EFA-670/2-73-077 <*
NOVEMBER 1973
Environmental Protection Technology Series
i a « n| ^ i 3 * ^ s
Combined Sewer Overflow
Papers
j *> j rt
,tf i t iit t H t
•f t, » ->;„,),
,J , 4> !'i*!^ |>1^
^ ^ ^) K ^
. V
yn- *» t
• *»" f •—•"»»
4 * I *
' t > J
^,M ^fe *** '** #1
* ,*JI^ 1?
4
_ ,—_4 *—^_wj,-t.
National Environmental Research Center
\ j
Office of Research and Development
i * ^
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
nil mil
'iiiiiiiiiiii
RESEARCH REPORTING SERIES
11 if ii
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
.......... been ........ grouped into five series. These five broad
categories were established to facilitate further
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a fna^i^'^7.1"WJ!i:!?.?)l^fi®~r.in. ,
"fields. The five series" .......... ajce'z [[[ ' .....
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
Socioeconomic Environmental Studies
5.
This report
PROTECTION
describes
demonstrate
... pollution.
.^.^.j-,. _.,_...._
:of pollution
has been assigned to the ENVIRONMENTAL
TECHNOLOGY series. This series
research performed to develop and
instrumentation, equipment and
to i repair ^ or prevent environmental
£^- ....... pQ-jjjj. ........... Q^g ............. 'nbn- point .................. source s of
This ............ work .......... provides ............... the ........ new ....... or improved
^_^__ ................ .£.„ ........... ^_.^ .......... _^__ ............ ,„.__ ..... treatment
sources .............. to .......... meet ....... environmental ...... quality
lUiililfc i
i;!!!! if fll!!!!!!!:!!"!!!!*!1!!!!!!!!!111'"1::!; I'Mii lii::!"1!;!: S:ii!il!l!i ''I1!1!:'!!";!!!! • ii'1;'"i!1,' !,J li* ill!|l!!!!|l!':"::i i1?*!!1'!1':!,'!',f': i!l!ij|l|l||lliii l!' !!!!• i1'1?!!':' Ii SRiL11:1;!!!1!!!!*'1!' I'1'1*!:!!!1" r"! • -
liiiikAl^'!,,':!'1! 1!!lll:llll|l!l"E I'll:!1!,,;"1! ilMiilKSlrri:;"'*,, ' H'WI":!!!1".!;1 i,!,1 <
si o i wsai. I SiS1* 7is f'l'fti1 K1 liiii; i iiu: s' i •'• ir; M1 si" fcifi: i', f • is'i,i n !J!,!i!':i'!i! s11::!!", fts im,i "s ifnir': 'i,1 • »i:::;, ••;, i ' i' • •":. .„
IMIiiJi;1"1!' .HI1 „ J||ill":1 ' „;>'J1,Jill1 '
-------
EPA 670/2-73-077
November 1973
COMBINED SEWER OVERFLOW SEMINAR PAPERS
A compilation of technical papers
and discussions presented at three
seminars in New York State given
jointly by the U. S. Environmental
Protection Agency and New York
State Department of Environmental
Conservation.
November 29, 1972
January 3, 1973
February 1, 1973
Edison Water Quality Research Laboratory
Office of Research and Development
National Environmental Research Center—Cincinnati
U. S. Environmental Protection Agency
Edison, New Jersey 08817
EERU-TIX
RECEIVED
APR 51989
EERU-TIX
-------
{, 10/a
73 _<>
0
-------
FORWARD
The.U.S. Environmental Protection Agency
in conjunction with the New York State
Department of Environmental Conservation
conducted three one-day seminars on the
problem of wet-weather flow pollution
abatement. Many facets of the problem
were considered including a brief over-
view of its magnitude and what the
federal government is doing to manage
and control this source of pollution.
Various management, control and treat-
ment techniques were described and the
most up-to-date information on design
and economics was presented. The audi-
ience consisted of consulting and muni-
cipal engineers from all areas of New
York State.
It is hoped that these seminars and this
compilation of papers will help solve
community problems or at least stimulate
new ideas as to how storm and combined
sewer overflow pollution abatement might
be approached.
-------
-------
CONTENTS
SECTION
PAGE
I
II
STORMFLOW POLLUTION CONTROL U.S.
PREVENTION AND CONTROL OF INFILTRATION
AND INFLOW. ' •
1
49
III : COMBINED SEWER OVERFLOW REGULATOR
FACILITIES
57
IV
V
PRESSURE.SEWERS
APPLICATION OF MICROSTRAINING TO
COMBINED SEWER OVERFLOW
65
89
VI
VII
HIGH-RATE MULTI-MEDIA FILTRATION
SCREENING/DISSOLVED-AIR FLOTATION
TREATMENT OF COMBINED SEWER OVERFLOW
115
129
VIII
HIGH-RATE DISINFECTION OF COMBINED
SEWER OVERFLOW .
153
IX
THE SWIRL CONCENTRATOR AS A COMBINED
SEWER OVERFLOW REGULATOR
171
THE EPA STORMWATER MANAGEMENT MODEL:
A CURRENT OVERVIEW
181
-------
-------
SECTION I
STORMFLQW POLLUTION CONTROL IN THE U.S.
by
Richard Field, Chief
Anthony N; Tafuri, Staff Engineer
Storm & Combined Sewer Technology Branch
Edison Water Quality Research Laboratory
National Environmental Research Center - Cincinnati
Office of Research and Development
U. S. Environmental Protection Agency
Edison, New Jersey 08817
-------
I. PREFACE
In an effort to introduce this seminar and tie the various dis-
cussions you'll be hearing today together, I thought it would be
appropriate to discuss the problem of stormwater discharges and
combined sewer overflows from the Federal Government's involvement.
The nation-wide significance of pollution caused by storm generated
discharges was first identified in a U.S. Public Health Service re-
port published in 1964. Congress, in recognizing this problem,
authorized funds under the FWPC Act of 1965 for the research, de-
velopment and demonstration of techniques for controlling this
source of pollution. Further authorization has been provided by
the 1972 Amendments to the Act.
Hence, the Storm and Combined Sewer Overflow Pollution Control Pro-
gram was originated and the problem of wet-weather flow pollution
was classified into three categories:
1. Combined Sewer Overflows
2. Stormwater Discharges
3. Non-Sewered Runoff
To date over 116 grants and contracts totalling over $82,000,000
have been awarded, the Federal Government's share being in the
neighborhood of $40,000,000 or 47.5%.
II. INTRODUCTION
The earliest sewers were built for the collection and disposal of
stormwaters, and for convenience emptied into the nearest water-
course. In later years, house sewage was discharged into these
-------
large storm drains, automatically converting them into "combined"
sewers. Subsequently, combined sewers came into widespread use
in communities because they represented a lower investment than
the construction of separate storm and sanitary sewers. (Fig, 1)
When the problem of pollution caused by sanitary or dry-weather
discharges became recognized, the engineer was confronted with
how to best separate the wet from the dry-weather flows to enable
proper treatment of the sanitary sewage portion. This was over-
come by designing overflow structures at selected points in the
sewerage system, so that combined sewage flows greater than a pre-
determined multiple of mean dry-weather flow were discharged
directly into the receiving stream. The diversion points were
usually chosen close to the receiving water for economy, and new
sewers were installed for intercepting and conveying the dry-
weather flows to the sewage works for treatment.
These overflow or relief points may also be integral to separate
sanitary systems. Initially, nominal allowances were made for in-
filtration and with pipe age this became more of a problem. Unauth-
orized connections compounded the problem, and reliefs in the "so
called" separate sanitary system were used as an immediate and low
cost solution. Studies conducted for the USEPA found that .separate
systems, with excessive infiltration and other inflows, act essent-
ially as combined sewer systems.
III. COMBINED SEWER OVERFLOW PROBLEMS
The basic difficulty with combined and "nominal" sanitary sewers
involves their "built-in" inefficiencies, i.e., their overflow
points.
Untreated overflows from combined sewers, particularly during wet-
-------
Figure 1. SCHEMATIC SYSTEM DRAWING RAINFALL THROUGH OVERFLOW
-------
weather, has proven to be a substantial pollution source in terms
of impact upon receiving stream water quality—even though the
percentage of sanitary sewage lost from the system by overflow is
small, that is, in the order of 3 to 5 percent.
Pollution problems stemming from combined sewer overflows are
widely distributed through the United States; the Northeast, Mid-
west, and Far-West being the principal areas .of concentration. In
a nation-wide survey performed by the APWA it was found that there
are over 3,000,000 acres of combined sewer drainage area contained
in more than 1,300 municipalities with a population of 54 million
served by some 55,000 miles of combined sewers. Of 641 jurisdictions
surveyed,
- 493 reported some 14,200 combined sewer overflow points,
- 340 reported infiltration problems during wet-weather and
- 96 indicated combined sewer overflows during dry-weather.
The magnitude of the overflow problem was exemplified by a 2-year
study conducted on a 229 acre combined sewer watershed in North-
hampton, England. This study showed that the cumulative yearly
biochemical oxygen demand (BOD) load in the combined sewer over-.
flows nearly equaled the BOD load contained in the effluent of the
local secondary treatment plant. .Suspended solids within the ." -/•..
overflows were three times the load contributed by the treatment
works effluent.
The relatively poor flow characteristics of combined sewers during
dry-weather when sanitary wastes alone are carried, encourages
settling and build-up of solids in the lines until a surge of flow
caused by a rainstorm purges the system. Studies in Buffalo, New
York have shown that 20 to 30 percent of the annual collection of
domestic sewage solids are settled and eventually discharged during
storms. As a result, a large residual sanitary pollution load,
-------
over and above that normally carried is discharged over a relatively
short interval of time, oftentimes resulting in what is Known as a
"first flush" phenomenon. This can produce shock loadings detri-
mental to receiving water life.
Aside from the raw domestic and industrial sewage carried in the
overflow, non-sanitary urban runoff in itself is a significant con-
tributor to the overflow pollution load. As the storm runoff drains
from urban land areas, it picks up accumulated debris, animal drop-
pings, eroded soil, tire and vehicular exhaust residue, air pollu-
tion fallout, heavy metals, deicing compounds, pesticides and PCB's,
fertilizers and other chemical additives, decayed vegetation, haz-
ardous material spills, together with many other known and unknown
pollutants. A study on a 1,067 acre drainage basin in Durham,
North Carolina has shown that the annual BOD contribution attrib-
utable to surface wash from storms is approximately equal to that
contribution of the secondary treated sanitary effluent and the
total organic matter exhibited by chemical oxygen demand was est-
imated to exceed the amount in the raw sanitary sewage from a res-
idential area of the same size.
It is important to note that there is no apt description of
"typical" combined sewage or stormwater runoff characteristics due
to the variable nature of the rainfafll-runoff patterns. Figure
2 illustrates some general concentration ranges of the wastewater
constituents listed. The major characteristic, i.e., qualitative
variability, is shown by these data. Quality may range from super-
strong sanitary sewage during the "first flush" to very diluted
sewage later in the storm. The composition is dependent on a
number of factors, including: length of antecedent dry weather,
local climatic conditions, condition of the sewerage system and
the nature of the drainage area.
-------
* FIGURE 2
GHARACTERISTICS OF COMBINED SEWER WERFLDWS
(SELECTED DATA)
'5
TSS
TOT, SOL,
VOL, Tor, SOL,
pH
SETTL, SOL,
OKG, N
Tor, €OLI,
PEG, COLI,
PEC, STREP,
30
20
3
5
,
2
TO
TO
TO
TO
TO
TO
TO
TO
TO
TO
TO
TO
600
1,700
2,300
820
8,7
1,53)
33,1
-12.5
6,2
goxio6/:
MG/L
MG/L
MG/L
MG/L
ML/L
MG/L
MG/L
MG/L
Id) ML
17X106/1GO ML
MO6/:
ICO ML
-------
As mentipned, urban stormwater in itself is a significant contri-
butor to the problem since it picks up a variety of known and
unknown pollutants as it drains from urban land area. Figure 3
illustrates some selective data on urban stormwater characteristics.
As noted, the extremely high chlorides concentrations have been
attributed to deicing salts. Our program has done some work in
this area resulting in the following conclusions:
1. Highway salts can cause injury and damage across a wide
environmental spectrum.
2. Practically all highway authorities in the U.S. believe
that ice and snow must be removed quickly from roads and
highways and that "bare pavement" conditions are necessary,
often resulting in excessive salt application.
3. Salt storage sites are persistent and frequent sources of
ground and 'surface water contamination and vegetation
damage.
4. The special additives, e.g., chromates and cyanides, found
in road deicers provoke great concern because of their
severe latent toxic properties and other potential side
effects.
5. A sufficient number of incidents and detailed studies have
been described to show adverse impact of deicing salts to
water supplies and receiving waters.
6. In less severe cases as salt intrusion into public water
supplies—salt free patients have been cautioned to change
their potable water source.
7. Deicing salts are found in high concentrations in highway
runoff.
Surveillance data is needed to clearly define the many in-
fluences of deicing salts upon the environment.
The majority of in-depth studies support the finding that
deicing salts are a major factor in vehicular corrosion and
roadway damage. The literature also indicates that rust
8.
-------
FIGURE 3
CmRACERISTICS OF URBAN STOfWATER
(SELECTED DATA)
GOD
TSS ; '
TOT, SOL,
VOL, TOT, SOL,
SETTL, SOL,
'Ore', N .
9m PO,
OUL, ru/i
TOT, P0i|
CHLORIDES
OILS
PhENOLS
LEAD
TOT, OoLi'i
FEC, COLI,
FEC, STREP,
1
5
2
TO
TO
TO
450 TO
12 TO
0,5 TO
0,1 TO
0,1 TO
0,1 TO
0,1 TO
2 TO
0 TO
0 TO
0 TO
200 TO
55 TO
200 TO
>700 MG/L
3,10) MG/L
11,300 MG/L
14,600 MG/L
L6QO MG/L
5,TO ML/L
16 MG/L
2,5 MG/L
10 MG/L
125 MG/L
25,0X3 MG/L*
110 MG/L
0.2MG/L
1,9 MG/L
MMO^/lOO ML
112xl06/100 ML
L2X105/10Q ML
*WlTH HIGHWAY DEICING
-------
inhibiting additives do not produce results to justify
their continued use. It is further noted that deicers may
attack and cause damage to telephone cables, water distri-
bution lines and other utilities adjacent to streets and
highways.
10. There is little doubt that road deicers can disturb a
healthy balance in soils, trees and other vegetation com-
prising the roadside environment.
Sewer Separation
Hhen considering combined sewer overflow problems, first attention
is generally given to the construction of separate sanitary and
storm sewer systems. In contrast, the 1964 PHS study stipulated
that alternative solutions be investigated to determine if means
other than sewer separation could be found at lower cost.
The previously mentioned APWA study of combined sewer problems in-
dicated that if all communities with combined sewers in this country
were to effect sewer separation, they would face an expenditure of
approximately 85 billion dollars at today's cost. Of this amount
New York State's share would roughly come to $18 billion, the
highest figure for any state in the nation. It was further esti-
mated that the use of alternate measures could reduce the national
figure to about 25-30 billion dollars.
It is again emphasized that urban stormwater runoff itself can be
a significant source of stream pollution. Sewer separation would
not cope with this pollution load. An EPA study revealed that if
separation were used, the reduction in wet-weather pollution would
be only 50 percent. The other 50 percent would remain in the un-
treated urban storm runoff.
10
-------
IV. CORRECTIVE METHODS
Program research, development and demonstration projects have pro-
vided significant results, and have illustrated that sewer separa-
tion in most cases is not the logical course of action. We.have
categorized three basic approaches other than separation: control,
treatment and combinations of the two.
Control
Control of combined sewer overflows can be obtained byreduction
or equalization of peak stormwater flows, increasing the effective
capacity of the sewerage system, minimizing infiltration and by
source prevention techniques.
For existing system control, the operator can attempt to maximize
wastewater treatment at the sanitary plant during wet-weather by
trying to contain as much flow or treat as much sewage as possible
during a storm flow occurrence. This would serve to reduce wet-
weather by-passing which at the beginning of storm flow can have
a high pollutant concentration, as previously described. It is
recognized this extra plant burden may decrease treatment effi-
ciencies somewhat, and create added sludge or solids handling
problems; however, these practices for only short periods during
storm flows are well worth the effort. If the operator deter-
mines that hydraulic loading will cause a serious upset of a unit
process then primary treatment plus disinfection should be con-
sidered as a minimum measure.
In, Detroit, where the prevailing direction of storms is known, ;the
operator receives advanced information on storms from a remotely
stationed rain gauge. The treatment plant pumping is increased,
thus lowering the surcharged interceptor gradient, allowing for
greater interceptor storage capacity and conveyance. This practice
11 '
-------
has enabled the city to entirely contain and treat many intense
spot storms plus many scattered city-wide rains.
The operator should also concern himself with improved regulator
inspection and maintenance, and preventive schedules to minimize
the occurrence of overflows. Overflows during dry as well as wet-
weather due to malfunctioning devices and clogged orifices can
thus be alleviated. Tide gate conditions allowing backwater in-
trusion can be corrected, and diversion structure settings can
be raised to obtain more interceptor carrying capacity.
Municipalities can also control combined sewer overflows without
large and costly modifications by concerning themselves with,in-
filtration and extraneous inflow. Excess flow caused by infil-
tration is a major thief of capacity that would otherwise be avail-
able to transport wastewater and can thereby affect proper operation
of sewerage systems and, consequently, the quality of streams.
Other adverse impacts caused by infiltration include: (a) sur-
charging and back-flooding into streets and private areas and need
for relief sewers ahead of schedule; (b) surcharging of treatment
plants and pumping stations, causing flow by-passing, decrease
in treatment efficiency, and higher treatment costs; and (c) diver-
sion of raw wastewater and greater incidence and duration of over-
flows. The APWA has reported that infiltration was a pronounced
problem during dry weather in 14 percent of communities surveyed and
in 53 percent of the communities during wet weather. The APWA also
indicates that other sources of extraneous inflow compounding the
problem include roof leaders; depressed manholes covers; cellar,
foundation, and yard drains; air conditioning and industrial cooling
waters; and other connections.
Control of infiltration should first take place during sewer pipe
installation. Better construction materials and proper installation
12
-------
techniques are necessary. The new methods of sewer-sealing and
lining should be fully evaluated before major rehabilitation or
replacement is undertaken.
Infiltration surveys should be undertaken when extraneous inflows
are suspected. Such surveys may use television and other visual
pipeline inspection, smoke tests, air and water pressure tests* and
various flow techniques. Undue deposits, partial blockages and
cave-ins causing premature surcharging and dry- and wet-weather
overflows (usually in older sewer systems) will also be pin-
pointed for subsequent corrective action.
Building connections to street sewers are a major source of infil-
tration. As much as 70 to 80 percent of the infiltration load can
occur in these lines. Accordingly, the aforementioned infiltration
control practices should be strictly followed here.
Before a municipality considers removing extraneous inflows, the
following basic factors should be considered:
1. Determination of what a "clean" or unpolluted inflow really
is. For instance, subsurface drainage may be contaminated
leachate or contain toxic material washed from basement
floors.
2. Sewer septicity and odor conditions that may arise because
of lowered flow from the elimination of long-standing in-
flow sources. .
3. Effect on the public of any sudden decision to eliminate
inflow sources and the associated problems of enforcement.
4. The strong possibility that communities will be forced tb
treat separate urban runoff sometime in the future .indicates
that the reconnection of certain so-called "clean" waters
from sanitary to storm drains may be done,in vain.
13
-------
Studies have indicated that it may be cheaper to remove solids from
the street surfaces by sweeping than by eliminating them via the
sewer system. One set of figures showed that street sweeping
costs $24 to $30/ton of solids removed as compared to $60 to $70/
ton of solids removed via the sewerage system. What may be even
more important is that the wet-weather overflow polluting potential
of these solids is eliminated by the urban surface removal practice.
Aside from abating the usual contaminants, a particular advantage
of effectively removing the dust and dirt fractions prior to sewer
entry would come from the reduction of major amounts of the more
exotic pollutants which include heavy metals (lead, zinc, cadmium,
mercury, copper, chromium), pesticides and PCB's, and nutrients
that commonly adhere to the surfaces of solids. Because of the
potential land and groundwater contamination, care should be given
to the solids disposal site selection and the fate and effects of
these pollutants. At this juncture it is appropriate to mention
that greater efforts should be applied in the area of non-routine
stormwater constituents. Their impacts and abatement measures must
be further researched, whether they be by surface "housekeeping" at
the source, or treatment of the storm flow itself.
It is recommended that the newer and more promising street clean-
ing equipment such as vacuum sweepers, air brooms and wet scrubbers
be further evaluated and employed as opposed to conventional
sweeping and flushing methods. The newer devices offer benefits
in picking up the dust and dirt particles rather than redistributing
them for aesthetic purposes as the conventional devices do.
Certain land use, zoning, and construction site erosion control
practices are other ways of alleviating the solids burde/i to 'the
receiving streams or treatment plants by surface source prevention.
14
-------
Cleansing of catch basins, sewer Tines, wet .wells and other appur-
^tenances^by flushing or dry mechanical means may reduce solid
loadings in wet-weather discharges and alleviate premature over-
flows during dry or wet periods due to partial or complete sewer
obstructions.- But here we must weigh the benefits of system
cleaning against "closing, the loop" by the installation of wet-
weather flow control and/or treatment facilities.
It is emphasized that before a community considers the establish-
.ment or continuation of the household garbage grinding practice,
it must be realized that increased solids deposition in both
combined and sanitary sewer lines will occur at times of low flow -
during dry weather which will be scoured put by the high storm flow
conditions. As a result, the overflows will create more severe
stream impacts. The jurisdiction's plans regarding future overflow
control and treatment will be an important consideration since
again the "loop" will be closed.
If there is insufficient carrying capacity in the "sewer lines,
polymer addition may serve to measurably reduce fluid friction.
Research has shown that polymeric injection can increase flow
capacity as much as 2.4 times at a constant head. This method can
be used as a measure to correct troublesome pollution-causing con-
ditions such as localized flooding and excessive overflows. Pre-
liminary cost comparisons have shown this procedure to be feasible.
', . , '"$ ' •• " - ','', •'"' : . ,-':'•',. .' - ' .''-"-,- L - ' f " .
Advanced Control /Systems
In this segment of the talk, some of the newer and more advanced
technology being developed by our Program will be described.
15
-------
Flow Regulation
.Several methods have been used to reduce operation problems asso-
tciated with the conventional regulator devices. Cincinnati utilizes
telemetered monitoring to detect unusual or improper dry-weather
overflows. More sophisticated approaches are being applied by the
Minneapolis-St. Paul Sanitary District and the Cities of Detroit,
'and Seattle. All three jurisdictions are making use of unused
storage capacity within the existing sewerage system for the purpose
Vpf reducing the frequency and volumes of overflows. For instance,
""in the period from 1969 to 1970, Minneapolis was able to reduce
overflow occurrences by 55% and the volume of overflow by B5%. The
general approach comprises remote monitoring of rainfall, flow levels,
'and sometimes quality, at selected locations in the network, together
with a centrally computerized control console for positive regulation
of the overflow structures. Figure 4 depicts the computer console
and strategy room in Seattle, and is a preview of what the operator
in 1980 may be contending with.
New types of regulators such as positive control gates and in-
flated rubberized-fabric dams (Figure 5) have been demonstrated
successfully. Another unique overflow device which has been
constructed for full-scale demonstration utilizes fluidic technology;
and requires no moving parts or external power since operation is
'.^entirely dependent upon motion of the wastewater. Improved regulator
capability and reduced operation and maintenance costs are anticipated.
Additional improvement in regulators is now in progress.
t
Storage
'Storage offers direct control by containing the wastewaters pro-
duced during wet-weather periods. The use of storage facilities
for controlling combined sewer overflows has been convincingly
demonstrated. The general procedure involves the return of re-
16
-------
Figure 4. Computer console for augmented flow
control system, Seattle, Washington.. ; .. • .
17
-------
t r s ft at
Figure 5. Inflatable Control Gate System
18
-------
tained overflows to the conventional treatment works for subsequent
treatment during low flow, dry-weather periods.
Concrete holding tanks are the most commonly used type of storage
facility. The storm stand-by tanks at Columbus, Ohio, shown in
Figure 6, constructed as early as 1932, were recently modernized by
installation of sludge collection and automatic flow control equip-
ment. The City of Boston has commenced operation of overflow hold-
ing tanks designed to provide 10-minute settling plus chlorination
for treating excess overflows of 233 million .gallons per day. New
York City and Milwaukee have similar facilities in operation. The
New York City plant has four storage, tanks which have a combined
capacity of 9.7 million gallons. Intercepted storm flow is stored,
degritted, and pumped, along with the sludge back to a nearby
Municipal Treatment Plant. Excessive overflows receive treatment by
sedimentation through the tank and are chlorinated and discharged.
The objective of the facility is to reduce coliform and solids
contamination of Jamaica Bay.
