CLEAN^ WATER POLLUTION CONTROL RESEARCH SERIES 11020 FDC 01/7
11/72
SCREENING/FLOTATION TREATMENT
OF COMBINED SEWER OVERFLOWS
U.S. ENVIRONMENTAL PROTECTION AdKNCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and progress in the control and abatement of pollution
in our Nation's waters. They provide a central source of
information on the research, development, and demonstration
activities in the water research program of the Environmental
Protection Agency, through in-house research and grants and
contracts with Federal, state, and local agencies, research
institutions, and industrial organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications Branch
(Water), Research Information Division, R&M, Environmental
Protection Agency, Washington, D. C. 20460
-------
SCREENINO/FLOTATION TREATMENT
OF
COMBINED SEWER OVERFLOWS
by
The Ecology Division
Rex Chainhelt Inc.
Milwaukee, Wisconsin
for the
Office of Research and Monitoring
Environmental Protection Agency
Contract 14-12-40
Project 11020 FDC
January 1972
For sole by tho Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1.50
-------
1-.?A Heviev; I.'otice
This rei.ort has bion rsviovccl by the Environmental Protection
Agency and a: : rev. L -"or :.ublication. Approval does not
signify thit'lh^ c-.:.':o-.-:As ivjcoscarily reflect the views and
policies oi' the: ;-:-:v::-o:v.-intal Protection Agency nor does
mention of trade- ).-.:.-j^ or co:;r;.ercial products constitute
endorsement or r-jcc::.::.'jndation for use.
-------
ABSTRACT
This report documents a study to develop a treatment system for combined
sewer overflows. The processes of chemical oxidation, screening,
dissolved-air flotation and disinfection were evaluated in the laboratory.
It was determined that chemical oxidation was not feasible. The majority
of the pollutants were of a particulate nature, which indicated solids/
liquid separation processes could possibly provide effective treatment.
A 5 MGD combination screening and dissolved-air flotation demonstration
system was designed, installed and evaluated.
The system was utilized to treat 55 combined sewer overflows. The
drainage area served by the facility was a completely developed 500 acre
residential area of Milwaukee, Wisconsin. Suspended solids and volatile
suspended solids removal in the range of 65-80% were consistently obtained
at influent concentrations of 150 to 600 mg/1. BOD and COD removals were
slightly lower at 55 to 65% for influent concentrations of 50 to 500 mg/1.
Addition of chemical flocculents (ferric chloride and a r.ationic poly-
electrolyte) was necessary to obtain these removals. Without the use of
chemical flocculents, removal of BOD, COD, suspended solids, and volatile
suspended solids were all in the range of 40-50%. The serening/flotation
system provided sufficient detention time (^15 minutes) for adequate
disinfection with hypochlorite salts. Cost estimates indicate a capital
cost of $21,056 per MGD capacity or $3,828 per acre for a 90 MGD screening/
flotation system. Operating costs were estimated at $3.09c/1000 gallons of
treated overflow, including chemical flocculent addition.
This report was submitted as partial fulfillment of contract number 14-12-40
under the sponsorship of the Water Quality Office Environmental Protection
Agency.
iii
-------
TABLE OF CONTENTS
SECTION I CONCLUSIONS
SECTION II RECOMMENDATIONS
SECTION III INTRODUCTION
SECTION IV LITERATURE SEARCH
Characteristics of Combined Sewer Overflows
Removal of Solids by Screening
Dissolved-Air Flotation
Disinfection of Combined Sewer Overflows
Ultraviolet Light Disinfection
Disinfection by Chemicals
Summary Remarks
SECTION V SITE SELECTION AND PRELIMINARY INVESTIGATIONS
Site Selection
Preliminary Investigations
Conclusions Preliminary Investigations
SECTION VI DESIGN AND CONSTRUCTION OF DEMONSTRATION FACILITY
Design of Screen
Flotation System Design
Design of Supporting System
Operation Methods and Test Plan
Methods and Operational Procedures
Sampling Procedures
Test Plan
SECTION VII OPERATING RESULTS AND DISCUSSION
Characterization of Raw Waste
Results of the Screening Operation
Operation of the Flotation System
Disinfection of Combined Overflows
Conceptual Design
Pumping System
Screening System
Flotation System
Economic Considerations
PAGE
1
5
7
9
13
15
18
18
18
20
23
23
27
36
39
39
44
45
48
48
51
51
53
53
59
61
72
72
74
75
78
81
SECTION VIII ACKNOWLEDGEMENTS
87
-------
TABLE OF CONTENTS (CONTINUED)
SECTION IX
SECTION X
APPENDIX I
APPENDIX II
APPENDIX III
APPENDIX IV
BIBLIOGRAPHY
PUBLICATIONS
CHEMICAL OXIDATION
LITERATURE SEARCH
Hydrogen Peroxide
Ozone
Oxidation by Chlorine
Oxidation by Oxy-Acids and Their Salts
Electro-Chemical Oxidation
Combination of Oxidants
LABORATORY INVESTIGATIONS
General Test Procedures
Special Test Procedures with Various
Chemical Oxidants
Ozone Oxidation System
RESULTS AND DISCUSSION
Oxidation with Hydrogen Peroxide
Oxidation with Chlorine
Oxidation with Combined System of Hydrogen
Peroxide and Chlorine
Oxidation with Ozone
Oxidation with Gamma Radiation
SUMMARY AND CONCLUSIONS
BIBLIOGRAPHY - APPENDIX I, CHEMICAL OXIDATION
ANALYTICAL PROCEDURES
Analytical Instruments and Apparatus
Analytical Procedures and Analyses
Dissolved - Air Flotation Test
Procedure
DEMONSTRATION SYSTEM COSTS
OPERATING DATA AND STATISTICAL PROCEDURES
OPERATING DATA
STATISTICAL PROCEDURES
PAGE
89
93
95
96
97
99
102
104
105
105
106
106
106
111
111
111
116
116
120
122
125
127
131
132
133
135
139
143
144
170
vi
-------
LIST OF TABLES
TABLE PAGE
1 QUALITY OF COMBINED SEWER OVERFLOWS 12
2 CHARACTERISTICS OF COMBINED OVERFLOW 30
3 SUMMARY OF PRELIMINARY SCREENING DATA 32
4 SUMMARY PRELIMINARY FLOTATION DATA 34
5 PRELIMINARY SCREENING/FLOTATION DATA 35
6 PRELIMINARY DISINFECTION DATA 37
7 VARIABLE COMBINATIONS UTILIZED FOR TESTING 40
8 SUMMARY FIRST FLUSH DATA 54
9 SUMMARY OF DATA - AT INTERVALS OF GREATER THAN 4 DAYS
BETWEEN OVERFLOWS 55
10 SUMMARY OF EXTENDED OVERFLOW DATA .58
11 PARTICULATE & DISSOLVED RELATIONSHIPS 58
12 POLLUTANT REMOVALS BY SCREENING 60
13 POLLUTANT REMOVALS BY SCREENING/FLOTATION 62
14 SUMMARY PARTICULATE AND DISSOLVED ORGANIC REMOVAL EFFICIENCIES 65
15 COMPARISON OF PRESSURIZED FLOW RATES 67
16 COMPARISON OF THE EFFECT OF OVERFLOW RATE ON REMOVAL 70
EFFICIENCIES
17 SUMMARY OF DISINFECTION DATA 73
18 OPERATING COST ESTIMATES 86
1-1 LABORATORY ANALYSIS OF COMBINED SEWER SAMPLES UTILIZED
FOR CHEMICAL OXIDATION STUDY 107
1-2 RESULTS OF CHEMICAL OXIDATION WITH HYDROGEN PEROXIDE 113
1-3 EFFECT OF UV LIGHT AND COBALT ON HYDROGEN PEROXIDE OXIDATION 114
1-4 CHLORINE OXIDATION TESTS 117
1-5 LIGHT CATALYZED CHLORINE OXIDATION 118
1-6 OXIDATION WITH COMBINED HYDROGEN PEROXIDE AND CHLORINE 119
1-7 SUMMARY 03 OXIDATION TESTS 121
1-8 SUMMARY 03 OXIDATION AND AIR FLOTATION TESTS 123
1-9 SUMMARY DISINFECTION DATA - ALL SPRING STORMS 124
1-10 SUMMARY OF OXIDATION OF COMBINED OVERFLOWS WITH VARIOUS
OXIDANTS AND COMBINATIONS 126
IV-1 OPERATIONAL DATA 144
IV-2 RAW WASTE CHARACTERISTICS - FIRST FLUSHES 148
IV-3 RAW WASTE CHARACTERISTICS - EXTENDED OVERFLOWS 149
IV-4 SCREENED WATER QUALITY - FIRST FLUSHES 151
IV-5 SCREENED WATER QUALITY - EXTENDED OVERFLOW 152
IV-6 EFFLUENT WATER QUALITY - FIRST FLUSHES 1.54
IV-7 EFFLUENT WATER QUALITY - EXTENDED OVERFLOW 155
IV-8 DISSOLVED COD AND TOC DATA .158
IV-9 SCREEN BACKWASH & FLOATED SCUM QUALITY 1969 DATA 1.59
IV-10 SCREEN BACKWASH & FLOATED SCUM QUALITY 1970 DATA 161
IV-11 FIRST FLUSH EVALUATIONS 162
IV-12 RST FLUSH REMOVALS IN PERCENT 164
IV-13 INTENDED OVERFLOW REMOVALS IN PERCENT 165
IV-14 SUSPENDED SOLIDS MASS BALANCE 168
-------
LIST OF FIGURES
FIGURE PAGE
1 RELATIONSHIP OF FREQUENCY OF COMBINED SEWER OVERFLOWS
TO INTERCEPTOR CAPACITY 11
2 PROJECT DRAINAGE AREA 25
3 RELATIONSHIP OF RUNOFF COEFFICIENT TO PERVIOUSNESS 26
4 RELATION BETWEEN WATER DEPTH AND FLOW IN HAWLEY ROAD SEWER 28
5 TYPICAL INTERCEPTOR DEVICE 29
6 COMPARISON TYLER MESH TO SIZE OF OPENING 33
7 SCALE DRAWING OF DEMONSTRATION SYSTEM 40
8 DEMONSTRATION SYSTEM 41
9 SCREENING SYSTEM 42
10 SCREENING SYSTEM 43
11 FLOTATION SYSTEM 46
12 FLOTATION TANK 47
13 HAWLEY ROAD OUTFALL 49
14 INTERIOR VIEWS OF THE CONTROL SHACK 50
15 RELATIONSHIP BETWEEN FLOW IN SEWER AND SUSPENDED SOLIDS 57
16 SUSPENDED SOLIDS REMOVAL-SCREENING/FLOTATION ' 63
17 BAFFLE ARRANGEMENT FOR OVERFLOW RATE TESTS 68
18 RECOMMENDED SCREEN ARRANGEMENT 79
19 RECOMMENDED SCREENING/FLOTATION ARRANGEMENT 82
20 OVERALL SYSTEM CONFIGURATION 83
1-1 HUMAN TOLERANCE FOR OZONE 100
1-2 OZONE ABSORPTION SYSTEMS 103
1-3 APPARATUS FOR ULTRA VIOLET LIGHT OXIDATIONS 108
1-4 APPARATUS FOR OZONE OXIDATIONS 108
1-5 SCHEMATIC OF OZONE TEST APPARATUS 112
vlii
-------
SECTION I
CONCLUSIONS
The following conclusions can be made based on the data obtained during
this study.
1. A screening/flotation treatment system is an effective method of
reducing the pollution caused by combined sewer overflows.
2. The combined sewer overflows monitored during this study had the
following characteristics in mg/1 at the 95% confidence level.
BOD 49110
COD 161±19
Suspended Solids 166±26
Volatile Suspended Solids 90114
Total Nitrogen 5.5±0.8
3. Approximately 20% of the overflows monitored exhibited first flushes
of the following characteristics in mg/1 at the 95% confidence level.
BOD 186140
COD 581±92 (
Suspended Solids 522±150
Volatile Suspended Solids 308±83
Total Nitrogen 17.6±3.1
When present these first flushes persisted for 20 to 70 minutes.
4. All first flushes occurred at a time interval of greater than 4 days
between overflows.
5. It was demonstrated that a screening/flotation system could achieve
the following percent removal efficiencies based on the raw waste
characteristics encountered(Hydraulic Loading:Screen 40 gpm/sq ft,Flotation
2.75 gpm/sq ft).
Without Chemical With Chemical
Flocculant Addition Flocculant Addition
BOD 35±8 60±11
COD 41±8 57±11
Suspended Solids 43±7 71±9
Volatile Suspended Solids 48±11 71±9
Nitrogen 29114 2419
6. The chemical flocculant addition required to achieve the above stated
removals was 20 mg/1 ferric chloride and 4 mg/1 of a cationic polyelec-
trolyte.
-------
7. An operating pressure of 50 psig and a pressurized flow rate of 20%
of the raw waste flow provided sufficient air bubbles for proper
flotation tank operation.
8. The average volume of waste residual solids (i^.e^ screen backwash
and floated scum) was 1.75% of the raw flow.
9. The average removal efficiencies for the screening/flotation system
decreased (as shown below in percent removals) when the hydraulic over-
flow rate was increased from ^2.75 to 3.75 gpm/sq ft.
Low Overflow Rate High Overflow Rate
BOD 59 52
COD 57 54
Suspended Solids 70 61
Volatile Suspended Solids 71 64
10. Utilizing a 50 mesh (297 p) screen at a hydraulic loading of 40 gpm
per sq ft of screen. The percent removal efficiencies for screening alone
were observed to be:
BOD 27 ± 5
COD 26 ± 5
Suspended Solids 27 ± 5
Volatile Suspended Solids 34 ± 5
11. Capital cost estimated for a full scale screening/flotation system
completely installed was $21,056 per MOD capacity or $3,828 per acre
for a plant capacity of 90 MGD. These costs were distributed as
follows: screening/flotation modules and accessories 71%; sewers and
outfall 4%, land costs 9%, engineering and related costs 16%.
12. The operating costs for a screening/flotation system were estimated
at 3.090/1000 gallons of treated overflow. These costs include chemical
flocculant addition. These costs were distributed as follows: chemicals
81%, power 17%, maintenance 2%.
13. The screening/flotation system provides sufficient detention time
(^ 15 min.) for effective chlorination.
14. Because of the intermittent operation and remote location of
combined sewer overflow treatment systems, completely automated opera-
tion is required.
15. The use of a 50 mesh (297 \i) screen eliminated the need for bottom
sludge scrapers.
16. It was not always possible to obtain optimum chemical flocculant
addition due to widely varying raw waste characteristics.
-------
17. The following conclusions were made from the chemical oxidation
study:
a. Chemical oxidation of combined sewer overflow is not
technically or economically feasible.
b. Ozone was the best oxidant evaluated.
c. Chlorine and hydrogen peroxide are not effective oxidants,
d. Particulate organic matter is very difficult to oxidize.
e. Ozone disinfection was found to be less reliable than
chlorine for combined sewer overflows.
-------
SECTION II
RECOMMENDATIONS
It is recommended:
1. that screening/flotation be utilized to reduce the
pollution from combined sewer overflows where treatment is the
preferred alternative.
2. that the design and operating procedures outlined in the
conceptual design section of this report be utilized for screening/
flotation systems treating combined sewer overflow.
3. to investigate the effects of the unit process of flocculation
on flotation removal efficiencies.
4. to investigate methods of control of chemical addition as
waste strength varies.
5. to evaluate the dewatering characteristics of the sludges
produced and various sludge disposal alternatives.
-------
SECTION III
INTRODUCTION
During recent years combined sewer overflow has been recognized as a
significant pollution problem. Immediate reaction was to separate the
sewer systems. It was found, however, that this solution was expensive
and sometimes impossible. Furthermore, it has also been shown that
stormwater could carry a large pollutional load, therefore, tending to
reduce the effectiveness of sewer separation. As research to find a
solution to the combined sewer problem began, it became apparent that
no one solution would solve all the problems associated with combined
overflows. The ultimate solution depends upon successful application
of many different approaches which may include separation of sewers,
holding tanks, and treatment systems.
The objective of EPA/WQO Contract Number 14-12-40 is to develop and
demonstrate a method of treating combined sewer overflows. It is
envisioned that the system or systems developed during this project will
be utilized to provide effective and economic treatment for those
combined sewer overflows where treatment is the best alternative.
Initially, the technical approach in this project was screening and
chemical oxidation of combined overflows. Early during the study it
was determined that chemical oxidation was not feasible and this approach
was abandoned. A summary of the work done on chemical oxidation may be
seen in Appendix I. During the investigation of chemical oxidation,
however, a promising alternative was uncovered. It was discovered that
the pollutants in combined sewer overflow consisted mainly of particulate
matter. This clearly indicated that an efficient solids/liquid separa-
tion process should provide a high degree of treatment for combined
overflows. -The work scope of the existing contract was modified to allow
study of a screening/dissolved-air flotation system. Later, after
evaluating the system for a short time, it became apparent that a more
detailed study of the screening process could provide important addi-
tional information and the contract was amended to include these studies.
The amended contract is divided into the following phases.
I. Literature search, site arrangements, preliminary investiga-
tions
II. Design and construction of a 5 MGD screening/flotation system
III. Operation of the screening/flotation system
IV. Design and construction of a 5 MGD series screening system
V. Operation of the screening/flotation and series screening
systems in various combinations
VI. Final report, system disassembly and restoration of demonstra-
tion site
-------
This report summarizes the work done to date on the above mentioned
phases I, II, and III. Work on phases IV, V and VI is now underway
and completion is expected by March of 1972.
-------
SECTION IV
LITERATURE SEARCH
It is assumed that the following process elements will be utilized to
accomplish the treatment of combined sewer overflows.
A. Solids/liquid separation
1. Screening
2. Dissolved-Air flotation
B. Disinfection of the waste flow
A search of the existing literature was made to determine the
characteristics of combined sewer overflow, and to obtain pertinent
information on the above mentioned process elements.
Characteristics Of Combined Sewer Overflows
More than 1400 U.S. communities serving 50 million people have what is
known as combined sewer systems which provide one collection system
for both sanitary sewage and stormwater runoff (1). During dry weather
the flow in such a system consists mainly of sanitary sewage and is
normally intercepted by the interceptor systems before it reaches the
outfall, where it otherwise would be discharged to a lake or river. As
long as the capacity of the interceptor system exceeds the flow in the
combined sewer, the flow is directed to the sewage treatment plant for
purification.
During a storm, the flow in the combined sewer can increase from 50 to
100 times the normal dry weather flow (2). Normal interceptor capaci-
ties are between 1.5 and 5.0 times the dry weather flow (hereafter
abbreviated DWF)(2)(3)(4). This would mean that at normal interceptor
capacities, the interceptors would be from 97.3 to 99.0% efficient in
collecting and transporting the sanitary sewage to the treatment plant.
These efficiencies are very misleading, since during periods of high
stormwater runoff, it is possible to have up to 96% of the sanitary
sewage bypassing the interceptor system and being discharged untreated
into the receiving waters (5). This discharge of untreated combined
overflows could cause severe pollution problems if the vater is reused
a relatively short distance downstream as a water supply or for
recreational activities.
The pollution caused by combined sewer discharge is directly related to
the frequency of storm runoff exceeding the interceptor capacity.
Three independent investigators (5)(6) (7) report that the normal dry
weather flow is approximately equivalent to a rainfall intensity of
about 0.01 inch per hour. With interceptor capacities of 1.5 to 5 times
the dry weather flow, rainfall intensities greater than 0.015 to 0.05
-------
inches per hour would be necessary to cause combined overflow which
would go untreated. Palmer (6), Johnson (7), and McKee (5) studied
tne rainfall records of Detroit, Washington, D.C., and Boston. The
results of their findings showing the number of combined overflows per
month for given interceptor capacities may be seen in Figure 1. With
an interceptor capacity of 5 times (DWF), the number of overflows per
month was between four (4) and six (6). This indicates the extent of
possible pollution of the receiving body of water.
The pollutional effect of combined sewer overflows will, of course, be
determined not only by the frequency of overflows, but also by the quality
of the overflow. The most important and commonly discussed pollutional
parameters associated with combined sewer overflows are suspended solids,
BOD, and coliform count. To establish a basis for comparison, the
average raw (primarily domestic) sewage contains 200 to 300 mg/1 suspended
solids (1)(8)(9), (50 to 60% of these suspended solids are settleable),
100 to 300 mg/1 BOD (1)(8)(9) and coliform counts of A3 to 150 x 10
per ml (9)(10).
Assuming pure stormwater to be essentially unpolluted, it would be
expected that the BOD, suspended solids, and coliform counts of
combined sewer overflow to be substantially lower than those of raw
sewage. This assumption could result in considerable error when
estimating the quality of combined sewer overflows. A limited amount
of data has been obtained which indicates the quality of pure storm
runoff is highly variable and can be quite polluted (4)(6)(11)(12)(13)
(14). BOD's ranging from 6 to 600 mg/1, suspended solids ranging from
100 to 2000 mg/1, and coliform counts as high as 200,000 per ml have
been reported.
Another factor which can greatly influence the quality of combined
sewer overflow is the amount of solids which settle in the sewer system
. ^^^ -'-^ -« ' - _ L
during periods of low flow. Since combined sewer systems must be
designed for a wide range of flow, the amount of solids which are
deposited during periods of dry weather can be significant (15)(16) (17).
Hopefully, these solids will be resuspended during the first flushes of
the storm and be intercepted before combined overflow begins. Published
data has shown, however, that with an interceptor capacity of only 1.5
to 5 times the dry weather flow and a sewer system designed to take 50
to 100 times the dry weather flow, most of the solids which have settled
are not resuspended until after the capacity of the interceptors has
been exceeded (11)(14) (15) (16)(17). Hence, these solids can contribute
greatly to the pollutional load in the stormwater overflow.
The above discussion has illustrated that combined sewer overflows can
contribute a significant pollutional load to receiving streams or lakes.
Table 1 summarizes data reported by others on the quality of combined
sewer overflows. It is obvious that the quality of combined overflows
is highly variable. Of the three pollutional parameters most mentioned,
i.e., BOD, suspended solids and coliform count, the coliform count,
a measure of pathogenic bacteria contamination i.8), appears to be
the most significant (10) (13). In a recent study (13), a survey was
made to indicate what water uses were most affected by combined
10
-------
I
01
D-
O
o
01
H
1
u
Id
01
Palmer (6)
McKee (5)
Johnson (7)
246
Interceptor Capacity (times DWF)
FIGURE 1
RELATIONSHIP OF FREQUENCY OF
COMBINED SEWER OVERFLOWS TO
INTERCEPTOR CAPACITY
10
-------
TABLE 1
Source
Keference BOD
Number mg/1
(6; JO
(14) 100
(13) 59
(13) 92
(13) 121
(12)
(11) 40-260
(11) 220-614
QUALITY OF
Suspended
Solids
mg/1
250
544
203
129
436
150
130-930
168-426
COMBINED SEWER
E. Coll
Density
per ml.
43,000
300,000
500
180,000
2300-24000
2,100
OVERFLOWS
Discussion
Detroit, Michigan
Average Estimated Values
Buffalo, New York (Bird Avenue)
1 Sample
San Francisco, California
(Average Value 14 Samples)
Chicago,. Illinois
Average 31 Overflows (1962
Buffalo, New York (Bailey Avenue)
Average Values, 43 Samples
1964 Data
Toronto, Ontario, Canada
(Eglinton Avenue)
Welland, Ontario, Canada
(Burgar Street)
-------
overflows. The survey indicated those uses most affected were commercial
(fishing), recreational (swimming), and public health (water supplies).
The main pollutant affecting these water uses is pathogenic bacterial
contamination (measured by coliform density) (1) (5) (15) (16) . Table 1
shows the range of coliform counts to vary from 0.5 x 10 to 300 x
10-Vml in combined sewer overflows. (A review of disinfection starts
page 170
The suspended solids concentration of combined overflow can be quite
large. Table 1 shows a range of suspended solids in combined overflows
of 129 to 390 mg/1. The values shown in Table 1 are averages. The
initial peak of suspended solids can be higher (11) (15) (14) . Romer and
Klashman (16) presented graphs which illustrate how suspended solids in
combined overflows vary as the flow Increases. Thev reported that, in
general, the suspended solids concentration and the flow reach a maxi-
mum value at approximately the same time. The suspended solids then
begin to fall off rapidly. Stegmaier (17) has shown that the main
Increase in solids is of inorganic nature. Initially, the BOD exerted
by the solids will probably be quite large, since the settled organic
material will be resuspended before the inorganics. Hence, the BOD of
combined sewer overflows will initially be quite high with a large
portion of the BOD in the partlculate form. This would seem to indicate
some sort of mechanical separation may be very effective in removing
BOD during the first flushes of a storm. Later, as the storm progresses
the BOD would be expected to decrease to levels much lower than those
of raw sewage. The ratio of dissolved to partlculate BOD will increase
and mechanical separation will probably become less efficient in removing
BOD.
Removal of Solids by Screening
Bar screens with openings of 1/2 to 1-1/2 inches are used extensively
in sewage treatment plants throughout the United States (18). The
amount of solids removed by these coarse type screens ranges from 0.5
to 6 cubic feet of wet solids per million gallons of sewage (19) (20).
On a dry weight basis, this represents 9 to 48 pounds of dry solids per
million gallons of sewage (19) (20) or a removal of about 1 to 6 ppm
from the sewage flow. Since raw sewage contains 200 to 300 ppm (1)(8)
(9) suspended s-l.ids, bar screens remove only a small fraction of the
solids, i.e., 0.5 go 2%. This fact is not surprising, since the primary
purpose of bar screens is to remove large and/or floating solids to
protect pumps and other treatment devices downstream of the bar screens
In general, the approach velocity in the screening channel should not
fall below a self-cleaning value (about 1.25 fps) or should it rise to
values high enough to force the screenings through the bars (about 3
fps) (8) (18). Headlosses through bar racks can be formulated as an
orifice loss (8).
4/3
h = B(w/b) h slnO
13
-------
where
h = loss of head in feet
w = maximum width of the bars facing the flow
b = minimum width of the clear openings between pairs of bars
hu = velocity head (in feet) of the water as it approached the rack
6 = the angle of the rack with the horizontal
6 = bar shape factor
Values for 8 are 2.42 for sharp-edged rectangular bars, 1.79 for
circular bars, and 0.76 for tear-drop shaped bars (8). Headloss
through bar racks is generally held below one foot by periodic
cleaning of the bars. This cleaning can be accomplished manually or
mechanically (8).
Fine screens 1/16 inch to 1/32 inch openings have been used in a
number of locations (19)(20)(21)(22) to treat municipal wastes. This
size of screen will remove 2 to 20% of the suspended solids in raw
sewage (8) (18)(19) (20) (21)(22). Openings of 1/16 inch, correspond to
a sieve size (Tyler series) of 12 to 20 mesh. Use of standard wire
sizes finer than these meshes tend to blind because of the build-up of
animal fats and greases on the screen (23). It has been reported by
Peterson (23), however, that the use of new synthetic materials for
screens has greatly reduced the blinding tendency and allows greases to
be discharged from the top of the screens. He also reports meshes as
fine as 300 have been used when filtering raw sewage. The allowable
rates of filtration, with these extremely fine mesh screens, however,
were not reported. The results of screening raw sewage through a 300
mesh screen showed a reduction in suspended solids from 196 to
101 mg/1 or a 48.5% removal (23). This removal rate is approximately
equal to the rates obtained by primary sedimentation of raw sewage (8)
(18).
As the openings in the screen are reduced, the headless increases and
screen blinding becomes very important. This necessitates automatic
mechanical cleaning, which is generally accomplished by backwashing
the screen with jets of water and/or air. Headlosses through fine
screens can be calculated from an orifice type equation.
where:
Q = flow (cfs)
A = area (sq ft)(actual open area)
g = gravity constant (32 ft/sec2)
C = screen coefficient
h = headless through the screen (feet)
Screen coefficients (C) vary with the size of the screen openings.
Values are generally between 0.3 to 0.6 for 2 to 60 mesh screens.
14
-------
Headlosses through fine mesh screens are generally held below 2.5 feet
(8) by varying the frequency of cleaning.
There is a very limited mention found in the literature on screening of
combined sewer overflows. Stegmaier (17) and Romer and Klashman (16),
however, have shown that a large portion of the suspended solids in
combined sewer overflows can be attributed to inorganic materials.
This tends to indicate that fine screening of combined sewer overflows
may show a higher percentage removal of suspended solids than is
obtainable when screening raw sewage. Boucher and Evans (24) have
discussed the microstraining process and its applications in removal of
suspended material. They have indicated that microstraining can be
effectively and economically used in polishing of sewage effluents.
Metropolitan Sanitary District of Chicago is reported to be utilizing
microstrainers for its Hanover Water Reclamation Facility (25). In a
recently reported study by Keilbaugh e^ al^ (26), up to 98% of suspended
solids were removed from a combined sewer overflow by using a nominal
23 micron aperture microstrainer screen. However, most of the data
published towards the use of microstrainers in waste treatment has been
of investigational nature only, and significantly more data is needed to
substantiate such findings.
Dissolved-Air Flotation
An active use of dissolved-air flotation (DAF) in the waste treatment
area has only been made in the last two decades, but the use of air-
bubbles to change the apparent specific gravity of suspended matter in
water has been practiced for nearly a century in the mining industry
for the beneficiation of ores. Selective adsorption was accomplished
by the use of flotation agents. Chase (27) and Rohlich (28) have
discussed the developmental background of dissolved-air flotation. The
process as it is used now is an outgrowth of air flotation methods used
in the recovery of paper fibers. The fiber recovery was achieved by
the adsorption of an air bubble on the fiber. The floated material
formed a scum and was collected by a suitable mechanism. The presence
of alum and addition of glues aided the separation of fiber by the
flotation process. Dissolved-air flotation utilizes a much smaller
diameter bubble (less than 100 microns) than that used in the. mining
industry. It is clear from the above discussion that the removal of
suspended matter is accomplished when there exists sufficient ad-
sorption forces between the air bubble and solid particle.
