'EPA
United States
Environmental Protection
"Agency
Research and Development
Municipal Environmental Research
Laboratory
Cincinnati OH 45268
EPA-600/2-79-015
March 1979
Dual Process
High-Rate
Filtration of Raw
Sanitary Sewage and
Combined Sewer
Overflows
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-79-015
March 1979
DUAL PROCESS HIGH-RATE FILTRATION
OF RAW .SANITARY SEWAGE
AND COMBINED SEWER OVERFLOWS
by
Hank Innerfeld
Angelika Forndran
New York City Department of Water Resources
New York, New York 10013
Dominick D. Ruggiero
Thomas J. Hartman
Nebolsine Kohlmann Ruggiero Engineers, P.C.
New York, New York 10022
Grant No. S-803271
Project Officers
Richard Field
Richard P. Traver
Storm and Combined Sewer Section
Wastewater Research Division
Municipal Environmental Research Laboratory (Cincinnati)
Edison, New Jersey 08817
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research
Laboratory, U. S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily
reflect the views and policies of the U. S. Environmental Protection Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
11
-------
FOREWORD
The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health
and welfare of the American.people. Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of that environment and the interplay between its
components require a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem solution
and it involves defining the problem, measuring its impact, and searching
for solutions. The Municipal Environmental Research Laboratory
develops new and improved technology and systems for the prevention,
treatment, and management of wastewater and solid and hazardous waste
pollutant discharges from municipal and community sources, for the
preservation and treatment of public drinking water supplies and to
minimize the adverse economic, social, health, and aesthetic effects of
pollution. This publication is one of the products of that research; a
most vital communications link between the researcher and the user
community.
The deleterious effects of combined sewer overflows upon the nation's
waterways have become of increasing concern in recent times. This
report presents the results of a two-year testing program of pilot plant
high-rate filtration for treating combined sewer overflows and sanitary
sewage.
Francis T. Mayo
Director
Municipal Environmental
Research Laboratory
111
-------
ABSTRACT
Pilot plant studies were,conducted in New York's Newtown. Creek Water Pollution
Control Plant from 1975-1977 to investigate the suspended solids (SS) removal
capabilities of the deep bed, high rate gravity filtration process on raw
sewage and combined sewer overflows.
The treatment system was composed of a rotating screen equipped with a 40
mesh (420 micron) screen followed by a dual media, high rate filter containing
48 in. (122 cm) or 60 in. (152 cm) of No. 3 anthracite (effective size 3.85 mm)
over 30 in. (76 cm) of No. 612 sand (effective size 2 mm).
A continuous series of tests on dry weather (raw sewage) flows demonstrated
SS removals across the filter averaging 67 percent at a flux range of
8-12 gpm/ft^ (20-30 m3/hr/m2) with an average effluent concentration of
44 mg/1 SS. BOD and COD removals were 39 percent and 34 percent, respec-
tively.
Tests on combined sewer overflow showed an average removal of 61 percent SS
across the filter and 66 percent across the system at a flux of 16 gpm/ft^
(40 m3/hr/m2) and an average effluent of 62 mg/1 SS. BOD and COD removals
across the filter were 32 percent and 42 percent, respectively. The addition
of cationic polymer (1-2 mg/1) in combination with alum (17-35 mg/1) Improved
filter removals to an average 72 percent for SS, 40 percent for BOD and 50
percent for COD for two tests.
Capital costs (ENR-2520) for a high rate filtration plant are estimated at
$55,225 per mgd for a 200 mgd plant (757,000 m3/day) . Total annual treatment
costs, including interest amortization, operation and maintenance charges,
range from approximately $396,450 to $1,794,050 for dual treatment facilities
in a 25 to 200 mgd (94,600 to 757,000 mVday) capacity range and $238,050 to
$1,175,900 for the same capacity range of facilities treating only CSO.
Comparison with alternate treatment systems show that high rate filtration
(HRF) is cost competitive with conventional sedimentation facilities for dual-
process or CSO treatment.yet HRF has only 5-7 percent the area requirements.
For strict CSO treatment, 'HRF is competitive with dissolved air flotation and
microstraining processes.
This report was submitted in fulfillment of Grant No. S 803271 by New York'
City, Department of Water Resources; and Nebolsirie Kohlmann Ruggiero Engi-
neers, P.C. under the sponsorship of the U.S. Environmental Protection Agency.
This report covers'plant construction and evaluation performed during the
period June 1, 1975 to June 30, 1977; all work was completed as of July 1,
1978.
IV
-------
CONTENTS
Foreword m
Abstract iv
Figures V1
Tables
Abbreviations and Symbols x^
Acknowledgements • • X-L1
1. Introduction • 1
2. Conclusions • • ^
3. Recommendations • • _**••
4. Pilot Plant Facilities 14-
5. Testing Program
6. Testing Results 32
Operational Experience 32
CSO Testing Results .. 34
RDWS Testing Results 42
Backwashing/Concentrate Evaluation 44
Metals Removal 50
Rotary Screening 52
Disinfection. j>2;
7. Full-Scale HRF Installations 56'
8. Cost estimates • 6&.
References. 80
oo
Appendices «
-------
FIGURES
Number
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Newtown Creek Water Pollution Control Plant drainage
area ...
HRF pilot plant location
HRF pilot plant process flow diagram
HRF pilot plant facilities
HRF pilot plant facilities.
HRF 30-in. pilot filter column
Discostrainer model DS-110
CSQ SS mass captures per unit filter area and time ..
RDWS suspended solids removal averages.
RDWS SS mass captures per unit filter area and time..
Filter backwash effluent SS vs.time
Disinfection vs. chlorine dosage
Dual-treatment HRF installation process flow diagram.
Dual-treatment HRF installation plan 25 mgd capacity..
Dual-treatment HRF installation plan 50 mgd capacity..
Dual-treatment HRF installation elevation 50 mgd
capacity
17 Dual-treatment HRF installation plan 100 mgd capacity.
Page
3
15
16 '
17
18
19
22
39
45
46
49
55
59
62
63
64
65
-------
FIGURES (Concluded)
Number
18
19
20
21
22
Dual-treatment installation elevation 100 mgd capacity. 56
Dual-treatment HRF installation typical filter section.. 67
Estimated capital cost vs. design capacity 69
Total annual cost vs. design capacity 72
Operating cost - benefits, CSO treatment 75
vii
-------
TABLES
Number
1
2
3
4
5
6
7
8
9
10
11
'12
13
14
15
16
17
Laboratory Analyses.
CSO Sampling and Analytical Schedule.
CSO SS Removals vs. Chemical Additions.
CSO SS Removals, Weekday-Weekend Comparison,
Average Mass Captures of CSO SS.
Settieable Solids Removals.
CSO, BOD and COD Removals
Page
26
29
36
37
38
40
41
RDWS SS, BOD and COD Removals vs. Flux and
Chemical Additions 43
Average Mass Captures of CSO SS.
Metals Removal Averages.
Discostrainer Sludge Analyses.
Disinfection Results.
Summary of Total Project Costs.
Estimated Project Costs for HRF Treatment Plants*
Summary of Total Annual Costs.
Estimated Annual Costs for HRF Treatment Plants.
47
51
53
54
68
70
71
73
Dual Treatment System Cost Comparisons 77
viii
-------
TABLES (Continued)
Number
Page
18 Dual Treatment System Area Comparison 77
19 CSO Treatment System Cost Comparisons. 78
A -1 CSO Testing Averages...».., 82
A -2 RDWS Testing Averages 84
A -3 Filter Performance - CSO 85
A -4 Filter Performance - RDWS 86
A -5 Filtration Composite Characteristics. 87
A-6 BOD, FBOD, UBOD Composites Comparison 88
A -1 FBOD Grab Sample Averages 89
A-8 VSS Composite Results. 90
A -9 Backwash Composite Characteristics , 91
A -10 Backwash Sludge Testing Results 92
A -11 Cadmium Removal by HRF Treatment 94
A -12 Chromium Removal by HRF Treatment. 95
A -13 Copper Removal by HRF Treatment 96
A -14 Lead Removal by HRF Treatment 97
A i 15 Mercury Removal by HRF Treatment 98
A -16 Nickel Removal by HRF Treatment 99
A -17 zinc Removal by HRF Treatment. 100
ix
-------
TABLES (Concluded)
Number
A-18 Arsenic Removal by HRF Treatment
A -19 HRF Removal of Coliform Bacteria
A -20 Test Storm Characteristics
A-21 Conversion Factors, English to Metric Units.
Page
101
102
103
104
-------
ABBREVIATIONS AND SYMBOLS
ABBREVIATIONS
BOD
COD
CSO
ENR
FBOD
FOOD
FC
HRF
mgd
PVC
ROWS
SDI
SVI
SS
TEFC
UBOD
VSS
SYMBOLS
Cl,
biochemical oxygen demand, five day
chemical oxygen demand
combined sewer overflow
Engineering News Record
filtrate BOD
filtrate COD
fecal coliform
high rate filtration, high rate filter
million gallons per day
poly vinyl chloride
raw dry weather sewage
sludge density index = mg/1 of suspended matter
ml of settled sludge x 10
sludge volume index = -=-^2.
SDI
suspended solids
totally enclosed fan cooled
ultimate BOD
volatile SS
chlorine
XI
-------
ACKNOWLEDGEMENTS
The authors wish to thank the following people for their timely and
valuable assistance on this project:
From the City of New York Department of Environmental
Protection: Norman Nash, P. E., Assistant Commissioner;
William Pressman, P. E. , Chief, Research and Development;
Jerry Degen, Senior Chemist-in-Charge of the Industrial
Wastes Laboratory and his staff for the laboratory
analytical work; Quentin Monahan, Superintendent of the
Newtown Creek Treatment Plant, and his staff for their
operating assistance.
From NKRE: Ross Nebolsine, P. E., President;
Harold Kohlmann, P. E., Vice President; Vincent
Stramandinoli, Senior Engineer; and Joseph Marsiello,
Senior Designer.
From the USEPA-Storm and Combined Sewer Section,
Edison, New Jersey: Hugh E. Masters and Chi-Yuan Fan,
Technical Advisors.
The authors further wish to thank Hycor Corporation for providing a
Discostrainer screening device and Kenics Corporation for providing
two in-line-mixers; all at no cost to the project.
This project was supported by demonstration grant No. S 803271 from
the United States Environmental Protection Agency.
Xll
-------
SECTION 1
INTRODUCTION
Combined sewer overflows (CSO) and non-point source runoff caused during
rainfall are significant sources of water pollution (1). These high volume,
short term events send concentrated pollutants into receiving waters,
degrading the water quality.
In recent years a variety of systems and treatment processes have been
tested for the control and treatment of CSO. Any system or facility
designed for retention and/or treatment of CSO should include automatic
startup, be flexible in capacity, and reliable. A system designed only to
control CSO would operate on a stand-by basis and be idle for significant
periods of time, tending to be less cost-effective than a system which
could treat both dry weather and wet weather flows throughout the year.
Such a dual-treatment system utilizing high rate filtration (HRF) techno-
logy was evaluated under the project discussed herein (USEPA demons-
tration grant S 803271) by the City of New York and Nebolsine Kohlmann
Ruggiero Engineers, P. C. (NKRE), using a dual media HRF.
The tests were performed on a 30 in. (76 cm) diameter gravity filter
column containing 48 in. (122 cm) or 60 in. (152 cm) of No. 3 anthracite
over 30 in. (76 cm) of No. 612 sand. The filter was preceded by a rotary
screening unit equipped with a 40 mesh (420 micron) screen to remove
grit, fibers and other coarse material which could clog the filter and
reduce the length of filter runs. Testing was performed both with and
without the use of chemical additives at a flux range of 8-24 gpm/ft2
(20-60 m3/hr/
m2).
HISTORY
HRF has been developed over the past 15 years and used for a variety of
treatment applications, chiefly in industrial wastewater treatment. A 2
mgd (7570 m^/day) plant is currently being used in sewage treatment for
effluent polishing in Ashvale, England (2) at a flux range of 8-12 gpm/ft2
(20-30 m3/hr/m2). Another HRF polishing system is under construction
-------
in Como Italy, as part of a new 10 mgd (37, 850 m3/day) secondary treat-
ment plant; the design flux is 15 gpm/ft2 (37 m3/hr/m2).
Several pilot plant studies of HRF for both CSO and dry-weather municipal
flows have recently been completed. Studies at Rochester, N. Y. (3) used
parallel 6-in. (15 cm) pressurized dual media filters at a flux range of
10-24 gpm/ft2 (25-60 m3/hr/m2) with and without chemicals. A primary
swirl solids concentration system treated the CSO ahead of the filters.
A previous study at Syracuse (4) used CSO pretreated by fine screening
and reported suspended solids (SS) mass removals of 43-91 percent at a
flux range of 10-13 gpm/ft2 (25-32 m3/hr/m2) with 50-220 mg/1 doses of
alum and 1-4 mg/1 doses of polymer filtration aids. A study in Minnesota
(5) utilized natural and simulated stormwater in 20-in. (51 cm) diameter
dual media pressurized filters and reported average SS removals of 63
percent with polymers and 51-58 percent without chemicals at a' flux range
of 16-24 gpm/ft2^ (40-60 m3/hr/m2).
The most extensive CSO filtration testing program to date was performed
in Cleveland during 1970-71 (6). Using 6-in. (15 cm) pressurized dual
media filters operating in a flux range of 8-24 gpm/ft2 (20-60 m3/hr/m2),
the reported removals were 93-98 percent of SS using polymer, and 50-75
percent without chemicals.
SCOPE OF PROJECT
The purpose of the study was to evaluate the HRF process on CSO flows,
and on raw dry-weather sewage flows (RDWS). The tests were performed
on a 30-in. (76 cm) diameter filter column to reduce the sidewall effects
which were experienced with the smaller columns at Cleveland (6) and thus
to closely approximate operations of a full scale HRF plant.
The pilot plant was located at the Newtown Creek Water Pollution Control
Plant in Brooklyn, New York. Construction began in June 1975, with field
testing and data collection extending from October 1975 to July 1977. The
pilot plant was mothballed both winters to protect the piping and mechanical
equipment from freezing.
The Newtown Creek plant receives flow collected by combined sewer
systems from parts of Brooklyn, Queens and Manhattan as shown on Figure
1. The drainage area covers 15,390 acres (6233 ha) serving more than
2 million people and the land use division is 54 percent (area) high-medium
residential, 23 percent commercial-business district, 13 percent light-.
heavy industry and 10 percent parks - cemeteries. Concerning the relative
contributions of dry-weather sewage flow (RDWS), the division is 50 percent
residential, 37 percent commercial-business district and 13 percent .
industrial.
-------
LEGEND:
DRAINAGE AREA
NEWTOWN CREEK PLANT
FIGURE 1 - NEWTOWN CREEK WATER POLLUTION CONTROL PLANT DRAINAGE AREA
-------
The Newtown Creek plant utilizes a high rate activated sludge process
preceded by bar screening and grit removal, and was designed for 60
percent removal of BOD and 70 percent of SS. The plant provides no
primary treatment; the flow passes from the grit chamber directly into
the aeration basins. Of the 12 sewage treatment plants in New York City,
Newtown Creek receives the highest daily flows and the largest industrial
component. The dry weather capacity is 310 mgd (1, 173, 000 m /day) but
the plant can provide effective grit removal and secondary treatment for
580 mgd (2, 195,000 m /day) during rain. Diversion structures (regulator
devices) in the combined sewers channel the excess storm generated flows
or CSO into nearby receiving waters to prevent hydraulic overloading at
the plant.
-------
SECTION 2
CONCLUSIONS
FILTRATION RESULTS - Csb TREATMENT
1. HRF treatment of CSO at 16 gpm/ft2 (40 m3/hr/m2) constant flux
provided overall average SS removals of 61 percent across the
filter and 66 percent across the system with an average influent SS
concentration of 182 mg/1. Average SS removals for the three
testing modes (no chemicals, polymer only, polymer and alum) and
test ranges were:
CSO: SS REMOVALS
No chemicals
Poly only
Poly & alum
Average
(weighted)
Range
Plant
Influent
(mg/1)
175
209
152
182
(99-311)
Filter
Influent
(mg/1)
Filter
Effluent
'(mg/1)
150
183
143
67
68
47
Filter
Removals
55
63
67
System
Removals
62
67
69
161 62
(94-266) (30-87)
61
(48-75)
66
(51-79)
BOD removals from CSO averaged 32 percent across the filter and
41 percent across the system with an average influent-BOD of
136 mg/1. The removals improved with chemical additions. Average
BOD removals for the three testing modes and test ranges were:
-------
CSQ: BOD REMOVALS
No chemicals
Poly only
Poly & alum
Average
(weighted)
Range
Plant
Influent
(mg/1)
164
143
92
136
(79-223)
Filter
Influent
(mg/l)
131
129
85
118
(67-139)
Filter
Effluent
(mg/1)
96
84
53
80
(49-99)
Filter
Removals
(% )
27
35
38
32
(26-45)
System
Removals
(% )
41
41
43
41
(30-50)
COD removals from CSO averaged 42 percent across the filter and
47 percent across the system with an average influent of 302 mg/1.
As with BOD, chemical additions improved removals. Average COD
removals for the three testing modes and test ranges were:
CSO: COD REMOVALS
No chemicals
Poly only
Poly & alum
Average
(weighted)
Range
Plant
Influent
(mg/1)
332
306
260
302
(223-384)
Filter
Influent
(mg/1)
285
295
242
278
(190-385)
Filter
Effluent
(mg/1)
193
159
120
160
(98-251)
Filter
Removals
( % )
32
46
50
42
(23-56)
System
Removals
( % )
42
48
54
47
(33-57)
High molecular weight cationic polymers appeared to provide most
improved solids removal during CSO testing. In the one test using
1.0 mg/1 of polymer (Betz 1150), 68 percent filter removal of SS
was obtained with a 266 mg/1 filter influent. Two tests with 1-2
mg/1 of the same polymer and 17-35 mg/1 alum feed obtained an
average 72 percent SS removal with an average 154 mg/1 influent.
The percent removal of SS in the filtration is largely dependent upon
the strength of influent flows.
-------
The CSO data was grouped on a weekday/weekend basis; SS influent
strengths were noticeably higher during weekday tests (attributable
to the higher background industrial component on weekdays).
RELATIVE SS CONCENTRATIONS
Weekday
"Weekend
Number
of Runs
5
4
Filter
Influent
(mg/1)
179
135
Filter
Effluent
(mg/D
58
66
Removals
68
51
From the data it is apparent that the variability in filter removals is
due to the influent strengths, since the effluent concentrations are
essentially the same.
FILTRATION RESULTS - RDWS TREATMENT
4. Treatment of RDWS at flux values of 8 and 12 gpm/ft2 (20 and 30
m3/hr/m2) with an average influent SS concentration of 138 mg/1
provided SS removals averaging 65 percent across the filter and 67
percent across the system. The average results and ranges for all
8 and 12 gpm/ft2 flux tests were:
RDWS REMOVALS
Overall
RDWS
SS
Average
Range
BOD
Average
Range
COD
Average
Range
Plant Filter Filter Filter System
Influent* Influent Effluent Removals Removals*
(mg/1) (mg/1) (mg/1) (••% ) ( % )
138 132 46
(113-170) (51-183) (15-75)
65
67
(50-77) (50-78)
170 152 98 36 42
(130-215) (55-230) (25-165) (12-64) (27-67)
348 290 194 33 44
(320-425) (170-340) (80-305) (20-53) (33-47)
* Results based on extremely limited plant influent data.