Chippewa Falls, Wisconsin has constructed an asphalt-lined basin
providing storage for up to 3.5 million gallons of overflow.
(Figure 7) During the 1969 - 1970 evaluation period, 50 river
discharges out of 62 storm overflows were eliminated.
Two basic problems encountered by conventionally-designed storage
facilities in urban areas are land cost and availability, and
adverse aesthetic impacts. In this regard, we are seeking new
concepts. A major demonstration in Chicago involves the-new
concept of "deep tunnels". The cost of the Metropolitan Chicago
tunnel storage system is estimated at over one billion dollars as
contrasted to over four billion for sewer separation. Additional
benefits of turtnel (or in-sewer) storage are a result of coverage
of an expanded area or length. Thus, storage is more readily avail-
19
-------
2S&BI
Figure 6. Storm Stand-by tank with upper portion
of sludge collection mechanism visible, Columbus,
Ohio.
Figure 7. Asphalt-lined basin providing storage
for up to 3.5 MG, Chippewa Falls, Wisconsin.
20
-------
able to remote areas, hydrographs may be smoothed or reduced for
treatment facility design because intense storms often are quite
localized, and overflows greater than storage capacity can be select-
ively and automatically'discharged!to the most suitable stream loca-
tions. Another subsurface storage idea, to be demonstrated in
Lancaster, Pa., is the underground "silo". The use of a 50-foot
diameter, 100-foot deep silo could afford over 1 million gallons of
storage. The preliminary design is shown in Figure 8.
Other designs requiring little or no urban land include offshore
storage and the use of natural underground formations. Two demon-
stration projects have evaluated the use of flexible neoprene-coated
nylon fabric material as underwater containers, for the temporary
storage of combined sewer overflows. Figure 9 presents a drawing
of such an installation. :
The engineer and operator will be interested in the sludge-handling
aspects of temporary storage. Two possibilities are the re-suspension
of solids by agitators and settling prior to pump-back. Re-suspension
can provide easier draw-off and is being evaluated. However, if
sludge is settled, on-site sludge disposal in lieu of solids pumped
back in stored flow should be considered.
Design criteria should be based on the pollution abatement results
expected. For example, Milwaukee used a mathematical model to
determine size and projected efficiency of its holding tanks.
Wherever possible, design of full-scale facilities should consider
the total environmental impact, including aesthetics. Figure 10 is
a conceptual drawing showing an off-shore site in Lake Erie at
Cleveland, Ohio
A concept worthy of note, which was successfully demonstrated in
21
-------
SCUM RING VORTEX fOUi. OUTLET
Preliminary Drawing — Elevation View of System, Lancaster, Pa.
PLAN AT EUVATJON 256'-0"
Figure 8. Preliminary Drawing - Plan View of System, Lancaster, Pa.
22
-------
FIGURE 9
UNDERWATER TANK
INTERIOR SECTIONED VIEWS
COATED FABRIC
STEEL FRAME-
FLUSHING JETS
23
-------
Figure 10. Conceptual design of combined sewage
retention-stabilization basin, Cleveland, Ohio
24
-------
London, England and Decatur, Illinois, is the conversion of existing
or abandoned sanitary treatment units, in this case sedimentation
tanks, to storm holding facilities as part of a plant expansion.
Also, plans have been proposed to use an abandoned trickling filter
as a storage tank for stormwater infiltration.
Porous Pavement •
Another feasible method to attenuate flows is the installation of
porous pavement; This pavement'is made of asphaltic-concrete and has
been developed for an ability .to, allow 60. or more inches per hour of
rainfall to permeate, through its. depth ^(Fjgure;OJ)vVri|;^$e^.|or
major highway, street, and parking lot paving-projects, it would have
the potential for reducing capacity and associated costs for both
sewer and wet-weather flow treatment systems, a feature attributable
to the porous pavement's ability to equalize flows entering or divert
flows away from the sewerage system. This type of pavement install-
ation can;lal so offer a'substantial benefit by recharging::water supplies.
Even more important are the safety 'features :whleft-could be-realized, .
i.e., an increased coefficient of friction which will help prevent
wet skidding or hydroplaning accidents, and enhanced visibility of
pavement markings due to more rapid removal of'rainwater and rougher
surfaces. However, when porous pavement is considered, we must real-
ize that such features as geographical area, temperature, subsurface
soil condition, and the possibility of groundwater contamination may
play an important part in design-and site selection.
New Sewer Systems
New types of sewer systems being demonstrated, based on vacuum and pres-
sure operation for the collection and conveyance of sanitary sewage,
can reduce the waste volume generated, reduce conduit sizes, eliminate
infiltration, minimize associated installation and treatment costs,
25
-------
Treatment methods which have been evaluated or are currently under
investigation by the Storm and Combined Sewer Pollution Control Pro-
gram include:
1. Fine-mesh screening and microscreening
2. Dissolved-air flotation
3. Rotating biological contactors
4. High-rate plastic and rock media trickling filters '*'"
5. High-rate, single, dual and tri-media filtration
6. Swirl and helical separators"
7. Advanced disinfection methods, e.g., high-rate application,
on-site generation, automated operation, ozonation, and use
of combined halogens (chlorine and iodine) and chlorine di-
oxide
8. Tube settlers
9. Powdered and granular activated carbon adsorption
10. Polymer and other chemical additives for improved settling,
microscreening, filtration and flotation
11. Chemical oxidation -
12. In-line or in-sewer treatment
13. Sludge handling and treatment
14. Regeneration of carbon and coagulants, and
15. Reclamation and reuse.
Time does not allow a detailed discussion of each of these methods
Some of the more promising treatment techniques will be discussed.
Since high throughput rates, are necessary for combined sewer over-
flows, the sanitary treatment processes are being studied for poss-
ible modifications. For example, the microstrainer is conventionally
designed for polishing secondary sewage plant effluent at an optimum
rate of around 10 gallons per minute per square foot, tests on a
28
-------
pilot microscreening unit in Philadelphia, Pa. have shown that, at
2
high flux rates of 35 to 45 gpm/ft , suspended solids removals in
combined overflows exceeding 99 percent can be achieved. Mr. George
Glover will speak about this in more detail this afternoon.
Increased flow rates greatly reduce capital costs and space require-
ments. Increased throughputs have also been obtained with other
fine-mesh screening processes, for example, fiberglass filtration
and dissolved-air flotation.
An EPA study in Cleveland showed high potential for treating com-
bined sewer overflows by contact coagulation and ultra high-rate
filtration. Figure 12 depicts the process flow diagram. With the
2
high loadings of 16 to 32 gpm/ft surface area, removal of solids
is effectively accomplished throughout the entire depth of filter
column. Test work showed suspended solids removal up to and
exceeding 90 percent and BOD removals in the range of 60 to 80 per-
cent. Substantial reductions, in the order of 30 to 80 percent of
phosphates, can also be obtained. Mr. Pat Harvey will discuss this
at length later on today.
Results from a 5.0 MGD screening and dissolved-air flotation demon-
stration pilot plant, in Milwaukee, indicate that greater than 70
percent removals of BOD and suspended solids are possible. Find-
ings also revealed 85 to 97 percent reduction in suspended,sol ids,
and better than 90 percent reduction in phosphate can be achieved
as an additional benefit, by employing chemical coagulants. Mr.
Gupta will give his presentation on this topic this afternoon.
A unique variation of the usual coagulation-adsorption, physical-
chemical treatment process has been demonstrated in Albany. This
29
-------
cc
1
Q
U.
CO
CO
UJ
O
O
or
Q.
i3
CO
cs
O
2
Q.
I
QC
uJ
O
CO
UJ
^
cr
x
S2
x
30
-------
system, shown schematically in Figure 13, is comprised of a 100,000
GPD trailer mounted pilot plant where both powdered carbon and coag-
ulants are added in a static mixing-reaction pipeline, and the result-
ant coagulated matter is flocculated downstream, separated by tube-
settlers and polished by multi-media filtration. The project also
demonstrated regeneration of alum and activated carbon by fluidized-
bed incineration. ,
At this point it is appropriate to bring out an important fact of
which future designers of storm overflow treatment facilities must be
cognizant--process efficiency should not be considered in the usual
terms of percent removal used in municipal treatment. It was found
during the microstrainer and dissolved-air flotation operation that,
due to extreme variation of the influent suspended solids concentra-
tion, removal efficiency would also vary while the more desirable
effluent concentration remained relatively constant. For example,
a typical effluent concentration of 10 mg/1 suspended solids would
yield a reduction of 99 percent for an influent concentration of
1,000 mg/1, whereas the suspended solids reduction would be only
50 percent if the influent concentration were 20 mg/1. This pheno-
menon is apt to reoccur in other physical-chemical stormwater treat-
ment operations.
Another project has studied a new biological process, described as
the rotating biological contactor consisting of a series of shaft-
mounted rotating disks. Similar in principle to trickling filtration,
a biological growth attaches onto the disks. Under steady loading
rates, efficiencies exceeding those of the trickling filter have been
attained, but a surge tank appears essential. Figures 14 and 15
give a close-up of the rotating disks and an overall view of the
pilot facility, respectively.
Another approach in overcoming the extreme variation in overflow rates
is to provide surge facilities prior to the storm treatment plant or
31
-------
H
ffl
g
H
u
H
K
CO
I
60
•H
32
-------
Figure 14. Close-up view of rotating biological
disks, Milwaukee, Wisconsin.
Figure 15. Overall' view of rotating
biological, disks, Milwaukee, Wisconsin,
33
-------
the municipal plant. The surge basin(s) (or existing combined sewers)
could furthermore serve a dual function in equalizing not only wet-
weather flows but dry-weather flows as well. In this way, a single
future treatment system can readily be designed for storm and sanitary
flow conditions. This could also assist presently overloaded sani-
tary plants in obtaining more uniform operation. Short-term storage
incorporated into the treatment plant would even out the daily cycle
of dry-weather flows allowing for more efficient use of the treatment
process over the entire 24 hours. Equalization would permit reduced
treatment process design capacity. Further analysis is necessary to
determine the most economical break-even point between the amount of
storage versus the treatment capacity. The designer should recognize
the wet-weather treatment plant's capability to draft stored flow con-
tinuously while it is raining in his evaluation of the optimum surge-
treatment system.
New Orleans has demonstrated the use of sodium hypochlorite for dis-
infection of storm flows as high as 11,000 cfs, to both reclaim and
protect public bathing beaches. In order to economically provide the
large quantities of disinfectant required, an on-site hypochlorite
batching plant was constructed (Figure 16). Figure 17 gives a view
of the massive-size chlorine contact basin in operation.
The disinfection of combined sewage entails certain differences,
which make the design and operation of facilities difficult when com-
pared to sanitary sewage. The highly varying qualitative and quan-
titative character of the storm generated inflows require disinfectant
dosages to be based on a predicted rather than an established tech-
nique. A decrease in temperature decreases disinfectant kill power.
This points to the importance of temperature in additon to the usual
(time and dosage) control parameters. As temperature is apt to have a
much wider range for runoff waters than it does for domestic sewage
flows, combined sewage may require disinfectant dosage to vary season-
34
-------
Figure 16. Stormwater disinfection project -
hypochlorite batching plant, New Orleans, Louisiana.
Figure 17. Stormwater disinfection project - chlorine
contact basin, New Orleans, Louisiana^
35
-------
ally or as effected by ambient temperature.
The Storm and Combined Sewer Overflow Technology Program is also
searching for high-rate disinfection systems to save on large
tankage requirements for the high storm flow rates encountered,
with the help of more rapid oxidants e.g. chlorine-dioxide, and, •
by imparting greater turbulence to the flow. Successful attempts
toward high-rate disinfection are being noticed at our Phila- !
delphia, Pa. and Onondaga County, New York demonstration sites.•••.>
The Philadelphia project also made an evaluation of ozone, gener-
ated on-site for disinfection purposes. Another study proposes .
the use of combined halogens (chlorine and iodine) to provide
more effective disinfection of viruses as well as bacteria in a
swimming lake. This study also supports dechlorination by.acti-
vated carbon or use of ozone, with a relatively short half life,
in lieu of chlorine to alleviate residual toxicity problems to fish
life. Mr. George Glover will present more on this subject.
Combinations
When a single method is not likely to produce the best possible ans-
wers to a given pollution situation, various treatment and control
measures—as previously described—may be combined for maximum flex-
ibility and efficiency. One such combination might be: in-sewer
or off-system storage for subsequent overflow treatment in specifi-
cally designed facilities, followed by groundwater recharge or
recovery for water sports and aesthetic purposes. Another combin-
ation might be flow retention with pump or gravity feed-back to the
sanitary sewerage system.
In all cases.the optimum abatement plan for stormwater overflow
pollution will have to be evaluated separately for the geographical
area in consideration., Aside from climatological conditions, terrain,
and land uses, choice of control and treatment will depend on the
36
-------
existing sewerage system configuration. For example, systems with
large contributory areas and .few overflow points present problems
and require design philosophies which differ from those in systems
divided into many subdrainage areas with individual combined waste-
water outfalls.
The temporary storage concept, previously discussed as a control pro-
cess, also provides for a certain degree of treatment by settling, for
excessive overflows greater than the design storage capacity dis-
charging directly to the receiving stream. Likewise, this settling
potential for flows less than design capacity, together with on-site
solids disposal usually overlooked, should be definitely considered.
The proposed prototype demonstration for Lancaster, Pennsylvania,
previously cited and shown schematically in Figure 18, will pre-treat
by a swirl device and microstrain and disinfect discharges greater
than the storage capacity of the "silo" structure.
Mr. Clemens, Michigan installed a system involving discharge of
combined sewage overflows into a series of three "lakelets" each
equipped with surface aerators. Effluents pass from one pond to the
next through microstrainers and filters, and the final effluent is
.chlorinated. This, control and treatment scheme is designed to have
no adverse aesthetic impacts, and the possibility of reusing these
waters for recreational purposes is being explored. Figure 19 shows
a schematic of the Mt. Clemens facility.
A conceptual engineering study for the Washington, D. C. area (Figure
20) has shown that it would be feasible to construct a control-treatment
facility to handle combined sewer overflows up to 3,000 cfs. A 175
million gallon storage facility is tentatively planned with an over-
head parking garage, coupled with a 50 MGD high rate filtration-
-.. adsorption-disinfection plant. This treatment complex is intended
to produce reclaimed waters suitable for swimming, boating, and fishing.
37
-------
-------
LU
< s
I— ^
2 '•%
5 I
!—
^ w)
S C
C^ ®
i- E
§^
CTv
H
01
H
60
•H
<
Z
O
i—
Z
UJ
i—
UJ
fid
c
g'
39
-------
O
E
tt
Li
** LLJ ^
o -5 <
«r O "-
O< u
2 "J
OSO ^
ui z QS
I- — ULI
n
UI
O
UI
O
cs
UJ
ex.
D
O
2
Ou
<
40
-------
Our Program, in conjunction with APWA, has refined and is demonstra-
ting the swir.l flow regulator/solids-liquid separator (Figure 21).
The device is;of simple annular-shaped construction requiring no
moving parts. It provides a dual function, regulating flow by a
central circular weir, while simultaneously treating combined sewage
by swirl action which imparts liquid-solids separation. The low-
flow concentrate is diverted to the sanitary sewerage system, and
the relatively clear liquid overflows the weir into a downshaft and
receives further treatment or is discharged to the stream. This
device is capable of functioning effectively over a wide range of
combined sewer overflow rates having the ability to effectively
separate settleable and light-weight organic suspended matter at a
small fraction of the detention time required for conventional sedi-
mentation. For these reasons serious.thought is now being given to
the use of swirl units in series and in parallel solely as wet-,
weather treatment plant systems. A helical or spiral type regulator/
separator has also been developed based on similar principles "as the
swirl device, and .we are looking for further refinement. Mr,. Richard
Sullivan will speak on this subject following my presentation.
Flow Measurement
The quantitative %and qualitative measurement of storm overflows .is
essential for process design, control, and evaluation. The "urban
intelligence systems" previously mentioned require real-time data
from rapid, remote sensors in order to achieve remote control of a
sewerage network. Conventional flow meters have not been developed
for the highly-varying surges encountered in combined sewers. Here,
a measuring device may be subjected to very low flow rates, submer-
gence, reverse flow, and surcharge, all during a single rainstorm.
These severe flow conditions rule out the reliable and accurate app-
lication of conventional devices, such as weirs and flumes at many
41 •
-------
inflow
foul
sewer
overflow
a
b
c
d
e
f
g
h
Legend
Inlcl Ramp
Flow Deflector
Scum Ring
Overflow Weir and Weir Plate
Spoilers
Floatables Trap
Foul Sewer Outlet
Floor Gutters
Figure 21
Isometric View of Swirl Regulator/Concentrator
42
-------
locations. Consequently, we are deeply Involved in the development
and demonstration of sophisticated and new flow measuring equipment
utilizing the various principles of: hot-film anemometers, concen-
tration of induced foreign matter, ultrasound, and electromagnetics
as applied to open channel flow.
Our Program has also contributed towards the development of a pro-
totype monitor capable of instantaneous, in-situ measurement of sus-
pended solids based on the optical principle of light depolarization.
USEPA Stormwater Management Model (SWMM)
The capability to analyse various component flows and pollution loads
throughout a sewerage system is one of the keys to better design of
control and treatment systems. Due to complexities of the rainfall-
runoff- flow phenomena past analyses have been less than adequate,
resulting in poor estimates of flow and predicted system responses to
a storm. By virtue of previous undertakings, we now have available
an operational "descriptive" mathematical model which can overcome
former analytical deficiencies. Figure 22 depicts a schematic over-
view of the model.
We are now in the initial phase of demonstrating the application of
this method for "decision-making", that is, its ability to analyse a
major combined sewer system to s.elect and to design control and treat-
ment approaches based on cost/effectiveness and to eventually design
a computerized means of overall management of the system during storm
flows. The model will be fully explained later on today by Dr. Wayne
Huber.
PROGRAM PROJECT NEEDS
Looking ahead, the Storm and Combined Sewer Pollution Control Program
43
-------
RUNOFF
(RUNOFF)
DECAY
(QUAD
INFILTRATION
(INFIL)
TRANSPORT
(TRANS)
EXTERNAL
STORAGE
(STORAG)
RECEIVING WATER
(RECEIV)
DRY WEATHER
FLOW
(FILTH)
INTERNAL
STORAGE
(TSTRDT)
COST
(TSTCST)
TREATMENT
(TREAT)
Note: Subroutine nalnes are" shown in parentheses,
I
Figure 22, OVERVIEW OF MODEL STRUCTURE
44
-------
needs are vast and numerous.
to the following:
At present, we are directing our efforts
1. A nation-wide assessment of sewered and non-sewered straight
urban runoff impacts, not combined sewage - a consideration
which has been stressed by the 1972 Amendments to the FWPC :
Act.'' ' """"" '" ' ' '•' ';"'-" •' ;;" ;
2. Dual use facilities for wet-weather and dry-weather treat-
ment. Wet-weather facilities built in conjunction with new
or existing sanitary.pi ants can demonstrate their synergistic
benefit by being utilized to take over during repairs, polish-
ing secondary effluents, or increasing dry-weather treatment
capacity during the vast majority of the time, i.e., when
it is not raining. . -. '
3. Land development making full use of runoff and natural drain-
age - aesthetically blending into the surrounding environment
rather than upsetting it.
4. Wet-weather facilities for treatment of dry-weather creek flow,
again making full use of these.facilities during otherwise
downtime.
5. A stormwater model monitoring/management system for dissemin-
ation, updating, and instructions on model application.
6. A functional evaluation of the need for catch basins today -
and development of new alternatives.
7. Establishment of uniform techniques for sampling and analysis
.of storm flow and for determining design volumes and flowrates.
8. Further development of flow measuring devices.
9. Fostering a stormwater survey course at the university grad-
uate level. Storm generated pollution ranks high along with
domestic and industrial sources and yet remains unstressed in
the schools. With wet-weather control requirements evident,
now is the time to encourage universities to cover the concepts
of stormwater runoff and combined sewer overflow pollution
in proper perspective in their graduate school water pollution
45
-------
control curriculum.
10. The swirl device applied for grit removal and primary separa-
tion of solids from combined sewage, stormwater, erosion run-
off, along with the optimization of its sister device, the
helical flow regulator/solids separator.
There are also certain major control methods requiring further.develop-
ment. "Upstream" storage or other control processes to decrease the
stormwater runoff effect on lower portions of the system is one case
in point. Aside from the main objective of controlling storm-generated
pollution, upstream control can preclude the need for additional down-,
stream sewer line capacity and associated construct!'o.n requirements,
alleviate shock loadings due to scouring velocities, relieve the often
occurring expense of constructing facilities downstream near water-
courses in unstable soil with high water table, while offering greater
flexibility for control and treatment. An example of this would be
the temporary storage or attenuation of stormwater at the building or
immediate area through the use of holding tanks, seepage pits (possibly
for recharge), rooftops, parks and playgrounds, backyard detention
facilities, porous pavement (previously discussed) or neighborhood
decentralized stormwater collection sumps including storage facilities
under streets. Upstream control systems should automatically regulate
discharge from storage to the groundwater, a watercourse, or a sewer
system. Plans for reuse of stored water for irrigation, street clean-
ing, sewer flushing, aesthetic and recreational ponds, potable supply,
and other purposes is also encouraged.
Many more ideas and concepts could be added - some may be more sig-
nificant than those discussed. Submission of ideas, project proposals
or grant applications to the USEPA is strongly encouraged.
46
-------
CONCLUSION .-• . • -. • •. • .-.. •-• •-
All facts point-to a real requirement for treating and controlling
stormwater runoff and combined sewer overflows. In view of the tre-
mendous quantities,,of pollutants bypassed during rainfall, from the
combined sewer,system, it does not seem reasonable to debate whether
.secondary treatment plants should be designed for 80, 85, or 901 BOD
or suspended solids removal, when in fact the small increments .gained
in this,range are completely overshadowed by the;bypassing occurring
at regulators, during wet-weather flow.
The multirbillion dollar treatment plant upgrading and expansion
.program now going on throughout the country-will do much to alle-
viate pollution of our waters. However, means of mitigating the
effects of combined sewers must also be found if we hope to abate'
the pollution in an optimal manner. Wet-weather standards are al-
ready being instituted by the federal government and, some states and
localities. Recognizing this, our Program will strive to be a prime
support for this real world application.
47
-------
-------
SECTION II
PREVENTION AND CONTROL OF INFILTRATION AND INFLOW
by
Richard H. Sullivan
Assistant Executive Director
American Public Works Association
49
-------
The American Public Works Association has had an active program
of research in the field of storm water pollution. Its program has
investigated such fields as the pollution of storm water, the extent
of combined sewer facilities, the design, operation and maintenance
of combined sewer overflow regulators and the prevention and correc-
tion of excessive infiltration and inflow into sewers. These pro-
jects were either conducted under contract with the U. S. Federal
Governement or as cost-sharing projects jointly financed by local
public agencies and the federal government. My remarks today will
be based upon the research findings of our Foundation.
I will briefly review some of the major findings of our report,
"Prevention and Control of Infiltration and Inflow". I will also
review with you guidelines for the establishment of a survey to
determine the nature and extent of infiltration, and some of the
factors to be used in making an economic analysis of desirable
corrective actions.
In our study of the problems of combined sewer facilities it
50
-------
became evident that infiltration plays a major role in many facil-
ities by either causing more frequent or prolonged overflow events.
With the assistance of some 34 local agencies and the Water Quality,
Office, we undertook a study of the prevention and correction of
infiltration. For ease of discussion we decided to consider the
"Two I's" of infiltration. The first "I" - infiltration - is in
the classic sense, that flow which enters the sewer through pipe
and joint defects and manhole covers, etc., and - inflow - is
surface water which is deliberately introduced into the system
'through footing drains, downspouts, area-way drains, and .such.
Infiltration and inflow both take up capacity within the'collection
system. However, the two have entirely different characteristics
as to time of occurrence, and means of correction and prevention.
If infiltration and inflow exist, why should we be concerned?
One of the most common problems associated with excessive infil-
tration or inflow is backups into basements, flooding of manholes,
treatment plant overloads, pavement and sewer failures; all are
common problems. Exfiltration may result in pollution of the
groundwater table.