Geinopolos and Katz (29) have discussed solids-liquid separation by
dissolved-air flotation as compared to sedimentation and have indicated
that a waste could be treatable by either sedimentation or flotation,
but higher separation rates and solids concentration may be possible by
dissolved-air flotation. This results in smaller basins, smaller
sludge volumes and higher water recoveries for treatment by flotation
processes as compared to sedimentation. The mechanism and driving
forces involved in the flotation process is similar to those
encountered in sedimentation. Stoke's Law illustrates the mechanism
15
-------
by the following equation:
V - gD2(Y8-Yi> /18u
Where:
V = Particle separation velocity, ft/sec
g = Gravity constant, ft/sec2
D = Diameter of particle, ft
Ys = Density of particle, Ib/cu ft
YI = Density of liquid, Ib/cu ft
U = Viscosity of liquid, Ib/ft/sec
In the flotation process, the density of the air-solid combination is
less than the suspending medium and value of V becomes negative, caus-
ing an upward particle velocity. Also, the effect of the air-bubble is
to increase the difference in densities between the particle and
suspended medium, which in turn increases the solids separation rate.
Van Vuuren et^ al (30) has reported the use of oxygen produced by algal
photosynthesis to accomplish flotation. He also cites the applicabil-
ity of dispersed air flotation in conjunction with chemical precipita-
tion in advanced treatment of wastes (31). Chase (27) has described
the types of dissolved-air flotation processes used for waste treatment,
These include pressure and vacuum types. However, pressure type appli-
cations in waste handling far outnumber those for vacuum flotation,
primarily because of inherent flexibility in a pressurized system.
The vacuum process is limited by the amount of reduced pressure that
can be effectively used. Geinopolos and Katz (29) state that the
process elements of dissolved-air flotation are:
(1) Flow pressurization
(2) Air Introduction
(3) Air Solution
(4) Pressure reduction and bubble formation
(5) Bubble/solids attachment
(6) Solids/liquid separation
(7) Separated solids removal
The literature describes variations in the flow pressurizing process
(27)(28)(32)(33). These include: total pressurization, in which the
entire volume of raw waste is pressurized, split flow, in which a
portion of the raw waste is pressurized and later blended with the
remaining raw waste stream; and effluent pressurization, in which a
portion of basin effluent is pressurized and later blended with the
entire volume of raw waste.
The dissolved-air flotation process has been applied to remove
suspended matter from many industrial wastes such as: paper wastes
(35), refinery wastes (28)(36), laundry wastes (37), soap wastes (38),
machine shop wastes (39), automobile wastes (40), and others (27)(28)
16
-------
prior to discharge in a receiving body of water or reuse. Vrablic
(34) in his discussion of the fundamental principles of dissolved-air
flotation mentions that hydrophobic solids will float much more easily
than will hydrophilic ones. In an investigation of the kinetics of
removal of organic matter by activated sludge, Rohlich and Katz (A)
demonstrated that activated sludge floe can be separated from water by
dissolved-air flotation. Ettelt (42) has also shown that dissolved-
air flotation can be successfully employed for activated sludge
thickening. There have been several other reports on the use of
dissolved-air flotation for dewatering of aerobic biological solids
(43) (44) (45)(46).
Proper performance of a flotation unit is dependent upon sufficient
air bubble/solids attachment to effect good flotation. Batch
laboratory flotation tests as described by Eckenfelder (47) have been
used to estimate the flotation characteristics of wastes for the
purposes of design. Howe (48), in a mathematical derivation of flota-
tion cell design also recommended that considerable experimentation
with each different waste precede the use of his equations in determin-
ing the exact criteria for flotation cells. For sludge thickening
application, air/solids ratio has been used to relate laboratory data
to prototype facilities. A ratio of 0.03 to 0.1 is considered
optimum for most sludges (34)(49). However, air/solids ratio for
clarification alone is generally not critical. Truck mounted
continuous flow, pilot scale units are used widely for obtaining
design information. The primary variables for flotation design are
operating pressure, ratio of pressurized flow to raw waste flow,
retention period, and the combined particle/air bubble rise rate.
Generally, the effluent suspended solids decrease and the concentra-
tion of solids in the floated sludge increase with increasing reten-
tion period. For clarification, a detention period of 15 to 30 minutes
is normally considered adequate in the flotation zone. Rise rates of
1.0 to 3.0 gpm/sq ft (0.13-0.39 feet per minute) are commonly
employed. The amount of pressurized flow usually varies from 15 to
50% of raw waste flow for separation. The recycle amount and
detention period are higher for thickening application depending
upon feed solids concentration. The process pressure may vary from
30 to 70 psig, with 40 psig considered normal. Air requirements
range from 0.5 to 1.0 cu ft of standard air per one hundred gallons
of pressurized flow (32). The air and liquid are mixed under
pressure in a retention tank with a detention time of 1 to 3 minutes.
Addition of chemicals may be employed depending upon the nature of
the waste and the degree of treatment required. Chemicals such as
alum, lime and ferric chloride are usually used in the chemical
treatment of a wastewater. The purpose of chemical treatment is to
precipitate and coagulate colloidal and finely suspended particles to
form large suspended floe particles. This increases the separation
rate of the solids by flotation. Recent efforts to improve flotation
efficiency by chemical addition have involved use of synthetic organic
polyelectrolytes. Generally, these polymers have been found to
17
-------
improve solids capture (clarification), to increase the concentra-
tion of solids in the float and to increase the capacity of flotation
equipment (50) (51) (52). An apparent advantage of polyelectrolytes
is the elimination of a separate flocculation tank, since the floe
particles are formed in an extremely short period of time. Cationic '
polyelectrolytes have been successfully employed as flotation aids
for thickening applications (46)(53). The results indicate that
desired float concentrations may be achieved with high solids
loadings when polymers are utilized. Bench scale procedures are
normally employed for determining the dosage and effectiveness of a
particular polymer application in dissolved-air flotation.
Disinfection Of Combined Sewer Overflows
Disinfection may be defined as destroying those bacteria that cause
infection or disease. Sterilization is defined as the destruction of
all living organisms (54). Methods commonly used for disinfection
include heat, ultraviolet light, and chemical addition. Chemicals
which have been used include chlorine, bromine, iodine, potassium
permanganate and ozone (8). The various methods and chemicals used
for disinfection which appear to be applicable to this project will
be discussed below.
Ultraviolet Light Disinfection
The effect of radiation on bacteria has been studied in detail, and
the relationship between wave length and germicidal effect is well
known. The optimum wave length is 2600 angstrom, hence, low
pressure mercury vapor lamps with their high output at 2537 angstrom
are effective bactericidal agents (55). Light of this wave length
falls in the ultraviolet section of the light spectrum. To insure
disinfection, the water should be relatively free of suspended
matter which might shade the organisms against the light (8). Time
and exposure intensity must also be adequate to insure disinfection.
One hundred microwatts per square centimeter of 2537 angstrom light
will produce high bacteria kills at contact times of less than one
minute (55) (56).
Use of ultraviolet light for disinfection has found limited appli-
cation, probably because other methods of disinfection are more
economical (8).
Disinfection Bv Chemicals
Chlorine and chlorine compounds have proven so economical and effi-
cient that they are widely employed for their bactericidal action.
The chlorination of wastes may be regulated to accomplish various
degrees of bactericidal action. Chlorine is a bacteriostatic agent
when applied in small concentrations, to prevent an increase in the
bacterial population. Chlorine is a disinfecting agent when
applied in larger amounts to destroy those bacteria that cause
18
-------
infection or disease. Chlorine may be, but seldom is, applied in
amounts so large as to be a sterilizing agent, i.e., to destroy all
living organisms (54).
Forms of chlorine which may be used for disinfection and other purposes
include gaseous or liquid chlorine, chlorinated lime, calcium hypo-
chlorite, sodium hypochlorite, chlorinated copperas, and chloramines.
Chloramines, however, are a slower acting and less active form of
disinfectant. Liquid or gaseous chlorine is the most commonly
utilized form.
The efficiency of disinfection using chlorine or other disinfecting
agents is dependent on (a) the contact time, (b) the type and
concentration of microorganisms, (c) the pH and temperature of the
water, (d) the presence of interfering substances, (e) and the degree
of protection afforded organisms from the disinfecting solution by
materials in which they may be imbedded.
Generally, a chlorine dosage, sufficient to give a 0.5 ppra residual
after a 15 minute contact time, is considered adequate for disinfec-
tion (54). The actual amount of chlorine added to obtain this
residual will vary with the waste being chlorinated. Fair and Geyer
(8) give values of 6 to 24 mg/1 for raw sewage.
Camp (10) indicates screening and chlorination of combined sewer
overflows would be an effective and inexpensive way of treating
these flows. Chlorine dosages were estimated to range from 1.6 to
8.5 ppm.
Symons (57) found the chlorine demand for Buffalo sewage during
periods of combined overflow to range from 6 to 7 mg/1. Considering
the possible quantities of combined sewer overflow, this represents
quite a substantial chlorine dosage, and will require chlorine
feeders operating over a very wide range of flows.
Iodine and bromine were also found to be effective in disinfecting
sewage, but required higher dosages compared to chlorine (58).
Fluorine was found to be so reactive that it was difficult to store
and apply (58).
Potassium permanganate has been used to disinfect water supplies (59)
(60), and is an effective disinfectant at relatively low dosages (59).
Use of potassium permanganate has not been widespread. This is
probably due to one or more of the following disadvantages of
potassium permanganate disinfection when compared to chlorine dis-
infection.
1. KMNO, imparts a color to the water which must be removed
before use (62).
19
-------
2. Most effective pH for KMNO^ is 5.9 which is below the pH of
many water sources.
3. Potassium permanganate is not a good post-treatment
disinfectant because of the color and pH criteria.
Ozone is a very effective disinfectant. This fact is well known and
has been proven over a period of 60 to 80 years (35)(46)(61)(62)(63).
Ozone is used extensively in Europe, especially in France (62).
Disinfection by ozone follows a slightly different pattern when
compared to chlorine. Generally, with increasing chlorine dosage the
number of bacteria progressively decrease, while with ozone little
reduction is noticed until a critical dosage is reached. At this
critical value, the bacterial population is rapidly reduced (35).
Unlike chlorine, the disinfecting action of ozone is little affected
by changes in temperature or pH. Ozone acts rapidly and is almost
instantaneous, where chlorine needs time to be effective (35)(61).
The reason ozone is not utilized to the same extent as chlorine is
probably twofold, i.e., cost and inability to carry an ozone residual
for any length of time. Hann (62) states that, in general, if
disinfection is the sole objective, ozone is more expensive than
chlorine. When using ozone as a disinfectant, it is not possible to
carry an ozone residual for a significant period of time (35), and
this is sometimes cited as a disadvantage when using ozone.
When using ozone proper precautions must be taken in applying the
ozone to insure disinfection (62). The water should be low in
turbidity, the ozone demand must be satisfied, and there must be a
residual of free ozone in the water for a definite period of time.
This usually is from 0.1 to 0.2 ppm ozone residual for 1 to 5
minutes (35)(62).
Summary Remarks
The material contained in this literature search was compiled in an
effort to uncover the known process elements applicable to the
treatment of wastes similar to combined sewer overflows. The informa-
tion gained from this literature search will be used as a guideline
for laboratory and bench scale tests to determine what combination(s)
of the various process elements is (are) best suited to the design
fabrication, and operation of a prototype combined sewer overflow '
treatment unit.
Although a literature search is technically concerned only with the
reporting of data found in published form, some of this data for
one reason or another can be eliminated from further consideration
without the benefit of laboratory testing. As a result, this report
comprises a certain number of engineering judgments and/or eval-
uations. Such judgments, however, have been made only wherethe
20
-------
material available or the investigator's experience have justified
making them. That data which appeared to have genuine value to the
completion of this project has been carried forward to the
laboratory studies portion of the project.
Significant, observations from the literature search are summarized below:
1. The flow in combined sewers ran increase 50 to 100 times the
dry weather flow,
2. Normal interceptor capacities are between 1.5 and 5.0 times
the dry weather flow.
.3. Up to 96% of the sanitary sewage can bypass the interceptor
system during heavy storms.
4. Rainfall intensity greater than 0.015 to 0.05 inches per hour
is generally sufficient to cause combined sewer overflow.
5. Limited data indicate pure storm water can be quite polluted.
6. The majority of the solids settled in thp sewer svstem during
dry weather are not resuspended until the capacity of the intercep-
tor system is exceeded.
7. It was indicated that the main water usages effected bv combined
sewer overflow were commercial fishing, swimming, and public
water supply due to bacterial contamination.
?1
-------
SECTION V
SITE SELECTION AND PRELIMINARY INVESTIGATIONS
Site Selection
A search was conducted to select a suitable demonstration site for
the installation and operation of a demonstration treatment unit.
The City of Milwaukee, as an available source of combined sewer
overflow, was solicited in regard to participating in the demon-
stration of the proposed treatment system. The City of Milwaukee
responded favorably by offering both access to a combined sewer as
well as providing sufficient land to install the demonstration
equipment. In addition, the City Engineer's Office was made
available for consultation and aid in selecting the most beneficial
demonstration site.
Six potential sites were investigated. These sites were located
in the City of Milwaukee as follows:
1. The Edgewood Avenue site, consisting of a 72 inch combined
dewer discharging into the Milwaukee River.
2. The 27th Street site (south end of viaduct), consisting of a
48 inch combined sewer discharging into the Menomonee River.
3. The East Kane Place site, consisting of a 72 inch combined
sewer discharging into the Milwaukee River.
4. The Bay Street site, consisting of a 90 inch diameter
combined sewer discharging into Lake Michigan.
5. The Russell Avenue site, consisting of a 120 inch diameter
combined sewer discharging into Lake Michigan.
6. The Hawley Road site, consisting of an 8*6" by 5'0" combined
box sewer discharging into the Menomonee River.
The potential sites were appraised and evaluated from the standpoint
of availability for project duration, access to sewer flow, access to
utility sources, space limitations, extent of structural and/or
topographical modifications, proximity to residential or developed
areas, drainage area covered, and the socio-economic nature of the
drainage area.
By inspection, the Edgewood Avenue, 27th Street, and East Kane Place
locations were eliminated as potential demonstration sites. The
Edgewood Avenue and 27th Street sites were inaccessible and were
limited in space. The East Kane Place site was located in a residen-
tial and developed area. In fact, the site consisted of a city-owned
23
-------
lot, located between two residences, which were about thirty feet
apart.
The Bay Street, Russell Avenue, and Hawley Road locations were
further investigated as potential demonstration sites. Maps of the
respective combined sewer areas were obtained from the City
Engineer's Office in which the drainage area covered and the socio-
economic nature of the drainage area were delineated. The areas
drained by each of these three sites is predominantly residential
with some local business and industrial activity. The drainage areas
covered by the three potential sites were as follows:
Bay Street 385 acres
Russell Avenue 465 acres
Hawley Road 495 acres
The Bay Street location was eliminated because an investigation
revealed that it would be difficult and impractical to impound the
flow in the sewer to provide the necessary raw sewage pump suction
conditions.
The Russell Avenue site was eliminated because of a restrictive time
limit placed by Milwaukee County authorities upon the use of the site
for project purposes. It was felt that the site may not be available
for the full duration of the project.
Since the Hawley Road site appeared to be the most desirable locatation
for a demonstration unit, additional information was obtained on this
sewer and the area served.
As mentioned above, the total area served by this sewer is 495 acres.
The area is located on the western edge of the Milwaukee City limits.
Figure 2 presents a street map of the City of Milwaukee. The outlined
portion depicts the drainage area served by the Hawley Road sewer. The
drainage area was studied using recent (1970) aerial photographs. It
was determined that approximately 42% of the area was impervious, i.e.,
roof tops, streets and parking lots. This value is within the range
presented in the literature (8) for North American cities in areas of
1 and 2 family dwellings. The relationship between runoff coefficient
and perviousness is presented in Figure 3. The data in Figure 3 is
based on the City of Milwaukee runoff curves. It may be seen from
Figure 3 that the runoff coefficient for the Hawley Road drainage area
is 0.40. While it is recognized that the runoff coefficient can vary
as the storm progresses (8), no attempt was made to refine the runoff
coefficient, since an exact number was not critical to this project.
The aerial photograph indicated that the area is a completely
developed residential area of one of the older sections of the city.
There is very little open area present such as parks or fields. All
dwellings are of the one and two family variety. Population density
based on number of dwellings per acre and assuming four persons per
family was estimated at 35 persons per acre. There are some small
-------
R : i 'UN
" %'
-
-
Onfe
'
s*»v T? r ' -
' C. t/» «« *-* -*- * . -
CM O *
r^. *X5 *^0 ^O v^ v^
-
* C W LiCfyd
M , wt,
-
.'acobi/J
FIGURE 2
PROJECT DRAINAGE AREA
-------
100.
80
Based on City of Milwaukee, Wisconsin
Runoff Curves
en
3
O
H
60
01
a
E
C
01
a
«-i
01
a.
40
Value for Hawley Road Drainage Basin
20
0.1
0.2
0.3 0.4
Coefficient C
0.5
0.6
0.7
0.8
-------
shops such as jewelry, hardware, etc. No industrial manufacturing
facilities are located within the drainage area served by the
Hawley Road sewer. The one industry which is within the drainage
area is Topp Oil and Supply Company. They are involved with blending
various oils and apparently do not actually process any oil at this
site. A discussion with personnel at the Milwaukee Sewage Treatment
Plant revealed that Topp Oil did not discharge any significant
pollutional load into the sewage system.
The combined sewer outfall serving the Hawley Road drainage area is
rectangular in cross section 8.5 feet wide and 5 feet high. The
relationship between water depth and flow rate is shown in Figure 4.
It may be seen from the figure that the sewer has a capacity well in
excess of 100 MGD. Because of the location of the sewer and the
rectangular shape, a retaining structure was easily added to impound
the flow. This allowed long test runs even during periods of low
rainfall intensity. Calculations show that if a 3 foot high retaining
structure is placed at the combined sewer outfall to retain the flow,
a volume of about 100,000 gallons can be retained for evaluation
after normal overflow has stopped. The retaining structure is a
movable dam which tips up as a result of water pressure if the sewer
begins to surcharge. In the normal untipped position the dam does not
seal the sewer completely and this insures that the sewer will always
be dry prior to any overflow. At a treatment rate of 5 MGD, this
reservoir would allow an additional one half hour operation of the
demonstration system. In actual operation, an additional operating
time of only 15-20 minutes has been realized. It should be noted
that the length of operation reported herein includes this 15-20
minute storage factor and hence the duration of overflow is shorter
than the length of run by this time period. The Hawley Road Sewer
is fed by two interceptor devices. These interceptor devices are of
the sump type. A sketch of the interceptor device utilized is shown
in Figure 5.
Since the Hawley Road Sewer and adjacent land met all the requirements
necessary to successfully complete the demonstration phase of this
contract, arrangements were completed with the City of Milwaukee in
February of 1968 to allow use of the sewer and land for this project.
Preliminary Investigations
After the site selection was finalized, 14 overflows were monitored to
obtain information on overflow characteristics, and to provide raw
overflow for subsequent laboratory testing. Results of the laboratory
testing program were then utilized to design the demonstration system.
Table 2 presents a summary of the data collected prior to system
design. As was expected, the range of values is quite large.
Suspended solids values from 2158 to 65 mg/1 were observed, while COD
values ranged between 1410 and 52 mg/1. The dissolved organic fraction
of the overflows as measured by COD was only 4 to 25% of the total
organic load present in the combined overflow. This indicated that an
27
-------
100
80
60
50
40
30
20
60
E
01
0)
in
o
-------
NJ
-O
Combined Sewer
Man Hole
Interceptor Sewer
FIGURE 5
TYPICAL INTERCF.PTOR DEVICE
-------
TABLE 2
CHARACTERISTICS OF COMBINED OVERFLOW
(Fall 1967 - Spring 1968)
Volatile
Date Total Suspended Suspended Total Dissolved
of Solids Solids Solids COD COD^' BOD (5) Coliform
Overflow mg/1 mg/1 mg/1 mg/1 mg/1 mg/1 ///ml
09/27/67 388 65
11/02/67
11/10/67 651 418
11/16/67 664
11/25/67 649 138
91 440
65 5000
520 110
2io
159 59 4260
12/07/67
12/21/67
01/29/68
04/03/68
04/03/68
04/17/68
04/20/68
04/23/68
04/28/68
04/28/68
05/20/68
232
242
2158
808
194
228
70
137
461
248
194
113
298
150
1410
889
188
178
52
134
476
311
251
50
30
60
95
48
17
47
13
26
111
65
54
2151
1860
5730
1388
421
1280
1850
6000
32000
26000
25600
Remarks
Extended Rain
Extended Rain
4 Days
First Flushes
First Flushes
L. Rain
End of Overflow
First Flushes
Extended Overflow
First Flushes
Extended Overflow
(1) See Appendix II Analytical Procedures
-------
efficient solids/liquid separation system should provide high removals
of pollutants from the raw waste. Colifonn density also varied
widely from 440 to 26,000 per milliliter.
Laboratory testing which was performed included screening, chemical
oxidation, flotation and disinfection. Chemical oxidation proved
impractical for use in treating combined sewer overflow. Complete
details of oxidants utilized, reaction times and efficiency may be
seen in Appendix I.
A summary of the preliminary screening data is presented in Table 3.
Various mesh sizes from 50 to 400 (297-37 w openings) were investi-
gated. The majority of the tests were run on a 50 mesh-297 w opening
screen. The laboratory screen test consisted of pouring the sample
through the screen into a beaker. The sample was poured gently to
avoid breakup of any particles, and in all tests a mat of solids was
not allowed to form on the screen. This eliminated the filtering action
which could have resulted if a layer of solids was allowed to form on
the screen. It was anticipated that in actual field operation the
screens would operate partially blinded and, hence the removal rates
would probably be higher in the field when compared to the laboratory
analysis shown in Table 3. All screen mesh sizes mentioned in this
report are Tyler series mesh. Relationship of opening size to mesh
size is presented in Figure 6.
Based on the data of Table 3, average removal of suspended solids was
18.6% for a 50 mesh screen, 32% for 100 to 200 mesh screen, and 51% for
a AOO mesh screen. An exception to these figures may be seen in the
data from 12/21/67 in Table 3. For this test a 120 mesh screen was
utilized and the raw waste was flocculated with 1 mg/1 of Reten A-l
polyelectrolyte. Removals of COD and suspended solids from this test
were 56% and 55% respectively. These removals were significantly
higher than those obtained on the same waste (12/21/67) using a 200
mesh screen, i.e. , 22% COD removal and 26% suspended solids removal
(Table 3). These results indicate that flocculation prior to screening
can improve the efficiency of the screens. More detailed data,
however, is needed to verify these results.
The results of the laboratory flotation testing is presented in Tables
4 and 5. Table 4 indicates those flotation tests which were run on the
raw overflow and Table 5 those tests run on screened overflow. The
results shown in Table 4 indicated extremely high removals of COD and
suspended solids may be obtained using flotation alone in conjunction
with polyelectrolyte addition. The bench scale flotation testing was
performed using a standard procedure which is detailed in Appendix II.
Removals of 73 to 95% COD and 88 to 97% suspended solids were obtained.
In Table 5 data on screening/flotation are presented. All the
experiments except 5/20/68 were performed without the addition of
flocculating chemicals. Based on the data presented in Table 5,
suspended solids removals of 41% to 72%, BOD removals of /-Q% f.o 62%
and COD removals of 42% to 72% were predicted at the 90% confidence
level. By comparing the two experiments on the overflow of 5/20/68,
31
-------
TABLE 3
SUMMARY OF PRELIMINARY SCREENING DATA
COD Data
Suspended Solids
Date
of
Overflow
11/10/67
11/16/70
12/07/67
12/21/67
01/29/68
04/03/68
04/03/68
04/17/68
04/20/68
04/23/68
04/28/68
04/28/68
05/20/68
01/29/68
12/21/67
11/10/67
12/07/67
12/21/67
01/29/68
12/07/67
01/29/68
Screen
Size
Mesh
50
50
50
50
50
50
50
50
50
50
50
50
50
100
120
170
200
200
200
400
400
Screen
Opening
Microns
297
297
297
297
297
297
297
297
297
297
297
297
297
145
122
87
74
74
74
37
37
Raw
mg/1
520
210
298
150
1410
889
188
178
52
134
476
311
251
1410
150
520
298
150
1410
298
1410
Dissolved
mg/1
110
50
30
60
95
60
30
110
50
30
60
50
60
After
Screening
mg/1
407
182
247
121
922
559
121
134
51
99
331
199
188
899
66
340
233
117
200
Removal
%
22
13
17
19
35
37
36
25
2
26
30
36
25
36
56
35
22
22
33
Raw
mg/1
418
232
242
2158
808
194
228
70
137
461
248
194
2158
242
418
180
242
2158
180
2158
After
Screening
mg/1
270
195
1526
707
183
207
67
122
331
177
153
1504
68
210
150
178
1368
100
900
Removal
%
35
19
29
13
6
9
4
11
28
29
21
30
55*
50
17
26
37
44
58
* Flocculated with 1 mg/1 Reten A-l (Hercules, Inc.)
-------
w
OJ
0)
N
01
0)
500
400
300
200
100
50
30
20 30 40 60 80 100 200
Opening (microns)
500
FIGURE 6
COMPARISON TYLER MESH TO ST7F OF OPF.NTNr,
-------
TABLE 4
SUMMARY PRELIMINARY FLOTATION DATA
Deten Chemical COD Data Suspended Solids
Date of Pressurized Time
Overflow Flow - % Min. Type
12/21/67
12/21/67
1/29/68
15
8
15
3
3
5
Re ten A-l/j
Reten A-l(]
C-31(2)
Dosage
mg/1
D 1
D !
10
Raw
mg/1
150
150
1410
Effluent
mg/1
37
40
98
Removal
75
73
93
L Raw
mg/1
242
242
2524
Effluent
mg/1
17
28
75
Removal
93
88
97
NO^ES: 1. Anionic polyelectrolyte-Hercules, Inc.
2. Cationic polyelectrolyte-Dow Chemical
-------
TABLE 5
PRELIMINARY SCREENING/FLOTATION DATA
4/3/68 4/3/68 4/17/68 4/20/68 4/23/68 4/28/68 4/28/68 5/20/68 5/20/68(1)
SUSPENDED SOLIDS
Raw mg/1
After Screen mg/1
Screen & flota-
tion mg/1
Overall Removal %
COD
Raw mg/1
After Screen mg/1
Screen Flota-
tion mg/1
Overall Removal %
808
707
279
65
889
559
194
183
114
41
188
121
228
207
73
68
178
134
69
61
70
67
55
21
52
51
44
15
137
122
65
53
134
99
64
52
461
331
65
86
476
331
119
75
248
311
199
93
70
194
153
68
65
251
188
114
56
194
153
36
81
251
188
92
63
BOD
Raw mg/1
After Screen mg/1
Screen & flota-
tion mg/1
Overall Removal %
181
148
85
53
36
33
12
67
47
33
21
55
13
10
10
23
26
20
12
54
111
92
36
68
65
58
27
58
54
49
39
28
54
49
29
46
(1) 10 mg/1 C-31 added to flotation test - all other runs without chemicals
pressurized flow - 20%; pressure 50 psig; detention time - 5 minutes
-------
the effect of chemical flocculant addition may be seen. Addition of
10 mg/1 C-31, a cationic polyelectrolyte, caused and increase in
suspended solids removal from 61 to 81%, while the BOD and COD removals
increased from 56 and 28% to 63 and 46% respectively.
A summary of the preliminary disinfection data collected during the
preliminary sampling period is presented in Table 6. Data from ozone
and chlorine disinfection are shown. In general, ozone disinfection was
not as reliable as chlorine. Ozone demand for these waters was quite
high, and since ozone is extremely reactive, a residual ozone
concentration could not be obtained except at extremely high dosages
(approximately 60 mg/1 or greater). Inability to obtain small
residual ozone concentrations for a short period of time resulted in
the relatively poor E. Coli removals.
Chlorine disinfection on the other hand provided improved disinfection
at a dosage of 10 mg/1. Since chlorine is less reactive compared to
ozone, some chlorine residual could generally be attained. Disinfection
efficiency was better than ozone, but in some tests (April 28, 1968)
with high coliform densities, the effluent had significant coliform
indicating incomplete disinfection. It appears from this data, as the
coliform density increases, higher chlorine dosages will be required.
Conclusions Preliminary Investigations
The following conclusions can be made based on the data taken during the
preliminary investigations.
1. Combined Sewer overflow from the Hawley Road sewer contains
primarily particulate pollution.
2. Chemical oxidation is not technically feasible for combined
overflows.
3. Screening/dissolved-air flotation is a relatively effective
method for treating combined overflow.
4. Addition of chemical flocculents greatly increases the removal
efficiency of dissolved-air flotation.
5. Disinfection with chlorine was found to be more reliable than
with ozone.
36
-------
TABLE 6
PRELIMINARY DISINFECTION DATA
E. Coli.
3
3
17
20
23
28
28
Date
April
April
April
April
April
April
April
E. Coli.
Type per ml
1968
1968
1968
1968
1968
1968
1968
FF
EO
EO
EO
EO
FF
EO
1
1
1
6
32
26
,388
421
,280
,850
,000
,000
,000
Ozone
Dosage
mg/1
80
59
40
"-30
<10
<10
<10
in
Effluent
per ml
21
4
74
17
3,200
9,500
13,700
Chlorine
Dosage
mg/1
10
10
10
10
10
E. Coli
in
Effluent
per ml
0.
1
2
20
8
5
FF - First Flush
EO - Extended Overflow
-------
SECTION VI
DESIGN AND CONSTRUCTION OF DEMONSTRATION FACILITY
The results of the preliminary sampling and laboratory analysis were
utilized in the design of a 5 MGD demonstration treatment facility in-
corporating screening and dissolved-air flotation. This section of
the report describes the design criteria utilized, and the features of
the demonstration system. A flow sheet for the system is shown in
Figure 7 and a photograph of the overall system is shown in Figure 8.