7
-------
In a continuous series of HRF tests on ROWS (interrupted only for back-
washing) with four filter runs at Sgpm/ft2 (20 m'S/hr/m?) without chemical
feed directly followed by ten runs at 12 gpm/ft"2 (30 ni^/hr/m2) with
0.5 mg/1 polymer feed, both modes provided SS removals of 67 percent
with average run lengths of 5.3 hrs. Both the BOD and COD removals
improved with the polymer addition. Subsequent testing with alum in
combination with polymer failed to improve removals at the 12 gpm/ft2
(30 m3/hr/m2) flux. The average results of the continuous series of
ROWS tests were:
Filter Filter
Continuous Series Influent Effluent Removals
ROWS (im'g/1) . (mg/1) (%}
8 gpm/ft2. no chemicals
12 gpm/ft2, 0.5 mg/1 poly
BOD
8 gpm/ftS, no chemicals
12 gpm/ftS 0.5 mg/1 poly
COD
8 gpm/ft ^ no chemicals
12 gpm/ft , 0.5 mg/1 poly
SUSPENDED SOLIDS CAPTURE RESULTS
129
125
225
140
270
265
42
41
157
86
195
174
67
67
30
39
28
34
5. A filtration system should be designed on the basis of total mass of SS
captured per unit of filter area or media volume per run under restric-
tions of head loss requirements and breakthrough of solids. Break-
thrbugh is indicated by a visible increase in effluent solids
concentration over a desired value.
From the testing the following mass removals data were obtained at a
flux of 16 gpm/ft'1 (40 mf/hr/m2) for CSO tests and flux values of 8 and
12 gpm/ft^ (20 and 30 nryhr/m^) for ROWS tests:
Capture Parameter
lbs/ftz/run
Ibs/ft2/hr
Ibs/ft3/run
Ibs/ft3/hr
SS CAPTURE BY HRF
CSO Tests
ROWS Tests
Average
3,7
Range
(1.7-9.4)
0.76 (0.36-1.45)
0.54 (0.27-1.45)
0.11 (0.06-0.22)
Average
2.9
Range
(1.5-5.2)
0.50 (0.21-0.76)
0.45 (0.22-0.79)
0.07 (0.03-0.12)
. 8
-------
BACKWASHING
6. The filter backwash concentrate composites of the CSO and ROWS tests had
the following average charateristies:
BACKWASH CONCENTRATE CHARACTERISTICS
SS (mg/1)
Average Range
CSO
ROWS
2353
2215
(770-4064)
(1030-3570)
SDI
2.8
2.9
Filter Leaf Test
(% dry solids)
35
34
The backwash concentrate appeared amenable to dewatering.
water used for backwashing were:
BACKWASH WATER VOLUMES
Percent of Flow Filtered
Volumes of
CSO
ROWS
Average
6.0
8.3
Range
(2.4-7.8)
(5.2-12.5)
7. Complete fluidization of the filter media is necessary for effective
filter backwashing. An air supply of greater than 5 scfm/ft2 (1.5 m3/
is required for 2-4 minutes of media scrubbing. Backwash water
min/m-; is required ror e.-n minutes or media scrubbing. Backwash water
feed appeared satisfactory at 35 gpm/ft2 (87 m3/hr/m2) for 6-10 minutes
of media flushing. Full-sized plants will utilize less due to the ab-
sence of all wall effects that hinder backwash.
PRETREATMENT
8. The cross-hatch mesh rotary screening device used (trade name "Disco-
strainer") proved effective in removing grit and fibrous solids from
ROWS and CSO. The unit discharged sludge cakes of 12-18 percent solids
content, treating influents with SS levels of 100-300 mg/1. Having such
a relatively high solids content in the sludge cake alleviates the need
for further dewatering prior to ultimate sludge disposal.
During CSO testing, equipped w.ith a 40 mesh (420 micron) screen the unit
appeared to effect a fairly consistent 10 percent SS removal with an
average influent of 190 mg/1; with a 70 mesh (210 micron) screen ap-
proximately 16 percent SS removals were obtained with an average influ-
ent of 171 mg/1.
The Discostrainer sludge cakes appeared to contain high levels of heavy
metals but insufficient data is available to draw conclusions. Mass
balances are not possible since sludge cake volumes were not determined.
-------
9. Applications of the HRF system to sewage with heavy loadings of fibrous
solids should not use a slotted element screening device, since the
fibers tend ,to pass longitudinally through the screen. A true mesh or
cross-hatch screen is required to prevent early plugging of the filter.
ADDITIONAL FILTRATION RESULTS
10. Settleable solids were removed (97-100%-).by.HRF treatment.
average for CSO and ROWS tests were:
SETTLEABLE SOLIDS REMOVALS
The overall
CSO
ROWS
Filter
Influent
(ml/1)
4.7
3.1
Filter
Eff 1 uent
(ml/I)
0.06
0.0
Removal s
98.7
100
11. No conclusions can be made on heavy metals removal from CSO by HRF
treatment. There were no evident correlations between metals removal
efficiency and SS removal, chemical dosing or influent metals concentra-
tion. The results showed significant variation.
12. HRF treatment was not effective in the removal of dissolved solids, fil-
trate BOD and filtrate COD.
DISINFECTION
13. HRF without disinfection did not appear to reduce bacterial concentra-
tion.
14. A disinfection unit with a well-baffled chlorine contact chamber provided
adequate disinfection (to less than 200 fecal coliforms per 100 ml) of
HRF effluent (treated ROWS) at contact time of 9 minutes and a sodium
hypochlorite dosage of approximately 7 mg/1 (as Cl2).
FULL-SCALE CAPITAL COSTS
15. Estimated capital costs of dual-treatment filtration., facilities for 25
to 200 mgd (94,600-757,000 m3/day) capacity plants,were: (Capital cost
estimates are given based on a design filtration flux of 16 gpm/ft2 (40
m3/hr/m2) and including equipment for alum and polymer addition).
10
-------
SUMMARY OF TOTAL PROJECT COSTS
HRF
Plant Capacity
(mgd)
25
50
100
200
FULL-SCALE ANNUAL COSTS
Total
Capital Cost
(ENR = 2520)
$ 2,731,000
4,362,000
8,111,000
13,981,000
16. Total estimates annual costs for HRF plants of 25 through 200 mgd capa-
cities were:
SUMMARY OF TOTAL ANNUAL COSTS*
HRF
Plant Capacity
(mgd)
25
50
100
200
Annual Cost (450 hr/yrj
CSO
Treatment Plant
Annual Cost (8760 hr/yr)
Dual-
Treatment Plant
$
238,050
382,350
693,200
1,175,900
$
396,450
600,100
1,053,600
1,794,050
;annual costs are in the categories of separate CSO treatment plants
operating only during the estimated 450 hours of CSO per year and dual-
treatment plants operating 8760 hours per year (365 days including 450
hours of CSO filtration). The costs include amortization, operation and
maintenance.
11
-------
SECTION 3
RECOMMENDATIONS
HRF treatment should be evaluated on the basis of mass removals per
square foot of effective surface area or per cubic foot of media volume
as opposed to length of filter run since the total mass captured
appears to be the major factor determining the length of filter runs.
The HRF system does not remove dissolved material. Analytical
determinations of dissolved solids, filtrate BOD, filtrate COD and
trace metals should be made on influent flows to determine their rela-
tive contributions to the total loadings. If the dissolved fraction is si-
zeable, the applicability of HRF may be limited for dissolved material.
The current testing was performed on RDWS and CSO separately; it
appears desirable to evaluate the HRF system in an actual dual
treatment capacity going directly from RDWS to CSO treatment.
Both the 1971 Cleveland (EPA report no. 11023 EYI 04/72) and the
present New York City studies were performed at sewage treatment
plants with significant industrial comppnents in the flows. Further
testing should be performed on sewage from, predominantly domestic
areas which have a smaller component of dissolved organic solids.
Efforts should be directed toward the development of mathematical
relationships for simulation of the HRF process to mathematically
evaluate filter performance under varying SS concentrations and other
appropriate parameters.
On-site and drainage area rainfall and flowrates should be established
for accurately tj.me.d startups and for more efficient operation in CSO
treatment.
Large fluctuations in pressure head appear to cause solids shearing in
the filter media resulting in breakthrough of solids to the effluent
of the pilot plant; therefore maintenance of a'regulated hydraulic head
should be provided in full-scale facilities.
12
-------
A full-scale HRF installation of at least 25 mgd (94, 600 m3/day)
capacity should be engineered and constructed to demonstrate the
capabilities for dual-treatment of RDWS and CSO.
To provide additional process and design data on the HRF system,
further testing programs should include data collection in the
following areas:
a. Cost optimization of chemical additives (e. g., backwashing with
polymers to precoat the filter media and provide more
efficient sludge handling; heavy chemical feed rates during the
onset of filtration for rapid precoat formation to possibly
eliminate the need for continuous chemical feeding) to the filter.
b. Effective mixing techniques to provide maximum benefits from
chemical coagulant additives.
c. Effective backwashing procedures.
d. Backwash sludge handling and disposal, evaluating both dewatering
techniques and methods of ultimate disposal.
e. Declining rate filtration with a pilot plant specifically engineered
for this purpose, i. e., to insure maintenance of a constant
hydraulic head on the filter.
f. Utilization of the Disco strainer to concentrate backwash water.
g. Other areas of potential HRF application (e.g., effluent polishing,
treatment of urban runoff).
h. CMorination of CSO for preventing growths on the HRF media,
extending media life and establishing effective disinfection
procedures.
13
-------
SECTION 4
PILOT PLANT FACILITIES
GENERAL DESCRIPTION
The pilot plant was set -up at the Newtown Creek plant's No. 8 grit chamber
(see Figure 2). The influent sewage was pumped from a channel situated
after the plant's bar screens but before the grit chambers and passed
through a rotary screen (for grit and coarse solids removal) and then
pumped onto the dual media high rate gravity filter. The filter effluent was
discharged to the plant drain. Two storage tanks, installed between the
rotating screen and filter, were used during the shorter storm flow condi-
tions to extend filter runs with retained CSO. The same influent source and
process flow routing were used for treating both CSO and RDWS. All
equipment was located outdoors, exposed to the weather. A process
schematic of the pilot plant is given in Figure 3. Figures 4 and 5 show
photographs of the facilities.
The HRF column was fabricated from transparent lucite plastic to permit
observation of the filtration process (Figure 6). The filter media was
48 in. (122 cm) and later 60 in. (152 cm) of No. 3 anthracite over 30 in.
(76 cm) of No. 612 sand. The column was designed to simulate a full size
filter in both filtration operation and backwashing for the treatment of CSO
and RDWS. The filter bottom was of the same type as used in the Cleveland
study, i. e., a solid disc containing tubular nozzle* spaced to allow an even
flow distribution. Two 6-in. (15 cm) diameter acrylic columns were
installed for supplementary testing at various flux values and coagulant
feeds. Two chemical feeding systems were used for polymer and other
chemical additions. An in-line static mixer was installed late in the project
in the filter influent line just after the chemical feed connections. A disin-
fection contact chamber was available to receive a sidestream of the pilot
plant filter effluent for disinfection with sodium hypochlorite.
The flow through each filtration column was controlled by flow meter
observation and manual regulation of valves on each filter effluent pipe.
Pressure gauges were located along the height of the columns to profile
head losses throughout the filter depth. Backwash water was obtained
from a city water hydrant; an air gap was provided between the hydrant
14
-------
7W
r— 5>" / -*s —
-.-r--' Sq T
•^-" UJ-J
*r
S3
z
1
S
Ld
Z
Z
<
2
1,
zm
o
o
H-
O
I
CSJ
UJ
DC
^3
(D
15
-------
CD
CO
CO
LU
o
LLl
1-6
-------
(b) (c)
Figure 4. HRF pilot plant facilities, (a) Storage tanks and filter columns
during testing, (b) Lower section of 30 in. column; influent and effluent
samples, (c) 30 in. column operation and water sampling; disinfection
chamber in center foreground.
17
-------
Figures. HRF pilot plant facilities, (a) Discostrainer showing effluent
pipe and sludge discharge, (b) 30 in. and two 6 in. filter columns with
30 in. column effluent pipe in foreground, (c) Discostrainer interior
•while in operation.
18
-------
A 1 _ ,, H
-H
_T
£2
'
•0
_i
b
t
si
1
b
*
=o
_i
"cvl
1
-' ,, K
1
N
n
^
*~r
f*)
t
^^\
i
v/
L COLUMN
j i
*•— v;
^
.
•••••^••••••IMB
111 i]
>.Pl
} 1
^^^^ — :
, I
\
\
\
/
"
pj
^— — i
\ G^
A, i
DISTRIBUTION TROUGH
LUCITE FILTER INFLUENT-BACKWASH
EFFLUENT PORT W/5V SLFP-ON FLQ
•30V LUCITE
SLIP-ON FLANGE
(JYR)
? LUCITE
PLASTIC PIPE
LUCITE MEDIA PLATE
* FILTER EFFLUENT-BACKWASH
INFLUENT PORT M/6V SLIP-ON FLQ,
.^y LUCITE
AIR PIPE
-COLUMN 4-0"x^-O'
LUCITE BASE PLATE
II
FIGURE 6 - HIGH RATE FILTER
COLUMN
19
-------
and the filters to prevent contamination. Backwash air was fed from the
Newtown Creek plant's air system.
Upon completion of each filter run, valves were readjusted to allow the
filter influent pump to send clean (city) water upward through the filter
column for a backwash cycle. During backwashing the media was fluidized
and agitated with low pressure air at 45 psi (3. 2 kg/cm^) with a flux of
5 cfm/ft2 (1. 5 m3/min/m2). The backwash flow was conveyed directly into
an adjacent plant aeration tank.
PILOT PLANT EQUIPMENT
Pilot Plant Influent Pump
The pilot plant influent pump is a self-priming centrifugal pump, Barnes
Manufacturing Company, model 105-Cue-E with cast iron body, open type
impeller, 6 in. (15 cm) suction and discharge. The pump was driven by an
1175 rpm, 10 hp (7. 5 kW) TEFC motor, operating at 230/460 volts and the
pumping capacity was rated at 500 gpm (32 1/sr). The pump and motor were
mounted integrally on a base plate located directly over the influent waste-
water channel and about 3 ft (0. 9 m) above the water surface.
Rotary Screens
Two screening units were successively tested, both continuous duty,
gravity flow, self-cleaning systems manufacturer by Hycor Corp. The
units were located atop the CSO storage tanks 10 ft (3 m) above the level
of the influent source and 12 ft (3. 7 m) above ground level.
Rotating Slotted Drum Screen --
This unit, identified by the trade name Rotostrainer, used during the early
stages of testing, had a cylindrical slotted straining element rotating on
its horizontal axis and accepting incoming wastewater on its outside
surface. The slotted cylinder was constructed of wedgeshaped stainless
steel wire oriented with the line of revolution and supported by cross rods.
Interchangeable cylinders were provided, with either 250 micron, 500
micron, or 750 micron openings between the wires. The straining surface
for the model RSB-2510 Rotostrainer was 25 in. (63 cm) diameter and
10 in. (25 cm) wide. A soft metal doctor blade, in contact with the
rotating surface, removed the trapped debris. The strained wastewater
passed through the cylinder to a drainage line.
Rotating Mesh Disc Screen (Figure 7)--
This second unit, identified by the trade name Discostrainer, was supplied
after disuse of the Rotostrainer. It consisted of two vertical disc screens
rotating in parallel against a water-tight seal while partially submerged
20
-------
in. a box chassis. The waste-water entered between the disc screens at one
end and flowed radially through them to the receiving chambers at either
side.. The screens were washed with recirculated screened effluent by a
jet spray directed on the outer surface of the screens at the end opposite
the influent. The trapped solids collected at this end were gradually lifted
over an adjustable weir into a collection, box. The discs for the model
DS-110 Discostrainer were aligned 7 in. (18 cm) apart and had a diameter
of 39 in, (100 cm), and were supported by a stainless steel axle, spokes
and rim. The interchangeable discs had true cross hatch stainless steel
wire screens, and 420 micron (40 mesh) and 210 micron (70 mesh)
screen opening sizes were evaluated in the testing program.
Filter and Backwash Influent Pump
This horizontal centrifugal pump, manufactured by Gorman-Rupp Industries,
Model 14A2B-4 with cast iron body and open type impeller, 4-in. (10 cm)
suction and discharge. The pump was driven by a 1740 rpm, 7. 5 hp
(5. 6 kW> TEFC motor operating at 230/460 volts and the pumping capacity
was rated at 200 gpm (13 1/s). The pump and motor were mounted integral-
ly on a base plate located at ground level between the rotary screen and
pilot filter columns.
Filter Columns
Pilot Filter Column (Figure 6)--
This column was of 0. 75 in. (1. 9 cm) thick lucite'plastic molded into a
30 in. (76 cm) diameter (o. d.) open top tube. The filter cross section
area was 4. 5 ft2 (0.42 m2). The column was 18 ft (5,5 m) high, consisting
of four 4-ft (122 cm) sections over a 2 ft (61 cm) base section, all connec-
ted by bolted flanges. The column was guaranteed by the manufacturer to
sustain 75 psig (5. 3 kg/cm2g) internal pressure. The filter media was
supported by a lucite bottom plate containing multiple lucite nozzles.
Above the plate, an 18 in. (45 cm) gravel layer was provided to support
the anthracite-sand media and prevent media loss through the nozzles.
Four pressure gages were spaced along the column depth to measure head
loss from the overlying water through the anthracite, sand and the filter
bottom. An impact rotameter and butterfly valve were installed on the
filter discharge pipe for measuring and controlling the rate of flow.
Supplementary Testing Filter Columns--
Two columns, each 6 in. (15 cm) diameter, were made of transparent
acrylic tubing in four sections with a total height of 16 ft (4. 9 m). A top
plate with air and water relief valves allowed the filters to be pressurized
up to 30 psig (2. 1 kg/cm2g). The filtration flux was controlled by an indi-
cating rotameter and globe valve on the effluent pipes of each filter.
21
-------
iv ii'iiiiihi'ii', ,• ,171,1, ' v"V "V „ ni ' i;' i MI iniiiim! !|!i,illij^
FIGURE 7 - DISCOSTRAINER MODEL DS-110
22
-------
Chemical Feed Systems
Both systems consisted of a polyethylene 40 gal. (150 1) chemical solution
tank, a diaphragm metering pump with 0. 25 hp (0. 19 kW) motor, and a
separate input line to each filter. A portable Lightnin mechanical mixer
was employed for solution preparation. Two in-line static mixing units
were installed in series in the influent line to the filters just after the
chemical inputs. Each in-line mixer consisted of a 36-in. (91 cm) section
of PVC pipe, with flanged ends, containing stationary PVC helical vanes
along its length. The in-line mixers were supplied by Kenics Corp. and
the metering pumps were manufactured by Wallace and Tiernan, Inc.
Storage Tanks
Combined Sewer Overflow Storage Tanks --
Two tanks were provided for filter influent storage. One tank was of carbon
steel plate with a 5,000 gal. (19 m^) storage capacity; the other tank was
of fiberglass with a 8,000 gal. (30 m ) capacity. Each tank was equipped
with a mixer for preventing solids settlement during storage. Each mixer
was equipped with dual propellers and was driven by a 3 hp (2. 2 kW)
totally enclosed motor which was integrally mounted on a gear reduction
unit providing a shaft speed of 350 rpm. The mixer was manufactured by
Lighnin Mixing Equipment Company. The two tanks provided approximate-
ly 2. 5 hrs wastewater supply to the filter at 16 gpm/ft^ (40 m /hr/m^).
Backwash Water Tank --
An open carbon steel tank of 1, 000 gal. (3. 8 m ) capacity was used for
holding potable city -water for filter backwashing.
Piping and Valves
All piping and connections between equipment were of Schedule 40 PVC.
All flow control valves (butterfly, gate and globe types) were of cast iron.
