c/EPA
Unitod States
Environmental
Acency
Municipal Environments! riosearch 1'PA GOO 'j '/;)() I 3
Lotwratbry Juiu.- I')/'.)
Cincinnati OH 45260
Resoofdi ond Development
Oxygen
Newtown Creek
-------
U.S. DEPARTMENT OF COMMERCE
National Technical Information Service
PB-297 657
Oxygen at
Newtown
New York City Environmental Protection Administration
Prepared for
Municipal Environmental Research Lab, Cincinnati, OH
Jun 79
-------
RESEARCH REPORTING SERIES
Research reports ol the Office ol Research ;ind Development U S I nviionionnt/ii
Protection Agency have boon grouped into nine ',(,"11;!; I hose nm<> bruad cnle
oories wore established to facilitate further dc'velopmont and application ol en
vironinontal technology (Jirnmation ol trnditionril rjroupiruj was consciuusly
planned to (osier technology Iransler and a n'axunu'n mlerlacf in K'l.ilc'd tic-Id';
The nine series are
I Environmental Health tllects Research
2 Environmental Prelection Technology
3 Ecological Rese-aich
4 Environmental Monitoring
5 Socioc'"onomic Environmental Studies
G ScienUic and Technical Assessment Reports (STAR)
7 Interagency Energy-F:'^':ronm(;nt Research and D(>vL'lor)menl
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 arid 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
-------
1 HF.f'ORT NO.
EPA-600/2-79-013
TECHNICAL REPOH1 DATA
(I'lcasc rent] tiisiniciiini! on ilie reverse lieftirt i
_
4. TITLE ANDSUOTITUG
OXYGEN AERATION AT NEWTOWN CREEK
&. RLI'ORT DATE
_ J.un e . 19.75._. lJ>; a u i n g Da t a)
C. I'Ll'FORMINO OHGANIXAI ION COO I:
Paul J. Krasnoff
7. AUTHOR(S)
Norman Nash
William B. Pressmen
9, PERFORMING OfiG '\NIZATION NAME AND ADDRESS
The City of New York
Departmenc of Environmental Protection
2358 Municipal Building
New York, New York 10007
0. PERFORMING ORGANIZATION REPORT NO,
10. PROGRAM ELfcMENT NO.
NO".~
Grant No. S802714
2. SPONSORING AGENCY NAME AND ADDRESS Cincinnati, Oil
Municipal Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final, 05/72 - Q3/75
HI. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer:
Richard C. Brenner, ( 513)-684-7657
16. ABSTRACT--A successful initial feasibility investigation of oxygen
aeration at the 0.11-m3/sec (2.5-higd) municipal wastewater treatment
plant in Batavia, New York, prompted a larger demonstration at New York
City's 13.6-m3/sec (310-mgd) Newtown Creek Plant. A 34-mo evaluation
was performed in a self-contained set of plant tanks using a 13.6-
metric ton/day (15-ton/day) oxygen generator with liquid oxygen backup
for oxygen supply and turbine mixers and spargers for oxygen
dissolution. For the 34-mo period, at influent flows of 0.44 to 1.53
m3/sec (10 to 35 mgd) , effluent quality averaged 19 ing/1. each of BOD
and suspended solids for removal efficiencies of 88 and 86 percent,
respectively. Removals were not impaired by intentional hydraulic
and BOD overloading of the oxygenation system. During the winter
months, a fungus in the influent sewage caused the oxygenation system
biomass to become filamentous, resulting in a deterioration of sludge
settling and thickening characteristics to varying degrees over the
three winters of the testing program. While operating difficulties
occurred, this condition had no significant effect on the plant
effluent quality.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Sewage treatment, *Activated sludge
process, *Oxygenation, Aeration
tanks, Sedimentation t^nks, *Oxygen,
Upgrading
b.lDF.NTIFIEHS/OPLN ENDED TERMS
DTs solution,
*0xygen activated
sludge system, *Pres-
sure swing adsorp-
tion oxygen gas
generation, Submerged
turbine sparger,
*Covered tank
c. COSATI i'lcld/Croup
13B
IS. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (TliiS Report)
Unclassified
21. NO. OF PAGES
2O. SECURITY CLASS (This page)
Unclassified
22. PRICE
- A® i
EPA Form 2220-1 (9-73)
. S. GOVERNMENT PRINTING OFFICE: W?-557-060/1663 Region No. 5-11
-------
EPA-600/2-79-013
June 1979
OXYGEN AERATION AT NEWTOWN CREEK
by
Norman Nash
William B. Pressman
Paul J. Krasnoff
Environmental Protection Administration
The City of New York
New York, New York 10007
Grant No. S802714
Project Officer
Richard C. Brenner
Wastewater Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
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.
-------
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. Nox-
ious 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 Environ-
mental 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 pol-
lution. This publication is one of the products of that re-
search; a most vital communications link between the researcher
and the user community.
As part of these activities, a pro^f-ct was undertaken in
New York City in mid-1970 to construct and demonstrate a large-
scale municipal oxygen activated sluage system as a follow-up
to successful feasibility studies completed several months
earlier at Batavia, New York. The information documented in this
report from that project should be carefully evaluated by design
engineers and municipal officials responsible for wastewater
treatment process selection.
F]arcis T. Mayo, rxrector
Municipal Enviro:,. nral Research
Laboratory-
-------
ABSTRACT
A successful initial feasibility investigation of oxygon
aeration at the 0.11-m3/sec (2.5-mgd) municipal wastuwater treat-
ment plant in Batavia, New York, prompted a larger demonstration
at New York City's 13.6-m3/sec (310-mgd) Newtown Creek Plant.
The U.S. Environmental Protection Agency desired to further
evaluate oxygen aeration on a scale sufficiently large to esta-
blish reliable engineering design and cost data, and the City
saw the process as a possibility for upgrading the plant's modi-
fied aeration process to step aeration efficiency. A 34-mo
evaluation was performed in a self-contained set of plant tanks
using :< 13.6-metric ton/day (15-ton/day) oxygen generator with
liquid oxygen backup for oxygen supply and turbine mixers and
spargers for oxygen dissolution. For the 34-mo period, at
influent flows of 0.44 to 1.53 m3/sec (10 to 35 mgd), effluent
quality averaged 19 mg/1 each of BOD and suspended solids (SS)
for removal efficiencies of 88 and 86 percent, respectively.
Removals were not impaired by intentional hydraulic and BOD
overloading of the oxygenation system. During the winter months,
a fungus in the influent sewage caused the oxygenation system
biomass to become filamentous, resulting in a deterioration of
sludge settling and thickening characteristics to varying degrees
over the three winters of the testing program. While operating
difficulties occurred, this condition had no significant effect
on the plant effluent quality. Power and oxygen requirements and
sludge production data are presented, and sludge volumes are com-
pared with those of the Newtown Creek diffused air plant and with
two New York City step aeration plants.
This report was submitted in fulfillment of Grant No.
S802714 by the New York City Environmental Protection Administra-
tion, Department of Water Resources, under the partial sponsor-
ship of the U.S. Environmental Protection Agency. This report
covers the operating period of May 14, 19/2, to March 11, 1975.
Also contained in this report in Appendix B is a description and
discussion of the current proposal by Union Carbide Corporation
for conversion of the Newtown Creek plant to an oxygen system.
IV
-------
CONTENTS
Foreword .iii
Abstract iv
Figures , vi
Tables vii
Acknowledgements ix
1. Introduction and Scope of Project. ..... 1
2. Conclusions and Recommendations 5
3. System Description 8
4. Analytical Methods 16
5. Summary of Operations 20
6. Equipment and Instrumentation Operating
Experiences 47
7. Discussion of Results 54
References 66
Appendices
A. Detailed Description of Equipment 67
B. Estimated Capital Cost for Conversion
of the Newtown Creek Plant to Oxygen
Activated Sludge 74
-------
FIGURES
Number
1 Plant Layout 2
2 Plan and Elevation Views of Oxygen Test Bay .... 11
3 Perspective Illustration of Oxygenation Equipment . 12
4 Photograph of Oxygen Reactor Deck . 13
5 Oxygen System Gas Flow and Control Diagram ..... 15
6 Comparison of Newtown Creek Oxygen System and Air
Plant BOD Removal Performance 57
7 Comparison of Nawtown Creek Oxygen System and Air
Plant Suspended Solids Removal Performance .... 58
8 Excess Solids Produced by Newtown Creek Oxygen
System Superimposed on Batavia, New York Sludge
Production Graph 63
VI
-------
TABLES
P aft fi
Number £2iit
1 Project Chronology • 21,22
2 Oxygen System Process Performance .......«• 34
3 Oxygen System Aeration Tank Performance 35
4 Oxygen System Final Tank Performance and Sludge
Settling Characteristics 36
5 Oxygen Supplied to Oxygen System 37
6 Power for Oxygen Generation and Dissolution 37
7 Oxygen System Sludge Production ... • 38
8 Oxygen System Nutrient Removals . . 39
9 Other Oxygen System Sewage Characteristics 40
10 Newtown Creek Air Plant Process Performance .... 41
11 Newtown Creek Air Plant Aeration Tank Performance . 42
12 Newtown Creek Air Plant Final Tank Performance
and Sludge Settling Characteristics 42
13 Power Consumption for Aeration at Newtown Creek
Air Plant > 43
14 Newtown Creek Air Plant Sludge Production 43
15 Power Consumption for Aeration at Jamaica Step
Aeration Air Plant 44
16 Power Consumption for Aeration at 25th Ward
Step Aeration Air Plant 44
17 Sludge Production at Jamaica Step Aeration Air
Plant 45
vii
-------
TABLES (continued)
18 Sludge Production at 26th Ward Step Aeration
Air Plant .................... 46
19 Comparison of Newtown Creek Wastewater Character-
istics Before and After Introduction of Manhattan
Flow ....... , . .............. 74
20 Design Operating Conditions for Full-Scale Newtown
Creek Oxygen System .......... . . , . . 78
21 Estimated Capital Cost for Oxygen System Retrofit
at Newtown Creek . . „ ............. , 79
viii
-------
ACKNOWLEDGEMENTS
The authors wish to thank Quentin P. Monahan, Superintendent
of the Newtown Creek plant, Jerome Degen, Senior Chemist in
charge of the Industrial Wastes Control Laboratory at Newtown
Creek, and their capable staffs for their invaluable contribu-
tions to this project.
Stephen Y. Arella, Assistant Civil Engineer assisted in field
supervision and data reduction; Hank Innerfeld, Principal Engineer-
ing Technician, and Angelika Forndran, Assistant Civil Engineer,
performed some tabulation and reduction of the data.
IX
-------
SECTION 1
INTRODUCTION AND SCOPE OF PROJECT
Plant-scale feasibility studies in 1969 and 1970 at the
O.ll-mVsec (2.5-mgd) activated sludge plant in Batavia, New
York, indicated that significant improvement of the process could
be achieved by the use of pure oxygen instead of diffused air,
and that considerabie capital and operating savings were possible
if oxygen was used in the upgrading of existing plants (1). To
the City of New York, this offered a means of upgrading its
Newtown Creek Water Pollution Control Plant, which was designed
to treat 13.6 m^/sec (310 mgd) by modified aeration from a pop-
ulation of 2.5 rr.xllion in the boroughs of Manhattan, Brooklyn,
and Queens. The plant began operation in September 1967, but
with only the flow from Brooklyn and Queens; delays in the con-
struction of a pumping station prevented the delivery of the
estimated 7,4-m3/sec (170-mgd) Manhattan flow through a force
main under the East River. During the 34 mo of operation des-
cribed in this report, the flow to the plant, including that to
the oxygen aeration segment, averaged only 7.5 m3/sec (171 mgd).
The Newtown Creek plant consists of 16 treatment modules
16.8 m (55 ft) wide and 190.5 m (625 ft) long, with flow-through
transverse baffles dividing each tank into ? pair of aerated grit
chambers, an aeration tank, and a settling tank, but with no pri-
mary settling tanks (Figure 1). After screening, the raw sewage
is pumped and divides north and south to enter channels leading
to two sets of eight modules. From this point on, each module
is hydraulically separate from the others, except that the return
sludge from each battery of eight tanks is combined and returned
to the same battery.
For its first 3 yr of service, the plant was operated at a
sludge age* averaging 0.35 day, but removals were only 37 per-
cent for BOD and 50 percent for SS, considerably below the res-
pective 60 and 70 percent design levels. In January 1971, the
City, upon the recommendacion of consulting engineers who had pre-
pared a report on the upgrading of some of the plant's facilities,
decided to venture into the usually hazardous sludge age range be-
tween 0.7 day and 3.5 days. The results were surprising: by
the use of four of the seven engine-generators and five of the
*Defined as kg mixed liquor suspended solids (MLSS) in aeration
tank/kg SS in aeration tank influent/day.
-------
STREET
STREET
tu
LU
1 SLUDGE '
DIGESTION
OXYGEN AERATION
TEST BAY
o o
O O
po o
IPO
_A IF c i
-^ If SLUDGE
" CONCENTRATION
AERATED GRIT
CHAMBERS
AERATION
TANKS
N
SECONDARY
CLARIFIERS
K
to
r
Figure 1. Plant layout.
-------
six blowers, by supplying an air to influent wastewater ratio of
9.0 m3/m3 (1.2 ft3/gal), and by maintaining mixed liquor sus-
pended solids (MLSS) concentrations at 1800 mg/,1 (equivalent to a
sludge age of 1.0 day), efficiencies were greatly increased to
average BOD and S3 removals of 81 and 74 percent, respectively.
However, this could only be viewed as a temporary success; the
inevitable arrival of 7.4 m3/sec (170 mgd) of additional sewage
from Manhattan would double the flow, halve the aeration time,
double the final tank overflow rate, and make it impossible to
maintain the higher air to influent wastewater ratio and sludge
age. Some method of improving the plant's efficiency was neces-
sary, and conversion to oxygen aeration offered a potential solu-
tion. Therefore, in June 1970, the City applied for Federal
support for a 0.88-m3/sec (20-mgd) test of the UNOX system*.
The unique layout of the plant was helpful. One module
could be isolated from the rest of the plant by the installation
of its own influent pumps and its own return sludge pump. Two
0.88-m3/sec (20-mgd) pumps were provided, one constant-speed
and one variable from 0.44 to 0.88 m3/sec (10 to 20 mgd) to take
suction from the grit chamber, along with a 0.66-mJ/sec (15-mgd)
variable-speed, return sludge pump to recycle settled sludge to
the head of the aeration tank. Thus, the oxygen activated sludge
system -.'ould be a separate plant within the Newtown Creek plant.
Provisions were made for the measurement of all important physi-
cal and chemical characteristics by an elaborate array of instru-
mentation. Alum storage and feeding equipment* was installed for
a test of phosphorus removal. The initially budgeted cost of
the project was $2,796,465, of which the Federal government pro-
vided $1,574,625 and the City $1,221,840. Additional overrun
costs of approximately $900,000 were shared by the City and
Union Carbide.
The program was designed to answer these questions:
1. Would the use of oxygen enable Newtown Creek to increase its
BOD and SS removals to those of conventional activated sludge
within the existing tank volume?
2 Would the plant's secondary clarifier design overflow rate
of 35 9 m3/day/m2 (880 gpd/ft2) be adequate? (As a precaution,
the adjoining Module 10 was kept out of service to be used for
additional settling capacity if necessary.)
*A proprietary oxygen aeration system of the Union Carbide Cor-
poration.
*This equipment was never used because a detergent phosphate ban
implemented by the State of New York reduced influent phospho-
rus at Newtown Creek to a level consistently below 3 mg/1.
-------
3. Were the claims of lower sludge production and lower power
requirements for oxygeneration** valid?
4. Could mixed liquor dissolved oxygen (DO) concentrations of
8-10 mg/1 be obtained without economic penalty?
5. Could mixed liquor volatile suspended solids (MI/VSS) con-
centrations of 4500 mg/1 or greater be maintained?
6. Would the activated sludge concentrate to 3 percent total
solids or more in the sedimentation tank and permit low sludge
return rates?
**0xygenation and oxygen aeration are used interchangeably in
this report.
-------
SECTION 2
CONCLUSIONS AND RECOMMENDATIONS
During the nearly 3-yr operating period of this project,
some problems were encountered which have made it difficult to
draw the type of broad conclusions that investigators hope for.
These problems included failure of the single return sludge pump
during two of the three winters, which required shutdowns and re-
starts of the process at low wastewater temperatures; the pre-
sence in the Newtown Creek influent of a fungal organism which
proliferated in both the air and oxygen aeration tanks; lower
removal efficiencies following the two cold-weather startups,
but satisfactory removals during the third uninterrupted winter
despite the continued presence of the fungi; a wide variety of
loading rates; and a number of mechanical, electrical, and in-
strumentation problems. Therefore, the data have been separated
into several groups with different operating conditions and are
discussed in the light of those conditions.
1. Over the life of the project, BOD and SS removal efficien-
cies were 88 and 86 percent, respectively. The effluent aver-
aged- 19 mg/1 for both BOD and SS.
2. During one phase of the test when the fungi were not present
(hereafter "non-filamentous"), and the flow was at the design
point of 0.88 m3/sec (20 mgd) (uniform rate, not diurnal), efflu-
ent quality averaged 9 mg/1 of BOD and 12 mg/1 of SS, for remov-
als of 94 and 92 percent, respectively.
3. During two non-filmamentous phases at 0.88-m3/sec (20-mgd)
diurnal flows, effluent quality averaged 21 and 22 mg/1 of BOD
and SS, for removals of 87 and 86 percent, respectively.
4. During non-filamentous phases at volumetric aerator load-
ings of 4.95 to 5.64 kg BOD/day/m3 (309 to 352 lb/day/1000 ft3)
and overflow rates of 47.3 to 65.6 m3/day/m2 (1160 to 1610 gpd/
ft2), the effluent BOD ranged from 21 to 24 mg/1 and the efflu-
ent SS from 17 to 26 mg/1, representing average removals of 90
and 85 percent, respectively. These removals were obtained at
BOD loading rates of 123 to 140 percent of design and influent
flows of 128 to 177 percent of design.
5. Slightly less successful efficiencies were obtained during
three phases, totaling nearly 6 mo, each following a shutdown of
the process and at low sewage temperatures. Effluent quality
-------
averaged 21 mg/1 of BOD and 22 mg/1 of SS, for removals of 86 and
84 percent, respectively. Operation during these three low-temp-
erature process startup periods was characterized by the presence
of the influent fungi which caused varying degrees of difficulty
with sludge settling and wasting and with maintenance of the
desired MLSS concentration. However, effluent quality was not
significantly affected.
6. Removals during a fourth low-temperature phase extending
over 3 mo averaged 87 percent for both BOD and SS, with effluent
concentrations of 18 mg/1 of BOD and 13 mg/1 of SS, at the de-
sign diurnal flow of 0.88 m3/sec (20 ragd). The fungi were pre-
sent in the influent, as they were during the preceding two
winters, but their concentration in the mixed liquor did not
appear to be as great. Fungi concentrations were determined
numerically during this last period, but were only visually
estimated during other phases.
This was the only winter operation that was not halted by
a failure of the return sludge pump. There is reason to believe,
therefore, that the lower efficiencies and larger volumes of ex-
cess sludge during the first two winter periods were caused by
the combination of startups of the process at low sewage temper-
.tures and the presence of the influent fungi.
7. Excess solids production on a total suspended solids (TSS)
basis during non-filamentous periods averaged 0.93 kg/kg BOD re-
moved, and 1.27 kg/kg BOD removed during filamentous periods.
For the Newtown Creek modified aeration plant at comparable
sludge ages, the averages were 1.27 and 1.71 kg TSS/kg BOD re-
moved, 37 and 35 percent greater, respectively.
Compared to the 1 available yr of data from the Jamaica
step aeration plant, oxygen produced 20 percent less solids dur-
ing non-filamentous periods, but 9 percent more during filament-
ous periods, at sludge ages about one-third of those under step
aeration. Compared to the first 17 mo of operation of the re-
cently-upgraded 26th Ward step aeration plant, oxygen produced
33 percent less solids during non-filamentous periods and 9 per-
cent less during filamentous periods, at half the sludge age.
However, it must be noted that these plants were operated at
significantly lower loading rates and on much weaker wastewaters
than the Newtown Creek plant.
8 During the three periods of above design BOD loading rates,
when the equipment was fully stressed, oxygen system power re-
quirements averaged 0.95 kWh/kg BOD removed (0.58 hp-hr/lb).
For the single year at the Jamaica plant, the range was 1.21 to
1 96 kWh/kg BOD removed (0.74 to 1.19 hp-hr/lb). In addition,
the first 17 mo of operation at the recently upgraded 26th Ward
step aeration plant produced a range of 0.71 to 1.05 kWh/kg BOD
removed (0.43 to 0.64 hp-hr/lb). Caution should be exercised in
-------
directly comparing these numbers since the wastewater character-
istics and plant loadings were different for these two facilities
relative to Newtown Creek.
9. At or above the design flow of 20 mgd, oxygen supply averaged
1.0 kg/kg BOD removed. At flows below design, oxygen supply
ranged from 1.2 to 1.6 kg/kg BOD removed.
10. The fungal organisms which affected cold weather operation
entered the plant in the influent sewage and concentrated in the
mixed liquor. The Newtown Creek air plant, in which the organisms
were also present, was less seriously affected during the first
two winters, but more seriously in the third.
