-74 is 75 MGD. Discrete grab samples were taken and
analyzed for pH, temperature, setteable solids, fecal coliform bacteria,
turbidity, chemical oxygen demand, five-day biochemical oxygen demand,
suspended solids, volatile suspended solids, ammonia nitrogen, organic
nitrogen total phosphorus, dissolved oxygen and chlorine residual. Thirty-
eight separate stormwater treatment events were analyzed over a thirteen-month
period beginning in January, 1975.
Of the screening methods studied, removal efficiences for suspended solids,
BODs, COD, and nutrients were not significantly different from zero at the
95% confidence level. No effect between hydraulic loading and removal
efficiency was found. The least overall cost method of screening was the
fixed vertical-type screen ($25.25/MG based on annual operations and main-
tenance costs). All the screens studied have hydraulic head requirements on
the order of 2 to 10 feet static loss.
A dual-use, 32-acre terminal pond was shown to be capable of meeting secon-
dary effluent standards (30 mg/1 as BOD5 and suspended solids) 95 percent of
the days studied in the four-year evaluation period. No perceptible effect
of the addition of storm-caused combined sewer overflows was found on any
parameter studied. Average monthly removal efficiency for BOD5 and suspended
solids was 75 percent for both. Loading rates greater than 200 pounds per
acre per day for both BOD5 and suspended solids were shown to result in
stable, relatively high efficiency performance for the pond at a hydraulic
detention period ranging from 1.1 to 2.0 days.
The construction of the Stormwater Treatment Facility resulted'in a reduction
of 82 percent of the by-passing of raw, untreated combined sewer overflows to
the receiving stream from the tributary area. A reduction of the quantity of
raw wastewater B0t>5 reaching the river due to combined sewer overflows
throughout the City of Fort Wayne was reduced by 54 percent. The stream
showed improvements in instream dissolved oxygen, 6005, and fecal coliform
bacteria. However, only the fecal coliform bacteria improvement was solely
attributed to the treatment of combined sewer overflows since upstream
dissolved oxygen and BODr also showed similar improvements.
This report was submitted in fulfillment of Grant 11020 GYU by the City of
Fort Wayne, Indiana under the partial sponsorship of the U. S. Environmental
Protection Agency. This report covers a period from April 1971 to February
1976, and work was completed as of October 1977.
iv
-------
TABLE OF CONTENTS
FOREWORD
ABSTRACT
LIST OF FIGURES
LIST OF TABLES
ACKNOWLEDGMENTS
SECTION 1 INTRODUCTION
SECTION 2 CONCLUSIONS
SECTION 3 RECOMMENDATIONS
SECTION 4 DESCRIPTION OF THE PROBLEM
COMBINED SEWER OVERFLOW TREATMENT REQUIREMENTS
AT FORT WAYNE
SELECTION OF SITE
DESCRIPTION OF THE COMBINED SEWER SERVICE AREA
DESCRIPTION OF THE STORMWATER PUMPING AND
SCREENING FACILITY
SECTION 5 EXPERIMENTAL PROGRAM
PRE-CONSTRUCTION EVALUATION PHASE
River Sampling Locations
Pre-Construction Sampling Program
POST-CONSTRUCTION EVALUATION PHASE
Screening Equipment Locations and Flow Measurements
Post-Construction Sampling Program
Screening Equipment Efficiency Evaluations
Stormwater Screening Facility Impact on Water
. Pollution Control Plant
Determination of Shock and Total Pollutional
Loads on the Receiving Stream
Screening Methods Cost Analysis
PAGE
iii
iv
vii
xi
xiii
1
2
8
10
10
11
11
15
24
24
24
25
26
26
31
35
36
37
37
v
-------
TABLE OF CONTENTS (CONTINUED)
PAGE
SECTION 6 RESULTS OF THE EXPERIMENTAL PROGRAM 39
STREAM WATER QUALITY IMPROVEMENTS 39
Introduction 39
Dissolved Oxygen 39
Biochemical Oxygen Demand 46
Fecal Coliform Group Indicator Organisms 47
Total Phosphorus 49
Ammonia Nitrogen 53
Suspended Solids (Nonfilterable Residue) 53
Stormwater Facility Evaluation Parameters 54
Settleable and Flotable Debris 54
Hydraulic Capacity 57
Stormwater Screening Equipment Comparative
Evaluation 62
Suspended Solids (Nonfilterable Residue) 62
Biochemical Oxygen Demand (BOD,-) 71
Total Phosphorus (Total P) 76
Ammonia Nitrogen (NHj-N) 80
Fecal Coliform Group Indicator Bacteria 81
Screening Efficiency Net Overall Treatment
Efficiency '• 81
Operations, Maintenance and Special Considerations 81
Stormwater Facility Conduits and Wetwell 81
Stormwater Pumps 82
Stormwater Distribution System 82
Bauer HydrasieyeQy . 82
Rex Rotary Drum Screen 83
SWECO CWC®Centrifugal Wastewater Concentrator 86
Stormwater Facility Chlorination 89
Effectiveness of the Stormwater Facility in
Dealing with Combined Sewer Overflows 91
Stormwater Effects on Terminal Pond 92
Dissolved Oxygen 99
Biochemical Oxygen Demand 100
Chemical Oxygen Demand 107
Suspended Solids (Nonfilterable Residue) 111
Total Phosphorus 116
SECTION 7 COST ANALYSIS FOR THE PROGRAM ' 118
CONSTRUCTION COSTS FOR THE STORMWATER FACILITY 118
ANNUAL COSTS OF OPERATION AND MAINTENANCE 121
REFERENCES 124
VI
-------
LIST OF FIGURES
FIGURE
3
4
8
9
10
11
12
13
14
15
16
AERIAL PHOTOGRAPH OF WATER POLLUTION CONTROL PLANT -
CITY OF FORT WAYNE, INDIANA
AERIAL PHOTOGRAPH OF STORMWATER TREATMENT FACILITY -
CITY OF FORT WAYNE, INDIANA
PAGE
12
12
SERVICE AREA LOCATION PLAN - CITY OF FORT WAYNE, INDIANA 13
14
WATER POLLUTION CONTROL PLANT - 84-INCH BYPASS
DIVERSION CHAMBER
WATER POLLUTION CONTROL PLANT - STORMWATER
TREATMENT FACILITIES
STORMWATER TREATMENT FACILITY - CITY OF FORT WAYNE,
INDIANA
CONSTANT HEAD TANK RAW STORMWATER DISTRIBUTION
DEVICE
STORMWATER TREATMENT FACILITY FLOW ROUTING
SCHEMATIC DIAGRAM .
BAUER HYDRASIEVE^AS ORIGINALLY INSTALLED
'AS MODIFIED WITH SPLASH GUARDS -
BAUER HYDRASIEVE1
EAST BAUER
REX ROTARY DRUM SCREEN
REX ROTARY DRUM SCREEN AUTOMATIC BACKWASH
COLLECTION TROUGH
SWECO CWC-60^CENTRIFUGAL WASTEWATER CONCENTRATOR
NORTH BANK OF SWECO CWC-60 ® SCREENING UNITS
MAUMEE RIVER MEAN STREAM DISSOLVED OXYGEN - ANTHONY
BOULEVARD BRIDGE - JANUARY, 1970 TO FEBRUARY, 1976
MAUMEE RIVER MEAN STREAM DISSOLVED OXYGEN - U.S. 30
BYPASS BRIDGE - JANUARY, 1970 TO FEBRUARY, 1976
16
17
17
18
19
19
21
21
23
23
43
44
vii
-------
LIST OF FIGURES (CONTINUED)
FIGURE
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
MAUMEE RIVER MEAN STREAM DISSOLVED OXYGEN -
NEW HAVEN BRIDGE AT LANDIN ROAD - JANUARY,
1970 TO FEBRUARY, 1976
STREAM FECAL COLIFORM GROUP INDICATOR ORGANISMS -
PRE-CONSTRUCTION EVALUATION PHASE - AUGUST, 1971
TO NOVEMBER, 1972
STREAM FECAL COLIFORM GROUP INDICATOR ORGANISMS -
POST-CONSTRUCTION EVALUATION PHASE - JANUARY,
1975 TO FEBRUARY, 1976
SCREENED MATERIAL ADHERING TO TRASH RACK
VERTICAL BARS
INTERIOR VIEW OF STORMWATER FACILITY TRASH RACK
SHOWING STRINGY DEBRIS
JOHNSTON 30 PS PUMP COLUMN AND SUCTION BELL
AFTER ELEVEN MONTHS OF OPERATION
CHLORINE CONTACT CHANNEL FLOTABLE DEBRIS COLLECTED
AT SCUM REMOVAL PIPE-- JULY 24, 1975
BAUER HYDRASIEVE
EVENT
'SCREENINGS AFTER STORMWATER
CLOSE-UP OF SEMIDRY SCREENINGS COLLECTED DURING
STORMWATER EVENT FROM BAUER'HYDRAS IEVE®
(R)
BAUER HYDRAS IEVE^ CENTER CONCENTRATE CHANNEL
SHOWING SEMIDRY SCREENINGS COLLECTED
BAUER HYDRASIEVE ®!OPERATING AT FULL HYDRAULIC
CAPACITY .
REX ROTARY DRUM SCREEN HEADBOX OVERFLOW POINT
CHLORINE CONTACT CHANNEL FROM OVERFLOW WEIR TO
TERMINAL POND SHOWING SLUDGE BANK DEPOSITS
CHLORINE CONTACT CHANNEL BOTTOM AT OVERFLOW
WEIR TO TERMINAL POND
SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR
SUSPENDED SOLIDS (NONFILTERABLE RESIDUE) AS A
FUNCTION OF HYDRAULIC LOADING RATE
PAGE
45
50
51
56
56
56
58
58
58
59
59
61
61
61
68
viii
-------
LIST OF FIGURES (CONTINUED)
FIGURE
32
33
34
35
36
37
38
39
40
41
42
43
44
45
SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR
SUSPENDED SOLIDS (NONFILTERABLE RESIDUE) AS A
FUNCTION OF SOLIDS LOADING RATE
SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR
BIOCHEMICAL OXYGEN DEMAND (BOD,.) AS A FUNCTION
OF HYDRAULIC LOADING RATE
SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR
BIOCHEMICAL OXYGEN DEMAND (BOD5) LOADING RATE
SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR
CHEMICAL OXYGEN DEMAND (COD) AS A FUNCTION
OF HYDRAULIC LOADING RATE
SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR
CHEMICAL OXYGEN DEMAND (COD) AS A FUNCTION
OF COD LOADING RATE
SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR TOTAL
PHOSPHORUS (TOTAL P) AS A FUNCTION OF HYDRAULIC
LOADING RATE
SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR TOTAL
PHOSPHORUS (TOTAL P) AS A FUNCTION OF TOTAL P
LOADING RATE
REX ROTARY DRUM SCREEN OPERATING ON RECIRCULATED
POND WATER
SWECO CWC ^ SCREENS IN OPERATION ON RAW STORM-
WATER OVERFLOWS SHOWING EFFLUENT SUDSING EFFECT
SWECO INFLUENT DISTRIBUTION MANIFOLD FOR SOUTH
BANK OF CWC UNITS
STORMWATER TREATMENT FACILITY FLOW ROUTING
SCHEMATIC DIAGRAM
RAINFALL HYDROGRAPH AND LONGITUDINAL HYDRAULIC
FLOW IN MGD - PART 1 AND PART 2
LONG-TERM PLANT RAW SEWAGE PUMPING DAILY
HYDRAULIC FLOW IN MGD
WATER POLLUTION CONTROL PLANT EFFLUENT DISSOLVED
OXYGEN LONG-TERM MEAN MONTHLY D.O. CONCENTRATION
PAGE
70
72
73
75
77
78
79
85
90
90
93
95
98
101
IX
-------
LIST OF FIGURES (CONTINUED)
FIGURE
46
47
48
49
50
51
52
53
54
PLANT TERMINAL POND EFFLUENT BIOCHEMICAL
OXYGEN DEMAND
WATER POLLUTION CONTROL PLANT EFFLUENT
BIOCHEMICAL OXYGEN DEMAND (BODj LONG-TERM
MEAN MONTHLY BOD5 CONCENTRATION
LONG-TERM TERMINAL POND MEAN MONTHLY ORGANIC
LOADING RATE
TERMINAL POND MEAN MONTHLY BODc LOADING
REMOVAL EFFICIENCY AS A FUNCTION OF POND
INFLUENT BOD- LOADING RATE
TERMINAL POND EFFLUENT CHEMICAL OXYGEN
DEMAND LONG-TERM MEAN MONTHLY CONCENTRATION
WATER POLLUTION CONTROL PLANT EFFLUENT
SUSPENDED SOLIDS LONG-TERM MEAN MONTHLY
SUSPENDED SOLIDS CONCENTRATION
LONG-TERM TERMINAL POND MEAN MONTHLY SUSPENDED
SOLIDS LOADING RATE
TERMINAL POND MEAN MONTHLY SUSPENDED SOLIDS
LOADING REMOVAL EFFICIENCY AS A FUNCTION OF
POND INFLUENT LOADING RATE
WATER POLLUTION CONTROL PLANT EFFLUENT TOTAL
PHOSPHORUS LONG-TERM MEAN MONTHLY CONCENTRATION
PAGE
103
104
108
109
110
112
114
115
117
x
-------
LIST OF TABLES
TABLE
3
4
5
6
10
11
12
13
14
15
PRE-CONSTRUCTION EVALUATION PHASE SAMPLING PLAN -
AUGUST, 1971 TO NOVEMBER, 1972
POST-CONSTRUCTION EVALUATION PHASE SAMPLING PLAN -
JANUARY, 1975 TO FEBRUARY, 1976
PRE-CONSTRUCTION EVALUATION PHASE STREAM DISSOLVED
OXYGEN - AUGUST, 1971 TO NOVEMBER, 1972
POST-CONSTRUCTION EVALUATION PHASE STREAM DISSOLVED
OXYGEN - JANUARY, 1975 TO FEBRUARY, 1976
PAGE
27
32
41
41
NUMBER OF DAYS STREAM DISSOLVED OXYGEN LESS THAN 4.0 mg/1 42
PRE-CONSTRUCTION EVALUATION PHASE STREAM BIOCHEMICAL
OXYGEN DEMAND - AUGUST, 1971 TO NOVEMBER, 1972 46
POST-CONSTRUCTION EVALUATION PHASE STREAM BIOCHEMICAL
OXYGEN DEMAND - JANUARY, 1975 TO FEBRUARY, 1976 47
PRE-CONSTRUCTION EVALUATION PHASE STREAM FECAL COLIFORM
GROUP INDICATOR ORGANISMS - AUGUST, 1971 TO NOVEMBER, 1972 48
POST-CONSTRUCTION EVALUATION PHASE STREAM FECAL COLIFORM
GROUP INDICATOR ORGANISMS - JANUARY, 1975 TO FEBRUARY, 1976 49
PRE-CONSTRUCTION EVALUATION 'PHASE STREAM TOTAL
PHOSPHORUS - AUGUST, 1971 TO NOVEMBER, 1972 49
POST-CONSTRUCTION EVALUATION PHASE STREAM TOTAL
PHOSPHORUS - JANUARY, 1975 TO FEBRUARY, 1976 . 52
POST-CONSTRUCTION EVALUATION PHASE AMMONIA NITROGEN -
JANUARY, 1975 TO FEBRUARY, 1976 53
STREAM SUSPENDED SOLIDS (NONFILTERABLE RESIDUE) 54
STORMWATER FACILITY WETWELL SEDIMENTS SAMPLING SURVEY .
RESULTS - DECEMBER 8, 1975 ' 55
SCREENING EQUIPMENT COMPARATIVE EVALUATION OVERALL
TREATMENT EFFICIENCY 63
xi
-------
LIST OF TABLES (CONTINUED)
TABLE PAGE
16 SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR SUSPENDED 69
SOLIDS (NONFILTERABLE RESIDUE)
17 SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR VOLATILE
SUSPENDED SOLIDS (VSS) 71
18 SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR BIOCHEMICAL
OXYGEN DEMAND (BOD5) 71
19 SCREENING FACILITY FILTERABLE BODg DETERMINATIONS -
EVENT NO. 207 74
20 SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR CHEMICAL
OXYGEN DEMAND 76
21 SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR TOTAL
PHOSPHORUS (TOTAL P) 80
22 SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR AMMONIA
NITROGEN 80
23 SCREENING EQUIPMENT NET OVERALL TREATMENT EFFICIENCY
IN PERCENT 81
24 WATER POLLUTION CONTROL PLANT TERMINAL POND EFFLUENT
DISSOLVED OXYGEN ' 99
25 PLANT TERMINAL POND EFFLUENT BIOCHEMICAL OXYGEN DEMAND 102
26 PLANT SECONDARY TREATMENT EFFLUENT BIOCHEMICAL OXYGEN
DEMAND . 105
27 PLANT TERMINAL POND EFFLUENT SUSPENDED SOLIDS
(NONFILTERABLE RESIDUE) 111
28 PLANT SECONDARY TREATMENT EFFLUENT SUSPENDED SOLIDS 113
29 DETAILED COSTS OF CONSTRUCTION OF FORT WAYNE
STORMWATER TREATMENT FACILITY 118
30 COMPARATIVE ANNUAL OPERATIONS AND MAINTENANCE COSTS
FOR THREE SCREENING METHODS 123
XII
-------
ACKNOWLEDGMENTS
No project of this scope and magnitude would be possible•without the assist-
ance of many dedicated people who have taken an interest in the conduct of
the work. We wish to extend our sincere appreciation and acknowledgment to
the following individuals who have been connected with this project through-
out its life:
Mr. Ron Bonar, Consultant and Former City Engineer
City of Fort Wayne
Mr. Phil Boiler, Pollution Control Engineer
City of Fort Wayne
Mr. Dick Koos, Water Pollution Control Plant Chemist
City of Fort Wayne
The following individuals at Howard Needles Tammen & Bergendoff, Henry B.
Steeg & Associates Division, deserve special credit for their interest and
cooperation during this project:
Mr. Jerry Lynch, Project Representative
Mr. Charles F. Niles, Jr., Special Project Consultant
Mr. George K. Erganian, Partner
The Contractor for the Stormwater Facility was:
Bowen Engineering Company, Inc.
Indianapolis, Indiana
The following companies are recognized for their continuing interest and
assistance throughout this project:
Bauer Bros. Division of Combustion Engineering, Inc.
Envirex, Inc., Water Quality Control Division
SWECO, Inc., Environmental Systems Division
Special thanks to Byron L. Anderson of B. L. Anderson, Inc., for his
technical assistance and special interest.
xiii
-------
-------
SECTION 1
INTRODUCTION
One of the major difficulties confronting most urbanized areas in this
country is the so-called "stormwater" problem. Every community in this
nation faces this problem at some time or another. The difficulty with deal-
ing with this problem is its quantity and quality variations. However, the
manner of dealing with storm-caused pollutant-laden wastewater has historic-
ally been a temporary, local solution. The net effect of these projects de-
signed to "solve" a local stormwater-caused problem has frequently resulted
in a transfer of the problem of storm-caused water pollution to a later time
frame and greater distance from the source.
So it is with one component of this stormwater problem that has been termed
"combined sewer overflows". The sewers were designed to carry dry-weather
sanitary wastes in the same conduits as wet-weather stormwater generated by
precipitation over and above the dry-weather flows. This storm-caused com-
bined sewer flow usually is greater than the capability of the treatment
plant to handle hydraulically, let alone treat to the equivalent of secondary
treatment removal efficiencies.
This work was intended to determine the effectiveness and cost of screening
combined sewer overflows from plant bypass wastewaters by each of the screen-
ing methods consisting of side hill vertical fixed screens, centrifugal waste-
water concentration and rotary drum screenings followed by two-day terminal
ponding of screened effluent plus secondary treatment plant effluent to eval-
uate the benefits to the receiving waters by the utilization of the methods
cited for capturing and treating plant bypass of combined sewer overflows,
and to quantify the effectiveness of the screening methods studied as
tertiary treatment units on the two-day terminal ponds.
-------
SECTION 2
CONCLUSIONS
1. There has been an increase in stream dissolved oxygen since
the diversion of the combined sewer overflows to the Stormwater Treatment
Facility.
a. The relative quantity of raw, untreated combined sewer overflows
originating from storm-caused bypassing at the Water Pollution
Control Plant has been reduced by 28 percent.
b. The mean stream dissolved oxygen observed showed a measurable in-
crease since the start-up of the S'tormwater Facility to an average
concentration of 8.9 milligrams per liter (mg/1), from pre-
construction mean of 7.4 mg/1, at a point 7-1/4 kilometers
downstream.
c. The frequency of stream dissolved oxygen levels less than 4.0 mg/1
measured at a point 7-1/4 kilometers downstream decreased from 23
days per year to 11 days per year since construction of the Storm-
water Facility. This dissolved oxygen improvement is not solely
attributed to the Stormwater Facility since upstream disssolved
oxygen levels less than 4.0 mg/1 also decreased from 6.9 days per
year to 0.04 days per year. These results are statistically
significant at the 95 percent confidence level.
2. There has been a decrease in the stream biochemical oxygen demand since
the construction and operation of the Stormwater Facility.
a. An estimated 54 percent reduction in the overflow quantity of BOD5
has been achieved since the start-up of the Stormwater Facility.
b. The range of values of stream BOD5 has been decreased since
the start-up of the Stormwater Facility - from 0.5 to 25 mg/1
in the pre-construction phase to 0.2 to 11.0 mg/1 in the post-
construction phase. However, the upstream BOD5 experienced a
similar reduction.
3. There has been a significant improvement in the stream water quality as
measured by membrane filter method for fecal coliform group indicator
bacteria.
a. The stream geometric mean fecal coliform was 9,000 organisms per
100 milliliters in the pre-construction phase and 1,000 per 100
milliliters in the post-construction phase.
-------
b. There must be measurable water quality improvements for fecal coli-
form upstream from the Water Pollution Control Plant to show any
additional improvements in stream fecal coliform downstream from the
plant.
Decrease in the stream total phosphorus after completion of the Water
Pollution Control plant chemical precipitation phosphorus removal and
Stormwater Facility are mostly attributable to the chemical precipitation
of phosphorus at the plant.
There was not distinguishable reduction in stream suspended solids over
the evaluation period in spite of an estimated 70 percent reduction in
suspended solids originating in storm-caused combined sewer overflows.
Among the screening methods evaluated, the fixed vertical side hill
hydraulic sieve (Bauer Hydrasieve®) was shown to be most effective in the
removal of coarse solids debris.
The Bauer unit produced a semidry solid residue that could be re-
moved and trucked off to a sanitary landfill without further
treatment at a hydraulic loading rate of approximately 25 gpm
per square foot (1,050
Of the three screening methods studied none of the pollutant removal
efficiencies were shown to be statistically different from zero at
the 95 percent confidence level.
SCREENING EQUIPMENT OVERALL TREATMENT
EFFICIENCY IN PERCENT*
Suspended Solids
Biochemical Oxygen Demand
Chemical Oxygen Demand
Total Phosphorus
Bauer
Hydrosieve*
11 Mesh
15.7 (4.7)
13.1 (5.7)
13.6 (6.5)
13.3 (6.9)
Rex Rotary
Drum Screen
100 Mesh
15.8 (9.2)
14,4 (3.4)
11.3 (1.5)
11.4 (2.0)
SWECO
CWC®
165 Mesh
26.6 (22.6)
16.5 ( 3.1)
19.5 (12.9)
18.9 ( 8.3)
^numbers in parenthesis are the overall treatment efficiencies
including negative efficiency data.
Negative treatment efficiencies calculated for screening units may
be due to sampling, analytical and compositing errors. Excessive
fluid shear may also account for a portion of the variation and
negative results. The negative removal efficiencies for suspended
solids, for example, are calculated to be as high as -85 percent
although they generally fall between 0 and -30 percent. Actual
negative removals are impossible for suspended solids since the
screening units are physical barriers.
-------
9. There does not appear to be a correlation between screening efficiencies
and hydraulic loading rates for any of the screens studied. There does'
appear to be an event-by-event variation in treatment efficiencies in-
dependent of the hydraulic loading.
10. There was no discernable trend in screening rate efficiencies as a func-
tion of pollutant unit loading rates for any of the parameters studied.
Rather, there appeared to be a wide dispersion of results from event to
event independently of unit loadings.
11. None of the screening methods studied appeared to have any effect on the
removal of fecal coliform indicator bacteria.
12. The fixed vertical side hill screening unit makes an excellent screening
device for removal of solid debris such as wood pieces, metals, plastic
materials, leaves and twigs, and other such gross solids:
a. As a screening unit in and of itself, the Bauer unit performance is
comparable to the other screens studied.
b. As a. pretreatment device, the Bauer was capable of reduction of net
overall pollutant loadings to subsequent treatment methods.
c. Only routine preventive maintenance was found to be necessary for
this unit.
d. Headless considerations are important for the Bauer because it re-
quires about two meters (6.6 feet) static headless to operate.
e. Minimum floor area requirements for this installation were 66.3
square meters or approximately 0.85 square meters per million liters
per day hydraulic capacity.
f. There were no auxiliary power or other services required for this
installation. :
g. Semidry screenings were possible by operating the Bauer unit at
about 3/4 of maximum hydraulic capacity.
