EPA-600/2-77-108
August 1977
Environmental Protection Technology Series
PILOT PLANT EVALUATION OF
ALTERNATIVE ACTIVATED SLUDGE SYSTEMS
Municipal Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
EPA-600/2-77-108
August 1977
PILOT PLANT EVALUATION OF
ALTERNATIVE ACTIVATED SLUDGE SYSTEMS
by
James A. Heidman
EPA-DC Pilot Plant
Washington, D.C. 20032
Contract No. 68-03-0349
Project Officer
Irwin J. Kugelman
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. Noxious air, foul water, and spoiled
land are tragic testimony to the deterioration of our natural environment.
The complexity of that environment and the interplay between its components
require a concentrated,and integrated attack on the problem.
Research and development is that necessary first step in problem
solution and it involves defining the problem, measuring its impact, and
searching for solutions. The Municipal Environmental Research Laboratory
develops new and improved technology and systems for the prevention, treat-
ment, and management of wastewater and solid and hazardous waste pollutant
discharges from municipal and community sources, for the preservation and
treatment of public drinking water supplies, and to minimize the adverse
economic, social, health, and aesthetic effects of pollution. This pub-
lication is one of the products of that research; a most vital communications
link between the researcher and the user community.
The study summarized in this report evaluates the process character-
istics and performance of alternative activated sludge systems commonly
used for municipal wastewater treatment.
Francis T. Mayo, Director
Municipal Environmental
Research Laboratory
-------
ABSTRACT
A step feed, a complete mix and a simulated plug flow activated sludge
system were operated over a wide range of process loadings. Information on
equilibrium characteristics and performance under steady state conditions at
the various loadings was obtained. Results obtained from additional complete
mix systems which were not operated as part of this study but which provided
relevant information were also summarized. All steady state information was
obtained under constant flow conditions although the particular flow rates
evaluated produced some variations in the hydraulic detention times. The
flow to each of the pilot plant activated sludge systems was generally
from 114 to 189 m3/day (30,000 to 50,000 gpd) depending upon the particular
unit and loading being examined. Information was collected on influent and
effluent concentrations of BOD, COD, TKN, (N02 + N03J-N, P, SS and VSS. The
relation of process loading to SRT was examined for all systems. Mixed liquor
settling velocities and process SVI's were also determined. The P and TKN
content and COD:SS and COD:VSS ratios of the process solids were measured in
selected cases.
This report was submitted in fulfillment of Contract No. 68-03-0349 by
the District of Columbia Pilot Plant under sponsorship of the U.S. Environ-
mental Protection Agency. This report covers the period from October 1973
to September 1975.
-------
CONTENTS
Foreword i -j j
Abstract iv
List of Figures vi
List of Tables viii
Acknowledgments xii
1. Introduction 1
2. Conclusions 2
3. Recommendations 4
4. Background Considerations 5
5. Experimental Systems 22
System Reactors and Clarifiers 22
Influent Flow 24
Sludge Wasting 24
Dissolved Oxygen Control 26
6. Methods and Procedures 27
7. Results 29
Laboratory Studies 29
General 32
Step Feed System 32
Plug Flow System 51
Complete Mix System 66
Related Complete Mix Systems 91
Soluble Effluent Quality 102
Sludge Settling Characteristics 108
a. Discussion 130
9. References 154
-------
LIST OF FIGURES
No.
1. Schematic Diagram of a Completely Mixed Activated Sludge Process. 7
2. Schematic Diagram of the Step Feed, Plug Flow and Complete Mix
Activated Sludge Systems. 23
3. Experimental and Theoretical Dilute-Out Concentrations of Na
in the Complete Mix System. 25
4. Variation of 4th Pass SVI's in the Step Feed System. 34
5. Photomicrograph of Foam Caused by Nocardia. 40
6. Influent and Reactor TOC and Suspended Solids Concentrations for
the Chemostat on June 4, 1975. 88
7. Influent and Reactor TOC and Suspended Solids Concentrations for
the Chemostat on June 17-18, 1975. 90
8. Solids Concentrations in the Chemostat and Plug Flow System on
August 12-13, 1975. 92
9. BOD, COD and TOC Concentrations in the Influent, Chemostat and
Plug Flow System Effluent on August 12-13, 1975. 93
10. Mixed Liquor Settling Velocities in the 4th Pass of the Step Feed
System as a Function of Suspended Solids Concentrations. 119
11. Mixed Liquor Settling Velocities in the Plug Flow System as a
Function of Suspended Solids Concentrations. 120
12. Mixed Liquor Settling Velocities in the Complete Mix System as a
Function of Suspended Solids Concentrations. 121
13. Mixed Liquor Settling Velocities During Steady State Operation
with the Step Feed, Plug Flow and Complete Mix Systems. 123
14. Batch Test Settling Velocities for the Step Feed System. 124
15. Batch Test Settling Velocities for the Plug Flow System. 125
16. Batch Test Settling Velocities for the Complete Mix System. 126
VI
-------
LIST OF FIGURES
(Continued)
No. Page
17. Effect of Suspended Solids Concentrations on SVI's for the Step
Feed and Plug Flow Systems. 127
18. Effect of Suspended Solids Concentrations on the SVI's for the
Complete Mix System. 128
19. F/M. Ratio Based on g BOD5 Applied per g MLVSS vs Process SRT. 131
20. F/M Ratio Based on g COD Applied per g MLVSS vs Process SRT. 132
21. F/M Ratio Based on BOD5 and COD Applied vs 1/SRT. 134
22. Variation in COD of D.C. Primary Effluent During a 5-day Period
(January 26-30, 1975). 139
23. Chronological Order of Process SRT's During Steady State
Operation. 145
24. Estimate of Soluble COD Residuals for Selected Systems During
August and September 1974. 147
25. Estimate of Soluble COD Residuals for Selected Systems During
November and December 1974. 149
26. Estimate of Soluble COD Residuals for Selected Systems During
June and July 1974. 150
-------
LIST OF TABLES
No. Page
1. Sludge Production Based on Results of Wuhrmann (9). 12
2. Solids Production as a Function of Process Loading and Type of
Flow. 17
3. COD and Soluble COD Values of D.C. Primary Effluent Before and
After Refrigerated and Acidified Storage. 30
4. Total and Filtered COD Values of D.C. Primary Effluent Grab
Samples Obtained at 0800-0900 Hours. 31
5. Clarifier Bed Levels and Stability With the Step Feed System. 33
6. Mixed Liquor Settling Velocities in the Step Feed System. 35
7. Process Characteristics for the Step Feed System at a 9.0 Day
SRT (Feb. 1 - March 7, 1974). 38
8. Influent and Effluent Characteristics for the Step Feed System
at a 9.0 Day SRT (Feb. 1 - March 7, 1974). 39
9. Process Characteristics for the Step Feed System at an 8.0 Day
SRT (April 16 - May 7, 1974). 42
10. Influent and Effluent Characteristics for the Step Feed System
at an 8.0 Day SRT (April 16 - May 7, 1974). 43
11. Process Characteristics for the Step Feed System at a 5.9 Day
SRT (Aug. 11 - Sept. 4, 1974). 44
12. Influent and Effluent Characteristics for the Step Feed System
at a 5.9 Day SRT (Aug. 11 - Sept. 4, 1974). * 45
13. Process Characteristics for the Step Feed System at a 4.1 Day
SRT (Sept. 18 - Oct. 19, 1974). 46
14. Influent and Effluent Characteristics for the Step Feed System
at a 4.1 Day SRT (Sept. 18 - Oct. 19, 1974). 47
15. Process Characteristics for the Step Feed System at a 3.7 Day
SRT (Nov. 12 - Dec. 12, 1974). 48
16. Influent and Effluent Characteristics for the Step Feed System
at a 3.7 Day SRT (Nov. 12 - Dec. 12, 1974). 49
viii
-------
LIST OF TABLES
(Continued)
No. Page
17. Summary of System Operation and Performance With the Step Feed
System. 50
18. Process Characteristics for the Plug Flow System at a 6.6 Day
SRT (April 7-25, 1974). 52
19. Influent and Effluent Characteristics for the Plug Flow System
at a 6.6 Day SRT (April 7-25, 1974). 53
20. Process Characteristics for the Plug Flow System at a 4.4 Day
SRT (June 1 - July 11, 1974). 55
21. Influent and Effluent Characteristics for the Plug Flow System
at a 4.4 Day SRT (June 1 - July 11, 1974). 56
22. Process Characteristics for the Plug Flow System at a 2.9 Day
SRT (Aug. 16 - Sept. 12, 1974). 57
23. Influent and Effluent Characteristics for the Plug Flow System
at a 2.9 Day SRT (Aug. 16 - Sept. 12, 1974). 58
24. Process Characteristics for the Plug Flow System at a 1.9 Day
SRT (Sept. 24 - Oct. 21, 1974). 59
25. Influent and Effluent Characteristics for the Plug Flow System
at a 1.9 Day SRT (Sept. 24 - Oct. 21, 1974). 60
26. Process Characteristics for the Plug Flow System at a 4.7 Day
SRT (Nov. 7 - Dec. 14, 1974). 62
27. Influent and Effluent Characteristics for the Plug Flow System
at a 4.7 Day SRT (Nov. 7 - Dec. 14, 1974). 63
28. Process Characteristics for the Plug Flow System at a 5.7 Day
SRT (March 14 - April 14, 1975). 64
29. Influent and Effluent Characteristics for the Plug Flow System
at a 5.7 Day SRT (March 14 - April 14, 1975). 65
30. Process Characteristics for the Plug Flow System at a 3.5 Day
SRT (June 15 - July 8, 1975). 67
31. Influent and Effluent Characteristics for the Plug Flow System
at a 3.5 Day SRT (June 15 - July 8, 1975). 68
32. Summary of System Operation and Performance With the Plug Flow
System. 59
ix
-------
LIST OF TABLES
(Continued)
No.
Page
33. Process Characteristics for the Complete Mix System at a 2.6 Day
SRT (June 7-30, 1974). 73
34. Influent and Effluent Characteristics for the Complete Mix System
at a 2.6 Day SRT (June 7-30, 1974). 74
35. Process Characteristics for the Complete Mix System at a 2.1 Day
SRT (Aug. 9-31, 1974). 75
36. Influent and Effluent Characteristics for the Complete Mix System
at a 2.1 Day SRT (Aug. 9-31, 1974). 76
37. Process Characteristics for the Complete Mix System at a 1.8 Day
SRT (Sept. 11 - Oct. 10, 1974). 77
38. Influent and Effluent Characteristics for the Complete Mix System
at a 1.8 Day SRT (Sept. 11 - Oct. 10, 1974). 78
39. Process Characteristics for the Complete Mix System at an 8.1 Day
SRT (Dec. 3, 1974 - Jan. 16, 1975). 80
40. Influent and Effluent Characteristics for the Complete Mix System
at an 8.1 Day SRT (Dec. 3, 1974 - Jan. 16, 1975). 81
41. Process Characteristics for the Complete Mix System at a 7.1 Day
SRT (Feb. 2-25, 1975). 82
42. Influent and Effluent Characteristics for the Complete Mix System
at a 7.1 Day SRT (Feb. 2-25, 1975). 83
43. Process Characteristics for the Complete Mix System at a 1.5 Day
SRT (April 25 - May 12, 1975). 85
44. Influent and Effluent Characteristics for the Complete Mix System
at a 1.5 Day SRT (April 25 - May 12, 1975). 86
45. Influent, Effluent and Process Characteristics for the Chemostat
(May 20 - June 24, 1975). 87
46. Change in Soluble Reactor TOC After Additional Aeration. 89
47. Summary of System Operation and Performance With the Complete Mix
System. 94
48. Process Characteristics for the 12.72 m3 Complete Mix System at a
3.9 Day SRT (March 19 - April 30, 1974). 97
-------
LIST OF TABLES
(Continued)
No. Page
49'. Influent and Effluent Characteristics for the 12.72 m3 Complete
Mix System at a 3.9 Day SRT (March 19 - April 30, 1974). 98
50. Process Characteristics for the 14.91 m3 Complete Mix System at
a 5.3 Day SRT (March 19 - April 14, 1974). 99
51. Influent and Effluent Characteristics for the 14.91 m3 Complete
Mix System at a 5.3 Day SRT (March 19 - April 14, 1974). 100
52. Selected Characteristics for the 12.72 m3 Complete Mix System
During March 19 - April 14, 1974. 101
53. Process Characteristics for the 14.91 m3 Complete Mix System
at an 8.4 Day SRT (June 23 - July 11, 1974). 103
54. Influent and Effluent Characteristics for the 14.91 m3 Complete
Mix System at an 8.4 Day SRT (June 23 - July 11, 1974). 104
55. Process Characteristics for the 14.91 m3 Complete Mix System
at a 2.2 Day SRT (Aug. 11 - Sept. 3, 1974). 105
56. Influent and Effluent Characteristics for the 14.91 m3 Complete
Mix System at a 2.2 Day SRT (Aug. 11 - Sept. 3, 1974). 106
57. Summary of System Operation and Performance With the 12.72 and
14.91 m3 Complete Mix Systems. 107
58. COD Analyses of Clarifier Effluent Grab Samples Obtained Between
0800-0900 Hours. 109
59. Relation of Filtered Effluent COD Values to Process SRT. 112
60. Mixed Liquor Settling Velocities in the Plug Flow System. 113
61. Mixed Liquor Settling Velocities in the Complete Mix System. 116
62. Relation of Process Loading Per Unit of MLVSS to 1/SRT. 133
63. Relation of Process Loading Per Unit of MLSS to 1/SRT. 136
64. Relation of Observed Yield Coefficients to SRT. 137
65. Summary of Sludge Production Data from Various Investigations. 140
66. Summary of Recycle Solids Characteristics. 152
-------
ACKNOWLEDGMENTS
A project of this magnitude, which covers nearly two years of pilot plant
operations, could not have been accomplished without the assistance of the
entire EPA-DC Pilot Plant staff.
Mr. Paul Ragsdale supervised the mechanics and instrumentation personnel.
Mr. Calvin Taylor served as chief operator. Laboratory analyses were
performed under the direction of Mr. David Rubis. The efforts of all the
mechanics, technicians, crew chiefs, operators and laboratory personnel are
gratefully acknowledged.
xn
-------
SECTION 1
INTRODUCTION
The study summarized in this report was performed to compare the
performance of step feed, complete mix and plug flow activated sludge
systems on a pilot plant scale under similar operating conditions with the
same wastewater. The process loading to each system was varied over a wide
range during the course of the investigation. Extended periods of steady
state operation at constant flow provided extensive data on effluent
quality, sludge yield, settling characteristics, etc. at several fixed F/M
loadings for each of the system configurations.
All systems were operated on primary effluent obtained from the
District of Columbia Blue Plains Wastewater Treatment Plant. The dry
weather flow to this plant is normally around 789 m3/min (300 mgd). The
industrial contribution to this wastewater is negligible. The heavy metal
concentrations in the raw sewage are quite low. Much of the sewer system
is a combined system. Hence the investigation was conducted with a munic-
ipal wastewater that is almost entirely domestic in character with substan-
tial storm water contributions during wet weather.
-------
SECTION 2
CONCLUSIONS
The step feed, plug flow and complete mix systems all demonstrated that
the variability in carbonaceous effluent quality was mostly influenced by the
suspended solids concentrations in the effluent over a wide range of process
loadings. Soluble BOD and COD residuals were low in all cases and also
about the same in any system at comparable loadings. Sludge production was
the same, within experimental error, in all systems at comparable SRT's. In
general the results are in agreement with numerous other literature
references on the same subject published over the last 25 or so years.
It was generally advantageous to relate system parameters to applied
loading because D.C. municipal wastewater contains substantial amounts of
colloidal material and because effluent quality was essentially determined
by the effluent suspended solids concentration plus a refractory component
over a very large range of loadings. Whenever effluent biological solids are
significant, the use of a Monod-type relationship relating substrate removals
to effluent quality can actually lead to kinetic constants which largely
reflect clarification efficiency unless the contribution of the suspended
solids to effluent quality is taken into account.
The complete mix system tended to have the poorest settling character-
istics because of excessive filamentous growth. There was no fixed relation-
ship in any of the systems between process loading and settling character-
istics or SVI.
Operation at low F/M ratios led to the development of excessive
Nocardia concentrations. This organism was not a problem at high loadings.
Temperature also influenced the F/M loadings at which this organism was
competitive.
Analysis of the aggregate data from all systems obtained over an 11°C
range of wastewater temperatures produced a yield coefficient of Q.79 g VSS
produced/g BOD5 applied and a decay coefficient of 0.064 day1.
The COD/VSS ratios of the biological solids as well as their TKN and P
contents exhibited insignificant differences among the various systems.