Design of Screen
The design of the demonstration unit involved three basic areas, i.e.,
screen design, flotation tank design and integration of all system
components. Based on removal efficiencies obtained in the preliminary
sampling phase, a screen mesh of 50 (297u openings) was selected.
This mesh gave fairly good removal of pollutants (20-30%) and allowed
high flow rates (50 gpm/sq ft) at reasonable headlosses. The screen
is an open ended drum into which the raw waste flows after passing a
1/2" bar rack. The purpose of the bar rack is to remove large
objects which may clog the screened solids removal system or damage
the screen material, which is 306 stainless steel. The water passes
through the screen media and into a screened water chamber directly
below the drum. The drum rotates and carries the removed solids to
the spray water cleaning system where they are flushed from the
screen. Those solids which will not adhere to the screen media are
picked up by 4 angle iron sections which act similar to a roto-dip
feeder and are thus removed from the flow. Screened water is used
for backflushing the screen. The drum rotation and spray water
cleaning are controlled by liquid level switches located in the
screen chamber. The switehes are set to actuate at a headless of 6"
of water through the screen. The solids which are flushed from the
screen along with the spray water are collected in a hopper inside
the drum. This hopper is connected to a drain pipe which forms the
main axis of the drum. The slurry is then routed to waste. Detailed
photographs of the screening system are shown in Figures 9 and 10.
These photographs show the screen, internal hopper, float switches,
and drive system. The basic screen is fabricated from mild carbon
steel. The screen backing material (Figure 9) is a pei'forated metal
plate which has proved to be adequate support for the screen. Hole
size for the backing material is 3/4" x 3/4" on 7/8" centers. This
provides 73% open area and only a minimum of flow restriction.
Rotation speed is controlled by a variable speed drive, which is
positioned manually. Rotation speed range is 0.5 - 5 rpm. The
screened water is sealed from the raw waste by compressing a tube
between the end of the drum and a stationary ring. The sealing
arrangement allows operation at a maximum headloss of about 12".
Any excess headloss will allow some unscreened water to enter the
screened water chamber and a possible raw waste overflow from the
bar screen chamber. During some runs with extremely heavy solids
loads, the headloss capacity of the screen has been exceeded. These
39
-------
Tank
Flow
5'
\
\
_ Q'
\3f
f-'l oration 7 one
*lf
'
i
I*
-------
FIGURE 8
DEMONSTRATION SYSTEM
-------
FIGURE 9
SCREENING SYSTEM
42
-------
FIGURE 10
SCREENING SYSTEM
43
-------
instances will be discussed in a later section of this report.
The drum screen as installed is an 8 sided drum with an effective
diameter of 7% feet. The length of the drum is 6 feet. The 8 panels
have dimensions of 3' wide by 6' long. Total screen area is 144 sq ft.
The wetted screen area was a minimum of 72 sq ft and a maximum of
90 sq ft depending upon the head loss across the screen.
Flotation System Design
The dissolved air flotation basin design was based on the criteria
utilized by the American Petroleum Institute (API)(64). The major design
parameters are overflow rate, detention time, horizontal velocity and
depth to width ratio. The design procedures were modified slightly to
allow the high flexibility required in a project of this nature.
The basic principle of dissolved air flotation (OAF) is to produce
extremely small air bubbles (<100u) which can be attached to the partic-
ulate matter in a wastewater and cause flotation and removal of the
particulate matter. To provide these fine air bubbles a liquid stream
is mixed with air under pressure. The pressure is then released through
a weir type diaphragm valve to form the bubbles. The bubble laden
stream is then mixed with the remaining wastewater to be clarified in a
contained mixing zone within the DAF tank. This mixing zone has a
detention time of approximately 60 seconds. The bubbles attach to the
solids in this mixing zone. The bubble formation system is termed the
pressurized flow system. Generally, the source of pressurized flow is
the process effluent or another relatively solids free stream. In this
project the screened water was used as the source of pressurized flow.
The advantages to this approach eliminate an increase in hydraulic
loading on the DAF tank and provides a simplified plumbing system.
Because of the possible presence of a significant amount of solids in
the pressurized flow stream, a pressure tank without packing material
was utilized. This avoided the potential clogging problems associated
with a pressure tank packed with some foim of tower packing to increase
the air water interface. The type of pressurlzation utilized in this
project has been termed sidestream-preissurization.
The following discussion will delineate the process steps occurring
in the flotation tank. Screened water is pumped into the pressure
tank. Air is mixed with the water at the inlet to the pressure tank.
Water level in the tank is controlled by a liquid level float switch.
A slight excess of air is always added. If the water level becomes
too low the air stream is vented to atmosphere to allow the water level
to rise. Hence, the water level in the pressure tank is positively
controlled. There is a deflector baffle in the tank which spreads the
incoming water and promotes a large air water interface for good air
solution. The pressure in the tank is controlled by an air operated
weir valve; this valve provides the proper back pressure as well as the
required shearing action to form the small bubbles. This valve is
44
-------
controlled by a pneumatic controller. Once the pressure has been
released and the bubbles formed, the bubble laden stream is mixed with
the remainder of the screened water flow. A 60 second mixing chamber
is provided in the tank prior to entering the flotation zone. This
allows time for bubble/solid attachment. When flocculating chemicals
are utilized they may be added either to the raw flow or to the pressur-
ized flow after the pressure has been reduced. Generally polyelectrolyte
type flocculants, which require little or no flocculation period are
added in the pressurized flow stream. Once the pressurized stream and
remaining raw waste have been mixed, they enter the flotation zone of
the tank where separation occurs. A skimming system skims the floated
scum into three separate scum troughs, which convey the scum to ultimate
disposal. Generally the solids concentration of the scum is 1-2% solids
on a dry weight basis. This concentration flows easily by gravity and
does not require screw conveying. Figures 11 and 12 show various details
of the flotation system.
The design of the DAF system was such that a wide range of selected
variables could be evaluated in order to be able to recommend design
procedures specific to combined sewer overflow. The following range of
variables is possible with the demonstration system.
Flow rate 1500 - 4400 gpm
Surface loading 2-10 gpm/sq ft
Horizontal velocity 1.30 - 3.75 ft/min
Pressurized flow rate 300 - 1100 gpm
Operating pressure 40 - 70 psig
Detention time 7-44 minutes
In addition to the above flexibility, the flotation tank can be divided
down the center. This in effect forms two separate flotation tanks.
The lengths of these tanks can be controlled by appropriate baffels.
This then allows the evaluation of two separate overflow rates on the
same storm, thus allowing direct comparison of the overflow rate variable
and elimination of all other possible interacting variables. The tank
was divided as described above at the start of the 1970 storm period
(April, 1970). The final as-built dimensions of the flotation tank were
18 feet wide, 8% feet water depth, and a 65' long flotation zone.
Design of Supporting Systems
The appurtenant equipment and tasks necessary to provide a functional
demonstration system included site preparation, construction of a manhole
and pumping sump, selection of the necessary flow metering equipment, and
design of suitable electrical and pneumatic control systems.
45
-------
PRESSURE TANK, CONTROL SHACK, AND SCREEN SYSTEM
PRESSURE TANK AND PRESSURE REDUCTION VALVE
FIGURE 11
FLOTATION SYSTEM (See also Figure 7)
46
-------
HEAD END OF FLOTATION TANK
OVERHEAD COLLECTOR AND EFFLUENT STRUCTURE
FIGURE 12
FLOTATION TANK
47
-------
As may be seen in Figure 8, the system was located under an existing
highway bridge which provided overhead protection. Site preparation
involved laying a concrete slab to provide support for the system
tankage. A manhole was constructed as well as a pumping sump, directly
in the combined sewer. A dam was also provided in the sewer to allow
impoundment of a limited quantity of overflow for subsequent treatment.
This dam was designed to tip under the hydraulic pressure of the overflow
to prevent surcharging the sewer. A photograph of the outfall is shown
in Figure 13.
Flow metering equipment is provided to measure the influent flow rate,
the volume of screen backwash water and the volume of floated scum.
All other process flows could be obtained by adding or subtracting the
proper measured values. The raw flow and screen backwash are measured
via venturi meters connected to differential pressure gauges which both
record and totalize the flows. The floated scum is measured with an
open channel float type meter, which records and totalizes the flow of
floated scum.
The electrical control panel provides all necessary controls for 100%
automatic operation with manual overrides on all systems. Operational
modes will be discussed in a later section of this report. A Merchants
Police alarm is connected to the system so that personnel will be alerted
when the system goes into operation. The system is always (24 hours per
day, 7 days per week) ready to operate, and hence, the maximum possible
number of overflows can be monitored. Photographs of the control shack
are shown in figure 14. Demonstration system costs are presented in
Appendix III.
Operation Methods and Test Plan
Methods and Operational Procedures
The demonstration system previously described was put into operation
in May of 1969. Data reported herein represent data taken during the
period May, 1969 through November, 1970. During this period 55 over-
flows were treated with the demonstration system. The system is put
into operation automatically when a float switch in the sewer senses
an overflow. The pressurized flow system is immediately put into
operation and the raw feed pump begins to prime. All runs were darted
with the tank approximately 80% full of water. The water in the tank
was that left from the previous run. The raw feed pump generally
primed in about 12-15 minutes. When primed the raw pump is activated,
and the flow meters, chemical feeder (if utilized), skimmers and all
other auxiliary equipment are put into operation. At the end of the
run the system shuts down automatically. All variables are then
selected for the next run and the controls positioned. Variables
associated with tank operation include pressurized flow rate, operat-
ing pressure, scum removal cycle, raw flow rate, and chemical dosage.
-------
FIGURE 13
HAWLEY ROAD OUTFALL
-------
FIGURE 14
INTERIOR VIEWS OF THE CONTROL SHACK
-------
Sampling Procedures
Sampling began when the raw pump primed. Raw waste and screened water
sample collection was started immediately. Effluent sample collection
was delayed for 15 minutes to allow purging of the water in the tank
from the previous run. This procedure insured collection of represen-
tative effluent samples. Screen backwash and floated scum samples were
taken during screen backwash and scum removal periods.
Sampling during 1969 runs was manual. Equal volumes of raw waste were
taken every ten minutes. Screened water and effluent samples were
taken continuously at a rate producing about 2.5 gallons of sample per
hour. Screen backwash and floated scum were composited in equal volumes.
An automatic sampling system was put into operation for all 1970 runs.
The system consists of two timers connected through the proper valving,
to automatically composite the raw waste, screen water and effluent
samples. The first timer controls the sample taking frequency (0-30
minutes). The second timer controls the duration of sampling time (0-60
seconds). The sampling valves are air operated weir valves. The
automatic system has proved much superior to the manual methods, since
the chance for human error has been eliminated and the operator is free
to monitor the remainder of the system. In general samples were compos-
ited every 5 minutes with the automatic system. Floated scum and screen
backwash sample taking was not automated due to the problems associated
with intermittent flows and heavy solids concentrations in these process
streams. The samples were refrigerated immediately after the run.
Analysis were then started within 0-8 hours. Sample analysis procedures
are discussed in Appendix II.
Test Plan
Variables associated with the operation of the demonstration svstem
include hydraulic overflow rate (gpm/sq ft of tank area), pressurized
flow rate, operating pressure, addition of chemical flocculants, and
the floated scum removal cycle. During the first few runs it was
determined that values for operating pressure and the removal cycle
for floated scum were not critical. A test plan was then initiated
with pressurized flow rate, hydraulic overflow rate, and addition of
chemical flocculants as the three variables. Eight possible combinations
of these variables are shown in Table 7. It was planned to obtain 7 to
8 runs for each variable combination requiring 56 to 64 separate runs.
The demonstration system was operated on 55 combined sewer overflows.
All variable combinations of Table 7 were fully evaluated except
Numbers 3 and 4. These combinations were abandoned based on the results
obtained from combinations 1 and 2 which indicated higher overflow rates
without chemicals was not feasible.
51
-------
TABLE 7
VARIABLE
Combination
1
2
3
4
5
6
7
8
COMBINATIONS
Pressurized
Flow as
% of Total
Flow
14-20
21-30
14-20
21-30
14-20
21-30
14-20
21-30
UTILIZED FOR TESTING
Overflow Rate
gpm/sq ft
2.5
2.5
3.8
3.8
2.5
2.5
3.8
3.8
Chemical
Flocculants
No
No
No
No
Yes
Yes
Yes
Yes
-------
SECTION VII
OPERATING RESULTS AND DISCUSSION
Characterization of Raw Waste
As was expected, the quality of the combined sewer overflow from the
Hawley sewer varied widely. In 12 of the 55 overflows reported herein,
extremely high pollutional values were observed. These first flushes
persisted for 20 to 70 minutes. A summary of the first flush data
may be seen in Table 8. The range of the data presented is at the 95%
confidence level. First flush occurrence appears to be associated with
the length of time between overflows and the intensity of the overflow.
The suspended solids concentration for those overflows occurring within
4 days of a previous overflow were calculated to be 151± 29 mg/1 while
the suspended solids for those overflows occurring at an interval larger
than 4 days were found to be 349 ± 80 mg/1. Comparison of the COD data
at intervals shorter than 4 days and longer than 4 days produced CODs
of 144 ± 21 and 394 ± 72 mg/1 respectively. These values are at the
95% confidence level. All the data was tabulated as a function of the
interval between overflows. (Table IV-10, Appendix IV-11.) This data
indicates that essentially all overflows which exhibited the first
flush phenomenon occurred at intervals of 4 days or longer between
overflows. However, all overflows occurring at an interval of greater
than 4 days did not exhibit the first flush phenomenon. Table 9
illustrates this point. All overflows which occurred at intervals of
greater than 4 days are presented in Table 9 along with data on rain-
fall intensity and total rainfall. A total of 23 overflows occurred at
intervals of greater than 4 days between overflows. Of these 23 over-
flows only 12 exhibited the first flush phenomenon. It may be concluded
frotn__tjvis data that the length of time between overflows is related to
the occurrence^f fj^rst~_LLush.&s_and.there are obviously other variables
which strongly influence this phenomenon. These variables could include:
the dry weather flow variation, the intensity of rainfall and runoff
and the sewer system interceptor capacity. Furthermore the data presented
herein may be biased by the 12-15 minutes which was required to prime the
raw pump for the treatment system in that extremely short first flushes
would not have been detected.
A satisfactory method for measuring the rate of flow in the sewer has
not been developed. Development of a method has been complicated due
to the presence of the dam, which was placed in the sewer to impound the
flow. This dam backs water up past one of the two interceptor devices
feeding the Hawley Road sewer. This makes single point gauging impossible.
Another complication which arises is the fact that the dam is not station-
ary.As the flow increases the dam tips to prevent sewer surcharge. This
causes large fluctuations in water level. Down stream of the dam the
sewer has a dog leg and extreme turbulence has been observed at the mouth
of the sewer. Because of the above discussed difficulties reliable data
on the total flow rate in the sewer was not obtained. However, lack of
total flow data does not in any way influence the operation of the demon-
stration system.
53
-------
TABLE 8
SUMMARY FIRST FLUSH DATA
Analysis Concentration mg/1
COD 581 - 92
BOD 186 * 40
Total Solids 861 - 117
Total Volatile Solids 489 * 83
Suspended Solids 522 - 150
Volatile Suspended Solids 308 - 83
Total Nitrogen 17.6 ± 3.1
Ortho Phosphate 2.7 - 1.0
pH 7.0 - 0.1
Coliform Density 142 - IQS x 103
per ml
Data Represents 12 Overflows
93% Confidence Level Range
-------
TABLE 9
SUMMARY OF DATA
AT INTERVALS OF GREATER THAN 4 DAYS BETWEEN OVERFLOWS
Run
No.
691
695
6911
6912
6914
6916
6919
702
708
7011
7013
7022
6918
6922
6923
6925
6928
707
7010
7014
7017
7019
7020
Days
Between
Overflows
18
15
8
5
6
12
24
26
9
14
12
19
11
17
6
11
19
7
11
6
8
18
15
First
Flush
Occurrence
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
No
No
No
No
No
No
No
No
No
No
Average
Rainfall
Intensity
in/hr
Total
Rainfall
(inches)
0.13
0.32
0.40
0.70
0.35
0.40
1.2
0.17
0.84
0.17
0.80
0.10
1.6
0.3
0.5
0.1
0.4
0.11
0.18
0.36
0.15
0.24
0.42
0.25
0.50
1.00
0.17
0.10
0.10
0.3
0.14
0.25
0.20
0.15
0.45
0.10
0.5
0.17
0.12
0.30
0.33
0.22
0.30
0.23
0.25
55
-------
Data on rainfall was obtained routinely and runoff rate estimates were
made using the rational method. These runoff rates were then plotted
against the suspended solids in the overflow. A result of this plot
is presented in Figure 15. It appears from Figure 15 that there was
no relationship between flow rate in the sewer and the suspended solids
in the overflow. This is as expected, since once the sexier system and
drainage basin has been thoroughly flushes, the rate of low flow should
not have a significant effect on overflow quality. However, since actual
sewer flow rates were not obtained, a positive conclusion cannot be made.
After the flushes had passed (if they were present), the characteristics
of the overflow became remarkably stable, considering the large variations
present in the first flushes. The end of the first flushes was determined
by visual observation of the raw combined overflow. The screen backwash
ran continuously during periods when high suspended solids were present
in the raw feed. Thus when the screen backwash pump began cycling the
end of the first flushes was indicated. The exact length of the first
flushes was not recorded. A summary of all data other than first flushes
(termed extended overflows) is presented in Table 10. The data of Table 10
show a relatively small range of values at the 95% confidence level. The
data compared well to data from other research on combined sewer overflow
(6)(11)(12)(13)(14)(65) (66). Jhe pollutant levels in the extended over-
flow are_about what_l£L.expected from a very weak domes.tlc sewaee. One
majbV~difference is the BOD value of 49 mg/1. This is quite low compared
to the COD value of 161 mg/1. Generally, the BODtCOD ratio of domestic
sewage is in the range of 0.6 (67). The dissolved solids in combined
overflow is also quite low. Total dissolved solids (TDS) based on the
data of Table 10, is 212 mg/1. Milwaukee tap water is about 160 mg/1.
This represents only about 42 mg/1 TDS increase as compared to -WOO mg/1
usually added when water is used and discarded as domestic sewage (8).
Of particular interest ot this project is the amount of organic material
present in the dissolved state. Table 11 presents data on this relation-
ship. The amount of dissolved COD ranges from 30 to 38% of the total COD.
The dissolved TOC ranges from 26 to 42% of the total TOC. These values
are somewhat higher than those obtained in the preliminary sampling phase
of 4-25%. The cause of this difference is not known. The dissolved
portion of the overflow will not be removed in the screening/flotation
system, unless it is chemically precipitated. Hence, lower removal
efficiencies will be expected as the dissolved fraction of the waste
increases.
A discussion of the results of the operation of the demonstration unit
follows. A discussion is divided into three phases: screening
operation, flotation tank operation and disinfection. Because of the
wide range of pollutional values obtained in the raw waste, the removals
through various processing steps are presented as percentage removals.
Absolute effuent water quality data may be seen in Appendix IV,
Tables IV-6 and IV-7. Absolute effluent quality can also be estimated
by applying the observed percentage removal figures (Table 12 and 13)
to the raw overflow water quality presented in Tables 9 and 10.
56
-------
BOO
600
iH
"So
CO
a
H
rH
w 400
01
c
a.
09
9
200
0
0
C
0 G
0
©
0
)
0
0
'0*
0 *
0
)
0
0
3 4
in Sewer (10 gpm)
FIGURE 15
RELATIONSHIP BETWEEN FLOW IN SEWER AND SUSPENDED SOLIDS
-------
TABLE 10
SUMMARY OF EXTENDED OVERFLOW DATA
Analysis Concentration mg/1
COD 161 ± 19
BOD 49 ± 10
Total Solids 378 ± 46
Total Volatile Solids 185 ± 23
Suspended Solids 166 ±26
Volatile Suspended Solids 90 ± 14
Total Nitrogen 5.5 ± 0.8
pH 7.2 ± 0.1
Coliform Density per ml 62.5 ± 27 x 103
Data represents 44 overflows
at 95% confidence level range
TABLE 11
PARTICULATE & DISSOLVED RELATIONSHIPS
Relationshln
Dissolved COD/Total COD
Dissolved TOC/Tota! TOC
Dissolved TOC/Dissolved COD
Total TOC/Total COD
No. of
Samples
34
13
14
17
95% Confidence Level
0.34 ± 0.04
0.34 ± 0.08
0.36 ± 0.02
0.33 ± 0.05
Range at the 95% confidence level
58
-------
Results of the Screening Operation
Table 12 presents a summary of the data on removal of pollutants by
screening alone. A listing of all data collected is presented in
Appendix IV. Generally, the percent removals during the first flush
were in the range of 30 to 40%. The confidence ban is wide due to the
relatively small number of first flush overflow occurrences. During
the first flushes a mat of solids sometimes covered the entire screen.
This mat acted as a fine filter medium and undoubtedly, increased the
removal efficiencies as compared to a clean screen. Unfortunately
the headless increased greatly as the mat formed and in runs 695, 6911,
6912, 6914, 6916 and 702, the headloss capacity of the screen was exceeded
for brief periods of time. Examination of these runs indicated that the
screen was removing about 200 mg/1 of suspended solids which was equiva-
lent to 4.9 pounds of dry solids per minute. Based on screen rotation
speed and area, this amounts to a solids loading of approximately 1.2
pounds of dry solids removed per 100 sq ft of screen area. This number
represents the critical level of solids loading at which a headloss of
14 inches will be exceeded.
This value of solids loading is specific for a 14 inch headloss across
the screen. The loading could possibly be increased by increasing the
allowable headloss differential. This could, however, cause a decrease
in removal efficiencies by forcing more solids through the screen and/or
cause a break up of solids which could affect the efficiency of the
flotation process.
The hydraulic flow rate was in the range of 40-45 gpm/sq ft. This rate
depending upon solids loading, could probably be increased.
The design of the screening system should be based on both hydraulic
flow rate and solids loading rate. Either of these variables could
control screen design depending upon waste characteristics. It is
obvious that proper screen design requires additional study. All facets
of screening combined sewer overflows will be studied in Phase VI of
this project which is scheduled for completion in March of 1972.
During the extended overflows and after the first flushes had passed,
removal efficiencies dropped to the 20 to 30 percent levels (Table 12).
The probable cause of this decrease is that essentially no solids mat
formed during extended overflows, and hence, no added filtering action
was obtained.
The screen was backwashed with water from the screened water chamber via
a pressurizing pump. Spray nozzles with 1/4" diameter orifices were
used to effectively distribute the water over the screen media. Washing
was performed from outside the screen drum. Some problems were encountered
with nozzle plugging due to an inefficient seal on the drum screen. The
seal has been improved significantly, but it is still recommended that
an inline small hydraulic cyclone be utilized on future designs to
eliminate any operational problems. The volume of screen water required
59
-------
TABLE 12
POLLUTANT REMOVALS BY SCREENING
Removal During 1 Removal During
Pollutant First Flushes % Extended Overflows %
COD 39 ± 15 26 ± 5
BOD 33 ± 17 27 t 5
Suspended Solids 36 ± 16 27 ± 5
Volatile Suspended Solids 37 ± 18 34 ± 5
1. Represents 8 overflows (see page 53 to 5f> for discussion)
2. Represents 46 overflows
Data at 95% confidence level
60
-------
is approximately 100 gallons per minute. Continuous washing of the screen
would therefore require about 3% of the raw waste flow. Actual spray
wash requirements were in the range of 0.7 to 1.0 percent of the raw flow,
since the screen wash did not run continuously during operation. Screen
wash quality was in the range of 500 to 3000 mg/1 suspended solids.
(Appendix IV - Table IV-9.) Good media cleaning was always obtained and
no permanent media blinding was experienced. In general operation of the
screen was very satisfactory. Tentative design criteria for the screen
system will be presented in a later section of this report and final
specifications at the completion of the project.
Operation of the Flotation System
Overall contaminant removals using the screening and flotation systems
are presented in Table 13. A listing of all data is presented in
Appendix IV. During the first flushes removals of BOD and COD were
in the range of 55 to 65 percent, while suspended solids removals were 70
to 75 percent. Removals of nitrogen were significantly lower at about
46 percent. In those first flush runs where flocculating chemicals were
added the optimum chemical dosage was generally not obtained, since the
pollutant levels were significantly higher compared to extended overflows.
This difficulty illustrates the need for some type of control system
which would automatically adjust the chemical dosage depending upon the
raw feed water quality.
Removals during extended overflows were generally lower than during first
flushes. Removals also varied widely depending upon whether flocculating
chemicals were utilized. Table 13 presents various pollutant removal
efficiencies for three different categories, i.e., without chemical
flocculant addition, with polyelectrolyte (Dow C-31) and clay addition,
and with polyelectrolyte (Dow C-31) and ferric chloride addition.
Figure 16 presents a probability plot of suspended solids removal for
the 1969 data. The slope of the probability line shows a distinct change
at about 50% removal. This change in slope indicates some operating
variable was affecting the removal rates and caused a change in the normal
distribution of the data. This change was chemical flocculant addition.
Examination of the data more closely indicated 10 of 14 runs without
chemical addition were below 52% suspended solids removal, while 10 of
13 runs with chemical addition were above 52% removal. Removals of
volatile suspended solids show trends similar to suspended solids removals.
The effect of optimizing the chemical treatment scheme is clearly illus-
trated in Table 13. Suspended solids and volatile suspended solids
removals increased significantly by varying the chemical addition from
polyelectrolyte and clay (1969 data) to polyelectrolyte and ferric chloride
(1970 data). The chemical dosage required to provide consistent removals
was found to be 20 mg/1 ferric chloride and 4 mg/1 polyelectrolyte. The
ferric chloride is added to the raw waste water flow prior to screening
61
-------
TABLE 13
POLLUTANT REMOVALS BY SCREENING/FLOTATION
Removal During Extended Overflows - %
(2)
Pollutant
COD
BOD
Suspended Solids
Volatile Suspended
Solids
Total Nitrogen
During First
Flushes %(!)
64 ±
55 ±
72 ±
75 ±
46 ±
6
8
6
6
7
Without Chemical
Flocculants
(1969-1970 Data)
41 ± 8
35 ± 8
43 ± 7
48 ± 11
29 ± 14
With Chemical
Flocculants
(1969 DataP>
40 ±
46 ±
59 ±
58 ±
19 ±
14
17
11
10
11
With Chemical
Flocculants
(1970 Data)(4
57 ±
60 ±
71 ±
71 ±
24 ±
11
11
9
9
9
All Data at 95% confidence level
Overflow Rate ^2.5 gpm/sq ft
(1) Represents 12 overflows
(2) Represents 38 overflows
(3) 2.5 - 3.5 mg/1 C31, 6 mg/1 Clay
(4) 3-6 mg/1 C31, 20-25 mg/1 FeCl_
-------
90,
80
70
I 60
50
I
o
01
a
(A
01
£ 30
20
10 20 30 AO 50 60 70 80 90
Probability of Occurrence (%)
98
FIGURE 16
SUSPENDED SOLIDS REMOVAL-SCREENING/FLOTATION
63
-------
and the polyelectrolyte is added to the pressurized flow stream prior to
mixing with the remainder of the screened waste flow.
The BOD and COD data also indicate increased removals when chemical
flocculants were added. An exception to this observation is the COD
removal data with and without chemicals for the 1969 overflows (Table 13).
It is felt that the small decrease in COD removals was due to an increase
in the dissolved organic fraction of the raw combined overflow. During
late summer and fall leaves and other decaying organic material can cause
a substantial amount of organic material to be converted to the dissolved
state. This was very evident in run number 6919 as the appearance of raw
waste after filtering was quite yellow. Run numbers 6928, 6929, and 6930
also had a very high dissolved fraction. This increase in dissolved organics
resulted in the small decrease noted when chemical flocculants were
added, since dissolved organic removals can be accomplished only by precip-
itation. Apparently, the dissolved organic material was not exerted as
BOD (5) possibly due to unacclimated organisms and BOD removals were
therefore not affected to the same degree as COD removals.
A summary of removal efficiencies for particulate organic material is
presented in Table 14. It may be seen that during the 1969 runs,
numbers 6923 through 6929, particulate COD removal ranged from a high
of 96% to a low of 34%. It is obvious from the spread of the data that
effective coagulation was not always being achieved. During the 1970
runs (700 Series) particulate COD removals were much more consistent
reflecting the improved chemical treatment system, i.e., ferric chloride
and polyelectrolyte addition. The average particulate COD removal
efficiency for the 1970 data was 76±8% at the 95% confidence level. This
value compared closely to the suspended solids removals for the 1970 data
of 71±9% (Table 13) thus indicating that effective chemical treatment
was generally obtained during the 1970 overflows. There are, however,
some overflows (7012, 7014, 7015, 7023) where the particulate COD removals
were below the 95% confidence range. This further illustrates the need
for some type control system for metering the chemical addition.
Removal of dissolved organic compounds can only be obtained by precipitation
of the material or adsorption on the flocculated particles obtained when
polyelectrolyte or other flocculating chemicals were added to the raw
waste stream. Removal of dissolved COD for those runs, where data is
available, is presented in Table 14. As expected removal of dissolved COD
was erratic and ranged from no removal to a high of 43% removal. Generally,
the removals were in the range of 20 to 25%. Other chemical flocculants
may give slightly improved removals, but 20-30% removal of dissolved
organics is all that can be expected in a screening/flotation system.
It has been determined from laboratory studies that a flocculation period
may further improve the removal efficiency of the flotation process.
Research is now underway to substantiate these laboratory results at
demonstration system scale.