13. Screening of combined sewer overflows by the rotary fine screening
method (Rex Rotary Drum Screen) resulted in the following conclusions:
a. The Rex rotary drum screen installed at Fort Wayne required more
corrective mechanical maintenance on the part of plant personnel
than any -other screening method.
b. The Rex unit experienced one screen failure in the evaluation
period. The SWECO unit had 142 screen failures in the same time
period.
-------
c. This particular unit required electrical power supply for the drive
motor and the backwash pump motors and controllers and potable water
as make-up for backwash.
d. The Rex unit minimum floor area is 38 square meters or 0.54 square
meters per million liters per day hydraulic capacity.
e. The Rex unit requires about 3/4 meter (2.4 feet) static headless to
operate on storm-caused overflows.
f. Approximately 225 liters per minute (60 gpm) screenings concentrate
was generated when the automatic backwash system was operating
properly.
g. It was not possible to operate this particular unit automatically at
the Fort Wayne Stormwater Facility because of hydraulic loading
problems.
14. The centrifugal concentration method of screening was observed to infer
the following conclusions:
®
a. It was not possible to operate the SWECO CWC unit automatically at
the Fort Wayne Stormwater Facility because the backwash sensing
circuits called for continuous backwashes, drastically reducing
screen through-put.
b. Manual backwash procedures were necessary to keep the units ready
for service at any time.
c. Screen failures from punctures adversely affected performance of the
CWC units because a single screen failure reduced screening unit
effectiveness dramatically.
d. At least 1.8 to 3.7 meters (6 to 12 feet) static hydraulic head is
required to operate these units.
e. Potable water was required at Fort Wayne to insure that there was a
clean source of make-up water available.
f. These units require electrical power, potable water, pneumatic air,
and during winter months environmental control (heating) to be oper-
ational.
g. Four, of these units required 48 square meters or 0.35 square meter
per million liters per day hydraulic capacity.
h. These units result in screenings concentrate from 10 to 20 percent
of raw flow to the unit.
-------
15. There x5 were
observed with the lower pond loading rates (i.e., less than 112 kg
BODs/hectare/day or 100 pounds BODs/acre/day). In contrast to
this, more stable, higher efficiencies of removal for 6005 were
observed with the higher pond loading rates (more than 224 kg
-------
BODg/hectare/day or 200 pounds BOD^/acre/day). Average removal
efficiency for the pond effluent BOD5 in the post-construction
period was 75 percent (a = ~t 4.8 percent).
d. The addition of storm—caused overflows resulted in shorter hydraulic
detention time and higher organic loadings which intended to
inhibit algal growth in the warm weather months.
e'. Short method chemical oxygen demand data did not follow the same
trend of values as BODc. No data regarding reflux method COD
are available.
f. Terminal pond effluent loadings for suspended solids were shown to
be a very heavily damped response to the pond influent loadings.
g. Pond effluent suspended solids concentrations correspond closely
to comparable BODc concentrations.
h. Average mean monthly removal efficiency for suspended solids was
74.6 percent (a = _ 14 percent).
20. The comparative annual operations and maintenance costs per million
gallons of treated flow for the three screening methods were:
a. Bauer Hydrasieve®, $25.25/MG
b. Rex Rotary Drum Screen, $50.00/MG
c. SWECO CWR® Centrifugal Wastewater Concentrator, $59.68/MG
-------
SECTION 3
RECOMMENDATIONS
1. Additional research into the nature of pollutant removals in stormwater
detention ponds with short holding times is needed. Specifically, more
research is needed to develop sound engineering design loading parameter
ranges for single purpose and dual-use terminal ponds intended to treat
stormwater and storm-caused combined sewer overflows.
2. Some provisions for sludge removal should be built in all piping, con-
duits, wetwells, channels and ponds designed to handle combined sewer
overflows to remove the build-up of sediments from these wastewaters.
3. Consideration of flow control to each individual screening unit is essen-
tial to maintaining maximum treatment efficiencies. Specifically, the
Bauer Hydrasieve® should have some way to differentially distribute the
flows to be treated to each module in order to optimize the hydraulic
flux rate to each screen module.
4. Some provision for direct collection and disposal of the screenings from
the Bauer Hydrasieve® should be considered. These screenings can be
directly collected semidry from the screen surface for ultimate disposal.
5. Some thought on the problems associated with the routine preventive
maintenance should be given for all components used in a combined sewer
treatment facility. Automatic oilers for all electrical motors and
bearings are a must. Intermittent service on storm-caused combined
sewer overflows requires sturdy, easily cleaned, checked, lubricated and
maintained components.
6. Some type of prescreening ahead of fine screens is necessary to prevent
premature failure of the screen cloths from stormwater-borne flotsam and
debris. The Bauer Hydrasieve® performed very well in this role.
7. Automatic controls for screens treating combined sewer overflows and
installed in remote facilities should be monitored using hard-wired re-
mote devices or real-time telemetry to a central operator control sta-
tion continuously manned year-round. Periodic equipment checks by
service and control operating'personnel are necessary to maintain maxi-
mum treatment efficiency of the units.
-------
8. Experimentation with heavier gage screens or tougher, more durable mate-
rials in the upper cage screens for the rotary centrifugal screens is
recommended to improve the screening efficiency and screen life of these
units.
-------
SECTION 4
DESCRIPTION OF THE PROBLEM
COMBINED SEWER OVERFLOW TREATMENT REQUIREMENTS AT FORT WAYNE
The City of Fort Wayne is located at the junction of the Maumee, St. Joseph
and St. Marys Rivers in northeastern Indiana. The Maumee River flows north-
easterly from Fort Wayne and eventually enters Lake Erie at Toledo, Ohio. A
report by the U.S. Federal Water Quality Administration in 1968, "The Lake
Erie Report1'^, identified the extensive combined sewer overflow system in the
communities on the upper reaches of the Maumee as a major source of highly
polluted raw wastewater. A major part of the lake pollution cited in this
report was attributed to the Maumee River loadings.
The City of Fort Wayne authorized a study of the entire collection and treat-
ment system of the City in 1969. The thrust of this study was to determine
the pollution sources, quantify them, and outline a Master Plan for correc-
tion of the major sources of pollutants.3 Within the scope of this city-wide
Master Plan was the investigation of the entire existing sanitary and storm-
water collection system, including combined sewer overflows.
As a result of the Master Plan, the urban drainage areas of some 12,000 acres
were found to have 150 separate trunk sewer systems. Approximately three-
fourths were found to be combined sanitary and storm sewers. These combined
service trunk sewers were interlocked throughout with more than thirty diver-
sions and regulators, resulting in more than twenty separate overflow points.
This Master Plan identified the largest sources of discharges by capacity and
pollution potential. A three-phase plan for combined sewer overflow deten-
tion of stormwater-induced hydraulic surges was recommended. A key feature
of the plan was the elimination of some combined sewer overflow points by
construction of a centrally-located stormwater treatment facility (identified
as Facility "A" in the Master Plan)3 near the Water Pollution Control Plant.
This facility was to be designed to handle all of the excess hydraulic flow
received at the Water Pollution Control Plant, apply the equivalent of
primary treatment to remove the flotables and settleables, and disinfect the
entire flow to reduce the pathogenic organisms before discharge to the
Maumee.
10
-------
SELECTION OF SITE
City Officials had selected a site across the river from the Water Pollution
Control Plant to construct a dual-use two-day terminal pond and chlorination
facility for disinfection and polishing of the Water Pollution Control Plant
flow (completed in 1971). This site was readily available for construction
of the proposed Facility "A" to be dedicated exclusively for treatment of
combined sewer stormwater overflows. As originally proposed in the Master
Plan, a large pumping station for combined sewer overflows was to be built
discharging raw stormwater-induced overflows at sufficient hydraulic static
lift to flow into conventional primary sedimentation basins.
However, the magnitude of the work and construction costs of full scale
installations elsewhere encouraged community interest in conducting a full
scale demonstration grant program at Fort Wayne using screening as an alter-
nate to primary sedimentation. An application for a Research, Demonstration
and Development Grant for evaluation of this alternate was submitted to the
U.S. Environmental Protection Agency in October, 1970. After several addi-
tions and revisions suggested in response to developments elsewhere, a pro-
gram plan of study was approved using several methods of fine mesh screening,
replacing the original concept of conventional primary sedimentation alto-
gether.
The City of Fort Wayne Water Pollution Control Plant on the City's east side
at Dwenger Avenue was scheduled for plant expansion to start in-1975 (shown
under construction in this recent aerial photograph, Figure 1). The 36-acre
body of water shown in the upper left-hand corner of Figure 1 is the dual-use
terminal pond described above. The new construction in the photograph shown
on the north side of the river is to become new Terminal Pond No. 2, another
dual-use pond.
Figure 2 is an aerial photograph of the Stormwater Treatment Facility showing
the entire terminal pond, the Pumping and Screening Building, the chlorine
contact channel, the pump wetwell overflow channel, and the terminal pond
final effluent channel.
DESCRIPTION OF THE COMBINED SEWER SERVICE AREA
Figure 3 is a map of the location of the Water Pollution Control Plant, the
Stormwater Facility, and other important service area landmarks and locations.
The area served by the stormwater facilities in this program included approx-
imately eight miles of combined trunk sewers in a drainage area of 995 acres
on the south side of the river. However, raw sewage flow could be diverted
from the Water Pollution Control Plant influent headworks at any time by
simply allowing the head to build up in the plant interceptors until the
overflow elevation was reached. See Figure 4 for a section drawing of this
diversion structure in the vicinity of the headworks of the Water Pollution
Control Plant.
11
-------
Figure 1 - Aerial Photograph of
Water Pollution Control Plant
City of Fort Wayne, Indiana
Figure 2 - Aerial Photograph of Storrawater
Treatment Facility
City of Fort Wayne, Indiana
12
-------
.(/
LEGEND
COMBINED SEWER SERVICE AREA
COMBINED SEWER WASTEWATER
TREATMENT ttCUTCS
A STREAM SAMPLING POINTS
O RAW OAOE LOCATIONS
O COMBINED SEWER OVERFLOW
MONITORMO POINTS
« — CITY LIMITS
FIGURE 3 - SERVICE AREA LOCATION PLAN
CITY OF FORT WAYNE, INDIANA
13
-------
EXUTIN* 7-0
MMULLIPTICAL
INFLUENT (EWER
TO WATER POLtUriON
CONTROL PLAIilT
PLANT BYPASS CHAMBER
PLAN
ELEV. 7SZ.07
..-W..1-. .-'..' 'V v>.». .-l^j."-v'..V-.i-".-"l -1'i.uV.'.,-•','.- '.ls'..-'.",i'.v.'^
-------
As the raw wastewater builds up in the plant bypass chamber, the diverted
raw wastewater flows- by gravity head to a river crossing structure connected
to twin 96-inch diameter conduits feeding the raw sewage wetwell at the
Stormwater Pumping Facility. Figure 5 shows the location of the Water
Pollution Control Plant, the connecting sewers, the WPCP chlorination facili-
ties, Terminal Pond No. 1, and the Stormwater Treatment Facilities. From the
84-inch plant bypass chamber, the raw wastewater flows to the twin 96-inch
river crossing structure. Also entering this river crossing structure is a
108-inch Stormwater relief sewer constructed as a part of this project carry-
ing the combined sewer overflows from the Glasgow Street interceptor sewer.
From the twin 96-inch river crossing structure oil the south side of the river,
the wastewater flows to the wetwell of the Stormwater pumping station.
DESCRIPTION OF THE STORMWATER PUMPING AND SCREENING FACILITY
The wetwell at the Stormwater Treatment Facility is sized at 550,000 gallons
(2,082 cu. m.) capacity. A mechanically cleaned trash rake bar rack is in-
stalled at the inlet end of the wetwell. A permanent metal building houses
the raw Stormwater pumps and the screening equipment.
Figure 6 shows the wetwell end of the Stormwater Facility Pumping and Screen-
ing Building. The trash rake mechanism is shown in the upper position in
this,photograph.
Two Johnston Pump Company mixed flow vertical turbine raw Stormwater pumps
provide the static lift needed to bring the wetwell contents up to the treat-
ment level. One pump, a Johnston Model 30 PS delivers 35,000 gpm at 55 feet
TDH, or 50 MGD [2.191 cubic meters per second (cumec)]. The Johnston Model
24 PS pump delivers 17,500 gpm at 55 feet TDH, or 25 MGD (1.096 cumec). Each
pump has its own separate suction column and discharge line with flap valve.
Both discharge to a common 14-foot diameter cylindrical head tank which pro-
vides the constant head necessary to operate the screening equipment.
Figure 7 is a photograph of the Stormwater pump discharge constant head tank
and flow distribution device. Not shown in this photograph is the six-foot
circular overflow weir inside the head tank which carries the raw Stormwater
overflow to the screening £oom effluent channel.
Figure 8 is a treatment flow routing schematic diagram showing the hydraulic
flow routing in the facility.
There are three types of screening equipment installed in the screening
equipment room. A side hill screen method, the Bauer HydrasieveB' , is in-
stalled in a pair of six-unit modules. The combined hydraulic capacity of
these modules is normally 3,500 to 10,400 gallons per minute (0.219 cumec to
0.651 cumec) with a peak loading of 13,825 gpm (0.865 cumec).
Figure 9 is a photograph of the east bank of the Bauer units taken shortly
before the facility became fully operational as a Stormwater screening
facility. Figure 10 is a photograph of the same screening devices showing
15
-------
TO ACCESS ROM)
TERMINAL POND NO. I
NORMAL WATER LEVEL 745.5 FT. M.S.L.
STORM LEVEL 748.5 FT.M.S.L.
STORMWTER
TREATMENT
FACSLrtY
WATER POLLUTION
CONTROL PLANT
WPCP
RIVER
CROSSING
STRUCTURE
FIGURE 5 - WATER POLLUTION CONTROL PLANT
STORMWATER TREATMENT FACILITIES
16
-------
Figure 6 - Stormwater Treatment Facility
City of Fort Wayne, Indiana
Figure 7 - Constant Head Tank Raw Stormwater
Distribution Device
17
-------
FIGURE 8 - STOKMWATER TREATMENT FACILITY
FLOW ROUTING SCHEMATIC DIAGRAM
18
-------
Figure 9 - Bauer Hydrasieve^
as Originally Installed
Figure 10 - Bauer Hydrasieve^ as
Modified with Splash Guards
East Bauer
19
-------
three of the units on one side of the east bank under full hydraulic loading.
This photograph also shows the field modification splash guards to the units
installed to permit greater flow distribution capability.
These screens operate on the hydraulic surface attachment principle of the
Coanda effect. The solids-laden fluid enters the headbox between the two
banks of modules and overflows a weir above the screens. The fluid is accel-
erated down the screen by gravity and the water tends to "slip through" the
screen openings set at 0.060 inch (1.52 millimeters). The screened effluent
drops into a. common channel in each module while the solids-laden screenings
fall into channels along the length of each module where it flows to a common
concentrate channel.
Another screening method evaluated was the Rex Rotary Drum Screen,, The
Rex screen operates on the principle that screening is accomplished by
forcing solids-laden wastewater through a rotating drum screen. The solids
first are deposited by sieving then by straining as the filter mat builds up.
The pressure gradient driving the fluid across the screen comes from the
differential head developed by the raw waste influent into a rectangular head-
box structure communicating into the interior of the drum. The flow moves
from here to the interior of the drum through an open end and passes through
the screen media on the drum circumference. After passing through the media,
the screened effluent overflows an effluent weir to the effluent channel.
Figures 11 and 12 show the Rex drum screen under operational conditions. The
rate of rotation and the hydraulic loading are variable for this screening
method.
Screenings collected by the media are carried up and out of the flow as the
drum rotates. When screenings build up on the screen media sufficiently to
cause the head differential through the unit to exceed 24 inches (61 centi-
meters) , a backwash spray header system washes the screenings from the media.
The screenings drop into a collection trough inside the drum and flow by
gravity to the common concentrated screenings channel.
The Rex screen installed was a 12-foot (3.65 meters) effective diameter drum
screen mounted horizontally within a reinforced concrete chamber. The flow
enters the screen inlet chamber in a direction parallel to the drum axis.
The drum length is 12 feet (3.65 meters). There are 24 separate rectangular
panels I1-6" by 12'-0" (0.45 m by 3.66 m) mounted on the circumference of the
screen for a total screen area of 432 square feet (40.1 square meters). The
screen material is No. 316 stainless steel tensile bolt cloth mesh No. 100
capable of screening particles down to a minimum size of 75 micrometers.
The screen rotating speed is variable over a range of 2.8 to 11.2 rpm. The
hydraulic loading rate is designed to be a maximum of 18.75 MGD (0.906 cumec),
or a hydraulic loading rate of 60 gpm per square foot per minute (21.1 liters
per minute per square meter'per minute), based on a submergence ratio of
0.54.
20
-------
Figure 11 - Rex Rotary Drum Screen
Figure 12 - Rex Rotary Drum Screen
Automatic Backwash
Collection Trough
21
-------
Eight SWECO CWC ® Centrifugal Wastewater Concentrators divided into two banks
of four units each are installed overhanging an open effluent channel that
collects the screened effluent from all the screening units plus the excess
of the head tank overflow weir. These units are each 60-inch (152 cm) SWECO
CWC® units. A rotating screen cage drive within a rigid circular outer
steel shell forms the hasis for the screening. Screening is accomplished
mostly by straining although some buildup of solids removed can be noticed
by excessive concentrate flows. During the operation of the screening units,
the screens are continuously "washed down" by the action of the directed
influent flow. This "washing" action produces the "concentrated screenings"
flow stream which is collected and conveyed to the common screenings.
Figure 13 is a photograph of one of the SWECO units. The utility connections
for hot and cold backwash rinse water are visible in the photograph. Figure
14 is a photograph of the manner of installation of the eight SWECO screens
along the facility screened effluent channel that are capable of operation in
tandem with the Bauer Hydrasieve ®.
The centrifugal wastewater concentrator has a rotating collar containing 36
individual screen panels each of approximately 2 square feet (0.19 square
meter) of No. 165 mesh (105 micrometer openings) screens. Each CWC unit is
nominally rated at 1,500 gpm (5,700 1pm) although they have been successfully
operated at up to 4,000 gpm (15,000 1pm) each. The unit is limited more by
the quantity of concentrated screenings generated and the structural limita-
tions of the screen material than other factors.
22
-------
Figure 13 - SWEGO CWC-60^Centrifugal
Wastewater Concentrator
Figure 14 - North Bank of SWECO CWC-60
Screening Units
23
-------
SECTION 5
EXPERIMENTAL PROGRAM
PRE-CONSTRUCTION EVALUATION PHASE
River Sampling Locations
To evaluate the effects of combined sewer overflow wastewater on the receiv-
ing stream, water quality and flow data of the Maumee River were monitored
from September 1971 to December 1972. This water quality monitoring program
was undertaken to evaluate the impact of untreated combined sewer overflows
on the river. More than fifty separate storm events were monitored for the
individual pollution loading characteristics of each event.
Three Maumee River flow measurement and sampling points were mbnitored in the
pre-construction 1971 - 1972 sampling program.
1. Maumee River at Anthony Boulevard (one-quarter mile, 0.4 km
upstream);
2. Maumee River at U.S. 30 Bypass (1-1/2 miles, 2.4 km downstream);
3. Maumee River at the New Haven Bridge (4-1/2 miles, 7.2 km down-
stream) .
A map showing these monitoring points is shown in Figure 3.
The stormwater events in the program were assigned sequential identification
numbers starting with 001 and running to 058 in the pre-construction sampling
program in 1971 - 1972. After completion of the construction program, a
similar numbering system started with EVENT 101 and ended with EVENT 216.
All intervening events between event 059 and 100 were not considered in the
study program.
A primary evaluation objective resulted from the monitoring of the impact of
storm-caused overflow on the river. To evaluate the rainfall intensity
factor on the overflows, rainfall data from five stations were monitored in
both the pre-test and post-test situation using recording rain gages. These
rain gage stations were identified with the following code locations.
24
-------
Station 1 - Brown Street Lift Station - Brown Street at Maumee River
on the City's west side at a distance of 3 miles (4.8 km)
west of the Water Pollution Control Plant.
Station 2 - City Utilities Park - Baals Drive at St. Joseph River
approximately 2-1/4 miles (3.6 km) north of the Water
Pollution Control Plant.
Station 3 - Deleted.
WPCP - Water Pollution Control Plant.
Weather - U.S. Weather Service information from Baer Field, Fort
Bureau Wayne. (This data is usually obtained from the U.S.
Weather Service monthly weather summary.)
The Maumee River elevation at the Water Pollution Control Plant is obtained
from the New Haven stream gaging station, applied back upstream. A reading
of this gaging station was taken at the start of the event, and daily as part
of the routine data-gathering at the plant.
The Morton Street Lift Station was selected as a major source of bypassing
stormwater overflows due to its large drainage area of mostly residential
neighborhoods. Elapsed hour meters were installed on the existing stormwater
pumps. As the head increased in the Morton Street Lift Station wetwell, the
stormwater pumps automatically started and actuated the running time clocks.
The amount of stormwater bypassed from the Morton Street station was esti-
mated from the total elapsed time and the average pumping rate derived from
the wetwell depth recorder.
Another major source of the storm-caused overflow at the Water Pollution
Control Plant was selected for monitoring in the pre-construction phase and
later for treatment in the construction program. Installed in the 84-inch
plant bypass structure was a water level recorder in which the depth of the
stormwater in the chamber triggered a depth recording chart in the Water
Pollution Control Plant control room. The depth of water in this structure
provided an estimate of the total raw stormwater being diverted to the river
in the pre-construction phase. Later, this same monitoring point was used
to indicate the start of a stormwater treatment event after the screening
facility became operational.
Pre-Cbnstruction Sampling Program
The objectives of this program called for determining the time-dependent
characteristics of the stormwater overflow. This was accomplished by taking
grab samples and flow measurements at predetermined time intervals. The
sampling program was intended to start with the initial storm-caused overflow.
This condition was termed "first flush", and the constraint upon the sampling
was to take a grab sample within the first 15 minutes after the overflow
began (determined from the 84-inch bypass depth chart at the Water Pollution
Control Plant).
25
-------
Sampling was usually continued to the point where the discharge to the river
ceased altogether and the pollutional loading of the river dropped to normal
levels. As a practical matter, on several events, it might have taken days
for the river discharge to return to the normal dry weather flow condition.
The decision to terminate an event sampling program was made when, at least
two sets of river samples had been obtained with one of them taken within the
interval of overflow bypassing. Sampling on occasion continued beyond the
end of the bypass condition in response to some evidence that the peak pollu-
tion loading was still building up. Normally, however, the end of the over-
flow in the pre-construction phase signaled the end of the event sampling.
River samples were taken at the three points every hour or so after the start
of the event. The method of sampling was to obtain a grab sample of the main
stream flow. The samples were field-analyzed for pH, temperature, fecal
coliform, and dissolved oxygen. Other analyses were run in the plant labora-
tory on turbidity, chemical oxygen demand, biochemical oxygen dema.nd, sus-
pended solids, volatile solids, ammonia nitrogen, and total phosphorus.
Staff gage observations were made at the same time as the physical and chemi-
cal samples taken at the Anthony Boulevard Bridge where the Weather Bureau
maintains a stream gaging station.
The duration of river sampling in the pre-construction program normally lasted
as long as the duration of the observed overflows at the Water Pollution
Control Plant, with one or more setp of samples taken thereafter.
Routine plant composite sampling on the basis of daily analyses for regula-
tory agency reports also continued [throughout the pre-constructiort phase.
The data for the terminal pond effluent represent the best estimate of plant
pollutional loadings to the river over this period of time.
Table 1 is provided to summarize the sampling locations used and the analyses
required for the pre-construction evaluation phase.
POST-CONSTRUCTION EVALUATION PHASE
Screening Equipment Locations and Elow Measurements
In the evaluation of the screening equipment as stormwater treatment devices
in the post-construction phase, the decision on when to start the stormwater
pumps at the test facility resulted from the indication on the 84-inch bypass
level indicator. When the water level in the 84-inch bypass structure was
noted as reaching the point where a depth of water showed on the bubbler
depth recorder in the control room, the Water Pollution Control PJLant oper-
ator notified the project director. At the point of incipient overflow of
the 84-inch bypass, the water level indicator at the Stormwater Facility wet-
well would read approximately 15 feet.
Specially-trained operators were alerted and directed to their work stations
for the duration of the stormwater treatment event. One such person, the
river sampler, was responsible for driving the road circuit taking river
samples at the Anthony Boulevard Bridge, the U.S. 30 Bypass Bridge and the
New Haven Bridge. The Stormwater Facility operator was stationed at the
26
-------
Table 1. PRE-CONSTRUCTION EVALUATION PHASE SAMPLING PLAN
August, 1971 to November, 1972
X
X
X
••a
X X X X X
X X X X X
X X
a
1
XXX
u
o
en
X X
X
x
O-jJ X X X X X X X
iSl
-I H
331
JS
•g
ftl
(U
XI
«
0)
4J
4J
cn
/-s
• I-l
O O
U O
iH
0) *-N
T> iH
*rt -**.
rH rH
0 g
CO N--
.