Disinfection Unit
This was of carbon steel plate and consisted of a contact chamber 4 ft x 2 ft
x 18 in. (122 cm x 61 cm x 46 cm) high, fitted -with panels (baffles) of cor-
rugated fiberglass spaced 1. 5 in. (3..8 cm) apart to provide a serpentine
path for the disinfected filter effluent. A seven gal. (25 1) mixing chamber
was separated from the contact chamber by a V-notch weir. Sodium hypo-
chlorite solution was metered into the mixing chamber and rapidly mixed by
a 0. 05 hp (0. 037 kW) portable mechanical mixer. At a flow rate of 10 gpm
(0. 63 1/s) the approximate detention times were 40 seconds for rapid
mixing and 8 min for chlorine contact. The resulting mean velocity
23
-------
gradient, G, within the rapid mixing chamber is 1050 sec"* or, combined
with detention time (Gtd), is 42, 000. The baffled contact chamber,, with
a reported head loss of 1. 5 in. (3. 7 cm) (7):, had values of G and Gtd
of 25 sec" and 12, 000, respectively.
24
-------
SECTION 5
TESTING PROGRAM
OBJECTIVES
The testing program had three basic objectives:
1. Extend the results of the Cleveland program. (6) demonstrating the
technological advantage and treatment performance of HRF in
treating combined sewer overflows.
2. Determine the feasibility of an HRF system for dual functions,
treating both CSO and RDWS.
3. Establish treatment design parameters for application in full scale
filtration facilities.
Evaluation of the filtration process, the heart of the HRF system, was
the focus of the study. The screening devices (the Rotostrainer and
Disco strainer) were not evaluated towards optimization but were used
only to provide reasonably effective pretreatment for extending filter
operation periods.
LABORATORY ANALYSES
The principal laboratory analyses performed for the study were: total
suspended solids (SS), five-day biochemical oxygen demand (BOD) and
chemical oxygen demand (COD). These and other analyses were performed
on the influent, effluent and backwash effluent to provide more detailed
information on process performance. Table 1 is a list of all analyses
performed and the procedures utilized. All analyses were performed
by the NYC Industrial Wastes Control Laboratory at the Newtown Creek
Water Pollution Control Plant within 24 hours after each test. Trace
metals were analyzed later after preservation by acidification and
freezing.
The major parameter for determining filtration effectiveness was SS,
25
-------
TABLE 1. LABORATORY ANALYSES
Analysis
Method Used
Solids
Suspended Solids (SS)
Volatile Suspended Solids (VSS)
Settleable Solids (Set. S.)
Organic
Biochemical Oxygen Demand (BOD)
Chemical Oxygen Demand (COD)
Filtrate Biochemical Oxygen
Demand (FBOD)
Ultimate Biochemical Oxygen
Demand (UBOD)
Filtrate Chemical Oxygen Demand
(FCOD)
Filtered through glass fiber filter
(Whatman GF/C) and dried at
103-105°C.
Filtered and ignited at 550°C.
Imhoff cone, by ml/1.
Unblended, diluted and incubated for
5 days at 20°C.
Oxidized by K^Cr^O^ solution.
Filtered through Whatman No. 1
paper, diluted and incubated for
5 days at 20°C.
Unblended, diluted and incubated
for 20 days at 20°C.
Filtered through Whatman No. 1
paper and oxidized by
s olution.
Trace Metals
Cadmium, Chromium, Copper,
Lead, Nickel, Zinc
Mercury
Bacteriological
Total Coliforms
Fecal Coliforms
Acidification, conventional atomic
absorption
Acidification, cold vapor atomic
absorption
Multiple dilution
Membrane filter (pore size 0.45 micron)
Special Sludge Tests
Gravity Thickening
Dewater ability
Sludge volume index (modified)
Buchner filtration vs time;
Filter leaf test.
26
-------
since the system is essentially a solids removal process. Insoluble
BOD, COD and trace metals are also removed to some degree by the
filters. Phosphate removal by the HRF system was not determined due
to the low levels found in the Newtown Creek plant influent. The state-
wide limit on phosphate content in detergents has greatly reduce
phosphorus concentrations in all New York City plant influents.
PROGRAM SCOPE
The testing program consisted of two separate phases. Phase I involved
filter testing of CSO for a minimum of 10 separate rainstorms over the
spring, summer and fall seasons. Phase II involved the continuous
filter testing of RDWS over a minimum seven day period of no precipita-
tion, the testing was performed 24 hours per day with interruptions only
for filter backwashing. The 30-in. (76 cm) HRF column was utilized
in all CSO and RDWS testing and in all testing for establishment of
full-scale treatment design parameters.
Selection of rotary screen mesh size, filter media, types of polymer and
coagulant were based on the results of the previous HRF testing at
Cleveland (6). The 6-in. (15 cm) filter columns were utilized only to
establish flux values and chemical feeds for the primary tests with the
large column.
PILOT PLANT
Screening
The Cleveland testing concluded that a 40 mesh (420 micron) screen
aperture was satisfactory when utilized in conjunction with dual media high
rate filters. Accordingly, the Rotostrainer was used with either of two
slotted cylinders having openings equivalent to 35 mesh (500 microns)
and 60 mesh (250 microns). The Discostrainer unit was equipped with
40 mesh (420 micron) and 70 mesh (210 micron) screen sizes.
Media
The filter media used was the same as that found satisfactory in^Cleveland
(6) - anthracite over sand - with the following specifications:
Material
Effective Size Uniformity Coefficient Depth
No. 3 Anthracite
No. 612 Sand
3. 85 mm
2.0 mm
1.52
1.32
48 or 60 in.
(122 or 152 cm)
30 in. (76 cm)
27
-------
Chemicals
Chemical filtration aids included alum and several types of organic poly-
electrolyes, equivalent to those found satisfactory during previous CSO
treatment studies (6, 7, 8).
Polymer
Purifloc A-23
HercoGoc 1054
Zetag 92
Betz 1150
Magnifloc 561C
WT 2575
CSO TESTING
Type
Manufacturer
Anionic
Anionic
Cationic
Cationic
Cationic
Cationic (liquid)
Dow Chemical Co.
Hercules Inc.
Allied Colloids Ltd, Engld.
Betz Laboratories
American Cyanamid Co.
Calgon Co.
The pilot plant was to be activated for testing when storm runoff increased
the flow to the Newtown Creek plant above the normal dry weather flow.
A minimum of 24 hr of antecedent dry weather were considered necessary
for a valid CSO test to allow deposition in the collector systems. CSO
was to be tested at any time of the day and, later in the project, any
time of the night. Filtration testing was conducted without chemical
additions, with polymer alone, or with polymer and alum in combination.
During filter operation, observations were made on the filter flux and on
gauge pressures along the depth of the filter column. Grab samples
were taken of the plant influent (Pi) ahead of the rotary screen and
before the plant grit chambers, filter influent (Fi) after the screen,
and of filter effluent (Fe). The observations and samples were taken
and laboratory analyses performed according to the schedule given on
Table 2.
Following the same schedule as the grab samples, composite samples
were taken of Pi, Fi and Fe and analyzed for SS, VSS, BOD, FBOD,
COD, FCOD, UBOD, Set. S. and trace metals.
The majority of the tests were conducted at a constant filtration rate.
Filtration was ended when the rate decreased by at least 50 percent (this
usually occurred suddenly within the last half hour of testing) or when the
filter effluent was of visually poor quality, indicating solids breakthrough
in the filter. However, two runs employed a declining filtration rate
in which the flux was not regulated by adjusting the effluent valve but
was allowed to decline under a constant, regulated head.
28
-------
TABLE 2. CSO SAMPLING AND ANALYTICAL SCHEDULE
Time After Start of Filter Operation
* 0
15
30
60
90
120
150
180
210
240
270
300
330
360
390
420
450
(Minutes)
Obs ervations
n
M
n
n
it
n
M
it
ii
it
n
n
n
n
ii
ii
Analyses
480
SS BOD FBOD COD
SS BOD FBOD COD
SS BOD FBOD COD
SS BOD FBOD COD
SS
SS BOD COD
SS
SS
SS
SS BOD FBOD COD
SS
SS
SS BOD
SS
SS
SS
SS
SS BOD
COD
COD
* The 0 minute samples were taken about 2 minutes after the start of
filtration to allow flushing of water remaining in column.
29
-------
BACKWASHING PROCEDURES
During filter backwash cycles attempts were made to determine the most
effective operational procedure, minimizing the volume of backwash water
necessary for filter cleaning. In general, volumes of low pressure air, 5
cfm/ft2 (1. 5 m^/min/m ) at 45 psi (3. 2 kg/cm2) were used to scrub the
O
media and loosen trapped solids early in the cycle. A low rate, 8-15 gpm/ft
(20-37 m /hr/m2) of water, fed with the air or in sequence with air would
aid in the scrubbing process. The air feed was stopped before the backwash
fluid reached the level of the discharge pipe, and water alone was fed at a
high rate, 22-35 gpm/ft2 (54-87 m^/hr/m2) until the media was flushed.
clean, with a distinct separation of the anthracite and sand layers.
Samples of the backwash effluent were collected at the point where the
effluent discharged into an aeration basin of the Newtown Creek Plant.
Samples were taken at maximum 30-second intervals throughout the
duration of dirty backwash discharge. Both composite and grab sampling
methods were used to estimate the total amount of material trapped on the
filter during each run and to profile backwash solids concentration with
time. Analyses of SS were made on all backwash samples, and the
composites were also tested for VSS, BOD, COD, Set. S. , trace metals,
gravity thickening and dewaterability.
The Discostrainer had a continuous backwashing system incorporated into
the unit. The sludge from the strainer was sampled and analyzed for trace
metals and dry solids content.
RDWS FILTER TESTING
This phase of the testing program utilized the same operating procedures
as the CSO testing. Due to the concern for achieving seven consecutive
dry-weather days for filter testing, as required under the project work
plan, the RDWS phase of the sampling program was scheduled to be
completed first in the project.
During RDWS testing, the plant influent, filter influent and filter effluent
were sampled at hourly intervals and analyzed for SS, BOD and COD.
FBOD analyses were also performed on samples taken every second
hour. Composite aliquots were taken with each sample and analyzed upon
completion of each filter run as was done under CSO testing. Each filtra-
tion test proceeded until the head loss reached 5-7 psi (0. 35-0. 50 kg/cm )
(bottom gauge pressure 0) or until a solids breakthrough occurred. The
filter was then backwashed and a new filter run begun. Testing occurred
24 hours per day over a 9-day period. Various flux and chemical feeds
were evaluated during this period.
30
-------
ADDITIONAL, TESTING
Metals
The 6-in. filter columns were utilized for special metals removal tests
early in the program to determine the extent of the precipitation and
coagulation of metals in the filter. In parallel operation, one filter
received influent flow adjusted to pH 8. 5 with lime addition and polymer,
the other filter received flow with only polymer addition. The sampling
followed the same procedures as in CSO testing. Analyses were
performed for copper, chromium., cadmium, zinc, nickel, lead and
mercury.
Bacteriological
Bacteriological testing was conducted to evaluate bacterial removals by
filtration and to evaluate the disinfection unit.
The filter testing involved sampling of plant influent and effluent during
selected RDWS and CSO tests. Analyses were performed for total and
fecal coliforms to provide estimates of removals across the HRF system.
The disinfection testing was performed during selected HRF tests by
diverting a 10 gpm (0. 63 1/s) sidestream of filter effluent to the disinfec-
tion unit. A controlled gravity feed of 12 percent sodium hypochlorite
solution entered the influent chamber and was mechanically mixed with
the filter sidestream before passing into the corrugated baffle contact
chamber. Dosages equivalent to 5, 10, 15 and 30 mg/1 chlorine were
added in succession, with each dosage period lasting about 15 min. After
an initial 10 min. for solution equilibrium, samples of contact chamber
effluent were taken at the 10 and 15 min. points. An influent
sample (filter effluent) was taken at the 10 min. point of each dosage
period.
Concurrent with each dosage period the chlorine residual was measured
with a visual chlorine comparator. Residual chlorine in the contact
chamber effluent samples was eliminated by sodium thiosulfate crystals .
present in the sterilized sample bottles. The influent and effluent
samples were analyzed for fecal coliforms immediately after the
•disinfection testing period, utilizing the membrane filter technique.
31
-------
SECTION 6
TESTING RESULTS
OPERATIONAL EXPERIENCE
The test period extended from October 1975 through June 1977. HRF
pilot plant evaluations were performed on 18 storm generated CSO .and
on 29 filter runs with RDWS. The complete data file can be found in the
appendix.
Since filter operation generally was effective for 4 to 1.0 hours before
backwashing was required, relatively few CSO events lasted long enough
to adequately evaluate CSO treatment. This, coupled with the uncoopera-
tive weather patterns of 1976-1977, mechanical breakdowns and the
misapplication of the original screening unit ahead of the filter, all
contributed to the project's rather lengthy test period.
During the first few weeks of testing, the Rotostrainer was used for
screening with retention of the screened flows in the storage tanks. The
retained flows were kept well mixed and then pumped to the filter columns.
Twenty-four tests on RDWS were performed during nine consecutive
days in October 1975 (tests Dl-24). The results are discussed later in
this section under RDWS testing. Three CSO events (Sl-3) were then tested
in November 1975,, but experienced poor filter performance, excessive
head loss and filter blinding within 0. 5 to 2 hours. The blinding
appeared to be caused by a mat of slimy and fibrous material which
formed on the surface and within the first few inches of the filter media.
When this layer was pierced, the filter was able to accept flow normally.
This problem did not occur during the continuous RDWS testing.
Because of the operating conditions of the pilot plant, after the RDWS
tests and between storm tests, some sediment and water (4-6 in. after
draining) remained in the bottom of the storage tanks in a light-deficient
environment ideal for slime propagation. This material, in combination
with the normally heavy fiber load to the Newtown Creek plant and the
heavier sediment load received during the first flush portions of storm
flow, may have encouraged the rapid formation of a relatively impervious
32
-------
mat on the filter soon after the screened CSO was pumped from the
storage tanks. The filter blinding was apparently caused by these
operating conditions of the pilot plant and would not occur in full-scale
HRF facilities where open storage tanks, if any, would be used.
To prevent a recurrence of the problem,, thereafter the storage tanks
were thoroughly flushed and the filters and tanks were chlorinated after
each complete CSO test. This procedure worked well in controlling, the
slime; however, while the duration of subsequent filter operations
increased to a degree, fiber matting continued to be a problem. It
later became evident that the slime problem had masked the ineffective-
ness, of the Rotostrainer in both solids and fiber removal..
In July 1976 a test was made with a 40 mesh screen (two-directional
mesh), placed at the Rotostrainer effluent pipe to trap any coarse solids
which may have passed through the 60 mesh slotted (unidirectional)
drum. The 40 mesh screen clogged completely in less than, a minute,
showing that the slotted strainer is not suited for treating sewage with
high concentrations of fibrous material since the fibers tend to pass
longitudinally through the slotted element.
Testing was then delayed until September 1976 when another replacement
unit (the Discos.trainer) became available.. The Discostrainer incorporated
a two-directional mesh screen and had the same piping arrangement as
the Rotostrainer. None of the ten CSO tests after installation of the Disco-
strainer experienced fiber matting and all lasted 3 hours or longer. These
teats (S9-S18) are considered representative of HRF performance and are
the most useful for process evaluation. Five tests on RDWS (D25-D29")'
were conducted after installation of the Discostrainer in an attempt to
correlate the HRF system performance with previous tests using the
R otostrainer.
In the fall of 1975, it became evident from the data that there was a lack
of correlation between the characteristics of the plant influent grab
samples, and the filter influent and effluent samples at any period during
a filter run since the filter influent came from the well mixed storage
tanks. To eliminate this disparity, the pilot plant process flow system
•was modified to provide piping directly from the screening discharge line
to the suction line of the filter influent pump. The standpipe by-passing
the storage tanks was kept flooded during filter operation and overflowed
into the storage tanks. Thusr the filter influent sample, with a few
minutes lag, represented the same sewage as the plant influent,
resulting in better correlation of grab sample data. The overflow to the
storage tanks continued to be used to prolong the filter runs.
33
-------
There were other operating problems resulting from the intermittent use
of the pilot plant. Growths of algae and slime in the filter column between
tests had to be controlled by a 5 gal (19 1) dose of 12 percent hypochlorite
solution prior to each test. Difficulties with motors, pumps and broken
piping caused several storm tests to be missed. The pilot plant was out
of service during the winters of 1975-76 and 1976-77 and breakdowns
were most severe during spring plant startup due to ruptured piping.
Two improvements to the pilot plant were made in March 1977. Additional
media was provided for a final composition of 60 in. (152 cm) of No. 3
anthracite over 30 in. (76 cm) of No. 612 sand. A static in-line mixer
was installed in the filter influent line to provide more thorough dispersion
of chemical feeds. The final CSO tests (S14-S18) and RDWS test D-29
were performed after these changes. To keep within the project budgets
until May 1976, the storm "alert" period, during which the pilot plant
would be manned and operated, was set at normal business hours. Since
this schedule resulted in only one in three storms being tested, the alert
period was extended in May 1976 to 24 hours on weekdays, and in June,
to 24 hours, 7 days/week, until the CSO testing was completed.
The Newtown Creek plant process, flow quantity and quality are atypical
of most large treatment facilities due to the relatively large industrial
flow component and the absence of primary treatment. The BOD and
COD removals effected by the HRF system reflect this high background
industrial component and should be cautiously reviewed in the design of
a full-scale HRF facility. The SS removals data, however, are repre-
sentative and should be considered in that light.
CSO TESTING RESULTS
The results of eleven CSO tests with filter runs of three hours or more
are given in Tables 3 through 7. All tests utilized constant rate filtration
at 16 gpm/ft2 (40 m3/hr/m ) except tests S-17 and S-18 which used
declining rate filtration and are discussed separately. All tests were
begun within two hours after the plant flow increased above dry-weather
levels.
The variability of flow of each storm event is an apparent and important
factor in the evaluation of CSO results. Based upon an analysis of storm
data records from 1948-1975 at the 16 recording rain gages in the
New York City-New Jersey metropolitan area, the average storm in
New York City has the following characteristics (9); duration - 6.5 hrs,
accumulation - 0.375 in. (0.95 cm), intensity - O."056 in./hr
(0, 14 cm/hr), and duration between storms - 77 hrs (3. 2 days). From
Table A-20, which gives characteristics of each storm during tests
S-5 through S-18, it is seen that storms S-13, 14, and 16 were of high
34
-------
intensity, and the other storms tested were moderate to low intensity
rainflows.
From the results it is apparent that a fairly consistent effluent suspended
solids quality can be effected by the filtration system and that efficiency of
removals is directly related to the influent SS concentration. Characteri-
zation of the sewage solids should be a major criterion in the design and
success of an HRF system.
The overall results for SS, when averaged for the nine tests using constant
rate filtration were: plant influent 182 mg/1, filter influent 161 mg/1, filter
effluent 62 mg/1, for SS removals of 61 percent across the filter and 66
percent across the HRF system.
Grouping the constant rate tests by modes of operation (Table 3)--without
chemical feed, with polymer alone, and with polymer and alum--the
average SS removals (across the filter/system) were 55 percent/62 percent,
63 percent/67 percent and 67 percent/69 percent, respectively.
Polymers enhance removals by flocculation, by more rapid formation of
the filter pre-coat or by a combination of both. The results of the constant
rate filtration tests show that the cationic Betz 1150 gave the best results
of the polymers tested. When used alone (test S-13) this polymer effected
68 percent removal of SS and when utilized in combination with alum (tests
S-14, 16), averaged 72 percent SS removal across the filter.
The influent concentrations should be evaluated in addition to removal rates
to provide an accurate picture of HRF process performance since higher
influent concentrations allow greater removal efficiencies. This can be
shown from the results of eleven representative CSO tests as given on
Table 4. The four CSO events occurring on a weekend averaged 51 percent
filter removal of SS while the remaining events, occurring on weekdays,
averaged 68 percent removals. The length of filter run and the filter
effluent quality (66 mg/1 weekend, 58 mg/1 weekday) averages were
essentially the same for each period. Since average background influent
SS loading on weekdays was much higher (179 mg/1) than that on weekends
(135 mg/1), much of the apparent difference in filtration efficiency can be
attributed to the higher influent strengths on weekdays. This finding is
supported by the results of the Cleveland CSO testing (6).