64
-------
TABLE 14
SUMMARY PARTICULATE AND DISSOLVED ORGANIC REMOVAL EFFICIENCIFS
Chemical Dosage
Run
//(I)
6923
6924
6925
6926
6927
6928
6929
704
705
706
707
708
709
7011
7012
7013
7014
7015
7016
7018
7019
7020
7021
7022
7023
7024
7025
C-31
mg / 1
6.0
0
0
2.7
0
3.0
3.0
6.0
6.0
6.0
6.0
6.0
4.2
5.3
3.9
4.6
3.5
4.5
4.5
4.5
4.5
3.8
3.8
0.5(2)
0.5(2)
0.5(2)
4.0
FeCl3
mg/1
0
0
0
0
0
0
0
0
0
0
0
0
30
17
16
16
21
21
21
21
21
18
18
15
15
25
25
Clay
mg/1
6
0
6
0
0
0
3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Particulate
COD
Removal
%
54
34
86
83
73
57
39
63
95
95
74
85
88
87
53
80
42
50
93
87
77
81
91
88
48
78
69
Dissolved
COD
Removal
%
+2
+14
+25
+22
+23
+23
+23
+11
0
-2
+10
-3
+20
+27
-19
+30
+31
+29
+4
+8
+33
+40
+24
+43
+21
-3
+22
1. Overflow Rate ^2.5 gpm/sn ft
2. Herco Floe 810 used for these overflows
-------
An important variable in the operation of a dissolved-air flotation system
is the volume of pressurized flow required to supply sufficient air bubbles
for proper process operation. For a given volume of liquid, the number
of bubbles and their size will be a function of the operating pressure.
Generally the greater the pressure, the larger the volumes of air dissolved
and the smaller the size of the bubbles. There is of course a limit to
the benefit to be gained by increasing operating pressure. Pressurized
flow volumes generally utilized are in the range of 15-50% of the raw
waste flow. Table 14 presents a comparison of pressurized flow ratios
utilized. Rates from 13 to 22% of the raw flow were considered low values
while rates from 23 to 44% were considered high values. The data is
further segregated into those runs with and without chemical flocculant
addition. A comparison of means test (68) was performed on the high and
low pressurized flow values of Table 15. The procedure for performing
this test may be seen in Appendix IV. The results of this comparison
indicated there was no statistically significant difference between sus-
pended solids removals for low or high pressurized flow whether or not
chemical flocculants were utilized.
It appears considering all data, that a pressurized flow value of 20%
of the raw flow will be sufficient for dissolved-air flotation operation
on combined sewer overflow. A slight increase in pressurized flow may
be required if the improved efficiencies predicted in the laboratorv are
obtained in the field, since a larger volume of solids will require
floating. Air requirements based on a 20% pressurized flow system would
be 1.4 SCFM per million gallons per day of raw flow capacity. An oper-
ating pressure of 50 psig provides sufficient air solution and a small
enough bubble size to be effective without requiring excessive pressure
drops at the pressure reduction value.
Another variable which can significantly affect the removal efficiencies
of a dissolved air flotation system is the overflow rate, i.e., gpm/ sq ft
of tank area. Removal data presented in Table 13 were all at the over-
flow rate of approximately 2.5 gpm/sq ft.
To allow for a study of higher overflow rates, the flotation tank of the
demonstration system was partitioned as shown in Figure 17 for all 1970
runs. All process elements up to the point of entering the flotation
zone were identical to the 1969 runs. A longitudinal baffle was placed
the entire length of the tank. This had the effect of producing two
separate flotation zones. Lateral baffles were then placed at one-half
the length of the tank on side one and at two-thirds the length of the
tank on side two. Effluent samples were taken just as the water flowed
under these baffles to insure collection of a sample representative of
the portion of the tank being utilized. The unshaded area of Figure 17
represents the unused portion of the tank. The raw flow as it enters
the flotation zone is split into equal volumes. Since the effective
length of each side of the tank is different, two overflow rates are thus
obtained on the same storm. This completely eliminates the effect of any
interacting variables like pressurized flow, chemical addition, and the
66
-------
TABLE 15
COMPARISON OF PRESSURIZED FLOW RATES
Low Pressurized Flow
Run
//
6911
6912
6914
6915
6917
6924
6927
7012
Average
6920
6921
6922
6923
6928
6930
707
708
7015
7016
7018
7019
7020
7022
Average
Pressur-
ized
Flow-%
17
21
13
16
16
16
18
17
16.8
21
17
16
20
19
19
22
22
19
20
20
20
19
19
19.5
Chemical
Flocculents
Utilized
No
No
No
No
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Suspended
Solids
Removal-%
30
55
80
53
64
48
61
52
55
70
57
66
52
49
40
68
76
37
85
86
71
85
85
66
Run
//
692
693
695
696
697
699
6910
6913
6917
6925
Average
6919
6926
6929
704
705
706
709
7011
7013
7014
7021
7023
7024
Average
High Pressurized Flow
Pressur-
ized
Flow-%
24
29
34
29
31
23
23
33
23
29
27.8
28
31
27
37
44
40
28
24
23
28
23
25
27
29.6
Chemical
Flocculents
Utilized
No
No
No
No
No
No
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Suspended
Solids
Removal-%
40
65
44
34
28
40
32
63
64
46
46
65
75
45
77
79
80
82
69
72
68
72
49
66
69
NOTE: Overflow Rate @ ^2.5 gpm/sq ft
-------
Pressurized
Flow System
Effluent Sample
Collection Point
30
Sidewater Depth 8.5*
Effluent Sample
Collection Point
9'
9'
FIGURE 17
BAFFLE ARRANGEMENT FOR OVERFLOW RATE TESTS
-------
differences inherent in any two combined overflows, allowing direct
determination of the effect of overflow rate.
Data taken on direct comparison of overflow rates are presented in Table 16.
The data indicates removal efficiencies decrease as the overflow rate is
increased. Data from each run is presented, as well as average values
for low and high overflow rates along with the 95% confidence range.
By comparing the mean values, removal efficiency decreased approximately
5-142 as the overflow rate increases from ^2.5 to 3.75 gpm/sq ft.
A paired comparison test (Appendix IV) was also run on the data of Table 16.
The results of this test indicated that there was a difference in the low
and high overflow rate data. Confidence levels at which the difference
was significant are as follows: COD - 90%, BOD - 95%, suspended solids -
99% and volatile suspended solids - 95%. It is anticipated that at
higher overflow rates the decrease in removal efficiencies will be even
greater. However, higher rates will be evaluated during this study.
Based on the data of Table 16 an overflow rate of 3.3 gpm/sq ft is recom-
mended as the basic design value.
Of considerable importance in the operation of a screening-flotation
system is the volume of residual solids produced during operation.
Volumes of floated scum generated during operation of the demonstration
unit ranged from 0.75 to 1.41 percent of the raw flow at the 95%
confidence level. Scum concentration was generally in the range of 1 to
2% solids on a dry weight basis (Appendix IV, Table IV-9). The floated
scum and screen washwater from this project are disposed of via an inter-
ception sewer which directs them to the sewage treatment plant for
ultimate disposal. Since gravity flow is utilized it is desirable to
limit the sludge concentration to about 2% as this concentration easily
flows by gravity. Higher concentrations can be obtained by skimming the
floated scum blanket at less frequent intervals. Sludge concentrations
as high as 4% solids have been obtained in this manner. Obviously the
higher the concentrations, the smaller the volume of sludge produced
for ultimate disposal. For this reason it is felt that in full scale
operations sludge volumes will be somewhat less than those obtained
during this study. Another factor, which will offset this reduction in
scum volume, is the addition of more effective chemical flocculants.
As the chemical flocculant addition is optimized, the volume of scum
will increase due to the additional chemicals added and the higher
efficiency of solids capture.
Ultimate solids disposal will dictate the desired solids concentration
in the floated scum and, hence, the volume of scum which requires disposal.
If scum is disposed of by way of an interceptor sewer a solids content in
the range of 1-2% will be desirable. This should result in an average
scum plus screen wash volume of 1.75 percent of the raw flow. If disposal
is by tanker truck or solids dewatering at the site, a high solids
concentration is desired (4%), and the volume should be less than 1 percent
of the raw flow. Additional discussion on solids handling is presented
in the full scale design consideration section of this report.
-------
TABLE 16
COMPARISON OF THE EFFECT OF OVERFLOW RATE ON REMOVAL EFFICIENCIES
Run
Low Overflow Rate -\-2.51
n
f
704
705
706
707
708
709
7011
7012
7013
7014
7015
7016
7020
7021
7023
7024
7025
Average
95%
Confidence
Limits
COD
35
72
76
52
67
66
63
41
56
36
43
74
74
58
37
61
50
56.5
49-64
BOD
69
69
75
61
59
51
60
36
55-
37
77
76
51
52
62
59.3
52-66
SS
77
79
81
68
76
82
70
52
72
68
52
85
85
72
49
66
59
70.2
64-76
VSS
90
79
73
72
77
76
76
58
80
68
47
84
86
70
49
61
61
70.7
64-77
High Overflow Rate ^3.75
COD
28
74
72
43
62
61
60
36
48
29
39
71
68
67
46
51
60
53.8
46-62
BOD
55
69
54
58
36
51
64
31
46
40
45
80
46
46
64
52.3
45-60
SS
33
79
66
50
66
76
74
36
66
66
48
83
63
67
35
56
68
60.7
53-69
vss
51
80
64
62
71
73
79
56
69
68
27
82
69
68
36
70
70
64.4
57-72
1. Actual Range 2.43 - 3.1 gpm/sq ft
2. Actual Range 3.4 - 4.5
3. Removal in %
NOTE: chemical flocculants added for all runs
-------
During the operation of the demonstration system some settled sludge
was noticed. The amounts of settled sludge were, however, extremely
small, with one exception. Runs number 691 and 692 were made without
using the drum screen. Upon draining the tank after run number 692,
about 4-6" of settled material was discovered. This consisted of grit,
twigs and other material which could not be floated. Later in the year
the tank was again drained after 13 additional runs which were all made
with the screen in service. Settled sludge volume amounted to only
2-3 inches at this draining. This clearly demonstrates the value of
the screen in removing heavy particulate matter which cannot be floated
in the flotation tank. The tank was again drained at the end of the
1969 season and sludge volumes were again only 2-3 inches. It may be
concluded that bottom skimming systems are not required when screening
precedes flotation. If, however, applications arise where flotation is
used alone, it is imperative that some type of bottom skimming be provided
or large sludge blankets will form which will eventually scour into the
effluent.
Material balances for suspended solids were made for almost all runs.
The tabulated data may be seen in Table IV-14, Appendix IV. The balance
of solids in and solids out of the system generally totaled within about
25%. In some cases, however, the error was higher. The probable
reasons for the material balance errors are believed to result from scum
sampling and collection procedures. At the end of a run, the floated
scum blanket which remained on the tank was lost with respect to material
balances. Generally this scum was skimmed off after the run when the
scum meter was not totalizing. It was also difficult to obtain a
representative sample of scum, since scum was scraped intermittently
and the solids concentration could vary widely from the beginning to the
end of the skimming cycle. Scum samples were taken manually and if other
operating problems arose, a scum sample from every skimming cycle may
not have been obtained. Because of the sampling procedures it is felt
that scum suspended solids concentrations were not representative. Since
suspended solids in the scum could account for a significant portion of
the solids balances, it is felt that these errors contributed to a large
extent in the material balance discrepancies.
Difficulties were also experienced with the raw flow meter totalizer.
The units on the totalizer were 30,000 gallons per count and thus could
not be read accurately. The totalizer has subsequently baen changed and
thus this problem eliminated. Another possible source of material balance
error is the fluctuating concentration of the raw overflow. Limited data
has been collected to indicate that the pollutant concentration can vnry
significantly during an overflow. This variation could introduce
errors in the material balance.
The maintenance of the demonstration system has been limited to lubrica-
tion of moving components. The following discussion will be limited to
those problems which were encountered and the method of solution. Some
difficulties were encountered with the float switches utilized in the
system. These switches are the reed type which are activated by a sliding
71
-------
float. They required frequent cleaning, once per week, as sand would
work its way between the float and the shaft and caused the float to
become inoperable. Replacement of this type switch with a conductivity
type should solve this problem.
Difficulties were also encountered with the back pressure control valve
on the pressurized flow system. This problem resulted from excess
moisture in the compressed air. The valve positioner apparently could
not tolerate the moisture level encountered. All problems were in the
positioner for the valve which accepts a 0-20 psi control pressure and
positions the valve accordingly. The positioner was utilized only because
of the design flexibility required in a demonstration system. In the full
scale applications when the pressurized flow rates are known, valve posi-
tioners will not be required and the problems incurred will be eliminated.
A minimum amount of difficulty was experienced with the chemical metering
pump, but this was a result of improper materials selection and the
difficulties were easily eliminated. As may be concluded from the above
discussion, the operation problems encountered were minimal and this
attests to the soundness of the system design.
Disinfection of Combined Overflows
Another important aspect in the treatment of combined sewer overflows
is adequate disinfection. The screening/flotation system provides
sufficient detention time (12-20 minutes) to obtain the necessary chlorine
contact time. The system also is flexible in that various points of
chlorine addition are available depending upon process needs. A summary
of the disinfection data taken during the project is presented in Table 17.
Hypochlorite salts were utilized as a source of chlorine. Dosage was
held essentially constant at 10 mg/1. It was noticed, however, that the
strength of the chlorine stock solution could decrease rapidly depending
upon environmental conditions. Therefore, some of the storms could have
had dosages less than 10 mg/1. In general, relatively good disinfection
was obtained. A trend seems to exist in Table 17 which indicates that
as the coliform density increases the absolute coliform density in the
effluent also increases. Nevertheless good disinfection can be obtained
in conjunction with the operation of a screening/flotation system.
Conceptual Design
This section of the report deals with the design concepts necessary to
utilize a screening/flotation system on a full scale basis for treating
combined sewer overflow. It is not the intention of this discussion to
provide specific answere to all design details. Sufficient information
will be presented to form a basis for process design. Various engineering
considerations and/or judgements must be provided to produce the final
overall system design. It should also be recognized that this project
is still underway and that later data may indicate changes in design
philosophy.
72
-------
TABLE 17
SUMMARY OF DISINFECTION DATA
Run
No.
695
696
697
698
699
6910
6911
6912
6913
6919
6920
6921
6922
703
704
706
707
708
PF
EFF
INF
Chlorine
Dosage
mg/1
10
10
10
8
10
10
10
10
10
10
10
10
10
10
10
10
10
10
= Chlorine added
= Chlorine added
= Chlorine added
Point
of
Addition
PF
PF
PF
PF
PF
PF
PF
PF
PF
EFF
EFF
EFF
EFF
INF
INF
INF
INF
INF
Detention Influent
Time Coliform per
min 100 ml x 10^
20
20
20
20
20
20
20
20
20
10
10
10
10
21
21
21
21
21
in pressurized flow line
to effluent from flotation
to raw waste
prior to bar
36
5.7
1.3
7.8
6.2
19
20
65
38
310
160
55
82
270
12.2
0.7
340
110
basin
screen
Effluent
Coliform
per 100 ml
<100
<4
<4
<4
2
<2
10
<5000
<5000
60,000
40,000
<7
15000
<1000
<100
200
3800
18000
Note: Samples were dechlorinated when necessary using Na2S03.
73
-------
The overall system can be divided into various subsystems and these
subsystems are listed and discussed below:
1. Pumping system
a. Self cleaning bar screen
, b. Pump pit
c. Variable rate pumping system
2. Screening/flotation system
a. Drum screens
b. Pressurized flow system
c. Sludge collection
d. System tankage
e. , Flocculators
3. Solids slurry storage
a. Mixers
b. Transfer system
c. Dewatering (if required)
4. Chemical Addition
a. Chemical storage
b. Mixing system
c. Metering pumps
d. Disinfection
5. Control Systems
Pumping System
The pumping system is obviously a vital subsystem in the overall treat-
ment concept. It is envisioned that a sump will be utilized to provide
the necessary pump suction conditions. A self cleaning bar screen will
be utilized just up stream of the sump to remove large objects which will
not pass the pumps and/or could cause problems down stream in the treat-
ment system in such areas as the drum screens or sludge removal systems.
Bar spacing is recommended at 1/2" as this value proved adequate in the
demonstration system. A medium duty bar screen will have sufficient
strength to handle the imposed solids loadings.
The pumping system must be able to handle the variable rates whi,ch will
be encountered during operation. The pumps therefore must be controlled
automatically to vary the pumpage as the raw flow varies. There are
numerous pumping system configurations of this type which have been
employed in the past for storm water pumping, municipal waste treatment
and industrial processing. Many pump manufacturers offer package pumping
systems (69) which would meet the required design consideratins. There-
fore no detailed information on pump station design will be presented herein.
-------
It should be noted however that open type screw pumps seem to be ideally
suited to combined sewer overflow treatment systems, since they can
provide variable flow rate from a constant speed motor. Since any system
must be designed for a specific hydraulic flow rate, an overflow structure
should be provided for the pumping system. Considerations involved in
overflow design should include utilization of the sewer system for maximum
storage of raw overflow. This will provide a damping effect on the raw
flow rate, and minimize control problems in the treatment system. Proper
design of the overflow will provide satisfactory treatment system operation
with a minimum of excess overflows. Provisions should also be made for
sump cleaning, since large volumes of gravel and grit are anticipated.
Overall pump system design should provide for ease of cleaning and maintain-
ing the pumps. No problems were encountered during operation of the demon-
stration syste, with regard to feed pumps. The demonstration system,
however, had only a single feed pump and the flow rate did not vary during
operation. Since the screening flotation system can function at overflow
rates higher than the design value, the pumping system should be designed
with this fact in mind and sufficient capacity should be provided. A
pumping capacity of one and one-half the flotation design capacity should
be provided, since the flotation unit can handle this additional volume
for short periods with only moderate decrease in removal efficiencies.
It may be possible in some applications to utilize gravity flow. In this
case only sluice gates will be required to direct the raw flow to the
various screening/flotation modules. The gates should be automatically
positioned to provide proper control. Consideration of head loss across
the screen becomes more critical when contemplating a gravity feed design.
It may also be possible in some applications to utilize gravity feed and
effluent pumping. This approach should provide the most maintenance free
pumping system since the effluent water will be low in suspended solids.
When gravity feed systems are utilized, a sump or stilling basin will
still be required to provide proper hydraulic control, and the bar screen
is also required for system protection. Grit accumulations are expected
to be greater in the gravity flow systems, since turbulance from the pump
suction will not be present.
Screening System
The screening system consists of basically two sections, the drum screen
and the backwashing system. The critical design parameters associated
with the drum screen include hydraulic loading, solids loading, and head
loss capability. The hydraulic loading is a function of the wetted screen
area. ' This area varies only slightly as the head loss across the screen
increases during operation. The hydraulic loading is not affected by
rotation speed. Generally a drum screen can operate with up to 70% of
the screen surface submerged. Higher submergence is not recommended due
to possible flooding of the backwash water removal hopper. Hydraulic
loading in the range of 25-45 gpm/sq ft of screen area was utilized in
the demonstration system for a 50 mesh, 297\i opening screen media.
75
-------
2
A hydraulic loading of 40 gpm/ft and a screen size of 50 mesh are recom-
mended at this time. Additional data on screen mesh size and hydraulic load-
ing are being obtained in Phase V of this project. Upon completion of this
phase, extensive information on screen design will be available.
Drum rotation speed controls the solids loading on the screen. It was
found during this study that a solids loading of 1.2 pounds of dry solids
removed per 100 sq ft of screen media produced a head loss of about 13 inches
of water. Drum rotation speeds were in the range of 1 to 5 rpm. Solids
loading may be calculated with the following equation:
Ls = FsR/rAD
where:
Ls = Solids loading per 100 sq ft
R = Screen removal efficiency (%)
FS = Feed solids into screen (Ibs per rain)
r = Drum rotation speed rpm
Ap = Total surface area of screen
As may be seen from the equation solids loading is directly proportional
to solids removal efficiency and inversely proportional to drum rotation
speed, A removal efficiency (R) of 35% is recommended based on the results
obtained in the demonstration unit. Drum rotation speeds of 2 to 12 rpm
are also recommended.
It is apparent that selection of drum area is controlled by two criteria;
hydraulic loading and solids loading. Either of these two variables may
control the screen area requirements. If a solids loading of 1.2 Ibs per
100 sq ft is utilized resulting head losses will be in the range of 12 to
14 inches of water. Higher head losses and subsequently higher solids
loadings may be possible without a significant effect on process efficiency.
Data to support this statement are now being obtained. The inside of the
drum should be fitted with an angle iron system (similar to a roto-dip
feeder), which will pick up any solids which will not adhere to the screen
media and would not be carried up to the spray washing system.
The screen cleaning system consists of a pump, a header system and spray
nozzles. A hopper inside the drum collects the material as it is flushed
from the screen. Required screen wash water rates amount to 15 gpm per
foot of drum length. This volume is sufficient to clean a completely
blinded screen. The spray nozzles utilized should provide a low mechanical
pressure on the screen media, but still provide good washing capability.
The nozzles should have as large an opening as possible consistant with
a good spray pattern. The nozzles utilized on the demonstration unit had
-------
egg shaped openings with the smallest dimensions being 1/4 of an inch.
They were positioned about 12" from the drum surface and provided
excellent cleaning capability. They exerted an average pressure of
less than 1 psi on the media.
Some problems were encountered with nozzle plugging due to an inefficient
seal between the raw and screened water. Plugging occurred mostly from
gravel or grtt with some plugging due to string and twigs. The drum seal
has subsequently been improved, but to eliminate a possible plugging problem,
a small wet cyclone (six inch diameter) should be utilized to trap any
material which could cause plugging. The dirty water discharge from the
cyclone should be routed back through the screen.
It is recommended that screen rotation be continuous. This eliminates
rapid change in head loss when the clean screen enters the water, which
could result in an incomplete washing cycle. If the drum were stationary,
the portion of media not in contact with the flow would initially be clean.
As the head loss reached the limit of the backwash actuating system and
the drum was activated, it would rotate only until sufficient clean screen
area was wetted to reduce the head loss to normal values and the washing
system would then be deactivated. This could leave portions of the screen
unwashed and cause rapid head loss flocutations. When the rotation is contin-
uous, a solids layer gradually builds up over the entire drum surface.
The back wash system is then activated and cleans the entire screen before
deactivating. It is recommended that the spray water cleaning system be
activated by monitoring differential pressure, i.e. head loss across the
screen. Conductivity probes can be utilized, and there are a number of
commercially available, inexpensive units on the market. The system
consists of two probes and an electrical circuit board with relays. The
probes are mounted in the tank at the desired differential elevation.
When the water contacts both probes, a relay is tripped which can be used
to activate the backwash pump. The relay remains closed until the water
level is below the lower probe. The differential chosen should be about
50-60% of the expected head loss across the screen, i.e. for a 2' head loss
the spray system should activate at 1 to 1.2 foot of head loss. This
prevents the screen from becoming completely blinded prior to activation
of the back wash system. The lower probe should be set two to three inches
higher than the head loss level when the screen is completely clean. This
will prevent unnecessarily long back wash cycles.
The media utilized in the demonstration system was type 304 stainless steel.
This media proved adequate. Because of the near neutral pH encountered in
the demonstration runs (6.8 - 7.5), brass screen which is considerably less
expensive could be utilized. During the two years of operation reported
herein some bacterial growth on the screen has been experienced. It was
found that a rinsing with chlorine solution would remove these growths.
The control of the feed water flow into the drum screen is extremely
important. Since the possibility exists, with this highly variable raw
waste, that the head loss capacity of the screen may be exceeded during
the heavy solids loading peaks. The system should therefore have an automa-
tic by-pass feature, which will allow relief of the screen if the head loss
77
-------
capacity is exceeded. The water which is bypassed during these times can
be routed to the flotation zone. It is also recommended that a head loss
capacity of 24 inches of water be provided.
Figure 18 presents a sketch of the recommended screen configuration. This
sketch illustrates the important features of the screen installation which
have been discussed above.
Flotation System
The design of the flotation system requires the consideration of the follow-
ing components: pressurized flow system, flotation basin and scum removal
system. Details on design recommendations will be discussed in this section
along with a proposed system arrangement.
The pressurized flow system is the heart of the dissolved air flotation
process. It includes a pump, air solution tank, pressure reduction valve,
source of compressed air and suitable control systems.
The pressurized flow system should be designed to provide a volume equal
to 30% of the raw flow rate into the system. This should provide an
adequate margin of safety since testing at the demonstration site indicated
20% was sufficient. The recommended design operating pressure is 50 psig.
The air solution tank should provide maximum air water interface to obtain
high air solution. Some air solution tanks are packed with a tower packing
to provide this interface. For treatment of combined overflows, however,
a packless tank is recommended. A packed tank would be very susceptible
to plugging due to the solids present in the waste stream. Packless tanks
are generally fitted with an internal baffle to promote air water interface.
Nominal detention time in the tank is generally in the range of one minute.
The tank should also have some method of controlling the water level, since
only about 20% of the tank volume is filled during operation.
The pressure in the system is controlled via an adjustable valve. Weir
type valves have been utilized successfully. The valve is positioned by
way of a pneumatic controller which allows automatic control of the system
pressure. The valve besides providing pressure control, provides the
necessary shear forces to promote proper bubble formation.
The sizing of the flotation basin is based mainly on the American Petroleum
Institute (API) standards (64). Based on operation of the demonstration
system, some modifications have been made to this procedure. The API design
criteria are listed below as follows:
L = 1.2 FVhd/Vt
where: L = effective tank length (ft)
Vt = rate of rise of particles (fpm)
78
-------
Backwash
Header
Lifting Flights for
Solids Not Adhering to Screen
Screen Backwash Header
Backwash
Actuators
A A A A
Screened
Solids
Hopper
Screened Solids Discharge
Screened Water
FIGURE 18
RECOMMENDED SCREEN ARRANGEMENT
-------
Vh = horizontal velocity in tank (fpm)
d = effective tank depth (ft)
F = (0.026 Vh/Vt) + 0.995
V^ maximum = 15 V or 3 fpm
tD (tank detention) = 10 minutes minimum
d (depth of tank) = 0.3 to 0.5 of tank width
minimum depth = 3'
The particle rise rate can also be expressed as a surface loading
rate, i.e.,
SL = (Vt)7.48
where S^ = surface loading gpm/sq ft and Vt is in fpm
The demonstration system design was based on the API procedures listed
above. A particle rise rate of 0.4 fpm (SL = 3 gpm/sq ft) was utilized
based on laboratory tests. The tank was also designed to allow changing
the overflow rate. It is obvious from the above discussion that the
particle rate of rise is critical in the flotation basin sizing. It may
also be seen that the horizontal velocity will be at the maximum value
(3 fpm) for all rise rates in excess of 0.2 fpm. The particle rise rate
is dependent upon particle diameter to the second power and the apparent
difference in density of the particle bubble combination and suspending
fluid to the first power. Little can be done to control the latter, but
the particle size can be significantly affected by proper chemical addition
and flocculating procedures. Rise rates as great as 5 fpm have been
obtained in the laboratory, indicating large floe particles can be
developed.
It is recommended that the above discussed procedures be utilized to size
the flotation basin. A particle rise rate of 0.45 (SL =3.3 gpm/sq ft)
should be utilized. This recommendation may be modified, depending upon
results obtained during the remainder of the project.
Overhead skimmers are provided to remove the floated scum. Bottom skim-
mers are sometimes utilized on flotation systems to remove any sludge
which may possibly settle. If screening (50 mesh or finer) is utilized
in the system, bottom skimming is not recommended since the small amount
of settled sludge expected can be removed while draining the tank between
storms. If, however, flotation is utilized without screening, bottom
skimming will be required. Removal of scum should be controlled by a
80
-------
timed cycle or the sensing of a sludge blanket. This will allow sludge
to be removed only when required and hold to a minimum the volume of
floated scum which will require ultimated disposal.
The recommended general arrangement for the screening/flotation system
is presented in Figure 19. It is felt that this configuration will
meet all process requirements and be the most economical svstem.
The additional subsystems required to complete the treatment system are
sludge storage and the chemical storage and feeder units. A possible
overall site arrangement including these subsystems is shown in Figure 20.
Sludge storage should be sufficient to handle 2% of the raw flow for the
design storm period. This should provide sufficient storage volume,
however, the ultimate sludge disposal method may dictate the desired
storage volume. Sludge disposal alternatives available include tanker
trucking, providing a truck mounted vacuum filter to reduce the slurry
to a cake for hauling, or disposal by pumping to an interceptor sewer
for transporting to the sewage treatment plant. If the latter method is
used, the interceptor obviously must not have possible overflow points
in its course to the sewage treatment plant. If disposal via interceptor
is utilized a sludge pump will be required and essentially no solids
storage pit will be necessary. On the other hand, if it is not feasible
to pump solids into the interceptor during a storm, they may be stored
at the site and pumped to the interceptor after the storm has passed.
Regardless of which method is selected, proper attention should be given
to the problems associated with handling sewage solids slurries.
The chemical treatment module (Figure 20) should provide for storage of
flocculating chemicals (ferric chloride and a polyelectrolyte) as well
as chlorine. A sufficient supply of chemicals consistent with the require-
ments of the system should be provided. Estimated chemical dosages are:
ferric chloride - 20-25 mg/1, polyelectrolyte - 5 mg/1, and chlorine -
10-15 mg/1. The chemical treatment module also houses the proper feeder
systems to accurately dispense the chemicals into the wastewater flow.
The entire system detailed in Figure 20 should be automated 100%. This
will allow remote monitoring and control of the treatment facility at a
central location. However, the maintenance involved with automatic
systems which are used intermittently will have to be tolerated. It
is felt that this maintenance will not be of such a magnitude to negate
the advantages of automation.
Economic Considerations
There are many factors which must be considered when estimating the
capital costs of a combined sewer treatment facility. The basic areas
of consideration for a screening/flotation system are listed below:
1. Screening/flotation system including
self cleaning bar screen
variable rate pumping system
81
-------
Drum Screen
Feed
Channel
cc
ro
Screw Convenor
Scum Collector
-> r
"~^T^
Mixing
Zone
Flotation Zone
o
Pressurized Flow Header
FIGURE 19
Effluent
Weir
RECOMMENDED SCREENING/FLOTATION ARRANGEMENT
-------
Automatically Cleaned
Bar Screen \.