B o
H ^
XXXXX X X X
X
XXXXX X X X
{gl X X X X X X X X
fi
g
•H
*J
«
O
3
on
c
•H
rH
0.
1
en
u la)
C 0) • CN a rH
•uo) Mcn-^Bs-H
OJtH W CO TJW&IJ^S;
3U-IT3 01 • >& fH4J CSJ
•HM-lfi ffl »J .VlrH3V< OS^^^S
U-J [£] O CX >, 4J *4-i> QJ > W 3 > | OJ
w >> d ea d -PC -HXtH •Hcdo -H -* M
w »H §
B O 04-) 'H *H 4JIW SC3 "^- S O 'rl S RJ O
•H o Bw iw MW 3<|,H 3
-------
stormwater pump control room to operate the stormwater pumps and screening
devices, communicate with the Water Pollution Control Plant, and direct the
sample-gathering at the facility. Normally, a third person was assigned to
the facility to a,ssist the operator in taking samples at the various loca-
tions in the sampling plan.
The project director usually supervised the Water Pollution Control Plant
operators in flow-routing and raw sewage pumping at the plant. This activity
was usually necessary because the plant personnel had been using the volume
of the plant influent sewers for raw wastewater storage and flow equalization
for years prior to this work. Careful flow-routing in the plant and sequen-
tial raw sewage pumping were necessary to avoid process problems with the
plant when stormwater-induced surges in hydraulic flow occurred.
The decision to begin an event rested with the project director exclusively.
Sometimes, the storage volume of the plant interceptors filled the river
crossing conduits enough to show on the 84-inch bypass bubbler cha.rt even
when there was no rainfall preceding the diversion. Until the wetwell level
at the Stormwater Facility reached 26 feet on the depth indicator there, it
was impossible to bypass raw sewage to the river without some kind, of treat-
ment .
The sequence of events preceding a pumping cycle typically involved manning
the facility as described above. The operator was instructed to start one
of the stormwater pumps when the wejtwell level reached 20 feet on the bubbler
level indicator. The decision of whether to use the 24 PS (17,500 gpm,
1.096 cumec) or 30 PS (35,000 gpm, 2.101 cumec) pumps usually was made on the
basis of equalizing the time of operation for each pump.
Once the pumping of the wetwell contents began, the project director decided
whether to sample the flow or to merely "pump down" to clear the wetwell and
river crossing conduits of the raw wastewater. In the post-construction
sampling period commencing in January, 1975 and ending in February, 1976,
there were 116 bona fide events triggered by the depth indicator in the 84-
inch bypass. There were samples taken in 53 events, not all of which were
stormwater-induced. Some of these events were initiated by diverting plant
raw sewage to the Stormwater Facility by shutting down one or more raw sewage
pumps and allowing the raw wastewater to fill up the river crossing to the
stormwater station. Some other evejnts were the result of recirculating
Terminal Pond No. 1 contents. Most events were the result of stormwater-
induced flows in the combined sewer collector areas, however.
The raw wastewater pumped by the stormwater pumps discharged into a 14'-0"
(4.27 meter) diameter head tank (32,500 gallons, 123,000 liters capacity)
before being distributed to the treatment units.
The head tank acted as a surge tank for the stormwater pumps, as a constant
head flow control device for the screening equipment, and as a completely
mixed sampling vessel for obtaining the raw wastewater sample aliquots
needed to perform the evaluations.
28
-------
Figure 8 shows a schematic diagram'of the flow routing to the different
screening devices as well as the overflow points necessary to protect the
screening room equipment from excessive flow surges and capacity. The head
tank has a 6'-0" (1.83 meter) diameter overflow weir discharging excess raw
wastes pumped to the head tank directly into the screened effluent channel.
By means of a water surface level indicator on the side of the head tank,
sampling personnel were able to determine the depth of wastewater in the head
tank at any time. This observation provided,the measurements necessary to
determine the quantity of raw wastewater pumped to the head tank and bypassed
over the 72-inch circular overflow weir.
The facility was designed to evaluate the relative treatment efficiency of
three different types of screening devices on stormwater-induced combined
sewer overflows. The flow split to these screening devices was determined
by dividing the total pumping capacity into four segments. The intent of
this arrangement was to apportion the available hydraulic flow equally to
each of the screening devices installed. The channels for this purpose were ,
designed in such a manner as to provide ample hydraulic head to run the Bauer
Hydrasieve® fixed screens in tandem with the Rex Rotary Drum Screen and the
SWECO Centrifugal Wastewater Concentrators. In other words, a dual-use
objective of using the Bauer screens as a pretreatment device for both the
Rex and SWECO screens was built into the flow diversion conduits.
Because determination of treatment efficiency was a prime objective from the
outset, accurate flow measurements for each screening device were essential.
The project sampling and flow measurement protocol established flow measure-
ment weirs, staff gages, and orifices that were to be calibrated with primary
flow measurement devices, and used thereafter for making field observations
that were treated as secondary flow measurements. A calibration deviation
of +10 percent from the actual measured flow was considered adequate to meet
the treatment evaluation program objectives within the evaluation phase bud-
get for the project.
Referring to Figure 8 again, there are five discharge points out of the head
tank. One is the 72-inch overflow weir discussed above. On the south side
of the tank, a 24-inch diameter line feeds a bank of four SWECO CWC ® Cen-
trifugal Wastewater Concentrators. A flow measurement of the head tank dis-
charge .feeding this bank was made using a 16-inch diameter sharp-edged ori-
fice installed at the flange in the line. Arrangements were made to equally
divide the flows to the individual units by balancing the influent flows
using a pressure drop measurement across the 12-inch inlet elbow to each unit.
The influent gate valve slide position was adjusted in the field to equalize
this pressure drop. Thereafter, measurements of flow to this bank of units
was made by observing a mercury-filled U-tube manometer attached to center-
line pressure taps upstream and downstream (at the vena contracta point)
from the orifice. Computations of the hydraulic discharge corresponding to
these measurements were done using the adjusted orifice equation, C = 0.60.
Another bank of SWECO CWC® units was installed on the other side of the
screened effluent channel. This bank had the capability of being operated as
primary treatment devices or secondary treatment behind the Bauer unit as a
29
-------
pretreatment device. Except for this feature, the measurements of flow for
this installation were identical to the south bank.
The Bauer Hydrasieve® units were arranged in two banks of 72-inch wide
screens, with six such screens on each side, east and west. There*, was no way
to vary the flow to either side except to throttle the influent gate valve
and thereby restrict the flow to the entire unit. However, it was observed
in the field that the velocity of the discharge from each bank was equal at
three different flow rates. Therefore, a rectangular weir was installed in
the discharge section of the west bank unit with a staff gage located approx-
imately four feet (1.2 meters) behind the weir. The operator was instructed
to take a staff gage reading at intervals set in the sampling protocol on the
west bank Bauer unit at the same time he was taking a reading on the head
tank water level. The assumption that there was no difference between the
(measured) west bank units and the (assumed equal) east bank units was vali-
dated by field measurements of discharges in the effluent channels of both
tanks.
The Rex flow measurements used the built-in 132-inch (3.35 meters) rectangu-
lar weir at the effluent side of the screening equipment. A staff gage was
placed behind the weir and a float sensor head recorder in a stilling well
was also installed.
To protect the drum from damage resulting from hydraulic overloading of the
unit, a 36-inch (0.9 meter) rectangular weir was installed in the headbox for
the unit to allow raw wastes to fill up the interior of the drum to the over-
flow elevation. A staff gage and head recorder in a stilling well were in-
stalled also. Flow control to this unit was possible by throttling the in-
fluent gate valve coming from the head tank.
Measurement of the screenings discharge of the treatment units was also
necessary. A series of rectangular channels in the floor of the screening
equipment room of the facility were utilized by placing standard smppressed
rectangular weirs at the ends of the SWECO collection channels with staff
gages. Operators were instructed to measure down from a convenient reference
point (the floor channel grating) to the water surface. The investigators
intended these observations to be used as screenings flow measurements by
converting these observations to staff gage datum. Unfortunately,, most of
the stormwater-induced events resulted in submerged weirs due to hydraulic
back head from the surcharged effluent sewer line. Therefore, few flow
measurements for the SWECO screening concentrate were considered valid enough
to be used in evaluation. Rather than using measurements that were likely to
grossly over-estimate these flows, the flow measurement protocol was altered
to use assumed screenings flows in place of these field observations.
The flow of screenings from the Rex drum screen was conveniently measured by
a weir box with a 60-degree V-notch weir installed. Operating personnel were
instructed how to measure the head :over such a device and the resxilts record-
ed on a field data sheet. These head measurements were converted to dis-
charge measurements by using the standard 60-degree V-notch weir tables.
30
-------
A 108-inch (2.74-meter) diameter circular weir was designed to operate as a
fully developed circular weir for flow measurement checks using a staff gage
and/or head recorder at the scum baffle a few feet upstream from the stage
monitoring point. This stage level measurement technique was field-cali-
brated by a current meter calibration procedure in April, 1975 using a cable-
suspended Price-type current meter. This calibration was necessary to pro-
vide an independent flow measurement as a base line data point to permit
accurate estimates of flow measurement error.
Two other staff gage measurement points were set in the screened effluent
channel. One point was set at the discharge end of the screening room open
channel exiting the building. Another point was located 550 feet (167.6
meters) downstream from the facility building. These two points were field-
calibrated at the same time as the circular weir at the end of the channel.
Raw sewage pumping"rates were determined from the pump head discharge curves.
The pump operating point was calculated from observed levels of the Stormwater
Facility wetwell indicator. The amounts calculated from this source were
compared with the sum of the amounts measured, with due consideration to the
problem of relative error. The chlorine contact channel calculated discharge
from the observed head gage readings was used as a last-resort check of these
quantities to help sort out discrepancies in the flow measurements. In this
way, an estimate of maximum probable error was made for those events where
the data show anomolous trends of behavior.
Other flow measurements used in this evaluation program came from routine
flow observations and records of the plant. Specifically, the terminal pond
influent coming across from the Water Pollution Control Plant was taken from
the raw sewage flow meter at the plant. The terminal pond effluent flow was
derived from the measured raw sewage flow at the WPCP and the measured storm-
water treatment flow at the chlorine contact channel, using the 8'-0" (2.44
meter) rectangular sluice gate opening as a conventional broad-crested sup-
pressed rectangular weir. Knowing the pond depth, and therefore, the head
over the "weir", a pond discharge was calculated.
Post-Construction Sampling Program
The parameters intended to be measured in the post-construction program were
selected from the experience of the pre-construction evaluation phase samp-
ling plan (See Table 1, preceding). Table 2 represents a summary of the post-
construction evaluation schedule of sampling and analysis. Note that there
were additional parameters added in the post-construction phase evaluation;
namely, chlorine demand, residual chlorine, odor, total organic nitrogen, as
well as ten additional sampling locations.
Evaluation of the screens was based on the difference between the raw waste-
water loadings and the treated wastewater loadings. Multiple samples were
taken at predetermined time intervals because the objectives of this program
called for following a quantity of stormwater overflow through the treatment
processes and eventually by discharge to the stream from the terminal pond
effluent. There, grab samples were mostly used in this phase.
31
-------
Table 2. POST-CONSTRUCTION EVALUATION PHASE SAMPLING PLAN
January, 1975 to February, 1976
I 60 AJ
I «3 *W
» 0-S
2-.
o «
S§S
(S O.t-1 __
IH m o s
O 3 W -—'
> CO
O.r-1 i
-------
Table 2. POST-CONSTRUCTION EVALUATION PHASE SAMPLING PLAN
January, 1975 to February, 1976 (Continued)
O P*t
> 0
(H 0)
O M
C « '
-H 3 i
M *a -
O i-l
H -3 « *
«rl (3 tJ »
U QJ t-1 *
« a i-r
IIJI
nj O
u y-i _
(U •>-( r-t
x x
X X
X X
X X
i-
oc uva •Hiuajenui u
CiH 3 a)O>SM gj UCMCQ
- - -
H i-i M-I 3 Cam 3.
.d o <4H ,b a) o u-i cti
o en w o H fw w £ •
33
-------
Some of the points where aliquots were to be taken in the screening facility
represented more than one physical point of sample collection. For example,
there were eight SWECO Centrifugal Wastewater Concentrators (CWC) located in
the screening equipment room of the building. Each CWC contributed its own
effluent and screenings to a common channel designed to carry all the screen-
ings and effluent away as they were discharged. Other treatment units also
discharged into these channels, and the individual sampling of these incre-
mental discharges would have presented an impossible analysis burden on the
plant laboratory personnel. Therefore, for specified sampling points, the
operators were instructed to take equal portion aliquots from each operating
unit and composite these as the "grab" sample of the location required in the
sampling plan.
The single raw sewage sampling point was located at a sampling port at the
head tank. A sample was drawn from this sample port into a clean sample
container. Field measurements of pH, temperature, D.O., and staff gage water
levels were made on the spot and recorded. Grab samples were taken according
to the Schedule of Laboratory Analyses worked out in the pre-construction
effort. Analyses run on these grab samples were fecal coliform, turbidity,
GOD, suspended solids and volatile suspended solids. A single composited
sample made up two or more equal-sized aliquots taken at the same time as
the grab samples were analyzed for BOD,., ammonia nitrogen, total phosphorus,
and pH.
Each of the screening equipment effluent samples and flow measurements was
analyzed separately, within the constraints of multiple sampling points out-
lined above. The following equipment was sampled for a single grab sample
per device for each sampling interval: Bauer East, Bauer West, Rex SWECO
South, and SWECO North. Parameters measured in the field were pH, tempera-
ture, D'.O. and staff gage and/or manometer readings, as appropriate. Grab
samples were also taken one per interval for the same parameters required of
the head tank. Similarly, for the composited samples, a single sample per
interval was obtained.
A single screening equipment concentrate sample was taken at each of the
Bauer and Rex units. A concentrate channel for the SWECO's was sampled by
composited aliquots in an attempt to obtain a representative sample for the
time interval involved. Individual samples and flow measurements were also
analyzed for the same scope of analyses as the head tank.
The combined screenings were also measured to determine the impact: of the
screenings collected on the Water Ppllution Control Plant. These measure-
ments were conducted along the same lines as the head tank above.
Another sampling point was located at the overflow weir of the chlorine con-
tact channel. This point, designated as "Chlorinated Screened Effluent
Channel", was monitored for all of the analyses covered under the head tank
plus the addition of residual chlorine.
34
-------
Routine plant performance monitoring was already being done by collection of
24-hour flow-composited information of the type being sought at the Storm-
water Treatment Facility. These data were intended to be used to identify
the impact of the screening facility on the plant and to determine the rela-
tive effect on the river by rerouting some previously bypassed stormwater
overflow to the plant treatment lagoons. The plant data collected were flow-
composited manually from aliquots taken at several sampling locations includ-
ing the raw sewage, primary effluent (in excess of the secondary treatment
capacity of 32 MGD) , secondary effluent, chlorine contact tank effluent
(consisting of secondary effluent plus the diverted primary treated secondary
bypass flow), and the terminal pond effluent. Daily analyses were run on
these samples for all of the analyses required of the head tank plus chlorine
demand and residual chlorine as appropriate.
One of the original objectives of the program was to determine the individual
screening units * effectiveness in dealing with the anticipated heavily-
polluted "first flush" overflows that were observed in the pre-construction
evaluation program/ Toward this end, each of the screening devices' indi-
vidual sampling protocol called for taking the first four samples not more
than fifteen minutes apart for the first hour after the start of the event,
not more than thirty minutes apart for the next five hours, and not more than
sixty minutes apart thereafter to the end of the events. However, in the
post-construction evaluation period commencing in January, 1975 and running
until February 6, 1976, there were not even two or three events that even
remotely resembled the characteristics of a "first flush" phenomenon. The
decision was made in June, 1975 to dispense with the sampling and analysis
burden of attempting to catch the so-called "first flush", and the post-
construction sampling and analysis protocol was modified accordingly.
Screening Equipment Efficiency Evaluations
Each of the screening equipment suppliers was required, as a part of the bids,
to submit a statement regarding the equipments' anticipated performance
screening combined sewer overflow wastewater. From these statements, a
method of evaluation consistent with the operation of each type of equipment
was devised.
Each of the screening equipment suppliers claimed a degree of removal
efficiency on the pollutant parameters, biochemical oxygen demand and sus-
pended solids. In order to evaluate removal efficiencies on a consistent,
objective basis it was necessary to use these pollutant loadings (and others)
rather than concentrations as the measure of before treatment - after treat-
ment. This resulted in the computation of influent pollutant loadings for
each screening method separately. For this computation, the raw stormwater
pollutant concentration was used along with the time interval measured
instantaneous flow rate to the individual screening equipment. A measured
effluent flow rate (where possible) times the sampling results pollutant
concentrations provided the effluent loading parameter. A computer program
was written to perform the incremental computations necessary to determine
these loadings and the incremental results were tabulated and summed in the
35
-------
computer output table. The computer also tabulated the pollutant concentra-
tions by time of sampling and location, and provided a table of sampling and
analyses results.
Efficiency of removal was calculated from the computer output mass loadings
by performing the following computation for each equipment location or unit
process considered:
Efficiency
of Removal
(percent)
/Influent \
I Pollutant 1
\ Loading /
/Effluent \
( Pollutant]
\Loading /
/InfluentX
( Pollutant]
\ Loading /
x 100
The pollutants for which this computation was performed include BOD,., COD,
total organic nitrogen, ammonia nitrogen, total phosphorus, suspended solids,
volatile suspended solids, chlorine demand and residual chlorine. The samp-
ling locations where the mass loadings were computed included the Maumee
River at Anthony Boulevard, U.S. 30 Bypass, and New Haven, the Stormwater
Facility head tank, Bauer screened effluent east and west, Rex screened efflu-
ent, Rex screening concentrate, combined screening facility concentrate, and
the chlorinated screened effluent channel, the Water Pollution Control Plant
primary effluent (bypassing secondary treatment), secondary effluent, and
Waste Stabilization Pond No. 1 effluent.
Stormwater Treatment Facility Impact on Water Pollution Control Plant
This evaluation result was necessary because the existing 60 MGD Water
Pollution Control Plant (WPCP) was expected to treat the combined sewer
overflow Stormwater Facility concentrate at the conventional activated sludge
plant on the south side of the"Maumee River. In addition, the screened
effluent was added to the contents of Terminal Pond No. 1 which had been al-
ready receiving the wastewater from the WPCP since October, 1971.
The evaluation of the impact of the Stormwater Facility concentrate on the
WPCP was anticipated to be done quantitatively by measuring the flow rate and
relevant pollutant concentrations at the facility itself, and comparing with
the WPCP raw sewage results for the same time frame. However, with the prob-
lems measuring the concentrate flow rate cited earlier and the observed low
pollutant concentrations at the Stormwater Facility concentrate channels, the
decision to proceed with this evaluation on a semi-quantitative basis was
made. The loadings indicated for this part of the evaluation are based on
assumed flow rates and should be considered as semi-quantitative.
The effect of chlorinated combined sewer overflow screening effluent was an
integral part of the overall treatment capability evaluations anticipated for
this project. Plant operators had been taking daily composited samples of
the terminal pond effluent and running the pollutant analyses for BOD^, COD,
suspended solids, volatile suspended solids, total Kjehldahl nitrogen, total
phosphorus, and pH. Grab samples for the pond effluent D.O., fecal coliform,
and residual chlorine were also taken since the pond start-up.
36
-------
Trend analysis o'f the pond data available from these routine plant operating
and monitoring functions was expected to provide some of the results needed
to evaluate the effect of the screened, chlorinated stormwater overflows on
the pond. In addition to these data, observations by the plant operators of
algae growth in the pond, qualitative operational variations such as floating
scum and grease, and other special tests on the pond effluent were expected
to provide added performance data.
Dry-weather screening of the pond to evaluate these screens on algae-laden
pond contents was also a requirement of the program. An 18-inch pond recir-
culation line was constructed as a part of the facility construction grant
to permit this dual-use mode.
Determination of Shock and Total Pollutional Loads on the Receiving Stream
The effect of the raw, untreated combined sewer overflows on the receiving
stream was the major objective of the pre-construction evaluation phase. In
the construction phase, the installation of conduits and structures to divert
the 84-inch bypass overflow to the Stormwater Facility was intended to essen-
tially eliminate this source of overflows directly to the Maumee River. The
post-construction evaluation was intended to quantitatively measure the same
stream water quality parameters after completion of construction to determine
if any improvement could be measured. A trend analysis approach should also
indicate any measurable improvements in the stream water quality, particularly
in terms of maximum and/or minimum pollutant parameters.
In substantiation of the anticipated improvements in stream water quality,
the effects of the terminal pond effluent itself was required to assess the
changes measured. If the Stormwater Facility effluent could be shown to have
minimal effect on the pond, and the pond could be shown to be a minor source
of stream pollutants before and after the construction program, a measurable
relative improvement in stream water quality downstream from the location of
the former 84-inch bypass for combined sewer overflows would indicate a sig-
nificant improvement in stream water quality that could be attributed in
whole or in part to the Stormwater Facility.
Screening Methods Cost Analysis
A primary objective of this program was to determine the costs of screening
operations treating the previously bypassed raw combined sewer overflows. In
order to do this, all of the construction and operations costs of the facility
were monitored to provide costs data for the evaluation of the Stormwater
Facility. However, to distribute these costs to the different screening
methods used in the evaluation required some estimations of power usage for
the stormwater pumps, the facility lighting and heating equipment, and the
different screening methods.
Pumping horsepower requirements for the flows delivered for each event were
to be allocated to the different screening equipment by the use of a flow
proportionally constant (determined from the ratio of the discharge treated
to the total discharge pumped). To this power requirement to simply deliver
the raw stormwater overflow to the individual screening devices was added the
37
-------
computed horsepower directly required to operate the screening device. All
such horsepowers were used to calculate the kilowatt-hour power requirement
which was summed and used to compute the operating costs of this screening
method. Added to this cost were the costs of potable, water usage, chemical
consumption, etc.
Labor costs were considered to be more of a reflection of the effort expended
to gather samples and make observations for completing the R/D evaluation of
this program rather than an accurate rendering of actual operational manpower
requirements. Therefore, labor costs are not included in costs derived for
this program.
Once all the costs for an event were estimated, a cost per million gallons
treated was calculated. This result was to be tabulated as an evaluation
parameter just as treatment efficiency was to be used. This resulting dollars
per million gallons figure was also summed and averaged, and the overall costs
estimated to be compared with the actual costs incurred. Any discrepancies
would have to be analyzed and explained under this program.
38
-------
SECTION 6
RESULTS OF THE EXPERIMENTAL PROGRAM
STREAM WATER QUALITY IMPROVEMENTS
Introduction
The Maumee River is a tributary stream to Lake Erie. The City of Fort Wayne
is the largest municipality located in the drainage basin of the Maumee River.
In the "Lake Erie Report"!, prepared by the U.S. Federal Water Quality Admin-
istration, the City of Fort Wayne was cited as a major source of wastewater
to the Maumee River. A significant part of the problem cited in this report
was reportedly the extensive municipal combined sewer system in the City with
multiple high flow diversion bypass points located throughout the City.
A major objective of this research, development and demonstration grant pro-
gram was to determine the benefits to the receiving water of diversion and
treatment of previously-bypassed combined sewer overflows. The manner of
assessing these benefits has been outlined in the discussion of the experi-
mental program, preceding. The results of this assessment are discussed for
the parameters undertaken in turn.
Dissolved Oxygen
The Maumee River downstream of the City of Fort Wayne was not particularly
threatened by low levels of dissolved oxygen (D.O.) in the years preceding
this construction program. Nevertheless, there have been times when the
stream D.O. was low enough to cause fish kills in the river. These occasions
usually were the result of sustained low flow periods where a significant
part of the stream flow originated from wastewater discharges upstream.
These discharges were usually not associated with stormwater-induced over-
flows in these periods. However, it seemed logical to assume that some
oxygen-demanding matter settling out of storm-caused overflows would adversely
affect stream D.O. at low flow conditions. If this were the case, the stream
D.O. after construction should show some improvement that might be attributed
to the-removal of storm-caused overflows of putrescent settleable solids.
This result should show its effect most noticeably in the low flow months of
the year. Once the deposited matter is settled out, it would remain in the
stream sediment long after the wet weather had ended. This settled matter
should exert an oxygen demand year-round, according to this thinking.
One of the objectives of the evaluation program was to attempt to show an im-
provement in the stream water quality resulting from the removal of a portion
39
-------
of the total combined sewer stormwater overflows for the City of Fort Wayne.
Only approximately 5 percent of the total quantity of stormwater-induced com-
bined sewer overflows were diverted and treated in this construction program.
Nevertheless, it was hypothesized that there still should be a significant
improvement in the stream water quality because this admittedly small propor-
tion of the total overall overflow discharge was thought to be the "dirtiest".
This assumption resulted from the observation that the initial rush of storm-
caused combined sewer flows received at the Water Pollution Control Plant
(WPCP) were frequently heavily laden with sediments, putrescibles, devoid of
any D.O., and highly polluted with pathogens and other contaminants.
