R             Filter




S          erf lows


                         [NISTRAT:

-------
WATER POLWJ ION CO 7TROL RISEARC SERTh
The Water Pollution Control Research Reports describe the results and
progress in the control and abatement of pollution of our Nation ‘a
waters. They provide a central source of information on the research,
developnent and demonstration activities of the Federal Water Quality
Administration, Department of the Iyiterior, through in-house research
and grants and contracts with Federal, State, and local agencies,
research institutions, and industrial organizations.
Triplicate tear-out abstract cards are placed inside the back cover to
facilitate information retrieval. Space is provided on the card for
the user’s accession number and for additional keywords.
Inquiriea pertaining to Water Pollution Control Research Reports should
be directed to the Heed, Project Reports System, Room 1108, Planning
and. Resources Office, Office of Research and Developaent, Department
of the Interior, Federal Water Quality Adinini.stration, Washington, D.C.
202142.
Previously issued reports on the Storm and Combined Sewer Pollution
Control Program:
WP-20-1l Problems of Combined Sewer Facilities and Overflows -
1967.
WP-20-15 Water Pollution Aspects of Urban Runoff.
WP-20-16 Strainer/Filter Treatment of Combined Sewer Overflows.
WP-20-17 Dissolved Air flotation Treatment of Combined Sever
Overflows.
WP-20-18 Improved Ses iRnts for Infiltration Control.
WP-20-2]. Selected Urban Storm Water Runoff Abstracts.
WP-20-22 Polymers for Sewer Flow Control.
ORD-li Combined Sewer Separation Using Pressure Sewers.
DAST-li Crazed Resin Filtration of Combined Sewer Overflows.
DAST-5 Rotary Vibratory Fine Screening of Combined Sewer
Overflows.
DAST-6 Storm Water Problems and Control in Sanitary Sewers,
Oakland and Berkeley, California.
DAST-9 Sever Infiltration Reduction by Zone Pumping.
DAST-13 Design of a Combined Sewer Fluidie Regulator.
DAST-.25 Rapid-Flow Filter for Sewer Overflows.
DAST-29 Control of Pollution by Underwater Storage.
DAST-32 Stream Pollution and Abatement from Combined Sever
Overflows - Bucyrus, Ohio.
DAST -36 Storm and Combined Sewer Demonstration Projects -
January 1970.
DAST-37 Combined Sever Overflow S n1r -r Papers.

-------
                  RAPID-FLOW FILTER

                FOR SEWER OVERFLOWS
              The Evaluation of Coarse Coal as a

                Filter Medium to Remove Large

                  Solids from Sewer Overflows
FEDERAL WATER POLLUTION CONTROL ADMINISTRATION

             DEPARTMENT OF  THE INTERIOR
                           by
               The Rand Development Corporation
                       13600 Deise Avenue
                        Cleveland,  Ohio

                     Contract No. WA 67-2

                         August 1969

-------
FWPCA Review Notice
This report has been reviewed by
the Federal Water Pollution Con-
trol Administration and approved
for publication. Approval does not
signify that the contents nece s-
sarily reflect the views and policies
of the Federal Water Pollution Con-
trol Administration.

-------
ABSTRACT
The concept of a rapid-flow filter able to remove large, visually
objectionable solids from the kinds of overflows which result from
combined sewer systems, has been shown to be feasible.
In a pilot installation at the terminus of an existing urban overflow loca-
tion, a filter using lump coal performed this pollution control measure
with a minimum of maintenance or difficulty. What difficulty existed
centered about obtaining representative samples of overflows contain-
ing the large solids the filter was intended to, and did, remove.
A preferred filter uses lump coal as the filter medium, preferably
sized to three-fourths by one and one-fourth inches, free of fines,
and about eight inches in depth. The overflow is directed onto the fil-
ter bed in such a manner that the filter bed is not displaced. When
plugged, or upon a routine basis, the filter bed is replaced; the spent
bed, composed of coal and solids, is incinerated or landfilled or dis-
posed of by whatever manner is locally in use which does not pollute
the atmosphere or surface or underground waters.
The work reported herein indicates that a rapid-flow filter can be used
for partial treatment of sewer overflows by removing large solids to
the extent of up to sixty-five percent removals. However, the poten-
tially most valuable contribution in the work was the finding that con-
ventional sewage sampling does not provide a representative indication
of the nature of large solids content. A recommendation for an
improved sampling procedure is made.
This report was submitted in fulfillment of Contract WA 67-2 between
the Federal Water Pollution Control Administration and the Rand
Development Corporation.
111

-------
FIGURES
FIG. NO.
TITLE
PAGE NO.
VI Some of the larger solids removed from a
combined sewer overflow by the Rapid-Flow
Filter
VII Rapid-Flow Filter pilot plant sheet
VIII Construction Costs
IX Yearly Operating Costs
X View of the solids sampling screens before
air drying
Acknowledgements
Bibliography
Appendix I Sample Handling and Analysis
75
77
22
23
77
79
81
83
vi

-------
CONTENTS
SECTION NO. TITLE PAGE NO.
Abstract . .iii
1 Conclusions . . i
2 Recommendations
3 Introduction . . 7
4 Site Selection
5 Filter Design ..ll
6 Operation and Maintenance . . 1 7
7 Costs . .19
8 Results and Discussion of Results . . .25
9 Discussion . . 33
10 Sampling ..35
TABLES
TABLE NO. TITLE PAGE NO.
I Chronological Tabulation of Natural and Simulated
Events 39
II Description of Natural Events 47
III Natural Events - ConventionalAutomatic Samples .49
IV Natural Events - Grab Samples 51
V Simulated Eve nts- Conventional Automatic Samples 53
VI Simulated Events - Grab Samples 55
VII Simulated Events - Stratified Bed 57
VIII Filter Bed Initial Run Results 59
IX Experimental Filter Screen Sampling Technique . . 61
X Total Segment Sampling Technique 63
XI Flow Rate Filter Media Size Tests 65
XII Filter Media Size Tests - Solids Removal 67
FIGURES
FIG. NO. TITLE PAGE NO.
I Twin outfall near Dorver Avenue and East 77th
- Street in Cleveland, Ohio 69
II View of typical houses on Beman Avenue 69
III Side spiliway type overflow structure 71
IV Topographical representation of Beman Avenue
drainage area 73
V Appearance of filter basket with filtered solids. . . 75
v

-------
SECTION I
CONCLUSIONS
1. The concept of a rapid-flow filter utilizing inexpensive, disposable,
lump media, such as coal, to remove unsightly floating and sus-
pended solids from combined storm and sanitary sewer overflows
has been demonstrated to be practical. Cost estimates for utiliz-
ing the concept can be based on data obtained during this program.
2. Conventional sewage sampling methods, where composite samples
are collected conventionally by the periodic dipping of small por-
tions from the stream, are not adequate for obtaining representa-
tive samples of the large solids transported in raw sewage or in
sewer overflows. This misrepresentation is then compounded in
the laboratory when the small aliquot portions are taken from the
composite for actual analysis.
In sampling, due regard must be given to means by which samples
large enough to be truly representative can be drawn. The design
of such methods must preclude effects of the sampler itself on the
stream flow pattern. These requirements can best be fulfilled by
the method described as segment-sampling, wherein a segment of
the entire stream, or of a portion of it that is deemed to be repre-
sentative, is taken periodically and analyzed in its entirety.
3. Optimum particle size of the filter medium for rapid filtration
purposes appears to lie in the range of 1/2 by 2-1/2 inches. A
standard commercial 3/4 by 1-1/4 inch stoker coal is entirely
satisfactory for the removal of gross solids at the design flow rate
of twenty gallons per square foot per minute.
The coal should be reasonably free of dust or fines before it is
placed into the filter, in order that the flow rate not be unduly
restricted; screening is satisfactory for this purpose. However,
washing of the coal is necessary in order to prevent residual dust
from being washed into the filtrate.
4. Filter bed depth does not appear to be critical. The degree of
solids removal does not materially increase at depths exceeding
eight inches.
5. For purposes of design a value of twenty gallons per square foot
per minute has been found useful for a bed eight to eighteen inches
deep composed of washed 3/4 by 1-1/4 inch coal. Variations in
1

-------
individual configuration for reasons of topography, convenience,
security and similar considerations can be expected. The
designed filtering area should of course accommodate the
maximum anticipated flow of the overflow structure.
6. Because the concept is based on the use of a disposable filter
medium, no backwashing or other processing is required; nor is
the attendance of an operator. Replacement of the medium can be
pre-scheduled, or determined by periodic inspection or by auto-
matic signalling.
It appears that replacement approximately six times per year
might be expected as an average; actual life expectancy of a given
filter bed is more a function of the character and load of the over-
flow than of the total volume of water passed. It was found in this
work that reasonable predictions of the nature of the overflow can
be based, in turn, on the season of the year, the nature of the
watershed, land use and individual circumstances. Intensity of
rainfall is especially influential, and it is important for design
purposes that the rainfall data for the microclimate of the indivi-
dual watershed be used, rather than regional rainfall statistics.
7. The rapid-flow filter is simple in design, and malfunction is
limited to overloading or to pluggage of the medium. In such an
event the filter itself merely overflows and the net effect is the
same as if the filter were not present.
8. Although no conclusive analytical data could be performed on the
subject, it is evident that a substantial amount of composting of
organic material removed by the filter takes place in the filter,
especially during warm weather. Organic material observed on
the filter surface immediately following an overflow visibly
degrades in size and consistency within a day or two.
9. At no time during this work was an odor of sewage solids or gar-
bage detected in connection with the filter operation as long as
unoiled coal was used and the filter was shaded from the sun.
The phenomenon appears to be a reproducible characteristic of
the process.
2

-------
10. In other recent work 0) coal has been found to be useful in sewage
treatment, partially because of a marked ability of some coals to
adsorb dissolved organic matter. Although this characteristic is
probably responsible for part of the inhibition of odor, its effect
is minimal with respect to sewage treatment in the rapid-flow
filter because of the short residence time and small surface area
of the large lumps. The rapid-flow filter process is essentially
one of physical removal of gross solids.
3

-------
SECTION 2
RECOMMENDATIONS
1. Sampling techniques for measuring the solids content of waste
streams containing gross solids can be improved.
In the usual technique the increments that make up the sample are
too small relative to the size of the solids in the main stream.
Furthermore 1 the sample gathering procedure disrupts the flow
pattern of the stream so that representative samples can not be
taken. The mere presence of the sampling dipper creates
streamlines which divert light solids around it.
Analytical techniques in general use are valueless for gross
solids. The physical size of the increment analyzed is smaller
than that of many of the solids in the waste stream; for example,
cans and dead rats.
A decided improvement would be the gathering of a large total
segment sample and then filtering it. The filtrate could then be
analyzed conventionally. This procedure provides a better repre-
sentation of what is actually in the waste stream, with regard to
both liquids and solids. (See section on Sampling.)
2. The rapid-flow filter concept may be used as a relatively inexpen-
sive measure for removing gross, visually objectionable solids
from sewer overflows.
This system should not be considered as a complete treatment
process or as a permanent solution to the problem of overloaded
sewers, or of the complete treatment of overflow f-rom them.
3. Automatically cleaned coarse screens with provisions for return-
ing the solids to the main stream of the sewer would probably be
better for large - over 10 MGD - overflows. Automatic bypass
arrangements would be needed for this type of device so that the
overflow would still be operable during power failures.
4. It was found that the 25 square-foot pilot filter failed to duplicate
the filtration capacity of the 1. 1 square-foot test filter in total
gallons filtered for filter area before pluggage. This suggests
that filter design modifications are in order for future installa-
5