The results of this investigation indicate that a step feed system
constructed with sufficient flexibility to divert or split the flow to any
segments of the reactor offers the best physical arrangement for secondary
treatment of District of Columbia wastewater. This conclusion would extend
-------
to most wastewaters where one is not confronted with quantitative, qualitative
or toxic shock loadings. Even under these conditions, however, the step feed
configuration offers the possibility of rapid adjustment.
-------
SECTION 3
RECOMMENDATIONS
Preliminary investigations of proposed sampling methodologies revealed
that large changes in the soluble/insoluble COD ratios occurred in acidified
primary effluent samples during 24 hours of refrigerated storage. Changes
in total COD during acidified/refrigerated storage were not detected. These
results indicated that an accurate determination of the soluble component
would have required immediate filtration of the grab samples prior to their
being composited. Insufficient manpower was available to investigate
alternative sample preservation or storage procedures in any depth. However
it was clear that the fairly common practice of filtering stored composite
samples can give misleading estimates of the soluble component. The general
extent of these changes in different wastewaters and with different storage/
preservation methods should be explored in greater detail.
Obtaining adequate solids-liquid separation in the final clarifier and
an effluent low in suspended solids produced a good secondary effluent
quality for the step feed, plug flow and complete mix process configurations
over a large range of loadings. The key to successful operation was main-
taining a system which settled adequately and did not produce adverse
bacterial growths which resulted in operating problems. Future efforts on
the secondary treatment of essentially domestic wastewaters by activated
sludge should attempt to better define the factors controlling biological
predominance. Fundamental examination of bacteria-bacteria interactions and
predator-prey relationships as a function of operating and wastewater
characteristics may provide clues for the more effective selection and
control of a desirable biomass.
-------
SECTION 4
BACKGROUND CONSIDERATIONS
The first municipal activated sludge treatment plant built in the United
States was constructed in 1916 (1). The origins of the process itself can be
traced back to 19th century England. Various modifications of this treatment
method have been developed over the intervening years, and at the present
time some form of the activated sludge process is used for municipal waste
treatment at thousands of locations throughout the world. Basic understanding
of the activated sludge process itself has also continued to evolve from
rule-of-thumb approaches to the present use of relatively sophisticated
computer simulation models to predict process performance. In many instances,
application of the more fundamental principles provides an explanation for the
past success of many of the rule-of-thumb approaches. Application of the more
fundamental approaches also provides systematic methodologies for relating
process loading, effluent quality and observed sludge yields.
Numerous studies have been conducted which relate the performance of
activated sludge systems treating domestic waste to various operating
conditions. Before some of this literature is reviewed, it is of value to
select a methodology for interrelating the various process parameters. A
large number of kinetic expressions and procedures have been proposed. One of
the most widely used approaches is that set forth by Lawrence and McCarty (2).
These are the expressions which will be used throughout much of the present
report. The basic equations are as follows:
dX _ yX _ Y d£ - bX
dt ~ ~ dt
d _ kSX
_
dt Ks+ S
where dX/dt = net growth rate of organisms per unit volume of reactor, mass
per volume-time; y = net specific growth rate of organisms, time'l ; X = micro
bial mass concentration, mass per volume; Y = growth yield coefficient, mass
per mass; dF/dt = rate of microbial substrate utilization per unit volume,
mass per volume-time; b = organism decay coefficient, time~^; k = maximum
rate of substrate utilization per unit weight of organisms, time-1; S = con-
centration of substrate surrounding the organisms, mass per volume; and Ks is
the substrate concentration when dF/dt is (l/2)k, mass per volume.
-------
A completely mixed activated sludge process with cellular recycle can be
represented as shown in Figure 1. Additional parameters not previously
defined include Q = influent flow rate, volume per time; S0 = influent
substrate concentration, mass per volume; Si = effluent substrate concen-
tration, mass per volume; V = reactor volume; w = flow rate of the underflow
waste sludge, volume per time; q = flow rate of the recycle solids, volume
per time; and Xr = recycle and waste solids concentration, mass per volume.
When equations (1) and (2) are applied to the complete mix process the
following steady-state solutions are obtained (2):
(3)
u = e - (4)
\+
K (1 + be.)
__ _
ec (Yk - b) -1
Y (sn - S,) e,.
Y - o 1 c
x ~ 1 + be e
C
YQ (S - SJ
° '
p
_
x 1 + be
where GC= biological solids retention time, time;e = mean hydraulic detention
time based on the incoming flow (Q), time; and Px = excess microorganism
production rate, mass per time.
An additional useful expression is to relate the observed yield, Y to
the biological solids retention time (3). The equation is: • '
Yobs Q (S ^~ST)
Alternatively the equation can be written as
Y
obs 1 + be
\f
-------
V, X, S
(Q + q)
x, s
i
(Q - w)
V
w, Xv
Figure 1. Schematic Diagram of a Completely Mixed Activated Sludge Process.
-------
which can also be conveniently written in the following linear form:
JL = * ec + 1 (10)
The food to mass ratio (F/M) or the process loading factor can te expressed
as follows:
_ 0 (S - S-i) (T\\
F _ o 1 UU
M " VX
This expression can be combined with equation (6) to produce the following:
1 = Y (£) -b (12)
o M
ec
Either this expression or equation (10) are convenient forms for determining
the system constants Y and b.
Similar equations were presented by Lawrence and McCarty (2) for a plug
flow system although an explicit solution for S-j was not developed.
When dealing exclusively with soluble substrates, it is a relatively
simple problem to evaluate either effluent quality or suspended solids
production. In fact, maximum utility of sludge production data can be
achieved if the sludge production can be realistically expressed per unit
of substrate removed. This is the basis for the expressions developed above.
However, when dealing with domestic wastewater a substantial portion of
influent BOD and COD consists of biodegradable colloidal material. As
indicated by O'Melia (4) about 80 percent of the COD of raw domestic sewage
is attributable to material which is colloidal or larger in size. Even after
primary clarification it is not unusual for at least half of the total
organic loading to the secondary system to consist of colloidal material.
Application of the traditional BOD or COD measurements to the secondary
clarified effluent will not distinguish between oxidizable colloidal
material which passes unchanged through the treatment system and biological
solids which are produced by growth on the colloidal or soluble material.
As a result, the determination of solids production or sludge yield cannot
readily be made on the basis of the quantity of substrate removed.
It is fairly common practice to assume that substrate removal can be
adequately represented by taking the difference between the influent and
clarified effluent BOD or COD values. In those cases where the effluent
suspended solids are low, this is a sound "fundamental" approach. The major
difficulty arises, however, in attemotinq to compare kinetic "constants" on
a substrate removal basis among systems with substantially different effluent
8
-------
characteristics caused by variations in effluent suspended solids concen-
trations. As the percentage of the total solids production passing over
the weir of the final clarifier increases there is a corresponding
deterioration in effluent quality; when calculations of cell yield are based
on the difference in influent and clarified effluent BOD or COD it is
possible to calculate higher and higher observed cell yields (Equation 8)
although the real cell yield from the system may not exhibit any change.
For any given treatment system this may or may not be important. However,
it must be considered when attempting to compare results from several
different studies. Failure to do so can lead to erroneous conclusions.
Middlebrooks et al.(5) reported on a study with model extended aeration
units operated at various detention periods on comminuted, degritted waste-
water from a local treatment plant. The data were analyzed according to the
Monod relationship between substrate removal and cell growth rate. The
clarifier effluent BODc values were used to measure the "substrate" escaping
into the effluent. Data were divided into three sets. The average BOD
removal for the set I data was 86%. When these data were analyzed the yield
constant was determined to be 0.65 Ib VSS/lb BODs removed and the decay rate
was 0.043 day~1. Both these values are consistent with a number of other
investigations with domestic wastewater. The third set of data analyzed
had high effluent suspended solids concentrations and consequently high
effluent BOD5 values. Naturally, a higher yield coefficient was calculated
since the effluent solids were considered to be unused "substrate" in the
set of equations chosen. In spite of this obvious fact, the yield value
obtained from data set III was considered to be impossible. In fact it was
stated that "the higher yield constants for the model studies probably can
be explained by the operational procedure of maintaining a constant flow rate
which provides adequate substrate to maintain the growth rate near the
maximum at all times. This would be particularly true for set III of the
model data, since the influent substrate concentration of BOD was maintained
at a very high level." This conclusion indicates that the authors did not
believe in the validity of the Monod relationship even though it was being
used as an underlying assumption in their model.
It is apparent that different yield constants based on the use of
clarified effluent quality as a measure of "substrate" remaining are
relatively simple to obtain. This in turn leads to distortions in an
attempt to measure the true growth yield or the kinetic constant, Kg.
A number of investigators have shown that the effluent quality of
domestic activated sludge plants is primarily determined by the quantity of
effluent suspended solids. As noted by Jenkins and Garrison (6), "Indeed it
should be very difficult with the accuracy of currently-used control analyses
to detect any variation in soluble degradable effluent COD for plants
operating below substrate removal rates of 3 Ib COD removed/day/lb VSS which
is approximately the upper limit of substrate removal rates currently
employed in the practical use of the activated sludge process! This is not
to say that effluent quality will not vary over this range of substrate
removal rates. It certainly will. But the variation in effluent quality
will be due largely to its content of activated sludge particles and its
content of nondegradable organic matter. The potential of the activated
9
-------
sludge process for the removal of biologically available organic matter there-
fore is exceedingly high. It is the inability to separate cell material from
the effluent and the presence of nondegradable (or refractory) COD that makes
the potential of the process unattainable in practice."
The presence of large quantities of degradable colloidal material in
domestic wastewater coupled with the difficulty of adequately defining the
quantity of unused substrate makes it advantageous to evaluate the yield and
decay constants on the basis of applied load except for cases of very high
loading. This approach makes it possible to compare the results from
numerous studies on a common basis. It avoids the problem of attempting to
"force" the Monod relationship to, in effect, serve as a model of clarification
efficiency. Where effluent suspended solids are low and the removal efficiency
is high, the equations previously presented (Equations 1-12) give yield and
decay coefficients which are nearly the same as those based on applied load.
One of the first studies which specifically addressed the relationship
between sludge production and applied loading was reported by Heukelekian,
et al.(7) in 1951. Laboratory experiments were performed with sewage from
different sources to measure the amount of excess activated sludge produced
under various operating conditions. All units were batch fed three times per
day. Variations in effluent quality were within the limits of experimental
error over the range of loadings studied (approximately 0.1 to 0.8 kg BODg/
kg MLSS/day). The BOD:SS ratio of the sewage was generally between 1.5 to
1.7:1. The investigators showed that sludge production decreased as the
loading decreased. It was also shown that at a given F/M ratio the sludge
accumulation was the same irrespective of the MLSS concentration. The authors
indicated that a literature search produced only one study with sufficient
information to relate solids production to process loading. Data obtained by
Sawyer (8) in 1940 were shown to produce the same relationship between process
loading and sludge yield as obtained by Heukelekian, et al. The following
formula was proposed to evaluate sludge accumulation at 20°C:
A = 0.5 B - 0.055 S
Where A = Ib of volatile suspended solids accumulation per day;
B = Ib of BOD5 fed per day; and S = Ib of mixed liquor
volatile suspended solids "
This is essentially the same equation as presented previously (Equation
12)_except that the relationship is based on applied loading in this case.
It is interesting to note that the basic expression between sludge production
and process loading was empirically developed without reference to currently
used mathematical models.
Wuhrmann (9) presented selected data from 29 experiments from a pilot
plant treating municipal sewage in Zurich, Switzerland with detailed data
presented from nine studies, "it was indicated that the results could not be
related to the equation developed by Heukelekian et al.(7). However it
10
-------
appears that Wuhrmann considered excess sludge production to relate to under-
flow wasting only with no attempt to include the solids lost in the clarified
effluent as a part of the total solids production. Sludge production was
shown to decrease with increasing sludge age as defined by Gould (MLSS in the
system divided by SS entering the system, kg per kg/day). Excess sludge
production was also shown to decrease as the BOD loading to the aerator
decreased. The detailed results presented by Wuhrmann are summarized
(except for the 20 ppm asbestos addition) in Table 1. Data were presented
on the effluent SS but not the VSS. Also the information on volatile solids
production in the excess sludge was expressed based on volatile solids rather
than volatile suspended solids. Effluent quality was shown to deteriorate
with increasing load but insufficient information was given on the effluent
suspended solids concentrations to further evaluate this observation.
Torpey and Chasick (10) summarized the operation of several step feed
systems in which sludge age (as defined by Gould) was related to process
performance. For the treatment of domestic sewage the lower limit of sludge
age was found to be about three days. Any attempt to operate at substan-
tially below this limit for a period exceeding a few days was indicated to
result in the formation of a voluminous floe with poor settling qualities.
Usually this resulted from large numbers of filamentous organisms. The
upper limit of sludge age was between 4 and 6 days depending upon the
particular treatment plant. Operation above this limit resulted in a breakup
of the floe with small discrete particles and effluent deterioration.
Because of these considerations operation was confined to a sludge age
ranging from 3 to 4 days for most plants. It was stated that the volumes of
sludge for disposal were about the same from step aeration or conventional
activated sludge. No quantitative data were provided.
In 1958, Garrett (11) summarized the results of 17 months of activated
sludge operation in which the growth rate was controlled by wasting excess
mixed liquor. Insufficient information was provided to thoroughly analyze
the data since influent BOD and effluent suspended solids information was not
provided. The SRT was expressed simply as a function of aeration volume to
waste mixed liquor flow. Since the effluent BOD values were generally
around 20 mg/1 the effluent solids were not a substantial part of the total
solids lost and, hence, the yield values based on BOD removal are probably
"reasonable." The yield coefficient was reported to be 1.1 Ib VSS per Ib of
BOD removed. Garrett noted that this was twice the value reported by
Helmers et al. (12) and Heukelekian et al.(7) and this increase was attri-
buted to the use of raw unsettled sewage.
In discussing the theory of extended aeration, McCarty and Brodersen
(13) separated sludge accumulation into the accumulation of biological solids
plus the accumulation of biologically undegradable suspended solids
originally present in the influent waste. The biological solids portion was
stated to represent the only solids accumulation with soluble wastes, but
could represent only about 50 percent of the accumulation when dealing with
domestic wastes. The following expression was developed for the accumulation
11
-------
TABLE 1. SLUDGE PRODUCTION BASED ON RESULTS OF WUHRMANN ( 9 ).
Test
No.
IX-a1
IX-o-,
xrai
xr°i
xr°2
X2"a3
X2-o3
XI-Q,
FlOW q
gpd/fr
83.5
20
75.5
75.5
75.5
164
164
167
MLSS
mg/1
3930
2830
3440
3100
3200
3460
3570
3460
F/M
g BOD^/day
g MLSS
0.35
0.12
0.49
0.54
0.45
0.88
0.85
0.74
'BOD
Inf. Eff.
mg/1 mg/1
123
123
167
167
142
139
139
115
13
8
16
16
10
24
25
32
'EXCESS SLUDGE
in underflow
g Vol. Solids
C
0
0
0
0
0
0
0
0
EFFLUENT SLUDGE
VSS* PRODUCTION
mg/1 g VSS
removed g
.55
.37
.46
.45
.56
.56
.62
.81
7.5
6.0
16.5
10.5
5.2
35.3
26.3
18.0
Inf. BOD
Inf. SS
BOD5 applied
0.
0.
0.
0.
0.
0.
0.
0.
55
39
51
48
56
72
70
74
2.0
2.0
2.1
2.1
1.9
0.92
0.92
1.2
* Assuming effluent SS are 75% volatile
-------
of the biological solids portion only:
A = aF - bMd
Where A = accumulation of volatile biological solids, mass per day; the
constant a represents the fraction of the mass of 5-day BOD removed per day,
F, which is synthesized into new biological solids; and the constant b
represents the fraction of degradable biological solids in the system, Mj,
which is destroyed per day, by endogenous respiration. The constant a was
estimated to be 0.53 for the degradable solids and 0.65 for mixed wastes such
as domestic sewage which results in both degradable and undegradable solids.
Based on results with synthetic sewage and acetate the decay constant b was
estimated to be 0.18 for the degradable biological solids fraction.
Hopwood and Downing (14) investigated several factors affecting the
production and properties of activated sludge from plants treating domestic
sewage from a residential district. The experimental results were related
to a sludge growth index which was defined as the weight of sludge formed
per unit weight of BOD applied. This was done since it was recognized that
results based on BOD removal would be largely determined by the concentration
of activated sludge remaining in the settled effluent and thus on the
efficiency of the settling tanks even though all the incoming organic matter
had been converted into sludge. The sludge growth index decreased
progressively from 0.9 g SS/g BOD5 applied to 0.38 g SS/g 6005 applied with
increasing period of retention of the sewage from 2 to 36 hours. The highest
loading studied was around 0.8 g BOD/g MLSS. The results suggested that the
sludge growth index passed through a maximum at a temperature of about 9°C.