This so-called "first flush" tended to load the plant with this hard-to-treat
polluted wastewater. These wastewaters adversely affected treatment plant
operations and operators sometimes allowed these heavily contaminated waste-
waters to overflow to the river via the WPCP bypass rather than attempt to
treat these wastes. It was. reasoned that to divert these raw wastes to a
treatment facility designed especially for stormwater treatment plus deten-
tion would greatly alleviate this loading on the stream.
The pre-construction phase sampling program provided for sampling of the
Maumee River upstream from the plant to establish a stream flow "baseline"
to evaluate the data taken downstream. This sampling point was located at
the Anthony Boulevard Bridge approximately one-quarter mile upstream from
the plant's eventual discharge at the Terminal Pond No. 1.
To assess the immediate effect of the pollutional loading of the terminal
pond effluent on the river, another sampling point was established at the
U.S. 30 Bypass Bridge at about one-and-one-half miles downstream from the
pond effluent. By the time the poncl flow had mixed with the river flow over
this distance, the effect of the plant discharge on the river could be
measured.
A third sampling point was located approximately four-and-one-half miles
downstream. The New Haven Bridge over the Maumee River was the sampling
point for the flow that would travel from 0.5 to 5 hours from the pond dis-
charge to the New Haven Bridge depending on stream flow.
The pre-construction evaluation plan was to sample at regular intervals dur-
ing the time when the plant 84-inch bypass was registering flow going to the
river. The plant interceptor depth was noted as WPCP Chart No. 8. At least
3.5 feet (1.07 meters) of depth in the interceptor was required to reach the
84-inch bypass sewer invert. The 84-inch bypass structure also had a bubbler
tube depth indicator set to indicatje when there was water in this sewer.
Normally, the river water surface elevation was also recorded because a high
water backwater to the plant bypass structure would sometimes fill the sewer
with river water.
Pre-construction sampling results throughout the period starting in August,
1971 and ending in November, 1972 demonstrated a progressive decrease in D.O.
from Anthony Boulevard to the New Haven Bridge. Dissolved oxygen data
averaged for these three sampling points as shown in Table 3.
40
-------
Table 3. PRE-CONSTRUCTION EVALUATION PHASE STREAM DISSOLVED OXYGEN
August, 1971 to November, 1972
Anthony Boulevard
(1/4 mile
upstream)
Arithmetic Mean
Dissolved Oxygen (mg/1)
Standard Deviation (mg/1)
Number of Observations
Range of Values Observed
(mg/1)
8.60
2.47
639
4.0 to 13.6
U.S. 30 Bypass
(1-1/2 miles
downstream)
7.90
2.76
629
4.2 to 13.1
New Haven
(4-1/2 miles
downstream)
7.40
3.12
653
2.1 to 13.1
The ranges of values of the stream D.O. also provide some insight into the
amount of oxygen-demanding materials in the period of sampling. Since stream
D.O. is dependent on temperature, and the most critical period for indigenous
aquatic life is during the warm weather months when the saturation D.O. is
lowest, a statistical distribution of.warm weather measurements shows an
interesting trend. The lowest value noted in the time prior to the construc-
tion of the Stormwater Facility was at New Haven where the warm weather, low
flow D.O. was 0.5 mg/1 on several occasions. See Figure 15 for this trend.
Post-construction sampling of the stream water quality at the
the period commencing January 1, 1975, and ending February 6,
in consistently higher dissolved oxygen measurements for this
apparent from these data is a decreasing trend of values from
vard to U.S. 30 Bypass to New Haven. For comparison with the
evaluation data, Table 4 displays the stream dissolved oxygen
same three sampling points:
same points in
1976, resulted
period. Still
Anthony Boule-
pre-construction
data for these
Table 4. POST-CONSTRUCTION EVALUATION PHASE STREAM DISSOLVED OXYGEN
January, 1975 to February, 1976
Anthony Boulevard
(1/4 mile
upstream)
Arithmetic Mean
Dissolved Oxygen (mg/1)
Standard Deviation (mg/1)
Number of Observations
Range of Values Observed
(mg/1)
9.86
2.53
240
6.1 to 16.0
U.S. 30 Bypass
(1-1/2 miles
downstream)
9.18
2.56
234
4.4 to 14.0
New Haven
(4-1/2 miles
downstream)
8.93
2.82
239
3.8 to 13.6
41
-------
It is apparent from the examination of Table 3 for the Pre-Construction
Evaluation next to Table 4 for the Post-Construction Evaluation that there
has been a significant increase in the average values for stream dissolved
oxygen from 1971-72 to the 1975-76 evaluation period. Figures 15 through 17,
inclusive, show the progressive stream average monthly dissolved oxygen for
the five-year period starting in January, 1971 and ending in December, 1975
for each of the stream sampling points in this program.
Indiana Water Quality Criteria applicable to the Maumee River and pertaining
to dissolved oxygen state, in part:
"...the following criteria are for the evaluation of conditions for the
maintenance of a well-balanced, warm water fish population...applicable
at any point in the waters outs.ide the mixing zone (of the effluent and
the stream):"
"(1) (Dissolved Oxygen) Concentrations of dissolved oxygen shall average
at least 5.0 mg/1 per calendar day and shall not be less than 4.0 mg/1
at any time."
From a statistical analysis of the pre-construction and post-construction,
the percent of the days exceeding the minimum water quality standards was
shown to be:
Table 5. NUMBER OF DAYS STREAM DISSOLVED OXYGEN LESS THAN 4.0 mg/1
Maumee River at Anthony Boulevard
(1/4 mile upstream)
Maumee River at U.S. 30 Bypass
(1-1/2 miles downstream)
Maumee River at New Haven
(4-1/2 miles downstream)
Pre-Construction
Evaluation
6.9
49.6
23.0
Post-Construction
Evaluation
0.04
0.37
11.0
Close examination of Figures 15, 16, and 17 indicates the long-term trend of
dissolved oxygen appears to be improving from the pre-construction phase
(1971-1972) to the post-construction phase (1975-1976). The range of values
for D.O. have consistently decreased over this time period. The same rela-
tive trend is shown at all three locations. Furthermore, the corresponding
long-term trend for the terminal pond D.O. shows a contrary trend to these,
one in which the terminal pond originally improved in effluent D.O. in late
1971 and throughout 1972 until a more variable average D.O. was apparent in
1975-1976. All of these factors tend to indicate that there has been an im-
provement in stream water quality over the period that may be attributed in
part or in whole to the diversion and treatment of combined sewer overflows.
42
-------
FIGURE 15 - MAUMEE RIVER MEAN STREAM DISSOLVED OXYGEN
ANTHONY BOULEVARD BRIDGE
JANUARY, 1970 TO FEBRUARY, 1976
43
-------
FIGURE 16 - MAUMEE RIVER MEAN STREAM DISSOLVED OXYGEN
U. S. 30 BYPASS BRIDGE
JANUARY, 1970 TO FEBRUARY, 1976
44
-------
10,000
FIGURE 17 - MAUMEE RIVER MEAN STREAM DISSOLVED OXYGEN
NEW HAVEN BRIDGE AT LANDIN ROAD
JANUARY, 1970 TO FEBRUARY, 1976
45
-------
Only one-year's data are presented for the post-construction evaluation
period. However, it should be apparent that there has been a significant
reduction in both the frequency of occurrence of low stream D.O. and the
minimum D.O. observed. Future data^gathering should determine whether this
trend is a permanent result of the diversion of some of the raw stormwater-
induced combined sewer overflows to treatment.
Biochemical Oxygen Demand
Pre-construction sampling results for average stream BOD,, increased down-
stream from the pond effluent to the U.S. 30 Bypass Bridge, as shown in
Table 6. The New Haven Bridge samples were found to have decreased slightly,
implying some stream recovery on the utilization of organic matter,, However,
this decreased exertion probably cannot be attributed to any property of the
stream recovery because the amount of difference between the samples taken at
New Haven and U.S. 30 Bypass could be attributed to experimental error alone.
Furthermore, this indication of some kind of stream recovery does no.t carry
over to the dissolved oxygen data described earlier. It is concluded that
the decrease in stream BOD,- from U.S. 30 Bypass to the New Haven Bridge dur-
ing the pre-construetion evaluation is an anomaly of these data and not an
indication of stream recovery over this short a distance.
Table 6. PRE-CONSTRUCTION EVALUATION PHASE STREAM BIOCHEMICAL OXYGEN DEMAND
August, 1971 to November, 1972
Anthony Boulevard
(1/4 mile
upstream)
Arithmetic Mean
Stream BOD5 (mg/1) 4.29
Standard Deviation
(mg/1) 2.82
Number of Observations 441
Range of Values 0.2 to 27.
U.S. 30 Bypass
(1-1/2 miles
downstream)
6.85
4.95
441
0.5 to 27.
New Haven
(4-1/2 miles
downstream)
5.51
3.31
443
0.5 to 25.
46
-------
Note that the range between maximum and minimum for each sampling site is
basically the same for all. The maximum stream BOD,, is 27 milligrams per
liter. ' 5
Post-construction evaluation phase data taken at the same three sampling
points show essentially the same trend as the pre-construction phase, con-
sidering a range of probable sampling and analysis error of +20 percent.
These data are presented in Table 7. The range of values between maximum and
minimum for each location has markedly decreased in these data. This de-
crease in range also resulted in a decrease in maximum stream BOD-. This
result is interpreted as a manifestation of stream improvements that may be
partly attributed to combined sewer stormwater overflow treatment at the
Water Pollution Control Plant. It tends to confirm the results presented in
the section discussing stream D.O. where a noticeable increase in the stream
mean D.O. was noted between pre-construction and post-construction periods.
Table 7. POST-CONSTRUCTION EVALUATION PHASE STREAM BIOCHEMICAL OXYGEN DEMAND
January, 1975 to February, 1976
Anthony Boulevard
(1/4 mile
upstream)
Arithmetic Mean
Stream BOD^ (mg/1)
Standard Deviation
(mg/1)
Number of Observations
Range of Values
Observed (mg/1)
2.01
3.21
201
0.1 to 11.0
U.S. 30 Bypass
(1-1/2 miles
downstream)
3.95
2.05
190
0.1 to 11.0
New Haven
(4-1/2 miles
downstream)
4.34
2.27
191
0.2 to 11.0
Maumee River chemical oxygen demand samples taken at these same three points
show roughly some of these same trends. However, there are fewer data in the
post-construction evaluation available for comparison.
As a result, no conclusions were drawn from these data compared to the pre-
construction phase.
Fecal Coliform Group Indicator Organisms
Bacteriological indicators were monitored in the stream sampling program in
both the pre-construction and post-construction evaluation work. The same
three sampling points were monitored in both evaluation periods. Upstream
at Anthony Boulevard (1/4 mile upstream), downstream at U.S. 30 Bypass (1-1/2
miles downstream), and at New Haven Bridge (4-1/2 miles downstream). Table 8
shows the summary results of the pre-construction sampling program for fecal
coliform group indicator organisms.
47
-------
Table 8. PRE-CONSTRUCTION EVALUATION PHASE STREAM
FECAL GOLIFORM GROUP INDICATOR ORGANISMS
August, 1971 to November, 1972
Anthony Boulevard
(1/4 mile
upstream)
U.S. 30 Bypass
(1-1/2 miles
downstream)
New Haven
(4-1/2 miles
downstream)
Geometric Mean
Fecal Coliform
Group (N/100 ml) 1,650
Standard Deviation
(N/100 ml) 116,000
Number of Observations 322
Range of Values 0 to
Observed (N/100 ml) • 29,000
9,000
392,000
340
0 to
1,400,000
4,800
476,000
343
0 to
6,400,000
It should be noted that there appears to be a bacterial "die-off" phenomenon
at work from the U.S. 30 Bypass to the New Haven Bridge sampling points.
With normal stream transit times of from 0.5 to perhaps 5 hours between these
two sampling points, this result may represent a declining population phase
for this class of organisms.
Total coliform indicator organisms sampled in the pre-construction phase
showed similar trends to the fecal coliform group. These data were not tabu-
lated because of insufficient numbers of events analyzed to statistically
evaluate.
Fecal coliform data were selected in the post-construction period as the most
reliable indicator of the bacteriolpgical contamination of the stream from
combined sewer overflows. Table 9 shows the comparable results from the post-
construction evaluation phase.
From the examination of the geometric mean distributions of the data for
fecal coliform group indicator orgahisms, it should be apparent that there
has been a significant improvement in the stream water quality as measured by
the fecal coliform MF procedure. Figure 18 shows these data plotted on log
normal coordinates for the pre-construction evaluation phase. Figure 19
similarly displays the data for the post-construction evaluation phase.
As shown in Figure 19, it appears as though stream background fecal coliform
measured at Anthony Boulevard must be decreased before any further improve-
ment in water quality downstream from the plant can be achieved.
48
-------
Table 9. POST-CONSTRUCTION EVALUATION PHASE STREAM
FECAL COLIFORM GROUP INDICATOR ORGANISMS
January, 1975 to February, 1976
U.S. 30 Bypass
(1-1/2 miles
downstream)
1,000
New Haven
(4-1/2 miles
downstream)
800
Anthony Boulevard
(1/4 mile
upstream)
Geometric Mean
Fecal Coliform
Group (N/100 ml) 920
Standard Deviation
(N/100 ml) 63,000
Number of Observations 131
Range of Values • 0 to
Observed (N/100 ml) 80,000
Total Phosphorus
Stream water quality measurements of total phosphorus have been regularly
taken by Water Pollution Control Plant personnel for several years. This
program included sampling and analysis of the Maumee River at the same three
points for total phosphorus during and after each stormwater event. The data
obtained for stream total phosphorus in the pre-construction program are
shown in Table 10 and the corresponding post-construction data are given in
Table 11.
Table 10. PRE-CONSTRUCTION EVALUATION PHASE STREAM TOTAL PHOSPHORUS
44,000
125
0 to
150,000
78,000
125
0 to
5,000,000
August, 1971 to November, 1972
Anthony Boulevard U.S. 30 Bypass
(1/4 mile
upstream)
Geometric Mean*
Total Phosphorus
(mg/1) 0.37
Standard Deviation
(mg/1) 0.21
Number of Observations 423
Range of Values
Observed (mg/1)
(1-1/2 miles
downstream)
0.78
0.90
427
New Haven
(4-1/2 miles
downstream)
0.90
0.72
413
0.1 to 2.0 0.1 to 3.5 0.1 to 3.4
*Log normal distributions provided the best fit of these data.
49
-------
10
9999 gg.g 99,5 99 98 95 90 8O 7O 60 50 40 3O 20 10 5 21 .5 .1
6 I III II i i i i i i i I ( I I *> i ,
.01
CC
e
8
a
UJ
CC
C/)
I0a
10'
10'
MAUMEE RIVER AT U.S. 30
BYPASS 1/2 MILE DOWNSTREAM
MAUMEE RIVER AT NEW HAVEN
4 1/2 MILE DOWNSTREAM .
MAUMEE RIVER AT ANTHONY
BLVD. 1/4 MILE UP STREAM
10' / tt If I I I I I I I I I ' I
.01 .1 .512 5 10 20 30 40 50 60 7O 80 90 95 98 99 99.5 99.9 99.99
FIGURE 18 - STREAM FECAL COLIFORM INDICATOR ORGANISMS
PRE-CONSTRUCTION EVALUATION PHASE
AUGUST, 1971 TO NOVEMBER, 1972
50
-------
99.99
O
o
2
o:
P
o
o
o
UJ
UJ
ioc
99.9 99.5 99 98
I ill
95 90
I I
80 70 SO 50 40 30 ZO
i I i i i i i
10
10°
102
10'
,01
.01
_MAUMEE RIVER AT ANTHONY
"BLVD. 1/4 MILE UP STREAM.
MAUMEE RIVER AT U.S. 30
"BYPASS 1/2 MILE DOWNSTREAM
• MAUMEE RIVER AT NEW HAVEN
20 3O 40 50 6O TO SO 90 95
ill I
98 99 99.5 99:9
99.99
FIGURE 19 - STREAM FECAL COLIFORM GROUP INDICATOR ORGANISMS
POST-CONSTRUCTION EVALUATION PHASE
JANUARY, 1975 TO FEBRUARY, 1976
51
-------
Table 11. POST-CONSTRUCTION EVALUATION PHASE STREAM TOTAL PHOSPHORUS
January, 1975 to February, 1976
Geometric Mean*
Total Phosphorus
(mg/1)
Anthony Boulevard
(1/4 mile
upstream)
0.40
Standard Deviation
(og/1) 0.25
Number of Observations 174
Range of Values
Observed (mg/1) 0.1 to 3.1
U.S. 30 Bypass
(1-1/2 miles
downstream)
0.37
0.32
139
0.1 to 1.7
New Haven
(4-1/2 miles
downstream)
0.40
0.31
143
0.1 to 2.3
*Log normal distributions provided the best fit of these data.
There are several plausible explanations for the reduction in stream total
phosphorus from 1971-72 to 1975-76. First, a State-wide detergent: phosphate
law went into effect in Indiana early in 1971. This legislation limited
household laundry detergents sold in the State to 8.7 percent as phosphorus
until 1973. Thereafter, these detergents sold in Indiana were required to be
"phosphate-free". Some of the change in the stream water quality at Fort
Wayne may be attributed to the effect of this legislation.
Another plausible alternate hypothesis to the diversion, treatment, and
detention of combined sewer overflows as the explanation for the improved
water quality as measured by total phosphorus lies in the manner of operation
of the Water Pollution Control Plant itself. In the summer of 1973, this
source was required by Indiana Stream Pollution Control Board regulations to
begin treatment of the plant effluent for removal of total phosphorus. The
results of this treatment is shown in the discussion on plant effluent data
taken at the terminal pond discharge.
These results may also be noted as .showing that the post-construction evalu-
ation data given in Table 11 show essentially the same trend of values up-
stream and downstream from the treatment plant and the combined sewer over-
flows. This result may be significant: it indicates that no additional
treatment of municipal wastewater at whatever origin will result in a further
reduction in the stream pollutional levels of total phosphorus, if the data
taken at the Anthony Boulevard Bridge truly represent the stream background
condition with respect to this pollutant. Additional upstream data-gathering
outside the scope of this work will be necessary to substantiate this conclu-
sion.
52
-------
Figure 54, following, depicts .the long-term plant effluent total phosphorus
levels have markedly decreased since the phosphorus removal facility start-up
in July, 1973. This result tends to substantiate the conclusion that the
stream water quality improvement in total phosphorus is mostly attributable
to phosphorus removal at the Water Pollution Control Plant.
Ammonia Nitrogen
Stream water quality monitoring in the pre-construction evaluation did not
include sufficient data for ammonia nitrogen to show any significant trends.
In the post-construction evaluation phase, these data were periodically taken
in dry weather as well as during storm events. The results seem to fit a log
normal distribution better than a normal distribution. Table 12 shows the
results for the distribution that best fit the data taken.
Table 12. POST-CONSTRUCTION EVALUATION PHASE AMMONIA NITROGEN
January, 1975 to February, 1976
Anthony Boulevard
(1/4 mile
upstream)
Geometric Mean
Ammonia Nitrogen
(mg/1)
Standard Deviation
(mg/1)
Number of Observations
Range of Values
Observed (mg/1)
0.42 ~
0,50
134
0.05 to 2.0
U.S. 30 Bypass
(1-1/2 miles
downstream)
0.84
0.63
135
0.2 to 4.0
New Haven
(4-1/2 miles
downstream)
0.90
0.84
140
0.1 to 4.0
Suspended Solids (Nonfilterable Residue)
Pre-construction and post-construction sampling of stream suspended solids
(nonfilterable residue) showed virtually no change in stream water quality
that could be attributed to the stormwater treatment program. The distribu-
tions obtained are tabulated in statistical summary form in Table 13.
Volatile suspended solids data gathered in the pre-construction evaluation
essentially showed a bimodal distribution depending on the season. Cold-
weather months seemed to have greater frequencies of low volatile data (10 to
20 percent VSS) while in contrast, warmer weather data showed a tendency to
cluster in the range of 25 to 35 percent VSS. This result is not considered
particularly significant because the values of VSS are dependent on the values
of total suspended solids. The data for total suspended solids gathered in
the preoonstruction phase essentially showed no difference in this parameter.
There were too few data gatherings in the post-construction evaluation stream
sampling to warrant any comparisons for pre-construction to post-construction
for volatile suspended solids.
53
-------
r
Table 13. STREAM SUSPENDED SOLIDS
Nonfilterable Residue
Anthony Boulevard
(1/4 mile
upstream)
U.S. 30 Bypass
(1-1/2 miles
downstream)
New Haven
(4-1/2 miles
downstream)
Geometric Mean
Suspended Solids
(rog/1)
Pre-C.
60
Post-C. Pre-C. Post-C. Pre-C. Post-C.
Standard Deviation 195
Number of Observations 266
Range of Values • 5 to
Observed (mg/1) 1,390
60
315
189
5 to
1,600
72
155
257
72
190
191
5 to 5 to
1,400 1,100
70
150
272
5 to
800
105
125
195
5 to
1,300
Note: The pre-construction evaluation period included August, 1971 through
November, 1972. The post-construction evaluation period began
January 1, 1975 and,ended February 6, 1976.
Stormwater Facility Evaluation Parameters
The principal Stormwater Facility parameters, which formed the basis of the
comparative evaluation for the treatment equipment used in this program, were
settleable and flotable coarse solids and hydraulic capacity.
Settleable and Flotable Debris—
A distinguishing characteristic of raw, untreated combined sewer storm flows
received at the Fort Wayne Water Pollution Control Plant has been the amount
of coarse solids and debris contained in the storm flow. These materials
have been screened continuously at the plant by a mechanical bar screen for
years, resulting in very large quantities of screened materials necessitat-
ing sanitary disposal. Since the Stormwater station was designed to handle
storm-caused overflows which would have gone to the river without treatment,
a trash rack with bar spacing of 2-1/4 inches was provided.
Figure 20 shows the nature of the screened materials that have been deposited
on the trash rack. In one location (not shown), a beer can was found firmly
lodged between the vertical bars in an upright position. Plant operating
personnel report that extraction of these materials using the motorized bar
rack mechanism has been mostly unsuccessful because these materials are
light and,easily dislodged when dry, and regularly pass through the bar
spacing.
54
-------
Figure 21 is a photograph of the fibrous, stringy debris hanging from the
horizontal members on the trash rack taken from a boat in the final weeks of
evaluation. This type of debris was also noticeable in the chlorine contact
channel after long, sustained events where the raw stormwater pumps operated
for a number of hours.
One of the reasons for the incursion by boat into the Stormwater Facility
wetwell was to inspect the condition of the wetwell and raw stormwater pumps.
Figure 22 is a photograph taken on December 8, 1975, in the wetwell at a
depth of 9 feet on the depth indicator.
The pump shown is the Johnston 30 PS pump. There was no evidence of external
wear or damage resulting from pumping raw combined sewer overflows.
Table 14 presents the results of wetwell soundings and sampling results taken
on December 8, 1975, after the facility had been in service for nearly one
year. The depth of sediment measured in this survey resulted from standing
water of depth from 7 to 11 feet (2.14 to 3.35 meters) in the wetwell and
river crossing conduits. All solids data are for total sample volume (about
one liter) except Sample No. IB which is sediment only. All depths were
taken by sounding with a graduated survey level rod and bottom sampling tube.
Table 14. STORMWATER FACILITY WETWELL SEDIMENTS
SAMPLING SURVEY RESULTS
December 8, 1975
Sample
No.
Location
1A CT 25 MGD Pump & CT
Wetwell 25'
Trash Rack
N of
IB
2A
2B
10' R+ & 10' N of C,
25 MGD Pump
NE Corner of Wetwell
NW Corner of Wetwell
3A Center Bay 25' S of
N Wall
3B W Side 35' S of N Wall
**Insufficient Sample Volume.
Depth of
Sediments
(feet)
1.5
1.5
0.5
0.25
0.5
0.5
Sediment
Specific
Gravity
1.002
1.062
1.016
1.026
1.020
Total
Solids
(Percent)
4.4
31.9
13.7
2.9
3.1
2.5
Volatile
Solids
(Percent)
27.0
31.0
26.0
49.0
. 29.0
31.0
55
-------
Figure 20 - Screened Material Adhering to
Trash Rack Vertical Bars
Figure 21 - Interior View of Storm-
water Facility Trash
Rack Showing Stringy
Debris
Figure 22 - Johnston 30 PS Pump Column
and Suction Bell After
Eleven Months of Operation
56
-------
Given the highly variable nature of the combined sewer overflows, it seemed
logical to assume that any removal of the gross solids would benefit the
screening devices. The debris shown in Figures 20 and 21 would likely cause
problems downstream if not removed somehow. Figure 23 is a photograph of the
chlorine contact channel scum removal pipe resulting from an event on July 19,
1975. This material normally was removed in screening unless the head tank
or Rex headbox overflows allowed it to bypass screening by overflowing to the
effluent channel.