-------
tions. The rapid flow (approximately 125 GPM/1t 2 ) through the
test filter kept small materials such as sand, etc. from settling
out of the stream and plugging the filter. The flow through a
given area of the pilot filter was somewhat dependent upon the
storm overflow rate, which varied from 0 to 50 GPM/ft 2 . The
average maximum pilot filtration rate during each of the 91
natural and simulated events was approximately 15 GPM/ft 2 .
During the 55 natural events this filtration rate ranged from zero up
to the maximum.
Therefore, a better filter would be one that was divided into adja-
cent sections so that during a low rate event only a few sections
would be used and the flow rate through a given section would main-
tain at a high rate until it became plugged. The unfiltered material
would overflow each section onto the next as each succeeding sec-
tion plugged or for other reasons was not able to accommodate the
full overflow.
This can be accomplished by having a series of baskets each fitted
with an internal honeycomb arrangement to hold the coal in place
against the force of the overflow. The sewer overflow would be
distributed from one end onto the first basket. Excess unfiltered
waste water would flow out the opposite end down and onto one end
of another filter basket. This may be repeated for three to five
baskets in series.
Further advantages of this improved design would be that only the
minimum number of baskets would be exhausted and need to be
changed at any given period while the far end baskets would still
be relatively clean and available for use.
6

-------
SEC TIION 3
INTIRODUCTION
There are over 1, 300 U. S. c ities serve i by combined sewers. (2)
Approximately one-fourth of the urban population is so serviced, or
perhaps 36 million persons. Overflow from sewers, including corn-
bined sewers, is an important pollutional source, although its real
importance was not widely recognized until the mid-1960’s. Not only
does the pollutional load contributed by sewer overflows include con-
‘taminants easily identified by standard tests for biochemical oxygen
demand, suspended solids, pathogens, coliform organisms and the
like, but overflows also contribute obvious esthetically objectionable
materials to surface waters. The American public is aware of these
effects, especially the visual ones, and supports efforts toward tl ieir
a batement and elimination.
roposed solutions to the overflow pollution problem are numerous
iizrxcluding various screening and filtering means, storage basins,
lau ger sewer:s and treatment plants, disinfection, sedimentation,
etkc.., any of wb ich will unavoidably add to the costs of protecting sur-
face water quality. Replacing combined sewers with separate conduits
ftor sanitary wa sthes and storm water could, it is estimated, (2) cost
a. s much as 48 bill]lion dollars. Despite this enormous cost, no treat-
m ent whatever of the storm waters would result.
Federal Water Poilllution Control Administration Contract No. WA 67-2
with the Rand Deveiloprnent Corporation, Cleveland, Ohio, iconcerned
one of many approaches to solutions to overflow pollution.’ It was
limited to the development and evaluation of a rapid-flow filter device
for removal of gross, esthetically objectionable solid objects from
combined sewer overflows. The device was intended to accept sudden,
large flows while removing relatively large objects such/as cans, plas-
tic items, glass objects and the like. Naturally a maxiri ium of removal
of pollutants is always sought, but it is obvious that coai se filtration
is limited in its ability to effect pollution abatement; nevertheless, the
existence of tons of thousands of overflow structures ju tifies the
examination of any approach which may feasibly - and perhaps quickly -
begin to alleviate water pollution.
No solution to any pollution is achieved if acceptable ultimate disposal
of the pollutants removed is not also achieved. In the rapid-flow filter
concept, a combustible filter medium may be used so that the medium
7

-------
with its load of removed pollutants may be properly incinerated for
disposal. A non-combustible filter medium may also be used where
supervised land-fill operations are used to dispose of solid waste in
acceptable fashion.
While coal is the preferred combustible filter medium wherever coal
is an item of commerce, the rapid—flow concept is not to be confused
with the coal-based sewage treatment process developed by Rand
Development Corporation on behalf of the U. S. Department of Interior,
Office of Coal Research. (3)
8

-------
SECTION 4
SITE SELECTION
A number of factors were considered in’the selection of the test site.
1. Type of outfall. Only combined storm and sanitary overflow out-
falls were considered for the pilot rapid—flow filter.
2. A clearly defined drainage area.
3. Convenience and accessibility. To keep total project costs as low
as possible locations which would’ require little modification and
also be relatively close to the contractor’s laboratory were given
prime consideration. /
4. Adapt4bility to construc-’tion. A location at which construction
costs’cou.ld be kept low is always a consideration. This was
es$eciaily importa nt for this pilot operation because it was
necessary to include instrumentation and other accessories not
needed for a simple rapid-flow filter structure.
5. Frequency of overflow. Records concerning overflows are limited
in the Cleveland area as in virtually all localities. Wherever
information was available it was referred to. Local inquiry was
also made.
6. Volume of overflow. An overflow conduit in the 18” to 24” dia-
meter size was judged to be a reasonable choice for an initial
test facility, able to provide it with at least 1.8 x 106 MGD.
7. Security. The protection of the Federally-owned equipment and
the data the instruments would provide.
An engineering report (4) which had recently been made to survey
Cleveland’s nearby southeast side sewers, suggested consideration of
the outfall just west of the blind intersection of Dorver Avenue and
East 77th Street. This is a twin outfall (see Figure 1) serving single
block areas of Beman Avenue (see Figure II) and Dorver Avenue. The
overflow structure serving the Beman Avenue combined sewer is of
the side spillway construction (see Figure III for a generalized draw-
ing). The Beman Avenue drainage area (see Figure IV) is approximately
1500 feet long and encompasses approximately 8 acres. Approximately
9

-------
40 percent of the total drainage area is comprised of street, sidewalks,
driveways, rooftops and other fast draining areas. The remainder is
lawn or natural cover. The drainage area is an established neighbor-
hood of mainly single residences with a population of approximately 300.
From inspection it appeared to meet the criteria imposed. Local infor-
mation (the only type available) indicated that the flow rate and duration
would meet requirements. The headwall was located on the southwest-
ern edge of a 40’ x 40’ parcel facing East 77th Street and owned by The
Cleveland Electric Illuminating Company. Immediately west of the par-
cel is the right-of-way of the Penn Central Railroad. Property to the
south is owned by one Harry Rock. Property to the north is owned by
The Cleveland Electric Illuminating Company and on it is located a high
tension transmission line tower supporting a 240, 000 volt power line.
The project site was leased from The Cleveland Electric Illuminating
Company. Approvals were obtained from the engineering and sewer
departments of the City of Cleveland and from Penn Central Railroad.
This particular parcel is zoned tire sidential” and it was necessary to
apply for a variance before any construction work could take place.
After a hearing, this variance was granted.
10

-------
SECTION 5
FILTER DESIGN
The concept of a rapid-flow, high capacity filter for sewer overflows
is based upon the well-known practice and art of removal of suspended
and floating material from fluid streams by causing the fluid to pass
through a bed of particulate matter sized and graded so that the filter
medium presents only a nominal flow restriction while retaining the
solids on and within the filter medium. Since sewer overflows contain
suspended and floating matter in quantity, removal constitutes a water
pollution abatement measure.
The following design criteria were used:
1. The device was to be capable of removing visually objectionable
floating or suspended solids, such as cans, bottles, rubber or
plastic articles and the like. (Figure V and VI)
2. The rapid-flow filter was to be capable of sustaining a flow rate of
20 gallons per square foot per minute, and accepting a flow whose
rate might increase from zero to 20 gallons per square foot per
minute within 15 seconds. Flows exceeding approximately 75,000
GPH were to be diverted to the existing overflow.
The literature reports a variety of filter bed configurations
to accommodate water containing relatively finely divided
solids. These filters are capable of flow rates of two to
five gallons per square foot per minute. These are generally
graded sand, sand and coal, or sand and gravel filter beds
up to several feet of thickness with the particle size of the
media generally in the 14-60 U. S. Standard Mesh range.
Prior Rand Development Corporation experience with filter
beds comprised of materials in the 18 to 80 mesh range
indicated sanitary sewage flow rate for this finer range of
media to be in the 0. 5 to 1 gallon per square foot per min-
ute range at a pressure drop of approximately twelve feet.
Using the available literature data, prior experience and
a limited number of exploratory trial runs, extrapolation
indicated that a filter bed comprised of 3/4 inch lumps, and
larger, could accept a flow rate up to twenty times that of a
conventional water treatment bed, and still remove reason-
11

-------
ably small objects from sewer overflows. Items as small
as the filter-tip of a cigarette could be expected to be
removed.
3. Provision was to be made for by-passing and measuring flow in
excess of rated filter capacity. This requirement was imposed
in order that the amount of overflow reaching the filter could
be determined. This is not an engineering requirement for an
operating rapid—flow filter, but was considered desirable for
this one particular pilot unit. (See No. 4)
4. Provision was to be made for an overflow from the filter itself,
if the filter could not accommodate the rated flow because of
plugging or of characteristics of the test media or procedures,
provision was to be made to permit any flow the filter could not
pass simply to discharge over or around the filter device and
thence to the existing overflow outfall. In any such case the
sewer overflow would operate ultimately as originally construct—’
ed and remove little if any of the solids from the overflow until
maintenance could be provided.
5. Means were to be provided for measuring the volume of flow to,
through, and by-passed around the rapid-flow.filter, via
Parshall flumes. This was a design factor used in this installa-
tion solely for the purpose of experimentation. It is not required
in a working situation except as information might be desired in
specific cases.
6. Provision was to be included for at least one demonstrable
method for easy maintenance of the overflow device.
Ultimately, rapid-flow filters will presumably be main-
tained on some kind of regular service basis, similar to
the manner in which solid waste pick-ups are made. One
simple method of loading and unloading an overflow filter-
used in this program - is to use a winch mounted on a
vehicle, which vehicle can also transport “spent” filters
to disposal as well as place “fresh” filters into the over-
flow. Any of a variety of methods could of course be
used; local option will no doubt be taken.
7. The design was to be flexible in regard to experimental usage.
The following were considered:
12

-------
a. The filter device was to be capable of accepting filter
media from 3/4” nominal diameter upwards to perhaps
4” nominal diameter.
b. The filter device was to be capable of accommodating a
maximum filter medium bed depth of four feet.
c. Sheet metal construction was to be used to permit changes,
especially in the filter inlet, where dissipation of consid-
erable kinetic energy during major overflows was expected
to be required. It was indeed found necessary to make
these modifications to the initial design.
8. Specific design requirements of the site were also to be met, in
consideration of topography, sewer line locations and elevations.
These included:
a. Rescriction of the height of any metal structure to nine
feet, a requirement imposed in this location by the
proximity of the 240, 000 volt power line.
b. Strict electrical grounding of all structures, again because
of proximity to the power line.
c. Restriction of the device and all appurtenances to the forty
foot square plot.
d. Inclusion of safety items wherever indicated.
9. Two intangible design factors relating solely to an experimental
demonstration unit were also to be considered:
a. Public relations.
A degree of activity not normally associated with
sewers was expected to accompany the operation of
the overflow demonstration. Special care was indicated
to protect what would undoubtedly become an object of
local curiosity, as well as to insure that the installa-
tion would present a workmanlike appearance.
One aspect of the installation which could not be
avoided as a result of the need for flow measurement
and sampling was that of apparent large installation
size in comparison with the headwall runoff structure.
13