The data suggested the occurrence of a maximum sludge growth index at a D.O.
concentration of about 0.5 ppm and further indicated that D.O. was not a
factor above approximately 1 mg/1.
Benedek and Horvath (15) conducted a series of experiments with the
sewage from Pecs, Hungary. A yield coefficient of 0.673 g VSS/g substrate
removed was obtained. However, substrate removal was measured by a perman-
ganate test with the following relationships stated to exist.
BOD. „ . = 43.7 + 1.237 (permanganate value)
influent r 3
BOD ff, . = -7.846 + 0.989 (permanganate value)
These relationships strongly suggest that the permanganate values are not
at all similar to the commonly used COD analysis although the coefficient
basis has been reported as COD by Lawrence and McCarty (2). Since all
substrate measurements are reported as permanganate values it is difficult to
evaluate the kinetic constants further.
13
-------
Jenkins and Garrison (6) also recognized the problem of measuring the
substrate remaining when dealing with clarified effluents. They reconmended
that the residual organic material from domestic wastewater be approximated
by use of soluble COD and this was the methodology used in their determination
of kinetic constants. As previously indicated, the soluble biodegradable COD
concentration in the effluent from standard rate activated sludge plants was
considered to be negligible. Data obtained with domestic wastewater were
compiled from the Pomona Water Renovation Plant, the Whittier Narrows Water
Reclamation Plant and SERL. Analysis of the combined data produced a yield
coefficient of 0.33 g VSS produced/g COD removed and 0.04 g VSS/day lost
endogenously/g VSS in the system. One months data from the Pomona Plant was
presented and the COD to SS ratio of the primary feed to the activated sludge
system was 2.4:1. The influent COD averaged 296 mg/1 with an average soluble
effluent concentration of 27 mg/1. Domestic wastewater from Richmond,
California treated at the SERL plant typically had an influent COD of
250 mg/1 with a soluble effluent COD of 37 mg/1. The San Ramon, California
Plant, which operates in the extended aeration range, was stated to produce
a soluble effluent COD which rarely falls below 30 to 35 mg/1.
Smith and Eilers (16) developed a generalized computer model to predict
the performance of activated sludge systems. Data obtained from the Hyperion
testing program showed almost negligible changes in the soluble effluent
quality over wide variations in process loading. The primary effluent at this
plant contained an average COD:BOD:SS ratio of 3.3:1.6:1. The sludge
production data were shown to be described by the following expression (17):
1 = 0.789 £ - 0.071
SRT u"u" M
The correlation coefficient was 0.972. In this expression, F is the
difference in influent and effluent BOD values and M represents the MLVSS.
Since the effluent BOD^ only averaged 12 mg/1 and never exceeded 20 mg/1,
the yield coefficient is only moderately higher than would be obtained on the
basis of applied loading. The average influent BOD was 152 mg/1 and average
influent COD was 314 mg/1. Filtered COD concentrations in the final effluent
averaged 40 mg/1.
Information relating process loading, effluent quality and sludge
production was summarized in a 1971 EPA publication prepared by the City of
Austin, Texas and the University of Texas at Austin (18). One of the
expressions presented relating sludge production and process loading was the
following:
,
V V
14
-------
where Xv = volatile mixed liquor suspended solids; t = hydraulic detention
time; and S0 and Se are the influent and effluent BOD5 values, respectively.
This is the same expression as Equation 12.. It was reported that Eckenfelder
obtained values of a = 0.73 and b = 0.075 per day for the activated sludge
process treating settled municipal wastewaters. Examination of the data
published by Garrett (11) produced a value for b of 0.08 per day. Since the
solids production values can be affected to a large extent by the concen-
tration of influent suspended solids the following alternative expression for
predicting sludge production was presented:
SRT
S - S *
_p e_
Xt
-b + X
on
Xt
where XQn is the concentration of nonbiodegradable suspended solids in the
influent, Se* is the filtered effluent BOD, and X is the mixed liquor
suspended solids concentration. Typical values of a* and b were indicated
to be 0.63 and 0.075, respectively. For municipal wastewater the equation
was further modified to the following form.
SRT
0.6 (S,
Xt
- 0.075
where X0 is the influent suspended solids concentration. Data obtained from
the Govalle Treatment Plant, Hyperion and reported by Wuhrmann were used to
show the correlation between excess sludge production and the combined
suspended solids plus BOD loading. Smith (17) has also indicated that the
Hyperion data alone fit this relationship very well. Analysis of the Govalle
and Hyperion data clearly showed that the soluble effluent BOD values
exhibited only very small changes over a wide range of loadings. Effluent
quality was shown to be largely determined by the effluent suspended solids
and these decreased with increasing sludge age in the 0-10 day SRT range
considered. It was also mentioned that results of laboratory-scale
experiments comparing contact stabilization and the conventional activated
sludge process indicated that the effluent BOD concentration for each process
was almost identical at organic loading rates of 0.40 to 1.90 g BOD/g MLSS.
Unfortunately, the work referenced (Water Pollution Research Laboratory,
1967, 1968) was not listed in the "References" so this information could not
be further evaluated.
Boon and Burgess (19) utilized sewage from an entirely residential area
to compare the performance of two pilot scale activated sludge units operated
in parallel. Diurnal flow was applied to one unit and steady state flow to
the other. The D.O. was maintained above 3.5 mg/1. The weighed average rate
of flow was the same for both units. The sewage feed had an average BOD^
near 300 mg/1, a COD near 600 mg/1 and suspended solids of about 160 mg/1.
Effluents of similar quality were obtained from the two units over the entire
range of sludge loadings (0.5-2.5 g BOD/g MLSS). Similarly the production of
15
-------
sludge was little Influenced by diurnal changes in flow of sewage but was
related to the daily average sludge loading. Relevant solids production data
are summarized in Table 2. The specific sludge production (g SS/g BOD
removed) was found to be described by the following empirical equation:
s r 0.24
Y = 0.("° °
XT
where the substrate concentrations are measured by BOD, the solids concen-
tration as mixed liquor suspended solids with T being the period of aeration
of the sewage in days. The investigators also noted that changes in settle-
ability of activated sludge measured during the periods of constant flow
rates could not be related to any measured parameter or condition of operation
The diurnal variation observed in settling characteristics was less than the
variation shown to occur when both plants were operated under constant
conditions. Sludges with the poorest settling characteristics were obtained
when the sludge loading was about 1 g BOD/day/g MLSS. Settling improved at
both higher and lower loadings.
Toerber (20) and Toerber, et al. (21) reported on a full scale compari-
son between parallel activated sludge systems operating under complete mix
and plug flow modes. The sewage was a weak domestic-commercial waste with a
limited industrial contribution. The total 6005 of the aeration tank
influent seldom exceeded 150 ing/1 and often was less than 100 mg/1. Under
normal loading conditions (detention time of 3.1-3.6 hours and F/M of 0.27-
0.82 g BOD^/g MLVSS), no appreciable difference could be found between
complete mix and plug flow on the basis of BOD or COD removal efficiency
from primary effluent to final effluent during 5 months of parallel operation.
Also effluent filtered BOD and COD values were insensitive to loading
variations between 0.2 and 1.0 g BODs/g MLVSS in the complete mix system, and
a comparison of soluble BOD removal efficiencies indicated no marked differ-
ence between complete mix and plug flow. Shock load studies were conducted
with a whey waste. In response to a severe shock load the complete mix
system demonstrated an overall removal efficiency about 10 percent greater
than the plug flow unit. However, since the plug flow unit was operated at
a D.O. level of about 0.2 mg/1 for 3 hours during the shock load the general
significance of this difference is minimal. During parallel shock l^ad
studies with adequate D.O., the response of the two units was very similar.
Solids production data were presented for the two units; the results are
quite scattered but do not reveal any obvious differences between the two
systems.
Gujer and Jenkins (22) developed a kinetic model of the contact stabili-
zation process and verified it with the aid of bench-scale activated sludge
units treating settled domestic sewage from Richmond, California. Infor-
mation on influent BOD:COD:SS ratios or concentrations was not provided.
Over the range of 0.5-1.6 kg COD removed/kg VSS/day only temperature and
process loading had a significant effect on sludge production. Of these two
variables, the temperature had a more pronounced influence on process
16
-------
TABLE 2. SOLIDS PRODUCTION AS A FUNCTION OF PROCESS LOADING AND TYPE OF FLOW
TYPE OF FLOW
PROCESS LOADING
AVERAGE TEMP,
SOLIDS PRODUCTION
Variable
Steady
Variable
Steady
Variable
Steady
Variable
Steady
g BOD/day
g MLSS
0.89
0.81
0.69
0.64
0.52
0.56
2.22
2.42
oc g SS
g BOD removed
20 0.76
0.76
19 0.65
0.68
19 0.64
0.67
16 0.84
0.85
q SS
g BOD applied
0.72
0.73
0.63
0.66
0.62
0.65
0.73
0.74
q SS
g COD applied
0.34
0.34
0.33
0.35
0.35
0.37
0.40
0.41
Data from Boon and Burgess ( 19)
-------
efficiency than did process loading. With substrate removal based on the
difference between influent and soluble effluent COD values and loading based
on reactor and stabilization tank VSS, the following relationships were
established:
1 0.48 F -0.07 at 11°C
SRT M
0.38 F -0.07 at 21°C
SRT M
The authors stated that the contact stabilization process and the con-
ventional activated sludge process produce approximately the same amount of
sludge and the fractional distribution of sludge between the contact and
stabilization basins did not appear to significantly influence sludge pro-
duction.
In 1970 and 1971, studies were conducted at the EPA-DC Pilot Plant with
air and oxygen systems treating primary effluent. Sludge production from
these two systems was compared (23, 24, 25, 26) and it was concluded that
the total production of solids in the oxygen system "... was significantly
less than the similarly operated diffused air system above an SRT of 6 days."
(25). This observation is not consistent with more recent information on air
and oxygen systems with adequate dissolved oxygen levels (27, 28). For this
reason, it appeared possible that the differences in reactor configuration
may have influenced the reported differences in sludge production. Since
virtually no data were provided on the operation of the so-called similar
air system, some of these data in the existing pilot plant files were reviewed.
In many instances the reported flow patterns and rates in the four pass air
system reactor bore no relationship to the measured MLSS concentrations thus
indicating large errors in measuring flow and, in fact, even knowing how the
flow was distributed. Furthermore, large quantities of BOD were observed to
"disappear" with essentially no production of biological suspended solids.
Solids production in the air system was indicated to reach a maximum at an SRT
of 9.5 days. At this point the solids production was stated to be 1 g SS/g
BOD5 applied. This observation is at variance with numerous other studies.
The numerous internal inconsistencies in the data from the air system strongly
suggest that the proposed relationship between solids production and SRT was
erroneous.
Drnevich and Gay (29) and Drnevich and Stuck (30) used the above refer-
enced data on air and oxygen systems in their sludge yield evaluations. It
was felt (30) that if the reported results for the air system had been
expressed on the basis of substrate removed instead of substrate applied, the
air system would have exhibited higher sludge production rates over the entire
range of SRT's tested. The problem was not in the use of applied vs. removed
loads, but the extremely poor quality of the air system data. This is not to
say that the oxygen system may not have produced a lower yield than the air
system but rather that the data on the air system are much too poor to be
18
-------
able to make any sort of meaningful comparison. Other studies on D.C. waste-
water (31) where parallel air and oxygen systems were operated have shown no
difference in volatile solids production.
The incremental increase in biological solids resulting from the growth
of autotrophic nitrifying organisms in secondary treatment processes where
nitrification is occurring will be quite small. For domestic wastewater
the additional sludge production would be exceeding difficult to measure in
most cases. Smith and Eilers (16) indicated that the small concentration of
nitrifyers present when conditions are correct would contribute about 1-2
percent of the total MLVSS. Similar estimates by Lawrence and Brown (32) led
to the conclusion that a very small fraction of the volatile suspended solids
are actually nitrifying bacteria. Gujer and Jenkins (22) stated that for a
typical domestic sewage, where the ratio of COD removed to nitrogen nitrified
is greater than 10, the contribution of the nitrifyers to the sludge pro-
duction is less than 5 percent. After reviewing the yield coefficients
reported in the literature, Poduska (33) concluded that the best estimates
were 0.05 g Nitrosomonas formed per g of NH^N oxidized and 0.02 g Nitro-
bacter formed per g of N02-N oxidized.
In addition to the reported studies on domestic wastes or municipal
wastes of largely domestic origin, a number of relevant studies have been
conducted with pure cultures, defined substrates or industrial wastes. A few
of the more pertinent studies will be summarized here.
Sherrard and Schroeder (3) operated a laboratory scale completely mixed
activated sludge unit under highly controlled conditions with sludge wasting
being the only process parameter varied. Bacto-peptone and selected inorganic
nutrients were used as the feed. Within an SRT range of 2-18 days only a
minimal change in effluent COD was observed. Sludge production, however, was
significantly altered. When the results were expressed as dry weight of SS
produced per unit of COD utilized the data were found to fit the following
expression:
Yh = 0.406 e(- °'067 9c>
obs
with a correlation coefficient of 0.98. When the data were plotted according
to the expression previously presented in Equation 12 the results were as
follows:
0.414 R - 0.093
I _ OX
ec = x
The correlation coefficient was 0.99.
19
-------
Sherrard and Lawrence (34) made a series of calculations to show typical
activated sludge performance using coefficients commonly found in the
literature. Using values of Y = 0.40 g VSS/gm COD, b = 0.09 day1 and
K- = 60 mg/1 of COD, it was concluded that for SRT values greater than 2 days
the changes in soluble COD would be insignificant with increasing SRT. These
coefficients also produced a maximum sludge production at an SRT of about
2 days.
Schroeder (35) indicated that in real systems the required operating
constraints force the result that both the ideal plug flow and the ideal
complete mix processes will effectively produce the same effluent quality.
It was indicated that in the cell residence time necessary for good cell
flocculation and process operation with respect to cell separation (SRT >
3 days) the differences in performance between complete mix and plug flow
would be very difficult to measure.
Muck and Grady (36) conducted laboratory studies with glucose substrate
and heterogeneous cultures to assess the effects of temperature upon the
microbial growth parameters. As was anticipated from the results of others
using pure microbial cultures, the maximum specific growth rate constant
and the bacterial decay rate constant were found to increase with increases
in temperature in accordance with the Arrhenius equation over the temperature
range of 10-30°C. The rate of increase of the decay rate constant was
slightly larger so that the ratio of the temperature characteristics for
decay to that for growth was 1.11. Furthermore the true growth yield was
found to increase as the temperature was raised from 10°C to 20°C but to
decrease with further increases in temperature.
Chudoba et al.(37) attempted to relate the amount of filamentous
bulking to the hydraulic regime of the aeration tank. Starch and peptone
were fed to four laboratory units which varied from complete mix to plug
flow. The sludge loading was maintained between 0.3-0.4 g BOD5/g MLVSS/day
and it was observed that the SVI values were highest in the complete mix
system and decreased as the degree of mixing was reduced. The high SVI
values were caused by a high content of the filamentous microorganisms
Leucothrix and Sphaerotilus. It was concluded that complete mix systems
cause excessive filamentous growth and that the filamentous growth could be
controlled by maintaining a concentration gradient of substrate along the
aeration system. However, subsequent studies at higher loadings (38)
revealed that the plug flow system also produced high SVI values* and high
proportions of filamentous microorganisms. Rensink (39) also compared the
SVI's in batch, plug flow and complete mix units as a function of loading
and hydraulic regime. A synthetic wastewater was employed. The greatest
tendency for bulking was in the complete mix unit although the particular
response was also observed to be dependent on process loading. The filamen-
tous organisms that induced bulking were Sphaerotilus natans, Flavobacter,
Flexibacter^ and Halisocomenobacter.
Bisogni and Lawrence (40) operated a continuous feed essentially
completely mixed reactor with internal cell recycle on glucose, yeast
extract and inorganics to evaluate the relationship between SVI and SRT.
20
-------
Based on total biomass in the effluent, the best overall solids removal
occurred at SRT values in the range from 4 to 9 days.
21
-------
SECTION 5
EXPERIMENTAL SYSTEMS
A. System Reactors and Clarttiers
A schematic diagram of the three types of activated sludge systems
operated during this study is presented in Figure 2. The step feed reactor
consisted of four completely mixed passes. Each pass was approximately
1.4 m X 1.4 m X 4.0 m liquid depth (4.5 X 4.5 X 13 ft). The total reactor
volume was 28.09 m3 (7,420 gal). All recycle flow was returned to the first
pass of the system. The influent flow was equally split into three streams
by use of splitter box with one-third of the flow going to the second pass,
one-third to the third pass and the remainder to the last pass. Flow from
pass to pass was through 10 cm X 15 cm (4 X 6 in) slots cut in the common
steel wall between the passes. Compressed air was supplied independently to
each pass and discharged through two 2.5 cm (1 in) diameter perforated PVC
pipes located at the bottom of each pass. All effluent was from the fourth
pass of the system and flowed by gravity to a circular center-feed clarifier
with a cross-sectional area of 8.92 m2 (96 ft2). Recycle solids were
returned to the reactor with a variable speed Moyno pump.