11. Foam (probably of Nocardia origin) twice developed in the
aeration tank, and once required a halt to the oxygen feed. Some
means of suppressing or removing foam should be provided.
-------
SECTION 3
SYSTEM DESCRIPTION
An oxygen aeration system is a high-rate activated sludge
process which employs oxygen instead of air for the biological
removal of organic pollutants from wastewater. In any system,
the rate of oxygen transfer is directly proportional to the con-
centration of oxygen in the aerating gas; thus, pure oxygen
should support a more concentrated biomass by making more oxygen
available. Because higher MLSS concentrations imply higher con-
centrations of microorganisms, reduced detention times and,
thus, smaller aerator volumes are claimed to result for a given
level of organic removal.
DESIGN FOR NEWTOWN CREEK
Tank No. 9 (Figure 1) was selected for the demonstration be-
cause its location adjacent to a wide plant road allowed room for
the installation of the oxygen generator, liquid backup tank, and
the control building. The existing aerator piping and diffusers
were removed, and three full-width, reinforced-concrete baffle
walls and a lightweight, precast-concrete, gas-tight cover were
installed. A 0.9-m (3-ft) gas space was obtained between the
mixed liquor surface and the cover, and separate openings in the
walls permitted the flow of mixed liquor and gas from one stage
to the next. Thus, four co-current gas and liquid stages were
created, each 15.2 m long x 16.8 m wide x 5.5 m high (50 ft x 55
ft x 18 ft) with a 4.6-m (15-ft) side water depth (SWD) and a
total liquid volume of 75,700 m5 (1.23 mil gal). The equipment
was designed for an average flow of 0.88 m3/sec (20 mgd) at an
influent strength of 250 mg/1 of BOD.
A system of reinforced concrete beams and girders was in-
stalled to support the mixers, and additional columns were in-
stalled in the tank to transmit the static and dynamic loads to
the existing foundation. The cover was designed to sustain an
internal positive pressure of 15.2 cm (6 in) of water and an
external negative pressure of 10.2 cm (4 in), and to support a
live load of 488 kgf/m2 (100 lb/ft2). The cover was free to
expand and contract with changes in temperature.
The primary oxygen supply for the UNOX system was a Union
Carbide pressure swing adsorption (PSA) air separation facility
which was designed to produce 473 std m3/hr (16,700 scfh) of gas
containing 425 std m3/hr (15,000 scfh) of pure oxygen. The unit
-------
consisted essentially of an air compressor, aftercooler and at-
tendant cooling tower, adsorption beds and valve skid, and auto-
matic flow controls. The PSA unit was backed up by a liquid
oxygen storage tank with an electric water bath vaporizer capable
of supplying up to 510 std m3/hr (18,000 scfh) of pure oxygen.
Flow from the liquid oxygen storage tank was automatically con-
trolled by the pressure in the line, which fell when either the
PSA was not supplying oxygen or when -.he system called for more
oxygen than the PSA could supply. Evaporation losses from the
liquid oxygen storage tank were recovered by piping gaseous
oxygen which escaped from the tank into the oxygen feed line.
Each stage contained two mixer-spargers suspended from the
cover for oxygen dissolution. A total of five compressors were
provided at the side of the tank, two for the first stage and
one each for the second, third, and fourth stages, to recirculate
the gas. The compressed gas was pumped through the hollow rotat-
ing shafts of the mixers and sparged into the liquid from the
bottom of the shafts. Three-bladed, 1.8-m (6-ft) diameter,
marine propellers attached to each shaft dispersed the oxygenated
wastewater and kept the solids in suspension. The mixers were
equipped with automatic controls for shutdown if the oil level
was too low, the vibration too extreme, or the seal air pressure
too low.
Degritted influent sewage entered Stage 1 of the tank from
the grit chambers, which acted as a well for the two 112~kw (150-
hp) influent pumps. A magnetic flow meter monitored the influent
flow, which was recorded and totalized.
The oxygen feed to Stage 1 was controlled by the pressure in
the stage gas space; if the pressure dropped below a pre-set
limit, an automatic valve opened further to admit more oxygen.
This flow was monitored by a temperature-compensated orifice
meter and recorded and totalized. The gas in Stage 1 passed
first through a trap to remove particulate matter and then into
a 18.4-std m3/hr (650-scfm) recirculation compressor. Two com-
pressors were provided for Stage 1; either or both could be used
if required. The recirculation systems for the second and succeed-
ing stages were the same as for Stage 1, except that manual ad-
justment of a bypass valve controlled the recirculation rate
in these stages. The gas leaving the fourth and final stage
was analyzed for oxygen content and metered,
In anticipation of denser sludge, the final clarifier sludge
collector drive was modified to increase its speed to 1.65 m/min
(5.4 ft/min) [before the alteration, two-speed operation was pos-
sible, 0.61 m/min (2.0 ft/min) and 0.91 m/min (3.0 ft/min)].
Later, to decrease wear on the mechanism, the speed was reduced
to 1.34 m/min (4.4 ft/min), the minimum possible after the modi-
fication.
-------
A 0,66-m3/sec (15-mgd), variable-speed, return sludge pump and a
1.14-m-Vmin (300-gpm) waste sludge pump were provided, along
with magnetic flow meters and recording and totalizing equip-
ment.
Section and plan views of the Newtown Creek oxygen test
bay, including both reactor and secondary clarifier, are shown
in Figure 2. A perspective illustration of the oxygenation
equipment and oxygen reactor are presented in Figure 3. The
reactor deck and turbire-sparger drives are shown photographic-
ally in Figure 4.
CONTROL SYSTEM
Sensing devices were installed in Stages 1 and 4 to
monitor the concentration of volatile hydrocarbon gases in the
reactor gas space. In the event of a hydrocarbon entering the
reactor with the influent wastewater and reaching a concentration
in the gas space of 25 percent of the lower explosive limit (LEL)
of hexane, an alarm would sound and automatically halt the feed
oxygen flow. The PSA compressor would then supply purge air at
a rate to completely renew the reactor gas space within 15 min.
If the gas space hydrocarbon concentration continued to rj.se,
another alarm would sound at 50 percent LEL and the recirculation
compressors would be shut down to eliminate them as a potential
ignition source. After any hydrocarbon alarm situations, the
equipment could be returned to operation only manually and only
after the hydrocarbon level decreased to below the 25 percent
LEL concentration.
Oxygen Feed
Oxygen feed flow to the system was automatically controlled
by the gas pressure in the first stage. If the organic load in-
creased, the oxygen uptake rate rose, the pressure in the gas
space decreased, and a pressure-regulating device compensated
to allow increased oxygen feed. If the PSA could not supply
enough oxyg=>,n to maintain system pressure, liquid oxygen was
automatically vaporized and added. If the organic loading de-
creased, the oxygen uptake rate would fall and the system pres-
sure would rise, causing suction throttling of the PSA compres-
sor and a reduced flow of oxygen to the system.
Oxygen Vent
As the organic loading to the system changed, the fourth
stage vent gas purity changed and adjustments had to be made
to return the purity to the desired level. An increased organic
loading caused the vent gas purity to decrease, and a decreased
loading caused the purity to increase. Adjustments to the vent
10
-------
PUMPS
RAW
SEWAGE
RAW
SEWAGE
^
FIVE GAS RECIRCULATING
COMPRESSORS
D D D D
EIGHT SUBMERGED
PROPELLER MillERS
O
O
O
O
200'
55'
SECONDARY
EFFLUENT
400'
( DRIVE
1 PROPELLER
ASSEMBLY I SPARGER
MIXER
n
n
PLAN VIEW
NO SCALE
n
'id
kr
<
L
<
cd
1
I
t
»
*.-..!-.» rrf
Bw
n — ~=S. — — O -m -Q ' — "~*
D~ — u~- O — *- Q
— E-
SECONDARY
EFFLUENT
GRIT 1 FOUR-STAGE OXYGEN AERATOR ' y /"^SECONDARY
CHAMBER!
WD = 15
I
^— SLUDGE
RECYCLE
T CLAR1FIER
_ ..] U/nr:19 '
ELEVAT ON 1
NO SCALE !
FT - 0.305 M
I
SLUDGE
WASTING
Figure 2. Plan and evaluation views of oxygen test bay.
-------
SEDIMENTATION
TANK No. 9
AERATION TANK No. 9
CHIT
CHAMBERS
PSA OXYGEN
GENERATION SYSTEM
Figure 3. Perspective illustration of oxygenation equipment.
-------
0
•p
C
fO
(1)
c
OJ
en
>i
x
0
4-1
O
fO
J-l
tn
O
4J
O
X
^
-------
gas purity were accomplished by controlling the flow rate through
the vent valve. If the purity fell below the desired level, the
vent flow was manually increased; if the purity rose above the
desired level, the vent flow was manually decreased. The entire
oxygen system gas flow and control network is diagrammed in
Figure 5.
Liquid Flows
The influent sewage flow rate was automatically maintained
at an operator-selected level. An automatic flow variation pat-
tern could be achieved by the use of a programmer to simulate
the diurnal flow variations of the existing air activated sludge
plant. The return sludge flow rate could be set at a constant
rate or at a pre-set fraction of the influent flow, and the waste
sludge flow rate was manually controlled using a variable-speed
pump.
14
-------
REACTOR
PRESSURE
CONTROL
^^^
NOTES:
I HP = 0746 KW
I PSIG = 703 KGF/M2
< I, i I
>^=j JLJ=| Ju^ JS
'S'"*'1*5'*:*15*5'*^
Figure 5. Oxygen system gas flow and control diagram.
-------
SECTION 4
ANALYTICAL METHODS
BOD
Total and filtrate BOD analyses, in two dilutions, were
performed daily on influent and effluent samples. Samples for
filtrate BOD were filtered through Whatman No. 1 paper. Begin-
ning in January 1973, dissolved oxygen readings for BOD were
taken with a DO meter and probe (Weston and Stack Model 300
meter with Model 3A probe, and Model 350 meter with Model 33
probe) instead of the Winkler titration method previously used.
Winkler titrations were used to calibrate the meter daily. DO
readings by meter saved time and proved to be reliable.
The procedures outlined in Standard Methods for the Examina-
tion of Water and Wastewater (2) were used, except in the deter-
minations of initial DO levels of the sample dilutions, which
were done according to the practice of the New York City Depart-
ment of Water Resources as described in the following paragraph.
DO measurements of the dilution water were taken on the day
the dilutions were prepared. The post-incubation DO levels of
the seed controls (5 and 10 percent seed incubated for 5 days)
were applied to those measurements to calculate the DO of the
dilution water (I percent seed) after 5 days of incubation. The
final DO levels of the incubated sample dilutions were then sub-
tracted from this calculated blank to obtain the depletions due
to the samples. An extensive comparison was performed of this
method and the standard method of taking the initial DO reading
of each sample dilution. BOD values obtained by the two techni-
ques were very close, well within the accuracy of the BOD test.
Although the final effluent often was supersaturated, handl-
ing, compositing, and shaking dissipated the supersaturation by
the time the BOD analysis was begun. The dilutions, therefore,
had essentially the same initial DO level as the dilution water,
despite the high percentage of effluent used in the dilutions.
Laboratory work generally was not performed on weekends;
Saturday and Sunday samples were refrigerated until Monday.
Dilutions which normally would have been removed from the incu-
bator on Saturday and Sunday were removed on Friday and Monday,
and the resulting BOD values adjusted to 5-day values according
to Strester & Phelps (3). The validity of applying the factors
16
-------
was confirmed by experiment.
SOLIDS
Suspended solids analyses were performed or, daily influent,
effluent, Stage-4 mixed liquor, and return sludge samples. Glass
fiber filters (Whatman GF/C with Gooch crucibles, permitted by
Standard Methods (2) as an alternate) were initially used for all
samples. Mixed liquor and return sludge samples were diluted
prior to withdrawing an aliquot for filtration.
Starting in May 1973, Buchner funnels were used instead of
Gooch crucibles for mixed liquor and return sludge samples because
their larger filtration area permitted larger aliquots to be
filtered v;ithout dilution. Some ceramic mateial was cut from the
funnel-: to keep their weight within the capacity of the analytical
balance. Volatile suspended solids were determined after ignition
at 550 C.
Total solids and total volatile solids determinations were
performed in accordance with Standard Methods (2). Suspended
solids and volatile suspended solids values were substracted from
these to obtain dissolved solids and volatile dissolved solids
values. Dissolved solids analyses were performed twice a week on
influent and effluent samples, but were discontinued in June 1973,
SETTLING DATA
Two 1000-ml graduated cylinders, one equipped with a 1-rpm
mechanical stirrer, were filled with well-dispersed, fresh,
Stage-4 mixed liquor. Readings of the volume of settled sludge
were taken after 1,2,3,5,7,10,15,20,25, and 30 min.
The settling rate in ml/min was determined for each time
increment, and the highest of these rates was taken as the initial
settling rate (ISR) and reported as m/hr (ft/hr).
The 30-min settled volume was used to calculate the sludge
volume index (SVI) and sludge density index (SDI) as specified in
Standard Methods (2).
pH AND ALKALINITY
pH analyses were performed on daily influent and effluent
samples and, starting in March 1973, on Stage-4 mixed liquor
settling test samples, using a Hach Model 2075 pH meter with a
combination electrode. Alkalinity determinations were performed
twice a week on influent and effluent samples by potentiometric
titration to pH 4.5.
Turbidity determinations were made twice a week on the
supernatant liquor from settled influent and effluent samples
17
-------
using a Hach Model 2100 turbidimeter, Doth samples required 15
min of settling to remove large particles which caused erratic
meter readings. Commercial Jackson Turbidity Unit (JTU) stand-
ards were used for calibration. These analyses were discontinued
in July 1973.
COD
COD analyses were performed on daily influent, effluent,
effluent filtrate (Whatman No. 1 paper), and return sludge
samples. Mercuric sulfate was added to the digestion flask to
inhibit the interfering effect of chlorides. Fifty-mi samples
or aliquots diluted to 50 ml were used. Influent and return
sludge samples were digested in the presence of a strong dichro-
mate solution (0.25N); a weaker solution (0.025N) was used for
the effluent samples. Refluxing for 1% hours was found to be
sufficient for complete digestion.
NUTRIENTS
Total Kjeldahl nitrogen (TKN), ammonia nitrogen, total
phosphate, and soluble orthophosphate analyses were performed
on influent and effluent samples, but not every day. The
analyses were alternated so that total Kjeldahl and ammonia
nitrogen analyses were run daily one week, and total and soluble
orthophosphate analyses were performed daily the following week.
The samples were preserved with 40 mg/1 of mercuric chloride and
refrigerated until analysis. Nitrate and nitrite nitrogen analy-
ses were performed on fresh influent and effluent samples twice
a month until March 1973, after which they were done only once
a month.
All nutrient analyses were performed with a Technicon Auto-
analyzer system. An Autoanalyzer I was used until November 1973,
and the Autoanalyzer II system thereafter.
Ammonia nitrogen analyses were performed according to the
automated method in Methods for Chemical Analysis of Water and
Wastes (MCAWW) (4). Samples were not distilled prior to analy-
sis .
Samples for total Kjeldahl nitrogen analysis were digested
and distilled according to Standard Methods (2), except that
0.02N sulfuric acid was user" for collection instead of boric
acid. Following distillation, the arr-nonia nitrogen analysis,
described above, was performed.
Total phosphate and soluble orthophosphate analyses were
performed by the automated stannous chloride method until
November 1973, when it was replaced by the automated single re-
agent method (ascorbic acid method). Both methods are described
in MCAWW (4).
18
-------
The automated hydrazine reduction method in MCAWW (4) was
used for nitrate and nitrite nitrogen analyses.
TOG
Beginning in October 1972, total organic carbon analyses
were performed on daily influent and effluent samples with a
Beckman Model 915 TOG analyzer using the procedure recommended
by the manufacturer. The samples were preserved by adding mer-
curic chloride to a concentration of 40 mg/1 and then refrigera-
ted. Before analysis, they wore acidified and purged with nitro-
gen to remove inorganic carbon and then homogenized in a blender
for 5 min for greater uniformity and reproducibility.
19
-------
SECTION 5
SUMMARY OP OPERATIONS
The Newtown Creek oxygen system was placed in operation May
14, 1972. The testing program was concluded on March 11, 1975.
In the intervening 34 mo, a broad range of operating conditions
comprising a total of 12 evaluation phases was imposed on the
system. Throughout, an extensive array of data were collected
daily.
The chronology of events is summarized in Table 1. Each
operating phase, shutdown, and restart is discussed individually
below. Operating and performance data generated on the oxygen
aeration system for Phases 1-12 are presented in phase-average
form in Tables 2-9 at the end of this section. For comparative
purposes, operating and performance data developed for the same
period of time for the Newtown Creek conventional air aeration
plant are also recorded at the end of this section in Tables 10-
14, along with power consumption and sludge production data for
two City step aeration air plants (Jamaica and 26th Ward) in
Tables 15-18.
STARTUP (MAY 14-SEPTEMBER 16, 1972)
The intent was to break in the process while collecting data
at a flow of 0.44 mj/sec (10 mgd) for 1 to 2 mo, and then at 0.66
m-Vsec (15 mgd) and possibly one other intermediate point before
reaching the design flow of 0.88 m3/sec (20 mgd) in late summer
1972. The summer, however, was characterized by many equipment
and instrumentation problems; these are discussed in Section 6
of this report. In mid-September 1972, with the flow at 0.88
mj/sec (20 mgd), it was found that most of the data already col-
lected to that point were unusable; large errors were discovered
in the metered influent sewage flow, return sludge flow, and
oxygen feed flow. The data collected through mid-September 1972
are, therefore, not presented.
PHASE 1 (SEPTEMBER 17-NOVEMBER 25, 1972)
With the flow at 0.88 m3/sec (20 mgd) and the instrumenta-
tion considered reliable, Phase 1 was begun. Waste sludge data,
however, are not presented until Phase 3, after the defective
magnetic flow meter was replaced (see Section 6). The flow was
kept at a constant 0.88 m3/sec (20 mgd) until the last week, when
a diurnal variation based on the air plant influent pattern and
20
-------
TABLE 1. PROJECT CHRONOLOGY
Dates Phase
5/14-9/16/72
9/17-11/25/72 1
11/26-12/9/72
12/10/72-
2/1/73 2
2/2-2/24/73
2/25-4/5/73 3
4/6-5/31/73 4
6/1-6/30/73 5
7/1-7/7/73
7/8-8/11/73 6
8/12-9/1/73 7
9/2-9/15/73
No. of
Days
126
70
14
54
23
40
56
30
7
35
21
14
Influent
Flow Rate
(mgd) * Remarks
11-20 Startup; faulty
meters
20.8 Warm weather; design
flow Cconstant)
Shutdown for return
sludge pump repair
17.7 Restart; winter oper-
ation; variable flows
(constant) ; fungus
Shutdown for tank
cleaning; restart
15.1 Winter operation;
75% design flow (con-
stant) ; fungus
20.3 Warm weather; design
flow (diurnal)
25.6 Overload; diurnal
flow
Flow increased; pro-
cess unstable; PSA
difficulties
30.0. Overload; diurnal
flow
35.4 Overload; diurnal
flow
Flow reduction due
9/16-10/8/73
to PSA compressor
motor bearing fail-
ure; process recov-
ery
23
10.1
Underload; diurnal
flow
(conti/nued on next page)
21
-------
_TABLE 1. (continued)
Dates
Phase
10/9-10/13/73
10/14-10/25/73 9
10/26/73-2/5/74
2/6-4/30/74 10
5/1-8/5/74 11
8/6-8/25/74
8/26-12/12/74
12/13/74-
3/11/75 12
No. of
Days
12
103
84
97
20
109
89
Influent
Plow Rate
(mgd)*
Remarks
Flow reduction due to
sludge collector re-
pair; process recov-
ery.
14.6 Foam problem; under-
load; diurnal flow
Shutdown for foam
problem and return
sludge pump repair;
restart
19.7 Winter operation;
design flow (diurn-
al) ; fungus
19.1 Spring & summer oper-
ation; design flow
(diurnal); recover
from fungus
Flow changed pre-
vious to shutdown
Shutdown for PSA
rehabilitation; re-
start
20.1 Winter operation;
design flow (diurn-
al) ; fungus
*1 mgd = 0.044 m3/sec = 3785 m3/day
22
-------
3
ranging from 0.61 to 1.05 m /sec (14 to 24 mgd) was introduced.
The flow for the period averaged 0.91 m3/sec (20.8 mgd). The
aerator detention time, including return sludge flow/ was 1.1 hr;
the volumetric organic loading v/as 2.61 kg BOD/day/m-J (163 lb/
day/1000 ft3); and the food-to-microorganism (P/M) loading was
0.63 kg BOD/day/kg MLVSS (see Table 3). The MLSS concentration
was 4860 mg/1, and the SDI averaged 2.2 g/100 ml. The final tank
detention time (Table 4) was 2.3 hr and, at an average return
sludge flow of 30 percent, the return sludge concentration was
16,200 mg/1. The overflow rate was 39 m3/day/m2 (950 gpd/ft2),
and the solids loading 244 kg/day/m^ (50 Ib/day/ft^). Process
performance (Table 2) was excellent: 94 percent total BOD
removal, 95 percent filtrate BOD, 83 percent COD, 92 percent
suspended solids, and 77 percent TOG.