-------
The rapid-flow filter occupied an area of only about
fifty square feet while the flumes and equipment
shed to house the instruments nearly filled the
1,600 square foot site,
b. Public safety
As a measure of protection to residents of the area -
especially children - the installation was to be fenced
and designated by sign to be a Federally sponsored
site. The fence was later found to be of equal value
in protecting the site from the children. A working
unit would probably need only normal provision for
safety such as those made in overflow locations in
conventional practice and those used in solid waste
handling.
These criteria were then incorporated into a set of specifications,
which was approved by FWPCA. The Cleveland engineering firm of
Trygve Hoff and Associates was selected to complete the engineering
according to the specifications, with the selection approved by FWPCA
in September, 1966.
Final approval of the plans and construction drawings by FWPCA was
given in January, 1967, and construction was begun in February, 1967.
The test facility went on stream on schedule, May 8, 1967.
Figure VII is a flow diagram of the device.
The main portions of the pilot rapid-flow filter as installed consisted
of:
1. Main flume to receive the total sewer overflow.
2. Filter inlet flume off the main flume and headed by an adjustable
gate to limit flow to filter to about 1200 GPM and further designed
to conduct the sewage to the middle of the filter through a sec-
tion that can be swung away so that the filter can be removed.
3. Bypass flume off the main flume to receive the excess flow.
4. Twenty-five square foot filter, a four foot high sheet metal
hopper with an expanded metal false bottom for supporting the
filter medium or shallow baskets. A hinged sunshade was fitted
14

-------
over the filter to eliminate fetid odors found to be produced by
the sun’s action on the sewage solids.
5. Filter baskets; expanded metal containers to contain the coal
(see Figure V).
6. Filter pit used to hold the filter hopper.
7. Filter outlet flume used to conduct the filtrate to the receiving
waters.
8. Flow rate indicator and recorders to monitor the flows in the
main, bypass • and filter outlet flume s.
9. Conventional dipper type samplers to sample the main and filter
outlet flumes. These were activated by a moisture sensor
located in the filter outlet flume.
15

-------
SECTION 6
OPERATION AND MAINTENANCE
A rapid-flow sewage overflow filter is simple to operate and main-’
tam. The device simply rests at the terminus of an overflow
conduit, requiring attention only for replacement of filter elements
and for routine inspection.
However, the pilot rapid-flow filter was instrumented to provide
information; moreover, efforts were made in its operation to have
personnel present during natural overflow events so that visu’al
observations could supplement recorded data. When it became appar-
ent that naturally occurring overflow events were so infrequent
during the test period available to provide a large amount of data,
simulated overflow events were conducted, In these, sanitary sewage
was pumped from a reservoir formed by sandbagging the sanitary
sewer serving the drainage area and sometimes supplemented by
hydrant water.
A moisture sensor located in the main flume terminus initiated and
stopped the sampler devices, one of which was located in the inlet
Parshall flume and the other of which was located in the Par shall
flume metering the effluent of the filter baskets. Liquid samples
were composited in five gallon containers and taken to the analytical
laboratory as quickly- as practicable after any event, usually within
three hours. In those runs in which segment samples were taken,
fifty-five gallon rubber storage containers were used to obtain the
samples; these containers were then transported to the analytical
laboratory in toto.
Flow recorders located in the influent flume, its overflow flume, and
the filter effluent flume recorded flows continuously upon activation of
calibrated float arms. A recording rain gauge located at the pilot
plant site also operated continuously.
Operation of the pilot filter included sample collection, interpretation,
analyses of samples, and evaluation of filter medium size. A stand-
ard flat-bed truck equipped with a conventional hydraulic hoist was
used to transport empty filter baskets to nearby coal yards, carry the
filled baskets to the pilot site, and lower the baskets into place. The
truck also was used to deliver the sp?nt coal to a local incinerator
for disposal. All of the pilot operations were easily handled by one
operator; no special skills were required.
17

-------
In actual rapid-flow filter operations routine inspection would of
course be important, as it is with any sewer collection system
device. Replacement of the filter elements may be made on the
basis of such inspection, or on a predetermined schedule.
18

-------
SECTION 7
COSTS
The rapid-flow filter concept is quite simple, comparable in many
respects to industrial solid refuse pick-up operations in which full
trash-bins are automatically hoisted by means of self-propelled mobile
mechanisms and dumped into a vehicular carrier for removal to a
suitable disposal operation. The emptied, or a replacement, bin is
left for reuse.
In the case of a rapid-flow filter for sewers, a similar operation is
entirely feasible. For example, a vehicle carrying replacement filter
elements (containers pre-loaded with the filter medium) and equipped
with a hoist arrives at an overflow location. Its arrival is either pre-
scheduled, or triggered by the occurence of showers, or by an electri-
cal signal. On arrival, the filter element is inspected. If replace-
ment seems warranted or is scheduled, the used element is removed
and a fresh one installed. The single vehicle deals with both delivery
of fresh, and with removal for disposal of spent filter baskets for
devices.
Any estimate of filter construction and installation cost must of neces-
sity be generalized and so considered. Overflow structures vary with
respect to size and capacity, location and accessability, special topo-
graphical consideration, and so on. Some uncertainty always exists
when operating costs are estimated. In this case, the frequency of
replacement of filter elements, local labor costs, materials costs,
disposal costs, and the like all can be expected to vary from city to
city. A contingency of Z5 percent is recommended for consideration
in the following estimates which include construction cost, frequency
of replacement of filter element, and labor costs including disposal;
the estimates cover filters handling overflow sewers up to twenty-four
inch diameter. For the purpose of this estimate, overflow rates up to
10 MGD discharge are considered. High rate overflows can probably
best be handled by automatically cleaned bar screens: an estimate of
the point where bar screens would be more practical is approximately
10 MGD.
Basis of Estimate
1. Apparatus to be capable of accommodating twenty gallons per
minute per square foot of filter surface.
2. Apparatus to consist only of filter element with a simple appara-
tus to direct overflows to it. That is, no analytical instrumenta-
tion or special equipment other than sun shade and elements of
safety.
19

-------
3. Filter element replacement to average six times yearly.
4. Disposal costs to be equivalent to solid refuse incineration dis-
posal costs per load, or $5. 00/truck load.
5. Spent filter pickup and exchange cost to be $35. 00 per station
pickup for the first basket plus $5. 00 for each additional basket.
6. Filter basket loading costs to be $5. 00 per basket.
Filter baskets 6 feet wide by 8-1/2 feet long and 2 feet deep weighing
approximately 2, 500 pounds loaded would be a practical size to handle
using standard trash bin handling trucks. This size filter basket could
handle 1, 000 to 2, 500 GPM. The sewer overflow discharge would be
diverted to the filter baskets through simple distribution channels that
would distribute the flow over the baskets. This was the method used
in the pilot project.
An improved distribution method might have the overflow water distri-
bution along one endof a basket which would handle 1,000 to 2,500
GPM. Excess flow would cascade off the opposite end of the filter bas-
ket. Additional filter baskets could be used in series or paralleled.
Approximately six baskets would be needed for a 10 MCD flow rate.
The baskets would be of a straight sided steel plate fabrication with an
expanded metal bottom. Internal baffles would minimize displacement
of the filter medium due to hydraulic action.
The rapid flow filter structure would consist of:
1. Filter baskets to contain the coarse filter medium.
2. Distribution channels or weirs, etc. to conduct the overflow
from the overflow sewer to the surface of the baskets.
3. Pit or excavation and walled area to contain the filter.
4. Security fencing.
5. Access roadway for servicing the filter.
Following are estimated costs for components of a representative rapid-
flow filter installation assuming the construction of several:
1. Filter baskets, each $ 350..00
20

-------
$ 8,
2,
$ 3,150.00
1,000.00
14, 000. 00
750. 00
2, 600,00
$21, 500.00
2. Distribution channel, per basket $ 500. 00
3. Pit (excavation and concrete work), per $ 250. 00
cubic yard.
4. Security fencing, per foot. $ 5. 00
5. Roadway Variable
6. Surveyandengineering $ 1,500.00
7. Contingency 25%
The estimated costs for a minimum sized filter (1.5 MGD)would be
1. Filter baskets - 1-1/2 needed $ 500.00
2. Distribution channel, one needed 500.00
3. Pit, 12 cubic yards 3,000.00
4. Fencing, 100 feet 500. 00
5. Roadway, average 2,000.00
6. Engineering 1,500.00
Sub-Total 000.00
7. Contingency 000.00
$10,000.00
Total
The estimated costs for a 10 MGD filter would b’ e:
1. Filter baskets, 9 needed
2. Distribution channel, 2 needed
3. Pit, 56 ctibic yards
4. Fencing, 150 feet
5. Roadway, average /
Sub- Total
21

-------
Sub- Total
Total
$21, 500.00
1, 500. 00
$23, 000.00
6, 000. 00
$29, 000.00
Dollars
Cost
Thou-
sands
32—
28—
24-
20-
16-
12—
8—
4-
I I I I I I I I I
1 2 3 4 5 6 7 8 9
MGD
Figure VIII. Construction Cost, Estimated
Annual operational costs for the minimum sized filter if the filter
medium is incinerated:
1. Pickup $
2. Coal
10
210. 00
75. 00
Sub-Total $ 285.00
6. Engineering
7. Contingency
22

-------
$ 285.00
3. Loading 30.00
4. Incineration disposal 5.00
Total $ 320.00
Annual operational costs for the 10 MGD filter would be:
1. Pickup $ 360.00
2. Coal 450. 00
3. Loading 200.00
4. Incineration disposal 30.00
Total $ 1, 050. 00
The above costs would be lower if the spent coal were burnt in a con-
ventional stoker being used for steam generation. This would probably
be satisfactory if oversized items were removed. Taking a credit for
the coal and estimating the disposal costs above would give operating
cost of $240. 00 and $570. 00 respectively.
Annual
Operation
Costs
Dollars
1000—
900—
800—
700
600—
500
400—
300—
ZOO—I
1001. 110
MGD
F’igure IX. Yearly Operating Costs, Estimated
23

-------
SECTION 8
RESULTS AND DISCUSSION OF RESULTS
Table I is a complete tabulation of analytical and rainfall data with
filter characteristics, representing 55 natural and 36 simulated over-
flow events. The data in Tables II through VIII are abstracted from
Table I. These latter tables group similar events and sampling tech-
niques for clarity in analyzing the results. Mixed events and incom-
plete sample events have been excluded from these latter tables.
Description of Natural Events
Table [ I summarizes pertinent recorded data obtained during those
storm events which caused overflows to the pilot filter during the
investigative period.
Accumulative rainfall was recorded on a seven-day time chart while
flow rates were automatically recorded on 24 hour time charts.
Total overflows were calculated by integration from the flow rate-
time records. Attempts to correlate total recorded rainfall and rain-
fall intensity with overflow rates or overflow totals were unsuccessful.
There were far fewer storms of sufficient duration or intensity to
cause overflows at the pilot plant site than were expected on the basis
of normal rainfall expectations for the region. Visual observations
made during many of the storms which did occur revealed that during
the first ten to twenty minutes of storm overflow gross solids are
numerous and identifiable as material being scoured from the combined
sewer itself as well as solids being transported as an immediate result
of the rain. Subsequent to this period, the overflow contains few
solids and in general lost its characteristic sewage coloration.
As a pciint of interest, there was a delay of almost precisely 12 minutes
from the start of a sharp shower or storm and the corresponding over-
flow at this particular overflow installation.
Conventional Sampling
Tables III and V list the percent reductions of suspended solids and
settleable solids in the natural and simulated events as indicated by
conventional sampling and analytical techniques. In general the results
indicate either no reduction, or an apparent increase. This behavior
would not be reasonable unless the filter was adding solids to the filtrate.
25