The plug flow system was simulated by constructing a reactor with
eight completely mixed passes connected in series. The flow pattern is
illustrated in Figure 2. Total reactor volume was 28.88 m3 (7,630 gal). Air
was independently supplied to the bottom of each pass and discharged through
a single 2.5 cm (1 in) perforated PVC pipe. Influent flow and recycle flow
were always added to the first of eight passes. The effluent flowed by
gravity from the eighth pass of the system to a circular center-feed clarifier
with a cross-sectional area of 8.92 m2 (96 ft2). A variable speed Moyno pump
was used to return the recycle solids to the head of the reactor.
The complete mix system consisted of a large rectangular tank which was
approximately 2.7 m X 3.0 m X 4.1 m liquid depth (9 X 10 X 13.5 ft). The
reactor capacity was 29.34 m3 (7,750 gal). Compressed air was supplied
through three, 1.5 m (5 ft) vertical spiral diffusers located at'the bottom
of the reactor and recycle flow entered approximately 1 meter (3 ft) below
the water surface. The effluent flowed by gravity to a circular center-feed
clarifier with a 8.83 m2 (95 ft2) cross-sectional area.
A dilute-out curve was run to evaluate the hydraulic regime of the
"complete-mix" tank. This was done by filling the reactor with sewage and
then adding 45 kg (100 Ib) of salt (NaCl). The system was then aerated for
15 hours. Next a steady state flow of sewage was started and effluent
22
-------
INFLUENT
RECYCLE
STEP FEED SYSTEM
INFLUENT
RECYCLE
PLUG FLOW SYSTEM
INFLUENT
RECYCLE
COMPLETE MIX SYSTEM
Figure 2. Schematic Diagram of the Step Feed, Plug Flow and
Complete Mix Activated Sludge Systems.
23
-------
samples were taken every 20 mtnutes. Furthermore, at every sample time, an
additional sample was taken near one of the three corners of the reactor
furthest from the effluent outlet. Influent sewage samples were taken once
per hour and the cumulative influent flow was also recorded once each hour.
The sodium ion concentration of all samples was determined by atomic
adsorption spectroscopy. It was found that the sodium ion concentration was
the same at any pair of sampling points (reactor comer and effluent) at any
given time. The Na concentration in the effluent is shown in Figure 3. Also
shown is the theoretical dilute-out curve. The theoretical curve accounts
for the influent sodium concentration in the sewage and was also solved for
hourly intervals to compensate for the very small variations in flow which
occurred. The steady-state form of the dilute-out curve used was
C. (e
i
Dt
Dt
where Ct = concentration at any given time, mass per volume; C0 = initial
concentration, mass per volume; Cj = influent concentration, mass per volume;
D = dilution rate, time -1; and t = time. The results of the dilute-out
study clearly indicate that the reactor was very close to being completely
mixed.
Two other complete mix reactors were also operated at the pilot plant
as part of another investigation. In some cases, data from these systems
can be advantageously used in conjunction with the results of the present
investigation. One of the reactors had a volume of 14.91 m3 (3,940 gal) with
the effluent going to a 6.93 m2 (74.6 ft2) clarifier. The other reactor had
a volume of 12.72 m3 (3,360 gal) and fed a clarifier with a 5.96 m2 (64.2 ft2)
cross-sectional area. Data presented from either of these reactors will be
explicitly referred to in the text. All other references to a complete mix
system will denote the 29.34 m3 (7,750 gal) reactor.
B. Influent Flow
In all cases, the influent flow to the various processes consisted of
District of Columbia primary effluent. The primary effluent was pumped to
a header tank in the pilot plant. Each individual process flow was pumped
from the header tank through a magnetic flow meter. The actual flow was
manually recorded once per day from a flow totalizer attached to the flow
meter. Once per week the flow was diverted from the process to a calibrated
drum to check the flow meter calibration and make any needed adjustments.
All processes were operated at a constant flow rate although the chosen flow
rate was varied from time to time.
C. Sludge Wasting
Sludge wasting was done automatically through the use of a timer and
automated valves. At preset intervals, the valve in the recycle line leading
to a given process would close and the recycle flow was diverted to a 210 1
24
-------
700--
THEORETICAL
A—-A MEASURED
1.0 2.0
3.0 4.0 5.0
TIME, hours
6.0 7.0 8.0
Figure 3. Experimental and Theoretical Dilute-Out Concentrations of Na
in the Complete Mix System.
25
-------
(55 gal) drum. A level control probe in the drum was used to reverse the
valve positioning. The signal from the level control probe was also used
to operate a waste totalizer. This system proved to be very reliable and
provided careful control of the waste volumes.
D. Dissolved Oxygen Control
The air flow rate was manually controlled to each pass of the step feed
and plug flow systems. Manual control was also initially tried with the
complete mix system, but it was found that this system always developed
excessive filamentous growth unless the D.O. level was carefully controlled.
Therefore the air flow control was automated with a Delta Scientific process
D.O. meter, an automated valve in the line leading to the three spargers and
either a digital or analog control loop. Although the digital control
sequence was quite effective, the continual hardware failures in the IBM
System 7 Computer eventually led to the use of an analog controller.
26
-------
SECTION 6
METHODS AND PROCEDURES
All processes were operated on a 24-hour a day, 7-day a week schedule.
The only interruptions in the normal operating sequence resulted from
mechanical malfunctions and these were normally of short duration. No
process data are presented in this report from any period with excessive
mechanical malfunctions.
Grab samples of reactor influent and clarified effluent were taken
every four hours, and grab samples of mixed liquor and recycle solids were
obtained every eight hours for the laboratory analyses described below. The
grab samples were composited over a 24-hour period on Tuesday, Wednesday,
and Thursday; samples collected on Friday-Saturday and on Sunday-Monday were
composited over the 48-hour period. The single exception to this was that
the samples for BOD5 measurement were always 24-hour composites and the
analysis was always started within a few hours (4-10 hours) after the last
sample for each 24-hour composite had been collected. All samples were
refrigerated at 2°C prior to analysts. In addition, all samples except
those taken for 8005 or suspended solids analysis were preserved with one
drop of H2S04 per 30 ml of sample while they were being held in storage.
All laboratory analyses (except BODg) were performed on a Monday through
Friday schedule.
The following analyses were performed in the District of Columbia Pilot
Plant laboratories according to the procedures specified in Standard Methods
(41): suspended solids, volatile suspended solids, BODs, COD and TKN. Also
BOD5 analyses were sometimes performed with nitrate production inhibited by
the addition of 0.5 mg/1 of ally!thiourea (42). The procedures specified in
the EPA Manual (43) were used for the determination of (N02 + N03)-N with a
Technicon autoanalyzer. The method of Gales et al. (44) was used for the
determination of total phosphorus. Occassionally the TOC concentration was
measured on a Beckman analyzer. The sodium concentrations in the dilute-out
study (Figure 3) were measured with a Perkin-Elmer Model 303 Atomic Adsorp-
tion Spectrometer.
The number of samples routinely analyzed for TKN or P04 varied period-
ically throughout the study depending upon the number of chemists/technicians
available and the total laboratory sample load resulting from this and other
ongoing studies. Hence some of the steady state operation to be summarized
in the next section does not include TKN or P04 data for all of the sample
locations. Since the biological growth was carbon limited with available
P and N always in considerable excess, the absence of the TKN and P04 data
27
-------
for some periods/systems was of no particular consequence.
In addition to collecting grab samples and compositing them for sub-
sequent laboratory analysis, the operating personnel also: (a) checked the
mixed liquor dissolved oxygen levels in the various reactor passes every
four hours with a portable YSI or Delta Scientific field probe and adjusted
the air flow rates as needed to attempt to maintain a D.O. concentration of
1-2 mg/1 in the step feed and plug flow systems and also the complete mix
system if the automated D.O. control loop was inoperative; (b) obtained
selected solids samples for 30-minute sludge volume determinations in
unstirred one-liter cylinders; (c) measured temperature, pH and alkalinity
of selected samples; and (d) measured the depths of the sludge blankets in
the clarifiers with a photoelectric cell.
Throughout the study samples of the various mixed liquor effluents were
removed periodically and settling tests were run in 2.3 m X 0.15 m (7.5 ft
X 6 inches) diameter stirred columns. The stirring mechanism consisted of
two 0.64 cm (1/4 inch) diameter rods which extended the length of the column
and normally rotated around the vertical axis at a rate of 15 rph. Settling
rates in which the recycle solids were mixed in varying proportions with the
clarified process effluent were also periodically determined.
Special studies or analyses were also occasionally performed. Since the
exact methods or procedures varied depending upon the purpose of a special
study, the details will be presented as an integral portion of the results.
28
-------
SECTION 7
RESULTS
A. Laboratory Studies
As indicated in Section 5, the presence of substantial quantities of
colloidal material in the sewage greatly complicates the evaluation of sludge
production and effluent quality. Examination of the soluble component in
both influent and effluent samples offers a means of partially overcoming
this problem. In an effort to learn whether the soluble COD of the regular
composited samples could be effectively utilized, the total and soluble COD
values of grab samples of D.C. primary effluent were compared when the
analyses were run on a fresh sample and when rerun on a portion of the
initial sample after it had been acidified with one drop of HoS04 per 30 ml
of sample volume and refrigerated for 24 hours. An aliquot of the stored,
refrigerated sample was filtered after storage to determine the 24-hour
value for the filtered COD. To eliminate leaching of organic materials from
the filter itself, only glass fiber filters were used. The filters used were
either Reeve Angel 934AH or Reeve Angel 984H without the cellulose paper
circles. The results from this brief study are summarized in Table 3. It
can be seen that there was no change in the total COD during acidified stor-
age. However, in every case there was a significant decrease (as much as
50%) in the filtered COD component after 24 hours of acidified storage.
This indicated that there was no value in performing soluble COD analyses on
composited influent samples.
BODr analyses were also performed on the samples from February 5 and 7
(Table 3). The BOD5 on the 5th was 108 mg/1 and the soluble BODs (934 AH
Filter) on the fresh sample was 59 mg/1. The BODs of the grab sample from
the 7th was 101 mg/1 with filtered BOD5's on the fresh sample of 34 mg/1
(984H Filter) and 43 mg/1 (934AH Filter). On the basis of these results and
the COD analyses in Tables 3 and 4, it is apparent that the colloidal material
in the District of Columbia primary effluent contributes 50-70 percent of the
influent carbonaceous material. Although many more analyses than the few
reported here would be required to reliably establish this percentage over a
24-hour diurnal cycle, it is apparent that an analysis of effluent quality
solely on the basis of soluble influent materials neglects more than 50%
of the total organic load to the system. Furthermore establishing the solu-
ble component of primary effluent precludes the use of the normal compositing
procedure because of the large changes which occur during storage.
Only one effluent sample was examined to evaluate changes in the soluble
component during storage. A grab sample of effluent from the step aeration
29
-------
TABLE 3 . COD AND SOLUBLE COD VALUES OF D.C. PRIMARY EFFLUENT
BEFORE AND AFTER REFRIGERATED AND ACIDIFIED STORAGE
SAMPLE
'INITIAL VALUES
* 984 H (Ultra) Filter
** 934 AH Filter
24 HOUR VALUES
DATE
1974
Feb.
Feb.
Feb.
Feb.
Feb.
Feb.
Feb.
Feb.
5
7
13
14
21
21
25
25
Total
cop
mg/1
201
183
207
211
222
222
186
186
Filtered
COD
mg/1
105**
64.9*
60.2*
76.7*
88.4**
69.8*
90.3**
72.0*
Total
COD
mg/1
189
186
191
222
232
232
192
192
Filtered
COD
mg/1
52.9**
40.6*
35.9*
55.4*
53.2**
53.6*
51.7**
39.1*
30
-------
TABLE 4. TOTAL AND FILTERED COD VALUES OF D.C.
PRIMARY EFFLUENT GRAB SAMPLES OBTAINED
AT 0800-0900 HOURS.
DATE
1974
4/22
4/23
4/24
4/25
4/26
4/30
5/1
5/2
5/7
5/8
5/9
5/13
5/14
5/15
5/16
7/5
7/10
Average
Std. Dev.
TOTAL
172
174
192
199
240
204
219
185
242
244
208
188
206
184
232
205
—
206
23.7
* 984 H Glass Fiber Filter
COD, mg/1
FILTERED*
63.8
72.0
77.5
75.2
76.1
83.6
93.3
55.1
73.2
81.4
82.2
72.0
76.7
75.1
85.8
75.0
76.2
76.1
8.5
31
-------
process had a total COD of 47.1 mg/1 and filtered COD's of 27.7 (934AH
filter) and 27.2 mg/1 (984H filter). After 24 hours of refrigerated and
acidified storage, the total COD was 47.3 mg/1 with the filtered COD's
decreasing to 23.9 (934AH) and 20.5 mg/1 (984H). This represents a
decrease of 14% and 25%, respectively.
The above results clearly indicate that analysis of the soluble com-
ponent of composite samples requires that each portion of the-composite
sample be filtered as soon as the sample is collected. If these filtered
samples were then composited, it would be possible to realistically deal
with soluble COD values in the interpretation of average system performance.
Since manpower did not exist to routinely filter a portion of the influent
and effluent samples immediately after collection, this aspect of the study
was limited to occassionally collecting and filtering grab samples.
B. General
According to the original research plan, the three activated sludge
units were to be operated at "identical" loadings during the periods when
steady-state data were collected. Because of adverse bacterial growths, a
need to change flow rates, occassional mechanical problems, etc., it became
apparent rather early in the study that a great deal of time was being spent
unproductively trying to bring all systems to the same set of equilibrium
conditions. Therefore the research approach was modified so that each system
was operated at whatever loading seemed advantageous at the time without
regard to the loadings or conditions prevailing in the other systems.
Because this was the predominant mode of operation during most of the study,
the basic results from each system will be presented separately in chrono-
logical order. Once the basic information has been presented, the inter-
relationships, similarities, dissimilarities, etc. among the various units
will be explored.
C. Step Feed System
The step feed system was seeded with D.C. secondary solids in November
1973 and the system was operated through December 1974. The major opera-
tional characteristic that made this system superior to the other two, is
indicated in Table 5. Throughout the 13 months of operation summarized, the
clarifier bed level was always near the bottom of the 3.35 m (11 ft) deep
clarifier. In addition to the obvious advantages of this situation, the need
to consider large changes in clarifier storage when calculating material
balances or sludge production was obviated. Thirty minute sludge volumes in
the fourth reactor pass were determined three times per day using one-liter
cylinders and averaged to determine a daily sludge volume. This was used
in conjunction with the laboratory suspended solids analyses of the fourth
pass to calculate an average daily SVI. The daily SVI's were averaged, over
5-day periods and these results are shown in Figure 4. The relatively small
changes over long time periods can be observed. Throughout the study, the
fourth pass mixed liquor settling velocities measured in the 2.3 m x 0.15 m
stirred column were satisfactory (Table 6). This brief summary of system
performance illustrates, in a very general way, that the step feed system
32
-------
TABLE 5 . CLARIFIER BED LEVELS AND STABILITY
WITH THE STEP FEED SYSTEM
MONTH
DEPTH OF BED
BELOW SURFACE
BED LEVEL
CHARACTERISTICS
meters
1973:
December
1974:
January
February
March
April
May
June
July
August
September
October
November
December
2.3-3.0
2.4-3.0
2.4-3.0
2.7-3.0
3.0-3.2
2.1-3.0
3.0-3.2
3.0-3.2
2.9-3.2
3.0-3.2
3.0-3.2
3.0-3.2
3.0-3.2
stable
stable
stable
stable
very stable
gradual increase followed
by gradual decrease
very stable
very stable
very stable
very stable
very stable
very stable
very stable
33
-------
300--
60
70 80 90 100 110 120 130 140 150 160 170
CO
-p.
O 300 T
LU
O 200
Q
3
100 --
180
200
220
H I—
240 260
DAY, 1974
280
300
320
340
Figure 4. Variation of 4th Pass SVI's in the Step Feed System.
-------
TABLE 6. MIXED LIQUOR SETTLING VELOCITIES
IN THE STEP FEED SYSTEM.