SHUTDOWN (NOVEMBER 26-DECEMBER 9, IJ72)
During the week of November 9, 1972, it became apparent that
the return sludge pump would have to be taken out of service for
replacement of its bearings. Since the repairs were expected to
take only 2 to 3 days, the aeration tank was not drained; raw
sewage addition was halted on November 26, but some oxygen feed
was maintained, about 3.2 metric tons/day (3.5 tons/day). The
job, however, required 10 days to complete because the bearings
in stock proved to be the wrong size and it took the manufac-
turer a week to locate and deliver the proper bearings. The re-
sult was that Phase 2 began with the handicap of anaerobic return
sludge.
PHASE 2 (DECEMBER 10, 1972-FEBRUARY 1, 1973)
The influent flow was resumed on December 6, 1972, initially
at 0.22 m^/sec (5 mgd) and increasing to the design 0.88 m^/sec
(20 mgd) by December 19. Performance during this period was not
as good as it had been before the shutdown. Total and filtrate
BOD and SS removals fell to 83-87 percent and COD removal to 76
percent. Sludge compaction also deteriorated with the SDI aver-
aging 1.8 g/100 ml compared to 2.2 before the shutdown
In January 1973, a microscopic examination told the story:
filamentous organisms had become established in the system. An
attempt was made to stabilize the process by reducing the influ-
ent flow, but the filamentous forms persisted. It was, there-
fore, decided to shut down the process, empty the tanks, and then
start up again and establish a new culture.
SHUTDOWN (FEBRUARY 2-24, 1973)
Advantage was taken of this otherwise unhappy event to in-
spect the interior of the aerator. No noticeable corrorsion and
only an insignificant accumulation of grit was found. No rags
were wrapped around the mixer blades and shafts; this was an un-
23
-------
expectedly happy development because Newtown Creek has a heavy
*nrt inar', ?fPa^S and minor Codifications also were made,
of
PHASE 3 (FEBRUARY 25-APRIL 5, 1973)
The purpose of this phase was to continue the collection of
/SSC (2° mgd) Which was be» in
ae2bn 1; c was e^»
Phase 2, but with diurnal flow variations. The defective waste
sludge magnetic flow meter had been replaced on March 2 1973
and so the data would be complete for the first time. The inten-
tion was to maintain a sludge age of 3.5 days and slowly move
from an initial 0.22-m3/sec (5-mgd) flow to 0.88 m3/sec (Tmqd)
favoTa hea?^d h^ &t ^ highe? Sludge ^ ' competition wSu?d
favor a healthy biomass over any intruding forms. However, al-
?Sf F I" ^tavtuP' filamentous organisms became apparent and, by
the first days of March, with the flow at 0.53 m3/sec (12 mad)
operation had become impaired . The f il nmentous f orSs , which were
Preval'ent'th^ f 1S^aS,a fUn
-------
PHASE 4 (APRIL 6-MAY 31, 1973)
n *i Jfi<:? ^S ^°W at the desi9n Point, a diurnal variation from
0.61 to 1.05 m-Vsec (14 to 24 mgd) and averaging 0.88 mVsec (20
mgd) was instituted on April 6, 1973, and Phase 4 was under way
For the next 3 wk, with the fungus still present and the sludge
age dropping even farther to 1.3-1.4 days, treatment efficiency
remained good. Total BOD, filtrate BOD, and SS removals hovered
around 90 percent, and COD removal was in the 75-80 percent
range. By the end of April, as the sewage temperature topped
60 F indicating the end of the winter temperatures, the fungus
in the sludge diminished and finally, in early May, disappeared.
But some questions remained. If the shutdown had not occur-
red and required a winter restart, would the healthy culture have
prevented the intrusion of filamentous forms? If the fungus was
indeed a cold weather occurrence that would have to be tolerated
could the design flow be treated throughout the winter? To answer
these questions, another full winter of operation would be requir-
ed. With the mutual concurrence of the City and EPA, the proiect
was, therefore, extended to the end of April 1974.
PHASE 5 (JUNE 1-30, 1973)
One of the program's objectives was to stress the process to
performance breakdown. Therefore, since the plant was perform-
ing well at 0.88 m3/sec (20 mgd), influent flow was increased on
June 1, 1973, to 1.10 m3/sec (25 mgd) with diurnal variations
from 0.74 to 1.31 m3/sec (17 to 30 mgd). The increased flow de-
creased the aerator detention time, including return sludge flow,
to only 0.81 hr. A sharp increase in BOD strength to 240 mg/1
more than doubled the volumetric organic loading to 4 95 kq
BOD/day/m-^ (309 lb/day/1000 ft3) and also doubled the F/M loadina
to 1.38 kg BOD/day/kg MLVSS. The final tank detention time drop-
ped to 1.85 hr, and the final tank solids loading and overflow
rate increased to 308 kg/day/m2 (63 Ib/day/ft2) and 47 m3/day/m2
(1160 gpd/ft^) , respectively. Process efficiencies for this phase
were very respectable: 90 percent total BOD removal, 89 percent
filtrate BOD removal, and 77 percent COD removal. SS removal,
however, fell appreciably to 26 mg/1 and 83 percent removal TOC
removal was 68 percent, about the average of Phases 1-4.
PROCESS STABILIZATION (JULY 1-JULY 7, 1973)
Flow was increased; process disruption was caused by PSA
problems, and efforts were made to stabilize conditions.
PHASE 6 (JULY 8-AUGUST 11, 1973)
Encouraged by the results at 1.10 m3/sec (25 mgd), influent
25
-------
flow was increased to 1.31 m3/sec (30 mgd) with a diurnal varia-
tion from 0.92 to 1.58 m3/sec (21 to 36 mgd). Aerator detention
time, including return, decreased to 0.74 hr; the volumetric or-
ganic loading increased to 519 kg BOD/day/m3 (324 lb/day/1000
ft3); and the F/M loading increased to 1.53 kg BOD/day/kg MLVSS.
The final tank detention time decreased to 1.6 hr, and the over-
flow rate rose to 55 m3/day/m2 (1360 gpd/ft2). Removals were 90
percent for total BOD, 88 percent for filtrate BODf 86 percent
for SS, and 71 percent for TOG.
PHASE 7 (AUGUST 12-SEPTEMBER 1, 1973)
After 5 wk and no process deterioration at 1.31 m3/sec (30
mgd), influent flow was increased to 1.53 m3/sec (35 mgd) on
August 12, 1973, but with a modified diurnal pattern. Since the
peak flow of 1.84 m3/sec (42 mgd) that would have been required
exceeded the combined capacities of both influent pumps, the
diurnal variation was altered somewhat to obtain the average
flow of 1.53 m3/sec (35 mgd) by means of a sustained peak flow of
1.62 m3/ sec (37 mgd), which was the combined capacity of the two
pumps, and a minimum flow of 1.05 m3/sec (24 mgd). This latest
flow increase represented the greatest stress that could be ap-
plied to the system. If the biological process could not be
broken at this flow, the maximum tolerable loading would remain
unknown.
Things went very well until September 1 when a bearing on
the PSA compressor failed, shutting down the oxygen generator.
Therefore, this phase, which was intended to last for a month,
had to be cut short. Nevertheless, the 3 wk of operation proved
very successful. At an average flow of 1.55 m3/Sec (35.4 mgd),
effluent total BOD and SS concentrations were only 21 and 17
mg/1, respectively, for removal efficiencies of 89 and 87 per-
cent. COD and TOG removals increased to 80 and 75 percent, res-
pectively. Filtrate BOD removal fell to 83 percent, but the
influent strength was 15 percent lower too. These results were
obtained at an aerator detention time, including return sludge
flov/, of only 0.67 hr; a volumetric organic loading rate of
5.64 kg BOD/day/m3 (352 lb/day/1000 ft3); and an F/M loading of
2.22 kg BOD/day/kg MLVSS. Settling tank detention time was 1.3
hr, and the overflow rate averaged 66 m3/day/m2 (1610 gpd/ft2).
PROCESS RECOVERY (SEPTEMBER 2-15, 1973)
The bearing failure which shut down the PSA on September 1,
1973, reduced the total oxygen supply capacity to the 16.3 metric
tons/day (18 tons/day) of the Driox vaporizer. Then on September
3, the motor operating the automatic liquid oxygen feed valve
burned out, preventing automatic control of oxygen feed to main-
tain a predetermined gas space pressure. The reduced oxygen
supply capacity and the crude manual control on the liquid oxygen
feed required a cutback in influent flow from 1.53 to 0.53 m3/sec
26
-------
(35 to 12 mgd). A return to 1.53 m3/sec (35 mgd) was not immin-
ent, and so on September 7, a diurnal pattern designed to yield
an average flow of 0.44 rn3/sec (10 mgd) was implemented in pre-
paration for Phase 8, On September 10, the PSA was returned to
service once again allowing automatic control of oxygen feed.
PHASE 8 (SEPTEMBER 16-OCTOBER 8, 1973)
By September 16, 1973, the procass had reached equilibrium
after the precipitous flow reduction from 1,53 m3/sec (35 mgd)
to 0053 m3/Sec (12 mgd) and then to 0.44 m3/sec (10 mgd) and
Phase 8 was begun.
Since the data collected during the first summer of opera-
tion could not be used because of inaccurate instrumentation,
this phase was an attempt to backtrack to collect data to fill
the void at low-flow loadings. At an average flow of 0.44 rn3/
sec (10.1 mgd), with diurnal variations from 0.31 to 0.53 m3/Sc-c
(7 to 12 mgd), the aerator detention time, including return,
averaged 2.4 hr; the volumetric organic loading was 1.44 kg BOD/
day/m3 (90 lb/day/1000 ft3); and the F/M loading was 0.39 kg BOD/
day/kg MLVSS. The final tank detention time was 4.7 hr. The
final tank overflow rate and solids loading were only 19 m3/day/
m2 (460 gpd/ft2) and 112 kg/day/m2 (23 Ib/day/ft2), respectively.
At these low loadings and at the high sewage temperature of 72
F, removals were 89 percent of total BOD, 87 percent of SS, 92
percent of filtrate BOD, 82 percent of COD, and 77 percent of
TOG.
An out~of-line slude-collector snaft threatened to cause
serious damage to the equipment. Repa.Lrs were begun on October
9. The process suffered somewhat during the 24 hr required for
the repairs, and the flow had to be reduced. It was not until
October 14 that the flow was up to 0.66 m3/sec (15 mgd) and the
next phase could begin.
PHASE 9 (OCTOBER 14-25, 1973)
Biological frothing developed in the aerator before the
phase was wel] established and required its early termination.
The data, while representing only 12 days of operation, never-
theless are presented as a phase because they represent the last
period of operation at low flow and loadings.
Some of the data obtained during this period must be viewed
with extreme caution because of the inordinately high MLSS con-
centrations observed (average 9040 mg/1). The samples undoubt-
edly were contaminated by the aerator froth. The aerator deten-
tion time (sewage flow plus recycle sludge flow) was 1.6 hr and
the volumetric organic loading 1.75 kg BOD/day/m3 (109 lb/day/
1000 ft3). -The final tank detention time was 3.2 hr, and the
overflow rate was 27 m3/day/m2 (660 gpd/ft2). Removals averaged
27
-------
91 percent of total and filtrate BOD, 07 percent of SS, 82 per-
cent of COD, and 31 percent of TOC.
SHUTDOWN (OCTOBER 26, 1973-JANUARY 1, 1974)
By October 26, 1973, the foam had become so pronounced that,
in order to protect the stage blowers, the oxygen feed to the
system was halted and air feed via the stage compressors was in-
stituted. However, beginning on October 27, frequent stage com-
pressor shutdowns indicated that they had indeed been affected.
It was then necessary to dismantle and clean all the compressors
and their related oxygen piping. While this was being done, the
influent flow was reduced to 0.44 m3/sec (10 mgd) and the air
feed was continued in an effort to sustain the process until the
cleaning operation was completed.
The foam which plagued the operation was light brown and
frothy, similar to the "head" on a chocolate milkshake. Its
appearance was markedly different from the white foam which
results from aeration of a wastewater containing detergents, a
common sight at many treatment plants. Bench-scale dosage of
the foam with a commercial defearning agent was not effective.
A similar foaming condition at the Jamaica step aeration plant
a year earlier resulted from an infestation or Nocardia, and an
EPA specialist (6) confirmed the presence of these organisms in
the UNOX mixed liquor. The foam appeared only in the oxygen bay
and never was evident in the parallel air modules. A means of
removing the foam was not provided in this installation, but
some mechanism for removal appears desirable.
By November 12, the foam had subsided, and so, with the
compressors and their piping cleaned, oxygen feed was resumed
The flow was increased gradually from 0.44 m3/sec (10 mgd) until
it reached 0.88 m3/sec (20 mgd) c > November 25. Throughout the
episode, filamentous organisms we. e present, although not as
predominant forms.
Early in December, further problems with the return sludge
pump bearings became apparent and on the llth the pump was taken
out of service again. To prevent a repetition of the conse-
auences of the first shutdown, the tanks were drained and the set-
tling tank cleaned. Repairs to the pump included replacing the
upper and lower pump bearings, building up the impeller wearing
ring to reduce excessive clearance, and replacing the worn shaft
packing sleeve. Some preventive maintenance also was performed
on the sludge collection mechanism, and a piping modification
was made to the sludge wasting system. An inspection of the
tank interior revealed no grit buildup, no rag accumulation
around the mixers, and no apparent concrete deterioration.
28
-------
RESTART (JANUARY 2-FEBRUARY 5, 1974)
On January 2, 1974, after all repairs were completed, opera-
tion was resumed. By the 17th, the fungus had reappeared.
Nevertheless, the flow was very gradually increased and by Feb-
ruary 6 the process had stablized at 0.88-m3/sec (20 mgd) diurnal
flow.
PHASE 10 (FEBRUARY 6-APRIL 30, 1974)
The intention was to operate during what was left of the
1973-1974 winter at 0.88 m3/sec (20 mgd) and to combat the fungus
by any means except reducing the flow. From the beginning, the
phase was characterized by severe fungal proliferation; in fact,
the dominant microorganism in the mixed liquor was the fungus,
often to the complete exclusion of the microbial forms. The
fungus, with its long-branching mycelia, interfered with the
settling and compaction of the sludge. Consequently, as during
the previous winter, the sludge blanket extended the entire
length of the clarifier and the wasting rate frequently had to
be increased to control it. The poor compaction resulted in an
average return sludge concentration of only 5220 mg/1 for the
phase and a low for 1 wk of less than 4000 mg/1.
These were in contrast to averages of more than 16,000 rng/1
for summer operation and were less than half of those obtained
during the previous winter. The weak return sludge concentra-
tions depressed the MLSS concentration, which averaged only
2400 mg/1 for the phase and experienced a weekly low of less
than 1700 mg/1. From mid-February to mid-April, the sludge age
averaged only 1.0 day, considerably less than the 1.3-2.0-day
range experienced the previous winter. The SDI slipped to a low
of 0.4 g/100 ml during the week of February 10 and 0.6 during the
next'2 wk compared to a weekly low of 1.0 during the previous
winter and averages of over 2.0 during summer operations and well
over 1.0 the previous winter. Although it appeared desirable to
increase the sludge age to the level of the previous winter, the
low MLSS concentration dictated by the filaments and the self-
imposed constraint of maintaining the influent rate made that
impossible.
The fungus was not unique to the UNOX system. During this
phase and also in Phase 3, the organisms were present in the
air plant, too; however, visual observation indicated that their
numbers were fewer and their effects far less severe. The aver-
age SDI of 1.4 g/100 ml for the air plant during this period was
only slightly less than the average of 1.7 for the preceding sum-
mer. At a sludge age of 1.3 days, the air plant MLSS concentra-
tion averaged 1700 mg/1 and the return sludge concentration 8800
mg/1 (Tables 11 and 12).
29
-------
Microscopic examination of the plant influent revealed that
the fungus was entering in the raw sewage. It may have origin-
ated in the discharge of some connected industry, or conditions
in the interceptor system may have promoted its development.
Once in the plant, if conditions were favorable, retention and
concentration could have caused it to proliferate. It was ob-
vious that further work would have to be done.
PHASE 11 (MAY 1-AUGUST 5, 1974)
Although the EPA grant-funded program ended on April 30,
1974, operations and data collection were continued. Phases
11 and 12, therefore, although not parts of the grant program,
are included in this report to present the most recent findings.
This phase was a continuation of Phase 10 into warm weather
at 0.88-m^/sec (20 mgd) diurnal flow. Although the fungi dis-
appeared as the wastewater temperature rose to 60 F and above,
a healthy biomass did not reappear as in the previous summer and
removal efficiencies averaged only 84 percent for total BOD and
82 percent for SC.
The PSA was taken out of service on June 17 for replacement
of a failed compressor motor bearing, and the biological process
was entirely dependent on the liquid oxygen backup system until
August 1.
FLOW REDUCTION (AUGUST 6-AUGUST 25, 1974)
Flow was gradually reduced preparatory to the approaching
shutdown.
SHUTDOWN (AUGUST 26-OCTOBER 31, 1974)
This 9 wk shutdown was a planned operation, primarily to
upgrade the performance of the PSA. Due to a malfunction of
certain of the PSA switching valves early in the test, two of
the four adsorption beds became contaminated by moisture, re-
ducing the output of the unit to 9.1 metric torn; of oxygen/day
(10 tons/day). Union Carbide could have rehabilitated the beds
at any time, but instead elected to supply liquid oxygen at its
own cost up to the PSA design rate of 13.6 metric tons oxygen/
day (15 tons/day).
The upgrading consisted of replacing the adsorbent material
in the two beds and portions of the material in the other two,
replacing all valves and rebuilding all valve activators, and
increasing the capacity of the PSA compressor and some piping.
The result of the refurbishing was that although an output of
13.6 metric tons oxygen/day (15 tons/day) was guaranteed, the
unit in fact thereafter delivered 15.2 metric tons oxygen/day
(16.7 tons/day) with virtually no downtime.
30
-------
In addition to the work on the PSA, inspection and preven-
tive maintenance were provided for all other oxygen system
equipment during this shutdown period.
RESTART (NOVEMBER 1-DECEMBER 12, 1974)
Because of the unanswered questions relating to the fungal
proliferation during winter operations, the City and Union
Carbide embarked upon an extensive investigation which lasted
throughout the 1974-1975 winter. The aims of the program were
to identify the fungus, determine whether it was peculiar to
Newtown Creek or was ubiquitous in New York City wastewater,
and find some means of eliminating it. To accomplish these ob-
jectives, Union Carbide provided a 1.89 I/sec (30-gprn) oxygen
pilot plant on which to test the effects of possible fungicides
before they were applied to the 0.88-m3/sec (20-mgd) system and
a consultant was engaged to identify the fungus and recommend
steps for its control. A second Union Carbide oxygen pilot plant
was stationed at the Wards Island step aeration air plant to
compare air and oxygen performances in cold weather at that
location.
The Newtown Creek oxygen module was restarted at 0.22~ro^/
sec (5-mgd) constant flow. Influent flow increased to 0.88-
mVsec (20-mgd) diurnal by December 12, 1974.
PHASE 12 (DECEMBER 13, 1974-MARCH 11, 1975)
The fungi persisted throughout this period, but their ef-
fect on the process was insignificant. Nevertheless, the fungus
control program was begun. It was soon found that, although
several species of fungi were present in the mixed liquor of
both the air and oxygen systems, the predominant form was the
arthrospore producer "Geotrichum" (10).
Sampling of the influents at three other New York City
plants and one drainage area not yet served by a plant revealed
the presence of the same organism in concentrations generally
similar to those at Newtown Creek. At Newtown Creek, the fungus
concentration increased moderately in the air plant mixed liquor
but significantly in the mixed liquors of both the full-scale
and pilot oxygen systems. At the three other plants sampled,
the fungi did not significantly proliferate in the mixed liquor.
The growth in the V7ards Island oxygen plant was about the same
as in the full-scale air piant. Apparently, Newtown Creek's
waste is unique in that it is conducive to the growth of these
organisms.
The pilot plant employed at Newtown Creek was a Union Car-
bide four-stage oxygen system equipped with turbine/spargers for
oxygen dissolution and an external circular clarifier. The
pilot plant treated an average of 1.1 I/sec (17.5 gpm) on a
31
-------
diurnally-varying pattern approximating those of the field oxy-
gen module and f {Ill-scale air plant. The f irst .^tempt at com-
batting the fungus was to operate at a high sludge retention
time (SRT), but this was not successful. An SRI of 4.3 days
maTnt lined 'for 2 wk did not diminish the fungi but al BO did not
harm the process, whose effluent averaged 11 mg/1 of total BOD
and 13 mg/1 of S3.
Since the phosphorus content of the waste during Phase 10
was onlv 2 4 mg/1, the lowest up to that time, it was considered
possible that nutrient deficiency might have slowed the growth
of Se normal biomass population and permitted the fungus to be-
duin which
of e norma o
come predominant. Therefore, after a short P«iod during which
the 4 3-day SRT was reduced to a normal level, 2-3 mg/l ot
phosphoruswas aOded to the pilot plant influent. However,
2 wk of phosphorus supplementation had no effect The ef-
fluent during this period averaged 17 mg/1 of total BOD and
15 mg/1 of SS.