-------
The unexpected lac of apparent solids reduction in runs monitored
by conventional sampling techniques can have several logical explan-
ations, including:
1. Fines from the filter medium being washed out of the bed into the
effluent. Experi nce has indicated that this can happen during the
initial use of a bed unless it has been hydraulically dedusted.
Except during the shakedown portion of the project, filter bed
washing was practiced. The effect of no or incomplete washing
would be most pronounced during the initial use of a bed, but
Table VIII indicates that the results from initial events on a given
filter bed were not significantly different from succeeding ones.
2. Filter medium spalling. Visual and microscopic examination of
the solids in the samples did not reveal coal fines in the effluent
from washed filter beds.
3. Reslurrying of collected solids. Once the solids have been re-
moved from the overflow stream and collected on and within the
filter bed it is possible that they could be returned to the filtrate
if the filter medium were disturbed and displaced by the force of
the stream, or if the collected solids were degraded so that they
broke down into pieces small enough to wash through the medium.
After the initial shakedown period, when it was found necessary to
install a baffle plate to break the force of the inlet stream, observ-
ation of the filter medium failed to indicate sufficient disturbance
of the bed to cause wash-through.
Degradation of the collected solids with time was an observed fact,
although no quantitative measure could be made. Organic mater-
ials - leaves, paper, fecal matter, etc. - were observed to di-
minish greatly in volume, or disappear completely within a few
days. This probably was caused by composting, a natural biolog-
ical decomposition which occurs under moist aerobic conditions.
Possibly some chemicals from the coal helped but this effect is
speculative. At no time was an appreciable odor from the entrapped
and shaded solids ever detected except during the one instance that
oiled coal was used. The oil may have isolated the coal surface
from the sewage solids, and thus diminished the known ability of
coals to adsorb odor. (‘)
26

-------
The actual mechanisms involved in the reduction of the collected
solids could not be investigated, but the fact that the solids did
tend to ‘disappear” was incontrovertibly observed. It is reason-
able to assume that some of the degradation products would have
appeared in the effluent. Initial runs through a new coal bed did
not demonstrate a significantly greater solids reduction, however,
which fact diminishes the strength of this argument.
4. Faulty analytical technique. Procedures given in “Standard
Methods for the Examination of Water and Wastewater” were used
by experienced analytical personnel, and this argument has no
validity, at least as it relates to the correct performance of con-
ventional analytical techniques.
5. Non-representative sampling. It was evident merely from visual
inspection of the operation test filter, performance was not being
accurately characterized by the analytical samples. The influent
and effluent scoop type samplers were of a standard type used in
waste-treatment systems, and were similar except that the former
was larger and operated less frequently than the latter. The influ-
ent and effluent channels differed in geometry, with differing
stream velocities, turbulence, etc. The influent channel was
necessarily- large to handle the peak overflows, so that the normal
influent overflow stream was relatively shallow. A well was
therefore required for the sampler; however, the well tended,
in service, to act also as a settling basin. This fact, plus the ten-
dency of the automatic sensor to keep the sampler operating for
some time after cessation of the overflow, resulted in an influent
sample containing proportionately fewer solids than the stream
being sampled. It was also visually apparent that classification
by density was taking place in the flumes, with heavy solids moving
along the bottom beneath the reach of the scoops, so that an add-
itional error was introduced into the technique.
Grab Sampling
Some of the data pertaining to natural and simulated events monitored
by grab samples are tabulated in Tables IV and VI. In general these
grab samples were taken by compositing four one-quart dipper-fulls
taken from different points across a section of the stream. Grab samp-
ling is subject to lack of consistency in technique, but it does permit
larger sample increments to be taken and also permits the use of judge-
ment by the operator. The grab samples taken in this work were there-
27

-------
fore considered to be more representative of the solid portion of the
stream. The inaccuracy of the scoop samp er, the well in the flume
and the segregation of heavy materials were not factors with the
grab sampling technique.
Based on the analysis of grab samples taken in seven of the nine natural
events, the expected reduction of suspended and settleable solids did
take place. For Events No. 53 and 63, in which the greatest flow rates
were measured, an addition of solids to the effluent was indicated. On
the other hand, the data for Events No. 15 and 80, in which low flow
rates were measured, abnormally high suspended solids removals were
attained. Evidently a high-flow turbulent filtration can wash out fine
solids which accumulate during less turbulent - or nonturbulent - fil-
tration of low rate overflows. The rapid-flow filter concept was not
designed for the removal of fine solids, but it was observed that dirt,
sand and fine organic solids did sometimes accumulate in the filter.
Grab sampling, and also the standard analytical procedures, still
suffer from a basic problem of being unable reliably to sample and
measure gross solids.
Initial Filter Bed Tests
Results of all runs are listed in the tables; however, careful discrimi-
nation is required in the interpretation of numerical data obtained dur-
ing the early tests. These runs included, to various degrees, shake-
down of the test equipment and analytical procedures, and of means of
communicating the existence of an overflow event from the test site to
the office. For the preparation of this report personal judgement has
been used where conclusions are drawn from the early runs.
Stratified Filter Media Experiments
The usefulness, in sewage overflow filtration, of filters composed of
layers of media of progressively smaller size ranges was examined.
The purpose of this group of experiments, reported in Table VII, was
to see if the additional cost of layered filter beds was justifiable in
terms of solids removals.
Tests were run in a 9 square foot filter set in the place of the filter
baskets in the pilot installation. The flow used was all sanitary sewage.
Z8

-------
The media consisted, top to bottom, of approximately 4 inches of
1—1/2” x 2”, 4 inches of 3/4” x 1—1/2”, and 4 inches of 1/4” x 3/4”,
washed coal except for Event No. 76 which did not use the smaller
sized coal layer.
The average suspended solids reduction using the stratified beds was
11 percent while the settleable solids reduction averaged 31 percent,
found using conventional analytical means. Similar runs using single
medium filter beds in the larger filter baskets gave average suspended
solids reduction of 11 percent and an average settleable solids reduc-
tion of 30 percent. (see Table VI) Therefore, it was concluded that
there is no discernable advantage found within the limits of this experi-
mentation to justify the added expense of preparing a stratified or
graded filter bed for the purpose of removing gross solids.
Experimental Filter Screen Sampling
During three separate simulated overflows wire mesh screens (Figure
X) were inserted into the influent and effluent streams for given times,
varying from one to four minutes depending upon the screen size used,
in such a manner that the total flow had to pass through the screen.
The screens were then air dried to constant weight and the increase in
weight noted. See Table IX for the results of the experiments with
this sampling technique.
The general approach indicated an improvement over conventional
sampling but suffered from two obstacles. The deposition of solids
upon the screens impeded the flow so that the total flow screened was
somewhat variable as a consequence. The screens were also unable
to retain large solids, such as rocks, cans, etc., that would not embed
in the screen and so be sampled.
Small scale filter media evaluation tests using raw sanitary sewage
were run to determine the ability of coal to remove solids. Nylon tulle
net bags with 1/16” holes were used to filter 1,000 gallons each of the
full influent and effluent streams. The collected solids were dried and
weighed (Table XII).
This procedure did collect all of the gross solids but relatively few of
the fine solids.
29

-------
Total Segment Sampling Experiments
Large samples were taken of simulated overflow runs by diverting the
entire influent stream into a 55-gallon drum and then diverting the
entire effluent stream into a second drum for the period of time
required to fill the drums. The samples were then filtered through a
series of screens, U.S. Standard Mesh Sizes 16, 30, and 35. The
solids were then air dried to a constant weight. A consistent reduction
was noted (Table X).
This approach is believed to represent the beginning of an improved
sampling procedure.
Filter Capacity
Tests were conducted to determine the probable life expectancy of a
filter medium. These runs were made using a 1. 1 square-foot filter
with an 18-inch depth of coal filter medium. In most cases the bench
tests were run at the very high throughput of 125 GPM/ft 2 of filter area,
using raw sanitary sewage. The criterion for calling a filter bed
plugged was when the bed pressure drop reached one foot of water
(Table XI).
For coal size consists ranging from three-fourths to two inches in
diameter the bench scale filter capacity per square foot ranged from
16, 300 to 49, 000 gallons before pluggage was attained.
Capacities of the pilot filters ranged from 2,200 to 8,100 gallons per
square foot. The pilot filters were changed whenever sufficient back
pressure to have caused the filter to overflow at rated capacity was
noted. A more economical arrangement would be to permit approxi-
mately one foot or more of pressure drop before considering the filter
to be exhausted.
Filter Media Effectiveness Experiments
In any filter design a compromise between effectiveness (in this instance,
removal of gross solids) and flow rate must be made. Capability of
sustaining a high flow rate is of considerable importance in the concept
of a filter for sewer overflows. Cost is always a primary variable.
The effectiveness of filter media comprised of various coal consists in
removing solidS were directly evaluated, and filter life and anticipated
30

-------
costs were judged in a series of experiments summarized in Table XII.
A 1. 1 ft 2 filter was used in these experiments. Coal particulate sizes
were varied from 3/4” minimum to 4” maximum. It was found that
solids removals in the order of 50 percent might be expected from coal
consists in which the minimum size was 3/4” and it was judged that a
satisfactory flow rate could be sustained. Coal sized between 3 /4”
and 1-1/4” is commercially available and except for having to be washed
is directly useful in a rapid—flow filter. This size range was most com-
monly used in the pilot runs to accommodate the requirements of cost,
effectiveness, and flow capacity.
31

-------
SECTION 9
DISCUSSION
The rapid-flow filter concept was evaluated at one location in Cleve-
land, Ohio. Sections of this report describe the pilot design assump-
tions, the site selected, data obtained, and problems encountered with
conventional sampling methods. A recommendation for improved
sewage sampling is included. Costs of construction and operation are
estimated for overflows to 10 MGD.
The most important overall findings, however, can be quite simply
stated and are supported more by observation than by the data returned
via conventional sampling. A rapid-flow filter for combined sewer
overflows will remove gross objects from large volume overflows. It
will operate on demand, without any need for labor except as occasional
or routine maintenance. If the removed sewage solids are protected
from direct sun, and if the filter medium is coarse coal, no objection-
able odor is associated with the operation. No flies congregate, no
nuisance factors are observed. From visual observation, it is believed
that aerobic digestion of organic materials occurs; if this be so, decom-
posable organics would be reduced in their ultimate demand for oxygen
in surface waters when the residues reach those waters.
The filter performance was not affected adversely by any extremes of
weather, including severe cold. If, on occasion, the filter bed does
become plugged or is otherwise unable to pass the entire overflow (or
is not in place) the situation merely reverts to current overflow prac-
tice. While such diversion of overflows directly to surface receiving
waters would be unfortunate in that no solids would be removed during
such period, it would be a temporary return to conventional practice.
A filter bed of 3/4” x 1-1/4” nominal diameter media has been shown
to have an extended life, extending for many months. In many situa-
tions, replacement every several months would be entirely adequate.
Naturally, any filter bed life is a function of flow and solids contained
therein, if any of the solids are captured. It is visualized that rapid-
flow overflow filters can be maintained either on some routine basis or
upon visual inspection. Inspection readily reveals whether a filter bed
is plugged with non-decomposable solids, i.e. cans, bottles, plastic
items, and should be replaced. Evidence that the filter device itself
had overflowed would suggest replacement; this information could be
telemetered if desired.
33