Among the screening methods studied during this program, the Bauer Hydra-
sieve® appeared to best deal with gross solids debris. The Envirex Rotary
Drum Screen and SWECO Centrifugal Wastewater Concentrator produce more liqui-
fied concentrate that required further treatment at the Water Pollution
Control Plant. In contrast to -this, the Bauer produced a semidry solid resi-
due that was usually hosed into a 55-gallon drum and trucked off to a sani-
tary landfill for ultimate disposal. This method of disposal is preferred
for stringy debris because these materials are less putrescible than other
organic solids. Furthermore, this debris sometimes is not caught in the
WPCP moving bar screen and winds up in the plant raw sludge feed to the an-
aerobic digesters. Stringy nondegradable debris has been a problem in the
anaerobic digesters at Fort Wayne in the past.
Figures 24 and 25 show photographs of the drying debris strained from the
storm flow. Note the plastic materials such as plastic bandages, plastic
drinking straw, and magnetic recording tape (upper left corner, Figure 25).
These materials are known to be nonbiodegradable using conventional sludge
handling and disposal treatment techniques. Once removed from the wastewater
by screening, it is possible to dispose of these dry solids at a sanitary
landfill. However, to operate these screens in this manner reduces the unit
effective hydraulic loading somewhat.
Figure 26 shows Bauer screenings collection channel after an event where the
unit hydraulic loading was limited to approximately three-fourths of design
flow to produce the semidry solids shown in Figures 24 and 25. This corres-
ponds to a surface hydraulic loading rate of approximately 25 gpm/sq. ft.,
or 1,050 lpm/m2.
At full hydraulic design capacity of 200 gpm per foot (2,500 1pm per meter)
of screen width (35 gpm/sq. ft., or 1,400 lpm/m2), the photograph of this
center channel shows the wastewater "skipping" along the surface of the
screens (Figure 27). Note that the unit at full capacity does produce some-
what more flow in the screening concentrate channels while giving up the
desirable characteristic of semidry solids screenings.
Hydraulic Capacity—
One of the consequences of operating a large scale facility for the treatment
of combined sewer overflows is that the facility must be capable of taking
everything that "comes down the pipe", so to speak. This necessitates some
compromises within the physical constraints of the system. The Stormwater
Facility at Fort Wayne was designed to pump and screen 75 MGD (3.286 cumec).
However, there were times when the nominal capacity of the screening equipment
57
-------
Figure 23 - Chlorine Contact Channel
Flotable Debris Collected at
Scum Removal Pipe - July, 24, 1975
Figure 24' -
Bauer Hydrasieve
Screenings After
Stormwater Event
Figure 25 - Close-Up of Semidry
Screenings Collected Dur-
ing Stormwater Event From
Bauer Hydras ieveO;)
58
-------
Figure 26 - Bauer Hydrasieve^ Center
Concentrate Channel Showing
Semidry Screenings Collected
Figure 27 - Bauer Hydrasieve^ Operating
at Full Hydraulic Capacity
59
-------
was Inadequate to treat all the wastewater received at the facility. In
order to minimize the amount of storm-caused overflows to the river, the de-
cision was made to pump as much of the flow to the screening room head tank
as possible and allow it to flow by gravity head either to the screening
units or to overflow to the screened effluent channel. In addition to this
raw unscreened waste overflow point, the Rex screen headbox has an overflow
point that leads to the screened effluent channel also (See Figure 28).
The overflows from the head tank were not metered in this program. The over-
flows from the Rex Rotary Drum Screen headbox were recorded by a circular
chart stage recorder. Regardless of whether stormwater from these two over-
flow bypasses in the facility received any type of screening, these wastes
were pumped to the elevation of the chlorine contact channel where they were
disinfected and run through the full detention pond period. The pond is an
inherent part of the treatment scheme and it should be recognized that a high
degree of treatment efficiency can usually be expected from such a pond.
The Rex Rotary Drum Screen overflow shown in Figure 28 is recirculated pond
water, and it appears in the photograph as clear and free of noticeable
solids. However, more typically, raw storm flow contained coarse solids and
all of the raw waste pollution. The chlorine contact channel is more than
1,100 feet (335 meters) long. In this length, the velocity of flow is pur-
posely less than 1.5 feet per second (0.46 meter per second) to avoid scour-
ing the bottom and undermining the fill material that makes up the levees
around Terminal Pond No. 1. A3 mil polyethylene channel liner was torn in
many places in the first few weeks of operation of the facility and did not
serve any useful purpose thereafter.
Figure 29 shows a longitudinal view of the channel after eleven months of
operation. Note the sludge banks near the scum baffle. Figure 30 is another
view of this same area. These photographs are typical of the channel bottom
along its length.
Similar sludge deposits formerly could be seen along the Maumee River at low
water. Since these deposits now occur at the facility rather than the stream
some method of treatment can be used to neutralize and/or remove these
settled solids.
The sediment deposits shown in Figures 29 and 30 are expected to continue
throughout the useful lifetime of the facility. These sediments atlso have
been observed to a depth of 1.5 feet (0.46 meter) in the terminal pond. A
small self-contained dredge has been operating in the terminal pond subse-
quent to the termination of the evaluation phase, beginning in March, 1976.
This dredging of these sediments has been responsible for removal of most of
these deposits in the pond. However, the .dredge unit cannot be used in the
chlorine channel itself because its draft is more than the channel depth.
A good technique for cleaning up these sediments would be to sluice the
settled solids with a high pressure water hose. However, the plastic channel
liner has deteriorated to the extent that any extensive amount of water
sluicing would result in excessive erosion of the channel banks and levees.
It is recommended that future channels designed for this purpose be lined
60
-------
Figure 28 - Rex Rotary Drum Screen
Headbox Overflow Point
Figure 29 - Chlorine Contact Channel Figure 30 - Chlorine Contact Channel
From Overflow Weir to Bottom at Overflow Weir
Terminal Pond Showing
Sludge Bank Deposits
to Terminal Pond
61
-------
with impervious bituminous or Portland cement concrete paving material with
underdrains to carry the sluiced sediments to the ultimate disposal point.
Stormwater Screening Equipment Comparative Evaluation
Table 15 presents the screening equipment comparative evaluation on the basis
of overall treatment efficiency for suspended solids, BOD,., COD, total phos-
phorus, and ammonia nitrogen (NliJ-N). Note that there were negative treat-
ment efficiencies determined for each of the screening methods considered in
this program.
Negative treatment efficiencies could be misinterpreted without some explana-
tion. Measurement error in either flow or concentrations may be responsible
for some of the negative efficiency parameters. Also, the time of sample
taking may account for some of the variation noted. The few number of events
(as few as 25 events for volatile suspended solids sampling) could also
account for some of the variation noted. The fluid shear observed in the
screening methods themselves may account for a portion of the variation and
negative results.
The manner of calculation of the screening equipment efficiencies was pro-
grammed by computer to perform the computations of loadings by increments for
which there were sufficient concentration data taken. These incremental load-
ings were then summed to obtain the sampling location event loadings. The
sampling location event loadings were then used to compute the treatment
equipment efficiencies. For the sake of relative comparisons, the negative
efficiency results so obtained are recorded in Table 15 as zero efficiency
with an asterisk (*) superscript to denote negative results obtained.
Suspended Solids (Nonfilterable Residue)—
Figure 31 is a graphical "scattergram" of the relationship between the three
screening equipment treatment efficiencies as a function of each screening
method rated hydraulic loading.* This approach was selected as the common
determinant among the different screening methods evaluated because it re-
lates directly to the hydraulic flux rate for the rotating screening devices
(i.e., Rex and SWECO units) operated at a constant rotational speed as well
as the fixed screening area Bauer Hydrasieve®. The screen drive for the Rex
unit was maintained at a constant rate of rotation throughout most of this
program and the variation in flux rate was obtained by varying the raw storm-
water flow.
*Rated hydraulic loading for the screens evaluated was based on the manu-
facturer's recommended hydraulic loading rate as determined in the equipment
bidding procedure. The nominal rated hydraulic loading for the Bauer
Hydrasieve® was 13,125 U.S. gpm (49,700 liters per minute), for the Envirex
Rotary Drum Screen was 12,600 U.S. gpm (47,700 liters per minute), and for
each SWECO CWC ®, the rated loading was 3,100 U.S. gpm (11,700 liters per
minute).
62
-------
TABLE 15. SCREENING EQUIPMENT COMPARATIVE EVALUATION OF
OVERALL TREATMENT EFFICIENCY
4.
H
c.
ftj
c
o
c
HI
U
•H
UJ
H
iJ
i
4-1
in
01
H
r-l
rt
IJ
01
3
O T3
«H (U C.
a £
H f C
cfl U t-
H
•H a
o a
I8
Q 4.
§
tH ^
10 O
4J J3
O A
H co
^ e
d.
3c -a
01 C
11 5? S
J3 O Q
U
M
o
r< a -a
g Q> C3
0) 00 flj
O X §
A
•o
•a to
c -a
(U -H
O.r-1
to o
9 CO
CO
CO
B
creeni
Units
CO
• en
G c
•H O
C -H
a) <->
01 a
U 3
a>
u
U 01
nj u
S §
S05
fncno OOOON ooo mcoo inmm o
^oincx CNCNO ^.^o> \o\o(T» m in m m i i~i
! i : i i i : i ! ! i i i : i i i i
'"srss:? MI iii vo^ScniS
t. «i ~ * * * ** *
S^iSl <=00 CNSTU-! CV100 Or-10 rHIN
^"^ vo in m rn cnicN
_ *** «««•»:««
ON sr 0> 000 OOrH 000 000 l^ICO
^o r- m cs | en
^J^"^ CNONO vomo r^cNO crvr^ov stloo
mvom minsr in-
-------
TABLE 15. SCREENING EQUIPMENT COMPARATIVE EVALUATION OF
OVERALL TREATMENT EFFICIENCY - Continued
S
O T3
rH a a
SS"
S
r- o
o% c^
o d
ro n
to r~*
O CM
o) c
•r4 Ot
C 6C
o o
a n
1 3
a
i i
I i
§
ra o
•u A
o a
H u
o
&
PU
° i
H
OJ C 13
0 S B
Tl M 3
§ j? S
js o a
C
r^ i
« 1
r-l
ra
u
•H C TJ
§01 C
oo n
X fc S
§5S
•H
m
•a
01
T3 O
c -o
QJ
o
u
w
o
u
w
o
JJ
w
64
-------
TABLE 15. SCREENING EQUIPMENT COMPARATIVE EVALUATION OF
OVERALL TREATMENT EFFICIENCY - Continued
g*
rH 0) O
ta to s
M , S
n
01
t3 W
C -0
OJ vH
a. tH
(0 O
3 Ul
C/3
00
c
•H Ul
01 C
J M
U 3
w O
01
in cu
nj o
3
S 0
03
u
Is8
•*J" CN CO O\ I^« OJ vO rH O O O\ ON \O OJ CT> OJ vO
COCNO\ -
-------
TABLE 15. SCREENING EQUIPMENT COMPARATIVE EVALUATION 01'
OVERALL TREATMENT EFFICIENCY - Continued
O\ VO CO
I
04
J3
PM
a
2
H
•te -K
o o
00
B
3 X W
rt a) 5
M US CO
8
X U
0) 3
CO
B C
•H O
B -H
a
•
E
o
I
66
-------
TABLE 15. SCREENING EQUIPMENT COMPARATIVE EVALUATION OF
OVERALL TREATMENT EFFICIENCY - Continued
o -o
rH 0) C
rH co
s h s
ai x a)
J3 6 «
o
rH
•H B -a
eve
o) oo cd
•H
pa
01
•a ra
5S
00
C!
•rl CO
a
CO
00
ci c
•H O
C -H
D u
QJ CO
CJ 3
to o
at
4-J
tfl QJ
w y
5 Q
S CO
05
C •
QJ O
1
coin CN m CN ^t- . cN ooo counm
CO CN iH CO
to OVJ O(J OJ-i O
| K g §«a §xM §xg
flj
•T3
&
cO
-K * M
O O -*3" 0
•5
S ™
w •§
H O to
a) u ,j
1 1 1 ^
&
§
•H
CJ
•rl
*W
01
4-1
S
§ | S
es i ^
aj cu co Q,
"o ZS b.
S ,- g -S
w c CD 11
EH 0 0! t5
00
S S
H
S *
CO
4J
•H
3
Ai
B
cd
rB
4J
1
0
tn
•a
B
cd
S
Cd
LM
O
T3
cd
-------
100
o
-h
30
-80
'
-60
•• —
-40
•
I
B
i
.
•"
i
•
•
* •
•
•
i
*
"
t
•
•
•
•
i — .
-20 0 20 4O 60 80 10
PERCENT REMOVAL EFFICIENCY
SUSPENDED SOLIDS
(I4«) (lit)
(l«4)
(M4)
s
— r~
00
-80
-60
•
•
•
•
•
0
f
f
•
*
,
•
**
-40-20 0' 20 40 60 80 IOO
PERCENT REMOVAL EFFICIENCY
SUSPENDED SOLIDS
•
(ITS) (144) •
50
O
v~j
00
•
«>
— v
•
•
4
•
(
*
•
>
*
|
I
•
*
-80-60-40-20 0 20 40 60 80 IOO
PERCENT REMOVAL EFFICIENCY SUSPENDED SOLIDS
FIGURE 31 - SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR
SUSPENDED SOLIDS (NON-FILTERABLE RESIDUAL AS
A FUNCTION OF HYDRAULIC LOADING RATE)
68
-------
As can be verified from the examination of this figure, there does not appear
to be a discernable correlation between removal efficiency for suspended
solids and hydraulic loading rate for any of the screens studied.
Another analysis parameter that was examined in this program was the rate of
suspended solids loading. It was hypothesized that the screens could be ex-
pected to perform differentially to the rate of solids loading for the units.
The "scattergram" results for this analysis are shown in Figure 32.
The average removal efficiency for the respective screens is given in Table
16.
Table 16. SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR
SUSPENDED SOLIDS (NONFILTERABLE RESIDUE)
Percent Removal of Suspended Solids
SWECO CWC®
Centrifugal
Wastewater
Concentrator
Arithmetic Mean @ 50%
Standard Deviation
f>
Number of Observations
Bauer .-^
Hydrasieve Uy
+4.7
34.8
27
Rex Rotary
Drum Screen
+9.2
27.6
26
+22.6
29.5
28
One complicating factor to a systematic analysis of these suspended solids
results is the frequency of screen failures of the SWECO, and to a lesser
extent, the Rex units. There were insufficient screen failure data taken in
this program to warrant the analysis of screen failure cases versus total
cases for some discernable trend based on screen failures. In a strictly
qualitative manner, there were several events observed where the fact of a
screen failure was shown to have resulted in a decreased quantity of screen-
ings concentrate for the SWECO CWC®units. This result is mentioned as a
possible plausible reason for the wide dispersion of efficiency observed for
this particular screen. However, refer to the later discussion pertaining
to the screen life of these SWECO screens.
The available data for the removal efficiency of the screens for volatile
suspended solids as a function of hydraulic loading rate show essentially the
same random patterns as Figure 31. Similarly, the screening efficiency for
the solids loading rate also follows closely Figure 32. Therefore, the vola-
tile suspended solids efficiency data arrays are not shown. Table 17 presents
the averages and estimates of dispersion for these volatile suspended solids
data.
There were no data taken for any of the screening methods utilizing chemical
flocculant or coagulant aids. However, data taken elsewhere indicate that
these aids may improve screen performance. It was not possible to investi-
gate these chemical aids under this program due to time and resource
69
-------
r
s
0.5
Vu
-KX) -80 -60 -40 T20 0 20 40 60 (JO 100
PERCENT REMOVAL EFFICIENCY
SUSPENDED SOLIDS
1.0
05
5
-K
30
-80
-60
*
1
«.
•
\
•
B
1
*
•
"
-40 '-20 0 20 40 60 80 100
PERCENT REMOVAL EFFICIENCY
SUSPENDED SOLIDS
1
1.0
1
£
X
°-
00
-80
-60
t
—
•t
•
•
t
l_i
f
i* •
• •
-40-20 0 20 40 60 80 WO
PERCENT REMOVAL EFFICIENCY SUSPENDED SOLIDS
FIGURE 32 - SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR
SUSPENDED SOLIDS (NONFILTERABLE RESIDUE)
AS A FUNCTION OF SOLIDS LOADING RATE
70
-------
limitation. There were no provisions for chemical additions at the Storm-
water Facility except for chlorination and cleaner solution addition.
Table 17. SCREENING EQUIPMENT REMOVAL EFFICIENCY
FOR VOLATILE SUSPENDED SOLIDS (VSS)
Percent Removal for Volatile Suspended Solids
Arithmetic Mean @ 50%
Standard Deviation
Number of Observations
Biochemical Oxygen Demand
Bauer ^~.
Hydra sieve ^§/
3.0
32
23
(BOD,) —
Rex Rotary
Drum Screen
1.0
36
23
SWECO CWC^i>
Centrifugal
Wastewater
Concentrator
24.0
30
25
The removal efficiency of BOD, was also examined. Shown in Figure 33 is the
BOD^ loading removal efficiency as a function of the hydraulic loading rate.
The dispersion of the results shown in this figure is similar to that for the
suspended solids data already examined in Figure 31.
Figure 34 is a figure depicting the BOD5 removal efficiency as a function of
BOD, loading rate. No apparent discernable trend can be identified from this
analys is.
Table 18 presents the summary results for the BOD loadings studied in this
program.
Table 18. SCREENING EQUIPMENT REMOVAL EFFICIENCY
FOR BIOCHEMICAL OXYGEN DEMAND (BOD,)
Percent Removal of Biochemical Oxygen Demand
Arithmetic Mean @ +50%
Standard Deviation
Number of Observations
Bauer ^
Hydras ieve vy
+5.7
26.9
26
Rex Rotary
Drum Screen
+3.4
32.3
25
SWECO CWCVSx
Centrifugal
Wastewater
Concentrator
+3.1
39.1
27
71
-------
<2
°=i
*
&Q
CC™^
GL*"^
§'
o
ga
I
i-_j
UJO
Bl
OL<5
al
§
1
8
8
8
• i
-100 -80 -60 -40 -20 0 20 40 60 80 100
PERCENT REMOVAL EFFICIENCY FOR BOD5
(I4t) HIT) (114)
• • •
(194)
-100 -80-60-40-20 O 20 40 60 80 KX>
PERCENT REMOVAL EFFICIENCY FOR BOD5
roo
50
°H
*
00
9
i •
•
•
•
t
••
•
•
4
1
(
.
m
4
It
II)
-80 -60 -4O -20 0 20 40 60 80 100
PERCENT REMOVAL F^FICIENCY FOR BOD5
FIGURE 33 - SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR
BIOCHEMICAL OXYGEN DEMAND (BODj AS A
FUNCTION OF HYDRAULIC LOADING RATE
72
-------
u
it
a:
So
§x
J 'if
f|
0
•
•
1
•
•
__J
•
•
f
•"
• 1
•
1
•
•
•
-KDO -80-60-40-20 0 ' 20 4O 60 80
PERCENT REMOVAL EFFICIENCY FOR BOD5
1.0
<
le-
in
0=S|
m
-100 -80 -60 -40 -2O 0 20 4O 60 80
PERCENT REMOVAL EFFICIENCY FOR BODg
-60 -40 -20 0 20 40 60 80
PERCENT REMOVAL EFFICIENCY FOR BOD5
FIGURE 34 - SCREENING EQUIPMENT REMOVAL EFFICIENCY
FOR BIOCHEMICAL OXYGEN DEMAND (BODj
AS A FUNCTION OF BOD LOADING RATE
100
100
i
o
3 x
« =0>5
i»*
CO
0
•
.
•
•
^
•f 1
*1
_?4
IL_j
4
f «
't
B
100
73
-------
These results are somewhat surprising because they imply that the Bauer
Hydrasieve®with a screen opening size of 1.5 millimeters (0.060 inch) is on
the same order of magnitude as the finest screen studied, SWECO CWC^with a
screen opening size of 105 micrometers (0.004 inch). However, it should be
noted that the difference between the mean efficiencies for the Bauer and the
others is less than 2.0 percent, when the sampling and measurement error
alone is not likely to be less than 25 to 35 percent.
Table 19 presents results for both total and filtered BOD5. These data are
based on event-composited samples taken every fifteen minutes over a 2-1/2-
hour sampling period. Note that the filtered and total screened effluent
BOD in every 'case except the chlorinated screened effluent channel shows a
gain in BOD,- concentration as compared to the raw stormwater concentration.
However, the magnitude of the gain in concentration is within the error of
measurement for all samples except the Rex effluent total BOD5-
Table 19. SCREENING FACILITY FILTERABLE BOD5
DETERMINATIONS - EVENT NO. 207
Event Composited
Biochemical Oxygen Demand
Raw Stormflow at Head Tank
Bauer Effluent
Rex Effluent
SWECO Effluent
Chlorinated Screened Effluent Channel
Rex Concentrated Screenings
SWECO Concentrated Screenings
Combined Facility Concentrated Screenings
Number of Aliquots
Chemical Oxygen Demand (COD)—
The "scattergram" for the removal of COD as a function of hydraulic loading
rate is shown in Figure 35. The dispersion of results shown in this figure
approximately equals that of Figure 33 for BOD5.
Total
(mg/1)
109
112
135
119
77
195
112
145
8
Filtered
(mg/1)
82
95
94
85
74
—
84
--
8
74
-------
BAUER PERCENT RATED
HYDRAULIC LOADING
0 8 I
i
•
1
*
•
•
*
.
i
M
i
•
1
•
•
*
*
t
9
'
•
-100 -80 -60 -40-20 0 20 40 60 80
to
PERCENT REMOVAL EFFICIENCY FOR COD
(162) (147) (117)
(114)
•
REX PERCENT RATED
HYDRAUUC LOADING
j° 8 I
— •—
1
4
•
0
•
B
•
•
•
•
•
.
•
•
•
)0 -80 -60 -40-20 0 20 40 60 80
K>
PERCENT REMOVAL EFFICIENCY FOR COD
nit) •
SWECO PERCENT RATED
HYDRAULIC LOADING
° 8 I
»
u
«
* •
M
••
i
<
i
t
• •
•
•
•
•
X> -80 -60 -40 -20 0 20 40 60 80
PERCENT REMOVAL EFFICIENCY FOR COD
K)
FIGURE 35 - SCREENING EQUIPMENT REMOVAL EFFICIENCY
CHEMICAL OXYGEN DEMAND (COD) AS A
FUNCTION OF HYDRAULIC LOADING RATE
75
-------
r
Figure 36 is a "scattergram" showing the dispersion of COD loadings as a
function of the unit COD loading rate. This result is similar to Figure 34
in that there does not appear to be any discernable trend between treatment
efficiency and unit loading rate for COD as well as BOD5<
Table 20 represents the statistical summary of these results for COD removal
treatment efficiency.
Table 20. SCREENING EQUIPMENT REMOVAL EFFICIENCY
FOR CHEMICAL OXYGEN DEMAND
Percent Removal of Chemical Oxygen Demand
Arithmetic Mean @ 50%
Standard Deviation
Number of Observations
Bauer (
Hydrasieve
+6.5
25.6
28
Rex Rotary
Drum Screen
+1.5
27.3
25
SWECO
Centrifugal
Wastexrater
Concentrator
+12.9
29.2
28
These statistical results run contrary to the trend of values for the
They show the SWECO units have the potential for a higher removal efficiency
on this particular waste parameter, which is definitely contrary to the re-
sult inferred from the data for BODg. This result may be due to the inter-
action of the raw stormwater wastes and the short-method COD test used in
this testing program or some other undetermined cause. No data on the
filtered COD-versus total COD was taken in this program.
Total Phosphorus (Total P) —
Data taken for the screening equipment removal efficiency for total phosphor-
us are presented in Figures 37 and 38 as a function of hydraulic loading rate
and total P loading rate, respectively. The dispersion of results for this
parameter imply a differential removal from event to event and for hydraulic
loading.
Table 21 is a tabular presentation of the statistics summarizing the removal
efficiency data for total phosphorus.
The Bauer and Rex results for this parameter were sufficiently well-distri-
buted to show a statistically significant result at the 95 percent confidence
interval. However, the SWECO result did not pass the test for statistical
significance at 95 percent confidence interval. Additional data is required
for the SWECO corroborating the hypothesis these data in reality represent
two separate subsets of data, one for the SWECO screens intact and another
for the SWECO screens ruptured.
76
-------
1.0
I
05
m
-IOO -80
1.0
fig
iX
O c
§ 11.0.5
M
1.0
i
o
8J
8°
-60 -4O -20 0 20 4O 60
PERCENT REMOVAL EFFICIENCY FOR COD
-IOO -80 -60 -4O -20 0 2O 4O 60
PERCENT REMOVAL EFFICIENCY FOR COD
80
-IOO -80 -60 -4O -20 0 20 4O 60 80
PERCENT REMOVAL EFFICIENCY FOR COD
80
FIGURE 36 - SCREENING EQUIPMENT REMOVAL EFFICIENCY
FOR CHEMICAL OXYGEN DEMAND (COD) AS A
FUNCTION OF COD LOADING RATE
100
KX>
77
-------
I
u
cc
1.0
-•«-
-IOO -80 -60 -40 -20 0 20 40 60 80
PERCENT REMOVAL EFFICIENCY FOR P
100
I
1.0
3P
§2 0,5
1
JT
-IOO -8O -6O" -4O -20 0 20 40 60
PERCENT REMOVAL EFFICIENCY FOR P
80
IOO
HI
ll
<•>
5
$«
c,2
X
s
*~d
§~
I.O
0.5
§
t-
n
|
f
w
r»
•«
.