Influent
Sewers
Pumping System
Flotation Tanks
(0
0)
01
>j
u
v:
(0
CQ
J3
H
Ui
4J
CO
1-1
Q
n)
oa
o
3
XI
H
M
H
Q
V)
0)
01
1-1
u
E
t-l
a
Solids
Slurry
Storage
c
0)
0)
4J
(0
^
to
Chemical
Storage
And
Pumping
System
FIGURE 20
OVERALL SYSTEM CONFIGURATION
83
-------
solids storage and disposal
instrumentation
control building
erection costs
2. Sewer interconnection (i.e. combining a number of overflow
points to reduce the number of treatment sites) and outfall
facilities
3. Land costs
4. Special design problems
foundations (soil problems)
ground water
special construction techniques
freezing problems
5. Engineering costs and fees
Generalized cost estimates for Item 1 can be made since these costs are
now known. Cost estimates for Items 2, 3, 4, and 5 are highly specific
for the selected treatment sites, and cannot be generalized. Total
installed costs for the screening/flotation system is estimated at
$19000 per MGD capacity for 10 MGD plants and $15,000 per MGD for 45
MGD and larger plants. This price includes system tankage in concrete,
all necessary hardware, pressurized flow system, variable rate pumping
system and scum collection and storage. The foundation is assumed to be
a simple slab, and solids disposal by gravity drainage into an intercep-
tor sewer.
The total installed cost for the demonstration facility was approximately
$90,000 or $18,000 per MGD capacity for the 5 MGD plant. It should be
noted that the demonstration system did not have a self cleaning bar
screen, a variable rate pumping system, or solids storage facilities
which would increase the cost per MGD capacity. On the other hand the
demonstration system was of all steel construction and only 5 MGD
capacity, larger capacity and concrete construction are expected to
decrease the cost per MGD capacity.
Based on the above generalized information, the cost of providing a
screening/flotation system to treat the entire Hawley Road drainage area
(^500 acres) is presented below. It was assumed that the design storm
was the once in 5 year storm. From Milwaukee rainfall records this
intensity is approximately 1.0 inch per hour for the estimated time of
concentration of the drainage area of 100 to 110 minutes. This amounts
to an overflow rate of approximately 120 MGD. Allowing a 33% surge
factor for this peak storm results in a design capacity of 90 MGD.
The following costs are estimated for this system capacity.
1. Screening/flotation system $1,350,000
2. Sewer interconnection and outfall facilities $ 65,000
3. Land costs (1.5 acres @ 100,000/acre) $ 150,000
4. Special design problems $ 90,000
84
-------
5. Engineering costs and fees $ 240,000
Total $1,895,000
Cost/MGD $ 21,056
Cost/acre $ 3,828
A detailed design and cost estimate for a midwest city indicated a
system cost of $22,000 per MGD capacity which compares favorably with
the above estimate for the Hawley Road drainage basin.
The operating costs for a screening flotation system include power,
chemicals, and maintenance. Table 18 presents a summary of the estimate
operating costs. The total operating cost is 3.09C/1000 gallons of
treated overflow. The majority of this cost is for chemical addition,
i.e. 2.Sic/1000 gallons. These costs are based on operation of the
demonstration system at Hawley Road and are not expected to vary signif-
icantly with plant size.
85
-------
TABLE 18
OPERATING COST ESTIMATES
Item
Power
FeCl3(1)
Polyelectrolyte/]\
Chlorine
Labor
,~N
Cost
C/1000
Parts
Quantity
15 KW/MGD
20 mg/1
4 mg/1
15 mg/1
30 hr/mo
$300/mo
Unit Cost
1.5C/KWH
4.5c/lb
35C/lb
4.7c/lb
$10/hr
gallons
0.54
0.75
1.17
0.59
0.02
0.02
TOTAL COST = 3.09C/1000 gal
(1) Based on carload prices
(2) Based on 90 MGD system 40 hours per month operation
86
-------
SECTION VIII
ACKNOWLEDGEMENTS
The cooperation of the City of Milwaukee on this project is deeply
appreciated. Messrs. Herbert A. McCullough, City Engineer, and
Gilbert Vosswinkel's assistance in site selection and review of
construction plans allowed timely completion of the construction of
the demonstration facility.
Many people at the Rex Ecology Division contributed to the success of
this project. Design of the demonstration system was made by J. E.
Milanowski. Operation of the unit at any time day or night was performed
by M. K. Gupta, F. Toman, and D. G. Mason. Excellent cooperation was
provided by the process equipment laboratory, headed by R. W. Wullschleger,
with regard to timely completion of the large numbers of analyses which
were required.
The principal author of the report is Donald G. Mason. Valuable assis-
tance in preparation of the literature search and the chemical oxidation
studies sections of the report was provided by M. K. Gupta.
Encouragement and assistance from EPA personnel, i.e. Messrs. William A.
Rosenkranz, Frank Condon, Ralph Christensen and Clifford Risley is also
deeply appreciated.
87
-------
SECTION IX
REFERENCES
(1) Romer, H. and Klashman, L., Public Works 94:3:100 (1963).
(2) Camp, T.R., Sewage and Industrial Wastes 31:4:381 (1957).
(3) Ben.les, H.H., WPCF Journal 33:12:1252 (1961).
(4) Palmer, C.L., WPCF Journal 35:2:162 (1963).
(5) McKee, J.E., Journal Boston Society Civil Engineering 34:2:55 (1947).
(6) Palmer, C.L., Sewage and Industrial Wastes 22:2:154 (1950).
(7) Johnson, C.F., Civil Engineering 28:2:56 (1958).
(8) Fair, G. and Geyer, J., Water Supply and Waste Disposal, John
Wiley & Sons, Inc., New York, 1961.
(9) Eckenfelder, R.B. and O'Conner, D., Biological Waste Treatment,
Pergamon Press, 1961.
(10) Camp, T., Journal ASCE, Sanitary Engineering Division 87:SA1:1 (1961),
(11) Dunbar, D. and Henry, J., WPCF Journal 38:1:9 (1966).
(12) Benzie, W. and Courchaine, R., WPCF Journal 38:3:410 (1966).
(13) U.S. Department of Health, Education and Welfare, "Pollutional
Effects of Stormwater and Overflows from Combined Sewer Systems",
November, 1964.
(14) Riis-Carsten, E.. Sewage and Industrial Wastes 27:10:1115 (1955).
(15) Romer, H. and Klashman, L., Public Works 94:4:88 (1963).
(16) Romer, H. and Klashman, L., Public Works 92:10:129 (1961).
(17) Stegmaier, R.B., Sewage Works Journal 14:6:1264 (1942).
(18) Steel, E.W., Water Supply and Sewerage, McGraw-Hill, 1960.
(19) Boruff, C.S. and Buswell, A.M., Sewage Works Journal 4:6:973 (1932).
(20) Flood, F.L., Sewage Works Journal 3:4:223 (1932).
(21) Rudolfs, W. and Hesig, H., Sewage Works Journal 1:5:519 (1929).
(22) Muldoon, J.A., Sewage Works Journal 11:6:1054 (1939).
89
-------
(23) Peterson, K., "Sludge Thickening by Screens", Presented at the
California Water Pollution Control Federation Conference,
April-28, 1966.
(24) Boucher, P.L. and Evans, G.R., Water and Sewage Works 109:11:R184
(1962).
(25) Lynara, B.T., Ettlet, G., Me Aloon, T., WPCF Journal 41:2;247 (1969).
(26) Keilbough, W.A. , et al., "Microstraining with Ozonation or
Chlorination of Combined Sewer Overflows", Preliminary Report by
Cochrane Division, Crane Company for presentation at FWPCA
Seminar, Edison, New Jersey, 1970.
(27) Chase, E.S., Sewage and Industrial Wastes 30:6:783 (1958).
(28) Rohlich, G.A., Industrial and Engineering Chemistry 46:2:83 (1954).
(29) Geinopolos, A. and Katz, W.J., Chemical Engineering 75:20:78 (1968).
(30) Van Vuuren, L.R.J., et al., International Journal of Air and
Water Pollution 9:823 (1965).
(31) Van Vuuren, L.R.J., £t al., Water Research 1:463 (1967).
(32) Katz, W.J., "Principles of Flotation Design", part of a panel
panel discussion on flotation, presented at the Fall Meeting of the
Committee on Waste & Disposal of the American Petroleum Institute,
Denver, Colorado, September 15, 1954.
(33) Katz, W.J., "Solids Separation Using Dissolved Air Flotation",
presented at the Air Utilization Institute, University of
Wisconsin. April 15, 1958.
(34) Vrablic, E.R., "Fundamental Principles of Dissolved Air
Flotation of Industr,ail Wastes", Proceedings of the 14th
Industrial Waste Conference, Purdue University, 1959.
(35) Berman, R.I. and Osterman, J., "Dissolved Air Flotation for White
Water Recovery", Proceedings of the 10th Industrial Waste
Conference, Purdue University, May, 1955.
(36) Katz, W.J., Petroleum Refining 37:5:32 (1958).
(37) Wollner, H.J., ejt al. , Sewage and Industrial Wastes , 26:4:309
(1954).
(38) Sessler, R.E., "Wastewater Use in a Soap and Edible Oil Plant",
Sewage and Industrial Wastes 27:10:1178 (1955).
90
-------
(39) Osterman, J., Waste Engineering 26:2:37 (1955).
(40) Balden, A.R., WPCF Journal 41:11:1912 (1969).
(41) Katz, W.J., and Rohlich, G.A., "A Study of the Equilibria and
Kinetics of Adsorption by Activated Sludge", Biological Treatment
of Sewage and Industrial Wastes by McCabe, J. and Eckenfelder,
W.W., Volumn I, p 66, Reinhold, New York, 1956.
(42) Ettelt, G.A., "Activated Sludge Thickening by Dissolved-Air
Flotation", Proceedings 19th Purdue Industrial Wastes Conference,
' p 210-244, 1964.
(43) Katz, W.J., Public Works 89:12:114 (1958).
(44) Hurwitz, E. and Katz, W.J., Wastes Engineering 30:12:730 (1959).
(45) Ettelt, G.A. and Kennedy, T.J., WPCF Journal 38:2:248 (1966).
(46) Katz, W.J., and Geinopolos, A., WPCF Journal 39:6:946 (1967).
(47) Eckenfelder, W.W., Jr., Industrial Water Pollution Control,
McGraw-Hill, New York, p 275, 1966.
(48) Howe, R.H.L., "Mathematical Interpretation of Flotation for Solid-
Liquid Separation", Biological Treatment of Sewerage and Industrial
Wastes, Reinhold Publishing Company, 2nd Edition, 1958.
(49) Eckenfelder, W.W., Jr., et al., "Dissolved-Air Flotation of
Biological Sludges", Biological Treatment of Sewage and Industrial
Wastes. Reinhold, New York, p 251-248, 1958.
(50) Garwood, J., Effluent and Water Treatment Journal 7:380 (1967)
(51) Mogelnicki, S.J., "Experiences in Polymer Application to Several
Solids-Liquid Separation Processes", Proceedings 10th Sanitary
Engineering Conference, University of Illinois, Bulletin 65,
115:47 (1968).
(52) "The Use of Organic Polyelectrolyte for Operational Improvement
of Waste Treatment Processes", Report prepared by City of
Cleveland, Ohio, for FWPCA, Grant No. WPRD 102-01-68, May, 1969.
(53) Braithwaite, R.L., Water and Sewage Works 111:12:547 (1964).
(54) "Chlorination of Sewage and Industrial Wastes", Manual of Practice
No. 4, Federation of Sewage and Industrial Wastes Association,
1951.
91
-------
(55) Koller, L.R., Ultraviolet Radiation, John Wiley & Sons, Inc.,
New York, 1952.
(56) Kelly, C.B., American Journal of Public Health 51:11:1670 (1961).
(57) Syraons, R.S., Sewage Works Journal 13:2:249 (1941).
(58) McKee, J.E., et. al^, WPCF Journal 32:8:795 (1960).
(59) Cleasby, J.L. , e± al._, Journal AWWA 56:4:466 (1964).
(60) Cherry, A.K., Journal AWWA 54:5:417 (1962).
(61) Hann, V.A., Journal AWWA 48:10:1316 (1956).
(62) Hann, V.A., Journal AWWA 35:5:585 (1943).
(63) Powell, M.P., et al^, Journal AWWA 44:12:1144 (1952).
(64) Manual on Disposal of Refinery Wastes, Chapter 5, Copyright
American Petroleum Institute, 1969.
(65) Christensen, Ralph, Private communication, EPA Project No. 11030
FOB, Detroit, Michigan, FWQA Chicago, Illinois, 1969.
(66) Gannon J. and Stueck, L., "Current Developments in Separate vs.
Combined Storm and Sanitary Sewage Collection and Treatment",
presented 42nd Michigan WPCF Conference, June, 1967.
(67) Eckenfelder, W.W. Jr., Principles of Biological Oxidation,
Pergamon Press, New York, 1966.
(68) Bennett, C.A. and Franklin, N.L., Statistical Analysis in
Chemistry and the Chemical Industry , John Wiley and Sons,
London, 1954.
(69) Bulletin 720B "Variable Capacity Pumping Systems", Aurora Pump
Company, Aurora, Illinois.
(70) Albrecht, A. E.. Turbine Flocculators. Rex Chainbelt Inc. Research
Report 3W-40 //I, December, 1969.
92
-------
SECTION X
PUBLICATIONS
The data reported in this report have been utilized as a basis for three
technical papers:
1. Mason, D.G., "The Use of Screening/Dissolved-Air Flotation for
Treating Combined Sewer Overflows," Presented at the Seminar on
Storm and Combined Sewer Pollution Problems, Edison, New Jersey,
November 4-5, 1969.
2. Mason, D.G., "Screening/Dissolved-Air Flotation for Treating
Combined Overflows," Presented at the Seminar on Combined Sewer
Overflow Abatement Technology, Chicago, Illinois, June, 1970.
3. Mason, D.G., "Treatment of Combined Sewer Overflows Utilizing
Screening/Dissolved Air Flotation", Presented at the WPCF
Conference October 3-8, 1971, San Francisco, California
93
-------
APPENDIX I
CHEMICAL OXIDATION STUDY
-------
LITERATURE SEARCH
Removal of organic contaminants from combined sewer overflows cannot be .
effectively and economically accomplished by conventional biological
oxidation methods (1)(2) for the following reasons:
A. Large and highly variable flows
B. Intermittent operation would be required because of the nature
of the flows, and this has a very detrimental effect on
biological oxidation systems (3)(4)(5).
C. Removal of the solids produced would require a large number of
sedimentation tanks (because of high flows) which would be used
only periodically.
Use of chemical oxidation on combined overflows has the potential
advantage of effective destruction of biological organisms and the
removal of organic contaminants without the production of residual waste
concentrates. Some of the technical difficulties connected with the
successful treatment of wastes by chemical oxidation include the rela-
tively dilute concentrations of organic materials present, the unknown
composition of a wide range of possible organic compounds present, and
the continually changing concentrations and compositions present in a
combined wastewater flow such as the one with this project is concerned.
These difficulties necessitate the use of gross parameters in evaluating
the effectiveness of a given chemical oxidant. Such parameters would
include knowing the chemical oxygen demand of the waste (COD) and the
concentration of the oxidizing properties in the chemical being utilized.
Having this information, the oxidation efficiency (6) can be calculated
as shown below:
Oxidation Efficiency = ACOD
Available Oxidation Equivalents x 100 (6)
Where: ACOD = change in COD brought about by the oxidation process
(in milligrams of Q£ per liter)
And: Available Oxidation Equivalents = amount of chemical oxidation
equivalents available from the oxidant used (in milligrams of
02 per liter)
Taking into account the variation in organic concentration and volume
of the waste flow to be treated under this contract, the "ideal" oxidant
should have the following characteristics:
A. It must be nonspecific in its attack on organic materials.
B. Its required contact and retention time should not be excessive
(30 minutes or less).
C. It should be easy to handle and dispense.
D. It should effectively oxidize the organic materials at the
prevailing pH of the waste
E. It should not produce any secondary pollutants which are
difficult to remove from the waste flow.
96
-------
From the literature (6) it was found that six types of oxidation systems
offer various degrees of potential for the treatment of organic impuri-
ties in wastewater. These systems are:
A. Oxidation by oxidants containing active oxygen
B. Accelerated molecular oxygen oxidation
C. Catalytic oxidation of adsorbed organics
D. Oxidation by chlorine and its derivatives
E. Oxidation by oxy-acids and their salts
F. Electrochemical oxidation
Of the six oxidation systems mentioned, only systems A, D, E, and F appear
to have potential for treating combined sewer overflows. These systems
will be discussed in detail below.
Hydrogen Peroxide
Considerable work has been done regarding the application of hydrogen
peroxide (H202) as an oxidizing agent in the treatment of wastewater
containing organic impurities (6)(7)(8)(9)(10)(11)(12).
In aqueous media, hydrogen peroxide decomposes to form the 'OH radical.
This radical is one of the strongest oxidants known in aqueous systems.
However, hydrogen peroxide does not, when used alone, oxidize organic
materials within a practical reaction time. The system requires a multi
valent iron salt as a catalyst.
The principle reactions involved in a hydrogen peroxide iron salt
system are as follows (11)(12):
1. Fe + H202 ->Fe + H02 + H
2. Fe"1"* + H202 -»-Fe"HH" + OH~ + "OH
Considering reactions 1 and 2, above, the iron salt acts truly as a
catalyst i.e., is not utilized during the reactions. Reaction 1 is the
rate limiting reaction. Once the ferrous ion is formed, reaction 2 is
extremely rapid (11)(12). The production of *OH radicals via reactions
1 and 2 using ferric salts requires two to three hours at 65°C (6). By
using ferrous salts, the production of *OH radicals is essentially
immediate, but the iron salt no longer performs as a true catalyst, since
it is not reproduced at the same rate at which it is being utilized.
There are other reactions associated with the H202iron salt system
which compete with the organic material being oxidized for the 'OH
radical. The following reactions illustrate the competition for *OH
radicals (11)(12).
3. Fe + 'OH -»-Fe + OH
97
-------
4. H02 + + 0 ~ +
5. Fe^ + 02~ -> Fe^ + 02
6. H202 + 'OH -» H02 + H20
Hence, only an unknown portion of the 'OH radicals produced are available
to oxidize the organic substrate.
The efficient production of 'OH radicals requires a specific pH range
of 3 to 5 (6)(7; (10)(11). Above or below this specific pH range,
production of "OH radicals is greatly reduced. During the reactions,
an excess of H+ ions are formed (9) so the pH will normally stabilize
in the optimum range. Eisenhauer (7) found that the initial pH of an
ABS solution could be as high as 11, before it would significantly affect
pH stabilization in the 3 to 4 pH range. The pH of the system can also
be lowered to the specific range by use of acid (11).
The reactions between *OH radicals and specific organic substrates were
summarized by Busch (12) after review of a number of papers on the
subject. Since the waste to be oxidized under this contract contains
many different types of organic substrates in unknown proportions, these
specific equations were not directly applicable and hence were not repro-
duced herein.
Use of the hydrogen peroxide method of oxidizing organic wastes has
been experimentally evaluated. Davidson found that phenols could be
chemically oxidized (90% complete) to carbon dioxide and water in ten
minutes (9). Eisenhauer demonstrated that in dilute aqueous solutions
of phenol, the reaction efficiency was considerably increased in the
presence of air-available oxygen (7).
Eisenhauer also found that the reaction between a hydrogen peroxide-
ferrous salt combination and ABS was rapid and 80 to 90% of the ABS was
destroyed in the first ten minutes. Further ABS removals were obtained
at slower reaction rates, with a 99% ABS removal effected after fifteen
to twenty hours. The optimum reaction pH level was found to be from
3.0 to 3.5. The optimum concentrations of the reactants were six moles
of ferrous salt per mole of ABS, and nine moles of ^C^ per mole of ABS.
Multiple incremental additions of the ferrous salt will decrease the
reaction time required (11).
Applying this information to a treatment of laundry wastes containing
ABS, Eisenhauer obtained ABS reductions in excess of 90%, both with the
raw waste and with effluent from a pretreatment process in which the
detergent builders were precipitated. For a laundry waste containing
50 to 80 parts per million ABS, it was found that chemical costs ranged
from $0.45 to $0.70 per 1,000 gallons for the raw waste treatment and
$0.15 to $0.25 per 1,000 gallons for the effluent pretreated with
ferric sulfate. This includes the cost of ferrous sulfate, ferric
sulfate, sulfuric acid, and hydrogen peroxide (7).
98
-------
Bishop et_
-------
Source: Prolonged Ozone Inhalation 8, Its Effects on
Visual Parameters, J.M. Langewerf, Aerospace
Medicine, 36, June 1963
Permanent
Toxic
Region
Temporary
Toxic
Regidn
Non Symptor
Reg i 01
0.1
10 100 1000
Exposure Time in Minutes
10,000
FIGURE 1-1
HUMAN TOLERANCE FOR OZONE
100
-------
Ozone can be produced in a number of ways. Among these are silent elec-
tric discharge, electrochemically, and chemically (22). Use of silent
electric discharge is the principal upon which most laboratory and
large scale ozonizers depend (13)(22). Production of ozone by this method
requires the use of cold, clean air or oxygen. Basic equipment necessary
include an air fan, equipment for filtering and drying the air and an
ozonizer (14). When applying the ozone to water, some type of diffuser
system and a contact tank will also be needed (14). Ozonizers produce
ozone by passing the dried air between two concentric electrodes separated
by a dielectric. High frequencies (500 cps to 1000 cps) and high voltages
(4000 to 30,000 volts)are utilized. Some oxygen in the air passing through
the ozonizer is broken into charged oxygen atoms, which recombine to form
03. Ozonizers operating on air under pressure produce far greater quan-
tities of ozone in higher concentrations than those operating in air it
atmospheric pressures (13)(17). Commercial ozonators produce ozone
concentrations of 1 to 4 percent by weight in air (17)(23). Power consump-
tion is 9.5 to 11.5 KWH per pound of ozone produced (23).
Ozone is very corrosive and hence the materials used in the piping system
carrying ozonated air must be carefully chosen. Porcelain has been used
with success, as have aluminum, stainless steel, pressboard, oak, vinyl
plastic and glass (14) (24). Polyvinyl chloride (PVC) has been used but
there is a possibility that this material decomposes ozone (14).
The exact oxidation mechanism of ozone on organic materials is not
thoroughly understood. Ozone decomposition may be a source of 'OH radicals
or the 0^ itself may be the actual oxidant (6)(21). Ozone is known, how-
ever, to be an effective oxidant (17)(18)(21). The exact oxidation
products and intermediates resulting from ozone oxidation have not been
defined. Evans and Ryckman (18) when ozonating secondary sewage treat-
ment plant effluents have found that ozone readily oxidized ABS to inter-
mediate compounds which were not detected by an ABS analysis. These
intermediates actually caused an increase in BOD and COD values of the
partially ozonated waste, and in fact destroyed the biological inertness
of the ABS. Increased ozonation of this partially ozonated waste then
caused a decrease in BOD and COD of the waste, apparently due to oxidation
of some of the organic carbon to C02 and t^O.
Andrews (16) was recently granted a patent on a new method of treating
raw sewage. The method involves removal of the solids from raw sewage
by a combination of settling, screening, and centrifuging. The resulting
liquor, esentially free of solids and containing the soluble BOD fraction
of the raw sewage, is ozonated to oxidize the BOD. The effluent from
ozone treatment is filtered before discharging into the receiving stream.
The patent claims this method of treatment produces an effluent which is
clear, colorless, free from bacteria and contains a high quantity of
dissolved oxygen. No reference is made, however, as to dosages of ozone
required and BOD content of the effluent.
Miller (21) reports use of ozone to control sewer odors at sewage lift
stations. He states ozone addition reduced the raw sewage BOD from
101
-------
120 to 32-39 mg/1. Septic conditions in flat sewers disappeared and
the plant influent contained 2 to A mg/1 of dissolved oxygen. No mention
was made of the quantities of ozone added. The ozone was injected
directly into the pump wet wells for conventional lift stations and into
the air compressors supplying air for the pneumatic-ejector type lift
stations.
March and Panula (15) reported on a study where raw sewage was ozonated.
They report that the BOD of the sewage which was ozonated decreased from
400 to 300 mg/1. Application rate was 6 mg/1 of ozone actually absorbed
by the sewage. This indicates that 1 mg of 0-j will oxidize about 16 mg
of BOD which seems quite high. The amount of 03 applied during this
study was greater than the amount actually utilized. Absorption effi-
ciencies averaged 22.6%. Hence, the actual ozone dosage was about 26 mg/1.
March and Panula (15) indicate that the biggest problem in full-scale
ozone treatment is efficient use of the ozone being produced. They
suggest collecting the unabsorbed ozone and reapplying it to the sewage
to obtain better absorption efficiencies. It should be noted that a
simple diffusion system was used during this study with a 3^ foot depth
of liquid over the diffusers. Bubble size was on the order of 0.01 inches
in diameter.
0'Donovan (14) stresses the importance of mixing techniques when intro-
ducing ozone into water. He stresses small bubble diameter and turbulence
to be important in obtaining good 0^ adsorption efficiencies. Two types
of adsorption systems which have been used are illustrated in Figure 1-2
(14) (25). Campbell (25) reports 90% utilization of ozone when using a
partial-ejection system.
The use of ozone to oxidize combined sewer overflow appears to hold
considerable promise. Ozone has many of the characteristics of the
"ideal oxidant" as previously discussed. It is apparently nonspecific
in its oxidation of organic material (21), the necessary contact time
is less than 30 minutes (14) (25), it oxidizes at the prevailing pH of
the waste (25), and is not difficult to dispense(14). Some disadvantages
involved with the use of ozone include precipitation of iron and manganese
when present in concentrations greater than 0.2 mg/1 Fe and 0.1 mg/1 Mn,
and inability to store ozone requires ozonators at each application site.
Recently, however, Matheson Inc. has announced it has a process which
will allow storage of ozone which could eliminate ,this latter disadvantage,
Oxidation by Chlorine
Chlorine or its derivatives are added to most municipal water supplied
in the United States to kill micro-organisms. Green and Stumpf (26)
have shown that chlorine compounds react with certain enzymes which are
essential to the metabolic process of living cells and that death results
from the inactivation of these enzymes. Dosages required for disinfection
are low, 0.2 to 2 mg/1 (1). At these dosages, probably little, if any,
oxidation of organic material is occurring.
Use of chlorine to reduce the BOD of sewage has been practiced. Four
kinds of reactions could be involved in reducing BOD by chlorination
(D(27).
102
-------
Injector
Ozonized All
Pump
Contact Tank
1
I
'
1
1
xv
" O-»- Main F
/
*
+^*" A * Ozonized Water
Partial-Injection
(A)
Mixer
Ozonized Air
Raw Water-
Turbulences
Chamber
Kerag System
(B)
FIGURE 1-2
OZONE ABSORPTION SYSTEMS
Ozonized Water
103
-------
1. Direct oxidation of organic materials.
2. Substitution of chlorine for hydrogen which produces compounds
which have a bactericidal power.
3. Substitution of chlorine for hydrogen which produces compounds
which are no longer biologically degradable.
A. Addition of chlorine to unsaturated compounds forming nonbio-
degradable substances.
Griffin and Chamberlin (28) have shown that chlorination of sewage does
result in a reduction of BOD. After an 18 hour contact period, about
35% of the BOD had been removed from screened raw sewage at a 200 mg/1
chlorine dosage. Increasing the dosage had little effect on further
removals.
Although chlorine and its derivatives do oxidize many of the organic
chemicals contained in municipal wastewaters, they do not usually convert
them to the readily acceptable forms of carbon dioxide and water (6).
Instead, the reaction of chlorine with organics may produce molecules
having considerable taste and odor, and complex compounds which serve
as secondary pollutants may also be formed. The addition of stoichio-
raetric amounts of chlorine will not effect oxidation of the organics,
and a large chlorine residual may remain in the treated waste (6)(28).
Since oxidation by chlorine is far less effective than by active-oxygen
elements and since the unknown by-products of chlorine oxidation may be
toxic, chlorine and its derivatives are not considered to be good
prospects for removal of large amounts of organic pollutants from waste-
waters by oxidation (6).
A method of making chlorine more reactive could, however, make the use
of chlorine more attractive. Use of ultraviolet radiation to catalyze
the oxidation of organic material by chlorine has been suggested (29).
No specific data on such a process was found in the literature, but
the process will be investigated during the laboratory phase of this study,
Oxidation by Oxy-Acids and Their Salts
The oxy-acid oxidants are usually added to aqueous systems as the salt
of the oxy-acid. Important oxy-acid salts with high oxidation potentials
include potassium permanganate (KMNO^), sodium ferrate (NaFeO^) and
potassium ferrate (K-FeO,). The oxidation potentials of the oxy-acids
are strongly dependent upon the acidity of the aqueous system, with the
potential increasing as the acidity increases (6).
Potassium permanganate (KMNO,) is used as an oxidant to disinfect water
supplies (1)(30) and as an oxidizing agent in the permangate oxygen
demand test (31). However, references in the literature report only
limited oxidation of organics with KMNO^, and a residual permanganate
persists in the water for more than twenty-four hours (30)(31). Vosloo
104
-------
(32) reports that the permanganate concentration necessary for maximum
oxidation should be about twice the concentration of permanganate which
will be utilized. The use of potassium permanganate would therefore
require removal of the excess MNO^-ion as a part of the treatment process,
and this would greatly increase the salt load in the aqueous system (6).
The potential for using KMNO^ as an oxidant in this project is not at
all promising, and hence, KMNO^ oxidation will not be considered.
Although ferrate salts such as potassium ferrate (I^FeO/) are not
commercially produced, their high oxidation potentials (33) and the
coagulating properties of their oxidation products, a highly insoluble
ferric hydroxide, make the ferrates a potentially attractive group of
oxidants for treatment of wastewaters (6).