-------
There appear to be two substantive limitations to the rapid-flow
filter concept for removing gross solids from combined sewer over-
flows. One consideration is that large diameter overflow situations
will require so large a filter bed area as to become prohibitive in
terms of available land, cost of construction or of servicing. The
second consideration is that the data obtained in this work are not
conclusive in showing numerically that pollution abatement is served
by filtering combined sewer overflows. Conventional sewage sampl-
ing techniques were not adequate to make the numerical evidence
available. However, visual observation and experimental sample
gathering and analytical techniques did make it unequivocal that
significant and useful quantities of solid materials are actually removed
from overflows. A recommendation for an improved sewage sampling
procedure is given in this report. The problem of sampling liquid
streams containing solids of a variety of sizes and densities has been
resolved in certain other fields of endeavor, and the recommendation
draws upon that experience.
It should be noted that if one wished to remove gross solids from over-
flows, such existing equipment as bar screens may be employed.
They are dependable and relatively compact, and in a sense competi-
tive to filters. It should be remembered, however, that operations
of conventional bar screens would require some positive and constant
means of dealing with the decomposable solids removed from the
overflow streams to prevent unpleasant odors and insect infestations.
In the coal filter, odor and insect problems did not arise. Further,
in the case of mechanically cleaned screens auxiliary electric power
should be provided in the event of power failure during a storm,
which is precisely when the unit will be expected to operate. A design
engineer will wish to consider all aspects of the several choices open
to him.
34

-------
SECTION 10
SAMPLING
“ Standard Methods for the Examination of Water and Wastewater ”
is the accepted authority in the waste water field with regard to the
physical and chemical examination of natural and treated waters.
In this text emphasis is placed on the analytical procedures, rather
than on sampling methods.
WPCF Manual of Practice No. 11 “Operation of Wastewater Treat-
ment Plants” points out the importance of obtaining representative
samples, and indicates further that in dealing with sewage this is
a difficult task. This manual also states: “Nearly everyone agrees
that laboratory analyses have little value or meaning if the material
analyzed is not fairly representative of the conditions or quality
which actually prevails.”
Difficulty in obtaining representative samples is increased when one
is especially concerned with raw sewage solids, and still further
when one is concerned with gross solids. The WPCF Manual on
Operations recommends that raw sewage samples “should be collected
preferably alter the waste has passed through screening and grit
removing facilities.
As a result there does not now exist recommended sample collecting
equipment or procedures for sampling sewage flows containing gross
solids.
That sampling difficulties exist for sewage overflows was recognized
by the contractor, his engineering consultants, and the sponsoring
Administration. A number of approaches to the sampling problem
other than conventional means were carefully considered in the
original pilot design. Finally, however, conventional dipper -type
sewage samplers and conventional analytical methods were selected.
This decision was based on a strong recommendation by FWPCA
that unless conventional equipment a.nd procedures were used, the
results derived might not be acceptable to the sanitary engineering
field.
The present work encountered the expected difficulties frequently
experienced in obtaining representative raw sewage or combined
overflow samples. Conventional samplers were found not to provide

-------
representative samples of grossly polluted raw sewage overflow
streams. Although it was not anticipated in the contract objective,
one of the most meaningful conclusions that can be drawn from this
program is the fact that sampling techniques for sewages can be im-
proved with respect to flows contaminated with gross solids. Rep-
resentative samples of such complex flows can be taken, and it is
believed that use of improved sampling techniques would be of con-
siderable benefit both to waste water treatment operators and also
to investigators and researchers in the sanitary engineering field.
The tables of analytical results contained in the appendix of this report
summarize the laboratory results obtained in this work and clearly
point up the unsuitability of using conventional sewage sampling de-
vices and analytical procedures for sewage overflows. Both are
unsuitable for providing meaningful analyses of streams containing
such contaminates as bottles, leaves, cans, fecal matter, contra-
ceptives, toilet tissue, sanitary napkins, plastic articles, and the like
that are found in quantity in such streams. Therefore, it must be
emphasized that analyses of influent samples taken during this work
have little validity except in those few instances in which segments of
the total flow were taken. Moreover, standard analytical procedures
fall short of numerically identifying these gross contaminants; if
they are to be identified, new procedures must be defined. By ob-
servation, gross solids were invariably filtered from the overflow,
yet conventional sewage sampling and analysis consistently failed
to reveal that fact accurately.
An improved procedure for taking samples of liquid flows containing
gross solids is to take the entire flow of the stream part of the time.
If the flow rate is too great to permit this, the stream can be split in
such a way that a side stream is formed which is representative of the
entire stream. All of that side stream can then be taken part of the
time as the sample. Periodic sampling by this method to accumulate
a composite can also be practiced, where its use is indicated. In
sampling combined or storm sewer overflows, where the composition
varies rapidly - especially during the first portion of overflow - fre-
quent samples would be necessary to provide a valid representation
of the contents of the stream.
The proposed general sample collecting and analytical procedure is
cumbersome. Sample volumes become large in order for the analyses
to be accurate, but they must be large because of the large and varying
sizes of the particles to be analyzed for. Also the design of the side-
stream piping must be considered carefully so that the side-stream is
representative of the main stream. Nevertheless, a generalized sew-
36

-------
age or overflow sampling technique can be recommended which, it is
believed, will represent an improvement over conventional sampling
methods.
Sample taking and analytical procedures for determining gross solids
in liquid streams:
1. Take all of the flow, part of the time. When the flow rate
is large, exceeding 1000 GPM, split the stream so that
the side sampling stream is about 1000 GPM and repre-
sentative of the main stream. Divert the flow to a
corrosion resistant deep-coned sample tank for a preselected
interval between 15-60 seconds. The diversion gate must
be quick acting and positive.
2. Filter the sample promptly through a filter screen after
noting the volume. Retain the solids. A 60 mesh stainless
steel screen is tentatively recommended.
3. Sample the filtrate conventionally as it is released at a
steady flow rate from the sample tank.
4. Repeat the above three steps periodically to obtain a repre-
sentation composite of the total flow.
Dry and weigh the solids and calculate the ppm of the gross
solids in the stream. Identify the solids if desired.
5. Analyze the filtrate for suspended solids and other contami-
nants conventionally.
The above must be considered as simply a preliminary suggestion.
Improvements undoubtedly will suggest themselves in practice.
37

-------
TABLE I
Rapid 01. flirar Co.ablad bor O.ooliaw 1’roa i d 1.530... aS stall.
O.n.alalnI TabalaSo. at NaM.al aS BaSalad LoaM.
—
L ast
liM.kr
Na* Obata-
nl 1*0.4
f l ’ ..
fall
r Io.
5 . Ia .
faU
rat.
14pj
FIlllr
j
Si .. of
Cal FlI r ar
Madle
TS1
Fllt.rad
IW o
J753L
Mate
O.orIIae

Acc -
lalira
TM.l
blat
Typa
at
La...I.
•If.74 3oiM.
%
Lads.-
S t.
Valawa
5a.p.S.d
Solid.
S a
16*2
I.
S
VaIad*
16*3
Solid .

IILJ
.mi#bl!M.rarIol
B
%.daa.
S . a
n—i
, at.
r
Nly—. ,,
r
— —
c..
BOG •j10 fl”an
r a S S
MPN
r
ceasa
‘ iafl ht.redbaa
—
—
1.1’
149
43
I 6.16.67
-
z 640 bra
a
‘ 0 . PEE
..aalt
. 14 K
Goab.Ial
IL?
0 P
-IL
.5?
.1±3
.22?
S
—
— —
a
Oil
J
7100
_3
Jfl
Grab-I NO.
46±3
.s.
‘ IL ?
at
s
.Js
ia
ii
2.4 ’
si
2.19
2.1
t.s t’
- -
L,

s
•
IS
°“
44,1
2.1 1
“
L_.
“
“
—
2.222..
12.2K
Gnb.i —
prab.iao
4080
no
0860
an.
I I
io
a
j p

—
—
—
°“
•
24’
LaS
2.1’

24
Li
—
—
—
t i _f ±3
— —
——
Z 1 1M_0
17 .9 -67
•
010
jjfr
“
“
Ja
070
21421
N/A
—
—
67-25-67
a
all
fla
“
j
15421
N/A
—
—
—
—
7 74547
a
!f r
3/4”al-l/4”
j 42f
n .M.
N /A
—
—
—
—
—
—
—
— ——-
—
keiM.redlaa
•
.211
.11°
.1_.
‘
..SP
. &.
4j4
N/A
— — — -
•
.21!
±1 °
.L_.
°
.S!4’
._J!L
44j
N/A
—
—
IN 0.9.67
•
030
100
‘
“
J4
304
fi
N/A
—
—
3 1 It”
AU
Baa -
. .
°
j
J
Grab
ILL?
.1±2. ?
5
k _ fL !
I L?
S
. W
iK!
IILI
—
%tffi
111 1
TILt
fl41
4444
— —
122.1
— .
- r
—
14
10 1-30.6’
lalaZalpl70
NaaIISy
501
III
Banal’
3/4 ”. 1-1/4’
iS
....J1L
.1421
IS -AM.
Eli -Grab
Ά2±3
.at
.2±.
S t
223
S
S
ft!
‘ S
jfl
110
02
2.10
flo
flo
4J0
aia
mat I L
—
if L
—
-
1n012*raiadle
J1_S±±’
—
....:: _
‘
. 12 5
4La
N/A
—
—
.__JJ
jj_ J J4°
•
jJj
1400
°
‘
i
JL.
A la
aSa
inj
....± .i
a.
a’
ALt
s’
s
122.1
121.
—
.LIP
.2.1 ’
—S
s it
uar
sa p
0
Inc
— —
ma. c..
— .
• ‘3 t
—
Ά1
..1L. SIk4 7
I I)
£24.
54’
“
J44’
QjQ
Grab
2444
lao
JJ
5
Sr
IL
1.11
fl
04
3 4’
I
j p
42.1. ‘ti_
J I I
..JL_JAL±’
—
—
—
0
425
66.700
AM.
160
J23
L I )
50
S t
. 560
000
9 44J
55
—
—
—
a
. 14.122
1Q44’
AM.
j)_)’
...53
±l_ 9
± 4 3
J44
. 11?
ILLS
Ά11.’
—
— —
— —
— .
— -
—
—
30 10-64’
!L±± 7
a
a
(2)
242..
jjf
°
“
°
37.740
45
430
Ά5W4’
Ά± JK
AM.
IS -AM.
L U-N/A
223
300
764
0300
!4.
2±3
S I
S I
0323
0050
11 1.-I
54’
—
a
.& _
— —
...t.
— -
0
—
10.15.67
a
Ba n.
2° 04’
5
oo
AM.
t143
j,154
0
flj
5
1050
5
—
—
SaIllNrM.dle
ii 16. 16-67
a
Oil
025
“
Q
AM.
. 2±3
,.j43
.59
343
j7f
12.
247
44 fl
‘1042
S I ?
.240
.11.2
— —
._4IJ t.
— -
“ e g
—
AM.
-
—
22 10.17.67
a
045
200
°
a
5Jg4
425
Oa io0
.323
2 1 . 9
Jj O
IL.
SM ’
52?
.22 4
J — -
—
—
.41.. ItS ”
24 11-1-67
•
a
J_
Boor
0a4”
. 11 . 222
444.
4244’
42440
AM.
AM.
43
1 S t
. 122..’