When we look at the extent of infiltration, we can conclude
that all sewers are combined, it is all a matter of degree. Where
even minimal amounts of infiltration and inflow are present, a
regulator device of some type will be used on the sanitary sewer
system to relieve the excess flow condition. Quite often this
is only a leader from a sanitary sewer to a storm sewer, or a
hole in the side of a sanitary sewer manhole which, under sur-
charge conditions, will allow excess flow to enter a creek or stream
bed. For such systems to be described as "separate" is ironic,
inasmuch as its volume of non-sanitary flow may reach 40 to 1, as
contrasted to the strict combined system where this could be 90
to 100 to 1. •
51
-------
Correction of infiltration problems can ,be categorized under
the dual headings of prevention of infiltration and inflow in new
systems and the correction of existing conditions. :;
With regard to new construction; tremendous advances have been
made in pipe and joint materials. Contractors;and pipe suppliers
who worked with the APWA in the preparation of-the report were
agreed that a construction standard of 200 gallons per inch-mile
per day was reasonable and could be met without additional cost to
the local agency. In practice we found that consulting engineers
had, in effect, an extremely wide array of,construction standards
.which they regularly cite for new construction. There was,little
agreement as far as to the unit of measure or how the standard
would be applied. In this regard I think it is important to :remember
the effects of a low standard for gallons per inch per mile applied
to lengths of 200, 300 and 500 feet. Allowable infiltration may be.
almost impossible to measure. Specifications using low infiltration
rates should spell out how compliance is to be measured. For example:
200 gallons per inch per mile per day allows 4.4 gallons in an 8-inch
pipe an hour between manholes 350 feet apart.
The detection of infiltration is a time consuming and generally
expensive process. I am not aware of any short cuts to the prepar-
ation of a comprehensive survey. Our report contains an outline of
a ten-point program as developed by the American Pipe Services Co.
of Minneapolis, Minnesota. . For purposes of our discussion.today I
have expanded this to twelve points, and would like to consider: ...
these steps briefly with you. , . . •• ,,
The steps involved in a complete infiltration-inflow.analysis
include:
1. SET OBJECTIVES: detemine what is the apparent problem,
.in what condition is the sewer system, is there an adequate
52
-------
maintenance program, how can sources of infiltration/
• inflow be determined, and at what cost. :
2, IDENTIFY SYSTEM: prepare plot plan of entire system,
identifying component drainage systems and key manholes
within the system. '
3. IDENTIFY SCOPE'OF INFILTRATION: make flow measurement,
install ground water gauges in manholes, and'meter flows
'- •••'<'• at lift-pumps. ; '"••' •"•" ' •'•• •
4. ^RAINFALL SIMULATION: flood the storm sewer and determine
•' " rif flowerite'rs the sanitary system - use when infiltration/
"' :inflow problems are identified as rain-connected.
: 5: .DETERMINE EXTENT OF SEWER CLEANING NEEDED: a TV camera is
'not effective unless a sewer Tine is very clean.
6, MAKE'-AN ECONOMIC & FEASIBILITY STUDY to determine which
portions of the system will be cleaned and physically
: , inspected. •' -•••"•' "-'•.'••'•''
7. CLEAN SEWERS'to be inspected. ;
8. MAKE TELEVISION INSPECTION. '
9. DETERMINE EXTENT & LOCATION OF,INFLOW.
10. MAKE ECONOMIC ANALYSIS: where should rehabilitation or
replacement work be conducted.
11. RESTORE AND REPAIR SYSTEM.
12. ESTABLISH TREATMENT PLANT DESIGN CRITERIA on basis.of
reduced flows.
I
' One of the important points that must be stressed again and
again is that if we are going to look for infiltration we must
look when it logically will be present. Thus, the use of ground-
water gauges to determine whether or not the individual pipe sec-
tions are belpw the groundwater table is a necessity. Second, the
sewer lines must be clean if they are to be inspected. By clean,
I mean that a full gauge tool must be passed through the line. This
is generally more than?the normal cleaning procedure of most agen-
53
-------
cies. The cleaning procedure will be expensive and time consuming.
Therefore, careful analysis must be made as to the capability of
the agency to clean sewers and this must be attached with the plan-
ned progress of the survey. Cleaning may be a deciding factor in
determining how much of a system may be actually investigated. It
may be necessary to contract for cleaning.
Properly timed television inspection in a well-cleaned sewer
is extremely helpful in analyzing the location and amount of infil-
tration waters entering the sewer line. Data obtained will include
an indication as to locations of many sources of inflow and build-
ing sewer infiltration. The latter, building sewer infiltration, is
a hard problem to approach, inasmuch as it is very difficult to
gain access to that portion of the sewer system. A rough analysis
of a community's total sewer system may indicate as much as half
of the total sewer system is building sewers. Should the ground-
water table be high, and the building sewers under the groundwater
table, a substantial portion of the total load may come from this
portion of the sewer system. Again, such lines if they are shallow
may be an important source of infiltration and inflow during periods
of precipitation. One community which experienced severe overload-
ing and basement backups during periods of rainfall found that roof
leaders discharged adjacent to a building allowed almost a direct
connection of the water from the roof into the building sewer.
This community required that roof leaders be discharged five feet
from the foundation, and the problem was corrected. In other
communities official practice may have allowed foundations drains
to be connected to the sanitary sewer. This again leads to a tre-
mendous increase in the flow. In a like manner, sump pumps, if
allowed to discharge into the sewer system, quickly cause over-
loading. Yet another source of inflow water is from manholes.
There are many conflicting opinions, however, with regard to using
watertight covers on manholes because of the buildup of gas within
54
-------
the system. However, if the manhole is to be located'in an area .-
where storm water may enter the system, many communities have gone
to watertight covers or have added plugs to the openings to keep
storm water out.
Detection of the location of inflow is perhaps the easiest
part of the battle. The real test is to attempt to change or
correct the conditions within private property. Residents of
built-up areas without storm drains in many areas are loathe to
have sump pumps discharge onto lawn areas. In fact, in many areas
-there may not be sufficient lawn area to take the flow. In like
manner, foundation drains must have a location and a way of carrying
off the flow or there will be backup into the basement. To reduce
erosion, roof leaders.may be discharged into the sanitary sewer.
The APWA report has recommended that agencies prior to fund-
ing reconstruction of paralleling of their interceptor sewer or
relief sewer and construction or additional treatment facilities,
make a .thorough infiltration study to determine the amount of flow
which might be eliminated by correction of inflow conditions or
improvements of the sewer line to eliminate infiltration.
From a dollars and cents point.of. view, this seems appropriate.
From a standpoint of controlling pollution, we are generally further
ahead in eliminating pollution if we clean up the source rather than
if we build additional facilities and then have continuing opera-
tional cost. _
For this reason, in our Manual of Practice, we attempted to
develop an outline of an economic analysis in order that the cost of
infiltration and inflow waters might be determined and so that an
agency could determine how much it could afford to spend for the
55
-------
control of infiltration and inflow. Very few examples were found
where such an economic evaluation had been made. While many of the
tools that are available at this time are not exact, because of
lack of adequate record systems by local agencies, we must have the
economic justification of our pollution control activities.
56
-------
SECTION' III
COMBINED SEWER OVERFLOW REGULATOR FACILITIES
by
Richard H. Sullivan
Assistant Executive Director
American Public Works Association
57
-------
There is a broad cross-sectional interest in the proper
design and operation of combined sewer overflow regulators.
Consulting engineers - general design of facilities; pollution
control personnel - monitoring facilities to determine the
nature and extent of the pollutional load to receiving waters;
industrial representatives - to design and build the actual
regulator; and local governmental officials - to bridge between
these three groups and to pay for the facilities. Payment is very
important inasmuch as for this portion of the pollution control
program, federal and state aid is not generally available to
assist local government in financing the construction and recon-
struction of facilities that will lead to a reduction of this
source of the pollutional load. Lack of such aid is somewhat
unique and, undoubtedly, is directly responsible for the fact
that relatively little work has been accomplished at the local
level to implement the types of pollution control programs which
have been advocated and demonstrated by the Water Quality Office
in the field of storm and combined sewers,. Construction grant
funds from EPA have been available for only a handful of facilities,
58
-------
where essentially primary treatment will be accomplished.
It is appropriate to consider the "of.ficial policy" re-
garding combtned sewers. For many years it appeared that the
official policy of the federal government was that combined
sewers would be separated. In 1967 the APWA completed its report
on the extent of combined sewer facilities with a cost estimate
of $48 billion in 1967 dollars to separating systems involving
some 36 million persons. It appears that generally the Washington
officials are now convinced that separation alone is not the
solution, though the word has not necessarily been reached, or
been adopted by the regional offices, as we still see results of
conferences which will require separation of combined sewers on a
wholesale basis. Other federal agencies such as DOT and HUD have
also geared their programs to further the separation of combined
sewers. This becomes particularly ironic as the extent of storm-
water pollution becomes evident and in some areas we begin to talk
or require treatment facilities for stormwater. A great deal of
rethinking appears warranted at this time before actually establ-
ishing a national policy. From the work that the APWA has accom-
plished, it has been shown that storm waters are polluted whether
or not they are carried in separate or combined sewers and that
to meet receiving water quality standards, treatment or control
facilities may be necessary.
Consulting engineers and local government officials in consid-
'ering the combined sewer overflow regulator facility problems
should begin by defining their needs, particularly in measurable
terms. For instance, a general need is to either reduce or elim-
inate pollution from combined sewer overflows. The need might be
based upon a requirement to improve receiving water quality, to
improve the value of land adjacent to the overflow, to improve or
make possible operation of treatment or control facilities, or to
59
-------
improve operation of the treatment plant. The need, then, must
be defined in terms of how much or the extent of actual improve-
ment required. Means must be available to determine whether or
not the desired goal has been achieved.
If our desire is to reduce pollution, we should determine
whether or not the economical solution is to reduce flow in the
combined sewer by a system of surface storage, in-system storage
or treatment of the overflow. The type and size of the regulator
will vary considerably depending upon the nature of the treatment
or control device.
Criteria for the operation of the regulator traditionally has
been to limit flow to the interceptor. I would like for you to
consider, however, the concept of the Two Q's, control of quality
and quantity of the overflow. Regulators can be classified as
either static or dynamic. If they are static, they perform in a
determined manner, and are unresponsive to changes in control levels
in the interceptor or changes in the quality of the sewage. Dynamic
regulators, on the other hand, can be designed to be responsive to
a variety of flow conditions and flow characteristics. The reg-
ulator must be responsive to flow both in the interceptor and
collector sewer, the maximum pollutional load should be diverted
to the interceptor sewer, there should be no dry weather overflows,
there should be low maintenance cost, and a low initial cost is
desirable. Operation of the regulator must be responsive to
changing conditions. Quality of overflow may be improved by
screening, use of secondary motion, or the mode of operation.
Choice of the individual regulating device to be used will be influ-
enced by space required, availability of access, outflow conditions,
head-loss within the regulator, and exterior power requirements.
All must be evaluated and considered.
60
-------
The major findings and recommendations of the APWA study were:
.' . • ..«,'.:.i ,
Efforts should be made by local jurisdictions to consolidate
minor overflow points into fewer locations, in which the
installation and maintenance of sophisticated regulator
devices and controls will be economically and physically
justified.
"Total systems" management of sewer system regulator-overflow
facilities should be instituted wherever this procedure can be
shown to be feasible and economical. This will involve the
use of dynamic-type regulator devices and the application of
instrumentation and automatic-automation control methods
which will be expedited by a reduction in the number of over-
flow points.
Dynamic-type regulators should be used wherever possible and
feasible for "traffic control" of combined sewer flows. This
could shunt surcharges of.portions of such a system into sec-
tions of sewers which are not simultaneously so affected. This
approach could be enhanced by the monitoring of precipitation
and sewer flows through an adequate network of stations, in
communication with a central control point from whence flow
routing decisions can emanate.
The type of regulator used should be determined on the basis
of its performance and potential reduction in overflow pol-
lutional effects.
Maintenance schedules and budgetary appropriations should be
planned on the basis of the specific need's of static, dynamic
and instrumented units in service. Each type of regulator
should be given the attention it requires to achieve maximum
61
-------
performance.
Regulator facilities should be situated in accessible locations,
provided with safe and dependable access facilities, be free of
other safety hazards, have adequate space for necessary mainte-
nance work and, when possible, be accessible from locations
other than the street or highway right-of-way.
Maintenance crews should be adequately staffed and crews should
be provided with all necessary service equipment and tools for
their work and for their protection. In-service training should
be provided and preventive maintenance schedules should be estab-
lished. Records of maintenance work must be accurate and com-
plete in order to assess properly the effectiveness of regulator
operations and to allocate budget costs for each specific main-
tenance and operation procedure.
Specifications must require the use of the most servicable
corrosion-resistant and moisture and explosion-proof materials
in the fabrication and installation of regulator devices and
control facilities." The number of movable parts and appur-
tenances should be reduced as much as possible, commensurate
with efforts to provide greater sophistication of regulator
facilities.
Where possible, tide gates should be located in adequate cham-
bers. In cases where system control of regulator-overflow
networks is provided by automatic-automated means, the prox-
imity of tide gates with regulator chambers will facilitate
the tie-in of backwater control with overflow control. State
and provincial water pollution control agencies should incr-
ease their regulatory control of this source of pollution and
provide standard requirements and the engineering personnel
62
-------
necessary for enforcing the control of overflows from combined
sewer systems. Further, such agencies must recognize the fact
that existing combined sewer systems must be upgraded if pol-
lution levels are to be reduced.
Efforts should be made to design regulators to minimize .clog-
ging and consequent pollutional overflows. Where clogging is
inevitable, maintenance schedules should be adapted to correct
this condition as expeditiously as possible.
As indicated, interest by various states in regulators and
overflow pollutional problems vary considerably. Few states have a
staff knowledgeable enough to give much guidance to local officials
or to even review plans. Many states appear to want to believe
th'at if they do not get too concerned about the problem, it will
go away. Many seem to be taking the textbook advice that combined
sewers are a thing of the past. Inasmuch as over 30 million people
are directly served by combined sewers with some 18,000 overflow
points, I doubt that this represents much more than wishful thinking.
At the close of the research project the APWA developed a
Manual of Practice. There is a great deal of heretofore unpublished
work in it, which represents good practice in the field. Certainly
you and the public agency which you serve should review the Manual
for information regarding requirements for the design, operation
and maintenance of facilities, as well as a description of'some
of the newer types of regulating devices. Many of you will have a
very difficult time convincing an agency that they should pay more
than the $2,000 to $4,000 cost of a static regulator device. How-
ever, if pollution is to be reduced, time and money spent on the
design and construction of adequate regulator facilities will do
much to enhance the local program.
6.3
-------
-------
SECTION IV
PRESSURE SEWERS
by
Italo 6. Carcich
Senior Sanitary Engineer
New York State Dept. of Environmental Conservation
Environmental Quality Research Unit
Albany, New'York. .
65
-------
Introduction
The pressure sewer concept has been around for a number of years.
When referring to pressure sewers, we are dealing with a wastewater collection
system that utilizes a newly developed Grinder Pump Unit and small diameter
plastic or metallic piping systems. It is by no means intended to replace
gravity sewers but only to supplement the wastewater collection system.
With financial assistance from both the State and Federal governments,
a 13 month study was completed in Albany, New York for the purpose of evaluating
the functional specifications of the GP Units and to gain first hand operating
experience on the mechanical performance, use pattern, operating cost, main-
tenance requirements, etc. on these units. The final report is available from
the U. S. Government Printing Office^ '. A full description of the installations, .
the monitoring equipment, the piping system, etc. was published previously^2'.
Therefore, it is not necessary to go into a detailed description of
the installation, with the exception of stating that the pressure sewer system
was very simple in design. The wastewater was diverted to the Grinder Pump
Unit's tank from which point it was discharged by means of a 1-4" plastic pipe
pressure lateral to an outside 1-j*' to 3" plastic pressure main. The pressure
main at a 4 foot depth received the macerated wastewater from all 12 houses
and simply discharged it into a gravity system within the city of Albany (Figure l).
Grinder-Pump Units
The GP Unit consists of the following mechanical components (Figure 2):
(a) Grinder, placed in an inverted position and operating at 1725 ppm with the
capability of handling foreign objects without jamming; (b) Pump, positive dis-
placement, progressing cavity type with an almost vertical H-Q curve and proven
66
-------
-------
FIGURE 2
CROSS SECTION OF GP WITH LOCATION OF
LEVEL AND OVERFLOW RECORDING FLOATS
STEEL TANK-
BELL SHAPE
PRESSURE SENSING
TUBE
GRINDER MECHANISM
DISCHARGE PIPE
ISULATED HOOD
3.8* O...IA.T.*
OVERFLOW RECORDING
FLOAT .:
68
-------
solids handling ability; (c) Motor, 1.0- horsepower, operating at 1725 RPM,
capacitor start, 'high : torque, '-squirrel cage induction motor with a 'built-in
thermal overload protector; (d) Check Valve^ swing check type with passageways
smooth .and free from roughness, and obstructions, and a unique flexible hinge
c-f small section without mechanical pins, rivets, .screws, .etc; (e) Controls,
an inverted diving bell system to -turn the motor on and off.
(3)
A li inch discharge pipe was selected as the optimum size capable
of not only handling the macerated wastewater without clogging but also
minimizing the frictional head losses (Figure 3).
Results • . • . , , /. .. •; . . , . . ' ; • . • ,
Thirty nine out of the 44 recorded malfunctions were contributed by
the Prototype GP Units. Nine of these Prototype units were replaced by
Modified GP Units (Figure 4) after only 6 months because of the large number
of malfunctions. The newer units performed .satisfactorily for the remaining of
the project. Loss of prime by pump and grease clogging of the 1" opening
within the bell-shaped pressure sensing tube was the major cause of the mal-
functions experienced by the Prototype Units. Corrective modifications were
incorporated in the manufacturing of the modified GP Units with considerable im-
provements in the daily operation.
One of the primary interests of this project was to extensively test
the reliability of the mechanical components in an actual field installation.
Pre-installation testing and post-installation Resting (Table l) was performed
in order to determine marked deterioration if any, in. the physical structure arid
performance of the GP Unit's components.
In addition to the 6282 'operations ^occurred during the so-called
69
-------
V-V = 4.5 f ps
to 3
O °
o
cc
u_
V=2.5fps
- ^=1.8
GP Units NORMAL OPERATING RANGE
J.
10 II 12 13 14 15
DISCHARGE- GALLONS PER MINUTE
FRICTION LOSS vs DISCHARGE
FOR THREE SIZES OF
POLYETHYENE PIPE
FIGURE 3
70
-------
4" INLET
I . H-P MOTOR
PRESSURE
SENSING TUBE
FIBERGLASS TANK
24.00" DIA.
CROSS-SECTIONAL VIEW
OF MODIFIED GP UNIT
FIGURE 4
71
-------
to 3
?>o
I
I
S
8 S
CO CO CO
J3SS
CN CO I
O O t-
882
atio
ter
ing
CM •-< *
co m (
co m CN
t in m
5*~
M *H
•frs-8
5 fe :
O 0) C JS •
2 »H in w |
SS'S
.
lllj
1-1 -PO. S
« +>
£ .8
W rH +3 ,
SI,
_^ « in
•H
Ǥis
j= >.
2 «H «H O ft
•f-fm
2"1 233
in o
Q cy in
o o *»i
CO
-------
de-bugging period, a total of 73,458 GP Units operations were recorded during
the remainder of the demonstration project (Table 2).
Even though the operating cycle varied greatly for the prototype
units, the modified units operated on a cycle between 57 and 74 seconds (Figure 5),
with the average operating time of 11^5 minutes -to . 27.5; minutes per day. Further-
more, based on the occupancy rate of 75 persons for the 12 town houses, a value
of 2.6 operating cycles per capita per day was calculated for-this particular
single family residential development. .
The documentation of the .operating cost was of prime interest, since
it was essential to verify the theoretical cost value of $2.12/year for a
(34) ,
family of 5V ' . Two watt-hour meters were installed to register only the
total power consumption of two individual GP Units. Based on the monthly oper-
ating time, proportional monthly power consumption values of 10.2 and 5.3 KW
were calculated. Applying an average incremental power consumption rate of
2.3$ per kilowatt hour (KWH), the monthly operational cost for Unit No. 1
amounted to $0.24 and $0.12 for Unit No. 2 (Figure 5), which is equivalent to
$1.18 for a family,of 3,up to $3.50 for a family of 9.
The GP Unit's usage varied greatly from day to day for any given unit.
An even greater variation was documented when comparing weekday versus weekend
usage. This is graphically illustrated in Figure 6 for two given units. The
total weekend daily usage exceeded ::,the weekday total .daily usage by 50-60
operations (an;increase of 35% over the weekday total).
As an indication of the improved performance record of Modified Units
versus the Prototype Units,; a value, known as;the, "downrtime", was computed for
each of the GP Units. The "down-time" value is based on the amount of time a
unit was non-operational over the total amount of time of possible operation.
73
-------
M
1
M
H
5
os
Ul
(X
o
p-<
o
Oi
w
5
z
3
t-.
g
CM
1
H
!-•
•P
C
(0
O flJ
(-• 0
CM
rH
rH
rH
O
rH
ON
CO
t^
NO
in
"5j"
CO
CM
rH
O
SI
•P r-!
•"•' ,£
rH
NO
•*
1
CO
CM
CM
CO
CO
CM
in
in
in
CM
CO
CO
i
CO
in
CO
CO
CO
CO
o
o
in
CO
rH
CO
t^.
8
I — 1
O
^
-p
o
o
•*
'
CM
^-
CM
8
CO
in
r-
m
r-
co
CM
p^.
CO
1
ON
vO
CO
CM
CO
§
in
8
CM
CO
m
C--
o
t
o
'**',
NO
vO
•vf
rH
rH
f-
rH
O
CM
vO
o
CO
in
o
CO
f-
o
CO
CO
CM
O
CO
CO
CO
in
CM
CO
CO
in
in
ON
CM
CO
t-
m
o
r-
o
fi
ON
s
in
t-
^
Ov
rH
O
m
CM
vO
NO
00
CO
CO
ON
m
CO
^
rH
CM
rH
00
CO
NO
iO
m
CM
CM
CO
Ov
rH
t-
c!
vO
in '
*
rH
0
vO
Ov
o
rH
CO
in
CM
?
CM
o
CM
CO
rH
rH
CO
CM
CO
CO
CO
CO
rH
00
CO
r-
CM
o
vO
rH -
r--
111
m
CM
rH
vO
CO
CO
Ov
,_,.
CM
rH
CM
"in
vO
in
•g
r*-
O«l
CO
CM
O
ON
CM
ON!
CM
ON
Q\
in
CM
r-
ON
CM
CO
CM
m
r-H
r-
rH
JO
vO
vO
CO
*"
CO
CO
^J.
h-
CM
rH
in
CM
CO
rH
*3"
CO
CO
CO
• fNJ
00
CO
m
NO
CO
^.
Q\
CO
s
in
SN
CM
ON
CO
rH
rH
in
m
VO
00
in
CO
vO
vO
vO
m
CM
m
t-
^*
Ov
Ov
ON
O
CM
rH
rH
CM
O
"*
CM
in
CM
rH
in
0
ON
NO
rH
£
vO
Ov
ON
in
8
vO
^.
CO
CM
Ov
r-
r-
in
rH
r-
m
o
•vt
i
rH
O
f
CO
rH
CO
CO
ON
|
•5}"
in
rH
rH
rH
r-
Ql
r.
D
CO
CM
vO
vO
vO
O
CO
CO
vO
O
CO
in
00
CM
CO
ON
m
in
i
•in
ON
CM
CO
00
CM
ON
CO
in
CO
CO
CO
ON
Ov
rH
f-
i — (
n
ON
CO
NO
O
vO
CO
vO
rH
f-
VO
in
•st-
oo
p—
CO
•vt
1
ON
CM
CO
ON
ON
in
vO
rH
f-
1
vO
CO
00
rH
f-
cn
CO
o
CO
NO
00
in
vO
NO
CM
CM
CM
rH
in
rH
in
CO
in
CO
f-
T^-
co
in
;
CO
CO
CM
m
t-
o
00
NO
NO
CO
00
NO
in
t"-
-p
O-
01
to
•o
r-
vO
8
vO
I
8
in
0
CO
NO
rH
in
^>
rH
00
1
00
CM
rH
VO
CNJ
CO
CO
CM
m
o
m
ON
rH
r-
-P
o
o
1
rH
O
vO
rH"
1
CM
CM
rH
00
vO
i — t
00
t-
Ov
rH
1
00
CM
rH
CO
CO
rH
NO
vO
rH
O
0
i — 1
Ov
o
CM
F^
*
>
rH
•P
O
H*
00
in
CO
c~-
f-
0
ON
CM
vO
O
ON
vO
vO
CO
m
00
rH
1 —
vQ
00
00
in
m
CO
in
rH
00
3
Ov
vO
CO
ON
CO
00
"?
s
H,
s
•P
•H
c
l~~i
o>
ex
rH
•P
o
H
ro
f-t
-------
•
O <
im ^^
UJ UJ
0. 0.
o
oo
• •
o
^
o
CO
K
*©
D?
m
m
m
O
0
o
m
K
P^
K
o
m
O
Z
<
0^
* l'"
UL uj
°d
55
o
z
UJ
CO
CM
CO
•
in
£
•
CM
d
CM
K
o
CO
00
d
in
^
^
^
m*
CM
K
CM
o z
Si
-5
UJ <
Q. Q
O
_J UJ
< 0,
0 !g
t— St
. p
CM
d
o
0
CM
o*
lf|
CM
d
CM
d
^~
d
. CM
5
CM
O
tN
d
CM
d
CM
d
CM
d
z^
OS:
QL X
II
^ S^
8a
LU .