Two wk of chlorine addition followed at a dosage of 9 mg/1.
Not only did this have no effect on the fungus, but it ^jured
the process, whose effluent deteriorated to 29 mg/1 of total
BOD and 32 mg/1 of SS.
A third additive, hydrogen peroxide, was assessed using jar
tests but failed to improve sludge settling even at dosages up
to 200 mg/1.
The final attempt at control was adjustment of the PH of the
mixed liquor, although it was expected to be very costly on a
full -plant scale. More than 2 yr of data indicated that the pH
of the influent, which ranged from 6 . 3 to 6 . 9 , was depressed by
about 0.4 unit during passage through the oxygenation aerator
£ was considered possible that this drop could have sparked the
growth of the fungus. However, 10 days of lime addition at 100
mg/T increased the pH of the mixed liquor to 7.0 but had no ef-
fect on the fungi.
The 1974-1975 winter operation in the 0.88-m3/sec (20-mgd)
oxygen system during Phase 12 differed from the past winters in
?wo respects. Firs?, it was not interrupted by a breakdown of
the return sludge pump, nor for any other reason. Second, al-
though the fungus was present from beginning to end its concen-
tration in the mixed liquor appeared to be less than those of
the first two winters. Effluent quality averaged 18 mg/1 of
total BOD and 13 mg/1 of SS for removals of 87 percent The
average sludge age was 1.6 days, and the SDI was 1.5 g/100 ml.
The reduced quantity of the fungus in the mixed liquor dur-
ing this phase may explain the improved overall process perform-
ance of the thirdwinter, but the reasons for the lower fungal
concentration are purely speculative. One possibility is that
32
-------
after the shutdowns and restarts of the first two winters, the
rapid growth of the fungus aborted the development of a proper
activated sludge culture, but. Phase 12 with no shutdown and
cold-weather restart began with a population that was able to
contain the fungi.
33
-------
TABLE 2. OXYGEN SYSTEM PROCESS PERFORMANCE
.,,-.-.._, . •• —
I'h-se
ll-i d'*; f
1234
19/17- 12/10/72- 02/25- 04/00- (
5
6
7
8
)9/16- 1
9
0/14- 0
10
•S/^M/74 0
11
12
2/13/74-
I3/11/7S
11/25/72 02/01/73 04/05/73 05/31/73 00/3U//i UB/I1//J uv/ui//o iu/uo/u iw--/-- _-y -..,.. --, - .
How 3
Sewage, m /day
Ueturn Sludge', m /day
(mgd)
". Sludge Return
B011
Influent (mg/1)
Effluent (mg/1)
°u Removal
filtrate BOD
Influent (mg/1)
Effluent (mg/1)
* Removal
COD
Influent (mg/1)
Effluent (mg/1)
°-a Removal
Suspended Solids
Influent (nig/l)
1 Volatile
Effluent (mg/1)
"» Volati le
°u Remova 1
TOC
Influent [mg/1)
Effluent (mg/1)
t* llemoval
78,700
(20.8)
23,800
(6.3)
30
156
9
94
84
4
95
356
61
83
79
12
76
90
21
77
67,000
(17.7)
26,900
(7.1)
40
157
21
87
78
13
83
365
88
76
146
77
22
87
99
35
65
57,200
(15.1)
28,800
(7.6J
50
151
17
89
88
12
86
364
75
79
145
74
18
89
99
30
70
76,800
(20.3)
35.200
(9.3)
46
168
17
90
96
11
89
365
77
79
159
76
18
83
109
54
50
96.900
(25.6)
41,600
(11.0)
43
.740
24
90
130
14
89
320
7i
77
149
75
26
77
104
33
68
113,600
(30.0)
38,600
(10.2)
34
215
22
90
98
88
290
64
78
119
75
17
71
96
71
134,000
(35.4)
34,100
(9.0)
25
198
21
S9
82
14
83
308
62
81)
131
83
17
82
lOh
75
3R.200
(10.1)
9,500
(2.5)
25
178
19
89
•Jft
3
112
363
65
82
136
84
18
83
87
103
24
77
55.300
(14.6)
14,400
(3.8)
26
149
:3
91
70
6
91
314
57
S2
135
SO
17
76
87
110
21
8!
74.600
(19.7)
43,100
(11. -I)
58
136
26
81
77
16
79
50'
87
72
126
72
27
70
79
104
.12
60
72,300
(19.1)
23,400
(7.5)
39
14ft
24
84
SO
13
S4
-
142
79
26
82
-
75.700
(20.0)
21 .600
(5.7)
29
140
IS
S7
87
It)
S3
2S9
70
77
101
81
13
S2
S7
-
•- — ' ...,.,
* Data for Phases 1-11 may
be in error because of air entrapment in return sludge magnetic flow meter.
-------
TABLE 3. OXYGEN SYSTEM AERATION TANK PERFORMANCE_
Phase
Dates
MUSS (mg/U
". Vo kit i 1 e
Sludge Age* (days)
Sludge Retention
Time*" (days)
Detention Time (hr)
(Q« !()***
0
OJ '
Ul
HOD Loading, kg/day/m,
(Ib/day/ I 000 ft )
roll Loading, kg/day/m,
(Ib/day/ 1000 ft )
H/M Loading
(kg HOD/day/kg M1.VSS1
K/M Loading
(kfi Crn/day/kg MI.VSSI
- -.
IrtDbC O. U A l vj i. i\ .j
1 2 ^
4860 4980 '1010
1.1 1-3 1-3
1.4 1.7 2.0
2 61 2.24 1 .84
(163) (MO) (115)
S Or. 5.21 4.42
(372) (3251 (276)
0.63 0.55 0.57
1.45 1.28 1.38
~. —
4 S 6
/06- 06/01- 07/08- 08y
3Q50 4530 4150
79 79 8?
1 5 1.5 1-4
14 1.3 1.2
1.0 0.8 0.7
1.5 1.2 1-0
1 74 4.95 5.19
(171) (309) (324)
5.9b O.bO 7.00
(372) (412) (437)
0.88 1.38 1.53
1.92 1.84 2.07
— —
7 8
mi/73 10/08/73 1
3130 4760
81 79
0.8 4.3
0.8 5.5
0.7 2.4
0.8 3.0
5.64 1.44
(352) (90)
8.78 2.95
(548) (184)
2.22 0.39
3.45 0.79
9 10
0/14- 02/06- OS.
0/25/73 04/30/74 OS
(9040)* 2400
80 78
1.2
0.9
1.6 1.0
2.0 1.5
1.75 2.)b
(109) (135)
3.6S 4.87
(250) (304)
I. IS
2.61
/Ol- 12/13/74-
/OS/74 03/1 1/7S
3870 2630
S4
1.8 1-6
1.2
1.1 1.2
1.6 1.5
2.24 2.2t>
(JJO) (141)
4.79
(2931
1.02
2. IS
Samples prohalily contaminated hy foam.
Oil 111 1 IL'S li l uuiH' * / "-"-•' • «• *••" • ••— •' 'i-i IA
' Defined as kg M1.SS ill reactor/kg SS in reactor int luent/day. ff]uenl/aay
-------
TABLE 4. OXYGEN SYSTEM FINAL TANK PERFORMANCE AND SLUDGE SETTLING CHARACTERISTICS
to
en
Phase
Dates
FINAL TANK PERFORMANCE
Return Sludge SS
(mg/1)
"o Volati le
Return Sludge COD
(mg/1)
Detention Time, Q (hr)
Overflow Rate, m'/dav/iu^
(gpd/ft")
Solids Loading, kg/day/m"
(lb/day/ft )
Weir Loading, m /day/m
(gpd/ lineal ft)
SI.IJUGI- SETTLING
CIIARACTI-KISTICS
Sludge Hensity Index
Unstirred (g/100 ml)
Stirred at 1 rpra (g/100 ml)
Initial Settling Rate (1SR)
Unstirred (m/hr)
Stirred at 1 rpm (m/hr)
Wastrwarer Temperature (C)
Amtiient Temperature (C)
1
09/17-
11/25/72
16,200
83
20,000
2.3
39
(950)
2-14
(50)
1,615
(130,000)
2.2
-
3.2
3.4
19
14
2
12/10/72-
02/01/73
13,000
81
17,900
2.7
33
(810)
230
(47)
1,374
(1 10,600)
1.8
-
2.6
2.7
14
3
3
02/25-
04/05/73
11,500
80
15,400
3.1
28
(690)
171
(35)
1,172
(91,400)
1.2
2.2
1.6
3.0
14
9
4
04/06-
05/31/73
13,200
80
17,000
2.3
37
(920)
220
(45)
1,576
(126,900)
1.9
2.7
3.8
4.1
67
17
5
06/01-
06/30/73
16,300
79
18,800
1.9
47
(1160)
308
(63)
1,987
(160,000)
2.3
2.9
4.3
4.3
21
25
6
07/08-
08/11/73
lb,200
80
22,r-00
1.6
55
(1360)
313
(64)
2,329
(187,500)
2.3
2.8
4.4
4.6
24
28
7
08/12-
09/01/73
13.000
81
16,200
1.3
66
(1610)
259
(53)
2,749
(221,300)
2.1
2.5
5.5
5.5
24
28
8
09/16-
10/03/73
17,300
79
21,400
4.7
19
(460)
112
(23)
784
(63,100)
2.7
3.0
5.0
5.0
tt
"
9
10/14-
10/25/73
18,600
80
22.500
3.2
,,
(660)
1 . 1 35
(91,400)
2 0
2.S
2.0
2.1
21
IS
10
02/06-
CW/30/74
5,200
78
6.700
2.4
37
(900)
137
(28)
1,529
(123.100)
1.2
1.6
S.4
•1.0
12
S
11 12
05/01- 12/13/74-
08/05/74 03/1 1/7S
12,700 9,400
77 SI
13,200
2.S 2.4
35 37
(S70) (910)
190 I"1"
(39) (26)
1,483 1.SS3
(119.400) (125,000)
I."* IS
-
21 13
-------
TABLE 5. OXYGEN SUPPLIED TO OXYGEN SYSTEM
Phase
Dates
kg/ra sewage
(tons/mil gal sewage)
kg/kg BOD removed
kg/ kg COD removed
kg/kg COD destroyed
Phase
Dates
kWh/m sewage
(hp-hr/mil gal sewage)
kWh/kg BOD removed
(hp-lir/lb BOD removed)
kWh/kg 02 supplied
(hp-hr/lb 02 supplied)
1
09/17-
11/2S/72
0.16
CO. 67)
1.1
0.6
-
TABLE 6
i
09/17-
11/25/72
0.22
(1117)
1.54
(0.94)
1.37
(0.83)
2
12/10/72-
02/01/73
0.16
(0.67)
1.2
0.6
-
3
02/25-
04/05/73
0.19
(0.79)
1.4
0.7
1.9
4
04/06-
05/31/73
0.16
(0.66)
1.0
0.6
2.0
5
06/01-
06/30/73
0.17
(0.70)
0.8
0.7
2.1
6
07/08-
08/11/73
0.16
(0.68)
0.8
0.7
3.5
7
08/12-
09/01/73
0.15
(0.62)
0.8
0.6
1.7
8
09/16-
10/08/73
0.24
(1.02)
1.5
0.8
1.3
9
10/14-
10/25/73
0.23
(0.93)
1.6
0.9
2.0
10
02/06-
04/30/74
0.13
(0.53)
1.2
0.6
3.5
11 12
05/01- 12/13/74-
08/05/74 03/11/75
0.16 0.17
(0.65) (0.72)
1.3 1.4
0.8
3.1
POWER FOR OXYGEN GENERATION AND DISSOLUTION
2
12/10/72-
02/01/73
0.27
(1371)
1.96
(1.19)
1.63
(0.99)
3
02/25-
04/05/73
0.33
(1676)
2.49
(1.52)
1.74
(1.06)
4
04/06-
05/31/73
0.24
(1218)
1.59
(0.97)
1.54
(0.94)
5
06/01-
06/30/73
0.20*
(1015)
1.01*
(0.62)
1.28*
(0.78)
6
07/08-
08/11/73
0.17*
(863)
0.95*
* (0.5S)
1.17*
* (0.71)
7
08/12-
09/01/73
0.15*
(761)*
0.90*
(O.SS)*
1.10*
* (0.67)*
8
09/16-
10/03/73
0.48
(2436)
2. 95
(l.SO)
1.89
(1.15)
9
10/14-
10/25/73
0.33
(1676)
2.42
(1.48)
1.48
(0.90)
10
02/06-
04/30/74
0.26
(1320)
2.38
(1.45)
1.89
(1.15)
11 12
05/01- 12/13/74-
08/05/74**03/11/7S
0.24
<_I21S)
1.S4
(I. IS)
1.39
(O.S4)
* Includes 0.37 kWh/kg (451 hp-hr/ton) additional power to generate purchased cryogenic oxygen above 13.600 kg/day US tons/day) PSA design capacity.
** 335-kW (450-Kp} PSA air compressor motor out of service for bearing repair from June 17 to August 1, 1974.
-------
TABLE 7. OXYGEN SYSTEM SLUDGE PRODUCTION
.,.-
riinso
UateS
1 2
3
09/17- 12/10/72- 02/25-
11/25/72 02/01/73 04/05/73
IVasTcd, i" /day
(gal/day)
Total SS, dry kg
(dry Ib)
Wasted/day*
In effluent/Jay
Total produced/day
Wasted, kg/kg
Rflll removed
Wasted, Kg/m' sewage
(Hi/mil gal sewage)
U) Produced, kg/kg
CO R01) removed
Produced, kg/m' sewage
(Ib/mil gal sewage)
Volatile SS, dry kg
(dry Ib)
Wasted/day*
In effluent/day
Total produced/day
has toil, kg/kg
RO[| removed
Wasted, kg/m' sewrifje
(Hi/mil gal •scv.agi-)
Produced, kg/kf.
HOP removed
Produced, kg/in' sewage
(Hi/mil gnl sewage)
(187
8
(17
1
(2
9
(20
-
0
(1
-
n
(i
6
(14
-
(2
7
(16
-
710
,000)
,100
,900)
,000
,300)
,1(10
.200)
1.0ft
.143
,190)
1.20
.161
,340)
,500
, 300)
900
,000)
,400
,3110)
O.R5
(1.11.1
-
-
0
(1
(950)
0 . 96
.ISO
,080)
4
04/Ob-
05/31/73
900
(239,000)
11,900
(26,300)
1,400
(3,000)
13,300
(29,300)
1.03
0.156
(1,300)
1.14
0.173
(1,440)
9,500
(21,000)
1,100
(2,500)
10,600
(23,500)
O.S2
0. 1J4
(1 .030)
0.92
0.139
(1 .It-Ill
5
06/01-
06/30/73
860
(228,000)
14,100
(51,100)
2,500
(5,600)
16,600
(36,700)
0.67
0.145
(1,210)
O.RO
0.172
0,450)
11 ,10(1
(24,500)
2 , 000
(4,30(1)
1 3 , 1 00
(2R.SOO)
0.53
0.115
(960)
0.62
0 . 1 36
(1.1 50 i
6
7/08-
08/11/73
890
(234,000)
14,300
(31 ,(,(10)
2,000
(4,300)
16,300
(35,900)
0.65
0.126
(1,050)
0.74
0.14.)
(1,200)
11,500
(25,300)
1 ,400
(3,000)
12,"00
(28,3001
0.52
0.101
(840)
0.59
0.113
(940)
7
08/12-
09/01/73
1.300
(343.000)
16,900
(57,30(0
2,300
0,000)
1«,200
(42,300)
0.71
0.126
(1 ,05(1)
0.81
0.143
(1,190)
13,700
(30,200)
1,900
(4,100)
15,600
(34.3001
0.5R
0.102
(850)
0.66
0.116
(970)
8
09/16-
10/08/73
190
(51,000)
3,400
(7,4(10)
700
(1,500)
4 , 1 00
(8,900)
O.SS
O.ORR
(730)
0.66
0. 106
(880)
2,600
(5,800)
600
(1,500)
3,200
(7,100)
0.43
0.1168
(570)
0.53
0.08'i
(700)
g
10/14-
10/25/73
350
(9.3,000)
6,500
(Il.JOO)
1 .000
(2,100)
7,500
(16.JOO)
0.87
0.119
(900)
0.99
0.136
(1.130)
5.200
(11,500)
7PO
(1,6(10)
5,900
(I-.IOO)
0.60
0.095
(790)
0.79
0.10S
(900)
10
02/06-
04/30/74
2,00(1
(529.000)
10,400
(23,0001
2,
-------
OXYGEN SYSTEM NUTRIENT REMOVALS
- _ . . — ,
Dates
Total I'liosjihatp
(mi:/ 1 as P)
Influent
l-fflucnt
", Itentova 1
Soluble Orthophosphate
(mn/l as P)
Influent
Kfflucnt
",. Removal
Ammonia NitroRen
(IIIR/I as N)
Influent
r.ffluent
TKN
dnR/1 as N)
Influent
P.ff luent
Nitrite Ni frozen
[mil/1 as N)
Influent
lU'Cluent
Ni l rate NitroRen
(mi;/ I as N)
Influent
lit fluent
1 2 3
09/17- 12/10/72- 02/25-
11/25/72 02/01/73 04/05/73
~"
4.3 3,9 3.6
<.C> 2.6 1.5
40 33 5R
TC •* *> 17
1.7 1.4 1.1
32 36 35
9.3 9.0 9.4
8.0 9.0 9.7
22.0 22.4 22.0
1 .! . 4 1 4 . 4 1 - 05/01- 12/U/74-
10/25/73 04/30/74 08/05/74 03/11/75
4.1 2.4 - 2.1
2.3 0.7 - n-6
44 71 - 71
3.0 1-6 - !••»
1.8 0.3 - "-->
40 SI - 79
11.3 10.6 - 10-*
8.8 R.5 - 9.0
21.1 20.7 - 1S..1
12.0 13.1 - 12.6
_
o.i- O.H
0.11 0.11
1.74 1.97
1.67 1.61
-------
TAELE 9. OTHER OXYGEN SYSTEM SEWAGE CHARACTERISTICS
j i 3 4 5 6 7 8 9 10 U 12
09/17- W10/72- 02/25- OJ/06- 06/01- 07/08- 08/12- 09/16- 10/14- 02/06- OS/01- 12/13/74-
11/^5/72 OV01/73 04/05/73 05/31/73 06/30/73 08/11/73 09/01/73 10/08/73 10/25/75 04/30/74 OS/OS/74 03/11/75
Alkulinity
lmg/1 as CaOXJ
influent «« 71 76 80 77 71 76 68 SO 76
Effluent 95 83 94 85 78 79 7, 74 74 /8
Dissolved Solids (.mg/1)
Influent U56 1003 1038 992 1544 ------
liffluiM.t 1087 847 991 976 1 2• 8 <'•' "-7 b-6 6'4 6'5 fS *'o I'* t*
>;...,, i 6.2 6.3 6.2 5.9 6.1 5.9 O.O 6.- to..
"fluent - 6.6 6.6 6.6 6.6 6.4 6.6 6.5 "-3 6.5 6.6
-------
TABLE 10. NEWTOWN CREhK AIR PLANT PROCESS PERFORMANCE
Phase
Lljtes
I'low (Per 3;iv)
Su«;ige (mud)*
lieturn Sludge (ingj)*
"a Sludge iiettirt)
UOI)
Influent (inc/M
l;f f 1 ueilt 1 nit;/ 1 )
•I keitiova 1
l-i It rjte Hill)
Influent (mjj/1)
ft' fluent, lnm/l)
'ii Keinox';il
Suspended Sol ids
Influent (nn;/ll
-. Vn 1 .1 1 i 1 e
1. ('fluent I ing/ 11
'u Ueinovj 1
1 2 3
09/17- 12/10/72- 02/25-
11/25/72 02/01/73 04/05/73
]0.6 10.4 11.1
2.0 2.1 2.2
19 21) 20
156 !57 151
24 33 41
85 79 73
84 7K 83
111 27 27
88 tv. <>9
1-19 14() MS
79 7^ 74
4(1 47 .10
73 t>8 "2
4 5 6 7 8 9 JO 11 12
04/06- 06/01- 07/08- 08/12- 09/16- 10/14- 02/06- 05/01- 12/13/74-
05/31/73 06/30/73 Oi/'l/73 09/01/73 10/08/73 10/25/73 04/30/74 OS/05/74 03/11/75
10.4 12.0 12.0 12.1 12.4 i?.0 9.7 11.1 11. S
2.4 2.4 2.4 2.4 2.4 2.3 2.2 2.1 1.9
20 20 l!) 20 19 19 22 IS 17
168 240 215 19S 178 14S 1 31) l-'fa 'JU
37 46 35 41 30 24 23 32 2'->
78 81 84 79 S3 84 ST. 7S 79
159 14'.; 119 131 13o 13S 120 142 H SI
45 50 45 41 49 4b 53 .is
7: „„ hj 09 o4 (-h 72 .^ 56.
-------
TABLE 11. NEWTOWN CREEK AIR PLANT AERATION TANK PERFORMANCE
Phase
totes
MI.SS '.me/')
ShiJgu Age" (days)
! 2 3 4 S 6 7 8 9 1C 11 12
09/17- 12/10/72- OV75- 04/06- 06/01- 07/08- 08/12- 09/16- 10/14- 02/06- 05/01- 12/13/74-
11/25/72 02/01/^3 04/05/73 05/31/73 06/30/73 08/11/73 09/01/73 10/08/73 10/25/73 04/30/74 08/05/74 03/11/75
2(IO°
1.5
jciOO
1.5
1800
1800
1.3
1800
1.2
1700
1.4
1900
1.4
1900
1.3
1800
1.3
1700
1700
1.2
1600
1.6
. tent '.on 'rime (hr)
Q » I?