DATE
1974
1-14
1-21
1-28
2-4
2-11
2-19
2-25
3-5
3-18
3-25
4-8
4-15
4-22
4-29
5-6
5-20
6-3
6-10
6-24
7-9
7-15
7-22
8-5
8-19
8-26
9-3
9-10
9-16
SUSPENDED
SOLIDS
rng/1
2250
2300
3150
2850
3150
3000
2800
2700
2450
2600
2150
2650
2700
2650
2600
1950
1600
950
1550
1600
1150
1350
1700
1650
1600
1400
1450
1300
TEMPERATURE
QC
14.5
15.5
16.5
16.0
14.5
15.5
15.5
18.0
17.0
17.0
18.0
19.5
20.5
21.0
20.0
23.0
22.5
24.0
24.5
26.5
27.0
27.0
27.5
27.5
27.0
27.0
25.5
24.5
STIRRING
SPEED
rph
0
0
13.5
13.5
13.5
15
15
15
14
15
15
15
15
15
15
15
15
15
20
15
15
15
15
15
15
15
15
15
SETTLING
VELOCITY
ft/hr
11.3
11.3
8.0
11.9
13.6
14.8
11.1
12.1
10.8
11.5
16.9
14.1
12.8
11.8
8.6
11.0
18.0
28.5
22.4
22.1
19.3
18.0
13.6
13.5
26.8
24.8
18.8
19.0
m/hr
3.4
3.4
2.4
3.6
4.1
4.5
3.4
3.7
3.3
3.5
5.2
4.3
3.9
3.6
2.6
3.4
5.5
8.7
6.8
6.7
5.9
5.5
4.1
4.1
8.2
7.6
5.7
5.8
35
-------
DATE
1974
9-26
9-30
10-10
10-15
10-22
10-29
11-4
n-n
11-18
11-25
12-2
12-9
12-16
SUSPENDED
SOLIDS
mg/1
1600
1450
1600
1600
1350
1300
1350
1350
1400
1450
1300
1250
1050
TABLE 6.
(Continued)
TEMPERATURE STIRRING
°C
24.0
22.0
23.5
24.0
22.5
21.5
23.0
21.5
21.0
20.0
17.5
18.5
18.5
SPEED
rph
15
15
15
15
15
15
15
15
15
15
15
15
15
SETTLING
VELOCITY
ft/hr
10.3
9.5
12.8
21.5
13.6
24.5
29.0
19.9
26.5
14.6
14.4
56.0
24.5
m/hr
3.1
2.9
3.9
6.6
4.1
7.5
8.8
6.1
8.1
4.5
4.4
17.1
7.5
36
-------
was quite simple to operate and relatively insensitive to the wide variations
in loading. A more detailed summary of system performance will now be pre-
sented.
The step feed system was started on November 8, 1973 by seeding with
secondary solids from the main D.C. plant. The biomass increased rapidly
and wasting was initiated on the 16th. The system presented no major oper-
ating problems during November or December. Waste rates were varied period-
ically since an attempt was being made to balance the loading to the three
activated sludge systems at this time. These periodic changes were made until
January 11, 1974 when a constant flow and hydraulic waste rate were set. The
reactor solids levels quickly stabilized, and the first steady-state operat-
ing data to be reported cover the period from February 1 through March 7.
These data are summarized in Tables 7 and 8. The reactor solids levels were
quite stable during this 35-day period. The system SRT based on the average
solids under aeration was 9.0 days, and this corresponds to a F/M ratio of
0.17 g BODc/g MLVSS/day. The influent mass of phosphorus divided by the
effluent plus waste total mass was 0.99. The influent mass of TKN divided
by the sum of the effluent plus waste mass was 1.03. These represent ex-
ceptionally good materials balances.
On the 8th of March large amounts of floating "scum" were observed on
top of the final clarifier. The material overflowed the top of the center-
well and floated across the clarifier surface. The material bridged the V-
notched weir around the clarifier periphery and after about a day completely
covered the clarifier surface to a depth of 3-5 cm (1-2 inches). During the
middle of March, Dr. Ron Lewis visited the pilot plant and examined a number
of slides of this material. The organism responsible for the copious
quantities of scum was visually identified as a species of Nocardia, an
actinomycetes. The scum or form consisted of a mass of hyphae of Nocardia
in which air bubbles were trapped. This mass also contained a number of
other microorganisms. A typical photomicrograph of this biomass is pre-
sented in Figure 5. The foaming problem resulting from Nocardia is not
unique to operation on D.C. wastewater, and has been observed at a number
of other plants (45).
The presence of large quantities of biological solids floating across
the clarifier surface effectively prevented measuring effluent quality or
solids production. Since the foam was primarily generated by overflowing
the top of the center-well in the clarifier, it was not possible to deter-
mine the quantity of solids lost by measuring the suspended solids in
clarifier effluent. If this material was left on the surface for a day or
so it became thick enough to "filter" the effluent leaving the clarifier and
an exceptionally good effluent quality could be obtained. Because of all of
these problems, no effluent quality data will be presented during any time
that the Nocardia problem was severe enough to produce foam on the clarifier
surface.
The hydraulic waste rate was increased a small amount on March 20 and
maintained at a constant value until May 8. No changes in influent flow
were made. Excessive foam was present throughout March and the beginning
37
-------
TABLE 7. PROCESS CHARACTERISTICS FOR THE STEP FEED SYSTEM
AT A 9.0 DAY SRT (FEB.l - MARCH 7, 1974).
Aeration Time, hrs.
Influent Flow, m3/day (gpd)
Recycle Flow Rate, 1/min (gpm)
Clarifier Overflow Rate, m/day (gpd/ ft2)
SVI, ml/gm
MLSS Pass 1, mg/1
MLVSS Pass 1, mg/1
MLSS Pass 2, mg/1
MLVSS Pass 2, mg/1
MLSS Pass 3, mg/1
MLVSS Pass 3, mg/1
MLSS Pass 4, mg/1
MLVSS Pass 4, mg/1
Average MLSS, mg/1
Average MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g BODs Applied
Solids Production, g VSS/g BOD5 Applied
Solids Production, g SS/g BODs Removed
Solids Production, g SS/g COD Applied
Solids Production, g VSS/g COD Applied
Solids Production, g SS/g COD Removed
Solids Production, g SS/g 6005 Removed
(excluding effluent solids)
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent,
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Recycle COD, mg/1
Recycle P04, mg/1
Recycle TKN, mg/1
38
Value or
Average
3.6
185.9 (49,100)
30 (8)
20.8 (510)
134
13,330
9820
5640
4200
3990
2920
3000
2210
6490
4790
9.0
0.17
0.85
0.64
0.97
0.42
0.32
0.55
0.82
0.47
0.35
107
14.9
14,740
11,130
16,290
1236
1145
Standard
Deviation
29
549
460
268
200
274
227
166
142
952
543
1180
114
122
-------
TABLE 8. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
STEP FEED SYSTEM AT A 9.0 DAY SRT (FEB. 1 -
MARCH 7, 1974)
BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, C
Influent Effluent
Mean Std. Mean Std.
Devtat. Deviat.
125
252
27.2
0
21.1
123
89
15.8
12.3
20.0
2.2
0
1.4
23.4
16.8
0.8
15.0
58.2
19.4
0
13.8
16
12
4.8
10.9
1.7
0
2.3
3.2
3.3
39
-------
Figure 5. Photomicrograph of Foam Caused by Nocardia.
40
-------
of April. By the middle of April the floating solids were no longer a prob-
lem, and a "measurable" effluent quality was obtained until May 8 when the
clarifier once again became covered with large quantities of foam. Data
obtained during the 22-day period of April 16-May 7 are summarized in Tables
9 and 10. The reactor solids levels during this time were quite stable.
The flow rate was not as stable as desired but varied from 114-132 1/min
(30-35 gpm). This had no measurable impact on effluent quality. The results
are quite consistent with the data obtained in the first period of steady-
state operation.
System operation during the later part of May, June and the first half
of July did not produce any usable steady-state data. Although the Nocardia
concentration was reduced below problem levels by the end of May there were
enough failures in the flow control system during June and the first half of
July to preclude considering any of the data as representative of steady-
state operation. Throughout this period the hydraulic waste rate was con-
stant.
By August 11 the system had operated under stable conditions for about
20 days and a new set of steady-state data were collected. These data cover
the period from August 11 to September 4 and are summarized in Tables 11 and
12. The flow during this 25-day period was very stable, and the reactor
solids level was also quite stable. Effluent quality was extremely good and
there were no traces of foam on the clarifier.
On September 5 the waste rate was increased considerably to move to a
new set of equilibrium conditions. The next steady-state operating data that
were obtained cover the period of September 18-October 19. The results during
this 32-day period are summarized in Tables 13 and 14. There were no foam
production problems during this period and hence no difficulty in character-
izing effluent quality.
The waste rate was increased again on October 20 and the final period of
steady-state operation with this system was obtained during the period of
November 12-December 12. The results obtained during this 31-day period are
summarized in Tables 15 and 16. Although the influent flow was stable, it
was somewhat higher than desired. Hence the process loading was only slightly
different than that observed during the previous steady-state operation. The
effluent parameters and the sludge production values are also essentially the
same. The phosphorus mass balance (ratio of influent to effluent plus waste
mass) was the poorest obtained from any of the five steady-state periods
and was 1.17.
The five periods of steady-state operation with the step feed system are
summarized in Table 17. A discussion of these data will be deferred until
the results from the other two systems are presented. However, it should be
noted that the effluent quality at the 8.0 and 9.0 day SRT values could have
been shown to be much worse by picking steady-state periods with excessive
foam production and consequently high effluent suspended solids.
41
-------
TABLE 9. PROCESS CHARACTERISTICS FOR THE STEP FEED SYSTEM
AT AN 8.0 DAY SRT (APRIL 16 - MAY 7, 1974)
Aeration Time, hrs.
Influent Flow, m3/day (gpd)
Recycle Flow Rate, 1/min (gpm)
2
Clarifier Overflow Rate, m/day (gpd/ft )
SVI, ml/gm
MLSS Pass 1, mg/1
MLVSS Pass 1, mg/1
MLSS Pass 2, mg/1
MLVSS Pass 2, mg/1
MLSS Pass 3, mg/1
MLVSS Pass 3, mg/1
MLSS Pass 4, mg/1
MLVSS Pass 4, mg/1
Average MLSS, mg/1
Average MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g BODc Applied
Solids Production,
Solids Production,
Solids Production,
Solids Production.
Solids Production.
r
g VSS/g BODs Applied
g SS/g'BODs Removed
g SS/g COD Applied
g VSS/g COD Applied
g SS/g COD Removed
Solids Production, g SS/g 6005 Removed
(excluding effluent solids)
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent, %
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Recycle COD, mg/1
Recycle PO^ mg/1
Recycle TKN, mg/1
42
Value or
Average
3.8
179.4 (47,400)
38 (10)
20.1 (494)
117
9930
7340
4770
3510
3460
2590
2530
1870
5170'
3830
8.0
0.20
0.84
0.63
0.96
0.40
0.30
0.46
0.81
0.39
0.29
100
14.9
10,390
7700
11,440
815
799
Standard
Deviation
22
535
469
297
267
298
220
220
182
393
292
792
46
91
-------
TABLE 10. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
STEP FEED SYSTEM AT AN 8.0 DAY SRT (APRIL 16 -
MAY 7, 1974).
BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, C
Influent Effluent
Mean
119
252
27.0
0
22.7
124
95
20.0
Std.
Devi at.
10.5
14.5
2.1
0
1.8
16.8
12.6
0.9
Mean
14.0
33.8
0.4
—
15
12
Std.
Deviat.
5.3
4.3
—
0.21
2.5
3.5
43
-------
TABLE n.
INSTEP FENSYST*
Aeration Time, hrs.
Influent Flow, m3/day (gpd)
Recycle Flow Rate, 1/min (gpm)
Clarifier Overflow Rate, m/day (gpd/ft )
SVI, ml/gm
MLSS Pass 1, mg/1
MLVSS Pass 1 , mg/1
MLSS Pass 2, mg/1
MLVSS Pass 2, mg/1
MLSS Pass 3, mg/1
MLVSS Pass 3, mg/1
MLSS Pass 4, mg/1
MLVSS Pass 4, mg/1
Average MLSS, mg/1
Average MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g BODc Applied
Solids Production, g VSS/g BOD5 Applied
Solids Production, g SS/g BOD5 Removed*
g SS/g COD Applied
g VSS/g COD Applied
g SS/g COD Removed
Solids Production, g SS/g BOD5 Removed*
(excluding effluent solids)
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent,
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, rng/1
Recycle COD, mg/1
Recycle P04, mg/1
* Calculated Using Inhibited BOD Values
44
Solids Production
Solids Production
Solids Production
Value or
Average
3.4
196.1 (51,800)
38 (10)
22.0 (540)
129
6540
4650
3100
2175
2175
1555
1650
1205
3365
2395
5.9
0.31
0.77
0.55
0.81
0.37
0.27
0.42
0.75
0.39
0.28
81.7
8.2
7470
5355
8180
485
Standard
Deviation
15.5
335
241
216
157
142
106
149
107
383
300
645
80
-------
TABLE 12. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
STEP FEED SYSTEM AT A 5.9 DAY SRT (AUG. 11-
SEPT. 4, 1974).
Influent
Effluent
Mean
Std.
Devi at.
Mean
Std.
Devi at.
BOD, mg/1
Inhibited BOD, mg/1
COD, mg/1
TKN, mg/1
(N09 + NOJ-N, mg/1
<— O
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, C
106
—
218
22.3
0
18.5
118
91
27.0
15.7
—
20.8
1.1
0
2.8
16.8
11.4
0.9
16.2
5.7
25.3
7.6
11.8
6.7
4.5
6.0
2.1
3.9
—
2.8
1.7
2.6
2.3
45
-------
TABLE 13. PROCESS CHARACTERISTICS FOR THE STEP FEED SYSTEM
AT A 4.1 DAY SRT (SEPT. 18 - OCT. 19, 1974).
Aeration Time, hrs.
o
Influent Flow, m /day (gpd)
Recycle Flow Rate, 1/rnin (gpm)
2
Clarifier Overflow Rate, rn/day (gpd/ft )
SVI, ml/gm
MLSS Pass 1, mg/1
MLVSS Pass 1, mg/1
MLSS Pass 2, mg/1
MLVSS Pass 2, mg/1
MLSS Pass 3, mg/1
MLVSS Pass 3, mg/1
MLSS Pass 4, mg/1
MLVSS Pass 4, mg/1
Average MLSS, mg/1
Average MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g BOD5 Applied
Solids Production,
Solids Production.
Solids Production.
Solids Production.
Solids Production.
g VSS/g BODs Applied
g SS/g BODs Removed
g SS/g COD Applied
g VSS/g COD Applied
g SS/g COD Removed
Solids Production, g SS/g BOD5 Removed
(excluding effluent solids)
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent,
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Recycle COD, mg/1
Recycle P04, mg/1
46
Value or
Average
3.4
196.1 (51,800)
38 (10)
22.0 (540)
210
5260
4030
2640
2015
1800
1400
1425
1105
2780
2140
4,1
0.41
0.77
0.59
0.85
0.37
0.28
0.44
0.77
0.40
0.30
96.8
9.3
6230
4770
7550
480
Standard
Deviation
60
371
298
135
118
114
95
78
75
403
359
647
71
-------
TABLE 14. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
STEP FEED SYSTEM AT A 4.1 DAY SRT (SEPT. 18 -
OCT. 19, 1974).
Influent Effluent
Mean Std. Mean Std.
Deviat. Devi at.
BOD, mg/1 126 13.5 12.7 3.6
Inhibited BOD, mg/1 --- 11-6 3.8
COD, mg/1 261 26.2 38.9 5.4
TKN, mg/1 28.5 2.5
(N02 + N03)-N, mg/1 0 0 0.4 0.50
P04, mg/1 21.1 2.7 13.0 1.2
SS, mg/1 116 15.9 9.1 3.7
VSS, mg/1 89 14.1 6.8 3.1
Temperature, C 22.9 1.1
47
-------
TABLE 15. PROCESS CHARACTERISTICS FOR THE STEP FEED SYSTEM
AT A 3.7 DAY SRT (NOV. 12 - DEC. 12, 1974)
Aeration Time, hrs.
o
Influent Flow, m /day (gpd)
Recycle Flow Rate, 1/min (gpm)
O
Clarifier Overflow Rate, m/day (gpd/ft )
SVI, ml/gm
MLSS Pass 1 , mg/1
MLVSS Pass 1, mg/1
MLSS Pass 2, mg/1
MLVSS Pass 2, mg/1
MLSS Pass 3, mg/1
MLVSS Pass 3, mg/1
MLSS Pass 4, mg/1
MLVSS Pass 4, mg/1
Average MLSS, mg/1
Average MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g BODs Applied
Solids Production, g VSS/g BODs Applied
Solids Production, g SS/g BODs Removed
g SS/g COD Applied
g VSS/g COD Applied
g SS/g COD Removed
Solids Production
Solids Production
Solids Production
Solids Production, g SS/g BOD5 Removed
(excluding effluent solids)
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent.