•»•
.
*
»<
•
™.
w -100 -8O -6O -4O -20 0 2O 4O 60 80
PERCENT REMOVAL EFFICIENCY FOR P
FIGURE 37 - SCREENING EQUIPMENT REMOVAL EFFICIENCY FOR TOTAL
PHOSPHORUS (TOTAL P) AS A FUNCTION OF HYDRAULIC
LOADING RATE
IOO
78
-------
BAUER PERCENT RATED
HYDRAULIC CAPACITY
0 8 I
9
•
«
."
•
•
e
•'
•
•
*
0
*
*
1 i
•
i
•
-------
Table 21. SCREENING EQUIPMENT REMOVAL EFFICIENCY
FOR TOTAL PHOSPHORUS (Total P)
Percent Removal of Total Phosphorus
Arithmetic Mean @ +50%
Standard Deviation
Number of Observations
Bauer ,
Hydrasieve
+6.9
23.5
24
Rex Rotary
Drum Screen
+2.0
27.8
23
SWECO
Centrifugal
Wastewater
Concentrator
+8.3
47.5
25
Ammonia Nitrogen (NH+-N)—
Data taken for ammonia nitrogen were not sufficient to warrant analysis by
dispersion analysis using a "scattergram". A mean value was determined for
the three screening methods studied and the appropriate statistical summary
parameters are given in Table 22. • .
Table 22.
SCREENING EQUIPMENT REMOVAL
EFFICIENCY FOR AMMONIA NITROGEN
Percent Removal of Ammonia Nitrogen
Arithmetic Mean @ 50%
Standard Deviation
Number of Cases
Bauer ,
Hydrasieve
-4.9
25.9
17
Rex Rotary
Drum Screen
-3.5
20.4
17
SWECO
Centrifugal
Wastewater
Concentrator
+15.3
26.4
17
Partly due to the few number of cases considered, ammonia nitrogen removal
efficiency data are difficult to interpret. The result for the Bauer and Rex
units is statistically'significant at the 95 percent confidence interval,
whereas the result for the SWECO does not come out as statistically signifi-
cant at this confidence level. Examination of the frequency of occurrence
analysis for the SWECO tends to indicate a bimodal distribution with the
modal category at +5 for one subset and +25 for the other. Whether this re-
sult is an inherent anomaly of the small number of cases or conversely an
indication' of the screen failure mode is unclear because these data are in-
sufficient to determine. However, it should be noted that the removal of
ammonia nitrogen by screening is not likely to be plausible grounds for sup-
porting the screen failure hypothesis because ammonia nitrogen is inherently
a soluble pollutant. This tends to discredit any inference that there is some
removal efficiency attributable to the physical process of screening.
80
-------
Fecal Coliform Group Indicator Bacteria—
This parameter was monitored briefly as a potential screening removal param-
eter but the results of an eight-week trial on this parameter did not indi-
cate any measurable removal of fecal coliform as measured by the membrane
filter method. Thereafter, the data-gathering for bacteriological sampling
was terminated as unproductive. Fecal coliform monitoring of other sampling
points in the treatment scheme continued for the duration of the program.
Screening Efficiency Net Overall Treatment Efficiency—
In summary, considering only the data resulting in a positive treatment
efficiency, Table 23 represents the net overall treatment efficiency summary
for the screening equipment evaluated.
Table 23. SCREENING EQUIPMENT NET OVERALL
TREATMENT EFFICIENCY IN PERCENT
(± 95% C.L.)
Suspended Solids
Biochemical Oxygen
Demand
Chemical Oxygen Demand
Total Phosphorus
Bauer
Hydrasieve ^-
15.7 + 19.0
13.1 + 18.4
13.6 + 16.2
13.3 + 16.4
Rex Rotary
Drum Screen
CWC
15.8 + 19.9 26.2 + 23.2
14.4 + 23.1
11.3+17.8
11.4 + 15.5
16.5 + 21.9
19.5 + 20.7
18.9 + 21.0
Operations, Maintenance and Special Considerations
From the original proposed Plan of Study for this work, the evaluation of the
functional value of the equipment installed was a principal objective of this
program. The Statement of the Work^ for the program specifically described
the scope of the study as:
6. "Determine the efficiency and effectiveness of the individual and
collective overflow treatment facilities with respect to pollution
abatement, cost effectiveness, reliability and other aspects of
evaluation..."
This section is intended to present those observations and results that fall
into the general category of "...effectiveness, reliability and other aspects
of evaluation...". These results specifically pertain to the Fort Wayne
Stormwater Facility and may be unique to this particular facility.
Stormwater Facility Conduits and Wetwell—
The Stormwater conduits constructed under this grant have been in continued
service throughout the duration of this program. Operations have been at
lower transit velocity than the design hydraulic capacity of these sewers,
81
-------
designed for 1990 service at five times the present day hydraulic loading.
This set of circumstances has contributed to the sedimentation of suspended
solids throughout the conduits and the wetwell.
Compounding this circumstance of future design loadings is the fact that the
wetwell has been partly full of wastewater since start-up in January, 1975.
The wetwell dewatering submersible sump pump was found to be unable to dewater
the wetwell and the conduits. Several events where the turbulence of flow in
the conduit churned up heavy septic, malodorous sediments were experienced in
the evaluation period. These sediments were sufficiently concentrated to
blind the mesh of the drum screen and the SWECO units. This usually backed
up in the flow distribution manifold to the head tank.
The water surface would increase in the head tank or the Rex headbox until it
exceeded the overflow discharge elevation, which then carried these raw, un-
treated flows directly into the screened effluent channel. These sediments
are thought to account for the sludge banks in the chlorine contact channel
shown in Figures 29 and 30, preceding.
Stormwater Pumps—
There were no exceptional operational difficulties experienced with either
the Johnston Model 24 PS (1.10 cumec or 25 MGD) or 30 PS (2.19 cumec or 50
MGD) Stormwater pumps. From calibrated flow measurement points installed in
the facility, these units performed adequately on a hydraulic basis through-
out the evaluation period.
Stormwater Distribution System—
This subsystem presents few operational problems except to note that the
overflow in the head tank passes untreated to the effluent channel, carrying
the screened wastewaters. The manually operated knife gate valves used in
the flow routing conduits as the flow throttling devices required little
maintenance and were sufficiently easily operated to adequately control the
equipment flows.
Bauer Hydrasievev-^--
The Bauer unit is inherently simple to operate because it has no moving parts.
However, the unit supplied at Fort Wayne experiences some hydraulic limita-
tions due to influent channel turbulence. A sudden enlargement hydraulic
jump phenomenon caused excessive loss of contaminated raw Stormwater to the
floor drain system by presenting a falling water curtain under the unit.
This hydraulic distribution problem was field-corrected by the manufacturer
after several months of operation.
The Bauer unit required so little in the way of routine maintenance after
each event that it gradually built up a layer of oily or greasy material on
the screen. This layer was gummy wet or dry and required cleaning by deter-
gent application and manual scrubbing from time to time.
82
-------
Adjustment of the flow distribution baffles at the top of the screen was pos-
sible for increasing the flow handling capacity. However, the unit hydraulic
loading was typically varied by the pumping condition and the desired flow
split from the head tank to all the screening units in service at the time.
Some type of flow control to each module based on an optimum hydraulic flux
rate per screen module would probably have improved the performance of this
unit in the end.
Some provision for the removal of semidry screenings directly for ultimate
disposal such as a solids collection hopper dumping into a containerized dis-
posal vessel would have been very desirable. These screenings are already
washed (like clean grit) and frequently could be raked off the Bauer screen
surface. Instead, operators were required to hose down the unit with plant
water and this created a lot of unnecessary splashing and mess.
The Bauer unit makes an excellent pretreatment device for the finer screens
such as the SWECO and the Rex. It removes solid debris such as wood pieces,
metals, plastic materials, leaves and twigs, and virtually any other coarse
solids. .These materials were shown to cause one screen failure after another
for the SWECO.
The Bauer screen modules were exclusively operated in parallel with each
other. Over the evaluation period, these units were responsible for treat-
ing 113.5 million gallons (430 million liters). The Bauer was always ready
to go into service and served as a type of "surge device" for distribution
of flows to the other units. It was operating a total of 276.1 hours for an
average duration of 3.2 hours during a total of 86 events.
No data were taken on the amount of screened effluent aeration that could be
attributed to the Bauer unit. However, like each of the other screens, the
effluent from this unit showed quantities of frothy suds coming out of the
effluent trough.
On the basis of floor area required, this unit takes up a space of 8.45 meter
by 7.85 meter (27'-9" by 25'-9"), or 66.3 square meters (714.6 square feet).
Since it is elevated above the grade level of the screening room, the floor
area beneath this unit is clear except for the support columns for the unit.
There are no utility services required for this unit except for cleanup after
an event. Headloss considerations are important for this method because the
nature of the design requires not less than about 2 meters (6.6 feet) of
static headless to operate.
Rex Rotary Drum Screen—
The Rex Rotary Drum Screen was operational in 50 events in the post-construc-
tion evaluation period (January, 1975 to February 6, 1976). The unit oper-
ated a total of 176.1 hours for an average running period of 3.0 hours. The
unit was inoperative for the first three months due to a bearing support
frame problem that was field-corrected in April, 1975. Thereafter, through-
out the duration of the evaluation period, this equipment was plagued with
mechanical problems.
83
-------
One of the most vexing problems with this specific unit was its tendency to
cease rotation under hydraulic loading. This problem occurred so frequently
and under such varied operational conditions that it was extremely difficult
to identify the precise reason for the problem. Sometimes the drum drive
would start to stall because the V-belt drive sheave was partly submerged.
However, at other times, under similar operating conditions, there were no
slippage problems.
Another source of slippage of the drum drive was identified as the increased
friction from the drum support roller wheels located at the open headbox end
and shown in Figure 12, preceding. These wheels are essential idler wheels
which transmit the drum weight to the support beam shown in the photograph.
A grease-lubricated roller-type bearing is provided in the hub of the wheel.
The roller bearing in one unit seized under load sometime after the tenth
month of operation and increased the rolling friction of the drum. This re-
sulted in wear to the roller surface, necessitating replacement, and was also
responsible for some slippage of the drive sheave.
At times, the drum would slow and slip without completely stalling. These
occasions were usually observed when high differential head between influent
and effluent flows occurred. This condition was also observed to be the re-
sult of a dirty drum screen. An overflow weir, shown previously in Figure 28,
was provided to relieve the hydraulic head that would result. Normal opera-
tional practice under such conditions was to throttle back the drum influent
flow to allow the automatic backwash system to clean the drum.
Problems with the backwash system itself were also experienced. The system
as installed used the screened effluent as a make-up source for the backwash
pump. However, the overflow relieves to the same channel as the effluent.
This condition resulted in plugging the pump strainer with debris at times or,
more frequently, plugging of the backwash manifold nozzles which had to be
manually cleaned. This problem was solved by putting the backwash pump make-
up onto the plant water system which was connected to the potable water
system through a break tank.
The Rex unit experienced one screen panel, failure in the evaluation period
resulting from a puncture or tear from a small wood block of scrap that had
become trapped in the drum. This unit required routine maintenance after an
event to clean up, inspect the backwash system, drive mechanism, etc.
The Rex drum screen also experienced two separate main bearing failures. The
main bearing for this unit serves as the main support for the drum on the
closed end as well as the thrust bearing to resist the hydraulic flow-induced
axial forces against the closed end. Both failures were attributed to exces-
sive deflection by the bearing support. This problem was corrected in the
field by the manufacturer.
On the plus side, the Rex drum screen takes up less floor space than either
of the other two screens. The power requirements to operate this unit are
modest by comparison x
-------
Figure 39 shows a photograph of the Rex Rotary Drum Screen operating on re-
circulated pond water. The screened effluent overflow point shown in this
photograph also represents the flow measurement weir for this unit, as cali-
brated in the post-construction evaluation flow measurement program.
The Rex unit at the Fort Wayne Stormwater Facility sits in a poured-in-place
concrete basin which takes up a floor area equivalent to 6.09 meters by 6.25
meters (20'-0" by 20'-6"), or 38.05 square meters (410 square feet). It re-
quires an electrical power supply for the drive motor and the backwash pump
motor plus controls, a clean source of backwash make-up .water, amounting to
378 liters per minute (100 gallons per minute) at a discharge of 45.8 meters
(150 feet) TDK. Manual cleaning of the screen wire fabric with scrub brush
and chemical cleaner may be necessary periodically to clean up the accumulated
greasy film that eventually builds up on the screen.
Figure 39 - Rex Rotary Drum Screen Operating
on Recirculated Pond Water
85
-------
The Rex unit can be operated at the least static lift headless among the
screens evaluated. A minimum headloss of about 0.75 meters (2.4 feet) is re-
quired for the unit at Fort Wayne.
Data presented elsewhere indicate a significant improvement in overall treat-
ment may be expected when a chemical flocculant was used in screening/flota-
tion processes. The chemical flocculant addition necessary to achieve the
removal efficiency improvements cited was 20 milligrams per liter ferric
chloride and 4 milligrams per liter cationic polyelectrolyte.5
These amounts would have required considerable quantities of these chemicals
to be added to the flows treated at Fort Wayne. For example, to add 20
milligrams per liter of ferric chloride to a flow of 17,500 gallons per
minute (66,200 liters per minute) would require the addition of 2.9 pounds
per minute (1.3 kg/minute), or the equivalent of one 135-pound (6.12 kilo-
gram) drum of anhydrous ferric chloride every 45 minutes. Since this mate-
rial is very hazardous to handle in either bulk liquid or anhydrous form, no
provisions were made to add chemical coagulants to the flows at Fort Wayne.
The Fort Wayne Facility combined sewer overflows are located at the Water
Pollution Control Plant (the 84-inch bypass) and the Glasgow Street inter-
ceptor regulator. Both sources are extensively used to store plant influent
flows under many operating conditions. The use of these conduits as storage
volume results in both flow and concentration damping of pollution surges
represented by storm-caused first flush conditions. In addition, there are
major industrial contributors on these interceptors, particularly the Glasgow
Street conduit. As a practical matter, some or all of the plant raw sewage
flow can be sent to the Stormwater Facility. The absence of the storm-caused
first flush overflow condition cited elsewhere is a direct resultant of these
factors. Consequently, addition of flocculant aids to these wastes probably
would not materially improve the performance of the screens, including the
Rex, based on past plant experience in treatment of primary influent with
coagulant aids.
SWECO CWC® Centrifugal Wastewater Concentrator—
The SWECO units operate on the basis of rotary fine mesh screening of a sheet
of raw wastewater hitting the collar screens. Particles suspended in the
flow are entrapped on the mesh of the screen while the screened fluid is
allowed to pass on through the screen and drops into a screened effluent
channel. The separated particles are washed down to produce the concentrated
screenings which are collected in a separate channel for conveyance to the
Water Pollution Control Plant for additional treatment.
The units installed at Fort Wayne were intended to be capable of automatic
operation with manual override feature. From the beginning of the evaluation
period, it was impossible to operate these units in the automatic mode with-
out causing problems with the flow distribution system because frequently all
eight CWC screening units began their backwash cycles simultaneously. This
reduced the flow treatment capability of the SWECO screens essentially to
zero for the duration of the backwash cycle.
86
-------
The automatic operation feature for these screens originally was triggered
individually by the level of concentrate in the screenings concentrate chan-
nel. Sometimes, the first flow of each unit experienced was so heavily laden
with heavy, greasy material that the units one by one would increase the
quantity of concentrate to such a level as to allow very little of the raw
stormwater to pass through the screens into the screened effluent channel.
With most of the influent flow from one or more screens going to the concen-
trate channel, the level in the channel would rise sometimes to overflowing
and all' units would sense the high level in the backed-up channel until the
sensor called for the influent valve to close automatically.
Another problem created by this simultaneous backwash demand of more than two
or three units at a time was the high peak short-term demand for compressed
air in the influent valve pneumatic operator circuits. The compressor reser-
voir would be exhausted and the pump could not catch up on its own. This
problem was corrected by adding additional reservoir capacity for the pneu-
matic air supply system.
The manufacturer's solution to the continuous backwash demand problem was to
rewire the electrical control panel to put a master cycle timer in the indi-
vidual backwash timer such that when a unit on one bank of CWC screens called
for backwash by timer, all other units were essentially "locked out" for cal-
ling for backwash until the backwash timer cycle on the unit being serviced
timed out. This arrangement limited the number of units in the automatic
mode calling for backwash at a time to one for each bank of CWC units. This
arrangement worked satisfactorily for the remainder of the duration of the
evaluation period as long as the units started with clean screens.
It became necessary at each event shutdown to ensure there was a clean CWC
ready for service at the next event start-up. To accomplish this, the unit
was operated in continuous cold water backwash all the time. This improved
the length of runs between backwash requiring hot water somewhat. The oper-
ating personnel were instructed to manually initiate a backwash cycle for one
unit at a time when it became obvious that the screened effluent had dimin-
ished and the concentrate had increased. At the termination of each event,
each CWC, whether operated or not, was manually put through a hot water back-
wash cycle three complete times as a part of clean-up standard operating pro-
cedure.
Frequently, the CWC units were operated with manual backwash cycles every
thirty minutes or so.
Another operational difficulty experienced with the SWECO screens was the
amount of concentrated screenings generated by the units. This problem was
aggravated by the fact that the Stormwater Facility concentrate drains were
located on the Lakeside interceptor, a combined sewer. This sewer was sub-
jected to stormwater surcharge just when the necessity of removal of con-
centrated screenings was greatest. The result was in effect to increase the
amount of head needed to "push" the screenings into the sewer or limit the
number of SWECO's on stream at a time. When this head requirement became
greatest, the concentrated screenings backed up the channels in the screening
room and overflowed the channels onto the floor. This result effectively
87
-------
reduced the weir flow measurements of concentrated screenings to nothing, and
the screening equipment evaluation required an assumed flow split of 15 per-
cent to concentrate, 85 percent to the effluent channel.
The most significant feature of operations and maintenance of the SWECO
screens stems from the screen failures experienced during the evaluation
period. There were 142 screen failures identified in the evaluation period,
in which the combined hours of operation of these units totaled 1,500 hours.
A failure of a single screen in a single CWC unit usually resulted in a
drastic increase in the effluent pollutant loadings for the remainder of the
screening event.
The computation of screen life is based on the principles of reliability
engineering. Reliability is defined as "the probability that a system or
device will perform without failure under given conditions for a specified
period of time". A computation of the Mean Time Between Failure (MTBF) was
performed for any one CWC unit, and the result is carried through the calcu-
lation of the expected useful life per screen; t = 30.5 hours.
This result is a direct consequence of the nature of the wastes and the man-
ner of distribution of the raw wastewater. For the first twelve months of
the evaluation period, the raw stormwater flow was distributed to the SWECO
screens directly from the head tank. This period represents the majority of
the screen life failures simply from the number of hours of operation. The
remaining six weeks in the evaluation period was the only period where the
Bauer Hydrasieve®was used as a prescreening device for the north bank of
SWECO screens. Screen life results for this period were not sufficient to
extrapolate a different useful life for this tandem operational mode. It is
expected that the failure mode for the north bank of units will shift to a
decidedly more normal curve distribution that approximates a classical wear
out failure rather than a chance failure mode. This result is expected to
extend the useful life of each screen to several hundred hours or more. How-
ever, it should be apparent that the south bank of units will probably con-
tinue in high chance failure mode because there is no provision for prescreen-
ing any of these units' influent flow.
The first few months of operation (without the continuous backwash spray in
service) resulted in heavily blinded screen panels from greasy deposits
thereon. These deposits adversely affected the hydraulic through-put of the
screens. It was necessary to manually scrub each screen panel (36 for each
CWC unit) with a diluted muriatic acid solution to restore the clean screen
porosity. This was recommended by the manufacturer prior to the installation
of the continuous backwash arrangement. After completion of this manual
cleaning and the continuous backwash system, manual scrubbing of these panels
was not required.
Figure 40 is a photograph of the SWECO effluent showing the sudsing action of
the screened stormwater. This sudsing came and went within events, and some
events did not show any sudsing action. As a general rule, once the sudsy
effluent left the Screening Building, whether chlorinated or not, the suds
disappeared when exposed to wind and weather.
88
-------
The eight SWECO screens installed at the Fort Wayne Storrawater Facility sit
overhanging the screened effluent channel. Counting the channel area re-
quired to carry the flows out of the Screening Building, these units as in-
stalled require an area of 5.5 meters by 17.8 meters (18'-0" by 58'-6"), or
97.7 square meters (1,052 square feet). This area can be oriented in several
ways. However, the overhead requirements of a SWECO installation require a
distribution channel capable of delivering the raw inflow at a recommended
static head of 2.5 to 3.0 meters (8 to 10 feet). Figure 41 shows one of the
two distribution manifolds for the SWECO units. This manifold is a pressure
conduit operating at 1.8 to 3.7 meter static head (6 to 12 feet) and de-
livering the flow to each of the units along its length in turn.
The SWECO units require more auxiliary services for operation than the other
screens. Each unit has a pneumatically operated 12-inch gate valve with
electrical solenoid valve actuators. Each unit also has a chemical feed pump
and a drive motor. An air compressor with receiver is required for the pneu-
matic valves' air supply. Some source of clean make-up water for backwashing
is required. At Fort Wayne, this was originally to be pond water but the
pumping requirements and solids plugging problems required the use of potable
water for backwash water make-up. A hot water heater for the screens' hot
water backwash is required.. Chemical cleaning solution concentrate is neces-
sary.
The necessity of all these auxiliary services for the SWECO required that the
Fort Wayne Stormwater Facility be kept heated during the cold weather months
to about 10 degrees Celsius (50 degrees Fahrenheit). At one time in January,
1975, the solenoid valves on several SWECO screens froze from moisture in the
air line and ruined the actuators. Heating the building prevented future
occurrences of this.
Stormwater Facility Chlorination—
The Stormwater Facility was designed with the capability of chlorinating at
several points in the treatment scheme. As shown in Figure 5, preceding,
it was possible to prechlorinate combined sewer overflows diverted to the
Stormwater Facility by opening the valve at the 84-inch bypass structure
or alternatively, the chlorination line drain valve at the facility wetwell.
The practice of prechlorinating the Stormwater overflows was discontinued
after several operational personnel assigned to the Stormwater Facility com-
plained of "tear gas" - like fumes in the screening room when either of the
prechlorination points were in service.
The post-chlorination point was operated for 18 events to assess the impact
of storm flow chlorination on the terminal pond. There are not sufficient
data to substantiate the effect of chlorination on the pond one way or
another. However, the pond effluent data taken since pond start-up in
October, 1971, tends to confirm that there is no adverse effect of chlorina-
tion of pond influent.
89
-------
Figure 40.- SWECO CWCV-' Screens in Operation
on Raw Stormwater Overflows Show-
ing Effluent Sudsing Effect
Figure 41 - SWECO Influent Distribution Manifold
for South Bank of CWC Units
90
-------
Chlorine demand for the post-chlorination point :of application was highly
variable, depending on the wastewater. A minimum chlorination rate of 6.0
kilograms per million liters (50 pounds per million gallons) was observed
during an event in June, 1975. The maximum rate of chlorine application was
21.1 kilograms per million liters (176 pounds per million gallons). The mean
rate was 13.3 kilograms per million liters (111.22 pounds per million gal-
lons), standard deviation of 6.79 kilograms per million liters (56.69 pounds
per million gallons).
Effectiveness of the Stormwater Facility in Dealing with Combined Sewer
Overflows—
In the pre-construction program, it was calculated that 1,040 million gallons
(3,935 thousand cubic meters) bypassed any sort of treatment by overflowing
the 84-inch bypass sewer in at least 58 separate storm-induced combined sewer
overflow events. This condition represents the baseline for comparison of
the effect of the Stormwater Facility on the combined sewer overflow bypass
diversions to the Maumee River. In the post-construction evaluation period,
there were eleven events identified where some raw, untreated storm flow was
relieved hydraulically by overflowing the Stormwater Facility wetwell. A
total amount of bypassed Stormwater overflow was estimated at 35.1 million
gallons (133 thousand cubic meters) for this period.
The reasons for these bypasses varied. There were seven events where the
duration of the bypass could have exceeded 60 minutes. The longest observed
bypass came during Event 192 on September 5, 1975, when the two Stormwater
pumps simply could not keep up with the wetwell inflow resulting from a
1.08-inch (2,74 centimeter) rainfall. There were six other times where the
bypass could have lasted as long as 4 hours. Because time required to man
the Stormwater Facility and start up the raw Stormwater pumps averaged less
than one hour, the average duration of bypass was limited to 1.6 hours when
there were problems with getting the facility going. There were four events
where the time of bypassing was 30 minutes or less.
On the basis of the decrease in the frequency of diversion of storm-caused
combined sewer overflows at the Water Pollution Control Plant alone, the
construction of the Stormwater Facility was highly'successful in controlling
overflows. Furthermore, it should be recognized that the 84-inch bypass
depth indicator at the WPCP recorded 117 separate diversions during the post-
construction evaluation period. It is concluded that the construction pro-
gram reduced the frequency of overflows to one-tenth of the number of poten-
tial overflows.