Q
2.27
2.71
^ 30
V?6
-> 17
2.(>0
i 3(1
2.77
2.00
2.40
2.03
2.43
1.98
2.37
2.41
2.96
2.13
2.53
2.15
2.50
HUH Uncling ,
(lb/ Jay/ 1000 ft )«
88
93
• Defined as kg MI.SS in reactor/kg SS in reactor influent/day.
" 1 Ib/day/I»00 ft = 0.016 kg/Jay/n .
154
138
118
96
71
T'XBLE \2. NEWTOWN GREEK AIR PLANT FINAL TANK PE RFORNHNfT
A N D S L I) n G. E S E T T L I N G G11A R A G T E R I S T I G S
Phase
Dates
Keturn Sludge SS
(ma/1)
". Volatile
Detention Time, Q Ihrl
Overflow Kate IgpJ/ft")
So 1 ids Loading,
(Ib/day/ft")'
We i r Load in^
(g(>J/ lineal ft)**
Sludge Density Index
Unstirred (jj/100 ml )
-.
_ _ .. . , ! — — '
, > 3 4 S 6 7 S 9 10 11 12
09/17- 12/10/72- 02/25- 04/06- 06/01- 07/08- 08/12- 09/16- 10/14- 02/06- OS/01- 12/13/74-
11/25/72 02/01/73 04/05/73 05/31/73 06/30/73 08/11/73 09/01/73 10/08/73 10/25/73 04/30/74 OS/05/74 03/11/75
,, sun 10500 10300 10 ooo 10.300 in.&on 11.000 11.400 10,200 s.soo 10,200 9,300
"8I, ' 87 ' ."II 80 78 79 80 79 83 79 78 79
4. .11! 4.50 4.22 4.50 3.91 3.93 3.86 3.77 3.8S 4.82 J.ll 4.117
.,.,,1 4SO 510 48" 550 550 560 570 ShO 450 520 550
,j.(, .j.ci 9.1 8.5 9.8 9.3 10.4 10.7 9.S 7.7 S.7 S.I
66,300 1-5,001) 69,400 05,000 75,000 75.000 75,600 77. ..00 75,0(=0 60,600 71.300 71.90!!
1.5 1.4 1... 1.4 1.6 1.8 1.7 1.5 1.3 1.4 1.4 1.6
....... _ .. __ . " — •••-'•
* I upJ/ft" = 0.041/m /Jay/itT
" 1 Ib/day/l't" = 4.88 kg/day/m*
•• I j!p
-------
TABLE 13. POWER CONSUMPTION FOR AERATION AT NEWTOWN CREEK AIR PLANT
i,7~ ~ i 2 ~~~34 5 6 7 8 9 10 11 12
I,.,' 09/17- 12/10/72- 02/25- 04/06- 06/01- 07/08- 08/12- 09/16- 10/14- 02/06- 05/01- 12/13/74-
11/25/72 02/01/73 04/05/73 05/31/73 06/30/73 08/11/73 09/01/73 10/08/75 10/25/73 04/30/74__Og/05/_7j_03/n,/7S___
' 85i 8fl5 87, 898 791 644
hp-hr/mil sal sewage" 898 Sal Bil
0.94 0.83 0.48 0.55 0.64 0.07 0.83 0.97 0.83 0.71
hp-hr/lh BOO removed**
1 hp-hr/mil sal = 0.197 kWh/HlOO in3
1 hp-hr/lli = i.M kWh/kg
TABLE 14. NEWTOWN CREEK AIR PLANT SLUDGE PRODUCTION
Tota I iirodtK-i-d/day/bay
W;isted/lh DOU removed
Wasted/mil jjal sew.-ijse*'
I'roduced/ Ih lull, removed
I'roJu-'e.l/mi 1 gal si-wage"*
3
" w/rT/73
16,500
20,200
1 . t>2
1 . .UK)
1.98
1 ,SJO
_ . .
4
OS/31/73
15,400
1 9 , 300
1.35
1 ,-lSll
1.70
1.850
-. .—
S
06/30/73 Of
14.400
19,400
0.74
1 , -Mill
1 . OU
1 ,620
:
6 7
i/11/73 09/01/73 1
12,600 12,100
4 500 4 100
17,101) 16,200
0.71. 0.76
1 ,1150 1 ,001)
0.95 1.03
1,45(1 1,340
„ • -
8
0/08/73 10
1 3 , 300
5,100
18,400
0.87
1 ,070
1.20
1.480
9
/2S/73
1 2 . 300
4 .600
16.900
0.99
1 ,1.30
1.55
1,410
02/06- 05;
04/30/74 OSj
12,900
2,800
15.700
1 .41
1,331.
J.72
1 ,620
11 12
'01- 12/13/7--
'05/74 03/11/75
14.200 11,200
S.faP:1 4. OHO
17,800 15.200
1.31 1-05
1.240 9~0
1 ,c>4 I ..13
! , 560 ! , 32(1
1 Ih - O.-IS kg
1 Hi/mil sal * 0.12 kg/1000 m*
-------
TABLE 15. POWER CONSUMPTION FOR AERATION AT JAMAICA STEP AERATION AIR PLANT
Month
Flow (mgd)*
BOD Removed in
Secondary System (mg/1)
hp-hr/mil gal sewage*
hp-hr/lb BOD removed**
03/72
95
82
619
0.90
04/72
95
83
577
0.83
05/72
93
79
619
0.90
06/72
97
70
577
0.97
07/72
89
61
619
1.19
08/72
91
72
577
0.97
09/72
92
98
619
0.74
10/72
92
66
644
1.19
11/72
99
80
619
0.94
12/72
94
92
SSO
0.74
01/73
92
99
619
0.74
02/73
88
97
619
0.74
1 mgd = 0.044 m3/sec = 3785 m3/day
1 hp-hr/mil gal = 0.197 kWh/1000 m3
1 hp-hr/lb = 1.64 kKh/kj
TABLE 16. POWER CONSUMPTION FOR AERATION AT 26th WARD STEP AERATION AIR PLANT
Month
Flow (mgd)*
BOD Removed in
Secondary System (mg/1)
hp-hr/mil gal sewage*
hp-hr/lb BOD removed**
06/75
79
86
349
0.48
07/75
88
75
295
0.47
08/75
90
43
215
0.60
09/75
89
52
201
0.47
10/75
82
54
215
0.47
11/75
82
66
241
0.43
12/75
78
58
255
0.54
01/76
85
48
228
0.56
02/76
81
54
228
0.51
03/76
80
S3
201
0.47
04/76
81
61
241
0.47
OS/ 76
87
47
2S5
0.64
06/76
8*
48
21S
O.S2
07/76
85
47
228
O.S6
08/76
87
•SO
201
0.59
09/76
SI
56
201
0.43
10/76
S3
47
201
O.S1
1 mgd = 0.044 m/sec = 3785 m/day
* 1 hp-hr/mil gal • 0.197 kWh/1000 m3
** 1 hp-hr/lb =1.64 kWh/kg
-------
TABLE 17. SLUDGE PRODUCTION AT JAMAICA STEP AERATION AIR PLANT
U1
Month
Row (mgd)*
BOD Removed (mg/1)
(Pri. * Sec.)
Effluent SS (mg/1)
MLSS (mg/1)
Sludge Age'* ((Jays)
Total SS (dry lb)*
Wasted/day (x 1000)
In effluent/day (x 1000)
Total produced/day
(x 1000)
Wasted/ lb
BOD removed
Wasted/mil gal sewage
Produced/ lb
BOD removed
I'nnluccd/rai 1 gal sewage
03/72
95
124
16
2900
4.8
118.6
12.7
131.3
1.21
1250
1.34
1380
04/72
95
122
17
2600
4.9
103.5
13.5
117.0
1 .07
1100
1.21
1230
05/72
93
120
15
2250
3.y
104.4
11.6
116.0
1.12
1120
1.25
125U
06/72
97
115
19
1800
3.1
98.8
15.4
114.2
1.06
1020
1.23
1180
07/72
89
103
22
1750
4.7
81.4
16.3
97.8
1.07
920
1.28
1100
08/72
91
115
22
3000
5.9
74.9
16.7
91.6
0.86
820
1.05
1010
09/72
92
109
29
2100
2.2
69.1
22.3
91.4
0.83
750
1.09
990
10/72
92
112
26
1700
3.8
65.0
20.0
85.0
0.76
710
0.99
920
11/72
99
102
26
2000
3.5
85.0
21. S
106. S
1. 01
860
1.26
1020
12/72
94
121
34
2400
3.7
78. *
26.7
105.3
0.83
S30
1.11
1120
01/73
92
125
22
1800
3.2
91.7
16.9
108.6
0.96
1000
1.13
11SO
02/73
88
138
25
1600
2.3
77.7
18.4
96.0
0.77
sso
0.9S
1090
* I mgd = 0.044 inVsec -- 3785 m /day
*• He fined as kg Ml.SS in reactor/kg SS in reactor influent/day.
' 1 lb = 0.454 kg
" 1 Hi/mil gal = 0.12 kg/1000 m3
-------
TABLE 18. SLUDGE PRODUCTION AT 26th WARD STEP
MI.SS (mg/l)
., r. n(,/7s u///b «»//3" 09/75 10/75 11/75 12/75 01/76 02/76 O.V76 (14/76 05/76 06/76 07/76
Motirn ' „„_ j.-^... -i--.--. • —— —"••"— * ~ ' — '
|:~o"w"'hlig~d)~™"«8 ^)0 M 82 *2 ?R 8S R' R" S' ^ M RS S? ^ "5
BOP RemoveiUmg/1) ^ ^ 5p 5f) ^ M ^ 6S „ M M ss S8 (-:
Km-cm ss (»g/i) 21 >o ^ -" lft 1S I6 22 1S 1? " 2' 1S '7
2700 2100 2200 2300 2300 1800 1900 25(10 230O 2100 2000 1900 1900 1700 19"C> 170n 2SOO
.. ,, , , j , -, 3 1 •> Q , 8 -1.0 3.9 5.3 5.1 5.0 3.7 S.R 3.6 3.5 4.6 3.S 4.7
Sludge Age*' (days) 3-'' -1-' J-' - •s
13.8 11.7 15.0 14.8 10.9 12.3 10.4 15.6 12.2 11.S 9.5 15.2 9.1 12.1 S.7 TO.! 13.S
Total produccd/dny „ ,000, 69.3 67.3 75.8 ,6.5 64.4 54.4 ,9.3 55.2 66., 64.8 69.4 70.3 6R.0 64.3 5S.9 «..' ,S.l
WastcU/lb BOO removed 0.92 1.20 1.37
Wasted/mil pal sewage
Produced/Ih ^ q_ \ n\ 1 ^0 1 >4 1 S9 1 .-19 1.67 1.67
BOH removed 1.14 1.46 1./1 1.-^_^^_^__n^
1.24 1.26 0.73 0.80 0.86 1.09 1.31 1.29 1.31 l.-tS 1.10 1.06 0.99 1.2S
0.70 0.63 0.68 0.58 0.65 0.51 0.50 0.47 0.67 0.67 0.74 O.,o 0.70 0.6, O.S8 O.M 0.65
* 1 ingil = 0.014 m'Vscc = 3785 m' /day
*• Defined as kg MISS in reactor/kg SS in renctor influent/day.
* 1 Ih = 0.45.1 kg
** 1 Ib/mil gal = 0.12 kg/1000 mj
-------
SECTION 6
EQUIPMENT AND INSTRUMENTATION OPERATING EXPERIENCES
A multitude of equipment failures and inadequacies were
experienced during the course of the test. Unless specifically
noted in the operational narrative, Section 5, these difficulties
did not significantly affect the overall performance of the pro-
cess. However, the great number of problems did require a large
cv.nount of attention by the technical and operating staffs. This
information is provided, not as a criticism of the original de-
sign and selection of particular products, but as a matter of
record.
PSA OXYGEN GENERATOR
This unit was rehabilitated in late summer 1974 after 27
mo of rather unreliable service. It operated practically flaw-
lessly after the overhaul, in contrast to its first 2 yr.
Most of the problems were associated with malfunctioning
valves beginning early in the program, which resulted in moisture
contamination of two of the four adsorption beds. Prom that
time forward, the maximum output of the unit was 10.4 metric
tons/day (11.5 tons/day) of 90 percent pure oxygen, instead of
the design rate of 15.2 metric tons/day (16.7 tons/day). To
protect the adsorption beds from further moisture contamination
from stuck valves, an oxygen purity analyzer was later installed
to continually monitor the PSA product purity. The analyzer
sounded an alarm when the purity fell to 85 percent and shut
down the PSA unit at 80 percent.
Continued valve problems required the replacement during
the course of the test of all valves, valve actuators, and
solenoids, and some pressure switches. The replacements, of
a different design, performed considerably better than the ori-
ginals, but they too failed to function properly from time-to-
time .
During checkout before the initial process startup, a de-
fective bearing was discovered in the PSA compressor motor, re-
quiring its removal and repair by the manufacturer.
In September 1973, the PSA oxygen generator was again shut
down for 10 days awaiting the return of the compressor motor,
which had been removed for repairs to the same bearing, and the
47
-------
repair cf a scored shaft.
After 2 mo of operation, the PSA cooling tower fan motor
burned out, requiring its replacement; 2 mo later, the replace-
ment motor was removed for repair of a defective bearing.
In the 27 mo prior to its overhaul, the PSA generator was
out of service for 3054 hr of a possible 18,504 hr, or 16.5
percent of the time. In the 7 mo following its rehabilitation,
there were only 57 hr of interrupted service, or 1 percent down-
time .
It must be noted that this PSA unit was the first of its
size ever manufactured, and so some bugs were inevitable.
LIQUID OXYGEN FEED VALVE
In September 1973, with the PSA generator shut down for the
second compressor bearing repair, the process was dependent en-
tirely on the liquid oxygen supply. At this most inopportune
time, the motor operating the liquid oxygen feed valve burned
out and had to be shipped to the manufacturer in California for
repair. Without the automatic valve, the oxygen feed rate could
no longer be controlled by the gas space pressure and only manual
control through the valve bypass was possible. This crude manual
control could not be adjusted for variations in organic loading
and resulted in less effective oxygen utilization. Although the
PSA unit was returned to service a week after the valve motor
failure, once again permitting PSA oxygen feed control, liquid
oxygen feed control was not possible during the 1 mo required
for the motor repair.
STAGE COMPRESSORS
During the first days of the test, bearing failures attri-
buted to improperly manufactured ball retainers required the re-
placement of both sets of shaft bearings on all five gas recir-
culating stage compressors and the substitution of a conventional
petroleum lubricant for the synthetic-based lubricant originally
supplied. Even after the replacements were installed, additional
bearing failures were experienced. It was then discovered that
rain was entering the bearing housings through the shafts, conta-
minating the lubricant and causing the failures. In October
1972, weather-proof shields were installed around the shafts
and some bearings were replaced for the second time. Still
another bearing failure occurred in August 1973.
In addition to the compressor shutdowns resulting from bear-
ing failures, there were nuisance shutdowns caused by excessive
vibration and low seal air pressure during the first 5 mo. Oc-
casional failures of the couplings between the seal air compres-
sors and their motors, as well as failures of the seal air sys-
48
-------
terns themselves, developed aCter a year of operation. Some of
these difficulties were repaired by plant personnel, while others
required the services of the manufacturer,
PROPELLER MIXERS
Oil leaks plagued the operation of the propeller mixers un-
til, after 2 mo of operation, gaskets were fabricated for the
speed reducers. Improperly sealed rotating unions resulting in
considerable oxygen leakage required replacement; some rotating
unions were still leaking at the end of the test program. On
one of the two first-stage mixers, the drive motor burned out,
some gears and bearings required replacement, and the motor
bearings failed. After 6 mo of operation, the mixer high-oil
temperature and vibration shutdown systems, which were causing
nuisance shutdowns, were doomed extraneous and were removed.
RETURN SLUDGE PUMP
The return sludge pump was the single most troublesome piece
of equipment. With no spare unit provided, bearing failures in
November 1972 and again in November 1973 required complete pro-
cess shutdowns to accomplish the repairs.
In addition to the bearing failures, another problem was
evidenced by the pump going out on overload at less than half
its capacity. The overload resulted from debris accumulating
between the shaft sleeve and the packing and between the wearing
rings, and also because of a restricted suction.
When the pump was taken out of service for the first bearing
replacement, the factory-supplied grease seal system was replaced
-with a water-seal system. The higher pressure water-seal system
prevented the further accumulation of foreign material between
the shaft sleeve and packing and also halted the excessive leak-
age that had been experienced through the packing. The first
plant shutdown provided the opportunity to modify the pump suc-
tion line; remove the accumulation of grit, rags and other for-
eign material from the sump, which had been responsible in part
for the reduced pump capacity; and install a water jet to pre-
vent future accumulations. With the improved seal system and
the suction modification, the pump output increased significant-
ly and overload shutdowns became less frequent. During the
second shutdown, it was discovered that the clearance between
the wearing rings was excessive, allowing debris to accumulate
and restrict the rotation of the impeller. The clearance was
reduced by building up the impeller wearing ring in the factory
because a new set of wearing rings was not readily available.
WASTE SLUDGE PUMP
The major difficulty with the waste sludge system was an
49
-------
improperly-designed pump suction. Originally, the suction line
consisted of a 7.6-cm (3-in.) pipe teed from the 51-cm (^0-in.)
return sludge pump suction line and operation was frequently
interrupted when a vacuum formed in the line and halted the
waste sludge flow. Wasting was then attempted by manipulation
of valves and the use of an alternate pipeline, but a constant
flow could not be maintained. During one of the shutdowns, a
piping modification was made to allow the waste sludge pump to
take suction from the discharge of the return sludge pump. This
also proved unsuccessful: with the waste sludge pump in series
with the much larger return pump, the waste flow v/as dictated
by the return pump and operation of the waste sludge pump vari-
able-speed controller had very little effect on controlling the
flow rate. Finally, during the third plant shutdown, a new 7.6-
cm (3-in.) sucti.cn line from the pump directly to the sump v/as in-
stalled. From then on, operation of the waste sludge pump com-
pletely independent of the return system was possible.
The only problem with the waste sludge pump itself was a
cracked casing, probably caused by water freezing in the casing.
During the second summer of operation, the plant force installed
a water seal system for the pump to replace the factory-supplied
grease seal system.
MAGNETIC DRIVES FOR VARIABLE-SPEED PUMP CONTROL
Before the initial process startup, the influent pump vari-
able-speed magnetic drive had to be removed and returned to the
manufacturer for the replacement of a defective bearing. During
the second summer of operation, the magnetic pick-up, which is
the mechanism that initiated the automatic control of the return
;ludqe pump, became inoperative, requiring manual control of the
pump for a short time until the pick-up was replaced by the man-
ufacturer. Two mo later, a faulty signal converter was discovered
in the return sludge pump variable-speed drive. Following a
flooding of the pump pit, the waste sludge pump lost its speed
control and the pump could not be used for the weeks until the
manufacturer replaced the tachometer generator and the defective
drive bearings.
WASTE SLUDGE MAGNETIC FLOW METER
In January 1973, after 8 mo of operation, a volumetric test
of the waste sludge magnetic meter showed that the meter v/as
reading low by almost 40 percent. The manufacturer's explanation
was that the indicated calibration voltage may have been incor-
rect. This error, coupled with additional problems encountered
with the power supply boards, required a non-linear correction
to the acquired data. Consequently, the first 8 mo of waste
sludge data were inaccurate and had to be discarded. The defec-
tive meter was replaced in February 1973 by one borrowed from
50
-------
another manufacturer, and in late March, the second mete: w< re-
placed by a third meter from a third manufacturer. The .second
meter was satisfactory, but it was only a loan and had to be
returned. By this time, the staff had become wary of magnetic
flow meters and weekly volumetric tests of the waste sludge meter
were begun. Therefore, the waste sludge data acquired from March
1973 until the end of the test are reliable. Although the per
manent replacement meter proved adequate, volumetric checks show-
ed that on two occasions it too required service.
RETURN SLUDGE MAGNETIC FLOW METER
In September 1972, the first routine service check of this
meter revealed an improper electrical connection and an incor-
rect calibration voltage. The return sludge flow data for the
first 4 mo of the project, before the meter was recalibrated,
were therefore dismissed. Because of the large volume of return
sludge, a volumetric check could not be performed in place, as
was possible with the much smaller waste sludge meter. The meter
was, therefore, removed for a factory bench flow test after al-
most a year-and-a-half of operation. Accuracy within 1 percent
was ascertained. Malfunctions of the electrical circuitry were
experienced on two occasions after September 1972 and were cor-
rected by the manufacturer.
In November 1974, after the conclusion of 11 phases of the
test program, it was discovered that the return sludge piping
had been installed in a manner which permitted the entrainment
of air in the meter. Since proper operation of a magnetic flow
meter requires full liquid flow, all return sludge data for the
first 11 phases before the condition was rectified must be viewed
with caution. If anything, recorded return sludge flow data dur-
ing this period were higher than actual.