The final two CSO tests were performed in a declining rate filtration mode;
test S-17 used Purifloc A-23 polymer alone and S-18 used Betz 1150 poly-
mer in combination with alum. The initial flux was 24 gpm/ft2 (60 m3/hr/
m ) dropping to 13 gpm/ft2 (32 m3/hr/m2) for test S-17 after six hours,
and to 7 gpm/ft2 (17 m3/hr/m2) for test S-18 after 10 hours. The poor
35
-------
TABLE 3. CSO SS REMOVALS VS CHEMICAL ADDITIONS
Chemical
Dosage Duration
Run Nbj
fmir/lV n
irl
Plant
Influent
(mg/l)/(lb)
Filter
Influent
(mg/lj/ (lb)
FUter
Effluent
(mg/1)/ (lb)
Removal
Filter
( % )
Removal.
System
C % >
Without Chemicals
S-4&
S-<)
S-10
S-15
Average
0
0
0
0
,3.0
3.0
5.5
,8.0
.5. 0
112/12.1
152/16.4
189/37.4
198/57.0
175/30,7
105/11.3
136/14.5
149/29.5
172/49.5
150/26.2
30/3.2
61/6.7
60/11.9
87/25.1
67/11.7
71
55
60
49
55
73
60
68
56
62
With Polyeleetrolyte
S-ll
S-17*i*
S-13
Average
With Alum and
S-12
S-14
S-16
S-18*
0. 5 polymer**'.
1.4 polymer
1.0 polymer
Polyeleetrolyte
15 alum
0.5 polymer
35 alum
2 polymer
17 alum.
1.3 polymer
6..0;
6.0
6.5
6.0
4.0
3.0
4.0
15 alum 10.0
1.0 polymer
99/21.4
276/74.6
311/72.8
209/47.1
133/19.2
185/20.0
147/21.2
174/62. 7
94/20. 3
Z48/67.0
266/62.3
183/41.3
122/17.6
192/20.7
126/18.2
158/56.9
49/10.6
110/29.7
85/19.9
68/15.3
55/7.9
58/6. 3
31/4.5
67/24. 1
48
56
68
63
55
70
75
58
51
60
73
67
59
-
79
61
Average
4.0
152/20.1
143/18.8
47/6.2
67
69
The average values were calculated by weighting each filter run by its duration. g.
* Declining rate filtration tests, S-17 had a 20 gprn/fT average flux, S-18 a 16 gpm/ft average flux;
these tests were not included in the averaging of SS data. All other tests were constant rate
filtration, 16 gpin/ft2 flux. " .
«* Polymers used were: for S-ll and S-12, Hercofloc 1054 {anlonic); for S-17, Purifloc A-Z3 (anionic);
and for S-13, S-14, S-16 and S-18, Beta 1150 (cationic).
36
-------
TABLE 4. CSO SS REMOVALS WEEKDAY-WEEKEND COMPARISON
RllTl NO.
Chemical
Dosage
(mg/1)
Duration
-------
results with declining rate filtration appear to be due to the design of the
pilot plant which did not allow maintenance of a constant pressure head
in the gravity filter. Sudden surges in head forced solids to pass through
the filter (to shear). An overflowing standpipe ahead of the filter would
have been necessary to maintain a constant pressure head. Tests S-17
and 18 therefore have not been used in calculating average results.
Figure 8 shows HRF treatment efficiencies based on SS mass removals
or capture per square foot of filter surface for the CSO tests. It is
apparent from the figure that test S-13 provided superior mass removals
as compared to the other tests, giving a strong indication of filter
operation under severe stress. The data from S-13 appears to be the
most significant collected from all the CSO tests since during this storm
the upstream sewer regulators remained open, channeling most of the
1. 15 in. (2. 9 cm) rainflow directly to the Newtown Creek plant.
Furthermore, S-13 was the only storm in which the precipitation was
constant and intense throughout the filter run. During all other
storms tested, the rains were of either short duration or of low
intensity preventing a thorough stressing of the HRF system.
Table 5 presents average SS mass captures obtained across the filter
during all CSO tests of at least 3 hours duration (excluding S-17 and
S-18) and the averages for tests S-13, 14 and 16 which used more
optimal chemical feeds and occurred during the storms of greatest
intensity.
TABLE 5. AVERAGE MASS CAPTURE OF CSO SS
Capture per Filter Surface Capture per Media Volume
lb/ft*/run*lb/ft^/hr lb/ftj/run** Ib/ft^/hr
3.7
5.2
0.76
1.2
0.54
0.76
0. 11
0. 17
CSO Test Nos.
S4B, S9-16
S-13, 14, 16
* 1 lb/ft2 = 4. 88 kg/m2
*# 1 lb/ft3 = 16.02 kg/m
Settleable solids removal was consistently better than 95 percent in the
filter (see Table 6). The influent levels ranged from 1.0 to 6. 0 ml/1
and effluents from 0. 0 to 0. 2 ml/1 in composite samples from seven
CSO and two RDWS filtration tests.
38
-------
K) | — l^^S?
>
^^
"•" 1 IX\.
,+
, 1 ft^SSisN
-y tO^
5 CO
•—
_., -,. f\S r^^
i\x\x I r VJLr
HJ "T" I J^jsS
1 — ^
' j ^^
ii- ™— ^" i '•^^s^
air **
UJ UJ CD,. . ..— .
no ' '- K^
H- H- ^O
LU til U>
rr fv TL R^5^
— — ' •«
Hi- xa
o O
^ ^ PQ
Uf 1 rt*r
u yr ft^
t/^ lO rf\
ta«a —
• i |^-
•
O LU
-in ^L
OJ *~*
Q
LU
CC
C£
LU
^-^ Kj
o "oS "-
-O C l-
CVJ v~> "
t/3 -^
" «* a:
L_ LU
- z °-
^ [3
O cc
-J =3
^- £
o o: w
"~IO UJ •*
_J 00
— oo
"•
00
oo
111
cc
ID
iT
g
HH/^d/qi)3Hnj.dVO
39
-------
TABLE 6. SETTLEABLE SOLIDS REMOVALS
Plant Influent
Run No.
S10
Sll
S12
S13
S14
S15
S16
S17*
S18*
S28
D29
5.0
3.0
4.0
5.5
4.5
4.0
4.0
7.0
5.0
2.5
Filter Influent
(ml/1)
2.0
0.5
2.0
2.2
6.0
2.0
1.2
3.5
1.0
1.7
4.5
Filter Effluent Filter Removal.
(ml/1) ( % )
0.0
0.0
0.0
0. 1
0.2
<0.1
0.0
0. 1
<0. 1
0.0
0.0
100
100
100
95
97
>95
100
97
^90
100
100
* Results from runs S17 and 18 (declining rate filtration) are not
used in data averaging (see pp.35-3-6)
S - indicates CSO test
D - indicates RDWS test
Organic pollutant removals across the filter are shown in Table 7, BOD
removal trends are consistent with the SS removals, 32 percent average
removal for all tests (29 percent for weekend flow tests, 35 percent for
weekday flow tests) or averaged for the three modes of operation (no
chemical, one chemical, two chemical aids), 27 percent, 35 percent and
38 percent, respectively. COD removals (42 percent general average)
also varied for the three modes of operation, 32 percent, 46 percent,
50 percent, respectively; and weekend versus weekday showed some
change, 37 percent and 41 percent, respectively.
The results of composite samples were not used in the calculation of
average values of SS, BOD and COD. SS results were calculated by flow-
weighted averaging of the results of individual grab samples. BOD and
COD results were from averages of the grab samples. Composite
sample results, for SS especially, were often at variance with the flow-
weighted averages. Composite sample results are given in Table A-5.
The variation of CSO characteristics is further indicated by the fact that
individual grab samples of filter influent often had greater SS concen-
trations than plant influent samples.
40
-------
TABLE 7. CSO BOD AND COD REMOVALS
BOD
Run
No.
Chemical Filter Filter
Dosage Influent Effluent
I mg/1) (mg/1 ) ( mg/1 )
Removal
( % )
Filter
Influent
(mg/1 )
COD
Filter
Effluent
(mg/1)
Removal
( % )
Without Chemicals
S-4B
S-9
S-10
S-15
With
S-ll
S-17
S-13
With
S-12
S-14
S-16
S-18
0 132
0 126
0
0 134
Average 131
Polyelectrolyte
0.5 polyme'r* 119
* 1.5 polymer 153
1.0 polymer 139
Average 129
Alum and Polyelectrolyte
15 alum 67
0.5 polymer
35 alum 97
2 polymer
17 alum 90
1.5 Polymer
* 15 alum 192
1 . o polymer
Average 85
98
91
99
96
79
111
90
84
49
53
58
106
53
. «*.« C 1 •?
26
28
26
27
34
27
35
35
27
45
36
45
38
297
226
350
267
285
205
456
385
295
190
312
223
280
242
192
161
251
170
193
132
263
187
159
98
136
125
182
120
35
29
28
3fi
32
36
42
51
46
48
56
44
35
50
S-18 a 16 gpm/ft2 average flux; these tests were not included in the
averaging of data. All other tests were constant rate filtration,
I6gpm/fl? flux.
**Polymers used were: for S-ll and S-12, Hercofloc 1054 (anionic);
for S-17, Purifloc A-23 (anionic); and for S-13, S-14, S-16 and
S-18, Betz 1150 (cationic).
41
-------
Ultimate (20 day) BOD were analyzed from composite samples only and
filter removals were comparable to BOD removals in the same
composites, ranging 27-39 percent (Table A-6). Chemical dosing
improved removals but there was no distinction among types of polymer.
The UBOD values were generally 2 to 5 times the corresponding BOD
(five -day).
Results for FBOD and FCOD indicate no significant removals of
dissolved organic material by filtration with or without-chemical aids.
The filtered values ranged 25-50 percent of the total BOD and COD
values; composites {Tables A-5 and A-6) and grab samples (A-7) showed the
same trends. In the CSO and RDWS tests there was no apparent relation
between filter removals of SS, BOD or COD .and proportion of dissolved
organic material.
VSS data appears on Table A-8. The percentage of volatiles varied
widely in the influent and effluent samples. The results appear
scattered and inconclusive. In future HRF studies, VSS removals
should be evaluated more thoroughly.
RDWS TESTING RESULTS
Twenty-four consecutive filter runs (Dl-24) operating continuously were
conducted with RDWS on the 30-in. filter column in, October 1975.
Table A-2 lists the results of each run, and Table 8 contains average
results grouped by chemical feeds and flux.
The results of the consecutive runs indicate that at a flux of 8 gpm/ft2
(20 m3/hr/m2) without chemical feed (Dl-4) and 12 gpm/ft2 (30 m3/hr/
m2) with 0. 5 mg/1 of polyelectrolyte (D5-14) average SS removals of
67 percent can be attained across the filter. Individual filter test SS
removals ranged from 60 to 74 percent at 8 gpm/ft2 (20 m3/hr/m2) and
50-77 percent at 12 gpm/ft2 (30 m3/hr/m2). BOD removals across the
filter at these flux values averaged 30 percent and 39 percent, respecti-
vely. Five filter runs (D15-19) were conducted at the flux of 12 gpm/ft2
(30 m3/hr/m ), with 0. 5 mg/1 polyelectrolyte and 10 mg/1 alum feed.
Filter removals averaged only 64 percent for SS and 23 percent for BOD.
Thus the alum-polymer addition did not provide improved percentage
removals over use of polymer alone. This conclusion is supported by
the results of the final RDWS test with the same flux and chemical feed
(test D-29 effected 62 percent removal of SS and 36 percent removal of
BOD).
2 3
Five tests (020-24) were conducted at 12 and 16 gpm/ft (30 and 40 m /hr/
m2} flux values. The results indicate lower SS removals at the higher flux;
42
-------
1
^_J
EH
i-t
P
P
^»
.,
O
i— i
H
ffl
U
P
£?;
-H
•J
o
O
CN
%
£
a
1
H
E
V
h
£
1
a.
n
S
£.
o
•f-l
to
rt
S
1
• H
fe
'.
.H
h
•n
1
ft
3
&
3
a
"g
61
,_
^
CO
CNl
in
o--
o
*
o
o
CO
t-
in
in
-H
CM —i O^OOCTN ^m CM in
1 CO CO CM^CO COCO CO CO
cr» co o in—^oo coo NO ON
NO o co o in o oo CM co —<
CO O CO -^O^P OO CO CM
O «-H CO O O O "^ ON O^ i-H
^ ^i CTN cocor- r-vo --H r*
com omcorOTjtin
^* 00 CO ON CO Is* CO ON
t*- ^ m co oo in ~* tM tn o
NO NO tnNOinm NO \o NO co
^-t co co r-i—tin CTNNO to o
^ in r- cocor* com^^ co
in t*- m Nocot»- ^HND ^H oj
CM-^J* t— fHONNO O^J* CM in
f
*
NOON fM^*^Tj»com
to Is- incocONO-N^r—
o o o m ON o oo o o
— 1 ^-* CM —1 ^H CO
* * t t 8 £ ' S
m 2k m ^ m *. o*. oJLm*^m*.
o* S o" 5 o fe o o .-: & o * £ o* 55. o E
CM fM NO CMNOCM CMCM NO — «
CO O O OOO OO O O
m %o tn corpso co tn T)1 co
^
*tT
'3
g
*•*-»
nt
U
o
in
N
4)
m
1
«
"u"
'c
o
•s
i
co
CNl
' ^J
"*< a
o g
O ^
SJ*"J
f^
H K
S|
A E
* 1
*tT
CQ
h
4^
*H
g
vS
0)
rt
J
PQ
-------
however, chemical feed and flux varied widely among tests. Figure 9
presents SS removals vs. flux for the RDWS runs grouped by chemical
feed.
Five subsequent tests (D25-29) were performed in 1976 and 1977
attempting to correlate the filtration data using the different screening
devices and to evaluate the screens themselves. The tests were
conducted at varying flux and chemical feeds; the filtration data appears
scattered and is therefore presented on Tables 8 and A-2 but not further
interpreted.
The data collected across the screening device (the Discostrainer)
indicate that SS removals of 18 and 23 percent were attained with the 70
mesh (210 micron) screen (tests D25 and 26) and 11-19 percent removals
were attained with the 40 mesh (420 micron) screen (D27-29).
Throughout the RDWS testing, the filter effluent SS concentrations averaged
49 mg/1 and within the range of 30-75 mg/1 for 86 percent of the tests.
All consecutive runs at 8-12 gpm/ft2 (20-30 m3/hr/m2) had filter effluent
levels below 75 mg/1 and averaged 44 mg/1.
•Figure 10 graphically shows HRF treatment efficiency based on SS mass
capture per square foot of filter surface for the RDWS tests. The data
is also presented, on Table 9, categorized by flux and chemical feed and
including mass capture per cubic foot of media volume.
COD results are listed in Tables 8 and A-2. Removals averaged 32
percent and ranged from 20 to 53 percent. As has been noted before, the
Newtown Creek plant influent has a large industrial component and
cannot be considered typical of municipal sewage with respect to COD.
BACKWASHING/CONCENTRATE EVALUATION
The most effective operating procedure tested for backwashing; required
the use of air in combination with water at a rate of 12 gpm/ft
(30 m^/hrVm ) for 2 to 4 min of scrubbing, and then water alone at
35 gpm/ft (87 m /hr/m ) for 6 to 10 minutes of flushing and backwash
discharge. This water flushing rate was the maximum obtainable using
the backwash pump. The specific duration of backwashing and volume of
water depended on the loading of trapped solids from each test. Effective
cleaning could be obtained by the use of backwash water volumes of
6 percent of the wastewater filtered in an average run. Tables A-3 and
A-4 give data for CSO and RDWS runs, respectively. Full-scale HRF
facilities would use filter effluent or secondary treatment effluent for
filter backwashing for most economical operations. This is the common
44
-------
100
1
LU
90
6O
50
I
NO CHEMICAL FEED
r—~ ALUM, NO POLYMER
• •••• ••• 0.5 mg/l POLYMER, NO ALUM
0.5 mg/l POLYMER, 10mg/l ALUM
(3) NUMBER OF RUNS AVERAGED
FLUX (gpm/ft2)
FIGURE 9-RDWS SUSPENDED SOLIDS FILTER REMOVAL AVERAGES
45
-------
or,
x t^
irp^z:
LUUJZD
DID UJ
E£b
en to
OJ
6
OJ
Q
51
i
10
P
cvl
o
LO
Q
O
HO
OJ
cry
J,
to
H
LU
O 3
-LO JJL
z
ft:
in
H
LU
Q
Z
LU
OH
LU
o:
LU
O.
CO
LU
0.
<.
O
en
to
GO
oo
oo
CD
^H
LU
CD
OJ
H.O
aanxavo ss
46
-------
TABLE 9. AVERAGE MASS CAPTURE OF RDWS SS
Capture per Filter Surface Capture per Media Volume
RDWS Test Nos. lb/ftz/run*
D-2 to 4
(8 gpm/ft2,
no chemicals)
D-26
(12 gpm/ft2,
no chemicals)
D-5 to 14
(12 gpm/ft2,
0. 5 mg/1 poly)
D-15 to 19'
(12 gpm/ft2,
0. 5 mg/1 poly,
10 mg/1 alum)
D-20 to 21
(16 gpm/ft2,
0. 5 mg/1 poly,
10 mg/1 alum)
2.3
3.0
2.5
4. 1
4.0
0.37
0.37
0.50
0. 57
0. 80
,.2 _
Ib/ft3/run**
0.36
0.46
0.39
0.63
0.61
Ib/ft3/hr
0.06
0.06
0.08
0.09
0, 12
* 1 Ib/ft^ = 4. 88 kg/rn ,
## 1 lb/ft3 = 16.02 kg/m'
47
-------
practice at HRF facilities used in industrial wastewater treatment and ;
for secondary effluent polishing in Ashvale, England (2).
The most important factor in effective cleaning was a complete agitation
of the media (bed fluidization) during the scrubbing and flushing opera-
tions. Problems early in the test were caused by breakage of the
equalizing nozzles in the filter bottom plate. This produced an uneven
distribution of water and air and prevented sections of the media from
being fluidized effectively. Repair of the nozzles (Spring 1976)
corrected this problem.
The supply of air was limited to approximately 5 scfm/ft (1. 5 m3/min/
m ) of filter surface throughout the program; this was the maximum flux
that could be supplied by the Newtown Creek plant.. The available air
supply appeared to be inadequate for optimal scrubbing; it was observed
that the air feed rate was uneven and only portions of the media were agi-
tated at any one time. A higher flux, 8-12 scfm/ft2 (2.4-3. 7 m3/min/m2),
would have been more effective in continuously agitating all sections of the
filter and should have allowed use of less water for scrubbing and flushing.
Wall effects could have hampered effective air scrubbing.
SS mass balances are given in Tables A-3 and A-4. The mass of SS
calculated as removed in the filter column during each run was
determined by flow weighting, assuming constant SS concentrations in
the filter influent and effluent for the time between each sample. The
actual amounts recovered in the backwash were determined by flow
weighting the SS concentrations in the backwash grab samples. When
only composite samples were available, these were used for the mass
balance determination. Generally there was agreement within 25 percent
between calculated filter removal and actual backwash recovery;
this approximates the error expected in sampling and analysis.
Profiles of backwash effluent SS concentration with discharge time are
given in Figure 11. The SS are on a logarithmic scale and show a fairly
continuous decrease after reaching maximums, during the initial 0. 5
minute, of 4,400 to 13, 100 mg/1. The average SS concentrations in the
backwash effluent ranged from 770 to 4, 060 mg/1, with a median of
2, 353 mg/1. Profiles of backwash effluent SS from the Cleveland testing
show similar values of SS maximum concentration, however, a greater
portion of the total SS was discharged during the first 3-4 min indicating an
an apparent greater effectiveness in media scrubbing and flushing.