is)
U±i
*2.’
64o
at
0670
I )’
3%
s
441
LL
j
a
4r
‘12?
0
1342
l O I S
—
—
.
—
— 6_
Jrj
043
25 1141-67
a
J4p
“
“
00
II
43440
AM.
1160
0000
143)
O
j4.0
J15
440
4jf
%ftr
—
—
ji
5!.
— Li
L I
—
LL12 4
•
.
“
a 1 .m
544.o
AM.
100
719
Il
57
460
703
040
1410
.ss
a
S.
‘L I ?
.2.22
—L
ti.
—
27 12-5-67
a
•
..
SI opa
550
154440
AM.
579
709
143)
jL
J r
44%
l 4j
li ii.
100
jj
1700
342
— L
U
—
20 12.12-67
a
BaW l
I ” aO”
42435
540
42435
Grab
300
4 ± 3
2.41
5
2±3
1?
.42’
‘ 12.!
U I?
n”
sa
...s.
‘a
‘a
—
—
—
— L
I 6 -
—
—
a ma edt
izio.r
30 22.4947
ii 12-20-67
a
—
a
4J3
1100
“
“
a
•.
1 1.22’
42430
425
43450
Grab
Grab
300
460
43J
350
.LL!1
00
11
16
.JL.°
000
.22’
410
41’
144
‘ J
650
Ά41.?
1100
fll

j
2.1’
..4..
.JL
011!
21!
7244
54’
—
— LI
— U
‘4 .
L i
—
‘
..
90
435
AM.
190
1430
L ID
06
fl•O
.J3
4.!
±4.?
1±43
030
21’
£52.
344
±1!
.23
1.12
ILL
ILL II
4
—
32 1-00—60
a
000
1000
‘
a
2400
30
3 2. 190
N/A
AM.
I
I
I
I
41
I
I,oo
-
—
—
55 1.09—60
a
000
1100
•
“
1 .000
40
15.194 04ff Iy)
‘ l i i
070
1420
244
60
39

-------
TflhL I d l
— — c d 11 O 1 I—— Ca- . .41
a ___ l _ J ___ l — a ., — l .d C m l.
I. . 4 4
- -
UI .. 1.44
41.71- 11. 1.11 711.
‘. ‘ 1. 4 -
1111.
01W
4.. .4
U.l 0111.
- . -
W1 U
11• .Cm
• !•‘
-
1.1W
T .1

T
.4
W
-
- - -- .
V. 1.44.
U.p.W.d
. . . . . 3
- - .
3044
I l_
. .
7 I . 1 W
7.44
..fr” •
-
. l .l
— — j—. j
j 4 044.
. . . — -
44I1 1331S”411
.
IL..
.
..

s,.
-•
.
JJ
. 2
—
FIIml
.
L±M
— —
——
—iL .L±±!
——
a
1113
I P
—-
——
——
.&J_
——
3I4I
a-
L’
L .& ’__
I OU
1
!2.
7l
3r —
!

—-
. .U. II• ..IIT
.11 .
s I . -7 4 4 1 W .7U

30 1-11.11
7
— A U
.... _ ..
•_j
174.140
a-
OTIj fj ZlIflll . 4 4 IIl 7 S4 7 lSII44
——
!o
L1
.1!_±
01
—
.12.. 1.40.11
—

44.
.
U.
441W
rE’
. !.
Ill” • I_Il l”
ilP o 1 .111
04
I I
L!!!
.
C..0
1111
4 7
jftf

•j
1441
.!L 3
II1
—

—
— —
—
1lI
j
—
—
4.4.41
‘
.t
44 1.41.11
— - - r
I
.!.Lk°
• .7l -I0
•
•
•
p

1
-
41
$44
ooo
Ill

L
L I!

Ill
I a

I.iJ
1I$
44.1.4
2.2:_
k2

lL 33.j
I..l
•
IlPo 1.114
-
. 14 1 1
•
S
.2.1
J
1.111
II I
1.111
Ill


.. .. . 21
12L
14.110
4.411
14.111


.2.1 !
a-
a-
£110
a-
a-
7.440-a-
.1*
3300
4411
I ,I

, VI1j_l___ l_
71II J 11114100
J! l4 ’ Ills
7 114$ 000
JIl I j 711 4317

J . 2L’AL°__
—
61 4.441
I II 030
144 337
jglIJ_
— —
D l j44
I II ‘!
103 1117
—
7.
‘3 44
411
(1W ii ’
——
— — — —
‘
.


110
. !
177 440
j j
. __
— —
— —
J ’ . ! - — —
- ‘.2 ‘ .. — - — —1—. -
III i ‘ — - — - s - i
j J ‘J ml - — —
. .La±!’JJ . . . 1 _. .._ - -
!i . I . 3 _ I
— — - — — —
. ___•_ _
— .! -
—- -
- . -
— - —
..L ! - — - — —
- .
— — - -
2 1 — — ‘!I1 - -
J J — .2 -
.I?J t — — — —
—. — — -
--
4.30.4 4
..a.. -‘-4
•
.
Ill
,4 .1

•
r
114471.1133

Iκ.N
044
j
I4.
a-

Cm
U I
-
ITS
I I I

. j
22_f
‘
1
I1J
- -—-
LI


•
c
.. .II_ J
.i .L12
•
•
L
1.1
‘
$43
21440
“ 4 ’-’r
,,..: ,. , .
•- I’I - 02-
J . 1
11.1
. .J .
111
i.
llJ
—
04 . A
‘22.’
300
.22.’
n
.2 1..
“
g7.
22J
. 0.1
A
.
Jr
r
iLl
.i
•ikI
.i
L!
L2
04
L!
.44
!il

J
24
— —
— — —
II P .I 222J.
..221.
V LIt .. J .
— — —
J IH 2
•
Lfl
-kM
b 2
1144. 1.1 144
J
j
.21!_’
.11.
•
•4 f
JIf
• ir
iLs
a.i
. . .JL .
._IL_
•
IS)
1_
L I!
.I 2
2LM
•
._ .....
•
.0.1
LII
.ji22L .
.._222_
- jj

a
Cm
MO
‘
047
.._22_
13
.I!..
.0_f
J.?
JIf
.1.0
..W
ff0_f
TLI
21.1
OP
i..r
LIP
LI?
I I
..E.
14
•
0_
L22
_
•
lola
•
2. 2&3 . I I_ _
•
.L1
.
I! .m
21J44
a-
f44 .f
j21f44J 1 IIS
——
.22.J. J......
.
. -..
-
- ._ —.
.t .2th9
II
ITS
.2 12
—i - -
. 2 . 1
.._JIL.
a-
I1
r
44_f
.1 . 0
.lfr
12_f
3 1 . 1
£2
Iii
.12_ ..0 .2. . 0

•
L . 0
.fl
.__
fl
•
.
—II
j
._.!.0_
iL.1E
U.—
lola
33.
131•
.il .
f
I L?
. 143
.144 ’
30 . 1
131_f
I I
—
13$
—
33
—
111
—
J
—
— -
—
—3 —
J
— -
—
—
—
u..lIII10 ..11 I
11
44 1.17.11 -
-
44
• I 4.5 1

........L
lIP. 1l -
II. s ,o—. z.,a—
Cm 1-I .j -
3 14
—
1W o
a-
“
21 . . ’
s - ’—
‘
130J
- -
.1L
—•

ILl
-‘-p
IL?
.144
JV
...
ILl
‘
!
—
i.!

—
. 2 . 2 ?
‘
.J&.
j ,
— - ‘- ° °
L2
‘
1.1


—
.11
•

—
—
. . .
. -t - j

—
— •
.
—
—
—
—
—
‘ ‘
I,
40

-------
TAILS I (C .5 d1
k.pld Il.. 71109 , C00.bth.d !4r ..fl.. T0 .4U .n5 CU? 1014n. d R...2 .
U?IO00001 Tlb414006 .3 NU?42 .44 5044424 E.. .t.
— —
0001
3 - 30
D ” ” l
PU?- 00 -
.11 IU?g
L0I.
0.36-
3611
o)8
L. ..
0 .3.-
1.11
T1 .

FII4

I T 431 1 A44 1-
036. .1 I 741e . . U? 1.44.
C. .l 7110., I 140. 0000. 1481
34.43 . 2 . J_&AA3.
‘o
.2
.-1.
06 . . .U?d 0.144.
04144
0.0.0- 0.34.
36 03.. I
TU?I
1.334.
I v44.02. 0.41.401. 1364.1.1
I 148 1 V. 14o 1
J 0.114. 0.444 •—•0’—

. .li&2 70A.

L_.. ..
—

1004
,
—b— .3 — . . ..
.
LL
. .Z
3 1
3/4 1.1/4 ’
..... JAf
._..1 L
...LiAi
. .lv
070
4J
•4 4 J
046
— —
330
00
— 6
0 —
- —
—
—
H ... .2: !:40
JL. 1 M
0
•
L
Wi
W i
W i
...
. ._
“
..L3 ..J!L
104
LiAi

AU?

1
“°
4000
J!LA
‘ilL
ALA
LW
lid
ALA
- I/f
m
-AM
°
lZL!
°°
04-
07
—
.L!!L
U
‘ I ’
LL°

—
—
iLL
,
.±LL.’.

•
- —
—
-
._LL. .A
..JL. L!LI
....AL. .i J!
•
P0
La
.LAj
La
.a

AL’
• •
3 14 ”I -II4-
3/4’ .1-Ill .
.aLlr_W.!
10.000
4 .. . .
500 ...J!L
.zIJI&
• . V..A f
._ .H!
0000
£48.40 Iy
N/A
l± 2
650
‘ ‘°
I3 2
60
ALA
314
.204
‘
0
°° “
— L I .°
.
O
1LQ L
•4-
H
‘ °
.LAP
1’
—
SO
....LL
406
—
—
060
-—
—
-
- —
—
,.iu , .00ai
.. th’
30-4-40
•
130
010
0.70
021
534
‘
“
.
..
300
170
AU?
A
3000
510
104.0
j )
.L1 L
000
ALA
735
.IIJ
206
.JI.
351
A1
32I -
AL
jjO
j
!Jf
‘L2I
‘
2.2
LH’
uP
..ILL
.ILL ..
LA
- —
- —
_fl_
o
I I I
—
3 00
•.

•
1/4 ’ 314’
•
34-
30
.!L
AU?
41.0
07 0
j
Mj
l
500 I
‘ 1j. 0
OJI
.
jQ
3
Jjj
j -
- —
lA iH
10- 1 4 -6t
“
•
—
—
. z -

‘-“ -
3/4-. 1-114-
..W?