INFLUENT MAGNETIC FLOW METER
The first routine service check in September 1972 also re-
vealed an incorrect calibration voltage in the influent magne-
tic flow meter. The resulting incorrect flow data collected
from May through September 1972 were, therefore, abandoned.
After recalibration, a volumetric check of the meter indicated
acceptable accuracy. No electrical problems due to circuitry
were experienced with the meter.
OXYGEN FEED FLOW METER
Difficulties were experienced with the flow totalizer
counter, the integrator, the square foot extractor, and the
multiplier-divider, all of which required service by the vendor.
Although the design combined oxygen supply capacity of the PSA
generator and liquid supply system was 30 metric tons/day (33
tons/day), the upper limit of the oxygen feed flow meter was only
51
-------
23 metric tons/day (25 tons/day). In the summer of 1973, when
system influent flow was increased to 1.31 m3/sec (30 mgd), the
resulting increased oxygen flow exceeded, the capacity of the
meter. During this period, oxygen feed was measured by adding
to the PSA generator output the amount of liquid oxygen trucked
in. in September 1973, the capacity of the meter was increased
to 36 metric tons/day (40 tons/day); however, incorrect calibra-
tion data were used to increase the range of the meter, and an
incorrect procedure was used during subsequent routine calibra-
tions. The extent of the resulting error was finally determined,
and all oxygen feed data acquired from September 1973 to April
1974 were corrected.
HYDROCARBON ANALYZERS
For the first 8 mo of the project, frequent failures of the
diffusion heads, as well as large span errors and drifts from
zero, were responsible for unreliable analyzer readings. Accord-
ing to the manufacturer, the poor performance resulted from ac-
celerated oxidation of the heads in the moisture-laden, oxygen-
rich environment. Performance improved after the heads were re-
placed with others of a design more compatible with an oxygen
environment, but for the remainder of the program, the zero and
spin Required weekly adjustments and the diffusion heads required
replacements, although infrequently.
OXYGEN PURITY ANALYZER
Although this unit performed satisfactorily, slight drifts
from zero and span errors required weekly calibrations.
DISSOLVED OXYGEN METERS
After 5 mo of troublesome operation, automatic DO measure-
ment was abandoned. It was impossible to keep the meters cali-
brated. Not only were the high DO levels accelerating the wear
of the probes, but with the probes located in the sparger bubble
stream, they were continually battered by turbulence, even after
deflectors were installed. The end of automatic DO measurement
also halted automatic control of the first-stage DO level.
Thereafter, the DO was determined manually with a portable analy-
zer at infrequent intervals since that measurement was not con-
sidered to be necessary for successful operation.
SAMPLE COMPOSITORS
The effluent automatic sampler operated successfully
throughout the project, but the influent unit was not as for-
tunate . The sampling line clogged so frequently that after a
little more than a year of operation, it was retired and grab
sample compositing was substituted.
52
-------
SERVICE CONTRACT
In order bo ensure data accuracy, as we.1.1 as to attend to
the frequent instrumentation problems, a service contract was
entered into between the City and an instrumentation company.
This contract called J:or the services of a technician 1 full day
each week, in addition to emergency calls.
53
-------
SECTION 7
DISCUSSION OF RESULTS
EPA's purpose in sponsoring the test was to obtain, evalu-
ate, and disseminate operating and performance data for us-* in
the design or upgrading of plants throughout the country. Eqr; il-
ly important to EPA were sludge production, power and oxyjem / -
quirements, and equipment performance. In the vJew of the C;^.
of New York, the prime requisite for success and r ; joptance f
the process was 90 percent removal of BOD and SS al; '.83 m-^, ; .-;„
(20 mgd) . A large body of data was obtained even ^i'cer siz.ible
portions had to be discarded because of defective flow meters.
However, although effluent quality can be readily assessed, only
qualified conclusions can be offered regarding sludge production,
power requirements, and oxygen consumption.
OXYGEN SYSTEM PROCESS PERFORMANCE
Effluent quality during the 12 phases averaged 19 mg/1 each
of BOD and SS and ranged from 9 to 26 mg/1 of BOD and 12 to 27
mg/1 of SS. The averages for the eight non-filamentous phases
were the same--19 mg/1 for BOD and SS—and for the four fila-
mentous periods, 21 and 20 mg/1, respectively. Thus, although
the fungi, were responsible for some operating difficulties and
for increased sludge volumes, they had no significant effect on
process efficiency.
No supportable relationship between sewage flow and effluent
quality could be discerned. During Phases 5, 6, and 7, when in-
fluent flow rates were 120 to 177 percent of the design flow and
the BOD loading rates were 123 to 140 percent of the design load-
ing, the effluent quality was only slightly below those at lower
flows, and, indeed, improved as the rate was increased.
Removals during Phase.1 (at constant ilow, not diurnal) were
higher than during any other period—94 percent of BOD and 92
percent of SS. Removals during Phases 8 and 9, at flows only 49
and 7C percent of Phase 1's 0.91 m^/sec (20.8 mgd), and during
the same season of the year, were 3-5 percent less. The reason
is not known.
Also unknown is the reason for the failure of the activated
sludge culture re return to its normal diversity in Phase 11
following the severely filamentous Phase 10. Effluent BOD and
54
-------
SS averaged 24 and 26 mg/1, respectively, just about the same as
they ware during Phase 10.
There was a sharp drop in the COD/BOD ratio during Phases 5
and 6 (June and July 1973) and a gradual return to normal in
later periods. Compared to the four earlier phases, influent:
COD strength was 16 percent lower during Phases 5 and 6 but the
BOD was 38 percent greater.
Average influent TOG ranged from 90 to 11C mg/1 and effluent
TOG from 21 to 54 mg/1. The 54 mg/1 occurred in Phase 4 and was
far greater than the usual average effluent range of 21 to 35
mg/l. Except for Phase 4, the fungus periods were highest in
effluent TOG.
There is strong but circumstantial evidence that restarts
of the Newtown Creek oxygen system during the first two winters
were responsible for the greater growth of the fungus experienced
in those winters as contrasted to the third winter. However,
complete shutdowns would not occur in a permanent plant system
equipped with the usual assortment of backup equipment.
CONSULTANT OBSERVATIONS AND PILOT PLANT STUDIES
The following conclusions are based on the fungus consult-
ant's observations (10) and the 1974-1975 winter oxygen pilot
plant studies at Newtown Creek and Wards Island:
1. The same fungus present in Newtown Creek wastewater was also
present in the four other wastewaters sampled, and, thus, it is
highly likely that it exists at all other City plants.
2. The fungus organisms did not significantly concentrate in
the mixed liquor of the three other plants examined nor in the
Wards Island oxygen pilot plant.
3. Newtown Creek's wastewater Js conducive to the growth and
proliferation of the fungus in activated sludge mixed liquor
during cold weather (wastewater temperatures below approximately
60 F) , more so itnder oxygen.
4. No fungicide or operating strategy was found which could
satisfactorily control growth of the undesirable organisms during
winter operation at Newtown Creek.
COMPARISON OF OXYGEN SYSTEM AND AIR PLANT PROCESS PERFORMANCE
It was expected that the UNOX process would provide removal
efficiencies greater than those of the Newtown Creek air plant,
for the air plant was being operated as a high-rate conventional
regime and the UNOX process was considered a substitute for step
aeration. Excepting the severe fungal Phase 10, the UNOX effln-
55
-------
ent averaged 15 mg/1 of BOD and 25 mg/1 of SS lower than the air
plant's, which operated at a flow of only 11.3 mgd. In Phase 10,
the air plant effluent BOD was 3 mg/1 higher but the effluent
SS was 1 mg/1 lower. However, during Phase 12 (the third win-
ter) , the UNOX effluent was better than the air plant's by 15
ivg/1 of BOD and 28 mg/1 of 3S.
In terms of percent removal, the oxygen system averaged
88 percent for DOD and 86 percent for SS: xor the air plant,
the averages were 81 and 68 percent, respectively. Oxygen sys-
tem flow ranged from 0.44 to 1.55 m3/sec (10.1 to 35.4 mgd) with
an average of 0.91 m3/sec (20.7 mgd), while the air plant single
module average was 0.50 m3/sec (11.3 mgd) with a range of 0.42
to 0.54 m3/sec (9.7 tc 12.4 mgd). Thus, the oxygen system_per-
formed considerably better than the air plant en nearly twice
the flow. In terms of filtrate BOD removal, the oxygen system
averaged 87 percent compared to 77 percent for the air plant.
Flow and process performance (both concentration and percent re-
moval) data were presented previously in Tables 2 and 10 for,
respectively, the Newtown Creek oxygen demonstration and full-
scale air systems. BOD and SS removals for the two systems are
compared graphically in Figures 6 and 7, respectively.
The two systems reacted differently during the three win-
ters. Oxygen system effluent quality was mildly affected during
Phase 3 (the first winter) , but BOD removal by the air plant fell.
to 73 percent, the lowest of all 12 phases. During Phase 10_(the
second winter), it was oxygen process performance that deterior-
ated; the air plant was not affected. During Phase 12 (the third
winter), however, the air plant was greatly affected and the
oxygen system only slightly.
Separate influent samples were collected and analyzed for
the air plant and the UNOX system, but the UNOX influent data
have been used for the air plant also because they are consider-
ed more reliable. The air plant samples were composites taken
at the entrances to the aerated grit chambers of two of the
14 tanks; the UNOX samples were taken at the end of the non-aer-
ated qrit chamber where, despite some settling of SS, the BOD
was only slightly reduced. There was a more serious hazard at
the air plant sampling location, where a tendency existed for
drawing a stream of return sludge into the aerated grit chambers
and contaminating the samples.
AERATION TANK PERFORMANCE
The oxygen diffusion equipment at first was designed for a
BOD loading of 15,150 kg/day (33,400 Ib/day), but was enlarged
before construction to 18,915 kg/day (41,700 Ib/day) because
the latest analytical data at the time showed the BOD influent
strength to be ubout 250 mg/1. Thus, the design volumetric or-
ganic loading '.or the 4656-iP3 (1.23-m.il gal) aeration tank was
56
-------
FLOW (MGD)" - OXYGEN SYSTEM vs. A!H PLANT
FLOW
(MGD)'
40-
30-
20-
10-
0
1 MGD = 0.011 M3/SEC
OXYGEN
PI AN I DESIGN FLOW PER BAY
NAIR (FLOW PER BAY)
-1—-—I—:—I—~—T
1 ' 2 ' 3 ' 4
5 ' 6 7
PHASE
9 ' 10 ' 11 ' 12
EFFLUtNT BOD (MG/L) - OXYGEN SYSTEM vs. AIR PLANT
MINIMUM
FEDERAL STANDARD
EFFLUENT
BOD 20 -
(MG/L)
23 4 5 6 /
9 10 1' 12
BOD REMOVAL (%) - OXYGEN SYSTEM vs. AIR PLANT
90-
30D
r.bMOVAL
MINIMUM
FEDERAL STANDARD
'2'3'4'5'6'7 ' 8 ' 9 10 ' 11 12
PHASE
Figure 6. Comparison of Newtown Creek oxygen system and air
plant BOD removal performance.
-------
FLOW (MOD)* - OXYGEN SYSTEM vi. AIR PLANT
FLOW
(MGD)'
40 1
30-
20
OXYGEN
.ANT DESIGN PLOW PER BAY
' 1 MGD = 0,044 M3/SEC
AIR (FLOW PER BAY)
1 ' 2 3 ' 1 ' 5 6 ' 7 ' 8 ' 9 ' 10 ' 11 ' 12
10-
EFFLUENT SS (MG/L) - OXYGEN SYSTEM vs. AIR PLANT
EFFLUENT
SS
(MG/L)
50
40
3° ^
10-
MINIMUM FEDERAL STANDARD
0 1 ' 2 ' 3 ' 4 ' 5 ' 6 ' 7 ' 8 ' 9 ' 10 ' 11 ' 12 '
PHASE
SS REMOVAL (%) - OXYGEN SYSTEM vs. AIR PLANT
90-i
MINIMUM
FEDERAL STANDARD
SS
REMOVAL
l'2345678'9 10 '11
60
PHASE
Figure 7. Comparison of Newtown Creek oxygen system and air
plant suspended solids removal performance.
58
-------
4.05 kg BOD/day/m (253 lb/day/1000 ft ). The volumetric or-
ganic loadings encountered during most of the test were consid-
erably lower than the design loading, but during the high-flow,
high-BOD Phases 5, 6, and 7, they averaged 4.95 to 5.64 kg BOD/
day/m3 (309 to 352 lb/day/1000 ft3), 123 to 140 percent of de-
sign. This was during the periods of 1.10-to 1.53-rn3/sec (25-
to 35-mgd) flows and BOD removal efficiencies of 90 percent. It
can be said, then, that the process performed exceptionally v/ell
during 3 mo of flow and BOD overloading at high sewage temper-
atures and in the absence of filamentous organisms.
The oxygenation tank detention time, which was 1.1 hr at
0.88-m3/sec (20-mgd) influent flow and 25-percent return sludge
flow, decreased to 48, 44, and 40 min, respectively, during
Phases 5, 6, and 7. These and other aeration tank data for the
oxygen system and air plant were shown previously in Tables 3
and 11, respectively.
FINAL TANK PERFORMANCE
Oxygenated return sludge was expected to average 3 percent
solids, but the highest phase average actually obtained was only
18,600 mg/1 in Phase 9. Excluding Phase 10, the average through-
out the test was 14,300 mg/1 compared to 10,500 mg/1 for the air
plant. The reduced compaction of the sludge in the settling
tank, chiefly during Phase 10, required the use of high return
sludge rates to maintain MLSS (which averaged 3930 mg/1 for the
entire test program) at desired levels.
The oxygen system final tank overflow rate ranged from 19
m3/day/m2 (460 gpd/ft2) at a flow of 0.44 m3/sec (10.1 mgd) to
66 m3/day/m2 (1610 gpd/ft2) at 1.55 m3/sec (35.4 mgd). The data
do not show any impairment of effluent quality due to increasing
overflow rate. The adjacent Tank No. 10, which was kept empty
as a standby if more settling area was needed, was not required.
The solids loading on the oxygen settling tank ranged from
.1.37 to 313 kg/day/m2 (28 to 64 Ib/day/ft ) , from two to seven
times the loading on the air plant settling tanks. Weir load-
ings reached a high of 2749 m3/day/m (221,300 gpd/ft), compared
to the maximum of 963 m3/day/m (77,500 gpd/ft) for the air plant.
The "Ten State Standards" (7) recommend a maximum overflow rate
of 33 m3/day/m2 (800 gpd/ft2) and a maximum weir loading of 186
m3/day/m (15,000 gpd/ft). Fo*- solids loadings, the normal prac-
tice is 59-88 kg/day/m2 (12-id Ib/day/ft2), with a maximum of
146 kg/day/m2 (30 Ib/day/ft2) (8).
Final tank performance data for the two Newtown Creek pro-
cesses were previously summarized in Tables 4 (oxygen) and
12 (air) .
59
-------
SLUDGE SETTLING CHARACTERISTICS
During non-fungus periods, the unstirred UNOX SDI averaged
t.t g/100 ml, and seldom fell below 2.0. During Phases 3 and
10, and less so in Phases 2 and 12, the Index slid severely to
averages of 1.2, 1.8, and 1.5 g/100 ml. In addition, there were
extended periods during Phase 10 when the SDI bottomed as low as
0.4 g/100 ml. A full plant-scale test under such circumstances
would have been afflicted with a formidable sludge disposal pro-
blem. In contrast, the Newtown Creek air plant Index was not
measurably affected by the fungus, averaging 1.5 g/100 ml
throughout the project and on a monthly averaae basis varvinq
only from 1.3-1.8 g/100 ml. Sludge settling data for the oxygen
ana air systems are included in Tables 4 and 12, respectively!
OXYGEN SYSTEM OXYGEN REQUIREMENTS
Throughout the test program, an unknown volume of oxygen
leaked through the oxygenation tank cover and through the mixer
unions. The vent gas was metered, but since the quantity lost
through the leaks could not be measured, the amount of oxygen
actually used could not be calculated. It is for this reason
that only oxygen supplied was listed in Table 5.
In order to minimize the leaks, the gas pressure under the
cover was reduced to 2.5-5.1 mm (0.1-0.2-in.) of water. This was
rar less than the design pressure of 25-76 mm (1 to 3-in.), but it
did not affect the process.
The amount of oxygen supplied averaged 173 g/m3 (1440 lb/
mil gal) and was somewhat proportional to the MLVSS and the
sludge age. Based on BOD removal, the oxygen requirements varied
from 0.8 to 1.6 kg supplied/kg BOD removed. At or above the de-
sign flow, the amount of oxygen supplied averaged 1.1 kg/kq BOD
removed; at flows below design, it ranged from 1.2 to 1.6 kq/kq
During the 1973 summer high-loading phases (Phases 5, 6 and 7)
the requirement was only 0.8 kg supplied/kg BOD removed. '
There is much less variation among the phase averages when
the oxygen requirement is calculated on a COD removal basis In
the 11 phases in which oxygen supply was monitored, the require-
ment was either 0.6 or 0.7 kg supplied/kg COD removed in eiqht
phases, and either 0.8 or 0.9 in the other three. In view of
the industrial nature of the Newtown Creek wastewater, COD re-
moval probably is the better parameter.
POWER REQUIREMENTS
The oxygen-generating and dissolution equipment, which in-
cluded the PSA unit and its compressor, the gas recirculating
stage compressors, the stage mixer spargers, and the liquid oxy-
gen vaporizer, was sized for a BOD loading of 18,915 kg/day
60
-------
(41,700 Ib/day), Unfortunately, however, it was not equipped
with turn-down capability to accommodate lower loadings; about
all that could be done was to throttle the suction !ine of the
PS?, compressor, which reduced the power only slightly. The re-
sult was that essentially the entire 746 kW (1001 hp*) of the
equipment was used at all times, regardless of the loading. The
power requirements at BOD loading rates below design (during all
phases except 5, 6, and 7), therefore, are not fairly repre-
sentative of the process. It was only during those three 1973
summer phases when the equipment was fully loaded the>t the power
required can be equitably compared with the work performed.
Tables 6, 13, 15, and 16 contain data on th
-------
For Jamaica, the power required by the air blowers during
the reporting period of March 1972"Febt:uary 1973 averaged 1./18
kWh/kg BOD removed (0.90 hp-hr/.lb) with the lowest rate of 1.21
(0.7<1) achieved dnriny four of the 12 mo. For 2Gth Ward, (;ho
average fj-om June l97D~October 1976 was Q.H4 kWh/kg BOD re-
moved (0.51 hp~hr/lb) with a j arnje of 0.71 to 1.0'3 "(0.13 to
O.G-1). Caution should be used when eonsi derimj these power data
relative to tlic Newtown Crock oxygena tion system data since the
parameter used here (kWh/kg BOD removed) does not account for
differences in the rate of oxygen consumption per unit BOD re-
moved or for differences in the plant loading relative to the
design point. Therefore, direct comparison of the.se data to
data from the Newtown Crook oxygenation system may not be valid.
SOLIDS PRODUCTION
As a practical operating consideration, more important than
the dry weight of sludge produced is its volume, for it is sludge
in its liquid form which must be processed. When sludge is dig-
ested, as at Newtown Creek, the volume can best be measured when
pumped from a thickening tank. Waste oxygen sludge was added to
the air plant's return sludge system because piping the waste
oxygen sludge and/or oxygenated mixed liquor to one of the
plant's eight thickeners would have been too expensive for what
was, after all, to have been only a 1-yr test. Thus, no informa-
tion was obtained on the gravity thickening characteristics of
UNOX Sludge.
Therefore, the only estimation that can be made of excess
sludge production with oxygen at Newtown Creek is on the basis
of the dry solids in the waste sludge and the final effluent.
Figure 8 compares UNOX sludge production at Newtown Creek with
that observed for the air and oxygen trains at Batavia, New York
(9), and relates kg volatile solids produced/day/kg MLVSS to kg
BOD removed/clay/kg MLVSS.
Four Newtown Creek points agree very closely with the P.ata-
via oxygen distribution, while three fungus points and one
transition point fall in the neighborhood of the Batavia air
distribution. Sludge production data for Phases 1 and 2 could
not be used because of the i.mccurately calibrated waste sludge
r.,eter in use at that time; Phase 9 data were om.i tied because the
mixed liquor samples were contaminated by the froth during that
period; and insufficient solids analysis were performed in Phase
11 to plot data from that phase. In comparing sludge production
data for Newtown Creek and Batavia, it should be remembered that
both plants operate wxthout primary sedimentation.
Newtown Creek oxygen system sludge production during normal
and fungus periods can also be compared with sludge production
during the same periods for the Newtown Creek air plant. As pre-
viously summarised in Table 6, during non-fungus operation, ex-
62
-------
1.2
CO
1.0
0.8
LLI
O
D
Q
O
QC n G
Q. 0.6
CO
Q
_J
O
CO
3 0.4
O
0.2
0
PHASE No. 7
PHASE No, 12
O
BATAVIA
AIR AERATION
BATAVIA
OXYGEN AERATION
PHASE No. 8
NEWTOWN CREEK
O FILAMENTOUS PERIODS
« NON-FILAMENTOUS PERIODS
0.4 0.8 1.2 1.6
KG BOD REMOVED/DAY/KG MLVSS
2.0
* INCLUDES BOTH WASTE SLUDGE AND FINAL EFFLUENT VOLATILE SOLIDS
Figure 8. Excess solids produced by Newtown Creek oxygen system
superimposed on Batavia, New York sludge production
graph.