Other average characteristics of backwash effluent composites for CSO
runs of 3 hours or more (Table A-9) are a 65 percent volatile component
of the SS, BOD 905 mg/1, COD 2, 445 mg/1, and Set. S. 69 ml/1. The
48
-------
'— 10,000
o>
E 5,000
•^-^
to
o
O 1,000
Q 500
UJ
Q
UJ
to 100
ID
4/3 50
in
..
N
/
r —
^
\
l^****!
•*-=
•~v
\
*
— 1
k-
s
1—
^ 10,000
E 5,000
^•""*
CO
0 ,,000
Q 500
UJ
o
UJ
to 100
4/5 50
IO
V^
\
v
x^,
\
\
•'^**
pssJ
^^^
X
^
0123456789 10 II 12
BACKWASH TIME (min.)
FILTER BACKWASH FROM RUN S-12
01 23456789 10 II 12
BACKWASH TIME (min.)
FILTER BACKWASH FROM RUN S-13
50,000
*- 10,000
g 5,000
to
Q
O 1,000
to
Q 500
UJ
Q
g
to 100
" 50
in
-V-
\
\
v>
-\
-V
1 •" »
r->
\
Sv
so.uuu
^ 10,000
E 5,000
CO
0 ,.000
Q 500
UJ
Q
UJ
to 100
Z)
W 50
in
A
"S~
— __^
« ^^
\ "~
^
V
>
vJ
^if
\
\
>
^
\
\
\
±>_
-~
OI23456789IOIII2
BACKWASH TIME (min.)
FILTER BACKWASH FROM RUN S-17
01 234 56789 10 II 12
BACKWASH TIME (min.)
FILTER BACKWASH FROM RUN S-18
FIGURE 11 - FILTER BACKWASH EFFLUENT SUSPENDED SOLIDS vs TIME
49
-------
average volatile component of SS in the filter influent (Table A-8) was
greater (68 percent) than the average volatile component of SS in the
backwash for the same CSO tests (62. 5 percent).
For the available RDWS backwash analyses, the average SS
concentrations were SS 2,215 mg/1, BOD 1, 150 mg/1, COD 2, 560 mg/1
and Set. S. 88 ml/1.
The sludge density index (SDI) and sludge volume index (SVI) were
determined for each backwash composite sample. The average SDI for
all samples was 3. 5 (SVI = 29). For RDWS tests the average SDI was
2. 9 (SVI = 35). The settled sludge from the SDI determinations was
separated from the supernatant and used in dewaterability tests. In
the Buchner filtration test, after 90 seconds of applied vacuum the
average volume passing the funnel was 110 ml. In the filter leaf test
(using 250 ml samples vacuum filtered through filter leaf cloth), the
resulting cake averaged 36 percent dry solids, with all but one filtrate
sample visibly clear. Data from the sludge tests are contained in
Table A-10.
METALS REMOVAL
Testing for metals removal became an important aspect of the HRF
program during 1976-77. Composite samples of influent, effluent and
backwash were analyzed for seven metals: cadmium, chromium, copper,
lead, mercury, nickel and zinc. Some arsenic analyses were also
performed. Discostrainer sludge samples were also analyzed for
metals. Appendix Tables A-ll through A-18 present the metals
results, including percent and mass removals.
In April 1976 several tests were conducted on the parallel 6 in. (15 cm)
columns to attempt metals precipitation at 24 gpm/ft2 (60 m^/hr/m )
with additions of lime and/or polymer. While SS removals of 79 percent
were obtained, neither polymer feed alone nor lime addition to pH 8. 5
combined with polymer, significantly affected metals removal.
For all tests, there is no indication of improved metals removal with
the use of polymer and/or alum aids. No correlations appear between
metals removal and SS removal or influent metal concentrations. For
each metal, low influent concentrations were treated as effectively as
higher concentrations, and the influents varied greatly. For all CSO
tests with filter influent and effluent sampling (S7-18), the average
results are given in Table 10, including mass removals per unit filte^
surface area per filter run (average 5. 1 hr duration).
50
-------
TABLE 10. METALS REMOVAL AVERAGES
CSO/RDWS Treatment Results*
Metal
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Zinc
Filter
Influent
(mg/1)
0.023/0.007
0.33/0.47
0.33/0.30
0.57/0.4
0.001/0.002
0. 19/0. 14
0.42/0.43
Filter
Effluent
-------
ROTARY SCREENING
There was no specific test program to evaluate or compare the effective-
ness o£ the Rotostrainer or Discostrainer. No analyses were made of
the effluent directly from the screens and the difference between plant
influent and filter influent samples cannot be effectively used to measure
screen efficiency for the Rotostrainer especially since, as previously
discussed, the time lag between samples prevented close correlation
between them. However, from the data on Table A-l, the Discostrainer
with a 40 mesh (420 micron) screen appeared to have effected a fairly
consistent SS removal of 10 percent from the plant influent which averaged
190 mg/1 during seven CSO tests. The Discostrainer with a 70 mesh
(210 micron) screen averaged 16 percent removal of SS during two CSO
tests in which the plant influent averaged 171 mg/1. The superiority of
the Discostrainer over the Rotostrainer for CSO treatment was apparent
from the improved HRF performance when the Discostrainer was used.
Samples of the solids (sludge) removed by the Discostrainer were
taken during four runs and were analyzed for dry solids and metals
(Table 11). The dry solids content of the sludge ranged 11.8 to 18. 7
percent, with the VSS proportion ranging 79. 0-87.4 percent.
DISINFECTION
Tests of the high-rate disinfection equipment were made during three
RDWS runs, using a filter effluent sidestream for about one hour during
the final two hours of the run. Table 12 presents the data for all tests.
The disinfection unit was reported to operate (7) at a flow of 20 gpm
(1.26 1/s) but, unfortunately, flows over 10 gpm (0.63 1/s) exceeded
the capacity of the V-notch weir at the end of the rapid mixing chamber.
The tests therefore had to be conducted at 10 gpm resulting in an
approximate 9 minute detention time within the baffled contact tank.
The hypochlorite dosage was fed by gravity from a small tank and mecha-
nically mixed with the flow in the mixing chamber (40 seconds detention
time) before overflowing into the baffled contact tank.
Figure 12 indicates that a dosage of approximately 7 mg/1 hypochlorite (as
chlorine) will provide satisfactory disinfection in meeting the New York
State standard of 200 fecal coliforms per 100 ml effluent.
During several tests of RDWS and CSO, samples of pilot plant influent
and filter effluent were analyzed for total and fecal coliforms using the
multiple tube technique. The results, given on Table A-19, show
inconsistencies in removals. This may be due to faulty sampling
technique or variability of background densities.
52
-------
O rHl . . .
£ \| cv o co r- o
•H. tn. oo in vo co so
to
i
5
<
W
0-
1
w
t— 1:
H
U
t-H
Q
i— I
9
CQ
^J
f-H
iH
OJ i~H
||
&
3—1
SH tn
E S
•Q rH
It
M
Q) rH
O, tn
se
e
3
•H H
O tn
°
g
j3
•iH rH
T) Cn
IS g
U
to
•a
0) -H
rH rH
•rH O
+J CQ Q
IS
1 1 ^.^
O M
EH Q
.
C O
j3 S
«
o ** ^3* co in
rH
00 O O H f~
O H; rH O C)
O O O O °
in o o o ^
r-~ co VD CM -^
oo vo CM in o
in H co H co
-=r o in CM „;
•* CM CM CM (CJ
00 CM CO
in o in ^* \^
o in o o -4
^ t*~ ^* ^* •— t
• • • •
in o _i CM cj
rH rH rH • _
r- oo „, CM oo
CO . ^u "el1 ^
i-l " J. rH ^
O
^3* r^ co c^ I-H
rH rH rH CM
CQ CQ CQ Q
3.
rt
rH
O
H
B
o
o
u
JS
«
2
53
-------
TABLE 12. DISINFECTION RESULTS
Test Date
11/4/76
11/4/76
11/18/76
11/18/76
11/18/76
4/12/77
4/12/77
4/12/77
NOTE:
Influent
FC/100 ml
1.0 x 106
0.9 x 106
1.0 x 106
1.4x 106
1. 1 x 106
1.2 x 106
C12
Dosage
(mg/1)
10
30
5
10
20
7
15
20
Contact
Time
(min)
9
9
9
9
9
9
9
9
C12
Residual
(mg/1)
0.7
1.5
0.3
6/0
12.0
0.2
7.0
12.0
Surviving
FC/100 ml
4
1
300
22
0
1000
2
1
1. All tests used filtered ROWS
2. FC is fecal coliforms
3. Bacterial analyses by membrane filter technique.
4. Dosage given as chlorine equivalents.
5. Chlorine residual determined by a -visual comparator
using orthotolidine.
6. In the rapid mixing chamber G = 1050 sec" and
Gt
-------
o:
O
u_
_i
O
o
o
UJ
U_
1000
500
E
O
9 100
or
UJ
a.
50
IO
•TEST 11/4/76
• TEST 11/18/76
A TEST 4/12/77
8min. CONTACT TIMf
10 2O
CHLORINE DOSAGE (mg/l)
FIGURE 12 - DISINFECTION vs CHLORINE DOSAGE
55
-------
SECTION 7
FULL-SCALE HRF INSTALLATIONS
ADAPTABILITY OF FUNCTION
The HRF system can be designed for existing or proposed sewer systems, both
separate and combined. In urban areas, control: of pollutants from- separated
systems is more expensive and less efficient than treatment of combined! sewer
flows (10). An HRF system designed to handle the storm flow of a separate
sewer system would be automated for immediate response to flow increases;..
The treatment scheme could include one central HRF pliant or several plants
located at strategic points on the storm sewer system. Since storms are
commonly short duration, high intensity events, the BRF plant should be de-
signed to capture and effectively treat a desired peak storm: flow.
Use of existing storage capacity or added storage facilities must be consid-
ered in the design for optimizing the size of the HRF system. The excessive
storm flow, if stored, would be treated by the HRF system after the storm
event or at a time in the day most cost-effective for operation, e.g., during
hours of off-peak electrical power demand. Selection of storm flow storage
capacity in the design of an HRF system is further discussed in reference 6.
For existing combined sewer systems, HRF can be used in a dual-treatment
capacity filtering ROWS between CSO events. During CSO the existing sewage
treatment plant will treat a portion of the flow and the HRF system, utilized
in parallel operation as a supplemental facility will treat the balance of
the flow. Using HRF systems in a dual capacity provides additional cost
benefits. The system would operate throughout the year, automatically
switching filtration modes, and since the HRF system: would be located adja-
cent to the existing sewage treatment plant, it would provide primary treat-
ment supplementing existing primary units. In overloaded sewage treatment
plants, the reduced loadings of primary effluent, resulting/from supplemen-
tary HRF treatment, would improve the efficiency of subsequent secondary
treatment and sludge handling. A discussion of area requirements and cost
benefits of using HRF vs. clarifiers in primary treatment of ROWS is con-
tained in Section 8.
HRF can also be used to upgrade existing sewage treatment plants by polishing
secondary effluent. This application of HRF has been successfully tested in
a pilot plant in Cleveland (11) and a full-scale filter plant for secondary
polishing is now in operation in .England as discussed in the Introduction.
A finer filter medium (No. 2 anthracite) has been found; more effective for
this function.
56
-------
In the design of new sewage treatment plants for combined sewer systems, HRF
can be incorporated as a dual-treatment system or for a triple function pro-
viding primary treatment and secondary effluent polishing in separate HRF
units during dry weather and, during CSO events, all or part of the units as
required would switch to CSO filtration. The HRF system for secondary pol-
ishing would be used only for treatment of that portion of CSO beyond the
capacity of the primary units.
FULL-SCALE DESIGN CONSIDERATIONS
Ira the .Newtpwn Creek plant process, flow quantity and quality are atypical .of
most large treatment facilities due to the relatively" Targe industrial flow
component and the absence of primary treatment. The BOD and COD removals
effected by the HRF system reflect this high background industrial component
and should be cautiously reviewed in the design of a full-scale HRF facility.
The SS removals data, however, are more representative and should be con-
sidered in that light.
From the results it is apparent that a fairly consistent effluent suspended
soil ids quality can be effected by the filtration system and that efficiency
of removals Is directly related to the influent SS concentration. Charac-
terization of the sewage solids should be a major criterion in the design
and success of an HRF system.
The overall results for SS, when averaged for the nine tests using constant
rate filtration were: plant Influent 182 mg/1, filter influent 161 mg/1,
filter effluent 62 mg/1, for SS removals of 61 percent across the filter and
66 percent across the HRF system.
The influent concentrations should be evaluated in addition to removal-rates
to provide an accurate picture of :HRF process performance since higher in-
fluent concentrations allow greater removal efficiencies.
The results of composite samples were not used in the calculation of average
values of SS, BOD, and COD. SS results were calculated by flow-weighted
averaging of the results of individual grab samples. BOD and COD results
were from averages of the grab samples.. Composite sample results, for SS
especially,, were often at variance with the flow-weighted averages. The
variation of CSO characteristics is further indicated by the fact that in-
dividual -grab samples of filter influent often had greater SS concentrations
than plant influent samples.
The data collected across the screening device (the Discostrainer) indicate
that SS removals cif 18 and ,23 percent were attained with the 70 mesh (210
micron) screen (tests D25 and 26) and 11-19 percent removals were attained
•with the 40 mesh (420 micron) screen (D27-29).
,As discussed in the testing results section, characterization of the sewage
should be a major criterion In the design of an HRF system for use in any of
these functions. The influent SS concentrations, industrial component, and
fraction of dissolved solids are major characteristics to be reviewed. The
.57
-------
selection of a screening device for effective pretreatment will also depend
on the specific wastewater characteristics. Therefore, all of these factors
must be considered in optimizing the capture capability of the HRF media in
full-scale design.
BACKWASH SOLIDS HANDLING
When HRF is used exclusively for treatment of urban storm runoff or CSO, the
filter backwash effluent containing the captured solids can be treated at a
secondary treatment plant. If the plant is at its daily peak capacity, the
backwash effluent could be retained in a holding tank and then added to the
sewage treatment plant influent at a controlled rate. A majority of back-
wash solids would settle in the primary basins as part of the primary sludge.
The backwash effluent from a dual-treatment HRF system should be added to
the sludge handling system of a secondary treatment plant, ideally the sludge
thickeners. The sludge is suitable for the further processing applied to
the plant sludge or processed separately on-site (i.e., dewatered). The ro-
tary screening concentrate would be added to the grit handling system of the
secondary plant.
PROCESS SEQUENCE
A conceptual schematic of the HRF system for dual-treatment of CSO and ROWS
is presented on Figure 13. The HRF system is considered to constitute the
primary treatment portion of a new or existing secondary sewage treatment
plant.
The plant influent first passes through bar screens for removal of coarse
debris, and is then conveyed by low lift pumps, if required, to a channel
leading to rotary mesh screens inside a treatment building.
The effluent from the rotary screens would be dosed with alum and polymer as
required for formation of floe. Only polymer would be added to the ROWS
flows. The sewage passes through mixers to effect complete blending and
flows along a channel leading to the gravity filters. The filter effluent
is then conveyed-to the aeration tanks.
Flows beyond the capacity of the aeration tanks overflow to a basin for dis-
infection and discharge to the receiving water. Backwash water could be
filter effluent or plant secondary effluent, and should be periodically
chlorinated to inhibit slime and algal growth in the filter media. Polymer
can also be added to the backwash water to aid in backwash solids settling
and to precondition the filter media for the next cycle. Low pressure blow-
ers would provide backwash air. The filter backwash effluent is conveyed to
the sludge handling facilities of the sewage treatment plant or can be pro-
cessed separately on-site (i.e., dewatered).
58
-------
1 a
! 5
1 1
•>»l
7|
(T
B
u
<0
i:
'-z
iig
3i5
tU
a:
CD
Q
3:
O
O
a:
a.
to
LU
a:
a:
Z3
CD
59
-------
MODE OF OPERATION
The design should incorporate automated operation, with personnel required
only for routine maintenance and periodic delivery of chemicals.
The mode of operation most amenable to automated control is "declining rate
filtration," in which the filter flux decreases without mechanical or instru-
ment control from a preset maximum initial value as the filter head loss
increases. Filter influent controls would maintain a constant hydraulic
head. When the filter head loss reaches a preset value filtration is termi-
nated and a backwash cycle begins. The operation cycles of the filter units
are staggered so that the backwashes occur in sequence. Declining rate fil-
tration is not a new operating technique; it is common practice in potable
water filtration and has been used for many years for industrial applica-
tions.
The filters would operate at different flux rates during dry and wet weather.
During dry weather the average flux would be 8-12 gpm/ft^ (20-30 nrvhr/m*)-
As increased (storm) flow enters the treatment plant each filter will auto-
matically adjust to achieve higher initial flux by sensor regulation of the
filter effluent valve. Adequate instrumentation would insure a smooth tran-
sition from dry-weather to storm-flow operation.
CONCEPTUAL DUAL-TREATMENT DESIGN
For design purposes, the low lift pumping facility and HRF system have been
incorporated into one site. The 25 to 200 mgd systems discussed below cover
the range of most potential dual-treatment applications. The designs are
based on an average wet weather filtration flux of 16 gpm/ft^ (40 md/fer/m*)..
The hydraulic capacities of these facilities should be set at 20 percent
greater than the design rates as determined by process considerations to
accommodate temporary conditions of eTevated flux as during the transition
to storm-flow operation with declining rate filtration.
Figures 14 to 18 present general plans and elevations of HRF installations.
The treatment systems have the same basic arrangements, the major differences
being'the size and numbers of the pumping and treatment units. The facili-
ties consist of a pumping station, housing bar screens and variable speed
low lift pumps, the head end of the treatment plant, housing the rotary mesh
screens, and the main treatment area, containing the HRF units.
In the control building, the first level contains the bar screens, lift
pumps, chemical storage area, alum and polymer feed equipment, air blowers
and the backwash pumps. Chiorination equipment can be located here or be
connected to similar equipment ih the secondary treatment area. The upper
level of the control building includes the rotary mesh screens, electrical
and instrumentation control areas, and space for office, service areas, etc,.
Figure 19 shows a typical cross section of the filtration portion of the
main treatment area, with the filter units arranged symmetrically along the
center line of the filter bay. Sewage flows from the central influent
60
-------
channel into individual receiving chambers or gullets and then to the fil-
ters. The effluent channel connects to a reservoir extending the entire
width of the filter bay, providing sufficient backwash water storage. The
treated effluent then flows by gravity in a channel to the aeration tanks.
One or more automated overflow chambers connect with the channel to receive
excess filtered CSO which is chlorinated and discharged.
61
-------
o
O-
O
Q
CD
CS1
•z.
O
CO
<
0.
UJ
111
o:
LU
CD
62
-------
Ill
<
Q.
O
<
D_
<
O
Q
CD
LTl
O
§
_J
I-
C/J
LU
LU
LH
UJ
cm
13
CD
i—i
U.
63
-------
o
•<
O-
-------
h-
1-^:
O
<
0_
Os-
i—f
-------
u
<
a.
<
o
Q
CD
CD
CD
O
i
UJ1
UJ
t/3
LU
o:
I—
(=1
I
oo
f—I
UJ
CD
66
-------
z
o
u
LU
CO
Cd
LU
I
z
o
to
LU
LU
Qi
(=)
LU
o;
=>
CJJ
67
-------
SECTION 8
COST ESTIMATES
GENERAL
Estimates of capital and operating, costs have been developed for the
designs discussed in Section 7 and are divided into two categories:
the pumping station and the HRF facility. The filtration plant estimates
presented in the summary curves (Figures 20~22)i can be used to
compare alternate technologies for separate or dual-treatment of CSO
during wet weather flow conditions and RDWS during dry weather.
CAPITAL CONSTRUCTION COSTS
Estimated capital costs of dual-treatment filtration facilities are presented
for 25 to 200 mgd (94, 600-757, 000 m.3/day) capacity plants. The total
project capital costs are summarized in Table 13. An analysis of cost
vs. capacity is shown on Figure 20. Detailed data on capital cost
estimates are given in Table 14 based on a design CSO filtration flux
of 16 gpm/ft (40 m^/hr/m.2) and including equipment for alum and
polymer addition.