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Recycle COD, mg/1
Recycle P04> mg/1
Value or
Average
3.1
214.3 (56,600)
38 (10)
24.0 (590)
189
4770
3570
2640
1975
1605
1220
1305
990
2580
1940
3.7
0.46
0.77
0.58
0.87
0.36
0.27
0.43
0.77
0.38
0.29
90.9
12.1
5600
4210
7225
478
Standard
Deviation
33
400
433
292
199
148
157
126
125
531
520
663
117
48
-------
TABLE 16. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
STEP FEED SYSTEM AT A 3.7 DAY SRT (NOV. 12-
DEC. 12, 1974).
BOD, mg/1
Inhibited BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, C
Influent Effluent
Mean Std. Mean Std.
Devi at. Devi at,
118
—
253
28.5
0
21.9
138
107
19.2
16.4
27.1
3.2
0
3.8
25.4
27.9
1.2
14.1
11.1
42.7
1.2
12.0
11.2
8.2
5.5
4.6
9.1
—
1.1
2.4
4.4
3.2
49
-------
TABLE 17. SUMMARY OF SYSTEM OPERATION AND
PERFORMANCE WITH THE STEP FEED SYSTEM.
STEADY STATE PERIODS
Days at Equilibrium 31 32 25 22 35
Average Flow, m3/day 214.3 196.1 196.1 179.4 185.9
Average Flow, gpd 56,600 51,800 51,800 47,400 49,100
Detention Time, hrs. 3.1 3.4 3.4 3.8 3.6
Average Temperature, °C 19.2 22.9 27.0 20.0 15.8
SVI, ml/gm 189 210 129 117 134
SRT, days 3.7 4.1 5.9 8.0 9.0
F/M, g BOD5 Applied/g MLVSS/day 0.46 0.41 0.31 0.20 0.17
Solids Production, g SS/g BODs Applied 0.77 0.77 0.77 0.84 0.85
Solids Production, g SS/g COD Applied 0.36 0.37 0.37 0.40 0.42
Solids Production, g VSS/g BOD5 Applied 0.58 0.59 0.55 0.63 0.64
Solids Production, g VSS/g COD Applied 0.27 0.28 0.27 0.30 0.32
Effluent BOD5, mg/1 14.1 12.7 16.2 14.0 15.0
Effluent Inhibited BOD5, mg/1 11.1 11.6 5.7
Effluent COD, mg/1 42.7 38.9 25.3 33.8 58.2
Effluent SS, mg/1 11.2 9.1 6.7 15 16
Effluent VSS, mg/1 8.2 6.8 4.5 12 12
Effluent P04, mg/1 12.0 13.0 11.8 — 13.8
Effluent (N02 + N03)-N, mg/1 1.2 0.4 7.6 0.4 0
Waste Sludge, % P 2.8 2.5 2.1 2.6 2.7
Waste Sludge, % TKN — — — 7.7 7.8
P Balance, Mass In ^ Mass Out 1.17 1.08 1.11 -- 0.99
N Balance, Mass In -e- Mass Out -- -- -- *-- 1.03
Volatile Solids, % 75.3 77.0 71.5 74.1 74.1
Recycle COD/Recycle SS 1.29 1.21 1.10 1.10 1.11
Recycle COD/Recycle VSS 1.72 1.58 1.53 1.49 1.46
50
-------
D. Plug Flow System
The plug flow system was started on November 8, 1973 by seeding with
secondary sludge from the D.C. treatment plant. The biomass was quickly
established, and sludge wasting was initiated by the 16th. Reactor solids
varied between 2000-3000 mg/1 during the later part of November and through-
out December. The SVI's increased to near 300 ml/gm by the end of December,
and \\2®2 (25 mg/1) was added on the 4th and 6th of January to reduce the
amount of filamentous growth (46). This was effective and the SVI's decreased
to 150 ml/gm by the 9th of the month. The flow and waste concentrations were
stable during the middle of January with a process loading of about 0.3 g
BOD5 applied/g MLVSS/day between January 9-21. By the 20th of the month the
Nocardia problem began to develop and it was particularly bad during the
last week of January.
The flow and waste rate were reduced considerably during the last 10 days
of January to move to low F/M conditions. During the first 13 days of Feb-
ruary the flow was maintained at 76 1/min (20 gpm). It was then gradually
increased in 19 1/min (5 gpm) increments until it reached 132 1/min (35 gpm)
on February 26. This caused the bed to overflow the clarifier whereupon the
flow was reduced. The MLSS varied between 4500 and 5500 mg/1 during the
month. The F/M ratio for the first half of the month averaged 0.13 g BODr/
g MLVSS/day and averaged 0.18 for the entire month. Throughout thei month
there were very heavy quantities of Nocardia on the surfaces of the final
clafifier. Large quantities of solids were pushed over the weirs as much as
4-6 times per day. Ho02 addition (25 mg/1) on the 6th and 8th of the month
had no impact on the Nocardia formation. During the first 19 days of March
the average flow was 148 nvVday (39,000 gpd) with an average process loading
of 0.14 g BOD5/g MLVSS/day. The Nocardia problem continued unabated. In
summary nearly two months of operation at a F/M loading in the range of
0.13-0.18 g BODs/g MLVSS/day resulted in so much Nocardia formation that there
was no way to realistically assess either effluent quality or suspended solids
formation.
Both the flow and waste rate were increased on the 20th of March to move
to somewhat higher loading conditions, and constant flow and volumetric
waste rates were maintained until the 30th of April. The reactor solids level
was very stable between April 7-25 and the Nocardia production was not suffi-
cient during this period to interfere with obtaining representative samples
of clarifier effluent. Unfortunately, the decrease in Nocardia production
was only temporary and by the 27th of the month the top of the clarifier again
had 3-5 cm (1-2 inches) of foam across the surface. Data obtained during the
19-day period of April 7-25 are summarized in Tables 18 and 19. The compre-
hensive effluent analysis was not .started until April 21 since the intent at
the time was to allow a longer transition period to the new steady-state
conditions before more extensive data collection. However, there were about
2.5 turnovers under the constant flow and waste conditions prior to the
steady-state period summarized. In view of the extreme stability of the re-
actor solids during this 19-day period, the data should be representative
of stable operation at a 6.6 day SRT. As with the step feed system, the SRT
calculation was based only on the average reactor solids concentration and
did not include the solids stored in the clarifier.
51
-------
TABLE 18. PROCESS CHARACTERISTICS FOR THE PLUG FLOW
SYSTEM AT A 6.6 DAY SRT (APRIL 7-25, 1974).
Value or
Average
Standard
Deviation
Aeration Time, hrs.
q
Influent Flow, m /day (gpd)
Recycle Flow Rate, 1/min (gpm)
Clarifier Overflow Rate, m/day (gpd/ft )
SVI, ml/gm
MLSS, mg/1
MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g BOD5 Applied
Solids Production, g VSS/g BODg Applied
Solids Production, g SS/g BODg Removed
g SS/g COD Applied
g VSS/g COD Applied
g SS/g COD Removed
Solids Production
Solids Production
Solids Production
Solids Production, g SS/g BOD,- Removed
(excluding effluent solids)
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent,
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Recycle COD, mg/1
3.5
198.0 (52,300)
61-68 (16-18)
22.2 (545)
147 26.3
4080 220
2980 198
6.6
0.26
0.82
0.60
0.90
0.38
0.28
0.45
0.70
0.35
0.26
90.5
22.0
12,610 290
9300 375
13,860 728
52
-------
TABLE 19. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE PLUG FLOW
SYSTEM AT A 6.6 DAY SRT (APRIL 7-25, 1974)
Influent Effluent
BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
111
238
25.4
0
20.8
113
87
19.0
Std.
Devtat.
10.5
29.4
2.8
0
2.7
23.1
12.9
1.0
Mean
10.7
38.0
0*
20
14
Std.
Devi at,
3.8
6.3
—
0
—
8.4
6.3
April 21-25 only
53
-------
Flow was shut off to the process for a few hours on April 30 and again
on May 13 to modify the center-well in the clarifier. The center-well was
covered and a seal installed between the cover and drive shaft. This pre-
vented the Nocardia from overflowing the top of the center-well. The first
seal did not work well, but the second proved quite effective. This immedi-
ately produced nearly a three-fold increase in the measured effluent suspended
solids concentration. Also on the 13th, the waste rate was increased sub-
stantially to move to higher loading conditions and attempt to eliminate the
Nocardia problem entirely. The MLSS dropped from 4000 mg/1 to a stable 2000
mg/1 during the last 10 days of May. The Nocardia problem also decreased
considerably and by the end of the month the effluent suspended solids were
only 15-20 mg/1. The clarifier bed level also declined considerably during
the first half of May, but remained very steady and about 0.3 m (1 ft) from
the bottom of the clarifier during the last third of the month.
Conditions were stable during the period of June 1-July 11, and this
steady-state operation is summarized in Tables 20 and 21. The clarifier bed
level was also very stable during this period. Nitrification slowly increased
and the (N02 + N03)-N levels rose from 3-5 mg/1 at the beginning of June to
8-10 mg/1 by early July. Throughout this period there was a gradual rise in
wastewater temperature. The partial nitrification is what accounts for the
relatively high effluent BODs values. Throughout this period of operation
there was some foam on the reactor surface, but the Nocardia concentration
was sufficiently reduced to pose no problems in clarification. The effluent
suspended solids levels were also typical of those normally encountered in
activated sludge systems. The ratio of the influent to effluent plus waste
phosphorus mass was 1.04 for the period of June 13-July 11.
The volumetric waste rate was increased on July 12 to move to somewhat
higher loading conditions. The clarifier bed level remained very stable
within 0.3 m (1 ft) of the bottom until July 25 when the bed began to rise
rapidly. It was necessary to reduce the flow to 114 1/min (30 gpm) to main-
tain the solids in the clarifier. The bed began to fall by August 3 and by
August 9 the bed level was again within 0.3 m (1 ft) of the clarifier bottom.
Operation at the reduced flow was continued, and the next period 'of steady
operation was obtained between August 16-September 12. This operation is
summarized in Tables 22 and 23. Throughout this period there was essentially
no change in the clarifier bed level. Nitrification was nearly complete for
the first 3/4 of the steady-state period, but declined to (N02 + NOaJ-N
levels of 1-2 mg/1 during the last week of steady-state operation. The drop
correlates with a decline in wastewater temperature. The ratio of the
influent mass of phosphorus to the effluent plus waste total mass was 0.99
which represents an excellent balance.
On September 13 both the influent flow rate and the waste rate were
changed to gain operating data under a new set of equilibrium conditions.
Operation in this low SRT range allowed for 3-4 sludge turnovers within
10 days, and a new period of steady-state operation was obtained from Septem-
ber 24 thru October 21. Throughout this 28-day period the clarifier bed level
remained within 0.3 m (1 ft) of the clarifier bottom but the SVI's cycled
considerably. The operation during this period is summarized in Tables 24 and
54
-------
TABLE 20. PROCESS CHARACTERISTICS FOR THE PLUG FLOW SYSTEM
AT A 4.4 DAY SRT (JUNE 1 - JULY 11, 1974).
Value or Standard
Average Deviation
Aeration Time, hrs. 3.8
Influent Flow, m3/day (gpd) 180.6 (47,700)
Recycle Flow Rate, 1/nrin (gpm) 53-61 (14-16)
Clarifier Overflow Rate, m/day (gpd/ft2) 20.2 (497)
SVI, ml/gm 117 21.8
MLSS, mg/1 2320 269
MLVSS, mg/1 1760 192
SRT, days 4.4
F/M, g BOD5 Applied/g MLVSS/day 0.42
Solids Production, g SS/g BOD5 Applied 0.71
Solids Production, g VSS/g BOD5 Applied 0.54
Solids Production, q SS/g BOD5 Removed 0.91
Solids Production, g SS/g COD Applied 0.37
Solids Production, g VSS/g COD Applied 0.28
Solids Production, g SS/g COD Removed 0.43
Solids Production, g SS/g BOD§ Removed 0.76
(excluding effluent solids;
Solids Production, g SS/g COD Removed 0.36
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.27
(excluding effluent solids)
Solids Production, mg/1 84.0
Suspended Solids Production in Effluent, % 15.8
Recycle Suspended Solids, mg/1 7060 701
Recycle Volatile Suspended Solids, mg/1 5345 481
Recycle COD, mg/1 7670 951
Recycle P04, mg/1* 498 65.6
Recycle TKN, mg/1* 564 81 .1
Recycle Suspended Solids, mg/1* 7270 674
* June 13-July 11 only
55
-------
TABLE 21. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE PLUG
FLOW SYSTEM AT A 4.4 DAY SRT (JUNE 1-JULY 11, 1974)
Influent Effluent
BOD, mg/1
COD, mg/1
TKN, mg/1*
(N02 + N03)-N, mg/1
P04, mg/1*
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
118
229
23.6
0
20.3
108
85
24.0
Std.
Devi at.
20.4
23.0
2.0
0
2.3
17.3
14.9
1.0
Mean
25.5
31 .9
6.0
6.7
14.5
13.4
10.1
Std.
Devi at
10.5
5.5
2.45
2.5
1.8
6.7
5.4
* June 13-July 11 only
56
-------
TABLE 22. PROCESS CHARACTERISTICS FOR THE PLUG FLOW SYSTEM
AT A 2.9 DAY SRT (AUG. 16-SEPT. 12, 1974).
Value or
Average
Standard
Deviation
Aeration Time, hrs.
Influent Flow, m /day (gpd)
Recycle Flow Rate, 1/min (gpm)
Clarifier Overflow Rate, m/day (gpd/ft )
SVI, ml/gm
MLSS, mg/1
MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g BOD5 Applied
Solids Production, g VSS/g BOD5 Applied
Solids Production, g SS/g BODr Removed*
g SS/g COD Applied
g VSS/g COD Applied
g SS/g COD Removed
Solids Production, g SS/g BOD,- Removed*
(excluding effluent solids)
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent,
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Recycle COD, mg/1
Recycle PO^, mg/1
Recycle TKN, mg/1
* Calculated Using Inhibited BOD Values
Solids Production
Solids Production
Solids Production
4.2
165.4 (43,700)
38 (10)
18.5 (455)
123 30.1
1550 97
1145 76
2.9
0.51
0.94
0.69
0.99
0.45
0.32
0.51
0.93
0.47
0.35
96.2
6.1
6140 311
4480 231
6875 631
408 57.6
468 46.9
57
-------
TABLE 23. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE PLUG
FLOW SYSTEM AT A 2.9 DAY SRT (AUG. 16-SEPT. 12, 1974)
Influent Effluent
Mean Std. Mean Std.
Devi at. Devi at.
BOD, mg/1
Inhibited BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
102
—
216
22.9
0
18.9
115
88
26.5
16.1
—
22.6
1.6
0
3.1
18.9
11.3
1.4
12.3
4.7
25.7
7.0
6.4
13.1
5.9
4.1
6.2
2.0
3.9
3.8
3.1
1.3
2.8
2.2
58
-------
TABLE 24. PROCESS CHARACTERISTICS FOR THE PLUG FLOW SYSTEM
AT A 1.9 DAY SRT (SEPT. 24-OCT. 21, 1974).
Value or Standard
Average Deviation
Aeration Time, hrs. 3.6
Influent Flow, m3/day (gpd) 191.9 (50,700)
Recycle Flow Rate, 1/nrin (gpm) 38 (10)
Clarifier Overflow Rate, m/day (gpd/ft2) 21.5 (528)
SVI, ml/gm 164 69
MLSS, mg/1 1375 78.5
MLVSS, mg/1 1065 73.8
SRT, days 1.9
F/M, g BOD5 Applied/g MLVSS/day 0.79
Solids Production, g SS/g BOD5 Applied 0.86
Solids Production, g VSS/g BOD5 Applied 0.66
Solids Production,, g SS/g BOD5 Removed 0.93
Solids Production, g SS/g COD Applied 0.42
Solids Production, g VSS/g COD Applied 0.32
Solids Production, g SS/g COD Removed 0.50
Solids Production, g SS/g BODg Removed 0.88
(excluding effluent solids)
Solids Production, g SS/g COD Removed 0.47
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.37
(excluding effluent solids)
Solids Production, mg/1 108.9
Suspended Solids Production in Effluent, % 5.7
Recycle Suspended Solids, mg/1 6200 296
Recycle Volatile Suspended Solids, mg/1 4840 286
Recycle COD, mg/1 8170 818
Recycle P04, mg/1 478 83.9
Recycle TKN, mg/1 604 81
59
-------
TABLE 25. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE PLUG
FLOW SYSTEM AT A 1.9 DAY SRT (SEPT. 24-OCT. 21, 1974).