63
-------
cess solids generated by the oxygen system averaged 0.93 kg/kg
BOD removed but rose to an average of" 1.27 kg/kg BOD removed,
37 percent higher, during Phases 3, 10, and 12 (the three fungus
phases). For the air plant, the comparable figures were 1.27
and 1.71 kg/kg BOD removed, respectively, as documented pre-
viously in Table 14. Oxygen, there tore, produced on a BOD re-
moval basis 27 percent less solids than the air plant during
normal periods and 26 percent less during fungus periods.
Data from the Jamaica step aeration air plant, which was
not troubled by filamentous organisms, were confined to a nar-
rower range, averaging 1.1.6 kg/kg BOD removed and varying from
0.95-1.34 during the 12-mo period of March 1972-February 1973
(see Table 17). Using these bases, Newtown Creek oxygen system
solids production when unaffected by the fungus was 23 percent
less than Jamaica's and during the fungus phases, 6 percent
more.
A comparison can also be made with the excess solids pro-
duced at New York City's 26th Ward step aeration air plant (also
not infected by filamentous growths) during the first 17 mo fol-
lowing its upgrading (Table 18). At an average sludge age of
4,0 days, twice that of the Newtown Creek oxygen system, solids
production averaged 1.39 kg/kg BOD removed. Oxygen system pro-
duction wr.s 9 percent less solids during filamentous periods and
33 percent less during non-filamentous periods.
OXYGEN SYSTEM NUTRIENT REMOVALS
One mo of the program had been set aside for a test of
phosphate removal by alum addition. However, in June 1973, a
state-wide ban on the use of high-phosphate detergents went into
effect, reducing the already low influent phosphorus concentra-
tion which had averaged 3.9 mg/1 for the previous 8% mo to 2.8
mg/1 for the month of June. Probably the ban had an effect be-
fore June because all high-phosphorus products had to be removed
from store shelves before that date. During the eight phases
following the initiation of the ban, the oxygen system without
alum addition achieved 59 percent total phosphorus removal from
2.P mg/1 down to 1.2 mg/1. The alum addition test, therefore,
was considered unnecessary.
Table 8 lists by phase the average concentrations of influ-
ent and effluent total phosphorus, soluble orthophosphate, and
the major nitrogen components. Influent soluble orthophosphate
was similarly reduced by the effects of the ban from 2.2 mg/1 to
1.7 mg/1 (both as P). The average effluent concentration from
June 1973 on was only 0.7 mg/1,
Ammonia nitrogen averaged 9.5 mg/1 in the influent and 8.2 mg/1
in the effluent. Removals were erratic, varying from 0 to 39
percent. The TKN content was relatively steady, averaging 21.0
64
-------
mg/1 in the influent sewage and 12.8 mg/1 in the effluent.
OTHER OXYGEN SYSTEM SEWAGE CHARACTERISTICS
Averages for alkalinity, dissolved solids, turbidity, and
pH were previously given in Table 9. The principal item of in-
terest here was the drop in the pH of the wastewater in its trav-
el through the covered oxygen aeration tank and its recovery in
the settling tank. This decrease in pH may have played some
part in the proliferation of the fungi.
Analyses for dissolved solids and turbidity were discontinu-
ed after Phase 5 to reduce the workload on the laboratory.
65
-------
REFERENCES
1. Albertsson, J.G., McWhirter, J.R., Robinson, E.K., and
Vahldieck, N.P., "Investigation of the Use of High Purity
Oxygen Aeration in the Conventional Activated Sludge Pro-
cess," Water Pollution Control Research Series Report No.
17050 DNW 05/70, U.S. Department of the Interior, Federal
Water Quality Administration, Washington, D.C., May .1.970.
2. "Standard Methods for the Examination of Water and Waste-
water," 13th ed., American Public Health Association,
Washington, D.C., 1971.
3. Streeter, H.W. and Phelps, E.B., "A Study of the Pollution
and Natural Purification of the Ohio River," Public Health
Bulletin 146, U.S. Public Health Service, Washington, D.C.,
1925.
4. "Methods for Chemical Analysis of Water and Wastes," U.S.
Environmental Protection Agency, Cincinnati, Ohio, 1971.
5. Private communication with R.F. Lewis, U.S. Environmental
Protection Agency, Cincinnati, Ohio, March 1973.
Private communication with R.F. Lewis, U.S. Environmental
Protection Agency, Cincinnati, Ohio, November 1973.
7. "Recommended Standards for Sewage Works," Great Lakes -
Upper Mississippi River Board of State Sanitary Engineers,
1971.
8. Dick, R.I., "Role of Activated Sludge Final Settling Tanks,"
Journal Sanitary Engineering Division, American Society of
Civil Engineers, 96, No. 2, pp. 423-436, April 1970.
9. Brenner, R.C., "EPA Experiences in Oxygen-Activated Sludge,"
Prepared for the U.S. Environmental Protection Agency
Technology Transfer Design Seminar Program, Cincinnati,
Ohio, October 1974.
10. Private communications (Progress reports to New York City
Department of Water Resources) with W.O. Pipes, Department
of Biological Sciences, Drexel University, Philadelphia,
Pennsylvania, February and March 1975.
66
-------
APPENDIX A
DETAILED DESCRIPTION OF EQUIPMENT
MECHANICAL EQUIPMENT
PS7. Oxygen Generation System
Oxygen gas was produced on-site by a Union Carbide pressure
swing adsorption (PSA) unit and its associated equipment, which
included an air compressor, cooling tower and pump, and instru-
ment air booster and dryer. Air at rates up to 4050 std mj/hr
(143,000 scfh) and pressures up to 28,125 kgf/nr (40 psig) was
fed to the PSA unit from a Clark Isopac-3 two-stage centrifugal
compressor, driven by a Louis Allis Co. 336-kW (450-hp) motor.
The PSA unit was designed to produce 473 std mj/hr (16,700 scfh)
equivalent to 15.2 metric tons/day (16.7 tons/day) of 90 percent
pure oxygen, or 13.6 metric tons/day (15 tons/day) of 100 per-
cent pure oxygen.
Liquid Oxygen Tank and Vaporizer
These units consisted of a Union Carbide liquid oxygen
storage tank with an equivalent gas capacity of 35,853 m
(1,266,000 ft ) and an electrically-heated, water-bath vaporizer
with a capacity of 510 std m3/hr (18,000 scfh) of gas. The tank
was a double-walled, vacuum-power-insulated unit with a liquid
capacity of 42 m3 (11,000 gal) and a working pressure of 45,700
kgf/m2 (55 psig). The vaporizer consisted of a l.Q-nH (500-gal)
water ballast tank including a vaporizer coil for the product
stream, smaller pressure-building coil for pressure maintenance,
and an electric immersion heater. The heater, rated at 72 kW,
maintained a temperature of 130 F. A sensor in the water bath
signaled an alarm if the temperature fell below 90 F, and another
sensor in the oxygen piping downstream of the vaporizer sounded
an alarm if the temperature of the oxygen fell below 0 F.
Mixers
The characteristics of the eight Philadelphia Gear mixers
were slightly different from stage to stage. The Stage 1 mixers
were driven by 37-kW (50-hp), 1750-rpm motoio, with 183-cm (72-
in.) propellers at a 267-cm (105-in.) pitch rotating at 77 rpm.
The mixers in Stages 2, 3, and 4 were driven by 22-kW (30-
hp), 1750-rpm motors. The propellers in Stage 2 were 183-cm
67
-------
(72-in.) diameter, 224-cm (88-,in.) pitch, and 70 rpm; .in Stage 3,
they were 183~cm (72~.in.) diameter, 183-cm (72-in.) pitch, and
74 rpm; and in Stage 4, .188-cm (74-in.) diameter, 188-crn (74-in.)
pitch, and 71 rpm. Eight-arm oxygen .spargers were rnountod to the
bottom of the hollow, stainless steel mixer .shafts 46-cm (18-in.)
beneath the propellers. Johnson rotary joints were used at the
tops of the .shafts as oxygen lead-ins to the rotating shafts.
Mo t or_Cp n t rp l_Center
The 460-volt motor control center (by General Electric Co.)
was located in the control building and contained motor controls
for all major system equipment, including the PSA air compressor,
cooling tower pump and fan, instrument air compressor, mixers,
oxygen recirculation compressors, influent pumps, return sludge
pump, and waste sludge pump.
Main Instrument Panel
The main instrument panel (MIP), located in the control
building, contained all the major instruments, controls, and
recorders, in addition to the annunciator system for signaling
abnormal operating conditions and alarms. The controls and
instruments on the MIP are described individually later in this
appendix.
Oxygen Recirculation Compressors
Five compressors (Hoffman Division of Clarkson Industries)
were provided, two for the first stage, and one each for Stages
2, 3, and 4. The Stage-2 unit was rated at 21 s td rn3/min (750
scfm), the other four at 18 std m3/min (650 scfm). Each was
driven by a 30-kW (40-hp), 3550-rpm motor and provided with an
integral seal air supply and necessary instrumentation for surge
prevention and vibration protection. The suction line leading
to each compressor was equipped with a V.D. Anderson Co. "Hi-eF"
separator to remove liquid entrained in the recirculated gas.
Influent Pumps
These units were Worthington 61-cm (24-in.) "Mixflo" horizon-
tal volute pumps. The variable-speed pump was driven by an
Electric Machinery Arnpli-Speed magnetic drive from a Westinghouse
119-kW (150-hp), 585-rpm motor. A Regutron II control and F.IS
signal transmitter were used to regulate the magnetic drive.
The constant-spaed pump was driven directly Ly a Westinghouse
119-kW (150-hp), 585-rpm motor.
R^oturn Sludge Pump
This variable-speed pump was a Worthington 51-cm (20-in.),
0.66-m3/sec (15-mgd) "Mixflo" vertical volute pump, driven by
68
-------
an Electric Machinery Ampii-Speed magnetic drive From a Westing-
house 75-kW (100-hp), 705-rpm motor. A Regutron II control and
an RIS signal transmitter regulated the drive.
Waste Sludge Pump
The waste sludge pump was a variable-speed, ITT Mar.Low,
vertical "Vane-Flow" solids-handling unit with a 0-1135-1/min
(0-300-gpm) range, driven by an Electric Machinery .\rnpli-Speed
magnetic drive from a 5.6 kW (7'-2-hp), 880-rpm motor. The Arnpli-
Speed drive was controlled by a Regutron II control, and speed
was manually regulated from the control building.
AUTOMATIC VALVES
Liquid Oxygen Feed Valve
The Driox feed valve was a 5-cm (2-in.) Masoneilan electron-
ically-controlled automatic globe valve with cast bronze body,
percent-contoured bronze trim, and Teflon seat. It was operated
by a General Controls hydromotor actuator in response to a 12-
to 20-ma increase-to-open input signal. The valve was set to
fail closed upon loss of power.
Oxygen Feed Valve
The PSA oxygen valve was a 5-cm (2-in.) Masoneilan auto-
matic bronze globe valve with a 3.8-cm (1^-in.) percent-contoured
trim. It was pneumatically-controlled, diaphragm-operated,
2100-6300 kgf/rn^ (3-9 psig) air-to-open, and equipped with a
positioner. An adjustable limit stop permitted presetting the
maximum open position of the valve. A solenoid valve in the
diaphragm supply normally was energized to allow the valve to
respond to its positioner signal. The solenoid valve war, de-
energized automatically in the event of a lower explosive level
(LEL) alarm, or manually by operation of a switch which inter-
rupted the signal, thereby closing the oxygen feed valve.
Stage 1 Mixer Bypass Valve
The bypass valve was a 10-cm (4-in.) Masoneilan electronic-
ally-controlled automatic globe valve with cast bronze body,
percent-contoured bronze trim, and Teflon seat. It was operated
by a General Controls hydromotor actuator in response to a 4-tc
20-ma increase-to-open input signal.
Air Bypass Valve
This was a 10-cm (4-in.) Norriseal automatic butterfly valve,
with case iron body, neoprene seat, Buna-N seals, and aluminum-
bronze disk. The valve was pneumatically-controlled, diaphragm-
operated, 2100-10,500 kgf/m^ (3-15 psig) air-to-open, with its
69
-------
air line interrupted by a solenoid valve. The solenoid was
normally energized, keeping the valve closed. The solenoid was
deenergized either manually by operating a sv/itch or automatic-
ally in the event of an LEL alarm.
Relief Valves
These valves were 30-cm (12-in.) Oceco regulators with
Pluorel (Viton) diaphragms designed to provide both pressure and
vacuum relief. Those for the first three stages were set to open
at 15 cm (6 in.) of water, the fourth-stage valve at 13 cm (5 in.)
of water. All were set to open at 10 cm (4 in.) of vacuum.
INSTRUMENTS
Stage 1 Pressure Controller
The Stage 1 pressure controller, located on the MIP. was a
Honeywell electronic deviation-indicating controller with a 4-
20-ma input signal and a 4-20-ma output signal. The controller
was reverse acting, that is, a decreasing input signal caused
the output to increase. The controller had both manual and
automatic control selection; the output signal was continously
displayed.
Stage 1 Pressure Recorder/Oxygen Feed Signal Recorder
A two-pen Honeywell strip chart recorder on the MIP recorded
Stage 1 reactor pressure and the oxygen feed signal on a single
0-100 chart.
Oxygen Feed Flow Totalizer
This instrument consisted of a Honeywell linear integrator
and a Honeywell totalizing counter with mechanical reset, both
located on the MIP. The integrator accepted a 1-5-volt-dc input
signal linearly proportional to the oxygen flow and produced
25-volt pulses to drive the totalizing counter.
Oxygen Feed Flow Recorder/Vent Gas Pressure Recorder
A two-pen Honeywell strip chart recorder on the MIP recorded
on a single chart the oxygen flow to the aerator and the vent
gas pressure at Stage 4.
Oxygen Purity Analyzer
This instrument, a Servomex Controls Ltd. (Adam David Co.)
oxygen analyzer, was also mounted on the MIP. It had two scales,
0-25 percent and 0-100 percent, and provided an output signal of
0-10 mv dc proportional to the scale, which was converted to a
4-20-ma signal. The analyzer normally monitored the purity of
70
-------
the vent gas stream from Stage 4, but connections permitted sam-
pling the PSA stream or the oxygen content of any of the other
stages.
Oxygen Purity Recorder
An instrument on the MIP recorded the oxygen purity on a
single 0 to 100 chart.
Stages 1 and 4 Combustible Gas Detection Systems
Two identical combustible gas detection system were pro-
vided to monitor Stage 1 and Stage 4 gas. Each system, by Mine
Safety Appliances, consisted of flow control components and the
MlP-mounted analyzer unit.
The flow control components consisted of the diffusion head
for detecting combustible gas concentrations and a flow meter
and necessary filters to protect the devices. A pressurized
stream from the recirculation compressor normally v/as sensed;
however, the low-pressure reactor gas space could be sampled
by the use of a pump and appropriate valves on the sample rack.
The analyzer contained a power supply, readout circuitry, alarm
circuits, and operating controls. Dual set points were provided
to signal the detection of combustible gas when it reached 25
percent of the LEL, and again when 50 percent of the LEL was de-
tected.
Alarm switches announced alarm levels of combustible gases in
the stages and triggered automatic responses in the control cir-
cuits .
Stages 1 and 4 LEL Recorders
A two-pen Honeywell strip chart recorder on the MIP recorded
LEL levels for Stages 1 and 4 on a single 0-100 linear chart, re-
presenting percent of LEL for hexane, the hydrocarbon for which
the analyzer was calibrated.
Vent Gas Flow Meter and Totalizer
An Eastech, Inc. 8-cm (3-in.) vortex shedding flow meter was
provided with an Eastech signal conditioner and totalizer to
measure vent gas flow from Stage 4. The meter body and bluff
body were of 316 stainless steel, and the 0-rings of Viton.
Influent Flow Meter and Transmitter
This instrument was a Brooks Inscruments electromagnetic
flow-sensing device with an integral signal converter and located
in the influent line leading to the aeration tank. The flow tube
was neoprene-lined stainless steel, and the 316 stainless steel
71
-------
electrodes were equipped with automatic mechanical electrode
cleaners. The converter produced a linear output signal of 4-20
ma. The meter had a flow measuring capacity of 94.6 mVmin
(25,000 gpm).
Influent Flow Controller
This controller was a Research, Inc. "Data-Trak" programmer
which could provide a 24-hr cycle by varying the output signal
according to the time of the day. It was equipped with a Re-
search, Inc. Model 607-R1000 "Match-Pack" signal transducer
which converted the 0-1000-ohm output to a 4-20-ma output signal.
Influent Flow Totalizer
This influent flow totalizer consisted of: a Honeywell linear
integrator and a Honeywell totalizing counter with mechanical
reset, both located on the MIP. The integrator accepted a 1-5-
volt-dc input signal linearly proportional to the influent flow
and produced 25-volt output pulses to drive the totalizing
counter.
Influent Flow, Return Sludge Flow, and Waste Sludge Flow Recorders
A three-pen Honeywell strip chai t recorder on the MIP re-
corded the flow rates of the three system streams on a single
chart.
Return Sludge Flow Meter and Transmitter
A 57-m3/min (15,000-gpm) Brooks Instruments electromagnetic
meter with integral signal converter was provided to monitor re-
turn sludge flow. The tube was neoprene-lined 304 stainless
steel and the electrodes 316 stainless? steel equipped with auto-
matic mechanical electrode cleaners.
Re turn Sludge Flow Controller
This unit was a Honeywell electronic deviation-indicating
cascade controller which accepted both a 1-5-volt-dc signal in-
put proportional to the return sludge flow and a 1-5-volt-dc,
remote, set-point input from a ratio controller for cascade
operation. The controller could be operated in manual, auto-
matic, or cascade modes.
Return Sludge Flow Ratio Controller
A Honeywell electronic ratio/bias device receiving a 1-5-
volt-dc input signal and producing a 4-20-ma output signal was
included with the return sludge flow control package.
72
-------
W=aste Sludge Flow Meter and Transmitter
The waste sludge flow meter was a 1.1-m /min (300-gpm) Honey-
well electromagnetic flowhead with integral signal converter and
a neoprene-lined 304 stainless steel tube and 316 stainless steel
electrodes equipped with ultrasonic cleaners.
Waste Sludge Flow Totalizer
The waste sludge flow totalizer consisted of a Honeyv/ell
linear integrator and a Honeywell totalizing counter with me-
chanical reset, both located on the MIP. The integrator accepted
1-5-volt-dc input signal linearly proportional to the waste
sludge flow and produced 25-volt output pulses to drive the
totalizing counter.
Sample Compositor System
Two Union Carbide flow-proportional samplers, one for the
influent from the grit chamber and the other for the settling
tank effluent, were used for collecting system influent and
effluent samples. During each sampling action, a 3-m.l portion
was taken from the respective stream and deposited into a re-
frigerated jug. The device varied the rate at which the portions
were taken in accordance with an internal rate control adjust-
ment and an external flow ratio control signal. The external
signal required was 5-20 ma with the 5-ma signal producing high
sampling rates (6 ml/min maximum).
Sample Compositor Flow Ratio Controller
This segment of the sample compositor system was composed of
a Honeyv/ell electronic ratio-bias station that received a 1-5-
volt-dc input signal and produced a 4-20-ma output signal. The
ratio of the output signal to the input signal was set by the
vertical thumbwheel and could be varied from 0.3 to 3.0.
73
-------
APPENDIX B
ESTIMATED CAPITAL COST FOR CONVERSION OF THE
NEWTOWN CREEK PLANT TO OXYGEN ACTIVATED SLUDGE
ThP EPA Project Officer requested that an estimate be _
added to this report of the capital cost that would be required
if the entire Newtown Creek plant were converted to oxygen
activate^ sludge. Design and estimated cost data are based on
?,he UNOX system and were provided by Union Carbide Corporation.
DESIGN BASIS
Since the termination of data collection on the UNOX
system demonstration module, the Manhattan pumping sta ion has
been completed, bringing an approximate 5.3 mVsec (120 mgd) of
very dilute wastewater to the Newtown Creek plant via a force
main under the East River. The net effect of this additional
contribution to the plant's influent has been a significant
reduction in wastewater strength. Raw wastewater character-
istics during the project and after addition of the Manhattan
sewage are compared in Table 19.
'TABLE 19. COMPARISON OF NEWTOWN CREEK WASTEWATER
CHARACTERISTICS BEFORE AND AFTER
INTRODUCTION OF MANHATTAN FLOW
Characteristics During
the Project:
Brooklyn and Queens
Flows Only (range of
monthly averages)
Current Character-
istics: Brooklyn,
Queens, and Manhat-
tan Flows (typical
values)
BOD (mg/1)
COD (mg/1)
Suspended Solids
(mg/1)
150 - 240
290 - 365
100 - 160
130
260
100
The current flow of 5.3 m3/sec (120 mgd) from Mannattan is
far less than the original estimate of 7.4 m3/sec (170 mgd) For
thi- design exercise, the average dry-weather hydraulic loading
selected fSr the full-plant conversion to oxygen at Newtown Creek
74
-------
is 13 4 mVsec (307 mgd). This allows for a modest increase in
the Manhattan flow to about 5.9 n^/sec (135 mcj«3) . Combined
with an average BOD of 130 mg/1,the design average BOD loading
is, therefore, 150,980 kg/day (332,850 Ib/day).