On the basis of the estimated quantity of raw combined sewer overflows di-
verted to the river, the Stormwater Facility was shown to have reduced the
potential overflow of 450.9 million gallons (1,707 thousand cubic meters) to
an estimated 35.1 million gallons (133 thousand cubic meters), an 82 percent
reduction. In the original design concept, outlined in the Master Plan for
Sewers, Part if it was estimated that the diversion of only 50 percent of the
combined sewer overflow quantities received at the plant would result in a
10 percent reduction of the organic loading going to the stream. In this
91
-------
program, the reduction in frequency and quantity of flow bypassed amounted to
a net estimated reduction of organic loading of 10,300 pounds of BOD5 (4,700
kg of BOD,-) to the stream, a percentage reduction of 54 percent from this
source alone.
Stormwater Effects on Terminal Pond
One of the objectives of this program was to assess the overall effect of
storm-caused combined sewer overflows on the Water Pollution Control Plant
terminal pond effluent. Figure 42 is a flow schematic diagram for the treat-
ment plant systems showing the origins of flows from the various sources.
The pond receives daily flows from the Water Pollution Control Plant on the
south side of the river. These flows, designated QWPCP» are ma^e UP °f t^6
treated secondary effluent and the remaining primary effluent that is un-
treated by the aeration tanks due to hydraulic flow limitations. These flows
are combined and chlorinated in the plant effluent diversion chamber at the
effluent chamber of the secondary settling tanks on the south side of the
river. By means of a 72-inch river siphon, these flows are conveyed to the
influent of the chlorine contact tank at the east end of Terminal Pond No. 1. •
The effluent from the chlorine contact tank flows through a 72-inch conduit
to the pond headworks. A drop inlet for the Stormwater Facility tees into
this pipe at the end of the screened Stormwater effluent channel just before
this conduit enters the terminal pond. This Stormwater Treatment Facility
effluent flow, already screened and chlorinated, is designated as Qstm on the
flow diagram. As can be seen, in the flow diagram, -the terminal pond influ-
ent is made up of 'the screening facility effluent plus the Water Pollution
Control Plant effluent
Once the combined Stormwater Facility effluent plus WPCP effluent entered the
terminal pond, the plant operating practice was to monitor the pond effluent
by 24-hour composite sampling at the WPCP for most of the pollutant parameters.
Bacteriological samples, dissolved oxygen samples and temperature were taken
once a day by plant operators on a grab sample taken in the mid-morning.
These routine plant operating data were used in the evaluation of the effect
of combined sewer overflow screening facility effluent on the pond perform-
ance. An analysis of terminal pond flow and detention time was a.n integral
part of this analysis.
The starting point for the sizing of the Stormwater Facility was in the
Master Plan for Sewers, Part l3, which discussed the necessity of stormwater-
caused combined sewer overflows, the benefits to be derived from construction
of relief sewers and centralized treatment facility, and the means for treat-
ment. The goals of the facilities, to be constructed for the treatment of
combined sewer overflows were to reduce the overall stream BOD,, loading by
10 percent. This goal was to be accomplished by diverting the maijor part of
the combined sewer overflow raw untreated wastewater to a stormwaiter treat-
ment facility. The flow that was anticipated for treatment in such a facil-
ity was estimated at 600,000 gpm ultimately. The initial construction phase
(undertaken under this project) was sized to handle 52,500 gpm or 75 MGD
(3.29 cubic meters per second). This quantity of flow was to be obtained
from the 84-inch bypass sewer overflow at the Water Pollution Control Plant.
92
-------
POND N9I-,
EFFLUENT /
TERMINAL POND N2
167.4 ACRE-FEET AT 5*
MEAN DEPTH
•QPond
STORMWATERv
FACILITY 7
Q
STORMWATER FACILITY-7
TREATED EFFLUENT /
QSt
""wpujJ-^
+Qstm \
m-
"rT
t- WETWELL
OVERFLOW
*-Qstmo
-TWIN 96" RIVER
CROSSING CONDUITS
CHLORINE
CONTACT
TANK
Qwpcp-
72 RIVER CROSSING
SIPHON
-84 BYPASS
-GLASGOW INTERCEPTOR OVERFLOW
OVERFLOW
WATER POLLUTION
CONTROL PLANT
Qsec-
FIGURE 42 - STORMWATER TREATMENT FACILITY
FLOW ROUTING SCHEMATIC DIAGRAM
93
-------
It was reasoned that to divert this overflow to primary treatment and deten-
tion in a holding basin would appreciably decrease the amount of raw un-
treated overflow passing to the river.
One of the key objectives of the program was to determine the shock and total
pollutional loading to the river for each significant rainfall event from both
treated (i.e., post-construction) and untreated (i.e., pre-construction)
sources that were part of this project. To assess the shock loading of treat-
ment provided by the Stormwater Facility and the terminal pond on these over-
flow quantities, a longitudinal hydrographic analysis was made of several
Stormwater treatment events. Figure 43 is the hydraulic mass flow hydrograph
for the longitudinal flow through the treatment, process drawn in the sche-
matic flow diagram. From this diagram, it should be apparent that the hydrau-
lic flow response for this event indicates that an accumulation of 0.15 inch
(0.38 centimeter) of precipitation was required to trigger the 84-inch bypass
diversion to the Stormwater Facility. This accumulation required about 2
hours to accumulate at this precipitation rate for this event.
As the stormwater-induced flow was received at the Water Pollution. Control
Plant, the raw sewage pumping rate was increased to keep up with the rising
interceptor levels. It took an additional 30 minutes to accumulate enough in
the interceptor sewers to start filling the Stormwater Facility wetwell.
Ninety minutes later, the wetwell water surface reached a depth of 26.0 feet
(7.92 meters). The 50 MGD pump was started and began pumping down the wet-
well. At this moment, the Stormwater Facility began operations as a treatment
process with these flows going to the screening units and then to the terminal
pond.
As can be verified from Figure 43, the pond inflow, as represented by the
summed flows of the Water Pollution Control Plant effluent, Q^pgpji plus the
Stormwater Facility flow, Qstm» shows a peak flow rate of 77,000 gpm (292,000
liters per minute) at approximately 30 minutes after the 50 MGD Stormwater
pump began operating. This flow corresponds to a 100 percent increase over
the instantaneous raw sewage pumping rate of the plant alone. In other words,
the Stormwater pump has essentially doubled the instantaneous peak flow rate
to the pond by taking the diverted raw sewage overflow through the screening
facility. This amount of raw, untreated wastes would have been bypassed to
the river directly starting with the 84-inch bypass flow at 2:00 a.m. if the
Stormwater Facility had never been built. Furthermore, this flow might have
began overflowing to the river at the point where the overflow elevation for
the wetwell was exceeded at about 4:00 a.m. if the Stormwater pumps had not
been operated.
Further examination of Figure 43 reveals that the terminal pond itself had a
noticeable effect on the hydraulic flow variable associated with this event.
The pond began receiving the Stormwater Facility effluent at 4:00 a.m. The
peak inflow already described arrived shortly thereafter. The pond volume
essentially absorbed this step input flow variable, dampened the peak flow by
its available volume, and allowed this flow to pass through gradually. The
peak flow for the pond effluent is shown to be at 8:00 a.m. on the 15th. This
peak flow was computed to be 50,300 gpm (190,000 1pm) representing a 28 per-
cent increase over the instantaneous raw sewage pumping rate at that time.
94
-------
o
o
o
o
2 o
§*
z
a o
8
VI
o
o
o
i
§
06
3 3
<• •». « 9
o o o o
CUMULATIVE RAINFALL
(INCHES)
°do - • * ~ -
_•
-------
I
STORMWATER FACILITY WETWELL
DEPTH INDICATOR IN FEET
CjSSSicJcviSS^^tiBinatoiM-O
S°FPEETtNT INTERCEPT°R
DISCHARGE RATE {6PM«IO«)
FIGURE 43 - RAINFALL HYDROGRAPH AND LONGITUDINAL
HYDRAULIC FLOW IN MGD - PART 2
-------
Furthermore, 7.1 million gallons (27 million liters) were the volume of storm-
water treated and pumped which might otherwise have passed on to the river un-
treated. Starting at 2:00 a.m., this flow was protracted in its discharge
rather than peaking at about 4:00 a.m.
This result is judged to be a significant benefit in and of itself because it
shows that the Stormwater Facility plus the treatment pond can essentially
reduce the short-term peak flow to the stream from the 84-inch bypass by as
much as 55,000 gpm (208,000 1pm) (maximum discharge capacity for the 84-inch
bypass). This would have amounted to about 5 percent of the average daily
stream flow for this date. Instead of receiving the raw stormwater bypass
directly, the treatment pond reduced this peak loading to 10,900 gpm (41,000
1pm) coming at a time six hours later when most of the rest of the short-term
peak flows had already passed.
From observations of the appearance of the terminal pond effluent and those
sampling data that are available, there was no evidence of pond treatment
short-circuiting or other hydraulic impairment of pond treatment efficiency
attributed to the inflow of screened combined sewer overflows. Wind and wave
action were observed to muddy the pond contents on occasion. However, there
did not appear to be any correlation with stormwater overflow pumping and pond
effluent turbidity.
In addition to the short-term hydraulic flow response given in Figure 43, a
long-term plant raw sewage flow analysis was performed on the plant operating
data available for the period from January, 1970 to February, 1976 as shown
in Figure 44. Plotted in this figure are the mean, minimum and maximum
monthly raw sewage pumping data available from plant operating records. Note
that the post-construction period flows do not appear to be significantly
different from the long-term trends recorded for the plant from the period
preceding start-up of the Stormwater Facility in January, 1975. This informa-
tion is not considered significant except to note that the peak flow days for
1975 plotted on this figure exceed 50 MGD (2.10 cumec) for eleven of the
twelve months in this period. In the preceding three years, there were only
eleven days when the plant raw sewage pumping rate exceeded this amount.
This higher frequency of maximum 24-hour raw sewage pumping rate is a conse-
quence of the construction of the Stormwater Facility because the plant inter-
ceptors must fill to a depth showing at 3.5 feet (1.07 meters) on the plant
bypass overflow depth recorder to divert to the Stormwater Facility. This
has improved plant operations from the standpoint of treatment of more raw
wastes because the operators are much more conscious of water depth levels in
the plant interceptors. Formerly, if the overflow elevation was exceeded
briefly, an amount of combined sewer raw untreated wastes was bypassed without
much attention. After completion of the facility, the operators are required
to alert the plant superintendent that some flow was diverted to the Storm-
water Facility which requires some form of pumping and clean-up at a minimum.
The resultant has been to use the available storage in the interceptors up to
3.5 feet (1.07 meters) depth on the indicator, increase the plant raw sewage
pumping rate to maintain less than 3.5 feet (1.07 meters) until the maximum
rate is reached, and continue to pump up to a depth of 6.0 feet (1.83 meters)
where the Stormwater Facility wetwell overflow is exceeded. This has usually
97
-------
r
CD
2
3
U.
UJ
I
UJ
CO
I
DC
Q.
O
Q_
PRECONSTRUCTION
EVALUATION
Note: 1 MGD = 0.0438 cumec
POST-CONSTRUCTION
EVALUATION
FIGURE 44 - LONG-TEEM PLANT RAW SEWAGE PUMPING
DAILY HYDRAULIC FLOW IN MGD
98
-------
resulted in much less frequent raw sewage overflows because the time of the
peak flow depth at the WPCP has been greatly lengthened by this interceptor
storage strategy.
Dissolved Oxygen—
Since the treatment pond was shown to be able to absorb the inflow of diverted
screened combined sewer overflow without apparent short-circuiting, the rou-
tine daily pond effluent sampling program results were analyzed for evidence
of pond performance deficiencies that may have resulted from the addition of
combined sewer overflow pumped and screened at the Stormwater Facility. The
terminal pond effluent data were separated into wet weather and dry weather
periods for both the pre-construction and post-construction periods. The wet
weather days were identified as those days where the plant 84-inch bypass
diversion recorded a diversion of raw sewage through this bypass conduit for
both the pre- and post-construction phases. When the post-construction data
were analyzed, the day following such an event was also included in the wet
weather category.
Table 24: is the presentation of the best-fit normal distribution on the data
available from daily routine plant monitoring records. As can be verified by
examination of these data, there is no perceptible effect on the pond effluent
dissolved oxygen that can be attributed to the addition of storm-caused
screened overflows to the terminal pond.
Table 24. WATER POLLUTION CONTROL PLANT TERMINAL
POND EFFLUENT DISSOLVED OXYGEN
Arithmetic Mean Pond
Dissolved Oxygen (mg/1)
Standard Deviation (mg/1)
Number of Observations
Range of Values Observed (mg/1)
Pre-Construction
Evaluation Phase
All Days
4.06
0.81
375
1.8
to
9.7
Post-Construction
Evaluation Phase
Wet Weather
Days Only
All
3.80
1.30
136
1.5
to
9.2
3.55
1.35
332
1.5
to
15.8
Another aspect of the operation of the terminal pond is the effect of the
stormwater on the maximum and minimum values of the pond effluent dissolved
oxygen. In one aspect, Table 24 represents the condition where the pond
effluent variation is shown probabilistically. Examination of the pond efflu-
ent dissolved oxygen over a longer term lends a slightly different view to
this trend.
99
-------
Figure 45 shows the long-term WPCP plant terminal pond effluent D.O. variation
going back to 1970. When the terminal pond went into operation in October,
1971, the pond originally responded gradually by showing an increasing trend
of effluent D.O. starting in May, 1972, peaking at a mean value of 8.7 mg/1
in December, 1972. This rising mean value was accompanied by consistently
rising maximum D.O. levels. Minimum D.O. samples analyzed follow the general
upward mean trend although the range of variation possible for these minimum
values is less than for either the mean or the maximum.
After this rapid rise from a mean D.O. of 2.8 mg/1 in May, 1972 to a mean
D.O. of 8.7 mg/1 in December, 1972, there followed an equally rapid decline
in pond D.O. levels to a mean D.O. low of 3.7 mg/1 in July, 1973. Thereafter,
there appear to be seasonal rises and falls within the range of mean D.O.
values of 2.1 mg/1 to 5.2 mg/1 through 1975 /
Part of the significance of the dramatic rise in pond D.O. for the first year
of operation as a plant effluent terminal pond lies in the likely algae
"bloom" of the summer and autumn of 1972. This bloom of oxygen-producing
algae could be partly responsible for the rise in pond effluent D.O. Indeed,
examination of the pond effluent suspended solids data for the same time
period tends to support this conclusion by exhibiting a seasonal increase in
effluent suspended solids in the warm weather months of July through Septem-
ber or sometimes October. Peak suspended solids values during this period
tend to also indicate an increase in some days' pond suspended solids data.
On the other hand, the pond suspended solids data also show high mean values
outside of the warm weather period. In fact, the highest monthly mean sus-
pended solids data for the pond effluent came in April, 1972, March, 1973,
and February, 1974. Corresponding peak suspended solids naturally followed
the mean monthly trend. Minimum values also were consistent with the mean
monthly trend.
Let it suffice to say that the addition of Stormwater Facility effluent to
the pond did not appear to adversely affect the range or mean values of the
pond dissolved oxygen, except to note that the lowest monthly mean value for
pond dissolved oxygen occurred in October, 1975 at 2.1 milligrams; per liter.
In any case, the monthly mean pond effluent D.O. cannot be said to have caused
the increase in average values for stream D.O. nor the demonstrated decrease
in numbers of days of stream D.O. less than 4.0 milligrams per liter.
Biochemical Oxygen Demand—
One would expect the terminal pond biochemical oxygen demand (BOD5> values to
show essentially the same trend as the dissolved oxygen data. The mean pond
D.O. data cited do not show any significant variation in the D.O. concentra-
tions measured in the pre-construction data as compared to the post-construc-
tion evaluation period (See Table 24).
Table 25 displays the BOD5 data results for the plant terminal pond before
the introduction of Stormwater and after the introduction of Stormwater.
Figure 46 gives the normal curve display of these data for comparison with
the tabulated results.
100
-------
ui
o
x
o
Q
LU
C/)
Q
LU
h-
2
<
CL
1973 1974
CONSTRUCTION
PRECONSTRUCTION^-
EVALUATION
POST-CONSTRUCTION^-
EVALUATION
FIGURE 45 - WATER POLLUTION CONTROL PLANT EFFLUENT DISSOLVED
OXYGEN LONG-TERM MEAN MONTHLY D.O. CONCENTRATION
101
-------
Table 25. PLANT TERMINAL .POND EFFLUENT
BIOCHEMICAL OXYGEN' DEMAND
Pre-Construction
Evaluation Phase
Arithmetic Mean Biochemical
Oxygen Demand (mg/1) 13.5
Standard Deviation (mg/1) 7.5
Number of Observations 463
Range of Values Observed (mg/1) 1 to 52
Post-Construction
Evaluation Phase
11.5
4.5
392
1 to 28
While there is no measurable difference statistically between the arithmetic
mean values cited in Table 25, it should be noted that the two distributions
do not appear to fall on the same line. The two standard deviations noted
in Table 25, approximated from the normal distribution curve, show different
trends. The explanation of this result is not apparent from examination of
the data for the two periods^ cited. Figure 46 is a graphical presentation
of these two probability distributions for the terminal pond effluent.
The degree of difference between the two distributions is decidedly and
noticeably different. Each distribution is very close to an ideal probability
distribution as determined from the raw data (not shown). It is interesting
to note that there is such a significant difference between the two distribu-
tions shown on the basis of concentration alone.
Upon examination of the terminal pond effluent data for a long-term period of
time, a very interesting trend is apparent. Figure 47 shows the long-term
plant effluent terminal pond data from January, 1970 through February, 1976.
In this six-year period, the effluent BOD5 is shown to vary greatly from
month to month in the range of mean monthly values from 5 to 36 milligrams
per liter. Since the pond start-up in October, 1971, the effluent mean BOD5
has tended to peak in the cold weather months and then suddenly drop at the
start of warmer spring weather to a minimum pond BOD5 value. This tendency
is very interesting because it tends to confirm the trend of temperature-
dependent biological kinetics for oxidation ponds for the entire period of
time prior to the post-construction evaluation period starting in January,
1975.
After January, 1975, note that the trend of terminal pond effluent BODS data
tends to remain more stabilized in-effluent quality in terms of mean, maximum
and minimum values of effluent 6005. These results might be attributed to the
addition of large volumes of warmer storm flow to the pond in cold weather,
tending to keep the pond temperature higher on the average. However, examina-
tion of the pond temperature data available (taken at the effluent structure)
does not tend to support or discredit this explanation.
102
-------
8 S
TERMINAL POND EFFLUENT
FIGURE 46 - PLANT TERMINAL POND EFFLUENT
BIOCHEMICAL OXYGEN DEMAND
103
-------
o
m
H
z
UJ
no
100
90
80
70
60
Ul 50
LJ
Z 40
O.
30
20
10
1970
a.
cc
o
a.
N
g>
-H-
o
o
I
1971
1972
/
1973 1974
CONSTRUCTION
PRECONSTRUCTION-
EVALUATION
POST-CONSTRUCTION-
EVALUATION
FIGUEE 47 - WATER POLLUTION CONTROL PLANT EFFLUENT
BIOCHEMICAL OXYGEN DEMAND (BOD5)
LONG-TERM MEAN MONTHLY BOD5 CONCENTRATION
104
-------
The more universal trend of increase in the mean pond effluent BOD,, during
the warm weather months does not seem deserving of an explanation Because the
post-construction trend seems consistent with the preceding years. However,
the lesser range of pond effluent values in 1975 deserves more comment. In'
previous years, the mean effluent values began an upward trend in July, August
and September. The corresponding maximum effluent BOD. values appeared to
come in this same time period. In contrast to this rising mean, greater var-
iation trend, the 1975 data show a rising mean, steady variation trend in
terms of maximum effluent BOD-.
Table 26 below shows the relationship between the pre-construction and post-
construction phase plant secondary treatment effluent BOD,, for the same time
periods. ->
Table 26. PLANT SECONDARY TREATMENT EFFLUENT
BIOCHEMICAL OXYGEN DEMAND
Arithmetic Mean Biochemical
Oxygen Demand (mg/1)
Standard Deviation (mg/1)
Number of Observations
Range of Values Observed (mg/1)
Pre-Cons truct ion
Evaluation Phase
15.33
9.16
483
1 to 47
Post-Construction
Evaluation Phase
29.48
20.15
399
0 to 102
From the^comparison of the two sets of data, it is evident that the terminal
pond efficiencies of removal based on BOD5 have improved measurably since the
completion of the Stormwater Facility. There are two explanations that come
to mind. First, the plant flow has increased over this period somewhat with-
out a corresponding increase in plant capacity. As the plant flow increased,
it follows that the secondary treatment effluent concentration should in-
crease somewhat because there has not been a corresponding increase in secon-
dary treatment efficiency over the same time span. In fact, the plant secon-
dary effluent in 1975-76 was substantially higher than the 1971-72 data.
This fact tends to support the contention that the pond efficiency has im-
proved since the addition of stormwater.
On the other hand, the quantity of flow treated by primary treatment.and by-
passing secondary treatment definitely increased as a consequence of the con-
struction of the Stormwater Facility. In the fourteen-month period of the
pre-construction evaluation, there were 1,954 million gallons (7,400,000
cubic meters) total secondary bypass flow sent to the terminal ponds. In
the thirteen-month post-construction evaluation period, there were 3,180
million gallons (12,040,000 cubic meters) of primary effluent that did not go
through secondary treatment at the Water Pollution Control Plant. In other
words, the construction of the Stormwater Facility has prompted an increase in
the amount of raw wastewater treated at the Water Pollution Control Plant to
the limit of the treatment capacity available.
105
-------
This contention is also supported by the data obtained in this program. The
plant secondary bypass flow is combined with secondary treatment effluent be-
fore sampling at the Water Pollution Control Plant. This means the additional
organic loading in the primary treatment effluent bypassing conventional
secondary treatment is integral to the composite samples obtained throughout
this program.
In spite of the much higher pond influent organic loading in the post-con-
struction period when compared to the pre-construction period, the terminal
pond effluent was shown to be quite stable even with the addition of raw,
chlorinated stormwater. This result was unexpected. Several hypotheses
were examined in detail to attempt to explain the apparent higher treatment
efficiency for the ponds when loaded with higher organic loadings.
The result was first thought to be a consequence of the slightly shorter aver-
age pond hydraulic detention period resulting from the addition of large
amounts of stormwater. Event 186 is a typical example of a relatively long
duration storm event (See Figure 43). Several other events were shown to have
a short-term increase in flow up to 50 percent more storm-caused tlow than the
Water Pollution Control Plant effluent flow. It was hypothesized that this
higher short-term flow would tend to shorten the pond hydraulic detention time
sufficiently to result in wash-out of the motile forms of algae and higher
forms of aquatic plants present. At the same time, the greater soluble load-
ing introduced in the ponds was stimulating the growth of bacterial forms
which tended to settle out in layers of sediments.
Examination of the pond effluent for biological classification of microbio-
logical organisms was not performed because of project time and budget con-
straints. Nevertheless, results obtained from physical and chemical analyses
of pond benthic deposits and effluent wastewater samples tend to support the
contention that algae and other motile forms were prevented from taking over
the pond by the short hydraulic detention period available. Volatile solids
content of pond benthic deposits showed these solids had physical and chemical
characteristics to primary settled raw sludge. This tends to confirm the hy-
pothesis that microbiological activity is responsible for the organic pollu-
tant removals observed for the ponds.
The result of the apparent beneficial effects'of the addition of raw screened
stormwater overflow was so intriguing to the investigators•that a more de-
tailed examination of the pond loadings seemed advisable. At.first it was
hypothesized that there may' be some kind of toxicity effect at work in the
pond that tended to suppress the effluent BOD5 concentration. This hypothe-
sis was advanced after the examination of the long-term mean month pond
effluent COD trends shown in Figure 50 following.
Since the data base from which these data were taken was shown to be con-
sistent over time, an alternative hypothesis stemming from the examination of
the pond effluent suspended solids was advanced. In this alternate hypothe-
sis, the pond was supposed to be responding to the Stormwater Facility BOD5
load resulting from the screened effluent BOD5 remaining plus the screening
106
-------
equipment overflow points raw untreated BOD- and suspended solids load. How-
ever, examination of the long-term terminal pond suspended solids concentra-
tions shows a similar trend for this pollutant parameter.
The investigators decided that an examination of the long-term terminal pond
organic and suspended solids loadings was necessary to further attempts to
explain these results. Plotted in Figure 48 are the data for the terminal
pond mean monthly organic loading rate in terms of pounds BOD- per acre per
day. These data were correlated with organic removal efficiency and plotted
in a bivariate manner using linear paper in a "scattergram", shown in Figure
49.