BOD, mg/1
Inhibited BOD, mg/1
COD, mg/1
TKN, mg/1
(N0? + NO,)-N, mg/1
£. O
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Influent Effluent
Mean Std Mean Std
Deviat. Devi at
127
—
261
29.0
0
21.1
117
89
22.5
12.3
28.1
2.3
0
3.0
16.7
14.8
0.76
9.9
8.7
42.3
19.4
0
12.7
6.3
4.2
2.8
2.3
6.1
1.5
0
1.5
2.7
2.5
60
-------
25. This was the highest loading investigated with the plug flow system,
and it can be seen that the carbonaceous effluent quality was very good.
On October 22 the waste rate was reduced substantially to move to a new
set of equilibrium conditions. After a transition period of 15 days (three
turnovers), the next period of steady-state operation was initiated on Novem-
ber 7. Stable operating conditions were maintained until December 16 when
the clarifier bed level rose to the top of the clarifier and began overflow-
ing the weirs. This did not reflect an accumulation of biological solids
since the bed level remained within 0.3-0.9 m (1-3 ft) of the clarifier
bottom throughout November and the first 9 days of December. It then rose
rapidly over a five-day period until it reached the top of the clarifier.
Operation during the 38-day period of November 7-December 14 is summarized
in Tables 26 and 27. The ratio of the influent mass of phosphorus to the
effluent plus waste total mass was 1.13. Since the phosphorus balance during
nearly the same period of the step feed system (Nov. 12-Dec. 12) was 1.17,
the deviation from unity probably represents analytical error in the phos-
phorus determination rather than failure to accurately measure flow or waste
volumes.
When the clarifier bed began overflowing the weirs on December 16 the
influent flow was reduced to 57-76 1/min (15-20 gpm). \\2®2 addition (200-
250 mg/1) was initiated on the evening of the 17th and continued for 24 hours.
This reduced the bed level considerably and the flow was increased to 95 1/min
(25 gpm) on the 19th and the waste rate was reduced to move to a higher SRT.
The bed level gradually rose and was again at the top of the clarifier by
December 31 when the flow was reduced to 76 1/min (20 gpm). The average SVI
from December 19-31 was 298 ml/gm. Reactor solids during this period were
stable and the average process loading was 0.30 g 8005 applied/g MLVSS/day.
The flow was maintained at approximately 76 1/min (20 gpm) until Febru-
ary 18, 1975. On January 18, 1975 the District of Columbia began bypassing
all elutriation and thickener water around the primary clarifiers directly
to the head of secondary. This produced a change in influent wastewater
characteristics to the pilot plant. In addition to this change, reactor
solids and bed levels varied considerably but it is impossible to attempt to
relate the changes to any one given factor. In any event, the SVI's were
back to 50-70 ml/gm during the first half of February and the clarifier bed
level was again within 0.3 m (1 ft) of the bottom of the clarifier.
The next period of steady-state operation was obtained from March 14-
April 14, 1975. Prior to March 14, the system had been operated at constant
hydraulic flow and waste rates for 23 days. Results obtained during this
32-day period are summarized in Tables 28 and 29. The carbonaceous effluent
quality was rather poor because of the high concentrations of effluent sus-
pended solids caused by the presence of Nocardia. Fortunately, it was poss-
ible to obtain reasonably representative effluent samples during this time
period. Throughout this time period the clarifier bed level remained within
0.3 m (1 ft) of the clarifier bottom.
The SVI's began to rise rapidly beginning on April 15 and reached 200 ml/
gm by the end of the month. The clarifier bed level also rose over 1.8 m
61
-------
TABLE 26. PROCESS CHARACTERISTICS FOR THE PLUG FLOW SYSTEM
AT A 4.7 DAY SRT (NOV. 7-DEC. 14, 1974).
Aeration Time, hrs.
Influent Flow, m3/day (gpd)
Recycle Flow Rate, 1/nrin (gpm)
n
Clarifier Overflow Rate, m/day (gpd/ft )
SVI, ml/gm
MLSS, mg/1
MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g BOD5 Applied
Solids Production, g VSS/g BOD5 Applied
Solids Production, g SS/g BODg Removed
g SS/g COD Applied
g VSS/g COD Applied
g SS/g COD Removed
Solids Production
Solids Production
Solids Production
Solids Production, g SS/g BODr Removed
(excluding effluent solids)
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent,
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Recycle COD, mg/1
Recycle P04, mg/1
Recycle TKN, mg/1
Value or
Average
3.7
188.9 (49,900)
38 (10)
21.2 (520)
219
2995
2225
4.7
0.36
0.81
0.59
0.86
0.38
0.28
0.43
0.81
0.41
0.30
97.8
6.3
12,990
9550
14,710
1478
1145
Standard
Deviation
67
261
197
840
733
1570
198
212
62
-------
TABLE 27. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE PLUG
FLOW SYSTEM AT A 4.7 DAY SRT (NOV. 7-DEC. 14, 1974).
BOD, mg/1
Inhibited BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, C
Influent Effluent
Mean
121
—
257
28.3
0
21.6
140
109
19.4
Std.
Devi at.
16.7
26.7
3.2
0
3.9
24.3
25.8
1.4
Mean
7.6
5.1
31.1
16.5
0.8
8.7
6.2
4.6
Std.
Devi at
2.3
2.0
5.2
3.4
0.56
2.8
2.7
2.2
63
-------
TABLE 28. PROCESS CHARACTERISTICS FOR THE PLUG FLOW
SYSTEM AT A 5.7 DAY SRT (MARCH 14-APRIL 14, 1975).
Value or Standard
Average Deviation
Aeration Time, hrs. 4.2
Influent Flow, m3/day (gpd) 166.9 (44,100)
Recycle Flow Rate, 1/min (gpm) 38-42 (10-11)
Clarifier Overflow Rate, m/day (gpd/ft2) 18.7 (460)
SVI, ml/gm 90 14.8
MLSS, mg/1 2370 260
MLVSS, mg/1 1670 222
SRT, days 5.7
F/M, g BOD5 Applied/g MLVSS/day 0.37
Solids Production, g SS/g BODg Applied 0.71
Solids Production, g VSS/g BOD5 Applied 0.47
Solids Production, g SS/g BODg Removed* 0.81
Solids Production, g SS/g COD Applied 0.36
Solids Production, g VSS/g COD Applied 0.24
Solids Production, g SS/g COD Removed 0.45
Solids Production, g SS/g BODg Removed* 0.51
(excluding effluent solids)
Solids Production, g SS/g COD Removed 0.29
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.20
(excluding effluent solids)
Solids Production, mg/1 76.1
Suspended Solids Production in Effluent, % 36.7
Recycle Suspended Solids, mg/1 9485 1021
Recycle Volatile Suspended Solids, mg/1 6655 894
Recycle P04, mg/1 631 68.4
Recycle TKN, mg/1 691 117
* Calculated Using Inhibited BOD Values
64
-------
TABLE 29. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE PLUG
FLOW SYSTEM AT A 5.7 DAY SRT (MARCH 14-APRIL 14, 1975).
BOD, mg/1
Inhibited BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Influent Effluent
Mean
107
—
213
21.2
0
16.9
95
65
16.5
Std.
Devi at.
20.0
38.9
2.5
0
3.1
26.5
22.6
0.9
Mean
23.7
12.7
45.8
2.4
11.2
11.3
28
17
Std.
Devi at
9.1
5.7
7.2
0.39
1.2
2.2
12.6
8.6
65
-------
(6 ft) between April 19-20. It remained high for 5 days and then decreased
just as rapidly as it had risen to within 0.6 m (2 ft) of the clarifier
bottom. The SVI's remained high throughout the remainder of the month.
Throughout this period of changes the flow and waste rates remained the same
as during the previous 32-day period of stable operation.
Special studies were undertaken on May 1 and May 9, 1975 to establish
whether nitrification would occur at the head of a plug flow process in the
presence of adequate dissolved oxygen. These results will be presented in
detail in another report (47). No nitrification inhibition was observed.
Following these studies the flow rate was increased to 132 1/min (35 gpm) and
the waste rate was reduced in the hope of gathering operating data with a low
F/M loading. The reactor MLSS increased to 3700 mg/1 and remained at this
level from May 11-17. Unfortunately denitrification in the final clarifier
led to a severe rising sludge problem and made continued operation in this
mode impossible. Consequently the waste rate was doubled on May 24 and
increased an additional 10 percent on June 6.
The final period of steady-state operation to be reported was obtained
from June 15-July 8, 1975. Results are presented in Tables 30 and 31. During
this period the clarifier bed level remained within 0.3 m (1 ft) of the
clarifier bottom and the SVI did not fluctuate widely. Nocardia production
was insignificant and effluent suspended solids were low. Since the clarifier
bed level increased nearly 2.1 m (7 ft) on July 10, the period of steady-state
operation was not as long as desired. As observed previously, the bed re-
mained high for 4 days and then again returned to within 0.3 m (1 ft) of the
clarifier bottom over a two-day period.
Operation of the plug flow system was continued until mid-August when
the process was shut down. The results from one special study performed on
August 12-13, 1975 will be reported later.
Results obtained with the plug flow system during the seven steady-state
periods of operation are summarized in Table 32. As was the case with the
step-feed system, the carbonaceous effluent quality was more influenced by
the presence or absence of Nocardia than any other factor. Again the effluent
quality could have been shown to be much worse at the higher SRT values by
characterizing periods of heavy Nocardia production.
E. Complete Mix System
The complete mix system was seeded with secondary sludge from tne u.C.
treatment plant on November 14, 1973. The biomass was quickly established and
a moderate sludge waste was initiated on the 20th of the month. Filamentous
growth began to predominate and by December 10 the clarifier bed level reached
the top of the clarifier at the 132 1/min (35 gpm) flow. Operation with inter-
mittent clarifier overflow continued until December 20 when the system was
drained and reseeded with waste sludge from the plug flow and step-feed
systems. Within nine days the system was back into heavy filamentous growth
with the clarifier bed level again coming over the weirs and process SVI's of
300-400 ml/gm. The flow was temporarily reduced and t^Og (about 25 mg/1) was
added to the system for 24 hours on the 3rd and 5th of January. The flow
66
-------
TABLE 30 PROCESS CHARACTERISTICS FOR THE PLUG FLOW SYSTEM
AT A 3.5 DAY SRT (JUNE 15 - JULY 8, 1975)
Value or Standard
Average Deviation
Aeration Time, hrs. 3.6
Influent Flow, m3/day (gpd) 191.5 (50,600)
Recycle Flow Rate, 1/min (gpm) 42-45 (11-12)
Clarifier Overflow Rate, m/day (gpd/ft2) 21.5 (527)
SVI, ml/gm 83 14.9
MLSS, mg/1 1820 109
MLVSS, mg/1 1340 80
SRT, days 3.5
F/M, g BOD5 Applied/g MLVSS/day 0.46
Solids Production, g SS/g BOD5 Applied 0.86
Solids Production, g VSS/g BODs Applied 0.62
Solids Production, g SS/g BOD5 Removed* 0.92
Solids Production, g SS/g COD Applied 0.42
Solids Production, g VSS/g COD Applied 0.31
Solids Production, g SS/g COD Removed 0.49
Solids Production, g SS/g BOD5 Removed* 0.80
(excluding effluent solids)
Solids Production, g SS/g COD Removed 0.43
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.31
(excluding effluent solids)
Solids Production, mg/1 80.2
Suspended Solids Production in Effluent, % 13.7
Recycle Suspended Solids, mg/1 7300 733
Recycle Volatile Suspended Solids, mg/1 5280 517
Recycle P04, mg/1 534 56
Recycle TKN, mg/1 557 83
Calculated Using Inhibited BOD Values
67
-------
TABLE 31. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
PLUG FLOW SYSTEM AT A 3.5 DAY SRT (JUNE 15 -
JULY 8, 1975).
Influent
Effluent
BOD, mg/1
Inhibited BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
93
--
190
17.6
0
14.6
73
55
25.3
Std.
Devi at.
15.1
--
26.2
2.2
0
1.5
10.5
7.9
0.7
Mean
12.8
6.1
27.4
1.6
8.6
9.7
11.1
8.0
Std.
Devi at
2.3
1.5
3.2
0.5
0.84
1.2
2.3
2.1
68
-------
TABLE 32 SUMMARY OF SYSTEM OPERATION AND PERFORMANCE
WITH THE PLUG FLOW SYSTEM.
STEADY STATE PERIODS
Days at Equilibrium 28 28 24 41 38 32 19
Average Flow, m3/day 191.9 165.4 191.5 180.6 188.9 166.9 198.0
Average Flow, gpd 50,700 43,700 50,600 47,700 49,900 44,100 52,300
Detention Time, hrs. 3.6 4.2 3.6 3.8 3.7 4.2 3.5
Average Temperature, °C 22.5 26.5 25.3 24.0 19.4 16.5 19.0
SVI, ml/gm 164 123 83 117 219 90 147
SRT, days 1.9 2.9 3.5 4.4 4.7 5.7 6.6
F/M, g BOD5 Applied/g MLVSS/day 0.79 0.51 0.46 0.42 0.36 0.37 0.26
Solids Production, g SS/g BOD5 Applied 0.86 0.94 0.86 0.71 0.81 0.71 0.82
Solids Production, g SS/g COD Applied 0.42 0.45 0.42 0.37 0.38 0.36 0.38
Solids Production, g VSS/g BOD5 Applied 0.66 0.69 0.62 0.54 0.59 0.47 0.60
Solids Production, g VSS/g COD Applied 0.32 0.32 0.31 0.28 0.28 0.24 0.28
Effluent BOD5, mg/1 9.9 12.3 12.8 25.5 7.6 23.7 10.7
Effluent Inhibited BOD5, mg/1 8.7 4.7 6.1 - 5.1 12.7
Effluent COD, mg/1 42.3 25.7 27.4 31.9 31.1 45.8 38.0
-------
Effluent SS, mg/1
Effluent VSS, mg/1
Effluent P04, mg/1
Effluent (NCL + NO,)-N, mg/1
<— O
Waste Sludge, % P
Waste Sludge, % TKN
P Balance, Mass In * Mass Out
N Balance, Mass In * Mass Out
Volatile Solids, %
Recycle COD/Recycle SS
Recycle COD/Recycle VSS
TABLE 32.
(Continued)
6.3
4.2
12.7
0
2.5
9.7
1.03
1.00
77.8
1.32
1.69
5.9
4.1
13.1
6.4
2.2
7.6
1.00
-
73.4
1.12
1.53
STEADY STATE PERIODS
11.1
8.0
9.7
8.6
2.4
7.6
1.00
-
73.0
-
_
13.4
10.1
14.5
6.7
2.2
7.8
1.04
-
75.8
1.09
1.43
6.2
4.6
8.7
0.8
3.7
8.8
1.13
-
73.9
1.13
1.54
28
17
11.3
11.2
2.2
7.3
1.17
-
70.3
-
_
20
14
-
-
-
-
-
-
73.4
1.10
1.49
-------
was increased to 132 1/min (35 gpm) on tire 5th and the clarifier bed level
held at mid-depth until January 15th when it again rapidly rose to the top
of the clarifier. During the period of January 5-15 the process loading was
about 0.5 g BODs/g MLVSS/day. The flow was reduced again on January 16 and
H202 addition ( r-> 25 mg/1) was initiated on the 17th and continued for
2.5 days. The combination of ^2 and reduced flow returned the clarifier
bed level to near mid-depth. The flow was maintained between 57-76 1/min
(15-20 gpm) for the remainder of the month and the wasting was temporarily
discontinued to increase the solids concentration. By the end of the month
the final clarifier was frequently covered with Nocardia.
Reactor solids during February and the first 13 days of March were main-
tained near 5000 mg/1. Several attempts were made during this period to
increase the flow from 76 1/min (20 gpm) to 95 1/min (25 gpm) but this always
resulted in the clarifier bed overflowing the weirs and a return to the 76
1/min (20 gpm) flow. Average process loading was 0.16 g BODs applied/g MLVSS/
day in February and 0.14 g BOD5 applied/g MLVSS/day during the first 13 days
of March. Throughout this period the Nocardia concentration was quite heavy.
This necessitated frequent cleaning of the final clarifier and negated
attempts to obtain representative effluent samples.
The waste rate was increased on March 14 and the flow rate was gradually
increased to 132 1/min (35 gpm) by March 19. This flow was maintained through
April 13 and throughout this period the clarifier bed level was stable and
remained 1.5 m (5 ft) below the surface. Reactor solids were quite stable
from April 1-13 and averaged 4,200 mg/1. The average process loading during
this 13-day period was 0.24 g BOD5/g MLVSS/day. Although Nocardia was pre-
sent during this time the concentration was sufficiently reduced on the
clarifier surface to obtain fairly representative effluent samples. The
effluent suspended solids averaged 48 mg/1 during this time.
On April 14 the clarifier bed level began to rise rapidly and it was
necessary to reduce the flow to 95 1/min (25 gpm) to maintain the bed level
below the clarifier surface. Periodic attempts to increase the flow during
the next several days always resulted in the bed overflowing the clarifier
surface. Also by the 24th of the month the clarifier surface was again
covered with large amounts of floating solids. Consequently the waste rate
was more than doubled starting on the 24th and the flow was gradually
increased back to 132 1/min (35 gpm) by the end of the month.