TABLE 13. SUMMARY OF TOTAL PROJECT COSTS
HRF
Plant Capacity
25
50
1 00
200
Total
Capital Cost
CENR = 2520)
2, 731, 000
4,362,000
8, 111,0,00
13,981,000
2 32
Design flux of 16 gpm/ft (40 m /hr/m )
Project costs include pumping statiqn and treatment plant.
1 mgd = 3, 785 m3/day
68
-------
14.0
12.0
10.0
•&
f-
V)
o
0
Q_
O
8.0
6.0
4.0
2.0
0
V
CAPITAL COST
"\/
y-
UNIT COST
NOTES
I. FILTER DESIGN FLUX= 16 g'pm/ft2
2.COSTS INCLUDE DISCOSTRAINER
- AND FILTRATION PLANT
EQUIPMENT AND STRUCTURE
EXCLUDING PUMP STATION.
50 100 150 200
DESIGN CAPACITY (MGD)
IOO
80
60
O
40
20
0
250
FIGURE 20 - ESTIMATED CAPITAL COST vs
69
DESIGN CAPACITY
-------
TABLE 14. ESTIMATED PROJECT COSTS FOR HRF TREATMENT PLANTS
. PUMPING STATION
Excavation and Backfill
Reinforced Concrete
Building
Bar Screen
Pump
Piping
Heating and Ventilating
Electrical
Plumbing, Lighting, Interior 8t etc.
Sub-total
Construction Contingency
Sub-total Construction Cost.
Engineering !t Administration
Project Sub-total, Conveyance Portion
H. FILTRATION FACILITY
Excavation and Backfill
Reinforced Concrete
Building
Diacostrainer
Static Mixers
Filter Media and Filter Bottom
Filter Backwash Pump
Air Blower
Piping
Polyelectrolyte
Alum Feed
Chlorination Equipment
Heating and Ventilating
Electrical
Instrumentation
Plumbing, Lighting, Interior & etc.
Sub-total
Construction Contingency (12%)
Sub-total Construction
Engineering & Administration (10%)
Project Sub-total - Treatment Portion
HI. TOTAL PROJECT COSTS
IV. PROJECT COST/MGD CAPACITY
$
$
$
$
$
$
$
$
$
$
25 MGD
7,000
66,000
85,000
35,000
140,000
10,000
20,000
70,000
35,000
. 468,000
56,000
524, 000
52,000
576,000
19,000
345, 000
160,000
375,000
20 000
50,000
35 000
35 000
275,000
35,000
35.000
50,000
25,000
90,000
120,000
80,000
1, 749, 000
210.000
I, 959, 000
196.000
2, 155,000
2,731,000
109,240
50 MGD
$ 7,000
66, 000
85,000
35, 000
240, 000
20,000
25,000
100,000
45,000
$ 623,000
75,000
$ 698,000
70, 000
$ 768,000.
$ 24, 000
628,000
210,000
750, 000
40,000
100,000
35,000 .
35,000
540,000
35,000
35.000
50,000
30,000
105,000
205,000
95,000
$ 2,917,000
350,000
$ 3,267,000
327. 000
$ 3,594.000
$ 4,3.62,000
$ 87i 240
100 MGD
$ 9,000
130,000
190,000
70,000
/ 440, 000
30,000
35,000
260,000
50,000
$1,214,000
146,000
$1,360,000
136,000
$1,496,000
$ 35,000
1,225,000
. 430,000
1.500,000
50,000
205,000
60,000
60,000
1,060,000
50,000
45,000
80,000
50,000
155,000
Z25.000
140. 000
$ 3,370,000
644, 000
$ 6,014,000
601,000
$ 6,615,000
$ 8,111,000
$ 81,110
200 MGD
$ 18,000
260,000
385,000
.140,000
820, 000
45, 000
50, 000
565, 000
100.000
$2,383,000
286,000
$2, 669,000
267.000
$2, 936,000
i$ 90,000
1,915,000
805,000
3,000,000
100,000
410,000
65,000
65,000
1,395,060
85,000
50, 000
110,000
70, 000
225,000
410,/DOO
170,000
'$ .8,965.000
1,076,000
$ 10,041,000
, 1. 004, 000
,$ 11,045,000
$ 13,981,000 ,
$ 69,905
NOTE: 16 gpm/ft2 (40 m3/hr/m2) Filtration Flux
Engineering News Record Construction Cost Index = 2520
70
-------
Capital cost estimates for the treatment facilities include: the cost of
equipment, installation and construction costs and a 12 percent allowance
for contingencies, plus a 10 percent allowance for engineering and
administration for the proposed construction. The cost estimates do not
include the cost of land, backwash sludge handling and interest during
construction.
TOTAL ANNUAL COSTS
Table 15 and Figure 21 present summaries of total estimated annual
costs for HRF plants of 25 through 200 mgd capacities. The annual
costs are in categories of separate CSO treatment plants operating only
during the estimated 450 hours of CSO per year and dual-treatment plants
operating 8760 hours per year (365 days including 450 hours of CSO
filtration). The costs include amortization, operation and maintenance.
TABLE 15. SUMMARY OF TOTAL ANNUAL COSTS
HRF
Plant Capacity
(mgd)
25
50
100
200
Annual Cost (450 hr/yr). Annual Cost (8760 hr/yr)
CSO Dual-
Treatment Plant Treatment Plant
$
Z38,050
382,350
693,200
1, 175, 900
$
396,450
600,100
1,053,600
1,794,050
2 •* 2
Design flux of 16 gpm/ ft (40 nvVhr/m )for CSO treatment and
8 gpm/ft2 (20 m3/hr/m2) for RDWS treatment.
1 mgd = 3, 785 m3/day
Table 16 gives breakdowns of these cost data. These costs are based upon
the following assumptions:
a.
b.
Interest at six percent for 25 years.
Maintenance at dual treatment plants being three percent of
mechanical equipment cost, two percent of electrical
and instrumentation cost and one percent of piping cost.
For CSO treatment plants, maintenance costs are set at
67 percent of the costs at the same size dual treatment plants
except for equal piping maintenance.
71
-------
DUAL TREATMENT PLANT
CSO
TREATMENT PLANT
. INCLUDES AMORTIZATION(6%,25yr)
AND OPERATING COSTS.
2. DUAL TREATMENT 8760 HR/YR
3.CSO TREATMENT 450 HR/YR.
0
100 150 200
DESIGN CAPACITY (M.GD)
250
FIGURE 21 - TOTAL ANNUAL COST vs DESIGN CAPACITY
72
-------
NO
PQ
O «-
(j
t
* §
n o
t 0
£ t--
r CO
Q S
l 3 -
<=> "8
in «
C o
1 S
h oo
o
TO «*
U
io
o
^
CO*
1
5 , «*
.n a 1
N 1 °"
IHJ
ol •»
8|
h
^
O
i co
*» S H
§ Sg 8
2 8 >• u
5 SS g
g 1 1
§o
LI W
•s 1 I?
3 ft
•< so O
o go
o o m
o ^ r—
0 so (s?
in -< ^
VI Tl,
JA.
••' ^« ..j
g 00
° § g
15 .«r oo"
«3- «• «.
O O o
0 0 o
o in vo
B i
CM ^O
§o o £ c
o o ° c
O OO f\J CO C
^ \o r- o c
^3. ^ .—
0 0 f,
in tn «
00
CO* *£
co r-
-?
•W- «• **
o o o o
s s I gg
"^ r*- (si o en
~" st T en
«••»«»
O O o ° °
O 0 o ° 0
O CO CO O O
"• f~ CSI CM 0
CM
«• -W- -W- -W- *5- «• s
I I § I § §s§
* \, ^-INO r*. vo-n-io
M 5 ^ '2* ^ «ooT
1-1 N fN r-l
**' «•«•«> ^ ^.
0 0' OO O 000
O O o o O O ^ ^
O •-* Cvji^ ^Jl CO^O
O N \o If) r^. — T'O o'
O C*> — " • —
«• <«•«.<
^' — H ^^ *— '
*~*
* *o- -ee-
0 g OO 0 000
o o mo o ooo
o «> ^s^i \o cocsir^-
o -- s* m- „- ..r^
•w- -w- -ce- -c
«• •» -w-
0 0 •*•
O O i-4
O^- O^ CO
o^ in*
t*- t-
N «-i
»H
«» ' «i. «»•
O O iTi
o o in
« »
CM CO
CM m
m o
•
-
ooo
o o r-
N o^ t*y
M cO
S §
«* ^ «•
r-
o o «o
o o *o
«o o
»»l O
CO •&
*• -w- «-
o o co-
in. in G-
^m- co Tjj
in* tsT
ON OO
- > CO
-CA." ^A- «v
O O- OO O oOO
O Oi o U
o in M t
n o o ° °
0 O rH 0
o co TJICSI r*^ in*ino
1-1
co in
•M- «> -W- -B3- «. «}.
§ g § 8 § § SS
° ^ OO t*- O QS rHCO
S tsT tvT N* 01 in* in**
**>• rHi
•• I
tj 2 «
S o* '43
M P ^
?l f ^1
!« |J |i *
A i! a I
i I
* •
-W-1 -W- ^5-
2* 17 17 rt
*» rH*^^- "*- ^ ^
^ f 86* •? 1
5- s§-S ^ f S
*•* *" 4J *"* W O U
3 1 S 8' 1 ° S
- J llP s> 1 I
r i°" ! 1 i
73
-------
c.
d.
e.
f.
g.
Labor at $25, 000 per man year, including overhead and benefits.
Filter flux averaging 16 gpm/ft2 (40 m3/hr/m2) during CSO
treatment and 8 gpm/ft2 (20 m3/hr/m2) during RDWS
treatment.
Chemical applications of alum (20 mg/1) and polymer (0.5 mg/1)
to the filter influent during CSO treatment and polymer
(0. 5 mg/1) during RDWS treatment.
After filtration chlorination (10 mg/1) of only CSO and periodic
backwashes.
Unit costs of chemicals are:
Polymer-$2.50/lb ($5.50/kg) (cost may vary greatly by
type used)
Chlorine- 15 £/lb (33^/kg)
Alum - 6.5^/lb (14£/kg)
The HRF plants have been sized according to CSO flux capacity. Dual-
treatment plants, operating mostly during dry weather, will be operating
at an average of one-third to one-half the filtration flux of storm flow
conditions. The CSO treatment plants, idle through most of the
year, will have much higher total annual costs per unit of treatment
(dollars per million gallons) than the dual-treatment plants which
operate full-time. From the above assumptions, it can be calculated
that the annual cost per miUion gallons treated would range from, $508
down to $314 for CSO plants of 25 to 200 mgd capacity. For dual-
treatment plants, these unit, costs would be from $83 to $47 over the
same capacity range (See Table 16).
Figure 22 presents a comparison of CSO operating cost-benefits for treat-
ment with doses of chemical flocculant aids (20 mg/1 alum and 0. 5 mg/1
polymer) and,treatment without chemical additions. Assumptions, based
on testing results, are: influent SS 200 mg/1; percent system SS removals
70 percent with alum and polymer addition, 62 percent without chemicals.
Operation costs per ton of SS removed from the storm water decrease dra-
matically with plant size increase from 25 to 50 mgd (94, 600 to 189, 000
m3/day) and show a slower but significant decrease with plant sizes up to
200 mgd (757, 000 m3/day). Treatment without chemicals has an increa-
sing advantage in cost benefits for all plants of 25-200 mgd capacity. This
indicates that the extra costs with use of chemicals outweighs the benefit of
74
-------
250
• 225
o
S 200
UJ
o:
CO
en
u_
O
ce
UJ
Q_
I-
co
o
o
175
150
125
100
NOTES'
I. FLUX I6gpm/ft£
2. 20mg /I ALUM
0.5mg/1 POLYMER
3. 450 hrs /yr. OPERATION
4. OPERATING COSTS ONLY
FILTRATION WITH
ALUM AND POLYMER
CSO FILTRATION
WITHOUT CHEMICAL ADDITIONS
O
50 100 150 200
DESIGN CAPACITY (MGD)
250
FIGURE 22 - OPERATING COST - BENEFITS, CSO TREATMENT
75
-------
extra treatment received unless environmental regulations require effluent
concentrations obtainable only by the use of chemical additives. The
operating costs for 25 mgd plants are inflated by the fact that labor costs
are the same as those for plants of 50 mgd capacity and at larger plants
reflect the economics of scale.
DUAL-TREATMENT SYSTEM COMPARISONS
HRF in dual functions could increase the capacity of an overloaded
secondary sewage treatment plant. During dry weather HRF could be
used to improve the primary effluent, and to treat CSO during storm
flow conditions.
Tables 17 and 18 present comparisons of HRF plants vs primary
clarification in capital and operating costs and area requirements for
25 to 200 mgd (94, 600-757, 000 m3/day) capacities. HRF facilities
include rotary screens and filters while the clarification facilities
include grit chambers and primary clarifiers. The comparisons exclude
cost of land acquisition,pumping stations, bar screens or chemical
additions and assume equivalent (60 percent) SS removals at a
16 gpm/ft2 (40 m3/hr/m2) RDWS flux and a 700 gpd/ft^ (28. 6 m^/day/m )
clarifier overflow rate. Bases for operating costs are as given previously
in the discussion of annual costs except amortization is excluded.
The compactness of HRF is immediately evident from Table 18. Land
requirements for the HRF units are only 5 to 7 percent of that
necessary for primary clarifiers of the same capacity. On Table 17,
it should be noted that the rotary screens are much more costly than
the grit chambers commonly used ahead of primary clarifiers; the
screen estimates were based on the costs of the largest Discostrainer
units now available. Since large scale treatment of raw sewage is a
relatively new application of this type of screen, costs of units designed
for 25 to 200 mgd size facilities should be considerably lower than
those indicated. The HRF units themselves have a significant cost
advantage over primary clarifiers; this cost advantage increases with
treatment capacity.
CSO TREATMENT SYSTEM COMPARISONS
Table 19 presents a comparison of HRF with several alternate CSO treat-
ment systems in terms of SS removal efficiency, capital and operating
costs, area requirements and'operating cost-benefits for a'25 mgd (94,600
ir>3/day) facility. The data for the alternate systems was taken primarily
from two recent comprehensive studies of CSO treatment technology (12)
76
-------
CO
5
5
<;
o
u
EH
CO
O
O
^
w
EH
CO
CO
EH
EH
W
EH
-i
M
§
Q
t^
W
PQ
EH
vv
*~-
c
c
c
l-H
^
•4-*
CQ
O
Q
\*s
b,
.s
"rt
/?
O
.
c
!>s ,c
rt re
C CJ
rH *P*
n! '*•
" n
i-H
U
vO vO Tf< O
CO vO Tj< ID
1—1 •— ' CM CO
«
in
CO i— 1 0s O
so o^ r^~ i"H
r-H fH CM Tf
4J.
o
o
o
-4-J
CQ
o
O
l-H
•4->
S
u
d
o
• rH
"(t
U
s
rt
l-H
U
>.
mar
•|H
ILf
PL,
"o
EH
«- < CM t- vO
CM vO vO so
r- I-H oo co
>— I CO ID O^
CQ
rH
£
• rH
in
rt
r-4
O
r— f
cd
-M >
• rH O
rv.
O CM r- ID
•— < , co I-H oo
vO O vO OO
1-1 CO ID GO"
•
CM O O i-H
O CO ID OO
I-H r-H (\J ^f(
r— 1
«
ffi
t— 4
rt
O
EH
|
4->
rt
• rH
cr^ r- ID o
O ID -^ (\3
^o r- ,-H vo
•*•*•*».
•-I CM ID OO
Tf t^ ID O
CO O •<* CM
CM O vO vO
•-I CM CO ID
bi
.S
'S
d)
w
JL,
U
CO
ID O O o
r- ID o o
co t^ LO o
« «
r-H OO
bo
d
• rH
*^f
0
cu
r-H
rt
CJ
a
0)
CJ
CO*
d
0)
CJ
CQ
rt
CQ
d
o
'-4-1
rt
CQ
bo
.s
Q_t
c
a
T)
d
rt
t — t
£3
0
• rH
CQ
CJ
rt
0
J3 •
i—* c*
H 0
£H -rH
!H "S
§ .2
"£!
O o
T3 c
CQ ,-j
O
S
.eq
PH
O
u
w
K
^
W
EH
CO
CO
H
S
s
EH
<;
K
EH
|
g
Q
00*
l-H
W
PQ
H
CM
g
rH
CM
0
O
1!
CM
i-H
-U
is.
rt
•
(T^
TO
a
•r-4
£
O O O O
o o o o
o o o o
in -o vO co
ID ^2 CT^ " 0s
i-H «— ! CO
r
§
O O O O
CM O ID O
ID ID CM O
CO If) O C>
I-H r-H
• rH
CJ
rt
ft— .
rt rQ
O W
a S
P IH
ho ^—
• rH
CQ
cu
Q
ID O O O
CM ID O O
i-H (\J
Q U £
If)
CM
O
in
o
o
o
o
CM
O
*
77
-------
? S "
CO
en
o
$ &o
us
rt
.S
u
I
fl)
I
I
.• 3
o bo
- .S
to
•s
•s
V
•o
rt B
14 .5
0 •
Cb O
a u
§•
o u
O
o
3
So
u 8
ss
— II
o
0
o
in
i^
CM
O
O
0
0
in
CO
o
0
o
ts
CO
rH
o
0
0
o
in
M
o
0
0
in
N
N
O
O
0
o
0
o
rt .
o
o
0
o
NO
fj
1
)K>
e
•rl
O.
i
"si
•« *.
S5
f-* U
bo c
S
n
S
§
I
u
H
HRF System with
DiscostralnerB
Flocculation=Sedimenta-
tion with Grit Chamber
Microstralners
(Horizontal Shaft Screens
Dissolved Air Flotation
with Drum Screens
Swirl Concentrator with
Swirl Degritter
Vertical Shaft Rotary
Screens (Drum Screens)
High Gradient Magnetic
Separation System
g
a
o
U
E-i
O
g
. U u
J; o B
£ 0 W
78
-------
and costs (13), supplemented by pilot plant studies of individual treatment
systems (7, 8, 14, 15, 16).
The comparison shows a wide range of costs, area requirements and treat-
ment efficiencies. Each system has inherent advantages and disadvantages
for specific applications. The swirl concentrator is easily the least
expensive and most compact system but it is adapted more for removal of
the heavier suspended solids fraction rather than total suspended solids.
High gradient magnetic separation has only recently been applied to CSO
treatment in tests with no more than 1 gpm (0. 063 1/s) process flow (16).
Flocculation-sedimentation is a well proven system for treatment of many
types of wastewater but has great land requirements. Dissolved air
flotation and microstraining have proven effective in other wastewater
treatment applications and pilot plant scale testing has shown them to be
competitive with HRF in CSO treatment.
79
-------
REFERENCES
1. Horowitz, J.R. and Bazel, L. An Analysis of Planning for Advanced
Wastewater Treatment. USEPA Contract No. 68014338.
July 1977. Draft Report.
2. Cox, G. C., et al. Recent Developments of Deep Bed Filters to
Produce High Quality Effluent from Sewage Treatment Works. In:
Proceedings of the Pro Aqua Vita 77 Colloquium on Sewage Treatment;
Basel, Switzerland, 1977.
3. Drehwing, F. J., et al. Combined Sewer Overflow Abatement Program,
Rochester, N.Y. Pilot Plant Evaluations. USEPA Grant No. Y005141.
1977. Draft Report.
4. Murphy, C. B., et al. High Rate Nutrient Removal for Combined
Sewer Overflows. USEPA Research Grant No. S-802400. 1977.
At Press.
5. Hickok, E. A. , et al. Urban Runoff Treatment Methods Volume II -
High-Rate Pressure Filtration. USEPA Research Grant No.
S-802535. 1977. At Press.