Exploratory laboratory tests using l^FeO^ to treat filtered municipal
secondary effluents in dosages supplying approximately 100 milligrams
per liter of available oxygen showed the combination of ferrate oxida-
tion and coagulation to be nearly as effective as the previously
discussed H-O- - iron salt system for removing organic materials from
wastewaters (o). The ferrate salts, however, also possess the same
disadvantages as the 1^02 iron salt system, (see previous discussion
on hydrogen peroxide oxidation) and these disadvantages must be consid-
ered when attempting to use the ferrate salts as oxidants in this
project.
Electro-Chemical Oxidation.
Any material that increases the electrical conductivity of an aqueous solu-
tion may enter into & chemical reaction at the surface of electrodes placed
in the Ablution (27)» Interest in electrochemistry as a possible technique
for purifying municipal wastewaters is based on the fact that many organic
chemicals take part in such electrode reactions, often resulting in the
complete degradation of complex organic molecules to carbon dioxide,
water and other oxides. After reviewing a recent study on electrochemi-
cal oxidation (34), it was concluded that electrochemical oxidation did
not hold potential for treating combined sewer overflows for the following
reasons:
1. High operating and maintenance costs.
2. Formation of precipitates which would have to be removed from
the waste flow.
3. Long reaction times are necessary 2 to 5 hours.
Combination of Oxidants
Although no specific data were found in the literature about combining
various oxidants, a combination of ozone and H?0~ or other oxidant
combinations may produce an effective oxidation system. For this reason,
105
-------
various combinations of oxidants will be evaluated during the laboratory
phase of this project.
LABORATORY INVESTIGATIONS
General Test Procedures
The literature search indicated that the oxiuants with the highest poten-
tial success were ozone, hydrogen peroxide, and chlorine. During the
laboratory investigation phase the above mentioned oxidants as well as
various oxidant combinations were evaluated.
Samples of seven separate combined sewer overflows from the demonstration
site at Hawley Road were collected during the Fall of 1967. These samples
were analyzed in the laboratory within twenty-four hours of their collec-
tion for BOD, COD, dissolved COD, total solids, suspended solids, volatile
suspended solids, coliform density and pH. The data from these analyses
has been reported and discussed in the main body of this report. Since
the likelihood of getting sufficient and frequent combined overflows
during the fall and winter seasons in Milwaukee is small, it was decided
to freeze some combined overflow samples and use them for chemical
oxidation evaluation during dry weather. Therefore, about twenty liters
of combined overflow samples from thre-3 separate storms was frozen in
individual containers having volumes of from 100 to 200 milliliters.
The laboratory analysis of the three storm samples frozen for chemical
oxidation studies is shown in Table 1-1. During dry periods, samples
were thawed as needed. Samples were mixed thoroughly in a Waring blinder
after thawing to provide a colloidal system characteristic of the
original unfrozen raw sample. Whenever a sample without the particulate
matter was needed for oxidation of the dissolved organic matter, the
thawed sample was not mixed. Instead, the supernatant was filtered
through a SS-597 filter paper and millipore filter discs to obtain a
dissolved sample. Standard Method (35) procedures were employed for
various analysis in the laboratory and are discussed in Appendix II.
Special Test Procedures with Various Chemical Oxidants
Chlorine and Hydrogen peroxide oxidations were performed with and without
ultra-violet (UV) light. Chlorine was applied in the form of calcium
hypochloride solution (HTH) and a 0.75% stock solution was utilized for
hydrogen peroxide oxidation. Procedures for determination of oxidant
dosages and residuals are discussed in Appendix II. Cobalt in the form
of CoClj'S H-0 was utilized in a number of tests to determine if it
helped to catalyze oxidation reactions of 1^02 and chlorine. The
apparatus consisted of an eight inch diameter flat bottomed dish on a
magnetic stirrer under a protective hood cover. A 300 ml aliquot of raw
waste was used. For tests performed with ultra-violet light, the UV
lamp was suspended above the water surface with ring stands at three and
six inch heights. The two UV lamps utilized were a single bulb Sperti
sunlamp and a double bulb G.E. germicidal lamp. Figure 1-3 shows a
photograph of the apparatus utilized for these tests. Samples were
taken at various intervals during the oxidation period for COD analysis.
106
-------
TABLE 1-1
LABORATORY ANALYSIS OF COMBINED SEWER SAMPLES UTILIZED FOR
CHEMICAL OXIDATION STUDY
Total
Solids
Date pH mg/1
11/25/67 7.4 649
12/07/67 6.6
01/29/68 7.1
Suspended
Solids
mg/1
138
232
2158
Volatile
Suspended
Solids
mg/1
>
113
COD
' mg/1
159
298
1410
Dissolved
COD BOD
mg/1 mg/1
59
50
60
Coliforms
per ml
4260
2151
5730
-------
FIGURE 1-3
APPARATUS FOR ULTRA VIOLET LIGHT OXIDATIONS
FIGURE 1-4
APPARATUS FOR C^.ONE OXIDATIONS
108
-------
Some difficulty was encountered in measuring the COD after oxidation
with various oxidants. Contrary to expectations, some samples
exhibited an increase in COD after oxidation. This observation could
have been caused by the break down of organic material which prior
to oxidation was not measured by the COD tests. The increase may also
have been caused by the excess oxidant remaining (the reaction kinetics
were not known and exact amounts of oxidants could not be utilized).
Since it was desirable to know if the COD was increasing due to
partial oxidation, the excess oxidants were removed prior to analyzing
for COD. Several trial and error procedures were investigated for the
removal of these excess oxidants. It was planned that excess oxidant
would be removed after oxidation tests by adding a small excess of
sodium sulfite. After the addition of sodium sulfite, the sample was
aerated until a galvanic cell oxygen analyzer indicated the presence
of dissolved oxygen in the sample. Also investigations were undertaken
to study the effect of excess oxidants and chemicals on the COD test
results. Theoretically, the reaction between hydrogen peroxide and
hexavalent chromium in a COD test may proceed as follows:
3H202 + K2Cr2°7 + 81l+ = 302 + 2Cr+3 + 7H202 + 2K+ [ORP = +0.65] ---- EQ
Then 1 mg H202 = 0.47 mg COD
Similarly, the reaction between sodium sulfite and dichromate may
proceed as follows:
3Na2S03 + K2Cr207 + 8H+ = 3Na2S04 + 2Cr+3 -I- AH20 + 2K+ ..... EQ
Then 1 mg Na2S03 = 0.127 mg/COD
109
-------
Experimentally it was found that the COD exerted by hydrogen peroxide
solution decreased with increasing strength of H202 solution as follows:
Concentration of
H202 Solution Ratio of mg COD/rag K2°2
50 0.42
100 0.41
150 0.39
200 0.38
250 0.37
The reason for such observation can be attributed to the redox relation-
ships of hydrogen peroxide, since H202 can also react with the trivalent
chromium produced as follows:
3H202 + 2Cr"1"3 + 6H+ + 7H20 = 6H20 + Cr207= + 14H+ [ORP = +.44] EQ III
Thus hydrogen peroxide solution was found to exert a significant COD
value and this was a factor in the increased COD results of hydrogen
peroxide treated samples.
Also, sodium sulfite was experimentally found to have a chemical oxygen
demand of 0.09 mg per mg of Na2S03. This is approximately 70% of the
theoretically calculated value of 0.127 and is not significant in light
of the small amounts of sodium sulfite utilized in these studies.
When calcium hypochlorite was used as an oxidant, the sulfite was added
in increments until a spot test with ortho-tolidine indicated that
there was no more free chlorine. The sample was then aerated until a
galvanic cell oxygen analyzer indicated the presence of dissolved
oxygen in the sample. Experimentally it had been found that a 715 mg/1
calcium hypochlorite solution exerted a nominal COD of 9 mg/1 which
could possibly be the result of experimental errors. This experimental
observation is also supported by possible theoretical reaction between
chlorine and hexavalent chromium, as:
3C12 + 5Cr20?= + 34H+ = 6C10-J- + 10Cr+3 + 17H20 [ORP = -.14] Eq IV
The negative oxidation reduction potential indicates that the above
reaction should proceed to the left and therefore, theoretically the
reaction should not take place. This leads to the conclusion that no
COD should be exerted by the chlorine solution.
Somewhat contradictory results were obtained when COD tests were
performed on samples from stormwater oxidation tests, whereas the COD
of calcium hypochlorite solution was negligible and the COD of oxidized
sodium sulfite was very small, in most cases the COD values of
110
-------
chlorinated waste samples which were dechlorinated with sodium sulfite
and aerated were 5 to 35% higher than the samples which were not dechlor-
inated. Because of this difference, an alternate method of dechlorination
was also used: the sample was acidified to a pH of one and stripped
of chlorine by aeration. Unfortunately, the stripping process was
extremely slow, often requiring several hours. In some cases, not all
the chlorine was removed by the time the COD analyses were performed.
In general, the COD values obtained by acidification and stripping
were lower than the values obtained by sulfite treatment. The exact
cause of this difference in values is not known.
Ozone Oxidation System
Ozone was purchased by the bottle and used in a pressure tank system.
A photograph of the apparatus utilized is shown in Figure 1-4 and a
schematic is shown in Figure 1-5. Ozone was supplied dissolved in
"Freon 13' under high pressure in stainless steel cylinders. Since
ozone has a half life of approximately three days at room temperature
it was necessary to keep the ozone cylinder packed in dry ice. A
three mole-percent vapor phase concentration was utilized for oxidation
studies. To maintain a constant ozone concentration supply, ozone was
utilized in the vapor phase. This was achieved by inverting the ozone
bottle and converting the liquid ozone to vapor phase by passing it
through a vaporizing coil.
Ozone was mixed with the waste at elevated pressures in the range of
40 to 80 psig. This pressure charged stream was then placed in a
graduated cylinder for the required reaction time. Samples were air-
stripped to remove excess ozone remaining after the reaction time.
Pressurization of waste could be total or partial. When the entire
waste was pressurized, it was called total pressurization. If only a
portion of the waste was pressurized, the amount was calculated as a
percentage of the nonpressurized portion, and was called a recycle
system. Use of a pressure tank system enabled large amounts of ozone
to be dissolved and when the pressure was released the extremely small
air-ozone bubbles stayed in contact with the liquid much longer than
the coarse bubbles produced from diffuser type systems. Such a system
was utilized because the literature search indicated that the ozone
contact time is critical and greatly affects efficient use of ozone.
Titr./,:.ions for ozone determination were performed in accordance with
Standard Methods(35).
RESULTS AND DISCUSSION
Oxidation with Hydrogen Peroxide
Results of chemical oxidation with hydrogen peroxide are presented in
Tables 1-2 and 1-3. Hydrogen peroxide requires a catalyst to oxidize
effectively within reaction times of one hour or less. Table 1-2
shows the results of the tests performed using iron as a catalyst.
Oxidation tests were run on the dissolved fraction of the overflov
111
-------
Gas Collection
For Analysis
Sampling
Port
Pressure Vessel
for Dissolving
03 in Waste
Sample
Ozone Supply
(Cylinder)
Gas Sampling
Access Ports
Vaporizing Coil
Pressure Regulator
Reaction Vessel for
Non-pressurized Oxidation
FIGURE 1-5
SCHEMATIC OF OZONE TEST APPARATUS
-------
TABLE 1-2
RESULTS OF CHEMICAL OXIDATION OF COMBINED
SEWER OVERFLOW WITH HYDROGEN PEROXIDE
Test
No.
1
2
3
4
5
6(3)
7(4)
8
9
Available
02
50
100
100
100
100
100
100
150
300
COD
Initial
54
67
54
54
65
65
65
216
216
Final
50
30
41
33
37
42
31
186
185
Percent
Reduction
7
55
24
39
43
35
52
14
14
Oxidation
Efficiency
Percent
8
37
13
21
28
23
34
20
10
Filtered
Prior to
Analysis
No
Yes
No
Yes
Yes
Yes
Yes
No
No
NOTES: 1. In Tests 1 through 7 only the dissolved COD was oxidized, i.e,
all solids removed prior to oxidation. Tests 8 and 9 were
run on combined overflow after screening through a 50 mesh
screen.
2. All oxidations were for 15 minutes at room temperature.
3. Aerated during the oxidation.
4. Cobalt added during the oxidation.
113
-------
TABLE 1-3
EFFECT OF UV LIGHT AND CObALT ON HYDROGEN PEROXIDE OXIDATION
Test
No.
Type of
Sample
H202
Dosage
mg/1 as
02
H202
Remaining
after Oxidation
rag/1 as 02
Catalyst Used
COD, mg/1
Reaction After % COD
Time Influent Oxidation Reduction
5
6
7
8
9
10
11
12
13
14
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Distilled
Water
Distilled
Water
Filtered
Filtered
Filtered
Filtered
110
110
110
110
110
110
110
110
110
110
110
110
110
110
None
29
15
66
54
60
62
93
106
78
66
79
68
1
UV Light
UV Light &
10 rag/1 Cobalt
UV Light 6.
100 rag/1 Cobalt
10 mg/1 Cobalt
100 mg/1 Cobalt
UV Light
UV Light &
UV Light
UV Light &
5 mg/1 Cobalt
UV Light
UV Light
UV Light
UV Light
15
30
90
15
30
90
30
30
30
92
92
106
106
62
75
175
149
161
138
100
98
33.0
19.0
30
30
30
30
30
106
106
58
58
119
114
34
39
80
81
5.6
7.5
41.0
33.0
15
30
15
30
114
114
112
112
76
107
59
80
34.0
6.0
47.0
29.0
NOTES:
1. Ultra-violet light wave length 2800 to 3200 A.
-------
(Tests 1 through 7) as well as the overflow after screening through a
50 mesh screen (Tests 8 and 9).
Test 1 through 7 (Table 1-2) indicated that a substantial portion of the
dissolved COD can be oxidized in 15 minutes and the oxidation
efficiencies were in the range of those reported in the literature.
When the effluent from the 50 mesh screen (Tests 8 and 9) was oxidized
with the hydrogen peroxide - iron salt system, the oxidation
efficiencies were lower than those obtained when oxidizing only dissolved
COD. This indicated the importance of removing a major portion of the
solids prior to oxidation, since particulate matter is more difficult
to oxidize than dissolved organic matter.
Use of hydrogen peroxide-iron salt system does not appear feasible.
Ferrous iron seems to work well, but relatively large concentrations
are required (on the order of 60 to 100 mg/1 as Fe ). This system
requires strict pH control (3 to 4) and a neutralization and settling
period to remove the iron precipitate which is formed. Organic
removal efficiencies are in the 30 to 50% range for dissolved organics
and were reduced to about 14% when particulate organics were present.
In an attempt to more efficiently utilize the hydrogen peroxide,
other catalysts were evaluated which would not be as restrictive as
ferrous salts.
The literature indicated that hydrogen peroxide is decomposed by
ultra-violet light. Also Test 7 (Table 1-2) indicated that cobalt
may help catalyze a hydrogen peroxide system. Therefore, tests were
run with both these catalysts to study any improvements in the
efficiency of hydrogen peroxide oxidation system. The results of
these tests are shown in Table 1-3. It is seen that the use of either
or both cobalt and ultra-violet light along with hydrogen peroxide
did not produce consistent results. From Tests 1 and 2 (Table 1-3)
it is clear that reaction time has an important bearing on COD reduc-
tions in the treatment of storm waters with hydrogen peroxide. In
Test 1, when a hydrogen peroxide dosage of 110 mg/1 as 02 was utilized,
a 33% reduction in COD was obtained in 15 minutes; the COD reduction
reduced to 19% in 30 minutes and even became negative at the end of
the 90 minute reaction time. A possible explanation for the increase
in COD in a certain initial reaction time may be the partial oxidation
of large molecular weight organic compounds in the raw storm samples
to smaller intermediate compounds which exhibited an increase in COD.
Increased oxidation at a higher reaction time may then cause a decrease
in COD of these samples, apparently, due to the oxidation of some of
the organic carbon to C02 and I^O. Similar explanations have been
mentioned in literature for ozone oxidation(18). When UV light was used
as a catalyst, results were contradictory as shown in Tests 2, 7, 11, 12,
13 and 14 (Table 1-3). Test 2 showed negative COD reductions for all
reaction times between 15 to 90 minutes while Tests 7 and 11 to 14 showed
COD reduction between 6 to 47% for reaction times of 15 and 30 minutes.
Similar contradictions in results were shown when cobalt was used as a
catalyst either independently or in combination with UV light (Tests 3
to 6, Table 1-3).
115
-------
The results presented in Tables 1-2 and 1-3 clearly indicated that a
Hydrogen Peroxide Reaction System was not technically feasible for treat-
ing combined sewer overflows. Thus work on such a system was abandoned.
Oxidation with Chlorine
Results of chlorine oxidation with and without ultra-violet light are
presented in Tables 1-4 and 1-5. Oxidation with chlorine and no ultra-
violet light resulted in a reduction of COD of about 20 to 25% at
available chlorine dosages of 280 to 560 mg/1 (Table 1-4). The amount
of chlorine destroyed was about 25% and hence a large chlorine residual
remained after the oxidation. Reducing the chlorine dosage of 56 mg/1,
reduced the percent COD oxidized to only 8% (Test 3, Table 1-4). These
results indicate that chlorine is not an effective oxidant for combined
sewer overflows. This conclusion is supported by the literature search.
Oxidation with chlorine and ultra-violet light in the 2800 to 3200 A"
range was not significantly different from those oxidations where ultra-
violet light was not used (Table 1-4). The reason for these low oxida-
tion efficiencies may have been due to the long wave length light which
was used i.e. 2800 to 3200°A or the low ultra-violet output of the lamp
which was used for the reactions. A review of the research conducted by
Midwest Research Institute (36) indicated that lower wave lengths were
more effective. Therefore, a lower UV light wave length of 2587 A was
used for further experiments. This wave length would also provide good
disinfection. The results of these experiments are shown in Table 1-5.
It should be noted that for light catalyzed oxidations to be effective,
relatively clear solutions were required. Therefore, the results of
the screening step could have a pronounced influence on the efficiency
of light catalyzed oxidations.
From the results shown in Table 1-5, it is seen that COD reductions
of 10% to 50% can be expected depending upon light intensity, chlorine
dosage and reaction time. The amount of chlorine required is extremely
large. Approximately 7 to 10 mg of chlorine is required per mg of COD
oxidized. Hence, for a 200 mg/1 COD waste, 1400 to 2000 mg/1 of chlorine
would be consumed. This also results in high chlorine residuals in the
effluent. With the use of an ultra-violet light catalyst, organic
removal does increase to the 25 to 50% level compared to 10-25% level
without the catalyst, but large chlorine residuals are still present in
the reactor effluent. This fact, coupled with relatively long reaction
times (30 to 60 minutes) and poor light penetration when the waste is
turbid appear to eliminate the possibility of using light catalyzed
chlorine to oxidize combined sewer overflows.
Oxidation with Combined System of Hydrogen Peroxide and Chlorine
Results of oxidation with the combined system of hydrogen peroxide and
chlorine with and without ultra-violet light catalyst are shown in
Table 1-6. The results of these tests were extremely discouraging as
-------
TABLE 1-4
Test
No.
I1
2
3
4
5
Chlorine
Concentration
mg/1
280
560
56
276
284
Reaction
Time
Min.
15
15
15
15
15
CHLORINE OXIDATION TESTS
COD mg/1 %
Raw
65
259
177
198
205
After
W/UV
211
167
169
151
Oxidation
No Light
51
194
163
164
161
COD
Reduction
W/UV2
19
6
15
26
No UV
21
25
8
17
21
Chlorine
Remaining
W/UV No UV
160 202
145 216
NOTES: 1. Sample filtered prior to oxidation
2. Ultra-violet wave length 2800 to 3200 X.
3. Ca (OC1)2 used as source of chlorine.
-------
TABLE 1-5
LIGHT CATALYZED CHLORINE OXIDATION
oo
Reaction
Date Type of Time
1969 Sample Min.
2/2 Filtered 15
2/2 Filtered 30
2/2 Filtered 60
2/2 Filtered 15
2/2 Filtered 30
2/2 Filtered 60
2/8 Filtered 15
2/8 Filtered 30
2/8 Filtered 60
2/9 Raw 15
2/9 Raw 30
2/9 Raw 60
2/13 Filtered 15
2/13 Filtered 30
2/13 Filtered 60
2/16 Filtered 30
2/16 Filtered 30
2/16 Raw 30
2/16 Raw 30
2/26 Filtered 30
COD mg/1
Influent
105
105
105
105
105
105
71
71
71
147
147
147
97
97
97
108
108
167
167
After
Oxidation
94
86
77
91
89
96
43
44
37
121
111
102
81
77
77
76
60
118
68
COD
Reduction
11
18
27
13
15
9
39
38
48
18
25
31
17
21
21
30
44
29
59
Dosage
mg/1
280
280
280
280
280
280
600
600
600
600
600
600
600
600
600
900
900
900
900
Percent Chlorine Chlorine U.V. Light Chloride Cl/COD
Remaining Intensity Cone. Ratio
93
72
23
300
mg/1
115
70
42
185
182
182
401
309
204
384
283
286
497
497
497
576
541
576
497
84
Watts
2.9
2.9
2.9
0
0
0
2.9
2.9
2.9
2.9
2.9
2.9
0
0
0
2.9
5.8
2.9
5.8
5.8
r
mg/1
45
82
253
37
382
502
397
487
352
15
11
8.5
6.
6.1
10.9
7.1
10.8
11.6
8.3
8.8
7.0
6.6
5.7
5.7
10.
7.5
6.6
4.1
10.3
NOTES: 1) U.V. Light Wave Length 2587 A - 2.9 watts intensity (bulb output).
2) Ratio is mg Cl utilized per mg COD oxidized.
-------
TABLE 1-6
OXIDATION WITH COMBINED SYSTEM OF HYDROGEN PEROXIDE
Chlorine H202
Test Type Dosage Dosage Reaction
No. of Sample mg/1 mg/1 as 0? Catalyst Used Time
, 15
1 Filtered-' no 100 None 30
60
2 15
2 Filtered 110 100 UV Light 30
60
Instant
15
3 Filtered no 100 UV Light 30
45
60
Instant
15
4 Filtered 110 100 UV Light 30
45
60
AND CHLORINE
COD, mg/1
After1
Influent Oxidation
81
93 95
122
115
93 117
127
SO
104
86 108
104
85
82
99
86 106
100
113
% COD
Reduction
13.0
7.0
4.8
:::
NOTES:
1. After stripping of the excess oxidant
2. Ultra-violet light wave length 2587 A.
-------
inconsistent and negative reductions in COD were observed. The importance
of reaction time was again demonstrated (Tests 1 to 4, Table 1-6) in the-
use of the combination of these oxidants for the treatment of storm wast.es.
These results were similar to the ones discussed earlier for oxidation by
hydrogen peroxide alone. A small positive COD reduction was shown when
reaction time was very small (instantaneous sample, Tests 3 and 4,
Table 1-6). For higher reaction times up to 60 minutes negative COD
reductions were obtained. Also the use of U.V. light as a catalyst did
not show any improvement in results (Tests 2, 3 and 4 compared to 1,
Table 1-6). Hence, investigations in this area were terminated.
Oxidation with Ozone
The result of the ozone oxidations are shown in Table 1-7. Using total
pressurization at 40 psig Tests 1 and 2 show the difference in COD
reductions for raw and filtered storm water samples. The higher COD
reduction of 46% exhibited for raw waste sample (Test 2, Table 1-7) was
probably due to the flotation of some of the suspended matter as well
as the higher solids content of the raw waste sample. When a 20% recycle
system was used with the 10 mg/1 dosage of polymer C-31 on raw sample
(Test 4) the COD reduction was extremely high at 96%. The reason for
such a high reduction was probably the flocculation of the suspended
matter with the C-31 flocculant and the flotation of floe on the surface.
Tests 5 and 6 (Table 1-7) show the effect of reaction time of ozone at
two different operating pressures of 40 and 70 psig. It was shown that
COD reductions increased from 27 to 52% when the reaction time was
increased from 1 to 20 minutes at 40 psig. There was no significant
improvement in COD reductions when the operating pressure was increased
to 70 psig. Therefore, a pressure of 40 psig was utilized for all
future tests. The ozone dosage utilized for all the tests discussed
above was 40 mg/1. When a 30 mg/1 ozone dosage was utilized the COD
reduction was 32% (Test 1-7) for a reaction time of five minutes as
compared to 43% COD reduction shown for 40 mg/1 for corresponding
conditions (Test 1-5). This means that higher ozone dosages result in
improved COD reductions. When ozone was applied in combination with
other oxidants (such as Cl£ and H^O^) or catalysts (such as Co and Ni)
(Tests 8 through 18), no significant improvement in COD reductions
could be obtained in most cases either with dissolved or raw storm water
samples. However, in one case where ozone was used with both chlorine
and hydrogen peroxide (Test 10), the COD reductions were observed to
be significantly improved at 70%. However, this efficiency was only for
the dissolved organic fraction and would be reduced significantly if
particulate organic materials are present. The use of all three oxidants
at their respective dosages (Test 10, Table 1-7) would result in extremely
high(>$l.00/1000 gal) operating costs.
Two benefits which are possible when using a pressure tank system are
the possibility of floating some of the lighter organic solids, and the
probability that the heavier grit particles will settle. These benefits
can greatly increase the removal efficiencies of the treatment system.
From the results shown in Table 1-7, ozone appears to be the best oxidant
among the ones evaluated in this study because it:
120
-------
TABLE 1-7
Reaction
Test Type Time
No. Sample Min.
1
2
3
4
5
5
5
5
6
6
6
6
7
8
9
10
11
12
13
14
15
16
17
18
Filtered
Raw
Filtered
Raw
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Filtered
Raw
Raw
Raw
Raw
Raw
Raw
3
3
3
3
1
5
10
20
1
5
10
20
5
5
5
5
8
8
8
8
8
8
8
8
SUMMARY 03 OXIDATION TESTS
°3
Dissolving
Raw
74
1680
73
1680
69
69
69
69
69
69
69
69
69
69
69
69
84
84
189
189
189
189
189
189
After
Oxidation
55
959
49
61
50
43
40
33
51
42
36
31
46
46
45
21
61
62
92
154
114
120
1414
116
% COD
Removed
26
43
33
96
27
38
42
52
26
39
48
55
32
33
35
70
27
26
51
19
40
37
25
39
Type
Recycle
Total
Total
Total
20%
Total
Total
Total
Total
Total
Total
Total
Total
Total
Total
Total
Total
Total
Total
Total
20%
Total
Total
20%
Total
Pressure
psig
40
40
40
40
40
40
40
40
70
70
70
70
40
40
40
40
i
40
40
40
40
40
40
40
Approx.
Oo Dosage
mg/1
40
40
40
40
40
40
30
30
30
30
30
30
10
Remarks
Excess 03
No Excess 03
Excess 03
10 mg/1 C-31
Excess 63
Excess 03
Excess 0-j
Excess 03
Excess 03
Excess 03
Excess 03
Excess 03
Excess 03
50 mg/1 H202
50 rag/1 Cl
50 mg/1 Cl and
5 mg/1 Co added
H202
25 mg/1 H202 and Cl
36
8
67
Actual measurement
3 mg/1 Ni added
actua
NOTE: Samples collected fall of 1967 and frozen for subsequent use.
-------
1. Provides the best oxidation
2. Requires reasonable reaction time
3. Can be used in a pressure system allowing effective introduction
of ozone
It should be noted that all the tests shown in Table 1-7 were obtained
with overflow samples which had earlier been collected and frozen.
Therefore, it was not possible to run disinfection tests because of the
adverse effect that freezing has on the coliform organisms. Hence,
further experiments were planned to study disinfection efficiency and
BOD and COD removals using ozone treatment and fresh overflows.
The results of these experiments are shown in Tables 1-8 and 1-9. A
recycle system was utilized and a value of 10% recycle was used to keep
the ozone dosage in 10 to 40 mg/1 range on all overflows except the
April 3 overflow (Tables 1-8 and 1-9). Since some flotation and/or
sedimentation occurs, air flotation tests were run as a control to
determine how much removal was due to oxidation and how much to
sedimentation and/or flotation.
Comparing the suspended solids removals (Table 1-8) indicates little
difference between ozone treatment and air flotation. This points out
that little if any particulate matter is oxidized. These results are
consistent with the oxidation results obtained with the frozen overflows.
The BOD removal was generally better with the ozone treatment compared
to air flotation (Table 1-8). The COD removals, however, followed the
opposite trend. Overall removal rates with screening and ozone treatment
or flotation are in the range of 50 to 80% for the f^rst flushes ahd
45-65% for the normal overflows providing means are available to remove
settled and floated solids.
The amount of disinfection obtained with ozone and chlorine is shown in
Table 1-9. At relatively high ozone dosages good disinfection is obtained.
When the dosage is dropped below 10 mg/1 disinfection becomes very poor.
Chlorine, however, will give good disinfection at a 10 mg/1 dosage
over a wide range of coliform concentrations. These results indicate
that ozone may not be satisfactory for disinfection because of its high
costs and nonattainment of residual ozone for a sufficient period.