15
c
o
O
i_
a
00
•— o
E o
a
U CO
•o a o
£ >• *t
o "o •"£
^ »- _
S a. Z)
_D
O
o
o
-------
a \< .. "I •;.
CD Z
O
h-
CD
tu
oe
SNOIlVHBdO
M38NHN 1VJ.O1.
76
-------
The Prototype GP Units produced a "down-time" of 2.69% for the first six
months in comparison to; only 0.27% for the Modified Units over the last ?•£
month period.
Discussion.
The pressure sewer system pipe sizing was based on the ASCE minimum
(5,6)
scouring velocity criteria of V =£•£ and on certain engineering assumptions
i ••••;' .2
regarding the estimated wastewateir flows from the 12 GP Units.
It must be understood that the flows in the different portions of the
pressure main were based strictly on an engineering estimate. There was no
data available on the frequency of GP operations for a multiple units system.
It wasi'possible to predict the peak usage hours of the GP Units, but since the
operating cycle per GP Unit is very small,: 57 sees, to 74 sees., it was almost
impossible to predict the number of units working simultaneously during this
peak period. It was, therefore, assumed that a maximum flow of 90 gpm would
flush regularly that portion of the pressure main serving all 12 GP Units. It
must be understood that the hydraulic characteristics of the pressure sewer
system is dependent greatly on the varying wastewater flows within that system.
Information on simultaneous occurrences was an essential phase of
the project. This type of data is critical for the design of future pressure
sewer systems. The maximum anticipated flows will dictate the size of pipe
within the pressure system. At the same time, the hydraulic gradient will
reach its peak slope. The engineer, therefore, must design a system optimizing
the sizes and scouring velocities and be certain that the upper recommended
working pressure of the GP Unit is not exceeded.
v •' - ,' . • ' ' '
During the last ten (10) months of5' the demonstration project, during
which time the 12 channel event recorder was in operation, a total of 58,823
77
-------
operations were recorded, which represent approximately 191 operations per day.
Therefore, in order to obtain a picture of the minimum and maximum flows within
the pressure system, the above mentioned data indicated that (a) on the average,
2 GP units ran simultaneously 20 times per day • (b) 3 GP units operated simul-
taneously slightly more than once per day, and (c) 4 GP units ran simultaneously
on the average of once every 14 days.
Also, by using all the automatically recorded data, total wastewater
flows were calculated, which ranged between 95 and 100 percent of the actual
water consumption (Figure 7).
The close relationship between the water and calculated wastewater
flow is a highly reliable indicator of the corresponding wastewater discharges.
Also, winter water flow records can be used to.estimate accurately expected
wastewater flows.
Pressure gages were installed in each basement so that the maximum
and minimum pressures occurring during any fifteen minute period might be
recorded. These pressure readings were indicative of the varying hydraulic
gradient line for each of the twelve GP units (Figure 8). The computerized
data indicate that pressures in excess of 30 psi were reached by a few GP
Units.
Once the demonstration phase of the project was completed, portions
of the pressure main and the 1^- in. pressure laterals were carefully excavated
and removed. Grease accumulation within most sections was evident. Reductions
of up to 40$ occurred in the pressure main.
The system was simply overdesigned. Where flows were expected to
reach 90 gpm regularly, flows of only 45 gpm were recorded (Figure 9). Therefore,
instead of a 3" pressure main, a 2" main would have been sufficient for the 12
town houses.
78 . .
-------
©
-J
yj
UJ
Ul
r^
iy
(SN011V9)
-------
CVJ
C>
00
CO
U (9
(9
O
<
(O
(O
UJ
o:
Q.
(9
<
o
(isd) soNiavaM awnssaud aaovs
80
-------
FIGURE 9 ;
ASSUMED AND ACTUAL FLOWS FOR THE PRESSURE MAIN
" : 1, V _ '" PROP. LINE
_ . -—
TRLRi
726 728 730 732
SAMPLING a
^^^ SECTION NUMBERS
'"72 II 10 9 B'l ~ 7 65 4 3 2 1 _
' II 1 1
12
1 1
IO
' 9. .;,
Ill 1
'•-••'•
7
,.6;
5
4
1
3
2
I
CONTROL BOX
734 736 738 740 742
STREET ft PROPERTY LINE
744 746 748
M.H.i
SECTION
NUMBER
1
2
3
4
5
6
7
8
9
!0
II
12
PVC-DWV
PIPE SIZE
1.25"
2.0"
,
1
3.0"
•
LENGTH OF
SECTION
(FT)'
19.7
20.0
59.2
19.2
19.5
19.6
1.9
58.5
1 7.4
" 19.5
2.7
81.0
ASSUMED
MAX.
FLOW
(GPM)
15
30
45
60
1
75
90
MAX.
FLOW
RECORDED
(GPM)
15
30
45
'
60
f
ASSUMED
MIN.FLOW
IN 24hrs
(GPM)
15
- |
\
30
45
60
f
DAILY FLOWS (GPM)
MAXIMUM
15
30
*
45
AVERAGE
15
30
MINIMUM
15
'
.
81
-------
TABLE g
SUMMARY OF COMPOSITE SAMPLE ANALYTICAL RESULTS
Parameter
5 Day Biochemical Oxygen Demand
Chemical Oxygen Damand
Soluble Total Organic Carbon
Total Solids
Total Volatile Solids
Total Fixed Solids
Total Suspended Solids
Volatile Suspended Solids
Fixed Suspended Solids
Total Dissolved Solids
Volatile Dissolved Solids
Fixed Dissolved Solids
Organic Nitrogen**
Ammonia Nitrogen**
Nitrate Nitrogen**
Total Phosphate***
Particulate Phosphate***
Filterable Phosphate***
Total Ortho Phosphate***
Methylene Blue-Active Substances
Grease
Settleable Matter & Hr.
Settleable Matter 1 hr. '
Chlorides
Hardness
Alkalinity
PH
Number
of
Samples
• 57
56
6
55
56
56
56
56
56
55
55
55
53
54
38
63
50
51
32
39
9
56
56
38
55
9
54
.. Mean*
330
855
140
681
476
205
310
274
36
372
201
171
29
51
0.1
15.9
2.8
13.1
8.7
12.4
81
14.5
15.0
52
65
198
7.8
Standard
Deviation
53
158
49
87
84
63
77
84
48
90
62
58
12
9
-
6.3
0.9
6.5
3.9
4.5
12.3
6.1
6.2 '
4
7.4
8.1
.3
Minimum
Value
216
570
21
526
336
57
130
78
0
195
22
27
7
34
—
7.2
0.4
5.2
1.3
4
31
4
4.5
41
46
185
7.1
Maximum
Value
504
1450
225
928
706
355
468
440
268
637
372 ;
353
76
68
49.3
4.2
47.9
17.9
24
140
37
38
61
90
209
8.7
* All values expressed as mg/1 except pH
** As nitrogen
*** As phosphorus
**** As linear alkylate sulfonate
82
-------
There are no existing standards for velocities dealing with the
grease accumulation problem, even though velocities in the range of 2 fps to
8 fps have been used by some in designing wastewater pressure conduits. However,
for a pressurized sewer system utilizing GP Units, a velocity range of 2 fps to
5 fps is hydraulically and economically preferable. = .
Extensive chemical analysis were performed (Table 3). The concentra-
tion of various pollutants in a pressure sewer system was found to be approxi-
mately 100% greater than those found -in conventional systems. On a gm/capita/day
basis the pressure sewer waste contained approximately 50% less contaminants
than reported for conventional domestic sewage. Settleability tests show no
significant differences when compared with conventional wastewater.
Therefore, the difference in the strength must be taken 'into account
in designing treatment facilities for a pressure system.
Conclusions
The pressure sewer system, which included the .usage of PVC Schedule
40 pipes and PVC-DWV fittings, functioned well for the duration of the demon-
stration project. Careful considerations must be given to the material used
in backfilling pressure main trenches. A good engineering practice is to en-
case the plastic pipe in sand.
As for the GP Units, the 'functional specifications have proven to
be appropriate. Even though the Prototype Unit exhibited low mechanical
reliability, the Modified GP Unit operated to its expectations. Pesign mpdi-
ficationS virtually eliminated all major malfunctions; that is, the 1" opening
of the pressure sensing tube was increased to 3" and the pump was relocated so
as to be positively primed.
83
-------
The service record coupled with the "down-time" performance of the
Modified Units was impressive, a 0.27% "down-time" value versus a 2.69%
"down-time" value for the Prototype GP Units.
Both the pump size and tank volume were more than adequate to handle
peak wastewater flows, so that no further design modifications are necessary
in this area.
Therefore, in order to summarize the operational performance of the
GP Units, a brief review of previously presented facts has been tabulated;
(l) Total Number of GP Operations for the duration of the project -
73,740 operations
(2) Average Operations per capita per day - 2.6
(3) Average Length of operating cycle - 57-74 sec.
(4) Electrical power consumption cost - 34
-------
ct*
T
a
g
>. '
%~
^
4-*
S &
8.3
t- 0
O ^"*
pp* £
g §
W 10 .
CO -P
< 10
s c
o
f— » Q
CO 0)
-p
til >•
S CO
w
CO H
W 3
^rf tD
3 CO
CO
CO fl>
CL. to
y)
P*
§ <
CO =
@ o
fe rH
8~
E 0.
•P -H tU
U) O r- )
o> c\.
h o o
a. o E
U) >
S 01 10
O C T3
0) T3 Q
>• 10 10
CO O O
-"^
(U c
ECO
0 O
•p
«-H fO
to Pi
•g-s
•H 0)
> 0^-i
•H C\
T5 O O
c a E
i— i
E
ID
4->
to >-
>- 01 to
W 5\
>i"O O-
-p to to
•H O O
> i-J^
to E
& § a
•H
r-l -P
(0 10
O -P
•H C
-P 0)
C 0 ,-H
!> 0^5
C O £
O
0
4
4-
G
E
0
d
D
lO O^ *^J"
CO ^^ CO tO CO ^" 1 ^0 lO C71
r-H
__
CO 00 vO ^ 00 00
vO
CO tO rH 00 VO
r-t (N 1 ' — 1 ' — 1 * ' * ^ ^
1 1 CO 1 1 •"*
^J" CM ' — i O ^r ' r"
^ 00 t — i ^O .^ - CO *-
Csl . q tn "O T3
OS ^H r-l l— IrH
c c
•HO ci)-PO>Ol
SO) rHT3>10OO
,(n'rl Cr-lS (H (H
_<~] ^ '^ .^ fl) Q 1 ' [ >
t_> O "H tO IX to O(-ii— if— iHcoo<;
'-^
co m a.
O^ r-l CO 10
r-i r-l
to
Ol
CN
CO
-. -^J ^_J
r-i in
ft-
•a
1 a
(0
o
^ 00>
-i. 1 r-H
• • CO
CM
5 -4 m
"* CM ^
' i ro
D r-l CO
•a
p.'
to
*^^
,—1
to
Ol
r-l
r-H r-l CO O
"0 1-1 CM ^ I
U)
cS
0) O
01 J3'
0 O.
•H JZ 0)
•H 0)
•— 1 r-l fH (0
to to o to 5
-P -P r-l 0) O
H t-H O O (i4
in
to
•O
CD
d)
(H
CX
85
-------
-------
SECTION V
APPLICATION OF MICROSTRAINING TO
COMBINED SEWER OVERFLOW
George E. Glover, P.E.
Research Engineer
Cochrane Division-Crane Co.
89
-------
Combined sewer overflow is a mixture of stormwater and sanitary flow.
The special problems of dealing with this flow are due almost exclusively
to the stormwater component. Thus, these remarks should apply equally well
to overflow of separate storm sewers.
The two components - stormwater and sanitary waste - are somewhat
similar in composition. Both contain suspended solids, BOD, and coliform
concentrations equal to many times the usual secondary effluent standards.
On an annual basis our eleven acre drainage area produces some 9,000,000
gallons of sanitary flow and about 3,000,000 gallons of storm runoff.
The flow rate of stormwater runoff, however, is very high and widely
variable. At our site, we have monitored several storms a year where the runoff
rate is over 400 times the mean dry weather sanitary flow. It is the flow rate
aspect of combined (and separate) sewer overflow that requires a totally
different approach when treatment is considered.
Only recently have we become aware of the magnitude of the possible
pollutional load from stormwater runoff and have considered treating it. It is
not surprising that there is a considerable difference in opinion as to what a
stormwater treatment facility should be able to do. The two basic dimensions
of a combined sewer (or separate storm sewer) overflow treatment facility are:
(a) The instantaneous flow rate it can handle, and
(b) the amount of each type of pollutant it can remove.
90
-------
In our studies we have used a flow rate of 2.0 cfs/acre (1.34 mgd/acre)
as the required instantaneous capacity of the treatment facility. This runoff
rate would require (at a runoff coefficient of 0.4) 4.5 inches per hour rain
intensity. At our site we have this intensity sustained for about 15 minutes
every 10 years. Analyses of very large drainage areas such as the Boston and
Chicago stormwater tunnels where rainfall does not occur over the entire area
simultaneously, and where there is tremendous surge volume within the sewer
(tunnel), have led to the adoption of a flow rate of.0.2 cfs/acre (0.13 mgd/acre)
based on the area of the entire basin. Less understandable is the adoption of
low (0.2 -0.3 cfs/acre) instantaneous design rate for the treatment of combined
sewer overflow from small drainage areas of 100 acres or so. Additional
experience will permit the selection of realistic design rates for each situation.
It has been suggested that flow equalization basins be included above
ground as part of the overflow treatment facility to reduce the peak instantaneous
flow rate. Above ground, flow rate equalization basins by themselves may be
an attractive scheme of treating overflows, providing space at low cost is
available. In this scheme, the peak overflow rate is reduced to a rate where
the existing interceptor sewer and sewage plant can handle it as an alternative
to an on-site combined sewer treatment facility. Although the annual stormwater
volume is some 35% of the sanitary volume, only some 15% additional flow rate
capacity would be required.
Flow equalization is most attractive where the subsequent treatment
techniques are very expensive on a dollar/cfs peak capacity basis. Flow
91
-------
equalization is essential where the subsequent treatment techniques cannot
accept sudden starts and stops or rapid changes in flow rate of several
hundred times the dry weather flow variation.
The extent of treatment to be required on combined sewer overflow is
at present not standardized. It is not certain what form regulations will take.
As will be seen later the familiar "percentage removal" type regulation would
be most inappropriate for this problem. Much more work and study must be
completed before it can be decided whether it is necessary or consistent with
the cost to design overflow treatment facilities for a 25, year return storm or a
5 year return storm.
With current practice, the combined sewer overflow regulator is adjusted
! ' •'
to overflow when the rate exceeds perhaps 3-5 times the mean dry weather flow.
Thus, the composition of the combined sewer overflow is 1 part sewage to at
least 1-1/2 parts of storm runoff. Frequently the composition is over 100 parts
of storm runoff to 1 part of sanitary flow. In any event, when significant over-
*
flows occur, the composition of the overflow water is determined almost
exclusively by the composition of the storm runoff.
The wide range of contaminant levels in the combined sewer overflow
reflect the breadth of the range in the storm runoff.
The contaminant level in the combined sewer overflow observed in our
site is shown in Table 1.
92
-------
Table 1
Contaminant
Suspended solids mg/1
BOD5 mg/1
Total coliform
cells/100 ml
.Minimum
. 15
8
1,000
Mean
100
800
Maximum
' 700
3,000
1,000,000 3,000,000
Previously we had found (during the fall and winter storms) that, in
general, the contaminant concentrations were higher on the bigger-storms
particularly in the case of the suspended solids. Recently, however, (during
spring and summer storms) we found little relation between storm intensity and
contaminant levels. The BOD and coliform content of overflow do not seem to
have any relation to storm intensity but do seem to have an annual variation.
Each drainage area has no doubt a unique combination of features which will
influence the character of the stormwater overflows. Our experience, however,
has been paralleled by the reported observations of others.' They find that
sustained higher contaminant concentration levels are as likely if not more
likely to occur in large overflows from the bigger storms as from the smaller
overflows from less intense storms.
Thus, the treatment design criteria and the regulations must, for
the present, assume that maximum overflow contamination concentration will
93
-------
exist at design peak flow rate. More work is needed on this aspect.
To attack a given combined sewer overflow situation, the first step is to predict
the peak rate-duration and frequency of the actual overflows. With these
predictions at hand a decision to treat all storms of less than a certain return
frequency must be made more or'less arbitrarily. One method of arranging the
storm flow data is that used by Dow (2). See Figure 2 from that report. Note
that treating about one-third of the peak,flow observed over an 8 year study
would treat some 98% the total annual flow.
The benefit of flow equalization can be evaluated for the "storms to be
treated. That is, the relation between equalization basin volume and the reduced
peak rate can be ascertained. This work might be extended to, say, 60 minutes,
which will be the residence time of some of the actual treatment techniques. We
will return to this flow rate consideration after we look at the degree of treatment
needed.
There is paucity of information regarding the impact of combined sewer
overflow contaminants on the receiving stream. It seems that the pounds of
suspended solids discharged per year would be an important criterion.
It is not known how much greater impact these solids would have when
they are discharged in slugs of approximately 40-60 hours annual duration.
If it is found that the instantaneous rate of solids discharge is significant, the "
regulations may be phrased in terms of maximum pounds per hour. This ;is a,
very complex problem and the methods of considering it have not been developed.
94
-------
UJ
H
<
u.
UJ
>-
I-
o
UJ
K\\\1
K:
•NIWV-'SSl - 1NVNIWVJ.NOO -
UJ
or
j-
cc
UJ
UJ
h-
I
z
o
<
IT
r>
o
UJ
S
16
-IS
s<8
u.
UJ
tr
UJ
X u-
oo
95
-------
09
(O
0>
i
O
0>
CO
LU
<
on
UJ :
a uj
o §
a:
o
u_
o
o
h-
CQ
Od
I-
5
96
-------
The potential pollutant load of untreated combined sewer overflow during
a big storm is: (Overflow Rate) x (Pollutant Concentration; e.g. , S.S.).
The potential load can be reduced by treatment to a lower level, depending
upon the design of the treatment facility as follows: •
(Overflow Rate-Peak Capacity) x (Pollutant Concentration) plus
(Peak Capacity) x (Pollutant Leakage).
Figure 1 is a preliminary attempt to illustrate this relationship in a
stylized manner. The bars represent overflows in increments of magnitude. The
height of the bar represents the magnitude of the flow (the left of the pair) and
of the instantaneous contaminant flow; e.g., pounds of suspended solids per
second. The width of the bar represents the duration of flow of the indicated
magnitude in minutes per year. The area of the bars then represent overflow
volume per year at indicated rate (left of pair) and the pounds of contaminant
per year/.; The shaded area at the bottom of the solids bar represents the solids
leaking through treatment facility and entering the stream. An arbitrarily selected
design peak flow rate for a treatment facility is shown. The shaded area on the
solids bars representing the biggest storms shows the additional solids,entering
the stream by direct bypass of the facility.
The amount of the annual contaminant load to the river of the design
parameters - peak flow capability of the facility and the leakage through the
facility can be seen. Also, the instantaneous rate of contaminant discharge can
be seen.
data.
Figure 2 shows another way to consider the overflow rate-annual duration
97
-------
In the previous application section, I have attempted to show the
importance of Peak Flow Rate Capability of a combined sewer overflow treatment
technique (s). Also I tried to show the importance of Contaminant Level Removal
Capability of treatment techniques at design (peak) rate and below design rates.
The announced subject of this paper is a description of the capability
of the Microstraining technique in this service.
of
Figure 3 is an isometric drawing^a micros trainer. A microstrainer is a
A
rotating drum fitted with fine screen. For stormwater the screen used is what
we call Mark 0, a stainless steel Dutch twill screen with 600 x 125 wires per
inch yielding about 23 micron (1-1/2 millions of an inch) apertures.
The stormwater enters the open end of the drum and passes through the
screen into the outlet chamber and then to waste. The suspended solids are
retained by the screen. As the drum rotates, the screen with a mat of retained
solids on the inside is brought up and under a row of backwash jets which wash
the solids off into a hopper and thence to disposal. The backwash water
requirement is about 1-1/2 gpm per foot of drum length which is a fraction of
a per cent of the thruput capability. The solids-rich backwash water stream
is small - less than the DWMF - and can easily be sent via the interceptor
to the sewage plant, for smaller CSO facilities, or disposed of locally. The
"•"lyj'
backwash water source can be re pumped microstrained CSO or preferably city
water on small unattended satellite facilities.
The flow of water through the screen is motivated by the difference in
level inside the drum over the level outside the drum. In conventional applica-
tions of Microstraining this differential is about 6 inches. At this differential
98
-------
.Figure'3
Isometric Drawing of a~ Microstrainer
99
-------
the Mark 0 screen will pass only about 6-8 gpm/ft2 of gross submerged screen
area. It might be noted here that the flow capability is not based upon the gross
area of the drum but rather upon the open submerged area. That is, that area of
screen unimpeded by hold-down straps which lie below the liquid level inside
the drum. There is considerable difference in the per cent submergence attained ,
and the per cent unimpeded area in currently available micros trainers and the
percentages vary a little from size to size. In the Current Crane design for a
10' dia x 10' long drum, the per cent submergence is 83% and the per cent
unimpeded area is 94%. The'Glenfield-Crane (older) design we are using has
only 83% unimpeded area and was adapted to achieve 83% submergence. Some
competitive designs have lower percentage submergence and unimpeded area.
For stormwater service we use much higher differentials, up to 24", and
have achieved flow rates of up to 45 gpm/ft2 of gross submerged area (i.e. ,
54 gpm/ft^ of unimpeded submerged area) with very high removals.
The following remarks will be based upon 35 gpm/ft2 of gross submerged
area (42 gpm/ft2 of unimpeded submerged area). Also, these remarks will be
based primarily on the use of a microstrainer as a satellite station for treatment
of CSO; i.e., located at the point of overflow so that no additional sewerage
is required.
Perhaps the best way to describe a microstrainer CSO facility is by an
example.
A present-day Crane 10 x 10 has 314 gross sq ft of screen area of which
245 sq ft is unimpeded and submergible. Such a machine can treat some
100
-------
10,500 gpm or 23 cu ft/sec of any of the combined sewer overflows we have
seen in 16 months of study. Our example will be a facility with two such
machines in parallel. (As previously mentioned, the 46 cfs (30 mgd) flow
capability of these two machines would be required by a drainage area of from
24 to 240 acres depending on many factors unrelated to the microstrainer.)
Any CSO treatment facility will require a coarse bar screen. The space
for and the cost of a travelling bar screen have been included in this example
«L
facility. Almost certainly any CSO'treatment facility will be sizeAto treat
something less than the peak storm that will occur in the life of the equipment.
Thus/ a bypass arrangement is. required to divert the flow in excess, of the peak • •
capacity of the treatment equipment without interfering with the capability of
the equipment to treat its peak flow. This consideration may be less important
with Microstraining than with other techniques . A microstrainer will flood;
i.e., untreated water will overflow the washwater hopper at inlet levels 3" or
so above the design level at peak design flow rate. The microstrainer cannot,however,
dump previously removed solids into the effluent under excess flow conditions.
The space for and the cost of a bypass weir and channel suitable to divert
excess flow equal to the design flow have been included in this facility.
That is, this facility can accept 92 cfs, treat 46 cfs without hinderance, and
' bypass the reaminder to the receiving stream, or rather to the disinfection
chamber, and then to the stream. , - ,
;The bar :screen-microstrainer facility with flumes and chambers for bar
screening of 92 cfs, Microstraining of 46 cfs, and bypass of 46 cfs will occupy
a ground area of 30 x 40 ft x 10 ft deep. The facility area of 1200 sq ft of ground
101
-------
area is 1/35 acre cr about 1/1.000 to 1/10,000 of the drainage basin. The liquid
volume of the facility is about 9,200 ft3, or 200 sec residence, at peak flow.
The head loss through the facility is about 3 ft during peak flow. While 3 ft -,
is the minimum head required during a storm, ideally there should be 10 ft of ~
head available so that the facility can be drained by gravity after the storm.
Otherwise, a small (3 hp) sump pump will be required. '
The chamber will be comprised of about 2,500 sq ft of concrete walls
and 1,200 sq ft of floor, and to put it below ground will require about 600 yards
of excavation. ...
The microstrainer section should be housed and kept above freezing.
The recommended building then would be about 16' x 40' x 18' high. The
individual microstrainer units weigh about 13,000 pounds and an I beam
craneway should be provided for installation and maintenance. An insulated
Butler Building of this size is included in the cost data.