6. Nebolsine, R. N., et al. High Rate Filtration of Combined Sewer
Overflows. 11023 EYI04/72, U.S. Environmental Protection Agency,
Cincinnati, Ohio, 1972. 340 pp.
7 Maher, M. B. Micro straining and Disinfection of Combined Sewer
Overflows - Phase IH. EPA-670/2-74-049, U.S. Environmental
Protection Agency, Cincinnati, Ohio, 1974. 82pp.
8 Gupta, M.K., et al. Screening /Flotation Treatment of Combined
Sewer OverflowTVolume I - Bench Scale and Pilot Plant Investigations.
EPA-600/2-77-069a, U.S. Environmental Protection Agency,
Cincinnati, Ohio, 1977. 281 pp.
9. Mueller, J.A., et al. Task 225, Areawide Rainfall-Runoff Model.
City of New York Section 208 Study, New York City Dept. of
Environmental Protection, 1978. 271 pp.
80
-------
10. Nebolsine, R. N. and Vercelli, G.I. Is the Separation of Sewers
Desirable? In: Proceedings of the National Symposium on Urban
Hydrology and Runoff and Sediment Management; University of
Kentucky, Lexington, Kentucky, 1974.
11. Nebolsine, R. N. . et al. Ultra High Rate Filtration of Activated Sludge
Plant Effluent. EPA-R2-73-222, U.S. Environmental Protection
Agency, Cincinnati, Ohio, 1973. 115pp.
12. Lager, J. A., et al. Urban Stormwater Management and Technology:
Update and Users Guide. EPA-600/8-77-014, U.S. Environmental
Protection Agency, Cincinnati, Ohio, 1977. 313pp.
13. Benjes, H. H. Jr. Cost Estimating Manual - Combined Sewer Over-
flow Storage and Treatment. EPA 600/2-76-286, U. S. Environmental
Protection Agency, Cincinnati, Ohio, 1976. 123pp.
14. Sullivan, R.H., et al. The Helical Bend Combined Sewer Overflow
Regulator. EPA-600/2-75-062, U.S. Environmental Protection
Agency, Cincinnati, Ohio, 1975. 125 pp.
15. Cornell, Howland, Hayes and Merryfield, Consulting Engineers and
Planners. Rotary Vibratory Fine Screening of Combined Sewer
Overflows. 11023 FDD 03/70, U.S. Environmental Protection
Agency, Cincinnati, Ohio, 1970. 70pp.
16. Allen, D. M. , et al. Treatment of Combined Sewer Overflows by
High Gradient Magnetic Separation. EPA-600/2-77-015, U. S.
Environmental Protection Agency, Cincinnati, Ohio, 1977. 117pp.
81
-------
S
T~> I/I
t— 01
APPENDICES
TABLE A-1
RATE FILTRATION S'
CSO Testing Averagi
S
0
o
o
Q
0
CO
QJ £
£ QJ *O
U S-
ina*
QJ-*. CM
u- o to
E| CM
I
i. Ot CM
E| *r*
SI D «tf-
El "3-
ta E
4J S r*.
OS- CM
4-»lSl
$_
£ E m
r- OJ CM
f->t
C
CO
U- O CM
•r— "vj CD
LU ra r**
E| r-
Si
OS- CO
4->%Q p«»
»r— |
O Ol
a. E| o
5 «^J
at E| O
CM
OJ *>J
"* S *°
"
JZ
^ _c-l °
r-^j r—
IJ 5
4-> 0 1
w cj to
^»r-» r^
a en i
'
in
CO
i
CM
cn
r-
cn
CM
in
CO
CM
O
CO
to
CM
to
CO
cn
CM
CO
*~
CO
r-.
^
to
0
CO
1—
^
r—
O
O
to
o
CO
m
^r
CO
JZ
.i
.in to
«3- in
r- in
**• in
r*. •—
co in
r- IO
CO CM
r*. •—
CM cn
in in
r-« co
to cn
m m
CO 1
CO
CO CO
co in
O 1
*$• co
r- CM
Sen
, m
CM CM
CM CM
CO CO
to r»-
S K
CM in
CM tO
i — O
CM CO
o cn
CO CO
o o
o o
to * to
o o
r- r-
CO CO
r- CM
.JL .1
CO ' 1
^f
O CD
^- vo
in i
s §
CM P—
S F:
*J- CM
CD CO
o to
in CM
1 I
*3- CO
CM CM
1 1
co in
i^ en
CO CM
CM CO
CM i—
CM P^
r- 1
1^.
o in
r^ 55
CM 1
oo in
r-. CM
CM i—
S S
CM •—
O
O r-^
CD O
to to
in
in r-*
r— r—
r^. co
i i
CO CO
r- CM
1 1
eh
*3-
cn
CM
en
CM
to
to
CM
CM
CO
CO
«3-
co
r^.
CM
00
cn
to
CM
cn
in
in
r-.
CM
to
,—
tn
•"*
0
CD
to
CD
CO
cn
CO
CM
i
s
CO
CM
CO
LO
CM
0
in
CO
CM
CO
CO
1
1
•1
1
cn
CM
CM
CM
CO
to
o
to
CM
CD
to
r—
CO
r—
0
0
to
in
in
o
i
to
o
*y
o
§!
to
0
CM
CO
S
CM
r*.
CM
CM
in
•*
«tf-
CO
r*.
o>
r*.
cn
r—
0
in
CO
^>
CO
cn
•*
en
in
CD
0
to
0
to
CO
*t
CM
O
r*»
in
CO
^~
to
CO
en
o
cn
r-.
CM
CM
CO
CO
r^
CM
in
cn
^J-
r*.
to
r>.
in
CO
in
CO
r-»
in
f—
co
*"
in
CD
in
to
CD
^j.
CM
J>
CO
O
in
.
in
i
S5
S
CO
•*
CO
CO
CO
CO
CM
CO
CM
o
cn
cn
CO
in
r-.
CM
r^.
p—
s
CM
S
CM
C3
0
to
in
to
CO
i
CO
r-.
i
CM
4-*
U
1
•o s-
S
0 C
c -5
in s-
in
2 8
"5
QJ QJ
£ 5
^o "^
J=
£fe -s
S'& 43
4-> in
^ QJ
O C 4->
4-> (O
QJ QJ
>» C71 3
r— (O CT
4-9 S- Q>
u o (n
OJ 4-> JD
S- W 3 •
•r- W 4J
-o -o • c
OJ r- 4-> (O
01X r— Cr—
c •!- 03 ma.
S E - 0. 01
oi— l. ^:
^ r— QJ CJ 4-*
O
4-» IB 4-»
C fO S- O
So S *"g
»— 4-S O 3! •—
**- C O O r—
cr 7 .*= «j o in M-
>i— O • 4-> "T3 4-> 4_
W--E J-W4J UJ
£M-£C rac
4-> O O U 4-> QJ S-
OCcnc *r- QJ O 4->
O 3 W S- JC OJ (n -r- 0)
s- o in 4J c u.
JC OJ 1- QJ S- -r- O
4J C E QJ 4->
S. r— tO f3 3 C
•O4JQ. 'COJT3 OJ
QJ W E W t- *r~ 3 13
4-> o td QJ -a in r—
v>4->tnjC4->tn s- if-
QJO 4J i — 4-> c aj JT
Sj=«4j4-> tOCO r—
•i- o3 -a 4-» cu QJ c
JC S C E 3 4-» 4J 1
0>* OJ 0 Ot=M-C
3-OJSU tf-(O(O -r-
2 -2> *** O r^ OJ "O *O ^
j; W C r— 4-> QJ QJ
4-> QJ QJ O Q 4-> 4J ••>
4-> QJ TJ QJ O (- S- 4->
1 W 4-> tO 'r- jQ .Q OJ
^J" (O QJ "O t tO tO 3
co S -fi en i—
in o in QJ •— i
O • *r- O 3 • • •
C 04-» Cr- 0 0 0 4->
£Z IO H— C C C C
« r— W »f- O4-»C:4J4->4-» 1
ii i iiicu
BB
•M
O
82
-------
8
&
g
o in 01
3
§
^| 1 *
O fS CM O r— r- r—
r— Cfl in O
O O
» r^^H CO CO
O O
1C O
°*-2 "
!/> C t-
« = 3 2
§C?: ?
» O 0)
q:
OJ 0) en
.+•» ** c :•-
o o
-------
CO O CO 00
co i i i i i
NO in co co
i i co t cv] i •<*« co
m in m tn o in
co in tn o co in
t t CM i-f t <-t I
o m m o in
,-H in NO co co
I N I CM PH CO CM
co in o TJ* t- in
CM CM I CO I CO I-H I tM
in m o in in in
*H O in ON CO ON
CO I I I CO I CO CM ^ CO
CO I I I I t t I I *1 t I CO I I t CO I
^H -H t* 00 OS
00 CO O sO O
co o in in co
in tn o
CM CM ^
tn o p-e co PH
3
«
K
CM CO N CM
CM O^ to ^t GO
I ^^ I I I I I
I I '- I I
CO CO CM t~ CO
CO CO .CO . CM CM
2 o o in J2 in
o o in o o o '
S o 2
* "-1 CO ^
o in o in o m o
o i-c in ca <-< o ^
CO I CM t I I I
o o o in
, , , , , a » . aa *s
o In
i i i i i i i i i i i r— i i i NO i
CM CO CM w <0
f ^-r co co oo
oo o N eo o
FH f-i P-l- I— I
r- t- N1 «*
t co in *$ co
< ,-i O (VI 03
^H O CO fH NO
« -s
55 L.
Jj
U
t p-1 I I I t I
Tj« O ON CM Tp
CM r- t^ eo co
ooooooomo
oooooooooooooo.
ininintnintntnininmininminininin o
OOOOOOOOOOOOOOOOOOOOOOO--H
co ON o ON NO
T|< co co NO u>
p-l i-c CM CO '
O O CO i-c i-H
in in o
o o o m I-H
S
ffi o
moor~oooooooooooooooooooo
QaOQQQQOQOOOQQQQQOQQQQQO
o o o o
co eo so in
(M N
-------
"OCX)
QJ ••— t—
oj o in
r- •— !•>.
(OCOOCMCMCDincMCOrC
f-t— r-COt— CMf— CM
to in
CM
in r-* jx. co co CM co
co in to *r *a- co rC
r-» co *s-
en o CM
5|£;S
•o H-UJ
c=cocooii-«-r-;cntopooim
— r- CO
8 § § § I I § g § i § I § § e
*a-cM*j-«a-ior-*cMcoinr--cocM
•-" CM CM ^- CM r—
O
S
E f^
CD O O CD in
r— CO i-^ r-^ r-^
i^» o in o
r— OO in »O
o in o o
** to co co
CD CD
tO CD
E "O (/) -
s- m 10 ^i-
(D CU O-^-
in co co to in co r—
I O CO CO to O ^J- CM
ooooooooo
CJ
S
I
J
in in in in in o o
^ o o o co oo co
8
° m in o in « o
oo ooo^-ooo d •-: CM o • t-: ,.: . ,J
totototoiotojo,^,^^^^^^^
in to ^* co
wootA^uocAcA^^jjVVTTTV
CM£ co
-*3*5 m 10
"^ E O) « O
co JK; o t—
ECO^ £
-------
86
-------
c
c
N.
(i
8
H
8
1
0
Ctf Q
a o
ffi O
o
H
ft
0
HI
B
'vc>vor:--co^H
1 ^P ^* oo-'sco o^ so oo**oooin^vomoo
a MNDt^-rv3tnr»- oor»-*c>oo*Dr«-f>j— *
jfM-,-,^^ , ^ - " S 2
1
i
1 ^^ ^ w
1 tMCMNCMCMrt icMill^cMl^CMZlcMeVI
j
HJ'^aS'no inNOinoococMNOco
^co^SSS . S^^ctlJS^S^S
ON-^CO^HONCM ONininCN-ooinincM-M
ON-^coinTp^H mr*-c*-ininONooNm
Tt"COT((N3 ^*N>0>OfMsONONOO
rO co tj* oo co co oo 0* co i— i f^ o^ »— i ^ w»
Ca^MCO(N3»H I^H ^H«M(NI^-(lH<5fV3
^* pq
>o oo oo m
>C (M O O^
—1 —4 I (\J — |
\O "^ ^ CO
*** O3 0s* O
p-4 r— { | p_| (VI
^o ^ co
^H ^ O
r-( ^ | | (M
CO %O ^i* O
r- NO r- ^
-H -< I CM CM
"•^ t^ NO O
CM CM 1 CO CO
NO t- ,H
0 O CM
CO CO 1 1 CO
NO in Tf o
Is- O CO ON
-H 1 -4
ON in o o
o rji in ^4*
-< ^ i ^ ^
•* m
-< in
NO •* CM ^
in m i IN- oo
Tt" CM CM ^1<
^J* co in o
NO 00 Tjt
'^ *-* t i P-I
"J NO t- 00 ON
'Y
-------
TABLE A-6. BOD, FBOD, UBOD COMPOSITES COMPARISON
Filter Influent
Filter Effluent
Removals
JJ 1J.1/OJ. J-AJJ-J-U.^J-J-*' •"- **** *"*•" — —..-.— —• ' "••—
Run BOD FBOD UBOD BOD FBOD UBOD BOD FBOD UBOD
No. (xng/1) (mg/1) (mg/1) (mg/1) (mg/1) (mg/1) (%) _I%) (%L_
S-4B 113
S-5 220
S-6 235
S-7
S-8
180
111
S-10 173
S-ll
S-12
83
68
S-13 123
S-14*
S-15 117
S-16 79
S-17 141
S-18
85
D-28 150
D-29 140
38
83
102
76
39
117
23
18
61
45
25
48
34
85
68
255
600
515
395
145
143
565
275
267
340
80
130
133
34
78
95
160 103
113
39
138 124
52
39
17
12
96 60
83
44
45 38
93
50
134
90
47
30
88
84
235
345
280
330
88
90
370
190
194
150
29
41
43
11
0
20
37
43
22
29
43
34
41
11
36
11
6
7
0
26
33
0
12
8
42
46
16
39
37
35
31
27
56
* Composite samples were contaminated for S-14.
88
-------
TABLE A-7. FBOD GRAB SAMPLE 3&3KERAGES
Plant Influent Filter Influent Filter Effluent
Run No.
S-4A
S-4B
S-5
S-6
S-7
S-8
S-9
S-10
S-ll
S-12
S-13
S-14
S-15
S-16
(mg/l)
86
51
100
107
111
47
53
108
45
30
88
67
63
31
(mg/1)
81
51
99
105
120
45
44
103
39
18
90
59
72
33
_ __V V.M .» _ f J,J. J, VfcVrfJl
(mg/1)
86
50
94
92
129
45 -
49
125
41
12
87
73
63
24
89
-------
W
a,
O
m NO P- 10 i
Tj< xO CO
W
£
oco^o •
W
ra oo co oo ra oo o
r- \£> «o rp co in *-t
_l —( N
WWWWOTWWWWWWWWWWtQ
90
-------
CO
H
H
EH
O
a
u
w
H
HH
8
i
U
to
O
«
pq
CO
o -
U Ml
u
o
p
1*1
W M
CO g
1 1 *
omot^-oooovo
OOsOvOlDsOOOOvOsO
o in o
oo oo o
°°
om
COt-
oooooo
xooooo
IT) (M 00
iiiiiiiiiii,,7,,
COCOCOCOCOCOCOCOCOIOIOCOCOCQCOCO
^ T}< m N
CO If) vO O
(M Is- .-< O
-< (VJ PO ^
1 1 1
o co co
m •* oo
mono
-4* -tf ^ o
•^t1 ro o oo
i i
c-
CM
O
CO
o
O O O
^ fvj t~-
•<# oo in
(NJ ~-i CO
DJD
0
o
t~H tjj M
S ^ -o
W fli _.
.zl /i) *^
h
JJ 4) TO
5 CO
cd C
co a
h (H
0) V
,C
a
O CO +5
-------
to
H
CO
O
to
W
H rH
_Q 6
C1 M
r-l rO
-P-H
•P r-l
0) O
to to
VO
•
00
CM
oo
in
CM
(U
00
oc
n
^< vo t*- CM n
CM r- m in n
in ^r CM CM en
CM
in
CO
O
vo
CM
CO CO
9
W
w
0)
4J
M-l
(0
0)
O
QJ
^4
ro
o
4J
Q)
oo
in
vo
o o r-
iH iH CO CM
vo
CO
00
CM
CO
vo
00
in
vo
00
CM
•
CM
CO 00
CM CM in O
oo
CTi
00
CD
co
CO
00
H
p
tn
CO
CM
VD
•
CO
00
CO
CO
CM
CO
in
•
CO
in
m
'•
CM
tn
Q) i
CM
00
CM
a\
CM
vo
to
oo
tn to
co
to
to to
to
vo
to to
92
-------
ESULTS
tf
i
H
a
EH'
H
0
n3
to
to
in
l
l
VO
^J*
1
rH
rH
1
CQ
CM
^J1
rH
in
CO
rH
CM
rH
f».
rH
rH
1
0
rH
in
VO
CM
rH
1
CO
o
rH
O
O
H
O
cn
l
in
r-
l
l
o
m in
rH in
o
CM
rH I
^r in
rH rH
1 1
CO CO
0
CM
rH
in
rH
rH
00
O
rH
CM
1
0
O
VO
rH
1
CO
93
-------
u
n
rt
O
u
u
c
rt
&
to
n
id
s
'0
O
•s
0
E
V
•o
U
g
3
rt
1.
e
c
u
c
o
U
•O
o
3 fr
a o
s il
n PS
u _,
u P
o 8 B
3 p*
-
(N &
«« "i
(xi b
J
W
•Sg
§ "~
a
si
u
65
|l
n
*H
£ "b
*—
^
"I
*•
SJ
»,!
•S E
S *
•HNOin mcoocMNO
dc-IM OOOCMrt
J loooi i i o o o o *«
*** ONI"- COONCO
^«MO*CO!l O OiH
! loeSo'cJoipJiescM
oincONOONCo ej; 25
CHdCMOcHCO •* OO
IIOOOO'OO IO IO»^
CO 0 ' 12 ! "9 2
tntncoNocMi ' ' iCNi-^m
: i ! : : : g a : » : ? -5
:: ; :::::: ° :::
coSrtpaiHi-i in ^2J
M 1 1-1 >•! -. 1 -1
t- O
t 1 1 I 1 I^«CO IOOONO
tllllf^iilICO CO
t- •* CO O 1-0
ininincococot»-co l^°°^2
co co co 1-1 ' f- N
jj |
•STSf? O-fMCO
if^PUwwwwwwwwwww
in co to o^
•-< r- ra to
o o o m
OO 00 f» 00
N «* O ^
o o o «n
co en •-< oo
fsj %£> N M t
o o o *o
in »-* CJ o
o •-• o o
t-i
in t*- \o in
o -^ •* t-
m ^ •-< eo
i N ^
^* in oo to
N ~«
i-*
t*- CO NO
IT) ^* • CO N
If)
m in i- o
^H CO CO N
iH 00
vO O O O
r- c^ in in
0*
^« w so r-
to to co in
^H
in
o
\o
^t
oo
o
m
ro
00
CO
\o
C4
CO
00
CO
0
o
to
CO
CO
1
o
sO
CO
o
CO
o
CO
%
sO
M
o
in
CO
CO
o
*
o>
CM
Q
(X 4?
.i g
a. a
94
-------
H
H
S
05
B
fa
!*
ffl
J
<
!>
i
w
K
S
5
5
o
g
o
•
CM
t-i
1
<
3
n
•n
TO
10
£
rt
C
<:
u
c
*rt
IS
(0
rf
1
h
U
1-4
o
o
K
0
|
rt
t-t
•4*
a
o
c
o
u
u
Sir
iii
W Si S
S P* "*
111
g « ""
t) 0
PN y
„
sl
a
•si
^
s S
r "~"
g-
W ^S
u
CO
j"?