Oxidation With Gamma Radiation
Use of gamma radiation for oxidizing organics was indicated in the
literature (38) (39).. Tests were run on gamma irradiation of combined
overflow using a cobalt source. The results of preliminary tests on
these were discouraging and investigations into this area were terminated,
122
-------
TABLE 1-8
SUMMARY 03 OXIDATION AND AIR FLOTATION
Date
4/3
4/3
4/17
4/20
4/23
4/28
4/28
Type
FF
EO
EO
EO
EO
FF
EO
SS
After
Screening
707
183
207
67
122
331
177
SS
After
Oxidation
279
114
106
48
86
102
109
SS
After
Air Flo-
tation
73
55
65
65
61
BOD
After
Screening
148
33
33
13
21
92
58
BOD
After
Oxidation
85
12
17
10
20
36
27
TESTS
BOD
After
Air Flo-
tation
21
10
12
42
27
COD
After
Screening
559"
121
134
51
99
331
199
COD
After
Oxidation
79
36
83
159
143
COD
Air Flo-
tation
69
44
64
119
93
NOTES: FF - First Flush
EO - Extended Overflow
all values in mg/1
-------
TABLE 1-9
Date
4/3
4/3
4/17
4/20
4/23
4/28
4/28
SUMMARY
Type
FF
EO
EO
EO
EO
FF
EO
DISINFECTION DATA -
E. Coli
per ml
1388
421
1280
1850
6000
32000
26000
°3
Dosage
mg/1
80
59
40
^30
<10
<10
<10
ALL SPRING
E. Coli
in
Effluent
per ml
21
4
74
17
3200
9500
13700
STORMS
Cl
Dosage
mg/1
10
10
10
10
10
E. Coli
in
Effluent
per ml
0.5
1
2
20
8
NOTES: FF - First Flush
EO - Extended Overflow
124
-------
SUMMARY AND CONCLUSIONS
Table 1-10 presents a summary of the various oxidants and oxidation
systems utilized. Hydrogen peroxide, chlorine, and ozone were evaluated
both alone and in combination. Various catalysts were also utilized
including ferrous iron, nickle and cobalt. The best system was a
mixture of all three oxidants, which provided up to 70% oxidation of the
dissolved fraction of a combined sewer overflow. Any single oxidant
did not produce greater than 50% oxidation of the dissolved fraction.
When screened overflow was utilized, the highest oxidation percentage
was 40%, hence particulate matter was oxidized only to a very limited
extent. This indicates that an efficient solids/liquid separation system
must be utilized prior to chemical oxidation. Even if this is done a
removal efficiency of only about 40% in the chemical oxidation stage
could be economically realized. Therefore chemical oxidation of combined
sewer overflows cannot be justified.
The following conclusions can be made based on the chemical oxidation
studies performed.
1. Chemical oxidation of combined sewer overflow is not technically
feasible.
2. Ozone was the best oxidant evaluated in this study.
3. Oxidation with chlorine requires extremely high concentrations
of chlorine and results in a high chlorine residual in the
effluent.
4. Particulate matter is extremely difficult to oxidize.
5. A combination of oxidants can provide increased oxidation of
organic material.
125
-------
TABLE 1-10
SUMMARY OF OXIDATION OF COMBINED OVERFLOWS
WITH VARIOUS OXIDANTS AND COMBINATIONS
Residual
Line No.
1
2
3
4
5
6
7
8
9
10
11
Required
, Oxidant Concentration
mg/mg(l)
H202
H202
Cl
Cl
Cl
Cl
°3
°3
H202
Cl &
°3
03
H202 &
Cl
3-5
5-10
17-20
30
17
17
1.2(2)
1.2(2)
1.0
1.0
0.6
1.1
30(3)
Contact Catalyst Oxidant
Times Used Remaining
(min) mg/1
15
15
15-60
15-60
15
15-30
5
5
5
8
60
Fe"1^
Ye*
None
None
UV light
UV light
None
None
None
3 mg/1 Ni
None
small
Large
180
500
400
300
Trace
None
Small
small
Large
Type of COD
Organics Removal
%
Dissolved
w/solids
Dissolved
Dissolved
Dissolved
w/solids
Dissolved
w/solids
Dissolved
w/solids
Dissolved
Dissolved
30-50
14
10-15
20
40
25-30
40
30-40
50-70
40
40
7-10
Other Limiting Criteria
Requires pH in 3-4 range
and removal of iron floe
formed during reaction.
Requires extremely high
chlorine dosages and
hence a dechlorinations
step would be required
03 equipment requires
extremely high capital
equipment costs at
these dosages
(1) Milligrams oxidant per mg COD oxidized to obtain stated efficiencies
(2) 03 dosages are approximate
(3) Equal amounts of Cl and
-------
BIBLIOGRAPHY - APPENDIX I. CHEMICAL OXIDATION
1. Fair, G. and Geyer, J., Water Supply and Waste Disposal, John Wiley
and Sons, Inc., New York, 1961.
2. Eckenfelder, R.B. and O'Connor, D., Biological Waste Treatment,
Pergamon Press, 1961.
3. Laboon, J.F., Sewage Works Journal 7:9:911 (1935).
4. Bolenius, R.M., Sewage and Industrial Wastes 22:3:365 (1950).
5. Mills, E.V., ejt al^, The Surveyor 104:79 (1945).
6. Summary Report, Advanced Waste Treatment Research, U.S. Department
of Health, Education and Welfare, Public Health Service Publication
No. 999-WP-24, o 69, 1965.
7. Eisenhauer, H.R., "The Chemical Removal of Aklylbenzenesulfonates
from Wastewater Effluents", Presented before the Division of Water
and Waste Chemistry, American Chemical Society, Los Angeles,
California, 1963.
8. Busch, A.W., et al., API Project for Research on Refinery Wastes
(Nonbiological Treatment) Progress Report No. 2, Rice University,
p 11, 1963.
9. Davidson, C.A. and Busch, A.W., "Catalyzed Chemical Oxidation of
Phenol in Aqueous Solution", Presented at 31st Midyear Meeting of
API Division of Refining, Houston, Texas, 1966.
10. Eisenhauer, H.R., WPCF Journal, 36:9:1116 (1964).
11. Bishop, D.F., et al., "Hydrogen Peroxide Catalytic Oxidation of
Refractory Organics in Municipal Wastewaters", Presented at the 148th
National Meeting of the American Chemical Society, September 2, 1964.
12. Busch, A.W., et al., API Project for Research on Refinery Washes,
Progress Report No. 3, Rice University, June 22, 1964.
13. Connell, G.F., "Application of Ozone", Refrigeration Service Engineers
Society, Chicago, Illinois, 1966.
14. O'Donovan, D.C., Journal AWWA 57:9:1167 (1965).
15. Marsh, G.R. and Panula, G.H., Water and Sewage Works 112:10:372 (1965).
16. U.S. Patent No. 3,276,994, issued to C.W. Andrews.
17. Sease, W.S. and Connell, G.F., Plant Engineering 20:11:486 (1966).
127
-------
18. Evans, F.L. and Ryckman, D.W., "Ozonated Treatment of Wastes
Containing ABS", Proceedings 18th Industrial Waste Conference,
Purdue University, o 141, 1963.
19. Buecher, C.A. and Ryckman, D.W. , "Reduction of Foaming of ABS by
Ozonation", Proceedings 18th Industrial Waste Conference, Purdue
University, p. 141, 1963.
20. Tyler R.G.,_et al., Sewage and Industrial Wastes 23:9:1151 (1951).
21. Miller, F.K., Water and Wastes Engineering 3:12:52 (1966).
22. Moeller, T., Inorganic Chemistry, John Wiley and Sons, Inc.,
New York, p 485, 1957.
23. "Welshack Ozone Equipment", Bulletin 109, September 1966.
24. Schevchenko, E.T. , ''Tests of a Semi-Industrial Apparatus for
Ozonization for Dnieper Water", Chemical Abstracts, Volume 65,
5220, 1966, Ozonirov. Vody i Vybor Rats, 1965, (Russian).
25. Campbell, R.M., Journal of the Institute of Water Engineers,
17:333 (1963).
26. Green, D.E. and Stumpf, P.K., Journal AWWA 38:11:1301 (1946).
27. Snow, W.B., Sewage and Industrial Washes 24:6:689 (1952).
28. Griffin, A.E., and Chamberlin, N.S., Sewage Works Journal .17:4:730
(1945).
29. Private Communication, Mr. Carl Brunner, Taft Sanitary Engineering
Center, Cincinnati, Ohio, November 14, 1967.
30. Spiker, R.G. and Skrinde, R.T., Journal AWWA. 57:4:472 (1965).
31. Burtle, J. and Buswell, A.M., Sewage Works Journal 7:9:839 (1935).
32. Vosloo, P.B., Sewage Works Journal 20:1:171 (1948).
33. Gould, E.S., Inorganic Reactions and Structure, Henry Holt and
Company, New York, 1956.
34. Miller, H.C. and Knipe, W., Report AWTR-13, U.S. Department of
Health, Education and Welfare, Public Health Service, 1965.
35. American Public Health Association, Standard Methods for the
Examination of Water and Waste Water - 12th Edition, APHA,
New York, 1964.
128
-------
36. Meiners, A.F., et al., "An Investigation of Light Catalyzed
Chlorine Oxidation for Treatment of Waste Water", Advanced
Waste Treatment Research Laboratory, FWQA,Report /'TWRC-3.
37. Sawyer, C.M., Chemistry for Sanitary Engineers, McGraw-Hill
Publishing Comapny p 287, 1960.
38. Touhell, C.J., et al.., WPCF Journal 41:2LR44 (1969).
39. Ballantine, L.A., et al^_, "The Practicality of Using Atomic
Radiation for Wastewater Treatment", WPCF Journal 41:3:447
(March, 1969).
129
-------
APPENDIX II
ANALYTICAL PROCEDURES'
131
-------
Analytical Instruments and Apparatus
pil Meter: Beckman Model H-2
beckman Instruments Incorporated
Fullerton, California
Incubator: Model 1483
Precision Scientific Company
Chicago, Illinois
Conductivity Bridge: Model RC16B2
Industrial Instruments Incorporated
Cedar Grove, New Jersey
BCD Incubator: Labline No. 3554B Incubator
Lab-Line Instruments, Incorporated
Melrose Park, Illinois
Analytical Balance: Type H5
Mettler Instruments Corporation
Hightstorm, New Jersey
Ozone Cylinder: Vaporizing Coil and Regulator
Matheson Company, Inc.
East Rutherford, New Jersey
Spectrophotometer: Coleman Model 14
Coleman Instrument Company
Maywood, Illinois
132
-------
Analytical Procedures and Analyses
The following analyses were performed according to "Standard Methods
for the Examination of Water and Wastewater," (35)
The page numbers of the analytical procedures used and the modifications
to the listed procedures are noted:
Total Solids - Method A, p. 423. The amount of sample used
was determined by weight rather than by volume;
approximately 20 grams of sample was used for
total solids.
Suspended Solids - Method C, p. 424. Reeve-Angel 934AH glass
fiber discs were used in place of the mats prepared
by using asbestos fibers.
Chemical Oxygen Demand (COD) - p. 510. The standard procedure
was used for the analysis of the raw combined over-
flow. The alternate procedure for dilute samples
(p. 513) was used to analyze the raw soluble and
oxidized samples.
Biochemical Oxygen (BOD) - p. 415. The BOD tests were performed
within several hours after collection. Samples from
laboratory oxidation tests were seeded. The raw
storm water was used as seed, with a volume of 1 ml
of seed. The azide modification of the iodometric
method (Method A, p. 406) was used for the analysis
of the dissolved oxygen content.
Chloride - Method A, p. 86. A 0.423N standard silver
nitrate titrant was used instead of 0.0141N because
of high chloride content in raw storm 1^0, thus
giving a more definite endpoint.
Chlorine. Total Available - Method A, p. 91.
Ozone Concentration - Method A, p. 220. The titration procedure
stated was followed, but the sample collection and
ozone absorption techniques are those as specified
by the Matheson Company (Matheson fias Data Book).
Nitrogen (Total Kjeldahl) - p. 44.
Phosphates (ortho) - Method A, page 231. Phosphate analyses
were performed after removal of suspended solids
by filtration.
133
-------
j^H - p. 226. A Beckman Model. H-2 meter was used.
The following analyses were performed according to the procedures
listed:
Hydrogen peroxide - This was determined hy the iodometric
procedure descrihed in Tahle 3-31 in Handbook of
Analytical Chemistry. Edited hv Louis Meites,
McGraw-Hill. Book Company, New York, 1963, p. 3-69.
Coliform Bacteria Count - The millipore filter technique
described in "Techniques for Microbiological Analvsis,
(ADM-40)", (Millipore Filter Corporation, Bedford,
Mass., p. 22) was used. Where necessary samples were
dechlorinated using Na-SO-,. (35)
Conductivity - Measurements were made at 25°C using a conductivity
cell having a cell constant of 2 cm . Measurements
were made with a Model RC16B2 Conductivity Bridge
made by Industrial Instruments, Inc. according to
the procedure described by the manufacturer.
Analysis for Dissolved Master
The analysis for dissolved matter(i.e.COD, BOD, TOC) was obtained by
filtering the sample through a millipore filter (0.47) and then per-
forming the appropriate analysis (COD, BOD, TOC) on the filtrate. If
the sample contained gross amounts of solids which would rapidly blind
the millipore filter disc, the sample was nrefjItered through SS-597
filter paper. The filtrate from this prefiItration was then filtered
through the millipore disc as described above.
134
-------
Dissolved-Air Flotation Test Procedure
A copy of the instructions, for performing dissolved-air flotation
tests is attached. This sheet described the test as normally used in
the REX Chainbelt Process Laboratory. This procedure was modified at
times during the chemical oxidation tests by using different detention
times and by using ozone instead.of normal air as the source of bubbles.
The rate of separation of the suspended solids from a waste is useful
in the design of industrial waste treatment,equipment. Rate of separation
data may be conveniently obtained in the laboratory from treatment tests
performed on the waste in question. The treatment processes which will
be considered are dissolved-air flotation and sedimentation. Generally,
the procedure used in obtaining rate-of-rseparation data is to observe
the solids-liquid interface and to.record its travel with time.
A. Dissolved-Air Flotation
In the tests using dissolved-air flotation, the rate of rise of
the major portion of the solids is recorded. At times the solids-
liquid interface may be vague and good judgment may have to be
. exercised in following this interface. Care should be taken
to avoid following the interface formed by the air bubbles alone.
In general, this interface lags behind the solids-liquid interface.
A suggested procedure for the performance of laboratory flotation
tests and the equipment needed is as follows:
1. Equipment
a. Flotation pressure cell
b. Graduated cylinder of one liter capacity containing
an effluent sampling arm
c. Tire pump or source of compressed air
d. Gooch crucibles for suspended solids determinations
e. Stop watch
2. Flotation Test Procedure
a. Record waste temperature, pH, operating pressure,
recycle rate, and flotation detention time
b. Record rate of separation data. The form shown below is
suggested in obtaining the rate of separation data.
The ultimate data desired is the position of the interface at various;
intervals throughout the test. The column below labeled "Volume" is
used as a convenient means of obtaining the position of the interface
at any given time. For example, in the hypothetical case shown below, a
liter graduate was used in the test. At the beginning of the test, the
solids-liquid interface is at the bottom of the graduate or at zero volume.
135
-------
As flotation progresses, the solids-liquid interface moves progressively
up the height of the graduate. The position of the interface at any
given time may be conveniently obtained using the appropriate graduation
mark on the liter cylinder as a reference. After the flotation test,
the graduation marks may be converted to the feed of height by actual
measurement.
Time Volume POI (Position of Interface)
(min) (ml) (feet)
00 0
1 100 0.115
2 350 0.411
3 500 0.589
4 650 0.766
5 800 0.946
6 950 1.122
7 950 1.122
8 950 1.122
The data obtained are plotted using Time as the abscissa and POI in feet
as the ordinate.
POI
(ft)
TIME (minutes)
d.
The slope of the straight line portion of the curve
represents the rate of particle rise.
During flotation it should be noted whether settling
of solids took place. Note observation.
Record the floated scum volume obtained immediately
before obtaining a sample of the effluent.
Obtain sample of effluent five minutes after flotation
is started for the appropriate analyses. Repeat the
flotation and obtain another sample of effluent
for analysis after an eight minute detention period.
136
-------
e. If possible, a small portion of the floated scum
should be analyzed for total solids content.
137
-------
APPENDIX III
DEMONSTRATION SYSTEM COSTS
139
-------
The following is a breakdown of the equipment and construction costs
incurred for completion of the demonstration facility.
RAW FLOW SYSTEM
Primer System
Pipe Fittings
Flanges
Elbows
Check Valve
Ballcentric Valve
Raw Feed Pump
$5557
PRESSURIZED FLOW SYSTEM
Pressurized Flow Pump $1080
Air Compressor 550
Pipe Fittings
Elbows
Flanges
Check Valve
8" Ballcentric Valve
Pressure Tank
Grinnell Pressure Release Valve
Air Control System, Complete
$4285
BACKWASH SYSTEM
Nozzles
Valve
Backwash Pump
Elbows
$ 372
DRUM SCREEN ACCESSORIES
Backing Plates $ 400
Screen Mesh 225
Drive Unit 829
Chain, Sprockets . 120
$1574
140
-------
SKIMMER ASSEMBLY
Drive Unit
Chains & Flights, Sprockets
CHEMICAL SYSTEM
Pump
Tanks
Mixer
Hoses & Fittings
ELECTRICAL SYSTEM
Float Switches
Panel, Complete Prewired
$1300
1400
$2700
$ 886
530
268
200
$1884
$1850
FLOW METERS
Raw Feed Venturi
Raw Flow Meter (Recorder)
Pressurized Flow Venturi
Pressurized Flow Meter
Backwash Venturi
Backwash Meter (Recorder)
Scum Recorder
PIPE FOR EFFLUENT AND SCUM
6" Flanges
6" Elbows
12" Flanges
12" Elbows
$ 465
653
200
335
103
646
1060
$3462
$ 115
60
125
45
MISCELLANEOUS
Tank Fabrication
Tank Erection
Tank Steel
$ 345
$13,418
20,225
4,830
$38,473
141
-------
MISCELLANEOUS (Continued)
Electrical Power $ 930
Electrical Wiring Labor. & Material 2,307
$ 3,237
Piping Labor $ 5,862
Manhole 2,015
$ 7,877
Foundation $ 4,150
Building $ 696
Catwalks & Steps $ 1,400
GRAND TOTAL $77,862
Engineering $12,100
142
-------
APPENDIX IV
OPERATING DATA
AND STATISTICAL PROCEDURES
.143
-------
TABLE IV-1
OPERATIONAL DATA - 1969
Run
No.
691
692
693
694
695
696
697
698
699
6910
6911
6912
6913
6914
6915
6916
6917
6918
6919
6920
6921
6922
6923
6924
6925
6926
6927
6928
6929
6930
Date
6/4/69
6/4/69
6/7/69
6/7/69
6/22/69
6/25/69
6/26/69
6/27/69
6/29/69
7/2/69
7/11/69
7/16/69
7/17/69
7/23/69
7/26/69
8/7/69
8/9/69
8/11/69
9/4/69
9/5/69
9/6/69
9/23/69
9/29/69
9/29/69
10/10/69
10/10/69
10/12/69
10/31/69
11/2/69
11/3/69
Duration
(min)
123
70
103
30
75
110
150
105
50
180
75
35
60
75
40
50
150
65
35
90
60
65
180
47
65
243
75
55
45
110
Raw
Waste
(gal)
378000
189000
302400
88200
201600
340200
441000
277200
151200
554400
226800
100800
163800
226800
100800
138600
395640
190680
100800
277200
176400
201600
453600
143640
176400
705600
226800
163800
113400
327600
Screen
Wash
(gal)
0
0
1600
800
2000
<600
600
600
<600
<600
800
3200
1200
5200
<600
4800
3600
<600
1200
600
<600
600
600
<600
<600
8400
1200
2400
1800
1800
Floated
Scum
(gal)
6200
3450
3400
900
1900
3500
6250
6700
850
4400
2100
400
700
1200
450
200
1700
460
450
1850
540
700
2200
310
3850
1700
750
600
250
1150
Overflow
Rate
gpm/sq ft
2.63
2.31
2.51
2.51
2.30
2.64
2.51
2.26
2.58
2.63
2.58
2.46
2.33
2.58
2.15
2.37
2.25
2.51
2.46
2.63
2.51
2.65
2.15
2.61
2.32
2.48
2.58
2.55
1.76
2.55
144
-------
TABLE IV-1 (Continued)
OPERATIONAL DATA - 1969
Run
No.
691
692
693
694
695
696
697
698
699
6910
6911
6912
6913
6914
6915
6916
6917
6918
6919
6920
6921
6922
6923
6924
6925
6926
6927
6928
6929
6930
Inches
0.42
0.17
0.20
0.10
0.25
0.50
1.75
0.50
1.00
1.50
0.27
0.12
0.45
0.17
0.40
0.10
0.30
0.45
0.10
0.05
0.20
0.14
0.70
0.17
0.40
0.12
0.10
0.50
Inches
Hour
0.13
0.68
0.15
0.10
0.32
0.50
1.17
0.40
0.70
2.30
0.40
1.50
0.45
0.35
0.35
0.40
0.10
1.60
1.20
2.10
1.20
0.30
0.15
0.15
0.10
0.30
Operating
Pressure
scfm
50
40
50
50
50
50
50
50
60
60
60
55
50
60
60
60
60
60
60
60
50
50
50
50
50
60
50
50
50
50
Chemical
Pressurized Addition
Flow Rate C-31
gpm
600
650
850
850
900
900
900
600
700
700
475
600
900
400
400
450
600
450
800
650
500
500
500
500
800
400
550
550
550
550
mg/1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
5
0
3
3
3
3
3
5
0
0
3
0
2.5
2.5
5
Cl
mg/1
0
0
0
0
10
0
10
8
10
10
10
10
10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
145
-------
TABLE IV-1 (Continued)
OPERATIONAL DATA - 1970
Run
No.
701
703
704
705
706
707
708
709
7010
7011
7012
7013
7014
7015
7016
7017
7018
7019
7020
7021
7022
7023
7024
7025
Date
4/13/70
4/9/60
5/11/70
5/12/70
5/15/70
5/22/70
5/31/70
6/1/70
6/12/70
6/26/70
6/26/70
7/8/70
7/14/70
7/15/70
7/19/70
7/27/70
7/31/70
8/18/70
9/2/70
9/3/70
10/23/70
10/26/70
10/27/70
10/31/70
Duration
Min.
90.7
47.6
49.7
138.6
52.3
73.7
90.1
90
44.1
75.1
68
49.5
40
49.8
97.2
66.6
34.1
36.4
51.8
95.6
38
64.5
149.1
95
Raw
Waste
Gal.
252000
126000
119700
286020
117180
153720
186480
178920
95760
142380
143640
131040
133560
118440
217980
147420
75600
.10200
152700
229320
118440
153720
322560
241920
Screen
Wash
Gal.
1200
4800
360
60
0
1380
3960
420
120
3300
460
2820
1080
480
60
0
240
1740
0
0
2160
0
60
4260
Floated
Scum
Gal.
1050
500
300
500
250
350
1000
2050
1300
1900
1700
1000
1000
500
1600
1050
150
100
600
1150
200
500
750
600
Overflow
gpm/sq
High
4.75
4.52
4.16
3.53
3.83
3.56
3.54
3.40
3.71
3.24
3.61
4.52
5.71
4.06
3.83
3.79
3.79
4.79
5.08
4.10
5.33
4.07
3.75
4.35
Rate
ft
Low
3.56
3.39
3.09
2.65
2.87
2.67
2.65
2.55
2.78
2.43
2.71
3.39
4.28
3.05
2.88
2.84
2.84
3.59
3.81
3.08
4.00
3.06
2.81
3.27
146
-------
TABLE IV-1 (Continued)
OPERATIONAL DATA - 1970
Run
No.
701
703
704
705
706
707
708
709
7010
7011
7012
7013
7014
7015
7016
7017
7018
7019
7020
7021
7022
7023
7024
7025
Rain
inches
0.7
0.3
0.07
0.57
0.25
0.28
0.14
0.5
0.33
0.25
0.10
0.20
0.22
0.10
0.25
0.30
0.45
0.23
0.25
0.35
0.15
0.70
0.40
0.32
in/hr
__
0.17
0.09
0.23
0.33
0.36
0.84
1.11
0.11
0.17
0.09
0.80
0.18
0.13
0.13
0.36
2.25
0.15
0.24
0.18
0.10
0.35
0.10
Operating
Pressure
psig
50
50
50
50
50
50
50
50
50
50
60
50
55
50
50
50
50
50
50
50
50
50
50
50
Pressurized Chemical Addition
Flow Rate
gpm
800
600
900
900
900
450
450
550
550
450
350
600
950
450
450
450
450
550
550
550
600
600
600
600
C-31
0
0
6
6
6
6
6
4.2
4.6
5.3
3.9
4.6
3.5
4.5
4.5
4.5
4.5
3.8
3.8
3.8
0.5*
0.5*
0.5*
4
(mg/1)
FeCl3
0
0
0
0
0
0
0
30
29
17
16
16
21
21
21
21
21
17.5
17.5
17.5
15
15
25
25
Cl
10
10
10
10
10
10
10
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Herco Floe 810 used for these runs.
147
-------
TABLE IV-2
Run
No.
COD
CO
691 676
695 615
6911 920
6912 472
6914 574
6916 531
6919 642
702 617
708 512
7011 390
7013 649
7022 372
BOD
172
180
180
180
89
170
224
160
212
330
145
RAW WASTE CHARACTERISTICS -
Suspended
Solids
642
538
1180
496
554
405
312
582
431
232
479
415
Volatile
Suspended
Solids
391
310
678
253
322
310
244
228
247
156
289
273
Total
Nitrogen
21.3
18.0
20.7
18.6
23.3
18.9
13.3
8.8
13.9
19.5
PH
7.20
7.08
6.80
7.2
6.7
6.9
7.0
6.90
7.1
7.1
7.0
7.35
FIRST F
Total
Solids
815
798
1322
779
833
677
869
913
746
732
984
Total Ortho
Volatile Phosphate
Solids as PO
431
480
780
300
488
443
536
494
387
433
604
3.92
2.47
4.24
88
24
96
00
15
0.68
1.06
2.46
Coliform
per ml
62,000
110,000
41,000
65,000
16,000
150,000
310,000
600,000
110,000
16,650
4,400
220,000
NOTE: All values in rag/1 except pH.
-------
TABLE IV-3
Run
No.
COD
691 108
692 220
693 241
694 173
695 207
696 87
697 66
698 47
699 59
6910 78
6911 188
6913 223
6915 118
6917 185
6918 137
6920 127
6921 114
6922 248
6923 163
6924 125
6925 118
6926 95
6927 94
6928 334
6929 195
6930 217
BOD
38
43
65
17
7
7
12
49
37
31
25
61
31
39
29
36
124
70
65
NOTE:
Suspended
Solids
112
365
169
158
168
88
192
57
148
208
276
176
116
104
192
114
102
165
104
129
70
141
56
180
87
87
All values
[ARACTERISTICS - EXTENDED 0
Volatile
Suspended
Solids
58
145
117
80
102
46
55
29
43
66
105
120
62
78
82
62
64
117
68
66
22
65
45
138
63
66
Total
Nitrogen
4.1
8.1
6.4
2.3
2.0
1.4
3.1
7.6
4.8
4.3
4.6
8.1
8.4
3.8
8.8
3.3
10.2
9.1
7.0
EM
7.10
6.80
7.60
7.60
7.10
6.80
7.20
7.45
7.00
7.10
6.40
7.45
7.30
7.10
7.00
6.85
6.80
7.20
7.18
7.18
7.60
7.10
7.20
6.90
7.30
7.40
Total
Total Volatile
Solids Solids
244
502
560
312
302
188
157
170
208
280
384
541
197
445
278
247
226
478
279
218
478
199
269
450
457
390
121
215
302
168
176
102
90
86
70
102
162
250
118
222
170
160
108
296
152
105
225
82
128
272
244
224
Ortho
Phosphate
1.18
1.76
0.88
0.60
0.30
0.37
0.36
0.52
0.24
3.26
0.50
0.30
0.78
1.00
1.20
0.91
1
24
0.36
0.72
12
88
2,
1,
2.26
in mg/1 except pH.
Coliform
per ml
9,500
39,000
36,000
5,700
1,300
7,800
6,200
19,000
20,000
38,000
1,300
1,500
160,000
55,000
82,000
1,400
330
55,000
12,000
78,000
200,000
180,000
110,000
-------
TABLE IV-3 (Continued)
Run
No.
701
703
704
705
706
707
709
7012
7014
.7015
7016
7018
7019
7020
7021
7023
7024
7025
COD
143
206
141
199
140
211
168
177
191
162
130
244
286
153
90
169
178
139
BOD
36
91
58
35
28
101
37
67
113
53
25
67
54
66
RAW WASTE CHARACTERISTICS - EXTENDED OVERFLOWS
Suspended
Solids
151
218
109
316
149
228
148
119
232
140
135
337
264
466
118
107
171
123
Volatile
Suspended
Solids
75
88
39
193
75
146
86
81
139
85
95
183
164
204
63
73
125
103
Total
Nitrogen
2.9
7.4
5.4
6.3
3.7
7.5
4.4
6.3
5.55
4.9
6.95
3.4
2.4
8.7
4.9
6.4
pH
7.45
7.1
7.4
6.8
7.7
6.9
7.35
7. .2
7.2
6.9
7.1
7.1
7.4
7.15
7.05
7.7
7.1
7.4
Total
Solids
295
412
676
511
466
533
343
577
441
415
304
566
574
673
266
- 1970 DATA
Total
Volatile
Solids
112
200
267
203
246
298
160
291
220
177
145
246
250
304
116
Ortho
Phosphate
as P04
0.83
1.34
1.08
1.1
0.5
1.17
0.92
0.96
0.06
0.47
0.71
2.28
0.74
0.77
2.73
0.74
1.78
Coliform
per ml
__
270,000
12,200
700
340,000
73,000
12,000
35,500
118,800
7,660
3,800
160,000
4,800
55,000
26,000
117,000
NOTE: All values in mg/1 except pH.
-------
TABLE IV-4
SCREENED WATER QUALITY - FIRST FLUSHES
Run
No.
695
6912
6914
6916
6919
702
708
7011
7013
7022
COD
168
449
281
432
510
321
306
254
611
262
BOD
51
120
71
169
133
93
141
295
113
Suspended
Solids
170
424
260
290
286
306
272
180
418
331
Volatile
Suspended
Solids
91
234
150
290
236
126
128
106
231
173
£H
7.75
7.0
7.0
6.9
6.8
7.30
7.3
6.8
6.8
Total
Solids
833
725
536
290
848
716
623
665
906
Total
Volatile
Solids
372
300
302
369
520
303
293
394
552
7.4
NOTE: All values in mg/1 except pH.