To keep the microscreen in condition to operate when needed it must not
be allowed to become dry while soiled. The recommended procedure for
combined sewer overflow service then is following a storm to drain the chamber,
continue the backwash of the slowly rotating drum using city water as washwater
for several hours and then stop the drum and the backwash water/
Also, for sustained dry periods the drum can be rotated slowly for short
periods at intervals under backwash jets and the UV lights. The program controls
for carrying out this maintenance operation automatically are included in the
cost data. . , .
102
-------
'- The'• cost of a complete facility installed, le'ss land and engineering, was
estimated to be $195,000 in 1969 dollars* This investment represents an annual
capital "charge of about $19 ,500/year ,to be "applied to the facility. •• This annual
capital charge is^,,-by far, the major.cost for Microstraining (or^ other techniques) -
for combined sewer overflow. This cost applied to the drainage area represents '
about $80 to $800 per acre at peak design rating of .0.2 and 2.0 cfs/acre respectively.
The effect of scale on the cost of a facility can be seen in. Figure 4.
The utilities required for the two machine facility include about 50 gpm of ;
city water. The electrical power demand is for two 5 hp drum drive motors,, a
3 hp sump pump, if required, a 5 hp drive for the automatic bar screen rake, and
for lighting and controls - about 25 kilowatt connected load in all. With 50 over-
flow events a year (we see only 40) , and several hundred short, dry weather
periods of operation, the running time then will be 280 hours a year so that the
annual power consumption will be 7,000 kwh/year or abour $140/year. Similarily,
the city water consumption will be about 14,000 gallons/year most of which is
consumed during rainy weather. '
' '' The microstrainer is automated. At onset of storm overflow the liquid
,.',... . . . _ ..jTi'... .•
level in the inlet channel rises and actuates a level switch which starts the
microstrainer drum motor, -the backwash jets, turns on the UV lights, and the
bar'screen-rake drive.
.,, ~ The'microstrainer drum speed controls regulate the speed of the drum in
accordance, with the difference in liquid level across the screen which is roughly
proportional to the flow rate. All of the combined sewer overflow passes through
103 :;
-------
• MICROSTRAINER a CONTROLS
o MICROSTRAINER INSTALLATION a BLDG.
O MICROSTRAINER INSTALLED
(LESS ENGR,, LAND, PROFIT)
O 120 SEC. CHLORINATION FACILITY
(LESS ENGR., LAND, PROFIT)
0 50 100
CAPACITY OF FACILITY- CU. FT. PER SEC.
FIGURE 4
104
-------
the drum. If the storm flow should exceed the peak'design rate of the machines
«(i.e., cause a differential in excess of 24,") the excess water overflows the
bypass weirs and flows directly to the receiving stream or to the disinfection
facility and then to the stream. At the end of the storm, the program controls
continue the operation of the microstrainer, sump pumps, etc. , until the chamber
'
is drained and the screen is clean and then shut them down. The instant
readiness and the very low residence volume of the Microstraining technique '
permits unattended operation with very simple controls. Our equipment ran on
all storms under automatic controls.. It was unattended during the first:part of
all storms. No trouble was observed.
The labor required for a facility would be weekly inspection and routine
maintenance visits (i.e.,, lubrication, etc..) and it is believed that a two man
crew could accomplish this in 2 hours. The labor cost would be the cost of
104 hours, or at $2.50/hour, $260/year.
Maintenance supplies, replacement parts, and maintenance labor (in
addition to operation-routine maintenance labor) should, not exceed 1% of the
facility cost per year. We have no long-term experience on the screen life
at high differentials/ however, it is believed that the original screen will serve
for 10 or more years in stormwater sen/ice. The cost of re screening a 10 x 10
is about $5,000. Our experience over a 3 year period has indicated a maintenance
cost of less than 1% of facility cost, even if a screen change every 10 years is
anticipated. ,
105
-------
In summary, the annual cost of a facility having 490 sq ft of open
submersible area (capable of treating 45 cfs) would be:
Capital charge @ 10% of installed facility cost
(less land and engineering)
Utilities - electric power and city water
Routine labor
Maintenance and supplies @ 1% of installed
facility cost
$19,500
200
250
1.950
$21,900
The annual cost of installing and operating a dual 10 x 10 microstrainer
facility is -$22,000/year. Such a facility will accept 92 cfs and treat 46 ,cfs .
Depending on conditions previously discussed, such a facility would serve a
drainage area of from 24 to 240 acres.
The suspended solids removal performance of a microstrainer on storm-
water follows a pattern that will seem strange to engineers accustomed to other
liquid-solid separation techniques such as settling or granular bed filtration.
A large portion of the first increment of solids applied to the screen leak
through before the mat is established. Most of subsequently applied solids are
retained as shown in Figure 5. Thus, those conditions that contribute to high
solids loading; i.e., high potential pollution make for high removals. These
conditions are high flow rate, high stormwater solids concentration and low drum
speed. It may be repeated that the higher the flow rate and the higher the
influent solids, the lower the effluent solids. This latter relation is shown in
Figure 6 and Figure 7.
The suspended solids in the stormwater at our site exhibited a surprising
106
-------
30
N
U.
O
O
O
CD
CO
O
O
03
O
111
2
<
UJ
a:
10
10 20
APPLIED SOLIDS LBS. / IOOO FT. 2
FIGURE 5
STORM WATER SOLIDS APPLIED VS.
STORM WATER SOLIDS RETAINED
107
-------
X
X
X
X
X
X
X X
x x
Xx *
, v X
X XV^
1 1 x
1 1
O CO C0 •* CM 0
_* * • • • ^^
0
o
o
o
to
o
o
CM
O
g
o
d3Nivaisoaow AS aswivisd sanos Q3aN3dsns jo i
x
o>
E
'
•ZL
UJ
•
_J
U.
•—
or
UJ
z
<
or
i—
Q
cc
o
s
z
CO
Q
I]
o
CO
o
UJ
o
Ul
a.
CO
CO
yiou
o
UJ
z
g
t~~
or
CO
o
_j
o
CO
o
h-
o
to ^
QC
lit ^~
UJ
a: (/)
I> Q
o --
!*- O
CO
Q
Ul
0
•^
SUSPEf
h-
z
Ul
^K
_J
u_
.OVMJ
108
-------
X
1 II !•"••-«
X
X
X X
X ~
X
X
X
X
X
*x ^
' '
- • x
I i
o»
E
P
z
UJ
o
u.
z
K
Ul
11
H-
tn
0
o
£K
UJ
J—
i
cc
0 2
o i-
— UJ
z
SOLIDS
o
UJ
o
z
UJ
a.
en
o tf>
o o o o o
CO ' (0 *5j* C^l
iN3nidJ3 ^3NivaisoyoiiAi NI sanos asaNSds
CO
0
o
CO
0
UJ
o
UJ
**•
D
CO
' ^
UJ
Z)
_J
u.
N- U.
UJ
UJ .
CK en
3 >
o ,
*• 5
0
CO
o
SPENDE
^3
CO
H-
UJ
_J
U.
^
ns
109
-------
characteristic. The greater the concentration of solids the easier they were
strained out. The permeability parameter is the flow rate possible at unit head
loss; i.e., one inch of water head loss per inch of mat thickness. The units
of this parameter, borrowed from oil well practice, are inconvenient for
Microstraining since buildup consists of mats of a few thousandths of an inch.
In any event, this permeability is a measure of the flow capacity of the machine
within the differential limitation imposed by the screen strength.
In summary, we have found in two studies totaling about 22'months of
operation at one site that the microstrainer will reduce suspended solids from
50-700 mg/1 down to 40-50 mg/1 at flow rates of 35 to 45 gpm/ft2 of gross
submerged screen area; i.e., 42-54 gpm of unimpeded submerged area. These
flow rates have been routinely achieved within an arbitrary limiation of 24"
of water differential between inlet and outlet liquid levels.
The removal of organic and other oxygen demanding material is shown
on Table 2 to be 25-40%. This removal is confirmed by BOD5, COD and TOG
measurements performed by the Standards Methods with and without a maceration
pretreatment in a Waring Blender. The advantage of this pretreatment is covered
in the formal report on this work.
The Microstraining had little or no effect on the coliform content of the
stormwater.
110
-------
Q
(0
.-.. o
CO O
OO r-H
CO
CO
CO
>,
CO
(0
• l-i
LO
•"o o o
1-1
g CO
"fo o
O-
'O
0)
M_,
(0
M-
CD
O
rO
S
CO
o
o o
CO LO
C-J CN!
CO
"CX3 CO
r-t CO
O
(0
to-
•
to
CSJ
JD
U3
(0
E-i
O 2
to
C CD
"3 ra
"rd jjj1
+J
co 13
O 0
o to.
T)
c
--O fO
0 ••
s-
Q
.LO
8
CQ
Ti
to.
s-.'
CD
O
(0
fO"
0
+->
(0
i_,
CD
O
fO
,'£.
•-I O
CS] o
CO
o
00
LO LO
(SI CM CO
o
CD
;o
o
o
CO Cv]
CO CO
. -CO
rM
LO LO
• CD
i-H t>-
LO CO CO
CT>
CM LO
CO '<-* CO
0'
fl
g'-g
11
to 0
fd. «
O
Q.
o '
to
o
o
S-,
to
-t-l
•I-J
(C5
to .0
Cn "F!
f S
o <5
ro
CO
to
'•,0;;.
-a
^jj •
*"•£'•
CO 1 — !
0 vi.,""
g 0
(d *•"
of
o d
•.•.•-.'ts>'-
" • • C3 xP
. ' •; .- H. P^1
to- ;CQ, CQ
- CD to -to'
Q CO CO
' - c! (0 <0'
522
•S--8 H
i-5 •— • •— '
0 "° ^ °
W ^ to to
to to Cn Cn
2 "a 2 2
o §, 2? ®
^< !> S> Zf
i— i I^H <3J <;
r— 1 S^
,.•'•• --.-->\ •• • •
' "-'•'• en w "
We 0 '
0 G CO
. O m (d
•C'-CQ <»' '
5-.S g
«. ® ®
£ o Q
0UPO
2 !-< to to
S 0 en en
2 J3 (0 fd
r\ B <-"«-•
CD 1 § S
Qh? *Si (2
f^i
-------
The advantages of the Microstrai-ning technique for suspended solids
removal are:
1. Instant readiness and low residence volume permit simple
automation for unattended facilities at remote locations. : ,",
2. Instant readiness and very high flow rate capability/unit
equipment cost permits installation without flow equalization
basins.
3. The low head loss - 3 ft - through the entire Microstraining
facility will generally eliminate the need for re pumping.
4. • The removal performance of Microstraining, where highest
removals, both absolute and percentage-wise, are achieved
at highest flow rates and highest suspended solids loadings',
is particularly suitable for the conditions existing in combined
sewer overflow servi ce.
5. The excess flow bypass is an integral part of a microstrainer
facility and eliminates the need for this necessary feature
as an appendage.
6. The very high flow rate capability and low residence volume
permit Microstraining to be the lowest cost solids removal
technique - less than $500/year per cfs capacity.
112
-------
-' • • . ACKNOWLEDGEMENT
This work was conducted with the City of Philadelphia in two phases,
(l) Under a contract from the Environmental'Protection Agency to the Cochrane
Division of the Crane Co., and (2) under an EPA grant-to the City of
Philadelphia. The efforts Of City personnel were under the general direction
of Carmen Guarino, Water Commissioner, with William Wankoff and M.
Lazanoff, serving,as Project Director and Laboratory Director. -J. Radzuil
headed the .City's R and D Department who also lent valuable' assistance.
The assistance and guidance of these people are gratefully acknowledged.
The overall guidance and helpful advice of Richard Field/ Project Officer,
EPA, Edison, New Jersey, were most valuable. "
113
-------
-------
SECTION VI
HIGH-RATE MULTI-MEDIA FILTRATION
Patrick Harvey
Environmental Engineer
U.S. Environmental Protection Agency
Region II, 26 Federal Plaza
.New York, New York
115
-------
GENERAL
The nature of combined sewer overflow, i.e., a highly pollut-
ional, high volume discharge, requires a relatively high rate treat-
ment process for economical pollution control. Deep bed, high rate
filtration, a new development in the field of industrial wastewater
treatment, has demonstrated favorable cost-efficiency, factors when
dealing with high volume wastewater discharges/ especially where •
suspended solids comprise one of the principal contaminants. Thus,
it was felt that such a process, which currently has significant
applicability and usage in the steel industry, might prove an effect-1
ive. and efficient solution to the treatment of combined sewer over-
flows .
To evaluate the applicability and effectiveness of the high rate
filtration process in removing contaminants from combined sewer over-
flows, a testing program was undertaken at Cleveland's Southerly
Wastewater Treatment Plant, beginning in 1970. The work was under-
taken by Hydrotechnic Corporation, Consulting Engineers, New York,
New York, under the sponsorship of the Office of Research and Monitor-
ing, USEPA.
The City of Cleveland ranks seventh in the nation in total area
served by combined sewers ,(44,000 acres), and is fourth in pppulation
served by combined sewer systems (1,000,000 persons). As can be
expected, Cleveland has a, very serious problem of combined sewer over-
flows .
,116
-------
TESTING PROGRAM '•••.'.'•-"•'
;'.,• The two major process units or equipment units in the proposed
treatment system are the drum screen'and the deep,bed, high rate .
filter.• The function of the -screen is to remove coarser material •
(fibrous type, etc.) that would impede the filtration operation.
Construction of a full scale treatment plant employing the, process-'
sequence under study would require design parameters for the screen
and for the filtration process. 'The major criteria,for the screen
are screen type, screen mesh, and hydraulic loading. "
, The filtration system, which is the heart of the overall process
sequence, can be characterized and described by the following para-
meters :
- ; Media'Composition • , . Length of filter run ,
•'..•.' ' .Media. depth •<,•': ...... , Head loss -• , .•-..
•: .Filtration rate , Backwash water volume ' . • ' ',•-•
, Coagulant -addition ; . Backwash procedure . , ,. ,
, A definition of these elements allows the construction,of a
full scale facility. .-•.-,
Testing'equipment at Southerly included a drum screen, two 5,000
gallon storage tanks, lucite filter columns of four (4) and six;'(6)
inch internal diameter, and chemical and polyelectrolyte feed equip-
ment. .(Figures 1 and ,2) . •• , , , - - ,-••-. -.- • • - •
The testing program evaluating the filtration components of the
proposed system was conducted primarily in two phases: 'first, evalu-
ation and selection of system media and filtration rates, and secondly,
optimization of the filtration process via coagulants and polyelec-
trolyte addition prior to filtration.
117
-------
Figure 1
Lucite Filter Columns
Figure 2
Drum Screen and Storage Tanks
118
-------
filter run, and backwash procedure.
TEST RESULTS
The recommended system is a drum screen (No. 40 mesh screen
element) followed By a deep bed, dual media filter (five feet of
No;.3 anthracite over three feet .of No. 612 sand). Sixty-nine
pilot filtration runs were performed in 1970 and 1971 utilizing
this system. Polyelectrolyte feed is an essential and critical
part of the system to achieve optimum treatment efficiencies.
Data utilizing coagulants ahead of filtration showed inconsistency
in treatment efficiencies and at the present stage of development,
polyelectrolyte feed alone appears optimum.
The proposed system,, with addition of appropriate polyelec--
trolyte, achieved t;he following treatment performance:
Filtration Rate Average Removals(%) .
(gpm.sq ft) Suspended Solids BOD Phosphorous
8
16
24
96
95.
93
43
40
40
66
57
46
average influent suspended solids concentration ranged
from 50 to 500 mg/1 and the average influent BOD concentration
ranged from 30 to 300 mg/1. Effluent 'levels at 24 gpm/s^ ft with
polyelectrolyte addition were 15 mg/1 suspended solids and 22 mg/1
BOD, respectively. '(Figures 3, 4, and 5)
HIgH RATE. FILTRATION -INSTALLATION - " - , '\
Combined sewer overflows would be conveyed from an automated
overflow chamber, or chambers (in case the centralized filtration
system is for many overflow points) , to a low lift pump station.
119
-------
Filtration media evaluated included: four or five feet of
anthracite over three feet of sand. The characteristics of the
media are indicated as follows:
Media Effective Size Uniformity Coefficient
No. 4 Anthracite 7.15 mm.
No. 3 Anthracite 4.0 mm.
No. 2 Anthracite 1.78 mm.
No. 612 Sand 2.0
No. 48 Sand 3.15 mm.
Screen meshes tested included:
Mesh Screen Screen Opening Tyler Screen
Designation microns/inches Scale Equivalent
(mesh)
No. 3 6350 0.025 3
No. 20 841 0.0331 20
No. 40 420 0.0165 35
1.42
1.5
1.63
1.32
1.27
Open Area
(%)
57.6
43.6
43.6
The filter tests were directed to determine the degree of
treatment that could be achieved by using different depths and
composition of filter media when operating at different flux rates,
with and without the application of coagulants and polyelectrolytes.
Using the results of the tests, criteria could be established to
determine design parameters of full scale installations.
The principal water quality parameters carefully observed and
recorded were: suspended solids, BOD, and COD. Measurements were
also made on pH, temperature, total solids, settleable solids, coli-
forms, and total organic, carbon. The laboratory analyses were per-
formed by a local laboratory in Cleveland.
Filtration operational factors measured and recorded were:
media depth and composition.,' flux rate, head loss, length of
120
-------
fc&UTofi
Figure 3
Filtration•System
Performance—
Suspended Solids
Removal
Figure 4
Filtration 'System'
Performance - ,-..-.
B.O.D. Removal
121
-------
|l^^
a ;v;yg:,: • ;,.i:. 10 3O 50
"''11"*
tiy!*'^
Figure 5
Filtration System Performance
Effluent Suspended Solids Quality
'• TOSEWSGE:
TREATMENT
PLANT
Figure 6
High-Rate Filtration Plant
Flow Diagram
122
-------
Figure 7
High Rate Filtration Installation
.-PLAN
Figure 8
Plan - High Rate Filtration Installation
(100 MGD)
123
-------
. fV'tts ."••!" I..- - .•. • ' i ' . •
Figure 9
Longitudinal Section High Rate
Filtration Installation (100 MGD)
Figure 10
Cross Section - High Rate
Filtration Installation (100 MGD)
124
-------
Before entering the pumping station, the combined sewer overflow
would pass through a bar rack (screen) for removal of coarse mater-
ials which might cause problems in the operation, maintenance or
wear of the low lift pumps. In certain locations, where consistent
with local topography and sewer invert, a low lift pumping facility
may not be required.,, ,
The combined sewer overflow from the low lift pump station
would enter a treatment building and be delivered to drum type
screening units. The wastewater would be introduced into the center
of the drum type screen and would pass through the screening mesh
into the influent channel to the filters. A gravity type design,
i.e., open filtration units, is proposed. The water would be intro-
duced at the top of the filter and flow downward through the filter
bed. The plant effluent could be discharged by gravity to the respec-
tive receiving water body.
Filtered wastewater would serve as a source of water for back-
washing filters after the overflow has attenuated to a sufficient
degree. The filtration building would be provided with low pressure
air blowers as a source of" backwash air. Backwash pumps would be
located in the filtration facilities to deliver water to the filters
for backwashing. The treatment building would also include a con-
trol area, office space, a polyelectrolyte feeding set-up, and a
system for adding hypochlorite to ^filter backwash water for the
prevention of slime growth on the filter media. The operation of the
high rate filtration facility would be completely automated, and could
be left unattended, except for routine maintenance and periodic de-
livery of chemicals. In full size .treatment systems, chlorine feed
for disinfection could be incorporated into the filtration facilities.
Dirty backwash effluent from the filtration facilities and /
125
-------
screenings would be directed into the interceptor running to the sani-
tary sewage treatment facility. The concentrated solids from the
drum screening units would be passed first through a grinder, and
then through a trash basket or classification device to insure that
very coarse, settleable material is not returned to the sewer system.
Sludge handling facilities should not be located at the filtration
site, as this would prove very costly. Centralization of material
handling facilities has always proved most economical; as an example,
the Southerly Wastewater Treatment receives sludge from another plant
in Cleveland.
For filter backwashing, two types of process control should be
considered: the first parameter would be total head loss through
filter bed, and the second would be effluent suspended solids con-
centration .
For measuring the filter head loss, each filter would be equipped
with a differential pressure transmitter to continuously sense the
loss of head across the filter and transmit a pneumatic signal line-
arly proportionate to this head loss to a central control panel.
When the filter head loss would reach a preset value, the differential
pressure switch associated with the filter would be actuated. A
contact in this switch would open a stepping switch circuit and the
filter would start to backwash.
An alternate, filter backwash control could be achieved with an
effluent suspended solids monitor. A continuous reading, light
scatter type suspended solids meter would be installed in each
filter effluent pipe to continuously measure the suspended solids
concentration and transmit the reading to a recorder at a central
control panel. When the filter breakthrough would suddenly take
place and the suspended solids concentration indicator would reach a
preset level, then a micro switch would be activated and an alarm
126
-------
would be initiated. The operator would check the filter performance
condition and start to backwash the filter.
1
Principal advantages of the proposed system are: high treat-
ment efficiencies, automated operation, and limited space require-
ments as compared with alternate flotation or sedimentation systems.
COST DATA
' • ' ' ' ',,•'• ' v- -"'••.' - *
Estimated total construction costs (ENR=1470) of a filtration
plant for treating combined sewer overflows range from $830,000
for the 25 MGD capacity to $3,754,000 for 200 MGD capacity at design
rate of 24 gpm/sq ft.
Estimated annual cost data ranges from $97,270 per year for a
25 MGD capacity plant to $388,210 per year for a 200 MGD capacity
plant. Annual treatment costs utilizing the high rate filtration
process are due primarily to interest and amoritzation charges, and
are less affected vby the volume of combined sewer overflow to be
treated annually.
These .costs do not include disposal of waste screenings and
filter backwash since the proposed system would discharge these to
the municipal sewage treatment plant. Assuming an average of 200
mg/1 of solids removed and a combined sewer overflow treatment plant
operation of 300 hours per year, solids processing and disposal
costs incurred by the municipal, sewage treatment plant could range
from 3 to 35 percent of the total annual' charges for the combined
sewer overflow treatment facility.
DUAL PURPOSE OF UTILIZATION OF HIGH RATE FILTRATION PROCESS
The selected media for combined sewer overflow treatment was
127
-------
also evaluated in terms of its capacity for polishing secondary
effluent under another research contract. Test data has confirmed
the applicability of this combined sewer overflow media to reducing
suspended solids, BOD, and phosphate to low residuals.
In Cleveland, the total duration of the overflows from the
combined sewer system is approximately 300 hours per annum. This
indicates the possibility of utilizing dual purpose treatment plants
based on the high rate filtration process. Such installations
would treat combined sewer overflows when they occur, and in be-
tween such periods, for over 95 percent of the time, the filtration
process would treat other wastewaters depending on the location of
the process.
For a high rate filtration process for combined sewer overflow
treatment located in the area of the domestic wastewater treatment
plant, the filtration process can be utilized for polishing the treat-
ment plant effluent as well as to protect the effluent quality
during plant overloading or process malfunction.
The economical benefits of such dual purpose utilization of
the high rate filtration process should not be overlooked.
128
-------
SECTION VII
SCREENING/DISSOLVED - AIR FLOTATION
TREATMENT OF COMBINED SEWER OVERFLOWS
by
Mahendra K. Gupta
Robert W. Agnew
Environmental Sciences Division
Envirex, Inc. (A Rexnord Co.)
Milwaukee, Wisconsin
129
-------
Introduction
The problem of combined sewer overflows (CSO) has been recognized
as a significant pollution problem in recent years (1). Large
amounts of untreated pollutants find their way into our water
courses through this route. The abatement methods dealing with
this problem are sewer separation, storage, treatment, or a
combination of these. The cost of separating the sewers is
prohibitive and this method is not considered as an economical
solution to the problem. A great deal of literature has been
published since 1964 which describes the characteristics of (CSO)
(2). Based on the data published, it has now been established
that a major portion .of the pollutional substances in CSO is
particulate in nature. This indicates that an efficient solid/
liquid separation process can be expected,to provide an effective
treatment of CSO. It was the mission of the Environmental Sciences
Division of Rexnord Inc. to develop an effective and economical
solid/liquid separation process under a program sponsored by the
U.S.Environmental Protection Agency.
A combined sewer outfall near Haw!ey Road in the west-central
portion of Milwaukee, Wisconsin was selected as a source of
combined sewer overflow for the bench scale studies. This
outfall services a 495 acre residential area. It was determined
that approximately 42% of the area was impervious, i.e. streets
and parking areas, house roofs etc. The calculated value of the
runoff coefficient was 0.40 and it compares well with the values
reported in the literature (3). The drainage area comprises of
mostly one and two family dwellings with an estimated density of
35 people per acre. No manufacturing industries are located
within the drainage-area except some small business shops.