-_. L
s°
.AL2
1 14 N 1A 48
• “‘ “ “
—
-
-
. —
- -
P 4 8 0 4 4 ,
.o o ,rn..
. .2L_ 0-Il-SI
•
030
II?
“
“
647
16 300

J
30
AU?
01 0
96 0
L!A1
0
26 0
19 1
000
66 0
60 I
070
•.. .
LAP
L I°
ALL
.ALL ‘
76 30.33—40
..JZ_ LLALM
•
•
LAP
L ’
T T


3,4’.. I.i ’.
1.1 13” .3•
l4 - • 1- 1/4”
115
35.300
5364
34.0
G..b
0.42
AU?
W
70 0

3660
• “ “W ”
00 0
•1T ’
3720
—e-
4
(40
Aro
04 0
AT4
ro
39
Th.J
‘ I I I
lfl
400
1T .
110
“ .
44f
“ “T1.
33?
iee .
142 1

$5 1 ’
.sz
64 o
iT.
loll
2 3
LA 0
0 3
7
I 00
330
071
971
30
25
1
24
3400
soo
009
0100
541
109
110
240
6
—
10 •
.A J
6 6 6
—
44
.
. —
.360.400.. 4841.
. 4 8 4 . 4 3 6 . . ..
2LAA
“
“
N/A
N/A
13 ’00 1
-
.
-
—
—
. -
11- 10.09
0
13)
5J
450
“
•
1009
100
00. 460
AU?
3029
1305
1
400
340
260
20/
7353’
149
—
057
300
—
350
—
I II
—
—
— -
Ill
— —
144 6
—
0 6$
- —
—
—
JL. 6lL. 02
jj
LAO
•
J
75.960
0.42 -
1740
1060
39
IL ”
.010
066
754
SSjj
0 (60
030
270
40
1409
114
23000
21001 1_
I 60
- -
LA1!-V
•
‘LAs
.
‘
J ”
JAIAL!
0000
0900
3400
04
4000
500
I
J
,00r
1441
_r

3500
150
570
97
3600
70
21100
030 6_
4 64
.JL. ALA .
SILl
iT’
.. .....
600
. .j .
..±Wf!
AU?
740
3130
j
IL°
353
1361

iSlO
o4
007
310
03
ISO
300 6
3 6_I
J ± ’
—
.
•
j o
...,JAL
A! .AL°
0 ’ .40025003
J!L
. . . .
II ’
‘ I L 9
JA±
600
Iso
•
6300
.3300
$100 7
0 70
/f .1.ILL’
.
j o
164
.i! .1L 0
0003403001.
L 0
J _ 0
•AL 1
l!L 7
Jf
. . . . . .
•
j•jO
0
t!°
‘1.2’
iAP L.
.!!±!.. t..
‘f
j
4
. .
J T’
250
74.360
0.44
9060
3060
3
1460
3360
036
024
3110
3000
750
95,
3
0000
j47 7
0 71
.iL. . ..2±r

.
13)
.LAL
LA’
. _
.
‘•
•
..L! .
.J j
APJ±!
0.42
AU?
1340
2300
1410
54360
jJJ

.o
L°
3770
jjs
433
I I
iior
j
1041
liii
sis
32o
111’
5100
.u
s
ij
ALA.
6
-Ji . _ L°
4 6_4 -
‘_‘
-
..._!L iAAP
•
LA!
±2’
•
.
370
004.360
AU?
3000
3120
i 2I.
i A . ’
A /A. 9
J O
jj p
r
jo
j
j••
j o
jo
..± .j.
t.
-
3 ‘.‘
.
. 2J !±?
!6IHII
3/4- - l-1/4
.J
.
3.700
a .ObIlS 003
074
970
j
ft .”
.1L’
‘4-
400
350
37
5001
-u
g-
-ALL.
-
I 6 I
,_ ,
133
LIt.
L!?
AP H
3/4”. 1-114”
lAP
AU?
I L 5
j
/ /j
121
314
700
1100
005
I l ?
000
509
—
°°
J 7
7 71
j J / f ’
LA!.
LAP
...
‘
J.f 2 !
.. Ja.
JILA!
£02.
0600
4110
1330
420
6A
1420
— —
.
-

43

-------
TABLE II
Beman Avenue Combined Sewer Overflow
Description of Natural Events
Rainfall
Event No.
Amount
Inches
Durat -
ion
Hour
Intensity
Inches
p r Hour
Total
Overflow
Filtered
Maximum
Overflow
Rate
GPM
Filter
Medium
Coal Size
Inthes
2
0.15
0.25
0.6
1,300
290
1x2
3
0.30
0.25
1.2
1x2
4
0.33
0.25
1.3
lic.2
5
0.10
-
1,200
270
1x2
6
0.25
0.25
1.0
3,000
660
1 x2
8
0.25
1.00
9
0.20
0.25
0.8
1,100
310
3/4x 1-1/4
10
0.30
1.00
0.3
3.400
300
3/4x 1-1/4
0.3
400
110
314x 1-1/4*
14
0.65
14.00
0.0
500
R0
314x 1-1/4*
15
0.60
9.50
0.0
1,400
fl O
3/4x 1-1/4
19
0.55
2.30
0.2
4.900
690
3/4xl-1/4
21
0.25
0.25
1.0
3.200
280
2x4*
22
0.45
2.00
0.2
1.100
180
2x4
25
80
15
Zx4
31
0.50
11.00
0.0
90
25
1 x2*
32
0.80
10.00
0.0
2.600
30
1 x 2
33
0.50
11.00
0.0
1.000
40
1x2
37
0.15
0.50
0.3
170
35
1 x 2
—
41
0.85
3.00
0.3
14. 100
1.100
42
0.50
1.25
0.4
4.400
.680
1 xZ*
43
0.50
0.25
2.0
11.800
1.300
1 x2
44
4.400
.100
45
0.35
19.00
0.0
320
20
46
1.40
24.00
0.0
2.600
230
48
0.35
1.75
0.2
1.500
290
0.65
7.00
0.0
3.600
320
49
0.50
3.00
0.2
3.600
250
51
0.30
2.00
0.2
3.000
250
53
0.75
1.50
0.5
9.200
960
54
3.60
3.75
1.0
155. 900
1.700
3/4x 1-1/4
55
0.55
19.00
0.0
2.100
.270
3/4x1-l/4
56
0.15
0.25
0.6
1.600
380
3/4x 1-1/4
57
0.25
0.25
1.0
3.100
560
3/4x1-1/4
58
0.70
0.50
1.4
8.700
510
3/4x 1-1/4
3/4 x 1-1/4
3/4 x 1-1/4
3/4 x 1-1/4
3/4 x 1-1/4
3/4 x 1-1/4
3/4 x 1-1/4
3/4 x 1-1/4*
3/4 x 1-1/4
45

-------
TABLE II (Cont’d)
Beman Avenue Combined Sewer Overflow
Description of Natural Events
* Filter changed with fresh coal.
Rainfall
Event No.
Amount
Inches
)urat-
ion
Hour
Intensity
Inches
per Hour
Total
Overflow
Filtered
Maximum
Overflow
Rate
GPM
Filter
Medium
Coal Size
Inches
59
0.30
0.75
0.4
500
100
3/4x 1-1/4
61
0.80
3.00
0.3
9,700
890
3/4x 1-1/4
62
0.75
3.00
0.3
7,900
530
3/4x 1-174
64
0.25
0.Z5
1.0
2, 900
69
360
0.08
65
0.50
0.75
0.7
6,200
870
3/4 x 1-1/4
66
0.30
0.25
1.2
3,600
550
3/4x 1-1/4
67
0.60
1.00
0.6
12,300
1,200
3/4x1-1/4.
68
3.60
35,000
3/4x 1-1/4
0.15
3/4 x 1-1/4*
•0. S
600
150
3/4x
1-1/4*
70
0.15
0.25
0.6
1,200
100
3/4x1-1/4
71
0.70
5.25
0. 1
22, 600
270
3/4 x
1-114
72
0.10
3.50
0.0
360
20
3/4x
1-1/4
75
0.20
1.00
0.2
600
30
3/4x
1-1/4
77
0.50
5.75
0. 1
—
4, 100
320
3/4 x
i-IJ
79
81
0.75
1.75
-
4.50
19.25
0.2
0.1
1,800
2,500
220
40
3/4x
3/4x1-1/4
1-1/4
82
0.25
4.00
0.1
600
60
314x
1-1/4
87
0..35
3.25
0.1
17,000
440
3/4x
1-1/4
88
0.50
3.50
0.1
1,000
270
3/4x
1-1/4
90
0.25
5.25
0. 0
0.20 0.50
0.4
5,600
550
500
20
3/4x1 -1/4*
3/4 x 1—1/4
46

-------
TABLE III
Rapid Flow Filter Combined Sewer Overflow
Treatment Results
Natural Events - Conventional Automatic Sample s
% Reduction
% Volatile
Event No.
Suspended
Solids
Settleable
Solids
Susp.
Solids
Total
Solids
Character
of
Overflows
In.
Eff.
In.
Eff.
14
58
49
65
48
37
Low Flow
21
(250)
2.5
43
60
34
25
( 55)
38
49
94
69
Low Flow
31
( 61)
(100)
18
41
52
56
Low Flow
41
1421) —
(160)
27
13
____
High Flow
42
(186)
33
43
31
70
18
43
7
72
43
25
42
32
High Flow
44
( 94)
( 43)
66
62
30
30
C 4)
34
Low Flow
46
20
35
45
49
(37)
46
22
51
41
38
41
48
46
54
63
55
25
12
37
40
High Flow
57
(264)
21
33
35
32
58
(118)
43
35
High Flow
62
44
66
47
48
High Flow
65
(127)
—
39
52
High Flow
66
( 83)
( 87)
25
37
28
70
(44)
18
39
38
45
48
71
(75)
53
49
39
49
43
72
( 78)
0
35
53
Low Flow
75
(16)
0
47
27
33
30
77
79
( 4)
2
24
(67)
51
67
46
42
47
49
82
(56)
50
32
25
33
37
87
(130)
(150)
46
32
44
35
88
( 29)
( 67)
2.7
34
34
32
90
( 80j
(160)
43
45
32
41
Low Flow
91
( 85)
( 36)
31
28
34
27
High Flow
47

-------
TABLE IV
Rapid Flow Filter Combined Sewer Overflow
Treatment Results
Natural Events - Grab Samples
% Reduction % Volatile
Event
No.,
Suspended
Solids
Settleable
Solids
Suspended
Solids
Total
Solids
Comments
In.
Eff.
In.
Eff.
2
12
54
45
67
25
3__
53
-
-
-
-
4
50
-
-
-
-
15
79
39
12
42
42
53
(40)
43
3?
42
40
High Flow
55
56
92
27
12
37
46
67
(2444
(100)
25
22
31
29
High Flow
77
57
24
63
28
83
77
81
84
97
47
41
60
50
Low Flow
49

-------
TABLE V
Rapid Flow Filter Combined Sewer Overflow
Treatment Results
Simulated Events (Diluted Sanitary Sewage) - Conventional Automatic Samples
% Reduction
% Volatile
Event
No.
Suspended
Solids
Settleable
Solids
Suspended
Solids
Total
Solids
Comments
In.
Eff.
In.
Eff.
16
(5)
77
71
54
38
17
28
89
92
33
35
18
19
85
91
32
30
20
(14)
83
59
23_
(40)
94
97
54
37
24
(16)
75
75
57
55
26
14
40
66
61
26
17
27
(23)
9
53
71
37
32
34
56
8
22
36
18
13
35
(20)
0
23
15
12
10
36
(21)
(36)
13
19
13
16
38
(52)
(11)
34
39
9
11
5].