-------
TABLE IV-5
Run
No.
693
694
695
696
697
698
699
6910
6911
6913
6915
6917
6918
6920
6921
6922
6923
6924
6925
6926
6927
6928
6929
6930
COD
132
165
159
49
26
51
49
60
144
163
89
141
103
76
91
184
138
105
122
73
78
260
171
200
BOD
__
50
15
7
5
9
37
31
21
54
27
41
20
28
90
59
60
ER QUALITY -
Suspended
Solids
123
124
148
68
152
40
132
156
249
111
67
62
139
45
88
99
95
124
79
96
47
105
72
71
EXTENDED OVER
Volatile
Suspended
Solids
69
64
77
36
42
24
32
50
107
38
39
47
55
32
51
74
60
57
26
45
35
55
54
44
- 1969 DATA
£H
7.70
7.40
6.92
7.00
7.20
7.40
6.90
7.25
6.80
6.90
7.10
7.12
7.05
6.93
6.90
7.25
7.22
7.10
7.62
7.10
7.30
6.90
7.40
7.30
Total
Solids
463
290
301
184
226
142
196
258
330
231
133
404
244
205
231
396
273
189
422
169
289
412
411
369
Total
Volatile
Solids
234
156
165
94
84
79
77
100
173
100
70
224
147
150
105
242
125
85
194
76
129
236
226
202
NOTE: All values in mg/1 except pH.
-------
TABLE IV-5 (Continued)
Run No.
COD
r-"
-n
U)
BOD
701
703
704
705
706
707
709
7012
7014
7015
7016
7018
7019
7020
7021
7023
7024
7025
133
183
126
121
90
156
84
123
183
137
80
169
201
122
57
136
82
132
35
75
15
26
18
51
18
49
71
35
13
58
29
47
?ER QUALITY -
Suspended
Solids
138
180
52
198
86
126
55
104
226
132
98
272
202
252
131
99
85
161
EXTENDED OVERF
Volatile
Suspended
Solids
66
95
16
122
44
61
39
53
114
63
56
125
108
86
156
73
64
101
7.5
7.2
7.6
7.2
7.6
7.2
6.6
7.1
6.8
6.7
6.6
6.8
7.4
7.1
6.9
7.6
6.6
Total
Solids
294
372
635
469
410
487
344
404
441
439
281
510
543
528
298
Total
Volatile
Solids
106
193
270
174
213
207
183
189
213
205
141
214
303
192
148
NOTE: All values in mg/1 except pH.
-------
TABLE IV-6
EFFLUENT WATER QUALITY - FIRST FLUSHES
Volatile
Run
No.
691
6911
6912
6914
6916
6919
702
702(1)
708
708(i)
7011
7011(1)
7013
7013(1)
7022
7022(1)
Suspended Suspended Total
COD
177
137
245
179
173
263
64
63
168
197
143
157
285
337
73
68
BOD
74
39
64
51
94
24
21
66
103
85
76
149
178
18
22
Solids
166
130
224
109
89
108
102
84
105
145
70
60
135
165
65
59
Solids
92
62
102
57
64
96
61
51
57
73
38
33
58
90
47
48
Nitrogen
15.4
9.0
9.0
13.5
2.7
2.4
7.4
8.2
4.0
4.0
7.5
8.1
5.4
4.6
pH
7.50
7.00
8.00
7.10
7.12
7.53
7.7
7.6
7.2
7.2
6.8
6.8
7.0
7.1
7.6
7.6
Total
Solids
714
411
902
533
627
959
1251
1183
440
519
545
412
690
676
Total
Volatile
Solids
320
221
437
289
314
460
481
410
206
248
297
253
406
389
Ortho
Phosphate
as PO,
8.10
2.51
1.84
1.10
3.02
0.46
0.22
1.47
2.79
0.08
0.03
0.07
0.06
1.32
1.01
Colif orm
per ml
26,000
40
16,000
80,000
8
1
1
180
1500
5700
5700
3150
500
1400
(1) Overflow rate ^ 3.75 gpra/sq ft
All values in mg/1 except pH & coliform
Overflow Rate ^ 2.5 gpm/sq ft except where noted
-------
TABLE IV-7
Run // COD BOD
691
692
693
695
696
697
698
699
6910
107
122
60
105
50
25
30
34
51
6911 102
6913 134
6915 69
6917 122
6918 67
6920
6921
57
63
6922 131
6923 99
6924
6925
6926
6927
72
54
30
45
6928 189
6929 149
6930 207
36
25
52
13
7
11
6
25
22
15
6
43
24
24
8
17
73
48
54
EFFLUENT WATER QUALITY - EXTENDED 1
Suspended
Solids
120
218
59
95
58
138
63
89
142
194
65
55
38
34
34
44
56
49
67
38
35
22
91
48
54
Volatile
Suspended
Solids
58
76
14
43
24
38
30
11
40
59
44
34
31
15
27
32
46
37
31
10
17
17
53
34
38
Total
Nitrogen
5.8
6.2
4.2
2.2
1.3
1.2
1.8
10.5
4.5
4.3
4.3
5.6
7.2
4.0
6.2
2.3
6.8
6.9
6.9
it
7.25
6.60
8.15
7.60
7.10
7.20
7.45
7.10
7.25
7.10
7.70
7.90
7. 20.
7.10
6.98
7.00
7.50
7.32
7.20
7.85
7.00
7.10
7.00
7.30
7.30
Total
Total Volatile
Solids Solids
320
380
725
606
258
234
174
194
300
286
412
434
429
475
134
222
544
250
156
457
115
221
335
358
342
147
172
306
296
140
84
95
77
125
206
220
203
240
185
73
98
282
109
56
214
46
100
184
176
182
Ortho
Phosphate
2.15
1.76
1.13
0.90
0.40
0.40
0.44
1.77
0.80
2.82
3.86
0.58
0.67
1.26
0.81
0.83
,22
1.
0.42
0.61
,58
.42
1.
1.
1.96
Coliform
per ml
9,000
61,000
<0.04
<0.04
<0.04
0.02
<0.02
0.1
<50
10,000
1,000
400
15
150
8
30
19,000
8,200
55,000
120,000
69,000
110,000
All values in mg/1 except pH and coliform
-------
TABLE IV-7 CONTINUED
ON
Run
No. COD
701 86
703 203
704 102
705 51
706 39
707 121
709 66
7012 113
7014 135
7015 99
7016 38
7018 67
7019 71
7020 56
7021 30
7023 92
7024 87
7025 57
BOD
24
74
26
11
13
42
18
46
23
29
5
36
29
24
Volatile
Suspended Suspended
Solids Solids
110
161
73
65
50
114
35
76
78
84
23
47
67
172
39
70
75
39
49
89
37
38
27
56
23
36
45
44
17
39
35
63
20
47
38
31
TENDED OVERFLOW -
Total
Nitrogen
2.9
6.4
5.4
4.5
3.0
5.7
3.4
4.8
5.4
3.6
2.5
2
1.8
6.1
3.6
4
pH
7.7
7.6
7.6
7.6
7.6
7.3
6.8
6.8
6.9
6.6
6.6
6.8
8
7.4
7.0
7.6
7.3
6.6
1970 DATA
Total
Solids
519
541
702
341
390
385
279
465
421
349
184
295
751
683
184
HIGH OVERFLOW
Total
Volatile
Solids
168
236
290
127
194
189
139
234
199
168
91
137
305
297
108
RATES
Ortho
Phosphate
.58
1.9
1.5
1.4
0.6
1.38
0.09
0.01
0.03
0.03
0
0.28
0.06
0.12
0.31
0.43
0.06
Coliform
per ml
5,000
<1
<1
3
.06
33
>10,000
11,200
8,200
49,500
1,960
600
73,000
1,700
49,000
.15,300
6,000
All values in mg/1 except pH and coliform
-------
TABLE IV-7 CONTINUED
Run
No.
COD
701
703
704
705
'706
707
709
7012 104
7014 122
92
98
184
92
55
33
102
57
34
61
7015
7016
7018
7019 102
7020 49
7021 38
7023 107
7024 69
7025 69
EFFLUENT WATER QUALITY -
BOD
23
72
18
11
7
39
18
A3
34
12
6
33
26
25
Suspended
Solids
94
141
25
65
29
73
26
57
74
88
20
46
76
71
33
55
59
51
Volatile
Suspended
Solids
40
79
4
40
20
41
21
34
44
41
15
30
37
29
19
37
49
40
EXTENDED OVERFLOW
Total
Nitrogen
2.9
2.7
5.3
4.5
2.9
5.2
3.4
4.4
3.9
2.6
3.1
1.9
1.8
5.3
3.8
3.9
r»H
7.8
7.6
7.7
7.6
7.6
7.3
6.8
6.8
7.0
6.7
6.7
6.8
7.7
7.2
7.0
7.4
7.1
6.6
- LOW OVER RATES
Total
Solids
535
508
616
311
445
343
274
499
373
350
200
285
621
567
198
Total
Volatile
Solids
174
229
241
123
237
170
152
243
183
163
91
116
274
248
116
Ortho
Phosphate
.37
2.0
1.1
0.4
1.29
0.06
0.05
0.03
0.03
0
0.95
0.18
0.09
0.15
0.43
0.06
Coliform
per ml
5,800
9
.02
38
>10,000
14,200
6,330
28,000
1,400
2,100
32,000
1,400
38,000
14,400
28,000
All values in mg/1 except pH and coliform
-------
TABLE IV-8
Run
No.
701
702
703
704
705
706
707
708
709
7011
7012
7013
7014
7015
7016
7018
7019
7020
7021
7022
7023
7024
7025
DISSOLVED COD AND TOC DATA
Raw Waste
Dissolved
COD
32
147
103
52
47
49
73
102
54
154
68
313
101
55
28
36
92
47
45
60
71
37
55
Total
TOC
35
237
75
42
71
41
249
79
139
38
57
Dissolved
TOC
12
57
39
18
16
17
128
36
13
17
25
Screened Effluent
Dissolved
COD
33
107
97
63
54
44
84
117
38
147
71
336
89
52
26
28
94
36
46
59
56
34
45
Total
TOC
31
121
53
38
36
29
235
61
91
27
49
Dissolved
TOC
12
41
37
20
15
14
137
31
12
17
18
System Effluent
Dissolved
TOC
27
64
122
59
48
34
66
105
43
113
80
219
70
39
27
33
62
49
34
34
56
38
43
Total
TOC
24
23
74
31
16
18
117
21
44
7
27
Dissolved
TOC
8
18
48
20
15
12
86
24
9
13
14
-------
TABLE IV-9
SCREEN BACK WASH & FLOATED SCUM QUALITY
1969 DATA
_ Screen Back Wash Quality
Run
£H
Total
Solids
Total
Volatile
Solids
Suspended
Solids1
693
694
695 6.8 1688 1388 1554
696 6.8 643 538 543
697
698 7.8 296 220 183
699
6910 7.1 1443 1025 1371
6911 6.5 2542 1814 2434
6912 7.0 2192 1450 1909
6913 7.2 1865 1100 1500
6914 7.0
6915
6916 6.9 3638 2486 3366
6917 6.6 1776 1406 1435
6918
6919 6.8 2303 1830 1746
6920 7.1 2431 1813 2298
6921
6922 7.1 4185 2533 3872
6923 7.2 890 694 715
6924
6925
6926 7.4 2901 972 2843
6927
6928 6.8 1600 1236 1330
6929 7.2 1623 1244 1253
6930 7.3 1322 1048 1019
1. Calculated based on dissolved solids in raw waste
All values in rag/1 except pH
159
-------
TABLE IV-9 (Continued)
SCREEN BACK WASH & FLOATED SCUM QUALITY
1969 DATA
Run
693
694
695
696
697
698
699
6910
6911
6912
6913
6914
6915
6916
6917
6918
6919
6920
6921
6922
6923
6924
6925
6926
6927
6928
6929
6930
£»
6.8
7.2
7.1
7.2
7.5
7.3
6.6
7.0
7.0
7.0
6.6
6.8
7.0
6.8
7.0
6.9
7.0
6.9
7.1
7.5
7.3
7.0
6.8
6.8
7.0
Floated
Total
Solids
2395
4005
584
686
385
1439
36860
14282
19081
11131
3705
22627
4687
7694
13650
12168
3111
2291
13228
6699
2508
11361
1705
3268
7027
2804
Scum Quality
Total
Volatile
Solids
1520
2390
330
332
202
647
19958
7550
10700
6919
2244
13403
3163
3544
8450
6644
1408
1848
8082
3471
1097
5002
1132
2100
5045
2020
Suspended
Solids1
2004
3879
484
613
272
1367
36720
13999
18716
10852
3624
22355
4346
7608
13093
12035
3087
1978
13053
6610
2100
11303
1492
2998
6657
2501
1. Calculated based on dissolved solids in raw waste
All values in rag/1 except pH
160
-------
TABLE IV-10
Run
No.
701
702
703
704
705
706
707
708
709
7011
7012
7013
7014
7015
7016
7018
7019
7020
7021
7022
7023
7024
7025
SCREEN BACKWASH AND
Screen Backwash
Screen
pH
7.7
7.2
6.9
7.1
6.6
6.7
6.9
6.7
7.1
7.1
Total
Solids
1655
3929
1891
3522
3976
3459
2817
1191
2945
2307
Total
Volatile
Solids
913
3017
1310
2723
3073
2387
2068
893
1987
1636
FLOATED SCUM
Suspended
Solids^)
1435
3599
1697
3142
3781
2959
2312
1022
2716
1940
6.7
(1) Calculated based on dissolved solids in raw waste
All values in mg/1 except pH
pH
7 5
7.3
7.2
7.4
7.2
7.5
6.9
7.2
6.6
6.6
6.8
6.8
6.9
6.6
6.6
7.0
7.5
7.2
7.4
7.1
6.6
Floated
Total
Solids
2547
8044
4431
7374
34981
8399
11379
21201
27416
11601
15675
13427
6014
15452
13812
17062
3138
16300
24064
12709
3782
24026
Scum
Total
Volatile
Solids
1455
4549
2284
4408
9835
4484
5157
9074
10406
5618
7476
6829
2606
7005
5747
6249
1589
4360
9781
6024
1807
11683
Suspended
Solids/-. \
2403
7713
4237
6807
34663
8082
11074
20866
27221
11101
15217
12922
5805
15177
13643
16836
2771
16152
-------
TABLE IV-11
FIRST FLUSH EVALUATIONS
(RAW COMBINED SEWER OVERFLOW QUALITY)
Days Since
Last
Overflow
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
3
3
Run
No.
703
692
694
6924
6930
705
709
7012
697
698
6913
6920
6921
6926
6927
7015
7021
7024
699
6911
706
704
706
7023
693
696
COD
mg/1
206
220
173
125
217
199
168
117
66
47
223
127
114
95
94
191
90
178
59
185
140
141
140
169
241
87
BOD
mg/1
91
43
31
65
35
37
49
7
7
31
25
29
36
25
54
28
58
28
67
17
Suspended
Solids
mg/1
218
365
158
129
87
316
148
119
57
176
114
102
141
56
140
118
171
148
104
149
109
149
107
169
88
Volatile
Suspended
Solids
mg/1
88
145
80
66
66
193
86
81
26
29
120
62
64
65
45
85
63
125
43
78
75
39
75
73
117
46
-------
TABLE IV-11 CONTINUED
FIRST FLUSH EVALUATIONS
(RAW COMBINED SEWER OVERFLOW QUALITY)
Days Since
Last
Overflow
3
3
3
3
3
4
5
5
6
6
6
8
8
11
11
11
12
13
14
15
17
17
18
19
19
24
26
Run
No.
6915
6925
7016
7018
7025
6910
6912
7014
6914
6923
707
6911
708
6918
6925
7013
6916
7011
7020
695
6922
7019
691
6928
7022
6919
702
COD
me/1
BOD
118
195
130
244
139
78
472
191
574
163
211
920
512
137
118
649
531
390
153
615
248
286
676
334
372
642
617
70
66
12
101
18
61
101
180
160
37
39
330
89
212
53
180
113
172
124
145
170
224
Suspended
Solids
mR/1 _
116
87
135
337
123
208
496
232
55A
104
228
1180
431
192
70
479
405
232
466
538
165
264
642
180
415
312
582
Volatile
Suspended
Solids
mg/1
62
63
95
183
103
66
253
139
322
68
146
678
247
82
22
289
310
156
204
310
117
164
391
138
273
244
228
163
-------
TABLE IV-12
Run //
691
695
6911
6912
6914
6916
6919
7021
702
708
7081
711
7111
713
7131
722
7221
COD
72.7
4.9
51.1
18.6
20.6
48.0
40.2
34.9
5.9
29.6
Screen
BOD
71.7
33.3
20.2
0.6
40.6
41.9
33.5
10.6
22.1
FIRST
FLUSH REMOVALS IN PERCENT
Overall System
SS
68.4
14.5
53.1
28.4
8.3
47.4
36.9
22.4
12.7
20.2
VSS
__
70.6
7.5
53.4
17.1
3.3
44.7
48.2
32.1
20.1
29.3
COD
73.8
85.1
48.1
68.8
67.4
59.0
89.7
89.6
67.2
61.5
63.3
59.7
56.1
48.1
80.4
81.7
BOD
57.0
78.3
64.4
42.7
44.7
90.6
89.2
58.8
35.6
59.9
64.2
54.9
46.1
87.6
84.8
SS
74.1
89.0
54.8
80.3
78.0
65.4
85.6
82.5
75.6
66.4
69.8
74.1
71.8
65.6
84.3
85.8
VSS
76.5
90.9
59.7
82.3
79.4
60.7
77.6
73.3
76.9
70.5
75.6
78.9
79.9
68.9
82.8
82.4
N
27.7
56.5
51.6
42.1
87.3
85.7
44.4
38.3
55.0
55.0
46.0
41.7
72.3
76.4
Overflow rate "^ 2.5 gpm/sq. ft. except where noted
1. Overflow rate ^ 3.75 gpm/sq. ft.
-------
TABLE IV-13
EXTENDED OVERFLOW REMOVALS IN PERCENT - 1969 DATA
Screen
Overall System
Run //
69-2
69-3
69-4
69-5
69-6
69-7
69-8
69-9
69-10
69-11
69-13
69-15
69-17
69-18
69-20
69-21
69-22
69-23
69-24
69-25
69-26
69-27
69-28
69-29
69-30
COD
45.2
4.6
23.2
43.7
60.6
__
16.9
23.1
23.4
26.9
24.6
23.8
24.8
40.2
20.2
25.8
15.3
16.0
3.4
23.2
17.0
22.2
12.3
7.8
BOD
23.1
11.7
0
28.6
25.0
24.5
16.2
32.3
11.5
12.9
5.1
31.0
22.2
27.4
15.7
7.7
SS
_._
27.2
21.5
11.9
22.7
20.8
29.8
10.8
25. G
9.8
36.9
42.2
40.4
27.6
60.5
13.7
40.0
8.6
3.8
12.9
31.9
16.1
41.7
17.2
18.4
VSS
___
41.0
20.0
24.5
21.7
23.6
17.2
25.6
24.2
43.8
68.3
37.1
39.7
32.9
48.4
20.3
36.8
11.7
13.6
18.2
30.7
22.2
60.1
14.3
33.3
COD
44.5
75.1
49.3
42.5
62.1
36.2
42.4
34.6
45.7
39.9
41.5
34.1
51.1
48.8
44.7
47.2
39.3
42.4
54.2
68.4
52.1
43.4
23.6
4.6
BOD
41.9
20.0
23.5
0
50.0
49.0
40.5
51.6
76.0
29.5
22.6
38.5
72.4
52.8
41.1
31.4
16.9
SS
40.3
65.1
43.5
34.1
28.1
10.5
39.9
31.7
29.7
63.1
52.6
63.5
82.3
70.2
56.9
66.1
52.9
48.1
45.7
75.2
60.7
49.4
44.8
37.9
VSS
47.6
88.0
57.8
47.8
30.9
3.4
74.4
39.4
43.8
63.3
45.2
60.3
81.7
56.5
50.0
60.7
45.6
53.0
54.5
73.8
62.2
61.6
46.0
42.4
N
23.5
34.4
4.3
35.0
14.3
41.9
6.2
6.8
30.9
14.3
29.5
30.3
33.3
24.2
1.4
-------
TABLE IV-13 CONTINUED
EXTENDED OVERFLOW REMOVALS IN PERCENT 1970 DATA - HIGH OVERFLOW RATES
Run Overall System
No.
701
703
704
705
706
707
709
7012
7014
7015
7016
7018
7019
7020
7021
7023
7024
7025
COD
39.9
1.5
27.7
74.4
72.1
42.7
60.7
36.2
29.3
38.9
70.8
72.5
75.2
63.4
66.7
45.6
51.1
60.0
BOD
33.3
18.7
55.2
68.6
53.6
58.4
51.4
31.4
79.7
45.3
80.0
46.3
46.3
63.6
SS
27.2
26.7
33.0
79.4
66.4
50.0
76.4
36.1
66.4
40.0
83.0
86.1
74.6
63.1
67.0
34.6
56.1
68.3
VSS
34.7
1.1
51.2
80.3
64.0
61.6
73.3
55.6
67.6
48.2
82.1
78.7
78.7
69.1
68.3
35.6
69.6
69.9
N
0
13.5
0
27.7
23.0
24.0
22.7
23.8
3.6
26.5
64.0
41.2
25.0
29.9
26.5
37.5
-------
TABLE IV-13 CONTINUED
EXTENDED OVERFLOW REMOVALS IN PERCENT 1970 DATA - LOW OVERFLOW RATES
Screen Overall System
Run
No.
701
703
704
705
706
707
709
7012
7014
7015
7016
7018
7019
7020
7021
7023
7024
7025
COD
7.0
11.2
10.6
39.2
35.7
26.1
50.0
30.5
4.2
15.4
38.0
31.0
30.0
20.0
37.0
19.5
53.9
5.0
BOD
2.8
17.6
74.1
25.7
35.7
49.5
51.4
26.9
-10.6
37.0
34.0
48.0
13.4
46.3
28.8
SS
8.6
17.4
52.3
37.3
42.3
44.7
62.8
12.6
2.6
5.7
27.0
19.0
23.0
46.0
7.5
50.3
vss
12.0
59.0
36.8
41.3
58.2
54.7
34.6
18.0
25.9
32.0
32.0
34.0
58.0
11.0
- 48.8
2.0
COD
31.5
10.7
34.8
72.4
74.6
51.7
66.1
41.2
36.1
43.2
73.8
75.1 '
64.3
68.0
57.8
36.7
62.1
50.4
BOD
36.1
20.9
69.0
68.6
75.0
61.4
51.4
35.8
69.9
77.4
76.0
50.7
51.9
62.1
SS
37.7
35.3
77.1
79.4
80.5
68.0
82.4
52.1
68.1
37.1
85.2
86.4
71.2
84.8
72.0
48.6
65.5
58.5
VSS
46.7
10.2
89.7
79.3
73.3
71.9
75.6
58.0
68.4
51.8
84.2
83.6
77.4
85.8
69.8
49.3
60.8
61.2
N
0
63.5
0
27.8
18.8
30.7
22.7
30.2
29.5
46.9
55.4
44.1
25.0
39.2
23.5
39.1
-------
TABLE IV-14
iOLIDS MASS
Pounds
SS
Out
372
337
146
157
163
499
141
111
650
362
181
88
46
124
79
64
94
184
80
54
203
41
122
45
146
BALANCE -
Pounds
SS Left
in Tank
80
145
39
63
39
92
42
59
94
86
48
43
37
25
23
29
37
33
45
25
23
15
60
32
36
1969 DATA
Pounds
SS
Screen Wash
0
0
26
16
51
15
43
12
--
19
4
199
13
19
15
Pounds
SS
Scum
.__
58
63
14
31
15
50
84
47
109
15
62
191
13
12
239
17
68
' 160
11
15
17
25
Pounds
Run SS
No. In
691 353
692 575
693 426
695 282
696 250
697 706
698 132
699 187
6910 961
6911 522
6912 417
6913 240
6915 97
6917 343
6920 263
6921 150
6922 277
6923 393
6924 154
6925 103
6926 829
6927 106
6928 246
6929 82
6930 238
(1) (+) Excess solids in; (-) Excess solids out
Balance
-28
+16
+57
- 9
+14
+12
-50
+ 9
+17
- 5
+22
- 6
- 1
+26
-16
+29
+41
-17
+ 8
-43
+29
+37
+14
-37
+ 7
168
-------
TABLE IV-14 CONTINUED
Run
No.
701
703
704
705
706
707
708
709
7011
7012
7013
7014
7015
7016
7018
7019
7021
7022
7023
7024
7025
Pounds
SS
In
316
228
108
750
145
291
667
220
274
142
481
257
138
244
211
224
225
408
137
458
247
Pounds
SS
Out
211
151
48
154
38
120
181
44
74
76
158
83
84
40
29
60
68
60
80
179
89
SOLIDS MASS
Pounds
SS
Out
211
151
48
ISA
38
120
181
44
74
76
158
83
84
40
29
66
68
60
80
179
89
BALANCE -
Pounds
SS Left
In Tank
68
100
33
43
26
63
80
20
43
43
100
50
57
15
31
48
24
41
42
44
30
1970 DATA
Pounds
SS
Screen Wash
14
67
36
38
81
54
5
54
28
Pounds
SS
Scum
21
18
17
144
17
32
173
463
175
215
107
48
63
181
21
2
154
40
53
24
120
Balance
+ 1
-47
+ 9
-1-55
+44
+14
+35
-157
-36
-135
+13
+30
-48
+ 1
+36
+35
- 9
+65
-28
+46
+ 3
(1) + Excess solids in
- Excess solids out
169
-------
STATISTICAL PROCEDURES
The following statistical procedures were utilized:
Estimate of a Mean Value
X = X ± tS
where: x - estimated mean value
x = mean of n values
t = student "t" value at n-1 degrees of freedom for some
confidence level
s = standard deviation of the data points
n = number of pieces of data
Comparison of Means Test
1. Perform "F" test
F = s
2
s2
If: FC < F Go to Step 2
If: F > F Go to Step 3
c t
Ft = Table "F" value
F = calculated "F" value
c
S1 = standard deviation of first set of data
s~ = standard deviation of second set of data
170
-------
2. Comparison of means "t" test when:
(x2)2 - ((Ix2)2/n)
-i + r\2 - 2
t = xl ~ X2
Degrees of freedom (DF) = n, + nj - 2
Read "t" table at t and DF to find confidence level at which a
difference exists
3. Comparison of means "t" test when
= (S2)2/n2
Complete DF (v) from:
= (l/v1)((Sx2)/(Sx2 + Sx2))2 + (l/v2 )((Sx2)/(Sx2 + Sx2))
,2^2
where: v-, = NI - 1
__ *T *\
t =
V Sx2 + Sx2
Read "t" table at t and DF to find confidence level at which a
difference exists
171
-------
Paired Comparison Test
1. Calculate S,.
di
where S, = /E(dj - d)
n-1
and d = x - y
dj = x - y
n = number of pairs of data
x & y = data pair
x & y = average of each set of data
2. Calculate t;
where t - d /n~
sd
Compare calculated "t" to table "t" to determine if there is a
significant difference in the two sets of data.
Corvfidpnce P.nnpv_
The Confidence Ranne was calculated bv:
p,e = x ± ts/ \/n-l
"here x = mean value
t = value from "t" table at desired confidence
range and decrees of freedom
s = standard deviation
n = number of pieces of data
172
-------
1
V
5
Accession Number
k
2
Subject Field & Croup
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Organization
Rev Chalnhpl t Tnr . Ml luaukpp . W^.QrnnR^n
Ecology Division
r;/;c
SCREENING/FLOTATION TREATMENT OF COMBINED SEWER OVERFLOWS
1Q Authors)
Mason,
Gupta,
Donald G.
Mahendra K.
16
21
Project Designation
EPA Contract
14-12-40, Project
11020 FDC
Wore
22
Citation
23
Descriptors (Starred First)
Sewage Treatment, Sewage Effluents, Sewage
25
Identifiers (Starred Firs')
Combined Sewer Overflows, Treatment Screening/Dissolved Air
27
Abstract This report documents a study to develop a treatment system for combined sewer
overflows. The processes of chemical oxidation, screening, dissolved-air flotation
and disinfection were evaluated in the laboratory. A 5 MGD demonstration system was
designed, installed and evaluated.
The system was utilized to treat 55 combined sewer overflows. The drainage area
served by the system was a 500 acre completely developed residential area of Milwaukee,
Wisconsin. Suspended solids and volatile suspended solids removal in the range of 65-80%
were consistently obtained at influent concentrations of 150 to 600 mg/1. BOD and COD
removals were slightly lower at 55 to 65% for influent concentrations of 50 to 500 mg/1.
Addition of chemical flocculents (ferric chloride and a cationic polyelectrolyte) was
necessary to obtain these removals. Without the use of chemical flocculents, removal of
BOD, COD, suspended solids, and volatile suspended solids were all in the range of 40-50%.
The screening flotation system provided sufficient detention time (^15 minutes) for
adequate disinfection with hypochlorite salts. Cost estimates Indicate a capital cost of
$21,056 per MGD capacity or $3,828 per acre for a 90 MGD screening/flotation system.
Operating costs were estimated at 3.09C/1000 gallons of teated overflow, Including
chemical flocculent addition.
(Mason-Rex Chainbelt Inc.)
Abstractor
Donald G. Mason
Institution
Rex Chainbelt Inc., Ecology Division, Milwaukee, Wis.
WR:102 IREV. JULY 19691
WRSI C
SEND. WITH COPY OF DOCUMENT. TO: WATER RESOURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON, D.C. 20240
°U.S. GOVERNMENT PRINTING OFFICE: 1972 484-485/2I7 1-3
------- |