Bench scale tests were conducted on 14 separate overflow samples
130
-------
to define the quality of the Haw!ey Road outfall and to evaluate
the various potential treatment proces$es. The evaluatory tests
included screening with various sized media, chemical oxidation,
flotation and disinfection. It was 'determined from tHese tests
that chemical oxidation of the raw CSO did not appear technically
and economically' feasible (4). However, the results of the
screening and dissolved-air flotation tests were encouraging.
These tests served as the design basis of a 5 MGD test facility
at the Hawley Road outfall utilizing screening and dissolved
air flotation. ' • '
Design of the Treatment System i ':1>""\
The process "schematic of the proposed treatment system is shown
in Figure 1. The raw overf1ow is pumped from the sewer to a
half inch manually cleaned bar rack.. The purpose of the bar
rack is to remove large objects which may clog or damage the
finer screen downstream. The flow then enters a 50 mesh
(approximately 300 micron) drum screen. The basic screen is
fabricated from mild carbon steel while the screening media is
a 304 stainless steel. The screen is an octagonal shaped drum
with an effective diameter of 7.5 ft', arid 6 ft. length. The
total screen area is 144 sq. ft. with wetted screen area rang-
ing between 72 and 90 sq. ft. depending'upon'the head loss
across the screen. The design hydraulic loading for the screen
is 50 gpm/sq. ft. and a maximum head loss capacity of 14 inches.
The drum speed can be varied in the range of 0:5 to 5.0 rpm.
Screened water is used to backwash' the screen. The solids
which are removed from the screen 'are collected in a hopper and
are then routed to1 the sanitary sewer. The screened effluent
is split into-two portions. ' A major portion of the flow goes
directly to the flotatipn tank while the.remainder of the flow
131
-------
CSO SOURCE
EFFLUENT tO
RECEIVING STREAM
CHEMICAL ADDITION
SCREEN BACKWASH SYSTEM
DRUM SCREEN
50 MESH
Vi" BAR RACK
TO INTERCEPTOR SEWER
SCREENED EFFLUENT
SCREENED
SOLIDS DISCHARGE
FLOATED
SCUM DISCHARGE
FLOTATION ZONE AND CHLORINE CONTACT
MIXING ZONE
AIR SOLUTION SYSTEM
CHLORINE AND CHEMICAL
FLOCCULANT ADDITION
PRESSURE
REDUCTION
AIR DISSOLVING TANK
c
AIR COMPRESSOR
SCREENING/FLOTATION FLOW DIAGRAM
FIGURE 1
132
-------
(approx. 20%) goes to a pressure tank where it is mixed with
air under pressure (approx, 50 psi). The pressurized air-
water "stream is then brought into contact with the bulk of the
raw flow at atmospheric pressure in a mixing zone. The dissolved
air comes out of solution in the form of tiny bubbles (50-100
micron size) in the mixing zone and these bubbles attach them-
selves to the suspended matter in the waste water. The mixed
flow then passes through a distribution baffle and into the
flotation tank where solid/liquid separation occurs. The scum
which floats to the top is then scraped into a trough via skim-
mers and is routed to the sanitary sewer. The treated effluent
is discharged to the Menomonee River.
The main details of the treatment system are shown in Figure 2.
Flexibility was,provided in the design so that the flotation
zone could be segmented for evaluating various hydraulic over-
flow rates. Chemical flocculants when utilized were added to
the raw waste as it enterad the drum screen or in the pressurized
flow stream after the pressure reduction valve. Chlorine was
also added in the pressurized flow stream for disinfection of the
CSO. '"Th'e entire system was automated and .was put,into operation
by sensing a pre-set level of the waste water in the sewer.
Operation of the Demonstration System
The system was operated on 55 separate combined sewer overflows
during 1969 and 1970. The quality characteristics of these over-
flows are seen, in Table 1. About 20 percent of the overflows exhi-
bited the first flush phenomenon, which was either caused by high
rainfall intensity or a length of time greater than four days
between overflows. After the first flush diminished, the quality
of the overflow was remarkably constant for each storm. The 95%
confidence ranges for the extended overflows were only about 10-
133
-------
PLAN VIEW
PRESSURE TANK
RAW
FLOW
•*- 5'-
FLOTATION ZONE
ELEVATION VIEW
BAR SCREEN
MIXING ZONE
DRUM SCREEN
CL^^:
SCUM COUECTOR
EFFLUENT WEIR
rr
^k
8.S' WATER DEPTH
PRESSURIZED FLOW HEADER '
PERFORATED BAFFLE
DEMONSTRATION SYSTEM DETAILS
FIGURE 2
134
-------
TABLE 1
COMBINED SEWER OVERFLOW CHARACTERISTICS AT HAWLEY ROAD3
Analysis
Total Solids (tng/1)
Total Volatile Solids (mg/1)
SS (mg/1)
VSS (mg/1)
COD (mg/1)
BOD (mg/1)
Total Kjeldahl Nitrogen (mg/1)
pH
First Flushes'1
861 ± 117
489 + 83'
522 + 150
308 18.3
581 + 92
186 + 40
17.6 + 3.1
7.0 + 0.1
3
Total Coliform (individuals/ml) 142 x
Dissolved COD/Total COD4
+_ 108
0.34 + 0.04
Extended
3
Overf1ow
378 + 46
185 +_ 23
166+26
90 +_ 14
161 ± 19
44 +_ 10
5.5+0.8
7.2 +_0.1
62.5 x 103 + 27
1
Ranges shown at 95 percent confidence level.
Represents 12 overflows.
Represents 44 overflows.
Represents 34 overflows.
135
-------
15% of the mean value as compared with 20-25% for the first flush
data. The dissolved organic fraction (measured as chemical oxy-'
gen demand) was approximately one third of the total organic load
in the raw combined sewer overflow. This. showed that a large por-
tion (2/3 of the total) of the organic pollutants was of a parti cu-
late nature which would be amenable to treatment via screening/
dissolved-air flotation.
The variables evaluated during operation included hydraulic load-
ing and drum speed for the screening operation, and surface over-
flow rate, pressurized flow rate, operating pressure, and floccu-
lant dosages for the flotation system. The optimum operating con-
ditions based on the treatment of 55 CSO are given in Table 2. The
optimum solids loading rate at a drum speed of 4.7 rpm and a head
loss of 12" was 1.2 pounds of dry solids removed per 100 sq. ft.
of screen area. This loading could possibly be increased by incr-
easing the allowable head loss differential. The hydraulic through-
put rate was in the range of 40-45 gpm/sq.ft. This rate again can
probably be increased depending upon solids loading. It was found
that no statistical difference could be shown in the removal effi-
ciencies by increasing the pressurized flow rate up to 45 percent
of the raw flow, or by increasing the operating pressure to 60 psi.
A pressurized flow rate of 20% of the raw flow at 50 psi was recom-
mended for future designs. The air usage was approximately one
cfm per 100 gpm of pressurized flow. The overflow rate at
which removal efficiencies were satisfactory and the capital
cost still reasonable was 3.3 gpm/sq.ft. Floated scum con-
centrations generally ranged between 0.7 and 1.4% of the raw
flow. The chemical flocculants utilized during this study were
FeClo and a cationic polymer (C-31, Dow Chemical Co.). The
selection of these chemicals was based on the results of a series
of bench scale jar tests. The optimum chemical dosages were
found to be 20 mg/1 FeClg and 4 mg/1 of C-31.
136 .
-------
.TABLE 2
OPTIMUM OPERATIONAL CONDITIONS
Characteristics
Screening
Operational Condition
Backwash
Head Loss
Rotation Speed
Submergence
Hydraulic Throughput Rate
0.7 - 1.0% raw flow
12 in. water
4.7 rpm
50 - 63%
40 - 45 gpm/sq. ft.
Flotation
Floated Scum
Pressurized Flow
Operation Pressure
Overflow Rate
Chemical Dosage
0.75 - 1.41% raw flow
20% raw flow
50 psi
3.3 gpm/sq. ft.
20 mg/1 Fed*
O
4 mg/1 cationic polyelectrolyte
137
-------
The performance of the 50 mesh screen alone is summarized in
Table 3. The pollutant removals (measured in terms of suspended
solids, volatile suspended solids, COD and BOD) ranged between
33-39% for the first flushes and between 26-34% for the extended
overflows. The slightly higher removal efficiencies for the first
flush overflows is probably a result of the screening-filtration
phenomenon that occurs during these high pollutant loading periods.
The total removal efficiencies for the combined screening/flota-
tion system are shown in Table 4. The pollutant removals ranged
between 35-48% without flocculating chemicals. However, the
removal efficiencies were significantly enhanced on the,addition
of flocculating chemicals and ranged between 57-71%. Removals during
the first flushes were similar to the results for extended over-
flows with chemical addition. The average effluent quality exper-
ienced with chemical addition and that can be expected via
screening/flotation treatment is shown in Table 5. These values
compare favorably with many secondary sewage treatment efflu-
ents.
Future Design Considerations
The data presented so far had been based on the results of two
operational seasons, 1969 and 1970. Research was continued on
this treatment facility during 1971 to obtain additional design
data for the optimization of the screening and dissolved-air
flotation processes in order to improve upon the effluent water
quality of the treated combined sewer overflows.
Laboratory bench scale tests have indicated that changing the
split flow mode of dissolved-air flotation to effluent recycle
mode of operation may enhance the effluent water quality signi-
ficantly. This change may require the operation of the flotation
138
-------
TABLE 3
PERCENT POLLUTANT REMOVALS BY SCREENING*
Characteristics
First Flushes
Extended Overflow
ss • ' '"
VSS
• COD
BOD '
36 + 16
37 i 18
39 +,15
33+17
27 + 5
34 + 5
26 +_ 5
27 + 5
* Values given at the 95 percent confidence level.
TABLE 4
PERCENT POLLUTANTS REMOVALS BY SCREENING/FLOTATION TREATMENT*
Extended Overflows
Characteristic ,
SS .
VSS
'COD
BOD
First
Flushes
72 +_ 6
75 +_6
64 + 6
55 + 8
Without
Chetoicals
43 + 7
48 +_ 11
41 + 8
35 + 8
With
Chemicals
71 + 9
71 +_ 9
57 + 11
60 + 11
Nitrogen (total
Kjeldahl)
46 + 7
29 + 8
24 + 9
* Values shown in a 95 percent confidence range.
v 139
-------
TABLE 5
EXPECTED AVERAGE EFFLUENT QUALITY AT HAWLEY ROAD
Analysis
SS
VSS
COD
BOD
Nitrogen (total Kjeldahl)
Value
"(mg/1)
48
26
69
20
4.2
140
-------
system at reduced overflow rates and could therefore increase the
flotation area requirements by approximately.20%...
Also, several other chemical flocculant combinations have shown
promise over the ferric chloride - C-31 polymer combination uti-
lized during the 1969 and 1970 operational seasons. Use.of pow-
der activated carbon along with screening/dissolved-air flotation
has also shown some merit. The economics of these concepts for ,
an optimum cost benefit relationship still, need evaluation. These
evaluations are a part of the proposed modifications to the Hawley
Road treatment facility. It is anticipated that these consider-
ations will be evaluated on the modified Hawley Road treatment
facility during the 1973 operational season..
Racine Root River Project
Encouraged by the promising results of the Hawley Road demonstration
facility, a search was made to find a site where the feasibility
of utilizing screening/dissolved-air flotation could be demonstrated
on a full scale for the treatment of combined sewer overflows. The
City of Racine, Wisconsin was indicated.to be an ideal site for
such a project. Racine is a city of approximately 100,000 people
located on Lake Michigan, approximately 30 miles south of Milwaukee.
The Root River, a stream having a mean annual discharge of approxi-
mately 100 cfs flows through the city and serves as a receiving body
for runoff from much of the northern half of the city. There are
approximately 700-acres of land having combined sewer systems in this
area. In the 3.7 miles of Root River through the city, there are
36 combined sewer overflow points and 17 storm water discharges to
the river. It was estimated that the cost of separation of the
existing combined sewer areas in Racine would be 10-13 million
dollars. The estimated cost of installing the screening/dissolved-
air flotation treatment plants at the various outfalls was 4
141
Two full scale SDAF systems have bieen. installed in Racine for
treatment of combined sewer overflow. The design criteria for
each of the various elements is shown, in Table,6. The systems
have been designed for completely automatic startup, operation
and shutdown.
The two systems are similar in function.and differ only in
design capacity. A schematic diagram of the larger system is
143 ,
-------
million dollars. Thus significant savings were evident in going
for the screening/dissolved-air flotation route for the treatment
of combined sewer overflow problem in the City of Racine.
In April of 1970 a grant application was submitted to the U.S.
Environmental Protection Agency. Under the terms of this proposal
the funds would be rendered by the federal government, State of
Wisconsin, and the City of Racine. The technical approach proposed
for meeting the project objectives includes the following elements:
1. Quantitativ_e_measurement^of^ the.
TABLE 6
DESIGN CRITERIA - SCREENING/AIR FLOTATION TREATMENT SYSTEM
RACINE, WISCONSIN
Item
Contributing area (acres)
Design Storm Intensity (inch/hour)
In-Sewer Storage (gallons)
Design Flow for Treatment System (MGD)
Site .#1 Site #2
82.5
0.5
14.13
364.2
0.5
600,000
44.4
Bar Screens
Mechanically cleaned and located
Just Upstream of Pump Sump
Yes
Yes
Drum Screens
Parallel Operation, automatic
bypass to flotation tanks should all
screens clog
Number of screens 2
Length (feet) 7
Diameter (feet) 8
Filter Media Stainless Steel -
50 mesh, .009 inch wire
Screen Backwash flow gpm
(when operating) 210
4
10
675
144
-------
TABLE.6 CONTINUED .
Item Site #1 Site #2
Flotation System
Operation-Each tank reaches 701 • : .- .,,.-• ... ;, ?, <
• , maximum flow before the next tank is
, put into use. -; ." ,.• ,- ?
Number of tanks 3 -8
Surface overflow rate - gpm/ft 3.5 3.5
,-..•'. Pressurized flow - gpm/tank ; :. , 650 770
Scum Removal - timer controlled
Surface skimmer to scum trough - • •
Screw conveyed to sludge holding tank
Chemicals
Chlorine - maximum concentration mg/1 20
Fed g - maximum concentration mg/1 25
Polyelectrolyte - concentration
•Dependent on specific polyelectrolyte
Sludge Storage ..•--; ,
..1.5% of design flow for 3 hour duration
Volume - cubic feet 3,500
Disposal to sanitary sewer by gravity
Drain following storm
20
25
11,030'
145
-------
shown in Figure 3. Upon sensing a high level in the overflow
sewer, the system is placed into operation. Raw overflow enters
the plant through a mechanically cleaned bar screen located in
the wet well. A by-pass weir is provided for storm flows in
excess of the design capacity. Flow entering the wet well is
pumped by means of a spiral screw pump through a Parshall flume
and into the screening chamber. The output of the flow recorder/
totalizers are used to provide a proportional signal for pacing
the chemical feed equipment. Ferric chloride is added to the
wastewater upstream of the screens. Chlorine and polyelectrolyte
are added downstream of the screens.
Each of the drum screens is equipped with 50 mesh stainless steel
screens. The screens are backwashed at a preset headless level.
Solids removed on the screen are conveyed to a sludge holding tank
by means of a screw conveyor which runs along the head end of the
flotation tanks.
Effluent from the drum screens is diverted to the flotation tanks
by means of a series of weirs and orifices. The inlet system is
designed so that the tanks are filled in series. This enables the
utilization of only as much tankage as is actually required by the
storm flow. Screened effluent is used as the source of pressurized
flow.
Scum produced in the air flotation tanks is skimmed to the head
end of the tanks where it is conveyed to the sludge holding tanks
by means of a screw conveyor. All sludge generated during a
storm is held in the holding tanks until after the storm subsides
and then is discharged to the interceptor sewer. At some future
date it may prove fruitful to provide onsite dewatering facili-
ties rather than return the concentrated sludge to the sewer
system.
146
-------
o
•§•
o
z
=>
z
2
SCHEMATIC LAYOUT OF THE TREATMENT SYSTEM FOR SITE NO. 1
FIGURE 3
147
-------
The flotation tank effluent which has been chlorinated will be
discharged directly to the Root River.
Following a storm all of the sludge, as well as the contents of
the flotation tanks will be discharged to the adjacent sanitary
interceptor sewer. The system will then be ready for the next
storm. .
Special Considerations
Certain special considerations have been made in order to insure
optimum use of the system. A floodgate was installed in one of
the overflow sewers to provide approximately 600,000 gallons of
in-system storage. This storage capacity will be utilized when
the treatment facility reaches full capacity.
In addition, the system has been equipped to be completely self-
draining. This will enable use of the system during peripds of
snow melt and cold weather. A roof has also been provided to
prevent floe breakup during heavy rains..
Costs
The cost for the Racine SDAF system is $30,000 per mgd installed
capacity. A detailed cost breakdown is given in Table 7.
Racine Program
A two year system evaluation and optimization is scheduled to
begin on April 1, 1973. The intent of this program is to fully
evaluate the installed facility, validate the EPA Stormwater
Management Model and determine the effect of the system on water
quality in the Root River.
148
-------
TABLE 7 : -
COST OF SCREENING/DISSOLVED AIR FLOTATION
Capital Costs
Cost per MGD Capacity
Cost per Acre*
$30,000
$ 3,900
* Based on 0.5"/hour runoff rate
Operating Costs
Power
Chemicals
Maintenance
TOTAL
Based on plant capacity .of more than 30 MGD
and 40 hours per month operation.
<£/1000 gallons
0.54 :•:
2.51 ...
0.04 :
3.09^/1000 gallons
149
-------
Acknowledgement
The work in this paper for the Hawley Road Demonstration Facility
was sponsored by the U.S. Environmental Protection Agency. The
implementation of the findings of the Hawley Road project as
applied to the Racine Root River Project were undertaken through
the joint sponsorship of the U.S.E.P.A., State of Wisconsin and
the City of Racine. Portions of this paper have been derived from
two publications: 1) Screening/Flotation Treatment of Combined
Sewer Overflows, EPA Project Report by Ecology Division, Rex
Chainbelt Inc., WPCR Series 11020 FDC, January, 1972, and 2) Treat-
ment of Combined Sewer Overflows by D.G.Mason, JWPCF, December,
1972.
150
-------
REFERENCES
1. Pollution of Stormwater;and Overflows from Combined Sewer
Systems - A Preliminary Appraisal,1 USPHS (November 1964).
2. .Combined Sewer Overflow Abatement Technology, U.S. Depart-
;.^ment of .the Interior, FWQA (June 1970).
3. Fair, G. and Geyer, 0., Water Supply and Waste Disposal,
John Wiley & Sons, N.Y, (1961). ;
4. • Screening/Flotation Treatment of Combined Sewer Overflows,
Ecology Division, Rex Chainbelt' Inc., Contract 14-12-40,
11020 FDC (January 1972),
151
-------
-------
SECTION VIII
HIGH-RATE DISINFECTION OF
COMBINED SEWER OVERFLOW
by
George E. Glover, P.E,
Research Engineer
Cochrane Division-Crane Co.
153
-------
The bacteria content of combined sewer overflow has been reported to
be as high as 30 million total coliform/100 ml and 3 million fecal coliform.
These levels are 1,000 to 10,000 times the allowable concentration in
secondary effluents and similar restrictions have been considered for combined
sewer overflows. The techniques used to remove suspended solids have
in themselves no ability to remove or kill coliform. Thus bacteria kills of
3 to 4 logs (that is, 99.9% to 99.99%) are required as a separate operation for
combined and separate sewer overflows.
As reported by others (1) (previous speakers) it may be possible to
achieve a suitable bacteria kill with high.chlorine dosages within certain types
of solids removal devices so that no separate contact chamber will be required.
Considerable more work needs to be done over a broad range of flow rates before
the proposed advantage of dual use of this volume can be utilized on full scale
plants. It is anticipated that required bacteria kills may not be obtained at.
low flow rates.
The special design considerations required to cope with the very high
instantaneous overflow rates previously mentioned (this morning) for removal
of suspended solids and organic matter hold for the disinfection equipment
as well.
Conventional chlorine contact chambers installed at sewage plants
are sized to provide 15 to 30 minutes detention which would require considerable
area (about 1 acre per 250 acres drained at 1.0 of s/acre). Operating close to
154
-------
their design rate as determined by'the 2 to 1 diurnal flow variation, these basins,
as often as not, fail to achieve the required bacteria kills. During the initial
filling, these sewage contact chambers do not, and are not expected to, perform.
A contact chamber,sized to'provide 15 minutes residence for a peak stormwater
overflow rate would never be filled to its operating level during most storms.
The operation of conventional 15-30 minute contact chambers in combined
sewer overflow would be uncertain at best. •
Our work on disinfection as well as the work of others (2) (3) was
performed in pilot size contact chambers at a constant flow rate. That is, these
chambers have not been tested at the wide (20 to 1) variations in overflow rate
anticipated for-a full scale chamber in stormwater service. As will be seen
later, the assumption that performance of a contact chamber will be as good,
If not better, at lower flow rates as it is at the higher rates is questionable
even though the contact time is longer.
We have made, five disinfections of combined sewer overflow while the
storm was in progress. We achieved 99 .99% kill (4 logs) with chlorine
dosages (10 ppm) in 120 seconds. The flow rate through our units -we have
two identical units - was 20 gpm.' In every case, both total and fecal coliform
were reduced to below 10 cells/100 ml. This performance was obtained on both
the raw overflow before Microstraining and the micros trained effluent. The
3 minute chlorine demand-was surprisingly uniform at about 3 ppm for the
microstrained effluent and somewhat higher for the raw stormwater.
One of these chambers is shown in Figure 1. They were designed to
155
-------
Figure 1
Intensely Mixed Chlorine Contact Chamber
156
-------
ensure that the hypochlorite was promptly and well mixed with the stormwater.
More important, or equally important, they were designed to ensure a high
degree of small eddy turbulence in the passages of the contact chamber.
We attribute the extraordinarily high kill rate of these chambers to
the turbulence during contact time.
The very recent literature (Collins et al (2), Kruse et al (3) and
the Dow work (4) ) reports several instances of laboratory studies on sewage
and stormwater disinfection where similar extraordinary kill rates have been
observed. Examination of the apparatus and the procedure used in these
studies reveals that very high turbulence existed during these studies as well.
In one case - a beaker study by Kruse et al (3), a high stirring rate was
used to demonstrate the advantage of prompt and thorough dispersion of the
chlorine. Very high (4 - 5 logs) kill rate of bacteria was observed in
2 minutes when the fast s'tirring rate (i.e., "fast mix") condition was sustained
throughout the whole study. Much poorer performance (only 1-2 logs in
2 minutes) was obtained at the same dosage when the more normal mixing
regime of a few seconds fast mix followed by 15 minutes slow mix was used.
It is of great importance that,in this study,virus were killed at high rate under
the sustained fast mix condition for a few minutes whereas there was minimal
virus kill even with prolonged slow mixing.
In the case of the Dow EPA (4) study, a long 1,500 ft tube was selected
as a flow thru contact chamber. This configuration was apparently selected
to permit precise collection of samples after a specified contact time and to
157
-------
a collision during the operation.
Several studies (7) (8) have shown that the reduction of the number of
particles (I.e., the formation of a single particle from two colliding particles)
is proportional to the GT product in secondary effluent flocculation. Special
hardware has been developed to enhance the flocculation of sewage-like
solids (9). -Design and calculation methods have been developed so that the
mixing intensities as measured by velocity gradient can be controlled in the
laboratory (1'P) (11) and also reproduced in full scale equipment (12).
The application of this already developed mixing intensity technology
to disinfection has been proposed by the writer (13).
The following will be a description of (a) the performance of the pilot
units, (b) the preliminary design scheme, and (c) of a 92 cfs chamber designed
according to this scheme.
Figure 2 shows the results of our disinfection studies to date on
combined sewer overflow in an intensely mixed chlorine contact chamber.
The kill is shown as the surviving fraction of,the total coliform on a log scale.
Note that almost 4 logs (99.99%) are obtained with 10 ppm'dosage at GT of
5,000 (2 minutes at G = 40). The contact time-mixing intensity scale is
dimensionless. It is based on the nominal contact time; that is, the volume
of the chamber divided by the thruput rate and is not corrected for short
circuiting. The value of 9,500, for example, is the product of the G = 40 sec
velocity gradient times 240 seconds (4 minutes) nominal contact time.
For comparison, the velocity gradient in the contact chamber of a local
-1
160
-------
evi
o
ui
tr
DC
f-
eo
n
CM
UJ
CO
O
o
iii
ONiAiAans
JO NOIlOVdd
O<
om
161
-------
sewage plant was calculated from observed velocity and head loss and found
to be about 6 sec . The nominal residence time in this chamber was 1,800
seconds (30 minutes) and the GT product then was 10,000. It might be noted
here that the nominal residence time is used although it has been shown (2)
(14) thatthe true residence time is often considerably less due to short
circuiting. Preliminary studies have indicated that the use of a true residence
time would improve this scheme but this refinement has not been incorporated
yet.