As can be verified from Figure 48, it appears as though there is a limiting
minimum organic loading that can be applied to the pond below which the treat-
ment efficiency as measured by BOD- is adversely affected. After the comple-
tion of the Stormwater Facility, BOD,- removal efficiency averaged 75 percent
(cr = 4.83 percent). In the preceding 39 months of operation, the mean monthly
organic removal efficiency was 47 percent (cr = 42.3 percent). Figure 49 shows
the bivariate analysis of the pond treatment efficiency as a function of total
pond organic loading. Note that there are no negative treatment efficiency
results reported for loading, rates greater than 200 pounds BOD^ per acre per
day (224 kg BOD,- per hectare per day). For loading rates in the range of 100
to 199 pounds BOD,- per acre per day (112 to 223 kg BOD- per hectare per day),
there are an equal number (7) of monthly mean values greater than 50 percent
than less than 50 percent (also 7). For loading rates greater than 200
pounds BOD,, per acre per day (224 kg BOD,, per hectare per day), there were
only two months with less than 50 percent mean monthly removal efficiency.
None of the post-construction evaluation period months (all of 1975 and the
first six weeks of 1976) had organic loading treatment efficiencies less than
60 percent.
These results tend to confirm the trend of values shown in the long-term
terminal pond effluent concentration (Figure 47).
Chemical Oxygen Demand—
Plant operators have been taking chemical oxygen demand data on plant influent
and effluent flows for several years. Monthly mean COD results are plotted
for the plant effluent in Figure 50 from January, 1970 to the end of 1975.
These results are included for comparison with the data for the plant effluent
BOD5 (Figure 47).
It should be apparent from a glance at each of Figures 47 and 50 that there
appears to be some differences in the trend of values for the monthly means.
Also, the much greater variation between minima and maxima in the two displays
of data tends to obscure any trend that may be observed.
Water Pollution Control Plant COD data are based on the so-called short
method procedure, rather than the reflux method. The difference between the
BOD- and COD data may be due to some organic residuum in the terminal efflu-
ent measurable by the COD test method used that does not contribute to five-
day biochemical oxygen demand. Whatever this difference is, it must be noted
107
-------
1100
1000
900
800
700
600
o
TJ
o> 500
u
o
V.
t/i
df
o
m
400
3OO
200
IOO
TERMINAL POND REMOVAL
EFFICIENCY
TERMINAL POND INFLUENT
TERMINAL POND EFFLUENT
PRECONSTRUCTION
EVALUATION
POST-CONSTRUCTION
EVALUATION
Note: Ibs./acre/day x 1.12 = ke/ha/dav
FIGURE 48 - LONG-TEEM TERMINAL POND MEAN MONTHLY
ORGANIC LOADING RATE
100
-100
108
-------
2!
IUU
90
ap
70
60
50
3 30
§ "°
(0
2^ 0
5
uj HO
cf "^
m -so
g -4O
-50
-60
-TO
-80
-90
HOO
,
•
q
«
J
•
«
"
*
^1
i
i|
ft
i
X)
9
«
«
e
B
4C
«
<
*
•
K)
LBS
i
«•
*
r~ coo
i.BODc/A(
,
,
9
*
•
1 800
;RE/ DAY
@
-------
220
PRECONSTRUCTION
EVALUATION
POST-CONSTRUCTION *
EVALUATION
FIGURE 50 - TERMINAL POND EFFLUENT CHEMICAL OXYGEN DEMAND
LONG-TERM MEAN MONTHLY CONCENTRATION
110
-------
that the 1975 mean trend values for COD are generally higher than the years
preceding. Whether this increase in trend mean values can be attributed to
the addition of stormwater to the terminal pond remains to be seen.
Suspended Solids (Nonfilterable Residue)—
Terminal pond effluent suspended solids appear to follow a similar pattern to
BOD5< The rises and falls in the monthly mean effluent suspended solids
correspond to the similar peaks and valleys for BOD,.. The range of values
observed in the time prior to the 1975 post-construction evaluation also seem
to show a cold weather peak range in mid-winter and another, lesser peak in
the warmer mid-summer months.
Figure'51 represents the long-term trend of mean monthly suspended solids
data obtained for the terminal pond effluent. A similar trend to that de-
scribed for BOD,, exists for the post-construction evaluation period of 1975-
1976 in terms of the ranges of monthly maxima and minima experienced. Also,
the amount of fluctuation from monthly mean to monthly mean is shown to be
much less for the post-construction period than for any time prior to this
period. Furthermore, the range of variation of the mean monthly suspended
solids measured is shown to be substantially reduced.
Table 27 is an attempt to quantify these observations of the data. The
statistics chosen for these data are based on arithmetic means and standard
deviations because the normal curve distribution was shown to provide the
best fit for the data.
Table 27. PLANT TERMINAL POND EFFLUENT SUSPENDED
SOLIDS (NONFILTERABLE RESIDUE)
Arithmetic Mean Suspended
Solids (mg/1)
Standard Deviation (mg/1)
Pre-Construction
Evaluation Phase
20.
31.
Po s t-Cons t rue tion
Evaluation Phase
11.5
11.0
Number of Observations
Range of Values Observed
(mg/1)
455
2 to 213
381
1 to 44
The corresponding Water Pollution Control Plant secondary effluent suspended
solids is shown in Table 28. These pre-construction and post-construction
data contain the plant secondary bypass as well as the secondary treatment
effluent.
Ill
-------
o*
E
to
Q
_1
O
CO
Q
UJ
Q
Z
UJ
CL
CO
(_
UJ
u_
U-
UJ
h-
z
0.
40
30
20
10
1973 1974
CONSTRUCTION
PRECONSTRUCTION-
EVALUATION
POST-CONSTRUCTION
EVALUATION
FIGURE-51 - WATER POLLUTION CONTROL PLANT EFFLUENT SUSPENDED
SOLIDS LONG-TERM MONTHLY SUSPENDED SOLIDS CONCENTRATION
112
-------
Table 28. PLANT SECONDARY TREATMENT
EFFLUENT SUSPENDED SOLIDS
Pre-Construction
Evaluation Phase
Arithmetic Mean Suspended
Solids (mg/1)
Standard Deviation (mg/1)
Number of Observations
Range of Values Observed (mg/1)
32.64
26.90
453
2 to 213
Post-Construction
Evaluation Phase
32.70
28.26
399
2 to 256
It should be noted that, in contrast to the plant BOD,, data summarized in
Table 25, above, these results do not indicate a measurable difference be-
tween pre-construction and post-construction results. This difference is very
interesting because the same primary treatment effluent that bypasses secon-
dary treatment is measured in the BOD,, and suspended solids data. This result
is unexplained and is merely noted as such.
A long-term terminal pond suspended solids loading rate analysis was per-
formed on the same data base as was completed for BOD,.. Figure 52 shows the
mean monthly variation in terminal pond influent and effluent suspended solids
loading rate as pounds suspended solids per acre per day. From these data,
it can be shown that the pond effluent loading rate is a very heavily damped
response to the influent loading. Furthermore, by comparison of Figure 52
showing the pond effluent mean monthly loading rate with Figure 51 showing the
pond effluent mean monthly concentration, it may be shown that for the monthly
mean values at least, the pond effluent suspended solids loadings correspond
very closely to the effluent suspended solids concentration. Similarly, com-
parison of Figure 48, pond effluent mean monthly organic loading rate, with
Figure 51, pond effluent mean monthly suspended solids concentration, shows a
close correlation in the peaks and valleys, particularly in the period since
the start-up of the Stormwater Facility. It should be noted that the pond
effluent mean monthly COD and NH^-N do not conform to the same trend of values
as the pond effluent mean monthly suspended solids.
Figure 52 also shows the mean monthly removal efficiency for terminal pond
suspended solids loading data for the period since pond start-up in October,
1971. The mean monthly suspended solids averaged 56.9 percent for the period
from pond start-up to the completion of the Stormwater Facility (December,
1974) with a standard deviation of 28.3 percent. The corresponding mean
monthly suspended solids loading removal efficiency for the 14-month post-
construction evaluation period commencing in January, 1976 was 74.8 percent at
a standard deviation of 14.0 percent.
Figure 53 is a bivariate "scattergram" of these mean monthly suspended solids
loading removal efficiency as a function of the suspended solids loading rate.
113
-------
o
"O
b
o
1100 •*
1000
900
800
700
600
500
to
o
~i 400
S
UJ
30O
200
100
TERMINAL POND REMOVAL
EFFICIENCY
TERMINAL POND INFLUENT
TERMINAL POND EFFLUENT
I
UJ
o
U_
b
UJ
o:
o:
UJ
o_
-80
-100
PRECONSTRUCTION-^
EVALUATION . .„
Note: Ibs./acre/riav
POST-CONSTRUCTION-
EVALUATION ,
.12= ke/ha/dav
FIGURE 52 - LONG-TERM TERMINAL, POND MEAN MONTHLY
SUSPENDED SOLIDS LOADING RATE
114
-------
100
90
80
TO
60
90
200
400 600 800 1000
LBS. SS./ACRE/DAY
I2OO
-30
-40
-SO
-TO
-80
-90
-00
Note: Lbs/Ac/day x 1.12 == kg/ha/day
FIGURE 53 - TERMINAL POND MEAN MONTHLY SUSPENDED SOLIDS
LOADING REMOVAL EFFICIENCY AS A FUNCTION OF
POND INFLUENT LOADING RATE
115
-------
Total Phosphorus—
Chemical precipitation of phosphate materials in the Water Pollution Control
Plant secondary treatment flows has been operational since July, 1973.
Figure 54 is a long-term trend of the mean monthly plant effluent total phos-
phorus. The trend of values for the post-construction period commencing in
January, 1975 and ending in February, 1976 does not appear to be significantly
different from the months preceding. Therefore, these results tend to sup-
port the contention that the addition of screened stormwater overflows to the
pond will not adversely affect the terminal pond effluent.
116
-------
1973 I 1974
CONSTRUCTION
PRECONSTRUCTION
EVALUATION
POST-CONSTRUCTION-
EVALUATION
FIGURE 54 - WATER POLLUTION CONTROL PLANT EFFLUENT
TOTAL PHOSPHORUS LONG-TERM MEAN MONTHLY
CONCENTRATION
117
-------
SECTION 7
COST ANALYSIS FOR THE PROGRAM
CONSTRUCTION COSTS FOR THE STORMWATER FACILITY
The Fort Wayne Storrawater Demonstration Project construction costs included
the cost of constructing the Glasgow Avenue and 84-inch plant bypass regu-
lator overflow sewers, the river crossing sewers and structures, the Storm-
water Pumping Facility wetwell and pumps, the Screening Building including
its piping and equipment, additional chlorination capacity for the Water
Pollution Control Plant chlorination system, the chlorine contact; channel at
the Stormwater Facility and the associated piping, fittings, electrical con-
nections., structures, appurtenances and conduits to tie the project together.
For the purposes of discussion of construction costs of the facility, only
the actual construction costs attributed to the pumping, screening and chlor-
ination of combined sewer overflows received at the facility were considered.
The costs of constructing the associated sewers and structures were excluded
from this evaluation because these costs were determined to be relatively
commonplace.
The total approved grant amount for the construction of the project was
$1,828,123 of which the total adjusted contract amount for the Stormwater
Facility was $1,321,026. It should be noted that additional direct costs of
construction were sustained by all parties concerned in the construction and
post-construction evaluation phase. Some of these additional costs are in-
cluded in this cost analysis because these costs would normally be incurred
in the grant-eligible portion of a construction project of this scope, and
should, therefore, be included in the total cost analysis of this evaluation.
Table 29 is a breakdown of the cost of construction of the Stormwater Treat-
ment Facility. It should be noted that these construction costs were in-
curred in 1972, 1973 and 1974.
Table 29. DETAILED COSTS OF CONSTRUCTION OF FORT
WAYNE STORMWATER TREATMENT FACILITY
Description
SITE DEVELOPMENT AND PREPARATION
Unit Cost
Basis
Excavation - 39,614 CY $1.57/CY
Wetwell Structure Construction L.S.
Backfill Wetwell Structure & Facility Site Work L.S.
Construction
Cost
$ 62,029.99
394,650.00
74.643.31
118
-------
Table 29. DETAILED COSTS OF CONSTRUCTION OF FORT
WAYNE STORMWATER TREATMENT FACILITY
(Continued)
Unit Cost
Description Basis
Subtotal Site Development & Preparation
STORMWATER PUMPING AND SCREENING BUILDING ERECTION
Mechanical Trash Rack Installation L.S.
Pumping & Screening Building Installation L.S.
Stormwater Pumping & Screening Building
Mechanical Services L.S.
Subtotal Stormwater Pumping & Screening
Building Erection
STORMWATER TREATMENT FACILITY PUMPING EQUIPMENT
Stormwater Pumps L.S.
Stormwater Pumps Installation & Start-Up L.S.
Stormwater Pumps Discharge Piping L.S.
Wetwell Sump Pump L.S.
Subtotal Stormwater Treatment Facility
Pumping Equipment
STORMWATER FACILITY FLOW DISTRIBUTION SYSTEM
Head Tank Fabrication L.S.
Influent Piping to Screening Units L.S.
Subtotal Stormwater Facility Flow
Distribution System
STORMWATER FACILITY SCREENING EQUIPMENT
Bauer Hydrasieve Qy
Equipment Bid Purchase Price L.S.
Materials & Labor to Mount Bauer Screens L.S.
Subtotal Bauer Hydrasieve® Equipment,
Materials & Labor as Installed
Rex Rotary Drum Screen
Equipment Bid Purchase Price L.S.
Materials & Labor to Mount Rex Screen L.S.
Materials & Labor to Install Rex Screen L.S.
Construction
Cost
$531,323.30
$ 92,948.00
60,900.00
34.922.50
$188,770.50
$ 54,056.00
5,200.00
9,741.00
10,695.00
$ 79,692.00
$ 10,900.00
69,020.62
$ 79,920.62
$ 60,500.00
14,486.83
$ 74,986.83
$ 39,400.00
8,600.00
3,900.00
119
-------
Table 29. DETAILED COSTS OF CONSTRUCTION OF FORT
WAYNE STORMWATER TREATMENT FACILITY
(Continued)
Description
Subtotal Rex Rotary Drum Screen Equipment,
Materials & Labor as Installed
SWEGO Centrifugal Wastewater Concentrator®
Equipment Bid Purchase Price
Materials & Labor to Mount SWECO
Materials and Labor to Install Mechanical
Services to SWECO's
Materials & Labor to Install Electrical
and Controls to SWECOTs
Subtotal SWECO Centrifugal Wastewater
Concentrator ® Equipment, Materials
& Labor as Installed
STORMWATER FACILITY ELECTRICAL
Sitework, Service and Grounding
High Voltage Switchgear & Transformer
Motor Control Centers & Equipment
Stormwater Facility Fixtures, Devices, Wire,
Conduit & Boxes
Subtotal Stormwater Facility Electrical
STORMWATER FACILITY CHLORINATIQN
Water Pollution Control Plant Chlorination
Building Modifications
Chlorination Equipment, Labor & Materials
Chlorination Building Piping & Fittings
Subtotal Water Pollution Control Plant
Chlorination Building Modifications
Stormwater Facility Chlorination Piping & Valves
Chlorine Solution Line Piping & Valves
Subtotal Stormwater Facility
Chlorination Piping & Valves
Unit Cost
Basis
L.S.
L.S.
L.S.
L.S.
L.S.
L.S.
L.S.
Construction
Cost
$ 51,900.00
$138,744.00
5,700.00
8,200.00
15,950.00
$168,594.00
L.S.
L.S.
L.S.
L.S.
$ 8,130.00
19,818.00
89,788.00
28,976.00
$146,712.00
$ 14,400.00
6,000.00
$ 20,400.00
$ 21,516.39
,$ 21,516.39
120
-------
Table'29. DETAILED COSTS OF CONSTRUCTION OF FORT
WAYNE STORMWATER TREATMENT FACILITY
(Continued) '
Description
STORMWATER FACILITY MISCELLANEOUS AND UNCLASSIFIED
Stormwater Facility Safety and Other
Subtotal Stormwater Facility
Miscellaneous & Unclassified
Unit Cost
Basis
L.S.
Construction
Cost
$ 10.744.87
$ 10,744.87
TOTAL CONSTRUCTION COST OF STORMWATER FACILITY
$1.374.560.51
ANNUAL COST OF OPERATION AND MAINTENANCE
Costs data obtained for the Fort Wayne Storrawater Facility over the opera-
tional evaluation period from January, 1975 to February, 1976 were apportioned
to each event for which sufficient data were available to warrant removal
efficiency comparisons. These data were compared to efficiency parameters
studied to attempt to isolate some correlation between operating costs for
chemicals consumed, etc., and treatment efficiency. No such meaningful cor-
relations were obtained for the data taken in this program. See the Appendix
material to this report for the breakdown of costs used in this portion of
the analysis.
Next, the overall yearly costs for treating the total quantity of Stormwater
were examined. In order to enable the researchers to meaningfully evaluate
the treatment facility power consumption, several categories of power usage
were identified with estimates made of the proportional usage for each factor.
Since the facility was electrically heated, a large heating and ventilating
load was noted for the building. There were approximately 6,200 degree-days
at 60 degrees Fahrenheit (15.5 degrees Celsius) during the study period. The
average temperature in the screening facility was kept at 50 degrees Fahren-
heit (10 degrees Celsius) throughout the cold weather after several of the
SWECO screens' flow control valves froze up in February, 1975. This power
usage,was distributed to each screening unit by its overall.flow proportion
for the .evaluation period. Total estimated power usage for heating was
428,000 kilowatt-hours, or the overall power consumption for ventilation
was similarly estimated at 2,565 kilowatt-hours. The lighting load was cal-
culated at'30,635 kilowatt-hours, and other miscellaneous power usage items
added another 86,935 kilowatt-hours. Raw Stormwater pumping accounted for
78,630 kilowatt-hours total. Summing all these power usage requirements re-
sulted in an estimated power cost of $13,703.65, very close to the actual
billings of $13,714.02.
121
-------
Water consumption for the facility by the City Utilities billings showed,a
net cost for potable water of $205.00 for the year. Consumption was esti-
mated on the basis of apportioning the observed water usage for the Rex and
SWECO screens' backwash make-up consumption. This usage totaled an estimated
4.67 million gallons (17.6 million liters) for these screens. An estimated
92 percent of the facility's total usage went for screening equipment back-
wash water make-up, based on a delivery rate of 150 gpm (570 1pm).
Chlorine usage was recorded for several treatment events. The cost of liquid
chlorine in this region has fluctuated considerably in the past few months.
As a result of this uncertainty of price, a cost of $0.16 per pound of chlo-
rine used was used to estimate this cost for comparative evaluation purposes.
Operating labor costs for the evaluation period are not considered typical of
the anticipated costs of a similar facility because of the tremendous number
of separate grab samples obtained for this evaluation program. As a result,
only maintenance costs are considered in the evaluation of costs shown in
Table 30, following.
Consumed resources utilized in the program such as chemical cleaners, elec-
trical power, potable water, replacement screens and such are considered in
the cost analysis undertaken as a part of this program. The total annual
treatment costs are used in the determination of the amount of treatment on
a cost per million gallons basis. Table 30 is a summary presentation of the
overall cost analysis for operation and maintenance costs, excluding the cost
of labor to operate the screens.
On the basis of consideration of the operating expenses incurred over the
lifetime of the screening units (estimated at 10 years for all units although
the Bauer units do not have the same electric motor-driven components as the
Rex and SWECO units), the cost per million gallons treated shows the Bauer
is clearly half the operating cost of the other two units. Note that the
major expense for this facility is heating and ventilating which was neces-
sary for keeping the SWECO pneumatic solenoid valves from freezing.
122
-------
Table 30. COMPARATIVE ANNUAL OPERATIONS ANP MAINTENANCE COSTS FOR
THREE SCREENING METHODS
Bauer .
Hydrasieve
SWECO
Centrifugal
Rex Rotary Wastewater
Drum Screen Concentrator
Annual Prorated Cost of
Pumping Raw Stormwater
Annual Prorated Cost of
HVAC Services
Annual Prorated Cost of
Lighting
Annual Miscellaneous
Prorated Utility Costs
Annual Prorated Potable
Water Usage
Annual Estimated Preventive and
Corrective' Maintenance Costs
Annual Estimated Replacement
Screens Costs
Total Annual Estimated Costs for
Operations arid Maintenance
Total Flow Treated in MG
Annual Operating Cost per
Million Gallons
$ 417.76 $ 348.15 $ 692.90
2,285.90,' .1,905.00 3,791.35
162.76 135.65 269.95
0.00
**
Subtotal Annual Utilities Costs $2^866.42*
0.00
0.00
$2.866.42
113.5 MG
57.90
43.60
1,079.35
42.70
$2,490.30* $5,876.25*
$ 735.00 $1,428.00
1,500.00 3,920.00
$4.725.30 $11.224.25
94.512 MG 188.1 MG
$ 25.25/MG $ 50.00/MG $ 59.68/MG
Note: 1 MG x 3.785 = Million Liters
*Utility costs are prorated on the basis of flow treated for each screening
unit. . . . ',
**Clean-up only.
123
-------
REFERENCES
1. "Lake Erie Report",: U.S. Federal ¥ater Quality Administration,
Washington, B.C., U.S. Department of Interior, August, 1968,
2. Middendorf, W. H. Engineering Design; Boston, Allyn and Bacon, Inc.,
1969, p. 286.
3. Master Plan for Sewers, Part One - Relief Sewers, Henry B. Steeg &
Associates, Inc., Indianapolis; City of Fort Wayne, Indiana, May, 1969,
135 p.
4. Grant Offer - Large-Scale Demonstration of Treatment of Storm-Caused
Overflow by the Screening Method, Statement of Work; Federal Grant No.
11020 GYU Modification to U.S. Environmental Protection Agency,
Chicago, June 14, 1972.
5. Screening/Flotation Treatment of Combined Sewer Overflows; Envirex, Inc.,
Milwaukee. U.S. EPA Project 11020 FED. January* 1972, p, 172.
124
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-79-085
3. RECIPIENT'S ACCESSION'NO.
4. TITLE AND SUBTITLE
5. REPORT DATE
COMBINED SEWER OVERFLOW TREATMENT BY SCREENING AND
TERMINAL PONDING - Fort Wayne,Indiana
August 1979 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Del mar H. Prah and Paul L. Brunner
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
City of Fort Wayne
One Main Street
Fort Wayne, Indiana
10. PROGRAM ELEMENT NO.
1Bg822 SOS #1 Task 17
46802
11020 GYU
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Municipal Environmental Research Laboratory—Cin.,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
Final 4/71 - 2/76
14. SPONSORING AGENCY CODE
EPA/600/14
15.SUPPLEMENTARY NOTES Hugh Masters, Phone (201) 321-6678
Project Officers: Clifford Risley, Phone (312) 353-2200
16. ABSTRACT
A 75 MGD combined sewer overflow treatment facility was constructed to obtain plant-
scale "data on the effectiveness and costs of screening combined sewer overflows (CSO)
by three types of fine screens. The methods of screening evaluated included fixed side
hill screens, rotary centrifugal fine screening, and horizontal rotary drum fine
screening. Also evaluated were the benefits to the receiving waters by pollutant re-
moval by screening, chlorination, and ponding previously by-passed CSO. Each of 38
separate CSO events were analyzed over a 13-month period commencing in January, 1975
for 14 pollutant parameters and flow rates.
None of the screening methods studied haye removal efficiencies for suspended .solids,
BQDc, COD and nutrients that were significantly different from zero at the 95% con-
fidence level. No effect between -hydraulic loading and removal efficiency was found.
The least overa'll cost method of screening was the fixed vertical-type screen. All the
screens studied have hydraulic head requirements on the order of 2 to 10 feet static
loss. The two-day terminal pond following screening was shown to be capable of meeting
30 mg/1 of BOD$ and suspended solids 95 percent of the days studied over the 4-year
evaluation period. No perceptible effect of the addition of storm-caused CSO was founc
on any parameter studied. Average monthly removal of BODs and suspended solids was
75 percent. With loading rates greater than 200 pounds BOD5 and suspended solids per
acre per day stable, relatively high efficiency performance was observed for the pond
at hydraulic detention of 1-2 days.
Construction of this project resulted |n 54 percent reduction in quantity of CSO to the
river. The stream showed significant improvements in instream fecal coliform.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Combined sewer, Overflows, Screening, Ponds,
Chlorination
Maumee River, Fort
Wayne, Indiana
Lake Erie
Combined sewer overflow
treatment
13B
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
UNCLASSIFIED
21. NO. OF PAGES
139
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 {9-73)
125
-U.S.GOVERNMENT,PRINTING OERCfc 1979 -6 57 -146 /5466
-------
-------
-------
«> "DO
COO) £<
§3 35
£.2.
(0
(0
(I)
m
2
01
o
O
no
i
o
00
en
5& =
«ll
|ff
=TO O
BO"*
3*6 <
°-*< |-
Hi
• OJ U
~ n tfi
Q* D"O
i.° 3
isl
^* ~r * u> _|
is U
3 wo
S§3
' §8
S«
Is
O 3
*• u
S"g
Is
g£
« m
= s
f°
3-<
sg-
3-S
0(5
5§
» 3
"1
2&
ZTO
28
CO
"d
(D
CO
o rt
0)
en
rt
>mc
in 3
ro <
'
3 '
TO
3
i?
OOr
3 " '
I
O
*.
01
to
en
oo
------- |