OXYGEN SYSTEM DESCRIPTION AND DISCUSSION
Since the original design and construction of the UNOX
System demonstration plant at Newtown Creek, the technology
base for oxygenation system process and equipment design hao
been greatly expanded through extensive research and development
programs, equipment testing and vendor development programs,
and design and operating experiences on the more than 100
oxvqenation systems currently in operation around the world.
in fact, the previously-cited problems experienced with equipment
and instrumentation during the Newtown Creek demonstration
project initiated many of the programs mentioned above. As a
result of these efforts, many improvements have been made in
equipment reliability and performance as well as in operating
cower efficiency of oxygenation systems over the last 6 yr. The
Sygenation system described below, which is proposed today for
upgrading of the entire Newtown Creek facility, bears little
resemblance to the original demonstration system in terms of
the specific mass transfer, oxygen generation, and control
instrumentation equipment included in the system. However the
functional definition of the oxygenation system and the process
parameLrs used to design the system remain entirely consistent
with the system tested and performance results obtained during
the Newtown Creek demonstration project.
Based on the organic loading defined above and the Process
performance data gathered during the testing program, a UNOX
svstem to treat the Newtown Creek wastewater would require
conversion of 14 of the 16 existing aeration tanks to oxygen
Service Each tank would be modified to contain four equal-
volume stages operating in series Each .^age would be 15.2 m
lona x 16.8 m wide x 4.6 m liquid depth (50 ft x 5^ ft x 15 It)
and contain 1168 m3 (308,590 gal) of mixed Ixquor. The total
oxyqenation volume would be 65,410 m3 (17.28 mil gal). One
surface aerator would be provided in each stage for mixing and
mass transfer. Each first stage would have an aerator with an
installed nameplate rating of 44.7 kW (60 nameplate hp (nnp)).
The second staqes would be equipped with aerators having name-
plate Stings of 29.8 kW (40 nhp). The third and fourth stages
would be equipped with aerators having nameplate ratings of
22 \ kW (30 nhp). Thus, each aeration tank would contain four
surface aerators with a total nameplate power of 119.3 kW (160
n"Pf and the total system installed power for oxygen dissolution
would be 1670 kW (2240 nhp) at 480 volts.
75
-------
The surface aeration mass transfer equipment proposed here
for upgrading the Newtown Creek facility is substantially
simpler and in much broader use for similar applications than
the submerged turbine system originally tested during the
Newtown Creek demonstration project. Employing surface aerators
would eliminate many of the pieces of equipment which were sources
of mechanical problems during the testing program, such as the
gas recirculation compressors and the rotary unions connecting the
oxygen gas piping to the mixer sparger unit. A substantial
amount of oxygen piping would also be eliminated. Furthermore,
use of surface aeration equipment would eliminate or minimize
the problems caused during the testing program due to foaming
since the high level of surface turbulence and mechanical
agitation would disperse the foam if it tended to form.
The existing UNOX system demonstration module (Aeration
Tank No. 9), which utilized the submerged turbine equipment,
would either be dismantled or "mothballed". This tank along
with one other aeration tank (not specified for this exercise)
would not be incorporated in the upgraded oxygen system design,
but their companion final tanks would be used to keep secondary
clarifier overflow rates as low as possible. The aerated grit
chambers mated with the two aeration tanks to be removed from
service would also be retained to maximize grit removal.
Figure 1 presented previously in Section 1 illustrates the
plant layout and the interrelationship of the various tank
complexes.
Modifications to the existing tankage would include: (1)
removal of all air diffusers and associated piping from the
aeration tanks, (2) installation of three interstage partitions
per aeration tank with openings to permit flow of mixed liquor
and gas from stage-to-stage, (3) installation of gas-tight
covers for the aeration tanks to retain the oxygen rich gas,
(4) installation of a new sludge recycle pump station, and (5)
improvements to the 16 aerated grit chambers and 16 final settling
tanks. The 0.9-m (3-ft) freeboard available in the 5.5-m (18-
ft) deep aeration tanks would provide sufficient gas space for
the surface aerators employed in UNOX designs.
A cryogenic oxygen generation plant having a production
capacity of 171 metric tons/day (188 tons/day) of 95-99 percent
pure oxygen would be the recommended unit to supply oxygen feed
gas to the oxygenation system. Two 189-m3 (50,000-gal) DRIOX
storage tanks with 434 metric tons (478 tons) of oxygen capacity
would be recommended for back-up liquid oxygen supply, which is
sufficient for 2.5 days of operation at design average-load
conditions when the cryogenic oxygen generation plant required
maintenance or repairs. At the design operating point, the
above cryogenic unit would be capable of supplying 1.10 kg
76
-------
oxygen/kg BOD removed at the design oxygen utilization rate of
90 percent and an effluent soluble BOD concentration of 10 mg/1.
This value is entirely consistent with the experience during the
demonstration project. The back-up liquid oxygen supply would
be used to meet diurnal oxygen demand peaks above 171 metric
tons/day (188 tons/day).
Although years of testing and design improvements have
corrected the ti>s-chanical problems and significantly improved the
performance efficiency of the Pressure Swing Adsorption (PSA)
oxygen generation system relative to the experiences with the
early version of this unit tested in the Newtown Creek demon-
stration, the magnitude of the oxygen requirements for a full-
scale upgrading at Newtown Creek make use of today's version of
the PSA system economically impractical. It is for this reason
that the larger capacity cryogenic oxygen generation system is
included in this cost estimating exercise.
The installed power of the oxygen supply system, including
that for liquid oxygen vaporization, would be 2610 kW (3500 nhp)
at 2300 volts. Adding the above specified connected dissolution
power plus 37 kW (50 nhp @ 480 volts) for miscellaneous power
needs, the total installed power is estimated at 4317 kW (5790
nhp) .
By way of comparison to the Newtown Creek demonstration
oxygen system which, produced a power efficiency at or above
design loadings of 0.95 kWh/kg BOD removed (0.58 nhp/lb BOD),
the system described above for the proposed upgrading at Newtown
Creek would have an overall power efficiency on the basis of
installed power and, assuming an effluent total BOD of 20 mg/1,
of 0.81 kWh/kg BOD removed (6.49 nhp/lb DOD). This represents
a projected efficiency improvement over the demonstration
system of approximately 15 percent at design loadings and is
the result of improvements in oxygen dissolution technology
over the last 7-8 yr and the use of the more efficient cryo-
genic oxygen generation system. It should further be noted that
the actual line operating power draw at design conditions for
this system would be approximately 4 to 5 percent less than the
installed nameplate power ratings of the equipment previously
discussed.
Based on the past oxygen system demonstration experiences
at Newtown Creek, it is anticipated that the return sludge sus-
pended solids concentration for the proposed system would average
about 15,000 mg/1. At an assumed return sludge f.ow equal to
40 percent of the influent flow, and MLSS concentration vould
approximate 4285 mg/1. During the demonstration project, the
volatile fraction of the mixed liquor suspended solids averaged
81 percent. Due to the decreased strength of Newtown Creek's
wastewater since addition of the Manhattan sewage, it is possible
77
-------
^ ^ UNITED STATES ENVIRONMENTAL'PROTECTION AGENCY
•*'•••-. . ' " •• - 8EHQ-0483-Q475
OATBi ADD O Q IQOO . . ' .. Page 1 of 4
lU IV £• J lijV^sJ
status Report* 8EHQ-0483-0475 Approved .
/"V, I Revision
MOHt Justine L. We^c^f/Wsam Leader . Needed
Chemical Selepfion and Profiles' Team/CHIB ..
t
/jflr \
TOi Frank D. KoveV, Chief •"'••'.'
Chemical Hazard Identification Branch/AD
Submission Description
The Eastman Kodak Company submitted a March 2, 1983, interim
summary report entitled "Inhalation Teratologies! Potential of
Ethylene Glycol Monobutyl Ether [EGBE; 2-butoxyethanol] in -the
Rat." Eastman Kodak reported that the stuuy is being conducted
under contract for the Chemical Manufactures Association (CMA)
and .is being sponsored by the Glycol Ether Program•(GEP) of which
the Tennessee Eastman Company, (a division of the.Eastman Kodak
Company) is a member. In its submission, Eastman Kodak noted
that the data contained in the interim report "have not been
subjected to .Quality Assurance and are incomplete since fetal
skeletal examination and statistics have not been finished."
According to the submitted interim report, 4 groups of 30 mated
female Fischer 344 rats were exposed via inhalation to EGBE vapor
concentrations of 0, 100, 200, or 300 ppm for 6 hours per day on .
gestation days (GD) 6 through 15. In addition, the interim
report presented the following summarized clinical.observations
and results of examinations performed as of March 2, 1983:
"Gestation Body Weight Changes; Gestational body weight gains
of animals exposed to 300 ppm were significantly depressed
(compared to controls) from GD 6-21. Significant reductions
in body weight gains (compared to controls) were also present
in the 200 ppm group from GD. 6-15 and in the 100 ppm qroup
from GD 6-12.".
Gestational Food Consumption; Food consumption was reduced
during the dosing period in all groups exposed to EGBE. Sig-
nificant reductions in food consumption (compared to controls)
were noted in the 300 ppm group from GD 6-15, in the 200 ppm
group from GD 6-12, and in the 100 ppm group from GD 6-12'.
. "Animals in the 300 and 200 ppm groups had a significant in-
crease in' their food consumption (compared to controls) from
the end of exposures (GD 15) until sacrifce (GD 21).
*NOTE: This status report is the result of a preliminary
staff'evaluation of information submitted to EPA. Statements
irade herein are not to be regarded as expressino final
Agency policy or intent with respect to this particular
chemical. Any review 6f the status report should take into
consideration the fact that it may be based on incomOrete
information.
, ^ 000015
-------
^ • - • • • 8EHQ-0483-0475 ^
* " - - Page 2 of 4
•'• - > \.
Gestational Water Consumption; Only animals exposed to 300
and 200 ppm had significant reductions in water consumption.
Those exposed to 300 ppra consumed less water than controls
from GD 6-12. Animals in the 200 ppra group consumed less
water than controls in the periods GD 6-9 and GD 12-15.
Water consumption prior, to the exposure segment of the study
and following exposures were similar in all groups.
Maternal Clinical Observations; Exposures to EGBE resulted
in a dose-related increase in the frequency of signs indica-
tive of hematuria. These signs included red fluid on cage
trays, and urogenital wetness, discharges and encrustations
of red and/or black color. In addition, animals exposed to
EGBE were hypoactive during the exposure segment of the study
when compared with controls.
*
Maternal Organ Weights; The weight of the uteri of animals
exposed to 300 ppm was significantly lower than those of con-
trols. This reduction in uterine weight is associated with
the high degree, of embryo and fetal lethality present at this
exposure level*
Maternal Uterine and Ovarian Examinations; Exposures to
EGBE, particularly at the 300 ppm level, resulted in a marked
increase in embryo and fetal lethality manifested by resorp-
tions of concepti. Exposures to 200 ppm produced a^more
moderate increase in resorption when compared to controls,
Fetal Visceral Examinations; Exposures to EGBE resulted in a
dose-related increase in fetal atelectasis [the incomplete
expansion of the lung(s) at birth]. In addition, fetuses
from groups exposed to EGBE had ventricular septal defects,
absent innominate arteries, and severely shortened innominate
arteries."
In addition to the March 2, 1983, interim report, the Eastman
Kodak Company provided the following background information;
"Nelson _et^.al. at the National Institute for Occupational
Safety and Health (NIOSH) studied the teratogenic potential
of three glycol ethers in rats; 2-biitoxyethanol (EB) , 2-meth-
oxyethanol (EM), and 2-ethoxyethyl acetate (EEA). They found
that EM and EEA were teratogenic, while 2-butoxyethanol did
not produce significant embryofetotoxicity. The exposure
concentrations of 2-butoxyethanol were 150 to 200 ppm and
Nelson et^ al^ reported that these concentrations were diffi-
cult to generate (vapor pressure of EB = 0.6 mm Hg at 20°C).
Maternal toxicity was evident at both dose levels and consis-
ted of hematuria onl;r after the first exposure. No other ad-
verse effects' were seen in the dams exposed to either level.
No major skeletal or soft tissue malformations were seen at
150 or 200 ppm of 2-butoxyethanol and the incidence of common
skeletal variants and minor soft tissue anomalies seen in the
treated groups were comparable to the controls. Potential
000016
-------
•£/--"'. - • 8HQ-0483-0475 /
»v - Page 3 of 4
reproductive effects of 2-butoxyethanol and other glycol
ethers were studied in male rats in [the submitter's] labora-
tory and in male mice by Nagano jst^ al^ in Japan. Administra-
tion was by gavage five times per week. In the former, doses
from 222 to 885 mg/kg/day, and in the latter, doses from 250
to 1000 mg/kg/day, did not produce testicular effects."
Finally, Eastman Kodak noted that the following factors should be
considered by EPA in assessing the CMA/GEP-sponsored inhalation
teratology study of 2-butoxyethanol in rats: "the severe maternal
toxicity produced, the lack of a dose-related response and the
low incidence of effects seen in the [CMA/GEP] study, as well as
the negative results from the NIOSH study, the known toxicity of
2-butoxyethanol in adult male and non-pregnant female rats, and
[the chemical's] vapor pressure."
Submission Evaluation
Although the submitted findings indicate that 2-butoxyethanol did
produce some degree of maternal tibxicity in the CMA/GEP-sponsored
teratology study, EPA does not be'lieve that 2-butoxyethanol pro-
duced "severe" maternal toxicity in the study. One of the most
common major manifestations of severe maternal toxicity in this
type of study is a significant weight change (usually seen as a
decrease) in exposed versus control dams. The summarized data-
that are presented in Table 4 (which contains measured maternal
organ weights and "corrected" maternal body weights (i.e., the
maternal body weight minus the weight of ths uterus and its
contents)), indicate that there were no significant changes in
the corrected body weights of exposed dams when compared to the
corrected body weights of control dams. There was, however, a
significant decrease found in the uterine weights of dams exposed
to 300 ppm 2~butoxyethanol when compared to controls. Stated in
another way, the apparent decreases in maternal body weights of
dams exposed to 300 ppm 2-butoxyethanol are due primarily to the
decrease in the weight of the uterus and its contents.
In order for EPA ho more properly evaluate the significance of
the embryo/fetotoxic effects (including abnormalities) observed
.1-n the CMA/GEP-sponsored inhalation study of 2-butoxyethanol, a
complete copy of the final report (including protocol(s), data
and the results of performed statistical analyses) should be ob-
tained. It should be noted at the present time, however, that
the embryo/fetotoxic effects observed in the CMA/GEP-sponsored
study (although in apparent contrast to previous findings), are
consistent with the adverse developmental effects found in labor-
atory animals exposed to two structurally related chemicals: 2-
methoxyethanol and 2-ethoxyethanol.
Current Production and Use
A review of the production range (includes importation volumes)
statistics for 2-butoxyethanol (CASJ No. 111-76-2), which is
listed in the initial TSCA Inventory, has shown that between 31
000017
-------
Jr . " - 8EHQ-0483-0475
" "k ' Page 4 of 4
million and 161 million pounds of this chemical were reported as
produced/imported in 1977. This production range information
does not include any production/importation data claimed as con-
fidential by the person(s) reporting for the TSCA Inventory, nor
does it include any information which would compromise Confiden-
tial Business Information (CBI). The data submitted for the TSCA
Inventory, including production .range information, are subject to
the limitations contained in the Inventory Reporting Regulations
(40 CFR 710).
2-Butoxyethanol is used as a solvent for nitrocellulose resins,
spray and quick-drying lacquers, varnish and varnish removers,
enamels, and drycleaning compounds. The chemical is also used
for preventing spotting during the printing and/or dyeing of
textiles and as an ingredient in certain pesticides.
Comments/Recommendations
EPA's Office of Toxic Substances (OTS) has received and evaluated
several Section 8(e) and "For Your Information (FYI)" submissions
concerning various glycol ethers. The Chemical Hazard Identifi-
cation Branch (CHIB/AD/OTS) has prepared CHIPs (Chemical Hazard
Information Profiles) on 2-methoxyethanol, 2-ethoxyethanol, and
their acetates. In addition, Priority Review Level-1 (PRL-1)
documents on 2-methoxyethanol and 2-ethoxyethanol have been pre-
pared by the Health and Environmental Review Division (HERD/OTS).
At present, the OTS "Existing Chemicals Task Force (ECTF)" is
considering various EPA options with regard to chemicals within
the class of .glycol ethers.
a) The Chemical Hazard Identification Branch will request the
Eastman Kodak Company to submit, when available, a complete
copy of the final report from the CMA/GEP-sponsored inhala-
tion teratology study of 2-butoxyethanol in rats. The sub-
mitter will also be requested to provide a complete copy of
the final report (including test protocols and data) from
the Eastman Kodak study reportedly conducted to determine
potential reproductive system effects of 2-butoxyethanol
administered to male rats by gavage five (5) times por week
at doses ranging from 222 to 885 mg/kg/day.
b) In view of EPA's general interest in company actions that
are taken on a voluntary basis in response to chemical tox-
icity/exposure information, the Chemical Hazard Identifica-
tion Branch will request the Eastman Kodak Company to des-
cribe the actions it has taken to warn workers and others
and to reduce and/or eliminate exposure to .2-butoxyethanol.
c) The Chemical Hazard Identification Branch will transmit a
copy of this status report to NIOSH, OSHA, CPSC, FDA, NTP,
0/..NR/EPA, ORD/EPA, OSWER/EPA, OW/EPA, OPP/OPTS/EPA, and the
OTS "Existing Chemicals Task Force (ECTF)." A copy of this
report will also lie provided to EPA's Industry Assistance
Office (IAO/OTS) for further distribution.
000018
-------
that the volatile fraction might drop to as low as 75 percent
in a full-scale system. The MLVSS concentration for design
purposes is, therefore, assigned a value of 3215 mg/1 (75
percent of MLSS), Using the above assumptions, expected oxygen
system operating conditions are summarized in Table 20 for a
design flow of 13.4 m3/sec (307 mgd) and a design BOD loading
of 150,980 kg/day (332,850 Ib/day). All other process design
parameters except as mentioned above are consistent with process
performance experiences during the Newtown Creek demonstration
project.
TABLE 20. DESIGN OPERATING CONDITIONS FOR FULL-
SCALE NEWTOWN CREEK OXYGEN SYSTEM
Oxygenation Tank Detention Time
Based on Q (hr) 1.35
Based on Q + R (hr) 0.96
F/M Loading (kg BOD/day/kg MLVSS) 0.72
Volumetric Organic Loading
(kg BOU/day/m3) 2.31
(Ib BOD/day/1000 ft3) 144
Secondary Clarifier Overflow Rate
(m3/day/m2} 36
(gpd/ft2) 872
Sludge Wasted
Based on Figure 8 (non-filamentous
curve) and assumed effluent sus-
pended solids of 20 mg/1
(kg/day) 94,350
(Ib/day) 208,000
Sludge Retention Time
"Based on Figure 8 (days) 2.4
Average Mixed Liquor DO (mg/1) 6
Average 02 Utilization Efficiency (%) 90
Design Oxygen Supply (metric tons/day) 171
(US tons/day) 188
Effluent BOD (mg/1) 20
Effluent Soluble BOD (mg/1) 10
Effluent Suspended Solids (mg/1) 20
78
-------
ESTIMATED CONSTRUCTION AND CAPITAL COSTS
The estimated construction cost of converting the Newtown
Creek plant to oxygen is $44,000,000. Adding anticipated fees
for engineering, legal, fiscal, and administrative services and
interest over an assumed construction period of 2 yr, the
estimated capital (total project) cost becomes $54,335,000. At
a design flow of 13.4 m^/sec (307 mgd), this capital cost is
equivalent to $46.76/daily m^ ($0.10/daily gal) of upgraded plant
capacity. The above estimate, which is applicable to the fourth
quarter, 1977, and includes provisions for major site work and
improvements to existing plant grit removal, clarification, and
sludge recycle facilities in addition to oxygen system equipment,
installation, and retrofit costs, is broken down by cost element
in Table 21.
TABLE 21. ESTIMATED CAPITAL COST FOR OXYGEN
SYSTEM RETROFIT AT NEWTOWN CREEK
Oxygen Dissolution Equipment $ 4,375,000
Installation 485,000
Oxygen Supply Equipment 3,225,000
Installation 1,025,000
Site Work (including pilings) 3,000,000
Aeration System Retrofit Modifications 16,000,000
Secondary Clarifier Modifications 6,000,000
Grit Chamber Modifications 2,890,000
New Sludge Recycle Pump Station 3,OOP , OOP
Subtotal 40,000,000
Contingency @ 10% 4,000,000
ESTIMATED CONSTRUCTION COST $44,000,000
Engineering, Legal, Fiscal,
Administrative 9 16.5% 7,260,000
Subtotal 51,260,000
Interest During Construction @ 6% for 2 yr 3,075,000
ESTIMATED CAPITAL COST $54,335,000
79
------- |