The design objective for our pilot chambers was to achieve a GT of
10,000. We arbitrarily selected 240 seconds (4 minutes) as the residence
time T so that we needed a G of about 40 sec"1. The velocity gradient is
defined (G) as:
G =
Energy Dissipation Rate/Volume
1/2
Viscosity
For open channel flow, it has been shown (12) that:
G = 1730 \ Velocity-fps x Channel Slope ft/ft
V viscosity-cp \
(Eq. 1)
The viscosity is known from the towest temperature to be considered in the
design; e.g., 1. 4 centipoise at 45° F.
The velocity can be arbitrarily selected at some level between 0.25
and 1.5 ft/sec, or possibly higher. The volume of the chamber has already
been determined by the selected nominal residence time so that now the
velocity selection also fixes the path cross-sectional area and path length.
162
-------
The depth of the chamber can now be selected based upon usual
considerations of soil condition, and land cost, etc., although, as-will be
seen later, shallower depths than usual are preferable. The remaining problem
is to ensure that the required slope is obtained. The required slope is
calculated from Equation 1. The slope in open channel flow can be calculated
by Mannings' equation:
I" T I"T /
Slope = (Velocity)2 j^J ||4, 3 (Eq<
where "n" is a factor relating to the obstruction to flow of obstacles at
walls and within the channel. This factor is historically called a : "roughness
factor" and the numerical value found in hydraulics handbooks is 0.011 for
steel or neat concrete and 0.03 for the situati6n where corrugated metal:froms
the wall of a channel whose width is several hundred times the corrugation
height. For our purpose, this could be considered a turbulence promotion
factor. Work is in progress to determine the effective turbulence promotion
effect of corrugated baffles in narrow passages where we believe it to be at
least twice the 0.03 value given above. The effect for other.configurations
is being studied as well. The term "roughness factor" will be used until a more
appropriate term is coined.
The hydraulic radius "R" is the ratio of the cross-sectional area of
the passage in ft2 to the wetted perimeter in feet.
Since the velocity has been fixed and the required slope calculated,
only the roughness factor relating to-iie type of wall and/or baffle surface
and the hydraulic radius relating to the wall area parallel to the flow path
163
-------
can be governed by the designer.
The combined effect of these two variables is calculated from
Equation 2.
For illustration in Figure 3, corrugated baffles parallel to the path are
shown. In this simplified sketch, the significant dimensions are shown.
The passage xvidth is fixed by the selected velocity and channel depth. The
number of the parallel baffles inserted determines the hydraulic radius. The
roughness factor is determined primarily by the surface of the baffle material
selected.
In spite of the undeveloped state of this design scheme, we Were able
to produce a chamber within 6% (9,400) of the design target (10,000), on our
first attempt. Also, additional baffles can be easily inserted at a later date
if required.
This design scheme yields considerable insight to the evaluation of
the performance of existing and future contact chambers. The disinfection
performance has been shown to be a function of the GT parameter.. In
conventional chambers the outlet weir is located near the design rate water,
level so that the water volume is nearly constant at all flow rates. As can be
seen by Equations 1 and 2, the G varies as the (velocity) *5. With constant
liquid level, the T varies as (I/velocity), thus the GT parameter will vary
as the (velocity) •5 or with (flow rate) • 5. This poorer performance at reduced
flow rate would escape attention under relatively constant rate conditions in
a sewage plant. However, under the widely variable rate conditions met in
164
-------
F-
\tf" ' ~*~: ~k
-™-3£BtJ
CO
Q
or o
LU y.
Q.
S
3
Q.
UJ
r-
o:
0
r
O CC -
O UJ
0. X
^ •"•
X 5'
' —
-
(
r
(
'
,
1
:
„/
-v
v_
-
1
' r
(5)-
\_y
^
t*~^r*~i.
,1
I
f
1
i
.f
i
•
i
:
~"N i
N '
> 1
.' 1
r
1
•
/
• , - ' . ••
• -. ."If -^ IT
5 J- . •.." ! i
1 J I j f j 1 fl I l{
l-i i ! * < -1 t • 1 'if
1 j 1 1 j 1 1. H .!••• Si-
1 i f ^ ? ^ ! M " 5 HJ t' <
• 'H * ;>i ! Mi .
! 1 <: i i- j j ] .-} ; j| ••'
I U| if !i.i.f.'l.!-i
jf ij.jj /} j.j.!.i |
^ j — -i-_j
j 5
< i
? |
J i
It
r
.
;
d
^-v-^-.
1
!
/ •
j
/
]
*
\
m-m
•~
i
'
>
«£f i^i
f ^ X CM
! Q 0 —
0 LU
"^ o 2.
O < iZ
CD h-
»^l|^j ™9
u? § I
0 LU
LU Q
0> — 0)
o < o g.
LU Q T
; _ LU X
. j *y - n
• Q- ^ IJL!
.L {-* ^ ~i~
f 1 1 * ».
1 ^~ _,
CO LU
2 r-
LU "^
h- ^
0
165
-------
combined sewer overflow service, it must be considered. The use of a Sutro
weir has been proposed to maintain a constant velocity at all flow rates .
A 92 cfs (60 mgd) Intensity Mixed Chlorine Contact Chamber has been
designed. This chamber was designed to follow a microstrainer facility with
46 cfs treatment capacity and an additional 46 cfs bypass capacity. The
chlorine contact chamber was designed to have 120 seconds residence time at
the 92 cfs rate and,since a Sutro weir is used the residence time at less than
the 92 cfs rate will be about 120 seconds also. The velocity is 1.5 ft/sec and the
amount of baffling and its configuration is such to yield a velocity gradient
G of 40,as in our pilot plant.
The chamber is 40' by 40' and has an average liquid depth of 7' at
maximum flow. Internal walls form a labyrinthine-like passage of 8' in width
and produce a velocity of 1.5 ft/sec. The internal walls are faced with a
commercially available corrugated asbestos siding having 1-1/2" deep
corrugations.
Two additional corrugated panels are mounted as parallel baffles in the
channels forming 32 inch wide passages. The baffles extend from liquid level
to within a foot of the floor. Ideally the floor would be similiarly corrugated*
but this is not necessary. The head loss through the chamber at peak flow
is about 8 inches. (See Figure 3)
The inlet to the chamber is equipped with a 3 hp mixer sweeping an
8' x 81 section of the channel (about 5 sec residence time). A mixer of this
••*
horsepower should be able to impart 1 hydraulic horsepower to the water to
166
-------
enduce a mixing intensity of about 200 sec"1 in this 450 ft3 volume, which
should be adequate for thoroughly mixing the chlorine chemicals. Such a
provision for mixing of chemicals is incomparably superior to the methods
usually used in sewage plants. The mixer should be of such a type that it
can operate at varying water levels from 7* down to 1'.
. The outlet of the chamber should be fitted with a relatively narrow
outlet weir placed as low as the available outfall head will allow, preferably
at the bottom. Further, the outlet weir should be of the Sutro type to maintain
the velocity in the chamber, at less than peak rate, as near the peak rate
velocity as possible. A Sutro weir at the bottom will maintain peak rate
velocity at all flow-thru rates. In the event the allowable outfall head will
not permit placing the weir at the bottom, a small pump must be provided to
empty the chamber at the end of the storm. ' . '
The installed cost of such a chamber has been calculated to be about
$53,000 (in 1969 dollars) less the cost of land, engineering and profit (1). It
is difficult to compare costs developed by different estimators. However, this
cost can be compared to the data developed by Smith (15) of $25,000 for an
$11, 000 ft3 basin, which is the volume of the basin described above. Also,
it can be compared to Smith's estimate of $90,000 for the 8Z,000 ft3 chamber
required to provide 15 minutes residence for 60 mgd in a conventional chamber.
The inherent advantage of increased turbulence economically induced in
this type of installation to enhance reaction rates can be used in many situations.
An obvious example would be to use it in chlorine contact chambers at sewage
plants with savings in construction cost, land, and the advantage of high
virus kill.and reliable bacteria kill.
167
-------
ACKNOWLEDGEMENT
This work was conducted with the City of Philadelphia in two phases,
(I) under a contract from the Environmental Protection Agency to the Cochrane
Division of the Crane Co., and (2) under an EPA grant to the City of
Philadelphia, The efforts of City personnel were under the general direction
of Carmen Guarino, Water Commissioner, with William Wankoff and M.
Lazanoff, serving as Project Director and Laboratory Director. J. Radzuil
headed the City's R and D Department who also lent valuable assistance.
The assistance and guidance of these people are gratefully acknowledged.
The overall guidance and helpful advice of Richard Field, Project Officer,
EPA, Edision, New Jersey, were most valuable.
168
-------
REFERENCES
1. Mason, D.G and Gupta, M.K. ,: (Rex Chainbelt Co.), "Screening/
Flotation Treatment of Combined Sewer Overflows ", Water Pollution
Control Research Series &lln?.n.Fnr; m /79 ; :
2. Collins, H.F., Selleck, R.E.y and White G.C., "Problems in Obtaining
Adequate Sewage Disinfection", Proc. Sanitary Engineering n»,i a,™
ASCE, October 1971. , - - ..
3. Kruse, C.W.., 'Olivieri, V. , and Katvata, K., "The Enhancement of
Viral Inactivation by Halogens ", Water and Sewage Works . June 1971.
4. The Dow Chemical Company (Daniels, S.L.), "Chemical Treatment
of Combined Sewer Overflows", Water Pollution Control R^^rr-.h
Series fl 1023 FDB 09/70. "' ~~ '- : : - ~
5. Cochrane Division-Crane Co. (Glover G.E. and Yatsuk'-P.), "Microstraining
and Disinfection of Combined Sewer Overflows (Phase I) ", Water Pollution
Control Research Series ftlln?.3 -pun ns/yn ~ ;
6. Camp, T. R. ..and. Stein, P.C,, 'Velocity Gradient and Internal Work in
Fluid Motion", -Journal, of the Boston Society of Civil Engineers r 30 219
' j
7. Parker, D.S., Kaufman, W.J., and Jenins, D., "Physical Conditioning
of Activated Sludge Floe", Journal of the Water Pollution Control
Federation . September 19.71.
8. Argaman, Y. , and Kaufman W.J., "Turbulence and Flocculation",
Proc. Sanitary Engineering Division ASCE . April 1970.
9. Cochrane Division -Crane 'Co. (Glover, G.E.), Patent Application for
Liquid Agitator to Produce Uniform Velocity Gradient.
10. Camp/ T.R. , and Conklin, G.F. ; "Towards a Rational Jar Test for
Coagulation", Journal of Boston Society of Civil Engineers. March 1970.
11. Camp, T.R., "Floe Volume Concentration", Journal of American Watsr
Works Association 6Q:Sfi.S (1968). ~ ~~~~ - -
169
-------
12. Cochrane Division-Crane Co. (Glover, G.E.), Series of Internal Studies
"Calculation of Velocity Gradient in Mechanically Activated Chambers "
in Open Channel Flow, in Pipeline Flow, in Flowthru Granular Beds, etc.,
(1964-67).
13. Glover, G.E., Discussion of Collins et al (Reference 2, herein), Journal
of Sanitary Engineering Division ASCE. August 1972.
14. Louie, D.S. and Fohrman, C.H., "Hydraulic Model Studies of Chlorine
Mixing And Contact Chamber", Tournal of Water Pollution Control
Federation, February 1968.
15. Smith, R., Private communication - Preliminary Results of Cost of
Wastewater Treatment Equipment (1969).
170
-------
SECTION IX
THE SWIRL CONCENTRATOR AS A
COMBINED SEWER OVERFLOW REGULATOR
by
Richard H. Sullivan
Assistant Executive Director
American Public Works Association
171
-------
A report by the American Public Works Association published
in 1970 gave the results of a study of combined sewer overflow regulator
facilities. Design, performance and operation and maintenance experiences
from the United States and Canada, and in s.elected foreign countries
were reported. It was evident that North American practice has
emphasized the design of regulators simply as flow splitters, dividing
the quantity of combined sewage to be directed to the treatment
facitilities, and the overflow to receiving waters. Little consideration
was given to improving the quality of the overflow wastewater.
Using hydraulic laboratory tests and mathematical modeling
strongly we have determined that it is possible to remove significant
portions of settleable and floatable solids from combined sewage overflows
by using a swirl concentrator. The practical, simple structure has"
the advantages of low capital cost; absence of primary mechanical parts
should reduce maintenance-problems; and construction largely with inert
material should minimize corrosion. Operation of the facility is . : •
automatically induced by the inflowing combined sewage so that operating
problems normal to dynamic regulators such as clogging will be very . ;
infrequent. ,
The device, as developed, consists of a circular channel in
which rotary,motion of the sewage is induced by the kinetic energy of.
the sewage entering the chamber. Flow to the treatment.plant is deflected
and discharges through an orifice called the foul sewer outlet, located
at the bottom and near the center of the chamber. Excess flow in storm
periods discharges over a circular weir around the center of the tank
and is conveyed to storage treatment devices as required or to receiving
waters. The concept is that the rotary motion causes the sewage to ,
follow along a spiral path through the circular chamber.
172
-------
A free surface vortex was eliminated by using a flow deflector,
preventing flow completing-'its'first'revolution in the chamber from
merging with inlet flow. Some rotational movement remains, but in
the form of a gentle swirl, so that water entering the chamber from the
inlet pipe is slowed down and diffused with very little turbulence. The
particles entering the basin spread over the full cross.'section, of the
channel-and settle rapidly. Solids are entrained along' the.bottom,
around the chamber, and are concentrated at the foul sewer outlet.
Figure 1, Isometric View of Swirl Concentrator; depicts the
final hydraulic model layout showing details such as the floatables
trap, foul outlet and floor gutters.
The swirl concentrator may have practical applications as a
degritter, or grit-removal device for sanitary sewage flows or separate
storm water discharges of urban runoff waters. It may have capabilities
for the clarification of sanitary sewage in treatment plants, in the
form of primary settling or, possibly, final settling chambers. It
might be used for concentrating, thickening, or elutriating sewage
sludges. It may be serviceable in the separation, concentration and
recycling of certain industrial waste waters, such as pulp and paper
wastes or food processing wastes, with reuse of concentrated solids and
recirculation of clarified overflow waters in industrial processing
closed circuit systems.
In water purification practices, it may find feasible
applications in chemical mixing, coagulation and clarification of
raw water. Other uses may prove to be realistic .and workable.
Complete reports describing the hydraulic laboratory study
and the mathematical modeling are included in the report EPA R2-72-008,
September 1972, published by USEPA. The body of the report details the
basis of the assumptions used to establish the character and amount of
flow to be treated and the design of a swirl concentrator based upon
the hydraulic and mathematical studies.
Although the study was performed for the City of Lancaster,
Pennsylvania, with a specific point of'application defined, all work
was accomplished in a manner which allows'ready translation
173
-------
inflow
overflow
Inlet Ramp
Flow Deflector
Scum Ring
Overflow Weir and Weir Plate
Spoilers
Floatables Trap
Foul Sewer Outlet
Floor Gutters
FIGURE 1
ISOMETRIC VIEW OF SWIRL CONCENTRATOR
174
-------
3.
4.
5.
application of the results to conditions which might be found at other
installations and for other purposes.
Consideration of the use of- a swirl concentrator as a combined
sewer overflow regulator facility requires an evaluation of many factors
which include:
1. hydraulic head differential between the collector
and interceptor sewers and head available in collector
sewer to allow insystem storage;
2. .hydraulic capacity of collector sewer;
design flow;
dry-weather flow and capacity of interceptor sewer; and
amount and character of settleable solids.
Although many of these items have been mentioned in the
preceding sections of the report, the importance of each will be
highlighted in order to emphasize the importance of. each point in a
preliminary evaluation of the use of the swirl concentrator.
Hydraulic Head Differential. There must be sufficient
hydraulic head available to allow dry-weather flows to pass through the
facility and remain in the channel. The total head required for
operation is shown in Figure 2, Hydraulic Head Requirements. Determination
of the maximum elevation in the collector sewer that can be utilized for
insystem storage and the differential elevation between the collector
and interceptor sewers is the total available head.
The head required will vary directly with flow and the outlet
losses in the foul sewer.
If sufficient head is not available to operate the foul sewer
discharge by gravity, an economic evaluation would be necessary to
determine the value of either pumping the foul sewer outflow continuously,
or pumping the foul flow during storm conditions and bypassing the swirl
concentrator during'dry-weather conditions, perhaps with a fluidic
regulator.
175
-------
Maximum elevation-
of flooding in
collector sewer
Overflow weir (side)_
height of flow
over weir
Overflow weir (central)
Collector sewer
invert —
Foul outlet
I chamber depth
losses due to outlet,
J gate, connecting
\ sewer and flow
Interceptor through chamber
sewer inlet 1—
FIGURE 2
HYDRAULIC HEAD REQUIREMENTS
hydraulic
head required
Hydraulic Capacity of Collector Sewer'System. The facility
must be designed to handle the total flow which might be delivered by '
the collector system. Thus a study of the drainage area must be made
to determine the limiting grade and pipe sizes which control the quantity
of flow. Solids removal from a peak flowrate may not be required.
If the chamber is not designed for such maximum flows, however, velocity
energies which could be developed at such full flow conditions should
be avoided by providing a bypass in the form of a side overflow weir.
Design Flow. Selection of the design flow for sizing the
chamber should be accomplished on the basis of a complete hydrological
study to determine frequency and amount of precipitation which can be
anticipated as well as runoff hydrographs. Computer models such as
developed by the University of Florida for USEPA can be of assistance
in determining the solids load which may be associated with various amounts
and intensity of precipitation. Provision of maximum solids removal for
a two-year frequency storm for the Lancaster, Pennsylvania, project was
made on the basis of engineering judgment and an evaluation of local
receiving water conditions. As the cost of construction will increase
in direct proportion .to design flow, an economical evaluation should
generally be used to select the flow capacity. The efficiency curve
176
-------
for the facility 'is rather flat over a wide range of ;flow^v resulting
in perhaps large increases in cost for marginal improvements in'
efficiency.
A major constraint in selecting large design flows is the
anticipated shoaling problems of solids at low flow rates in large
facilities. Self cleaning is enhanced by reduced diameters. 'This
consideration may make it desirable to design for lower flows,'
particularly where some form of overflow treatment is to be provided.
Again the computer model can be used to determine the magnitude .of
the solids carry-over problem to the-secondary device. ->•'-..
A third consideration is, the maintenance of low-inflow •
velocities, with turbulence minimized. At the design flow the inflow
velocity should be in' the range of three to five fps. The inflow ' """""
velocity may require reduction by enlarged pipe sections or other
means to achieve this rate. ..".-.••-.,_- ' " . :'''::
,Dry Weather Flow and Capacity of Interceptor Sewer. '.Sizing
.of the foul sewer, the foul outlet and the gutter .depend upon a
de-termination of the dry-weather, flow in addition, the capacity of the
interceptor sewer- to handle the foul flow.must be .known. , The .fbul ,sewer
must be large enough to maintain and not be subject to blockage—
usually a minimum 12-inch diamter. however, the head on the outlet
during overflow conditions will allow considerable variations in the
foul discharge if it is not controlled.
The efficiency of the chamber is affected by the ratio of
foul flow to overflow although there appears to be a broad operating
range over which reasonable removal efficiencies can be maintained.
Maximum advantage should be taken of capacity in the interceptor
system, particularly during the period when the chamber is being'drawn
down. Thus, sensing of the flow in the interceptor and the use of a
control gate on the foul sewer appear desirable to obtain maximum results
from the use 'of the chamber. ' " " ' ' " ' ' :
177
-------
Amount of Character of Settleable Solids. The sewer system
must provide capacity to handle the increase in settleable solids which
will be captured from the combined sewer overflow and discharged to
the treatment plant. In the case of Lancaster, Pennsylvania, this
could amount to mote than a ton of solids from one device in a very
short period of time. Additional grit removal and sludge processing
equipment may be necessary. Should the foul flow be pumped, sumps and
pumps should be designed to handle the anticipated high solids content.
If the settleable solids which can be anticipated in the
combined sewer overflow can be defined by the amount, specific gravity,
and particle size, the mathematical and the hydraulic model may be
used to determine the size of the chamber required to achieve desired
levels of solids removal* Ordinarily this will not be feasible and
the flow criteria developed by the hydraulic model will be used to
design the facility and predict removal efficiencies.
In order to evaluate the efficiency of the chamber, facilities
should be provided for sampling the .inflow, foul sewer flow and overflow.
Settleable solids should be delineated in all of these flows. The
quantity of inflow and foul sewer flow should also be measured.
Difficulties in obtaining representative samples from any of the flows
may make evaluation difficult. However, the treatment plant or
combined sewer overflow treatment facility , if used, should provide an
excellent means of making a gross evaluation into the effectiveness
of the chamber.
Provision of a means to measure the depth of flow over the
weir should act to give a reliable measurement of the flow when added
to the quantity of flow to the foul sewer.
Data from many full-scale operations, operating with various
flow conditions and solid loadings will be necessary to properly
evaluate the usefulness of the swirl concentrator as a combined sewer
overflow regulator.
178
-------
Cost of Facility. The cost of construction of the swirl
concentrator will vary with the length of inlet pipe which must be •
reconstructed, the depth of the chamber and the nature of the material
to be excavated, the need for a roof, and the'general site conditions
under which the work will be conducted. The materials of construction
will usually be concrete and steel and elaborate form work will not
be required.
For the Lancaster, Pennsylvania, application where a
36 foot diameter chamber in limestone is contemplated, the preliminary
estimate of cost was $100,000 in 1972 costs. This cost estimate included
a roof, foul sewer outlet control and a wash-down system. Site
construction problems are minimized in as much as the construction will
.be off of the street right-of-way.
179
-------
-------
SECTION X
THE EPA STORMWATER MANAGEMENT MODEL
A CURRENT OVERVIEW
by
Wayne C. Huber, James P. Heaney, Hasan Sheikh
Department of Environmental Engineering Sciences
University of Florida
Gainesville, Florida 32611
181
-------
1, INTRODUCTION
A, COMBINED AND STORM SEWER OVERFLOWS
An enormous pollution load is placed on streams and other
receiving waters by combined and separate storm sewer overflows.
It has been estimated that the total pounds of pollutants (BOD
and suspended solids) contributed yearly to receiving waters by
such overflows is of the same order of magnitude as that released
by all secondary sewage treatment facilities (Gameson and Davidson,
1964; Field and Struzeski, 1972). The Environmental Protection
Agency (EPA) has recognized this problem and led and coordinated
efforts to develop and demonstrate pollution abatement -procedures
(Field and Struzeski, 1972). These procedures include not only
improved treatment and storage facilities, but also possibilities
for upstream abatement alternatives such as rooftops and parking
lot retention, increased infiltration, improved street sweeping,
retention basins and catchbasin cleaning or removal. The com-
plexities and costs of proposed abatement procedures require that
care and effort be expended by municipalities and others charged
with, decision making for the solution of these problems.
B, THE STORM WATER MANAGEMENT MODEL
It was recognized that an invaluable tool to decision makers
would be a comprehensive mathematical computer simulation program
that would accurately model quantity (flow) and quality (concen-
trations) during the total urban rainfall-runoff process. This model
182
-------
woyld not only provide an accurate representation of the physical
system, but also provide an opportunity to determine the effect of
proposed pollution abatement procedures. Alternatives-could then
be tested on the model and least cost solutions could be developed.
As a result, the University of Florida (UF), Metcalf and Eddy,
Inc., Engineers (ME) and Water Resources Engineers (WRE) were
awarded a joint contract for the development, demonstration and
verification of the Storm Water Management Model (SWMM). The re-
sulting model, completed in Ociober> 1970, has been documented
(EPA, 1971a, b, c, d) and is presently being used by a variety
of consulting firms and universities.
: The present SWMM is descriptive in nature and will model most
urban configurations encompassing rainfall, runoff, drainage,
storage-treatment, and receiving waters. The major components of
the SWMM are illustrated in Figure 1-1. However, it does not
define nor determine any decisions for the system or consider alter-
native methods for efficient economic comparisons.
C, 'DECISION MAKING
In recognition of the need for improved decision making
capabilities, the University of Florida submitted a proposal to EPA
titled "A Decision Making Model for the Management of Storm Water .
Pollution Control" in which it was intended to provide a sys-
tematic procedure which, could be applied to a wide variety of
.specific circumstances in support of intelligent management decisions,
The work required to obtain a least cost solution would be considerably
183
-------
RUNOFF.
(RUNOFF)
DECAY'
(QUAL)
INFILTRATION
JFILTRAT
(INFIL)
TRANSPORT
(TRANS)
EXTERNAL
STORAGE.
{STORAG)
RECEIVING WATER
(RECEIV)
DRY WEATHER
FLOW
(FILTH)
INTERNAL
S"TORAGEV
(TSTROT)
TREATMENT
(TREAT)
COST
(TSTCST)
COST
(TRCOST)
Note: Subroutine names are shown in parentheses.