. rt -^
n
« 8
£ M
-
-1
N £
!-)
S|
cocoo^ msor-^oor-in-^r-^4
i icococo i i irNiocONOtncomrMcM'^
•"•• CM r-4 >-H CM CM
* * * *
, )NOcofMoovco ovo r-oocoinco
-< -^ fM CM
ij1 CO 'tf .CO NOO^ OT|* CO (M M* "-^ ^*
i tino-t-H-^coin i i in vo i r- CM so co in
•~* •"• CM T(< in
tn*M oo oocMNor-
1 ' ' i i 1^*001 looimo^co^vo
1 '.' ' • ifi i _i 1-1 t-i ificoSioh-
mooococM^t*cMfM i 10 i it^-inosoo
cr-ND^^MincMco.i iini iinmt^-^in
J I! ! ' it^oi locoirMco^sOco
1 * ' ' i t cot imcoiin^NOco^
I'llJ'^-fO'iitirt^-iininco
'till 1 fl IllllftCUfl.^^!
•-t
oo^oooNomooo m in
• ••ininNOinin-^fCo^cMcMcooot^o
ONfoooxOsOi-i inmo mr-omo
CNioNcocMt-cMincxa coOfiinfifificocM
Ot-tOOOOOO IOOOOOOOOO
oo m o m fiocoinm
t ( ( ( t NOco (M*-4fiincococMinco
1 ' ' ' ' 'OO IOOOOOOOOO
^«*£inovDc>3'"11 ooo incocoino
ooincoomt^co cM*-Hi-tincococoNO^
NONOoo"o"ooc3 1000000*0*00
i
t-
t-
°:
o
N
O
00
N
rH
m
00
N
•s
* I
o
h
S
MH
CO 1
1
^. 5
m "3
° h
S
d
fw g
° 1 fi
S|
o -o -g
0 ^"p-
s >
.§' 1
•a -s
£ £
s s
NO
W
0 ^H
»-H i-H
w w
IM
*-l
w
CO
I-H
W
-i -; sj
in m Ct
95
-------
H
S
K
K
In
§
J*
«
»J
<
>
5
s
U
o:
§
1
CO
f-«
I
<,
3
n
S
K
c-i
C
^
V
U
c
rt
u
rt
S
3
U
"rt
|
K
pj
U
e
_o
14
'S
V
U
c
o
U
»
U
•Si?
a o
& ° "i
^ U O
5«*~
•O _,
** P
I ° "C
K P C
'5
«?
S I
^^
d
SI
I*
tfr
^
o *—
p —
W
fj
ss
to
«1
o <
1
•H "**
C
"
|
^£
0
II
en ^p o * i •<*< i i i i i i
* 1 t 1 l 1 0 O
en en CM in Tf en
o o o o o o
^t* in o t»- co ^
co co co co w Q
)
^^•MMM
e
1
5
a
&
4>>
d
s
1
s
4) fe
1 •**
1 1
3 ^
^ §
1 *
•s •
U £
*
^H
-------
w
n
"rt
C
u
e
a
"a
K
m
K)
JJ
2
2
^*
M
«
fa «
g 0
«
« «
1 £
o
2
w
«
Q
S
.
•7
w
3 §
S 1
£
V
• u
e
o
O
_Q
&
srr
rt t>
Efe * *~
n &
Hi
111
Rl Qj
o
Be
K
w „
s 1
1 —
** IS
l5a*
C >-•
+* ^
§ ""
8-=
U
to
(0
• '« 1
tf J
« 5
h n
f-H
w C
**—
•3
., g
-
i
v c
a ;
2«
O CM t*" CO PJ NO 00 ^ ^ "(f ^ CO O
i iNor-NO i i ico«-HNOp4.-MNOcONONOo i
"t ' t-Ht-i-H^pdrONO
* * * *
^*ooo^o^ in coNOt-it^cONOt^r-^-i
NO r-1 in «-* CO CO
omo^in NO coinNONOcorHf-fi-4^
1 1 PI] *H ^< NO I in 1 1 O^ >O tj* O^ CM CO O"* 0s 00
tj< i-t in •-* i# *# 1-4
fH NO^-tr-coNor-cocoin
i i i i i i ir-i i lococoT^Tjipjmi— (-^
I 1 1 1 1 1 |,-( i If-HiMt-icOOCOCOCOrM
rH r-4
•s
»H. NO^Tt*oor-o i tooro'cooorHooco
vOvOvocoin^i— cin i immm-^mr— mmvo
i t i i i i t o i looococooocom
i i i i i i t m i itnm'NOtpcoNo^^pij
II 1 1 1 1 1 1 1 1 1 I 1 1 (O ON CO t— O
iiiitiiiiitiiicocMi-Hi-iin
S^NOo^coNoS^NoSNOinxowOinNooo
inpooO'"-t^-i^Hi— i o o ^H pj »-i m ^* co t*~ o co
f-l fH «-H f-l
O-^fMcOOOOO OOOPJ
OOOOOO»-iO,l OOOOOOOOOO
O O O O O NO
O-* pacMfMor-coint^in^
III III ••! •••••*..•*
I I 1 1 1 li-iOIOOO«-tOOOOO O-
COCOCOCT^OOOO OOOCO
CMCMCOr-tsOOCMTt" CMCMfMCOr-fMt^CONOCO
i-Hi-tOOOi-tf-^
&* ft TO TO (0 CQ tO tO tO TO TO IO tO TO TO TO TO TO Q
•w
fl
F-4
*s
n
V
a
V
&
'o
1
S
3
i
f-Ej
1 4J
Ji
* -g
i2 «
?a
97
-------
s
K
o
u
U
o;
•Sfr
IS?
gel
ll
1 Jl
g*~
si
si
•sg
CO
h
o —
d
gg
u
TO
a
is
li
m
q
•2
*** "u
Sec
3
J
" 1
gl
>o co !•• -0 IH ^* m o^t
1 1 CO O O 1 1 IOOOOO
1 1 O 0 O 1 1 IOOOOO
ooo ooooo
* #
O CO CO 00 t^ ^
1 Ir-tl 1 N •-< O 1 1 lot-
1 IOI 10001 1 100
o ooo o o
CO CO ON rH CO CO
1 IOI t CO CO rH 1 I | ,_| T*
1 IOI IOOOI 1100
O OOO 0 O
CO ON O t-
III 1 1 1 *4* <-H 1 1 1 CO ^-1
IlllllOOIIIOr-l
O O O O
^?
iNtnoiotncoi i icoo
icor-tn imT}
-------
G
'e
*>
(i
c
4)
U
a
fl
GQ
0
fl
g
H
1
H
I
H
h
g.
« ""
J !
2 §
0 «
2 *7
W g
P3
w
s
1
<;
w
-J
(X)
H
A
c
o
U
fl
o
U
g
|£
1 si
to «
•o „,
•V ^
"3 P 1
f 1"
U
« 1
EH .3
••^i
c
*H C
fn 5
g
|.i
n
0
o ^«
0 ~"
ra
J3
|s
rt ^
fP *
rj ^
~
~.
%e^
J.
^.
•-•
• i "i * ^
1 1 O O M
*
00
11.11
• 101.
00
1 1 .1 1
1 1 CO 1 1
1 1 1 1 1
Tjt ON 00 1 1
CO N^ ,_( | |
1 1 1 1 1
1 1 1 1 1
'
1 1 t 1 1
1 t 1 1 1
co co in ^ o
•-* —« O 0 O .
co i-t ro rj t-
o o o o o
I 1 1 1 1
1 1 1 t 1
0 0 00 « 0
co oo rsj ra in
• 1* • • - •
o o o o o
.
1
1
, j
)
1
,
iNOON^«C>^,0000rt<^
1 OOOrtcOOrtrtrt
moNCMinoNO *^
' ' CO *)" rt 00 <0 CM 1 NO
in NO pH NO 0s ON M
' 1 CO rt in 0 CM 0 1 CM
i-l CM rt CM rt
O in co >H ON in Tf
' i r** NO %o ON m co i M
' 1 CM CM « CM ! 2
co J200 'cMooini-i
CO 1 Tf 1 t^ * N CM f-
J
n
3
4
. ' | O ON ON Tjl o rt 1 CO
i ' i in PI 1-1 »* in PH i co
I-
I
}
2 ' ' ' 10 i o oo t-
•* ' « ' i in i p-< co in
in
o o in o 0s NO N^< oo ^* o
•'^'^•"'•"'p'HincMineoTti
OOOO'OOOOOO
o mcMr-ONONOcMo
•-< i-HCMOrtOrt.wS
o 100000000
IN. NO
r- o|N-ONOOcMoooin
O COCMOlHpHpHpHpH
o 'oooooooo
«n tn«44r^sOoosO^
-• |CMCMOCO-HCM-4CO
O 1 OOOOOOOO
in
ft 1
CM in .
00 CM
t- P-<
CM NO
CO
ON NO
• •
O OO
O 00
ca m
o o^
CJ N
4J
!. 2 I
S
0
a
S
in NO *"
CM T)i 1
o" o' «
1*1
*.
d
-------
B
1
H
«
H
e
S
i
s
w
K
O
ft
ts
t*-
rH
I
w
J
m
<
H
B
n
>
CS
B
<
O
o
c
rt
"3
«
to
B
1
d
N
em oval
K
c
N
e
o
3*
rt
£
'c
«
o
Q
O
U
c
N
•Sir
Us
o £•
m fS """
i§ O -~
o § J
O
•p
r B
-
-
Si
Ig
to
U
s ~
a
u *-~
2£
o
to
•C
5 ^»
i C
si
ft
v "5
fe c
^m"
^
^ "S
•—
bi
K J
0
14 "
II
t| •••III
i i T}< tn oo i t i
vo in o NO o o
i i ^ co co t^- CO(NJ
o>Tt»'Ni««-*co incoco
I t 1 1 i i O CO
t 1 1 1 1 1 CM CO
iii i i i r- i
iii i i i f-i i
COeOCOTj«ONONOO
NOO-vff-CjNOCOON
CO ^ O O O i— * O O
\o in o* o^ C3 CJ (\i o
omtNjcococococo
oooooooo
O CO
1 1 f 1 t 1 ...
1 1 1 1 1 1 O O
oor-vOococoM*
oom^int»-^»Tj«
i-li-*OOOOOO
1 1
•s -s <; n
v V f i
^ ^ ^j< ij< m NO t^' oo
PtPUtOtOCOlOlOCO
^ -^ in M NO co in
^ O t- •-* in ON Tf
f-l CO (VI
O NO Tf ON CO
1 1 CO NO CO t*- CO
(V] f~t
oo ^ r- rt co
- ^C - CO
OO t^ ^H O NO
t i i-i t— o^- r^ co
t-l rH NO oj m
i i in ^ o o t*-
1 1 CO i-H N M "?
1 1 in Is- ^ ON Tj1
1 1 CN! CO CO CM CO
1 CO 1 1 1 1 O
1 1 1 1 1 CM
o r*- o
r- o oo -H t*-
p-H rH
CM t1- in o
•* I- VO vO
00 N , •-< t-
O 0 0 O
in o •-» co
0000
vO O t*- vO
co co in ^
o o o o
2 ^ 2 S
co co co Q
t
•s
o
1
V
IH
o
•5
!H
1
4)
b
t?
c
1
**"*
^
Ji
r-4
**H
, 1
•d
« "
1 •£
ft §
ii
rt ™
•3 P.
^i 1
rt .M
U ft
#
100
-------
H
W
H
K
ffi
ffl
O
w
o
!»
L
-i
h«
o
U]
o
o
O rH
O N
O O
0)
m ^-*
rt ^
rt *"*
m
«<
h |
K|
EI
PO in in co
•-* o o o >— t \o
N ON
N 1 -* t- rH 2
^O O ^
.1 . . » .
^* t Tj< sO t^~ i— 1
m f-< Tf
• ' «...
CO 1 CO ^O ^H ^<
1
.s
101
-------
TABLE A-19. HRF REMOVAL, OF COLIFORM BACTERIA
Total/Fecal Coliforrns
Filter Flux
Run (gpm/ft
No.
D-5
D-8
D-ll
D-14
D-17
D-20
D-23
S-1A
S-1B
S-2B
S-2C
S-3A
12
12
12
12
12
16
16
16
16
16
16
16
, Poly
') Feed
(mg/1)
0.5
0.5
0.5
0.5
0.5
0.5
0
0
0
0
0
8
(liquid)
Alum Plant Filter
Feed Influent Effluent
(mg/1) (106/100 ml)(106/100 ml)
0
0
0
0
10
18
19
0
0
0
0
0
.0,37
11.07
3.3/
l.l/
160. /
35. /
54. /
240. /4. 5
22. 11. 1
240 /4.
- /6.2
= _
2.8
6.3
4.9
4.9
240.
240.
35.
35 /14.
1.7/1.8
240. /I. 3
- /0.5
- /3.8
Total %
R emo val
-
43
-
-
-
-
35
85/-
137-
-161
-/92
_
NOTES: 1. Composite samples taken for coliform analysis only;
chemical samples were taken separately.
2. Coliforrns analyzed by multiple tube technique
(confirmed results), fecals incubated at 44. 5°C.
3. Fecal coliforms not tested in RDWS runs (D5-23).
102
-------
TABLE A-20. TEST STORM CHARACTERISTICS*
Storm No. **
S-5
S-6
S-7
S«8
S-9
S-10
S-ll
S-12
S-13
S-14
S-15
S-16
S-17
S-18
3
1
14
7
13
21
6
9
13
16
13
19
4
25
Duration (hr.) Accutnulation(in.) Intensity (in/hr)
0.07
0.17
0.48
0.52
0.10
1.54
0.19
0.57
1.15
3.15
0.74
1.92
0.24
1.68
0. 02
0.17
0,03
0.07
0.008
0.07
0.03
O.Ob
0.09
0.20
0.06
0.10
0.06
0.07
* Data obtained from the National Weather Service, La Guardia Station,
New York City
** Dates of each storm tested are indicated on Table A-l.
1 in. = 2. 54 cm
103
-------
I
o
•a
«
s
hectare
meter(s)
to
"JT
U
3
cubic meter (s)
cubic meters per da;
square meter
3
0
it
o
Q.
01
it
I-H
o o co co •*
TJ* co r^ r^ o
o o co co C3
m
o
CO
P-
O
4)
n
y
*
r1-* —. -i-* M
H (X IM .H
S
•s
w
a>
o<
(It
13
5 « £
D. jj 3
II 111
S g g «
il
s2 g
,5 m
a-s-
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing}
1. REPORT NO.
EPA-600/2-79-015
2.
3. RECIPIENT'S ACCESSI
4. TITLE AND SUBTITLE
DUAL PROCESS HIGH-RATE FILTRATION OF RAW
SANITARY SEWAGE AND COMBINED SEWER OVERFLOWS
5. REPORT DATE .
March 1979 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Hank Innerfeld, Angelika Forndran,
Dominick D. Ruggiero, Thomas J. Hartman
8. PERFOR
9. PERFORMING ORGANIZATION NAME AND ADDRESS
New York City Department of Environmental
Protection: Water Resources-R&D
40 Worth St.
New York, N.Y. 10013
10. PROGRAM ELEMEN1
1BC822, SOS 1, Task 34
S-803271
12. SPONSORING AGENCY NAME AND ADDRESS Cin. , OH
Municipal Environmental Research Laboratory—
Office of Research & Development
United States Environmental Protection Agency
LfiJ
13 TYPE OF REPORT AND PERIOD COVERED
Final, 6/1/75-6/30/77
14. SPONSORING AGENCY CODE
EPA/600/14
IB suPPLEMENTAfiY NQjES Supplement to "High Rate Filtration of Combined Sewer Overflows,
Report"No. "11023EYI04, NTIS-PB 211 144. Project Officers: Richard Field, Chief, SCSS,
MERL, USEPA, Edison, N.J. 08817 (201) 321-6674 and Richard P. Traver, SCSS, MERL,
USEPA, Edison, N.J. 08817 (201) 321-6677_ .__,... . _—
oan studies were conducted at New York's Newtown Creek Water Pollution Control Plant from 1975-1977 to
to investigate the suspended solids (SS) removal capabilities of the deep bed, high rate gravity filtration
process on raw sewage and combined sewer overflows.
The treatment system was composed of a rotating screen equipped with a 40 mesh (420 micron) screen followed by
a dual media, high rate filter containing 48 in, .(122 cm) or 60 in. (152 cm) of No. 3 anthracite (effective
size 3.85 mm) over 30 in. (76 cm) of NO. 612 sand (effective size 2mm).
A continuous series of tests on dry weather (raw sewage) flows demonstrated SS removals across the filter
averaging 67 percent at flux ranging 8-12 gpm/ft2 (20-30 m3/hr/m2) with an average effluent of 44 mg/1. BOD
and COD removals were 39 percent and 34 percent, respectively.
Tests on combined sewer overflow and average removal of 61 percent SS across the filter and 66 percent across
the system at a flux of 16 gpm/ft2 (40 m3/hr/m2) and and average effluent of 62 mg/1 SS. BPD and COD removals
across the filter were 32 percent and 42 percent, respectively. The addition of cationic polymer (1.3,2 rag/1)
combination with alum (17,35 mg/1) improved filter removals to an average 72 percent for SS, 40 percent for BOD
and 50 percent for COD for two tests.
Capital costs (ENR-2520) for a high rate filtration plant are estimated at $55,225 per mgd for a 200 mgd plant
(757,000 m3/day). Total annual treatment costs, including amortization, operation and maintenance charges,
range froin approximately $396,450 to $1,794,050 for dual treatment facilities in a 25 to 200 mgd (94,600 to
757,000 m3/day) capacity range and $238,050 to $1,175,900 for the same capacity range of facilities treating
only CSO.
Comparison with alternative treatment systems show that HRF is cost competitive with conventional sedimentation
facilities for dual-process or CSO treatment yet HRE! has on 5-7 percent the area requirements, "-«- »*-•!'•*•
CSO treatment, HRF is competitive with dissolved air flotation and mlcrostraining processes.
For strict
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
COSATI Field/Group
*Combined sewers
*Overflows
Filtration—sewage
Runoff
Cost estimates
treatment
^Combined sewer over-
flow
*Deep bed
Dual media
High-rate filtration
Rotary screens
Urban runoff
Sewage treatment
plants
13B
13. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (ThisReport)
unclassified
117
20. SECURITY CLASS (Thispage)
unc1assified
22. PRICE
EPA Form 2220-1 (9-73)
105
ft U.S. GOVERNMENT PRINTING OFFICE: 1979 -657-060/1645
-------
-------
-------
-------
-------
s-
rn
o
o
ro
us
i
o
i—"
en
.
rrj ~;
'•I
tb ?",
^ §
O CD
li.
CD •?
ct
CL co
-, 01
CD 01
I?'
-* :a'
o o
?§
CD CD
o) a
§•;
ca a*
Q> o^
§:^
» 3-
CO O)
?> 5
§
3
CD
8.
Q
§
8-
j.,
Z
rn
D
C
r
O
PPORT
C
z
H
m
TJ
r
O
m
TJ
m
Z
Tl
0
•%
-o
<
J>
H
rn
C
0)
m
V)
CO
o
o
O
Tl
-n
O
r
CD
C
CO
z
rn
0)
0)
O
^
5'
03
i— t-
O
u°
o'
en
ro
O>
00
m
§
CD
(— *•
CD
33
CD
C/)
CD
ID
0
_}
3
— *l
i-^
jrmation
o
CD
<-*•
g
— +*
o'
CD
o
33
esearcf
_/
0)
Q.
Develop
0
^
c
C/5
m
z
55
0
Z
<£
ENTAL
T)
3)
O
m
o
H
O
z
^>
O
m
0)
T)
CD
O
Tl
CD 0
o c
o 3.
?r 3-
i
O
en
en
33
s.
oT
m
03
o
G
cn
m
z
5 T
o o
z
2
m
z
cn
H
Q
m
> >
r z
-D O
O m
H m
m to
H 5
i °
0
m
Z
O
------- |