-------
TABLE VI
Rapid Flow Filter Combined Sewer Overflow
Treatment Results
Simulated Events (Diluted Sanitary Sewage) - Grab Samples
% Reduction
% Volatile
Event
No.
Suspended
Solids
Settleable
Solids
Suspended
Solids
Total
Solids
Comments
In.
Eff.
In.
Eff.
1
23
31
75
29
22
11
25
26
31
No dilution
28
(3)
0
89
92
32
33
29
(11)
0
83
92
42
43
30
28
43
78
85
40
32
40
25
38
57
62
38
36
.
1 ft. filter
No dilution
50
( 3)
25
29
43
65
9 ft. 2 filter
52
18
72
67
46
40
9 ft. 2 filter
60
47
33
40
66
38
41
9 ft. 2 filter
63
12
37
41
45
46
35
9 ft. 2 filter
76
6
28
73
80
39
42
Sample from early part
of run. 9 ft. 2 filter
(14)
25
77
36
41
Sample from latter
part of run.
80
39
48
87
95
68
67
85
0
37
79
73
39
37
86
(5)
37
77
71
61
40
89
(11)
37
79
77
41
37
53

-------
TABLE VII
Rapid Flow Filter Combined Sewer Overflow
Treatment Results
Simulated Events - Grab Sample a
Filter Media: Stratified Bed using 1/4 x 3/4; x 1-1/2; and 1-1/2 x 2 Coal
%Reduction
% Volatile
Event No.
Suspended
Solids
Settleable
Solids
Suspended
Si ,ids
Total
Solids
Comments
In.
Eff.
In.
Eff.
50
(3)
25
29
43
65
52
18
72
67
46
40
60
47
33
40
66
38
41
Same filter medium
as for Event No. 2
63
12
37
41
45
46
35
76
6
28
73
—
80
39
42
Sample from early
part of run.
(14)
25
77
74
36
41
Sample from latter
part of run.
Note: The filter bed
coal layer
for Event No. 76 did not contain the 1/4 x 3/4
55

-------
TABLE VIII
Rapid Flow Filter Combined Sewer Overflow
Treatment Results
Results of Initial Runs through a Filter Bed
Event
No.
Descr.
of
Event
Type
of
Sample
% Reductton
—
% Volatile
Filter
Medium
Coal Size,
Inches
Comments
Sus.
Sol.
Sett.
Sol.
Sus.
Sol.
Total
Solids
In.
Eff.
1
Sim.
Grab
23
-
31
75
29
22
1 x 2
7
Sun.
N.A.
16
34
29
47
3/4 x 1-3/4
12
Sun.&
Nat.
Grab
-
49
3/4 x 1-1/2
20
Sun.
Auto
(14)
-
83
59
-
-
2 x 4
28
Sun.
Grab
0
89
92
32
33
1 x 2
40
Sim.
Grab
25
38
57
62
38
36
1/4 x 1-3/4
41
Nat.
Auto
421)
(160)
27
13
-
-
1/4 x 1-3/4
First significant
run with this
filter
42
Nat.
Auto.
1 j
33
31
70
1• _
1 x 2
44
Nat.
Auto.
(43)
66
62
30
30
3/4 x 1-1/4
50
Sun.
Grab
(3)
-
25
29
43
65
1/4 x 2
Coal in 3 graded
layers
51
Nat.
Auto
41
-
38
41
48
46
3/4 x 1-1/4
52
Sim.
Grab
18
12
-
-
72
41
-
67
45
-
46
40
35
-
3/4 x 2
Coal in 3 graded
layers
63
Sun.
Grab
37
46
3/4 x 2
Coal in 3 graded
layers
64
Nat.
Mixed
-
-
3/4 x 1-1/4
Sample results not
comparable
69
Nat.
N.A.
-
-
-
-
-
-
3/4 x 1-1/4
73
Sim.
N.A.
-
6,
J
-
-
-
80,
74
-
-
42,
41
1/4 x 2
Coal in 3 graded
layers
76
Sun.
Grab
28.
25
73,
77
39,
36
3/4 x 2
Coal in 2 graded
layers
89
Sun.
Grab
J
37
79
77
41
37
3/4x 1-1/4
57

-------
TABLE IX
Rapid Flow Filter Combined Sewer Overflow
Treatment Results
Simulated Events - Experimental Filter Screen Sampling Technique
% Reduction
Event No.
Mesh Size
in Screen
Filterable
Solids
Suspended
Solids
Settleable
Solids
Comments
74
4
57.7
76
2
36.8
6
28
Grab Samples
89
8
38.0
(11)
37
Grab Samples
59

-------
TABLE X
Rapid Flow Filter Combined Sewer Overflow
Treatment Results
Simulated Events - Total Segment Sampling Technique
Weight of
Dried Solids
% Reduction
Event No.
Influent
Effluent
Filterable
Solids
Suspended
Solids
Settleable
Solids
Comments
78
23.2
13.0
43.9
80
29. 7
17. 1
42. 5
39
48
Grab Sample
84
49.1
31.0
36.9
16
28
Grab Sample
S
89
47.2
25.1
46.9
(11)
37
Grab Sample
61

-------
TABLE XI
Rapid Flow Filter Flow Rate Tests
Raw Sanitary Sewage — 1. 1 ft 2 Coal Packed Filter
Total Flow
before
Flow Rate Filtration Coal Size pluggage
Test No. Rate -GPM inches gal. Comments
1. 100 3/4 x 1-1/4 1,500 Sewage was
c omrninut e d
2. 140 3/8 x 6 > 20, 000
3. 140 3/8 x 1-3/8 2,380
4. 140 1-3/8x 2 48,000
5. 140 4 x 6 >230, 000
6. 140 3x4 > 80,000
7. 140 3x4 > 50,000
8. 140 2 x3 > 50,000
9. 140 1-3/8 x 2 48, 000
10. 140 1-3/8 x 2 36, 000
11. 140 1 x 1-3/8 48, 000
12. 140 lx 1-1/2 36,000
13. 140 3/4x 1-3/8 46,000
14. 140 3/4 x 1-3/8 36, 000
15. 140 3/4 x 1 37, 500
16. 140 3/4 x 1 26, 000
17. 140 3/4 x 1 18, 000
18. 140 3/4 x 1 and 28, 000 2 layers of coal
1 x 1-3/8
19. 140 3/4 x 1 and 19, 000 2 layers of coal
1 x 1-3/8
20. 140 lx 1-3/8 and 48, 000 2 layers of coal
1-3/8 x 2
21. 140 1 x 1-3/8 and 43, 000 2 layers of coal
l-3/8x 2
22. 140 1-3/8 x 2 and 50, 000 2 layers of coal
2 x 2-3/4
63

-------
TABLE XI (Cont’d)
Rapid Flow Filter Flow Rate Tests
Raw Sanitary Sewage — 1. 1 ft. 2 Coal Packed Filter
Total Flow
before
Flow Rate Filtration Coal Size pluggage
Test No. Rate-GPM inches gal. Comments
23. 140 1-3/8 x 2 and 38, 000 2 layers of coal
2 x 2-3/4
24. 140 2 x 2-3/4 and 50, 000 2 layers of coal
2-3/4 x 4
25. 140 2 x 2-3/4 >50,000 2 layers of coal
4x6
26. 140 3/4x 1 46,800
27. 140 1 x 1-3/8 49, 800
28. 140 1-3/8 x 2 54, 000
29. 140 3/4 x 1 and 28, 000 2 layers of coal
1 x 1-3/8
64

-------
TABLE XII
Rapid Flow Filter Solids Removal Tests
Raw Sanitary Sewage — 1. 1 ft 2 Coal Packed Filter
Solids
Removal Filtration Coal Size % Solids
Test No. Rate-GPM Inches Reduction* Comments
1. 140 3/4xl 66
2. 140 1 x 1-3/8 40
3. 140 1-3/8 x 2 27
4. 140 2x3 23
5. 140 3x4 18
6. 140 3/4 x 1 and 68 2 layers of coal
1 x 1-3/8
7. 140 1 ,c 1-3/8 and 53 2 layers of coal
1-3/8 x 2
8. 140 1-3/8 x 2 and 40 2 layers of coal
2x3
* The amount of solids in the influent and effluent streams was measured by
filtering 1,000 gal through a nylon tulle net bag containing 1/16” holes and
then weighing the dried solids.
65

-------
Figure I.
Twin outfall near Dorver Avenue and East 77th
Street in Cleveland, Ohio
OCI
Figure II.
View of typical houses on Beman Avenue
67

-------
OVERFLOW SEi ’ER
FLOW ____ - /
FLOW
FI _ o,/ _
PLAN
Figure III. Side spiliway type Overflow Structure
SECTION A - A
69

-------
Topographical representation of Beman Avenue drainage area. Pilot filter site
is 30 et from west end of Dorver Avenue, l’ 170’
71
Figure IV.

-------
Figure V.
Appearance of filter basket with filtered solids.
Figure VI. Some of the larger solids removed from a combined
sewer overflow by the Rapid-Flow Filter.
73

-------
I
From existing
combined sewer
side overflow Flow Detector
structure Flo 7a:;; : der
Rapid -Flow
Filter
Baskets
Flow Recorder
Sampler
Receiving
Over flo
J owWater
To
Receiving
Waters
Figure vii. Rapid-Flow Filter Pilot Plant Flow Sheet
Figure X. View of the solids sampling screens before air
drying. Effluent sample on left, influent on right.
.‘ -:- .—-- -
7
g 4 S
75

-------
ACKNOWLEDGMENT
The cooperation of all of the individuals, agencies, corporations, and
officials involved is gratefully acknowledged.
77

-------
BIBLIOGRAPHY
1. “Investigation of the Use of Coal for Treatment of Sewage
and Waste Waters”. Office of Coal Research, U. S. Dept.
of Interior, Research and Development Report No. 12
Rand Development Corporation October 1965.
2. “Problems of Combined Sewer Facilities and Overflows”
American Public Works Association Research Foundation
FWPCA Publication No. WP 20-11, 1967. Pxii.
3. “Process Using Coal in Sewage Treatment”
U.S. Patent 3,401,114 September 10, 1968.
4. “Cleveland, Ohio, Master Plan for Pollution Abatement,
Part 2”. Havens and Everson Consulting Engineers,
July, 1968.
79

-------
APPENDIX I
Sample Handling and Analysis
Sa’rnples were gathered automatically and collected in plastic five gallon
cans. Each five gallon can was then stirred and a two gallon bottle was
filled from it. The two gallon bottles were then stored in a refrigera-
tor (5° C) until analyzed during the next weekday day shift. The sam-
ples were stored at ambient temperatures at the collection site for vary-
ing lengths of time if a natural overflow event had occurred. They
were refrigerated within two hours if a simulated event was sampled.
Simulated runs were all made during the day shift so that the amount
of time that these samples were not refrigerated before being analyzed
was fairly uniform. This type of handling was also characteristic
of all the grab samples from the natural events.
In the laboratory the samples were transferred to a two gallon polyethy-
lene vessel equipped with a 500 RPM mixer. Sample increments for
the various analyses were taken from this agitated vessel through a
1/2 inch I.D. spigot.
The 12th Edition of Standard Methods for the Examination of Water and
Wastewater was used for all analyses with minor modifications:
1. Suspended Solids
Eight-micron porous plastic (Millipore) filters were used in
all cases. A sample volume was selected that would pass the
filter in less than five minutes. Duplicates were run in all
cases.
2. Total Solids
Twent -five milliliter samples were dried in tared crucibles
at 105 C overnight. Duplicates were run.
/
A Corning Model 7 pH meter with a glass electrode was used.
A pH 7. 05 phosphate buffer was used for calibration before
each use.
81

-------
4. Volatile Solids
Samples in tared crucibles were heated for twenty minutes at
6000 C. Duplicates were run.
5. P0 4
Phosphates were run by the amino naphthol sulfonic acid
method as outlined in Standard Methods.
FEB
82

-------