Figure 1-1
Overview of Model,Structure
184
-------
reduced by means of determining the origin of the most severe ••"—
pollution load, consideration of all upstream and downstream"pol-
lution abatement procedures and associated costs, and through the
possible use of mathematical optimization techniques. ;,
The project was funded as part of an EPA Demonstration
Grant to Lancaster,,,. Pennsylvania (Federal Grant No.. 11023GSC)-,
in which an underground "silo,"1 a swirl concentrator and a micro-
strainer were to be installed at the outfall of the Stevens Avenue
Drainage District to control overflow into the Conestoga Creek *
(details are presented in the next section). ',
Results of the decision-making methodology and other aspects
of the research have recently been formulated .(Heaney and Huber,
1973). Decision-making for urban storm water management is
presented in the broader context of urban water resources management.
Pollution sources and control options are inventoried and accompanied
by economic data. Performance standards are considered and the
importance of automobile-related facilities (e..^.., streets, parking
lots, curbs and gutters) as contributors'to storm water pollution
and quantity is emphasized. Finally, a,linear programming and game
theory approach is used to develop efficient and equitable control
strategies.
This paper presents an overview of the SWMM by illustrating its
use in Lancaster; the following section is "taken from the Final
Report (Heaney and Huber, 1973) from which other details are available.
Major revisions to the Model have been made to include urban erosion
185
-------
prediction, modeling of new treatment devices and'biological treat-
ment facilities, monitoring of significant pollution sources,
flexibility in modeling new areas, new and improved cost functions
for treatment and storage options and a modest hydraulic design
capability as well as minor programming changes and slight format
revisions. The SWMM has proven to be a useful and economical
tool in the assessment of urban storm water problems. Incjividual
runs described in the following section, for instance, could be
accomplished using less than three minutes of CPU time on the
IBM 370/165 at the University of Florida Computing Center, for a
Runoff-Transport-Storage/Treatment-Receiving simulation. Although
computational changes vary, they are well within reasonable bounds.
186
-------
2, TESTING IN LTOSTER, PENNSYLVANIA
The/City of Lancaster, Pennsylvania, population 79,500, is
.situated in a drainage area of about 8.24 square miles (5,274 acres),
The receiving stream in the Lancaster area is the Conestoga Creek
which drains an area of approximately 473 square miles into the
Susquehanna River. The average flow is 387 cubic feet per second
with a maximum recorded flow of 22,800 cubic feet per second.
There are two sewage treatment plants within the city, both of
which discharge into the Conestoga Creek. The North Plant with a
capacity of 10 mgd serves a population of 36,000 people, and the
South Plant recently expanded from 6 mgd to 12 mgd and is designed to
serve 69,000 people. Both plants provide secondary treatment. About
one third of the flow to the North Plant is derived from areas with
separate sewers outside the city serving an estimated population of
17,500 people and some industries. The remaining two thirds of the
sewage flow to the North Plant is derived from the combined sewers
serving the north part of the city plus about 250 suburban acres
estimated to have 18,500 people.and many water-using industries. In
addition, most of the year the water table is high resulting in con-
siderable infiltration. An overflow line diverts excess flow to the
Conestoga during wet weather. The North Plant drainage,area is esti-
mated at 3.72 square miles,
The South Plant is designed to handle a population of 34,500
served by combined sewers and,in addition,up to an approximately equal
187
-------
amount from separated sewers throughout the surrounding area. The
South Plant drainage area encompasses 4.52 square miles and is
comprised of four districts. Stevens Avenue district which is the
subject of EPA demonstration grant is one of the four districts
connected to the South Plant, Three of the districts, including
Stevens Avenue, pump the sewage from a receiving station within the
district to the South Plant. All locations have overflow arrange-
ments that discharge into the Conestoga Creek when the capacity of
the system is exceeded.
The total drainage area of the Stevens Avenue district is 227
acres which, while only about 4.3% of the total Lancaster
drainage area served by North and South treatment plants, is 17% of
the drainage area designed to flow into the South Plant from combined
sewers. The population within the Stevens Avenue district is esti-
mated at 3,900. Figure 2-1 illustrates various drainage districts
within the city.
1. DEMONSTRATION GRANT DESCRIPTION
In order to remedy the situation resulting from combined sewer
overflows, the City of Lancaster decided to explore means other than
sewer separation. Construction of several underground silos at
various locations within the city is contemplated for retention of
overflow during wet periods and subsequent pumping to the treatment
plants during low flow periods,
Stevens Avenue district was selected as the demonstration site
for evaluation of the effectiveness of a silo in combating combined
188
-------
<0
CO
§
•H
4-1
60
•S
0)
u
Ml
H -a
i c
PJ ea
l-i -H
00 §
•H >
(U
PM
CO
1
O
CO
O
•H
0)
60
CD
— <0 £J
189
-------
sewer overflows. The sewer layout for Stevens Avenue district is
shown in Figure 2-2. During normal dry weather periods, the dry
weather flow is pumped to the South treatment plant. During wet
periods, when the incoming flow to the pump station exceeds the
capacity of the station, the overflow discharges directly into the
Conestoga Creek through a 60 inch sewer located at point 6 on
Figure 2-1.
The City of Lancaster also authorized APWA to develop design
parameters for a full-scale swirl concentrator for removal of
solids prior *to the retention of flow in the underground silo.
Location of the demonstration site is shown in Figure.2-2. A
flow diagram of the proposed swirl concentrator-silo treatment is
presented in Figure 2-3. In order to fully evaluate this treatment
the city decided to include chlorination and microstraining as a
part of this demonstration project. The capacity of the silo is
expected to be 160,000 cf.
The tasks assigned to the University of Florida were as follows;
1) Conduct further verification and testing of the Storm
Water Management Model based on active overflow
measurements on selected storm events and to make
refinements to the Model;
2) Provide results of simulations to the APWA in order
for it to develop design criteria and sizing of the
swirl concentrator;
3) Simulate, the effect of the swirl concentrator-under-
ground silo treatment; and
4) Simulate the effect of combined sewer overflow from
the entire city to the Conestoga Creek.
190
-------
CONDUIT USED IN STUDY
© SUBCATCHMENT MANHOLE
9) MANHOLE
—— CONDUIT NOT USED IN STUDY
STEVENS AVENUE DRAINAGE AREA
LANCASTER,PA.
SCALE IN FEET
0 500
_„ A Figure 2-2
Stevens Avenue Drainage Area with Runoff-Transport Numbering System.
1000
191
-------
MICRO-
STRAINER
n
CO
I
CC
UJ,
o
UJ
o
UJ
cc
3
Li.
4J
•H
CO
§
•H
01
a
8
p
4-1
n)
w
en c!
I O
IN -H
60
cfl
J-l
o
4-1
w
4J
I
4J
n)
0)
I
192
-------
2. DESCRIPTION OF> THE STEVENS AVENUE RUNS
A total of four studies comprising nine storms were simulated.
The city and its engineers provided input data as well as two overall
measurements. The Stevens Avenue district was subdivided into 41
subcatchments. A description of each study and its results are given
below:
Study No. 2.—The first study was based on a/series of storms
between July 29 and August 3, 1971. This six-day period deposited
a record amount of precipitation throughout the Lancaster area -
(variously measured between 7.3 and 9.46 inches); During;four of the
six days, the storms were very,intense over short periods; in one case,
being'the second heaviest of record! For purposes of simulation,
Study No. 1 was divided into six storms. The amount and times of
precipitation assumed for each of these six storms are shown in
Figures 2-4 through 2-9 and results of computer simulations for
each of these storms are shown in;the same figures. These figures
show the expected quantity and quality of the overflow from the
Stevens Avenue district for a given rainfall. These runs indicate
that an overflow/as high as 400 cfs may be expected for a storm event
similar to Storm No. 6. ,-•-,.
\. These computer runs also indicate that .total suspended solids and
BOD discharges expected in the overflow aay be on the order of magnitude
of 778 pounds and 635 pounds respectively for Storm No. 5 and 849 pounds"
and 768 pounds respectively for Storm No. 6. Unfortunately, since'
actual flow measurements were not taken during this study, it was not pos-
sible to determine the actual overflow quantity and quality. However,
193
-------
.S
>-
55
ui
CM'
^COMPUTED FLOW FROM TRANSPORT
g
<
ee
v>
CO
300"4locTsoo6=00 7=00
TIME, HOUR OF DAY
CM
.COMPUTED BOD FROM TRANSPORT
3=00
8=00
700 8--00
TIME, HOUR OF DAY
COMPUTED SS FROM TRANSPORT
3:00
4:00
5=00
6:00
7=00
TIME, HOUR OF DAY
8--00
Figure 2-4
Runoff-Transport Simulation for Stevens Avenue.
Study 1, Storm 1.
194
-------
.5
i ~"
111
CM
fi-
42°
o
s,
U.O.
RAINFALL
V-— COMPUTED FLOW FROM
TRANSPORT
o
o
m
3=00
COO.
QCM
O
CO
So-l
o1"
LU
Q.
10
CO
00
300 5:00 r=oo 9:00
TIME, HOUR OF DAY
• COMPUTED BOD FROM TRANSPORT
5=00 7=00
TIME, HOUR OF DAY
9=00.
COMPUTED SS FROM TRANSPORT
5=00
7:00
TIME* HOUR OF DAY
• Figure 2-5
Runoff -Transport Simulation .for Stevens Avenue.
Study 1, Storm 2.
195
-------
— CM-
>-
111
Z
2
to-
i 00
co-
Q
O
CD
3'-00
-------
. o-
,-c
>-
10-
ui
U. O-
2
3:00
3:00
1:00
RAINFALL
COMPUTED FLOW FROM TRANSPORT
4:00 5=00 5=00
TIME , HOUR OF DAY
7:00
COMPUTED BOD FROM TRANSPORT
4:00 . 5'00 6=00 .
,TlME, HOUR OF DAY
7=00
COMPUTED SS FROM TRANSPORT
4=00 5=00 6=00
TIME, HOUR OF DAY
7=00
Runoffs-Transport Simulation, for Stevens Avenue.
Study 1, Storm 4.
197
-------
results of subsequent studies indicate that actual overflows are
generally predicted adequately by computer runs. Quality predictions
are more variable.
Results of this study were used by'APWA in sizing the swirl con-
centrator. A design flow of this device was established at 150 cfs.
Computer simulation studies were also conducted for all six
storms to evaluate the effect of the swirl concentrator-underground
silo facilities on the combined overflow quality. The results of
Storm Nos. 5 and 6 are shown on Figures 2-8 and 2-9 respectively.
As illustrated in these figures, the quality of the overflow is
significantly improved through the installation of the swirl concen-
trator-underground silo.
Study No. 2.—This study consisted of a storm that began in the
morning of August 27, 1971 and continued almost 30 hours to the
morning of the next day. It resulted in varying amounts of rainfall
throughout the city averaging more than 3.5 inches. The results of
the computer simulation were similar to those obtained from Study No. 1,
and for this reason are not included herein. Again, no measurements
were taken during this study.
Study No. 3.—This study is-based on a relatively minor rainfall
event of March 22, 1972. This study is of special importance, however,
because it is one of the types most frequently experienced in terms of
intensity of rainfall. It is also one for which relatively complete
verification data such as rainfall, flow readings and samples were
collected. The rainfall is shown in Figure 2-10 along with results
of the computer simulation showing overflow quantity and quality.
198
-------
\
.S
co-
en
z
<
o:
tn
f--
ro
•So
in
"c\r
^-RAINFALL
•COMPUTED FLOW FROM TRANSPORT
-SILO STORAGE
COMPUTED FLOW TO RIVER
^COMPUTED FLOW TO SOUTH
PLANT WITH SILO
3=00
4:00
TIME, HOUR OF DAY
5=00
u>
-COMPUTED BOD FROM TRANSPORT
-SILO STORAGE
^COMPUTED BOD TO RIVER
COMPUTED BOD TO SOUTH PLANT WITH,
3;00 4:00 5:00
*,"..' ' ., ' ' •
TIME, HOUR OF DAY
COMPUTED SS FROM TRANSPORT
'SILO STORAGE
:OMPUTED SS TO RIVER
/COMPUTED SS TO SOUTH PLANT WITH.
/ SILO
> ^-~^
3:00
4=00
5:00
TIME, HOUR OF DAY .. '.
,.,. Figure 2-8 \
Runoff-Transport Simulation for Stevens Avenue with Silo an4 Swirl Concentrator,
Study 1, Storm 5. .
-------
O-,
CO
55
LJ
z
< CM
CM
CO
RAINFALL
-COMPUTED FLOW FROM TRANSPORT
COMPUTED FLOW TO RIVER
•COMPUTED FLOW TO SOUTH PLANT
WITH SILO
3=00
to-
4!00
TIME, HOUR OF DAY
-SILO STORAGE
COMPUTED BOD TO RIVER
5=00
/!.__ y
3=00
/COMPUTED BOD FROM TRANSPORT
/COMPUTED BOD TO SOUTH PLANT WITH SILO
4iOO 5:00
TIME, 'HOUR OF DAY
IOMPUTED SS FROM TRANSPORT.
•COMPUTED SS TO RIVER
SILO STORAGE
COMPUTED SS TO SOUTH PLANT WITH SILO
3:00 4=00
TIME, HOUR OF DAY
5=OO
Figure 2-9
Runoff-Transport Simulation for Stevens Avenue with Silo and Swirl Concentrator.
Study 1, Storm 6.
200
-------
Shown in the same illustration are the actual quantity and quality
measurements of the overflow. It can be seen that agreement between
the computer simulation and the actual measurements of flow is fairly
good considering the degree of accuracy of the input data as well
as that of the measurements. The agreement between the computed and
measured quality parameters is not as good as for flows.
Computer simulations were also conducted on this study to deter-
mine the effect of the swirl concentrator-underground silo system.
These results are also shown in Figure 2-10. With the silo system, the
Model indicates no overflow in the Conestoga .Creek.
Study No. 4.— This study is based on a storm that occured on
November 29, 1971. This study is also of importance from the stand-
point of Model verification as overflow measurements were conducted
during this storm. The rainfall and results of the computer simu-
lation for this storm are presented in Figure 2-11 along with the
actual measurements for comparison. Again, it/can be seen that
agreement between the actual measurements and predicted results is
fairly good. The predicted results of the swirl concentrator-under-
ground silo system are also shown in Figure 2-11.
3.' RUNS IN THE NORTH AND SOUTH DISTRICT
_ Limited computer simulations were also conducted for the North
and South drainage districts. The North district was subdivided
into 66 catchments and the South district into 104 catchments. The
sewer layouts for the North and South districts are shown in Figures
2-12 and 2-13.
201
-------
J=
\
.c
I
1
cc
OMPUTED, FLOW
SOUTH PLANT
WITH SILO
EASURED FLOW
-COMPUTED FLOW FROM
TRANSPORT
11:20
4:20
I2':20 1:20 2i20 3=20
TIME, HOUR OF DAY
/MEASURED BOD
BOD FROM TRANSPORT
•COMPUTED BOD TO
SOUTH PLANT WITH SILO
11:20
12=20
4=20
•§•*,
•r CM
co
Q
_i
o
CO
n
: co
CO
CO
|:20 2=20 3:20
TIME, HOUR OF DAY
MEASURED SS
COMPUTED SS FROM TRANSPORT
MEASURED SS
.COMPUTED SS TO
SOUTH PLANT WITH SILO
11=20
12=20
1:20
2:20
3'20
4:20
TIME, HOUR OF DAY
Figure 2-10 .-,.'•,.•
Runoff-Transport Simulation for Stevens Avenue with Silo and Swirl Concentrator .
Study 3. No Overflow to River Since Silo Capacity not Exceeded.
202
-------
.£
I-?
I
i
RAINFALL
COMPUTED FLOW FROM TRANSPORT
c°MPUTED FLOW TO RIVER
MEASURED FLOW
/COMPUTED FLOW TO SOUTH
j \\( PLANT WITH SILO
I * »\\i • • '
,}...„ ..\%VP... - ... ^. ..........
5;I5 T^Ii"
HOUR, TIME OF DAY
^COMPUTED BOD FROM TRANSPORT
9=15
o- MEASURED BOD
/•COMPUTED BOD TO SOUTH PLANT WITH SILO
-COMPUTED BOD TO RIVER
IMS 3'!5 5:15 TI59=15
HOUR, TIME OF DAY
-COMPUTED SS FROM TRANSPORT
o-MEASURED SS
COMPUTED SS TO SOUTH PLANT
WITH SILO
IMPUTED SS TO RIVER
1=15
3:I5
5=15
HOUR, TIMS OF DAY
, Figure 2-11
Runoff-Transport Simulation for Stevens Avenue with Silo and Swirl Concentrator.
• ' ' - . Study 4.
' 203 ;'
-------
\r
204
-------
205
-------
Results of computer simulation for Study No. 3 for the North
district are presented,in Figure 2-14 and for the South .district
in'Figure 2-15. The North district outfall is located at point 1
while the South district outfall is located at point 12 as shown
in Figure 2-11.
An examination of these figures shows that for a rainfall event
equivalent to Study No. 3, overflow from the North district would
be about 100 cfs and from the South district, 160 cfs. The BOD
and SS discharged to the river would be 7,075 pounds and 9,696
pounds from the North District and 4,468 pounds and 10,006 pounds
respectively for the South district..
4. EFFECT ON RECEIVING WATER
To simulate the effect of the overflow on the Conestoga Creek,
Receiving Water Model was run on the entire city for the Study No. 3.
The manner in which various districts were combined is shown on
Figure 2-16^ In conducting this run, the swirl concentrator
was used at Stevens Avenue while Refined Storage and
Treatment Model, as described elsewhere* was utilized to simulate
the existing biological treatment at the North and South plants. The
silo was deleted in order to have an overflow at Stevens Avenue
outfall since the installation of the silo prevents any overflow for
rainfall event equivalent to Study No. 3.
The reaeration coefficient for the.Conestoga Creek was computed
from a formula by O'Connor and Dobbins (1958). Results of the Receiving
Water Model are shown in Figures 2-17 through 2-20. Figure 2-17 shows
206
-------
COMPUTED FLOW
FROM TRANSPORT
11=20
I2:20 |i20 2=20 3-:20
TIME, HOUR OF DAY
IMPUTED BOD FROM TRANSPORT
4:20
COMPUTED BOD
FROM TREATMENT
12:20
1=20 2=20 3=20 4=20 5=20
TIME, HOUR OF DAY
^COMPUTED SS FROM TRANSPORT
-COMPUTED SS FROM TREATMENT
12:20
4:20
TIME, HOUR OF DAY
. Figure 2-14
Simulation of North Drainage District.
Study 3.
207
5=20
-------
O-i
£
ID
-------
SOUTH
A
RUNOFF a
TRANSPORT
SOUTH
B
RUNOFF a
STEVENS
AVE.
RUNOFF 8r
TRANSPORT
RUNOFF
TRANSPORT
TRANSPORT
STEVENS
AVE.
TREATMENT
COMBJNE
SOUTH
SOUTH a
STEVENS
SILO
SOUTH
TREATMENT
PLANT
NORTH
TREATMENT
PLANT
COLLATE
SOUTH, NORTH
& STEVENS
OVERFLOW
RECEIVING
' Figure 2-16
Combination of SWMM Runs for Overall 'Lancaster Simulation.
209
-------
DO profiles 24 and 48 hours after the storm inception, while
Figure 4-28 shows the BOD profile for the same period. Suspended
solids and coliform are shown in Figures 2-19 and 2-20 respectively.
Initial values used to simulate the Receiving Water Model are listed
in Table 2-1:
TABLE 2-1
PARAMETERS USED FOR SIMULATING RECEIVING WATER MODEL
Dissolved Oxygen in Conestoga Creek (all junctions)
BOD in Conestoga Creek (all junctions)
Suspendid Solids in Conestoga Creek (all junctions)
Coliform in Conestoga Creek (all junctions)
Decay Coefficient (BOD)
Reaeration Coefficient
Flow in Conestoga Creek (entering junction 1)
10.0 mg/1
5.0 mg/1
10.0 mg/1
50/100 ml
0.20/day
1.50/day
700 cfs
5. SUMMARY
The above discussion can be summarized as follows:
1) The SWMM was able to predict fairly accurately
the quantity as well as quality of the combined
overflow for the Stevens Avenue district in
Lancaster.
2) The installation of the swirl concentrator and
the silo will result in substantial improvement
in the quality of the overflow at Stevens Avenue,
provided the full-scale performance of the swirl
concentrator is comparible to the results obtained
in laboratory studies by APWA.
210
-------
-------
CM
CO
in
o ±
r> 10
O)
00
(0
to
to
CM
OS
l/Ooi 009
01
co
o
LU
UJ
CO
x oo a'
0) CO
ui
o
•8 1
CVI
212
-------
o
tH to
O)
CO
s-
CD
m
o
•CM
H
O
CO
-a
-------
0)
(0
3 —
"O
0)
09
(0
tfl
to
O
(0
o
(0
o
ID
O
ro
Ul
IU
u.
o
z
<
(0
o
I
U
60
O
4-1
0}
(1)
g
CO
111
o
1
a)
d
CM
o
M
O
OOS3 0003 0091
iiu ooi/siMyojnoo jo
OOOI
OOQ
214
-------
3,
American Public Works Association, "Water Pollution Aspects
of Urban Runoff,"• Federal Water Pollution Control Admini-
.stration Contract WP-20-15, 1969.
Environmental Protection Agency, "Storm Water Management Model,"
Water Pollution Control Research Series, Washington, DC, 1971:
a. "Volume I, Final Report," Rept. No. 11024DOC07/71
b. "Volume II, Verification and Testing," Rept. No. 11024DOC08/71
c. "Volume III, User's Manual," Rept. No. 11024DOC09/71
d. "Volume IV, Program Listing," Rept. No. 11024DOC10/71
Field R., and E. J. Struzeski, Jr., "Management and Control
of Combined Sewer Overflows," J. Water Pollution Control
Federation., Vol. 44, No. 7, 1972.
Gameson, A. L., and R. N. Davidson, "Storm Water Investi-
gations at Northhampton," Institute of Sewage Purification,
Conference Paper Np. 5, Annual Conference, Leandudno,
England, 1962.
Heaney, J. P., and W. C. Huber, "A Decision-Making Model for the
Management of Urban Storm Water Pollution," Final Report to
Lancaster, Pennsylvania and Environmental Protection Agency,
Department of Environmental Engineering Sciences, Gainesville,
Florida, June, 1973.
O'Connor, P. J., and W. E. Dobbins, "Mechanism of Reaeration
in Natural Streams," Trans. ASCE3 Vol. 123, 1958.
215
-------
Simulation results presented herein were performed for the
City of Lancaster, Pennsylvania as part of Demonstration Grant
No. 11023GSC from the Environmental Protection Agency. We are
grateful to many people in Lancaster, EPA, Meridian Engineering,
Inc. (consultants to Lancaster), and APWA for their cooperation
and suggestions. Our systems analyst, Mr. W. Alan Peltz "bore the
brunt of the computational effort. Computations were performed
on the IBM 370/165 at the Northeast Regional Data Center at the
University of Florida.
*US. GOVERNMENT PRINTING OFFICE: 1974 546-317/Z96 1-3
216
-------
SELECTED WATER
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
*31 3. AccessionNo.
4. Title
Combined Sewer Overflow Seminar Papers
7. Authors)
USEPA Storm & Combined Sewer technology Branch
9. Organization jj.g. Environmental Protection Agency
Edison Water Quality Research Laboratory
, National Environmental Research Center- Cinn.
Edison, New Jersey
10. Project No.
11. Contract/'Grant No.
IS. Supplementary Notes
U.S. Environmental Protection Agency Report No. EPA-670/2-73-077
November 1973.
16. Abstract
The U.S. Environmental Protection Agency in conjunction
with the New York State Department of Environmental Conser-
vation conducted three one-day seminars on the problem of wet-
weather flow pollution abatement. Many facets of the problem
, were considered including a brief overview of its magnitude
and what the federal government is doing to manage and control
this source of pollution. Various management, control, and
treatment techniques were described and the most up-to-date
information on design and economics was presented. The audi-
ence consisted of consulting and municipal engineers from all
areas of New York State.
This publication is a compilation of the papers presented
at the seminar.
17a. Descriptors
Combined sewer overflow management and control
Identifiers
Infiltration/Inflow, Regulation, Pressure Sewers, Microstraining,
Filtration, Dissolved Air Flotation, Disinfection, Storm Water
Management Model.
I7c. CO WRR Field & Group
18. Availability
aifc
B,2^V-*yP .%?>•; ^t*"1 -W-Tis;-; U.S. DEPARTMENT OF THE I
KfJjifeiiitpVito: ;,';i%,^;.;::-,:-•;-!«': WASHINGTON. D. c. 20240
Send To:
Abstractor USEPA Storm and Combined I Institution U.S. Environmental Protection Agency
WRSIC 102 (REV. JUNE i97» Sewer T.eclinolagy Brantsh
-------
-------
-------
3
------- |