By the end of the first week of May, the Nocardia problem with the
clarifier no longer existed. During the first two weeks in May the MLSS and
bed level were relatively stable even though the SVI's were varying between
350 and 400 ml/gm. The SVI's declined to around 250 ml/gm starting at mid-
month and the reactor solids increased somewhat. This was only a temporary
change since the SVI's were back to about 430 ml/gm during the last 10 days
of the month. Also the clarifier bed level increased sharply beginning on
the 24th of May and it was necessary to reduce the flow to 95 1/min (25 gpm)
to stop the bed from overflowing the weirs. The average process loading dur-
ing the first 23 days of May was 0.42 g BOD5/g MLVSS/day. The wasting was
increased again on the 26th and the flow returned back to 132 1/min (35 gpm)
by the 28th of May.
71
-------
The reactor solids, flow and clarifier bed level were very stable during
the period of June 7-30. Results obtained during this period are presented
in Tables 33 and 34. Since the SRT was only 2.6 days, there were over three
turnovers under the new operating conditions prior to the initiation of
steady-state data collection. Reduction of the SRT to 2.6 days eliminated
the Nocardia problem and resulted in an entirely satisfactory carbonaceous
effluent quality with low suspended solids.
The waste rate was increased on July 1 and again on July 7 to move to new
equilibrium operating conditions. The reactor solids declined to around 1400
mg/1 and the clarifier bed level was very stable for the first 17 days of the
month. Process loading from July 10-17 was about 0.65 g BODs/g MLVSS/day.
The clarifier bed level began to rise on the 18th and by the 21st it was
necessary to reduce the flow from 132 1/min (35 gpm) to 76 1/min (20 gpm) to
maintain the bed within the clarifier. At this time the SVI was 700 ml/gm.
HoOp ( r~> 250 mg/1) was added for 24 hours during July 22-23. This decreased
the SVI's to about 200 ml/gm and reduced the bed level considerably. The
132 1/min (35 gpm) flow was resumed on the 24th and stable solids concentra-
tions were achieved by August 8.
The next period of steady-state operation was obtained between August 9-
31, and these results are summarized in Tables 35 and 36. The clarifier bed
level held very steady during this time but the SVI's decreased from 400-450
ml/gm during August 9-19 to 150 ml/gm during the last 7 days of the month.
Since the SRT was only 2.1 days, the system experienced 7-8 turnovers under
constant operating conditions prior to the start of the steady-state period.
The waste rate was increased on September 4 to move to new equilibrium
conditions. Since the SRT was only 2 days, the system underwent three turn-
overs under the new operating conditions by September 11. The next period
of steady-state operation was from September 11-October 10 and data obtained
during this period are summarized in Tables 37 and 38. Throughout this per-
iod the clarifier bed level held steady but the SVI gradually increased
from 300 ml/gm on the 9th to 800 ml/gm by the last third of September. The
TKN and P analyses were somewhat erratic during the later part of September
and this is reflected in the relatively high standard deviations for recycle
TKN and P04.
The clarifier bed level started rising on October 10; it increased over
1.5 m (5 ft) and began overflowing the weirs by October 11. The-influent
flow was reduced to 76 1/min (20 gpm) and H202 (^ 200 mg/1) was added for
24 hours. Also the waste rate was reduced to move to operation at a higher
SRT. The flow was increased back to 132 1/min (35 gpm) on the 13th but the
bed was back to the clarifier surface by the 15th and continued to overflow
the weirs even though the flow was reduced to 76 1/min (20 gpm). H20? was
added again for 24 hours on the 17th. The bed dropped considerably and the
flow was increased back to 132 1/min (35 gpm) on the 18th. The bed held
about 1.5 m (5 ft) below the surface until the 23rd when it again rose
rapidly and started overflowing the clarifier weirs. The flow was reduced
to 57-76 1/min (15-20 gpm) and the clarifier continued to overflow inter-
mittently until the 27th when the bed began to drop. On the 29th the flow
72
-------
TABLE 33 PROCESS CHARACTERISTICS FOR THE COMPLETE MIX SYSTEM
AT A 2.6 DAY SRT (JUNE 7-30, 1974)
Value or Standard
Average Deviation
Aeration Time, hrs. 3.8
Influent Flow, m3/day (gpd) 187.8 (49,600)
Recycle Flow Rate, 1/min (gpm) 30-34 (8-9)
Clarifier Overflow Rate, m/day (gpd/ft2) 21.3 (522)
SVI, ml/gm 256 76.0
MLSS, mg/1 1875 87.8
MLVSS, mg/1 1440 70.8
SRT, days 2.6
F/M, g BOD5 Applied/g MLVSS/day 0.53
Solids Production, g SS/g BOD5 Applied 0.93
Solids Production, g VSS/g BOD5 Applied 0.71
Solids Production, g SS/g BODs Removed 1.04
Solids Production, g SS/g COD Applied 0.48
Solids Production, g VSS/g COD Applied 0.37
Solids Production, g SS/g COD Removed 0.59
Solids Production, g SS/g BODs Removed 0.93
(excluding effluent solids)
Solids Production, g SS/g COD Removed 0.53
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.40
(excluding effluent solids)
Solids Production, mg/1 111
Suspended Solids Production in Effluent, % 10.7
Recycle Suspended Solids, mg/1 9180 538
Recycle Volatile Suspended Solids, mg/1 6895 438
Recycle COD, mg/1 10,420 843
Recycle P04, mg/1 799 105
Recycle TKN, mg/1 747 56
73
-------
TABLE 34 INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
COMPLETE MIX SYSTEM AT A 2.6 DAY SRT (JUNE 7 -
JUNE 30, 1974).
Influent
Effluent
BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
120
232
24.1
0
21.3
114
89
24.2
Std.
Devi at
9.4
24.3
1.5
0
1.87
13.6
13.0
0.70
Mean
13.5
43.8
16.4
0
12.4
12.0
10.7
Std.
Devi at.
4.2
8.6
1.9
0
2.3
4.4
3.9
74
-------
TABLE 35. PROCESS CHARACTERISTICS FOR THE COMPLETE MIX SYSTEM
AT A 2.1 DAY SRT (AUG. 9-31, 1974)
Value or Standard
Average Deviation
Aeration Time, hrs. 3.4
Influent Flow, m3/day (gpd) 205.5 (54,300)
Recycle Flow Rate, 1/min (gpm) 34 (9)
Clarifier Overflow Rate, m/day (gpd/ft2) 23.3 (572)
SVI, ml/gm 322 142
MLSS, mg/1 1480 94.6
MLVSS, mg/1 1120 78.0
SRT, days 2.1
F/M, g BOD5 Applied/g MLVSS/day 0.68
Solids Production, g SS/g BOD5 Applied 0.97
Solids Production, g VSS/g BOD5 Applied 0.72
Solids' Production, g SS/g BODs Removed 1.09
Solids Production, g SS/g COD Applied 0.48
Solids Production, g VSS/g COD Applied 0.36
Solids Production, g SS/g COD Removed 0.58
Solids Production, g SS/g BODs Removed 0.99
(excluding effluent solids)
Solids Production, g SS/g COD Removed 0.53
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.39
(excluding effluent solids)
Solids Production, mg/1 104
Suspended Solids Production in Effluent, % 9.0
Recycle Suspended Solids, mg/1 7220 651
Recycle Volatile Suspended Solids, mg/1 5340 497
Recycle COD, mg/1 8080 1085
Recycle P04, mg/1 606 68.7
Recycle TKN, mg/1 579 63.7
75
-------
TABLE 36 INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
COMPLETE MIX SYSTEM AT A 2.1 DAY SRT (AUG. 9 -31,
1974).
Influent Effluent
BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
108
217
22.0
0
19.1
116
90
27.0
Std.
Devi at.
14.8
21.1
1.3
0
2.5
16.0
12.6
1.0
Mean
12.2
36.3
14.4
0
11.5
9.5
7.3
Std.
Devi at
5.5
7.1
1.0
0
1.7
5.1
4.3
76
-------
TABLE 37 PROCESS CHARACTERISTICS FOR THE COMPLETE MIX SYSTEM
AT A 1.8 DAY SRT (SEPT. 11 - OCT. 10, 1974)
Aeration Time, hrs.
Influent Flow, m3/day (gpd)
Recycle Flow Rate, 1/min (gpm)
o
Clarifier Overflow Rate, m/day (gpd/ft )
SVI, ml/gm
MLSS, mg/1
MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g BOD5 Applied
Solids Production, g VSS/g BOD5 Applied
Solids Production, g SS/g BOD5 Removed
g SS/g COD Applied
g VSS/g COD Applied
g SS/g COD Removed
Solids Production
Solids Production
Solids Production
Solids Production, g SS/g BOD5 Removed
(excluding effluent solids)
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent,
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Recycle COD, mg/1
Recycle P04, mg/1
Recycle TKN, mg/1
Value or
Average
3.9
182.8 (48,300)
34 (9)
20.7 (508)
694
1220
970
1.8
0.76
0.94
0.73
1.06
0.44
0.34
0.55
0.94
0.49
0.38
112
11.5
5650
4440
7200
532
597
Standard
Deviation
170
71.8
64.0
494
356
1000
113
179
77
-------
TABLE 38. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
COMPLETE MIX SYSTEM AT A 1.8 DAY SRT (SEPT. 11 -
OCT. 10, 1974).
Influent Effluent
BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
119
256
27.7
0
21.2
117
92
23.4
Std.
Devi at.
16.8
28.9
3.0
0
2.6
16.3
12.6
1.5
Mean
13.8
53.4
19.6
0
13.2
13.1
9.8
Std.
Devi at
5.2
13.9
2.6
0
2.5
4.2
3.2
78
-------
was increased from 57 to 76 1/nrin (15 to 20 gpm). Reactor volatile solids
from October 30-November 6 averaged 1580 mg/1 and the resulting process
loading was 0.35 g BOD5/g MLVSS/day. On November 7 the clarifier bed was
again overflowing even at the 76 1/min (20 gpm) flow. The flow was reduced
to 57 1/min (15 gpm) and H202 was added on the 8th and 9th. Flow was in-
creased back to 76 1/min (20 gpm) on the llth.
The clarifier bed level remained stable the rest of November, December
and January. The reactor solids concentration also reached equilibrium by
December and the next period of steady-state operation to be reported was
from December 3-January 16, 1975. Results obtained during this 45-day period
are summarized in Tables 39 and 40. The return to operation at an increased
SRT was paralleled by increased concentrations of Nocardia. Although the
Nocardia concentration was sufficient to cause brown foam over much of the
reactor surface, it was not sufficiently troublesome to cause major difficul-
ties in measuring the effluent quality. There was negligible nitrification
in the system during the first two-thirds of December, but then the N03-N
concentration began to increase until it reached 10 mg/1 by mid-January.
This is the reason the TKN and (N02 + N03)-N concentrations in Table 40 show
such a large standard deviation. The high effluent BOD's reflect both nitri-
fication in the BOD analysis and the increased effluent suspended solids
resulting from the Nocardia.
On January 17, 1975 the waste rate was increased by a small amount to
move to new equilibrium operation. The clarifier bed level declined a small
amount during the remainder of January, but was very stable throughout Feb-
ruary 2-25, 1975. These data are presented in Tables 41 and 42. Because of
a temporary change in influent wastewater characteristics on February 26, the
steady-state operation was considered to cover this relatively short time
period. Although this only allows for two sludge turnovers following the
waste change on January 17 (and the influent change beginning on the 18th),
the reactor solids were very stable and the process loading was not substant-
ially different than in the previous steady-state period. Only minor amounts
of Nocardia were present and this is reflected in the low effluent suspended
solids levels.
The influent water characteristics returned to normal by March 2-3. On
March 7 the flow rate was increased to 132 1/min (35 gpm) and the waste rate
was also increased substantially. The MLSS declined to 1500-1600 mg/1 during
the last third of the month. From March 20-29 the process loading was 0.60 g
BODs/g MLVSS/day with an average SVI of 85 ml/gm. There were several pump and
meter failures over the next few days, and all problems were not resolved
until April 4. At this time the SVI was still < 100 ml/gm and the bed level
was very stable. The bed held 0.9 m (3 ft) off the clarifier bottom until
April 13 and then rose to the top of the clarifier within 24 hours. The
underflow wasting was increased on April 14 and the effluent suspended solids
were 70-80 mg/1. The SRT was f*> 1 day under these conditions and the SVI was
varying from 900 to 1100 ml/gm. The flow was decreased to 114 1/min (30 gpm)
on the 21st and the volumetric waste was increased further. The flow rate
was decreased to 95 1/min (25 gpm) on the 22nd. The bed level decreased be-
low the effluent weir on the 24th and the effluent suspended solids decreased
79
-------
TABLE 39. PROCESS CHARACTERISTICS FOR THE COMPLETE MIX SYSTEM
AT AN 8.1 DAY SRT (DEC. 3, 1974 - JAN. 16, 1975)
Aeration Time, hrs.
Influent Flow, m3/day (gpd)
Recycle Flow Rate, 1/min (gpm)
Clarifier Overflow Rate, m/day (gpd/ft2)
SVI, ml/gm
MLSS, mg/1
MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g 8005 Applied
Solids Production, g VSS/g BOD5 Applied
Solids Production, g SS/g BODs Removed
Solids Production, g SS/g COD Applied
Solids Production, g VSS/g COD Applied
Solids Production, g SS/g COD Removed
Solids Production, g SS/g BODr Removed
(excluding effluent solids)0
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent,
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Recycle COD, mg/1
Recycle P04, mg/1
Recycle TKN, mg/1
Value or
Average
6.1
115.8 (30,600)
49 (13)
13.1 (322)
247
2990
2160
8.1
0.22
0.78
0.56
0.98
0.37
0.26
0.44
0.74
0.33
0.24
94.0
24.2
7950
5725
8500
612
599
Standard
Deviation
42.9
272
170
490
283
992
128
100
80
-------
TABLE 40. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
COMPLETE MIX SYSTEM AT AN 8.1 DAY SRT (DEC. 3,
1974 - JAN. 16, 1975).
Influent Effluent
BOD, mg/1
BOD, mg/1 (Jan. 7-16)
Inhibited BOD, mg/1
(Jan. 7-16)
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
121
--
--
257
28.1
0
22.2
134
101
17.6
Std.
Devi at.
16.5
--
--
22.6
2.9
0
3.5
17.5
14.5
0.7
Mean
25.4
35.3
18.1
42.0
13.7
3.2
14.9
23.0
16.3
Std.
Devi at
10.8
5.1
3.7
7.3
6.2
3.6
2.8
10.0
7.2
81
-------
TABLE 41. PROCESS CHARACTERISTICS FOR THE COMPLETE MIX SYSTEM
AT A 7.1 DAY SRT (FEB. 2-25, 1975)
Value or Standard
Average Deviation
Aeration Time, hrs. 5.6
Influent Flow, m3/day (gpd) 125.7 (33,200)
Recycle Flow Rate, 1/min (gpm) 45 (12)
Clarifier Overflow Rate, m/day (gpd/ft2) 14.2 (346)
SVI, ml/gm 92 11.1
MLSS, mg/1 2400 181
MLVSS, mg/1 1760 142
SRT, days 7.1
F/M, g BOD5 Applied/g MLVSS/day 0.26
Solids Production, g SS/g BODs Applied 0.77
Solids Production, g VSS/g BOD5 Applied 0.55
Solids Production, g SS/g BOD5 Removed* 0.82
Solids Production, g SS/g COD Applied 0.37
Solids Production, g VSS/g COD Applied 0.26
Solids Production, g SS/g COD Removed 0.43
Solids Production, g SS/g BODs Removed* 0.72
(excluding effluent solids)
Solids Production, g SS/g COD Removed 0.38
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.27
(excluding effluent solids)
Solids Production, mg/1 80.5
Suspended Solids Production in Effluent, % 12.6
Recycle Suspended Solids, mg/1 6995 514
Recycle Volatile Suspended Solids, mg/1 5095 357
Recycle COD, mg/1** 7940 901
Recycle P04, mg/1 500 80
Recycle TKN, mg/1 539 94
* Calculated Using Inhibited BOD Values
** Feb. 2-Feb. 13 only
82
-------
TABLE 42. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
COMPLETE MIX SYSTEM AT A 7.1 DAY SRT (FEB. 2 -
FEB. 25, 1975).
Influent Effluent
BOD,
Inhi
COD,
TKN,
(N02
P04,
SS,
VSS,
Temp
mg/1
bited BOD, mg/1
mg/1
mg/1
+ N03)-N, mg/1
mg/1
mg/1
mg/1
erature, °C
Mean
105
--
220
19.5
0
17.7
88
62
16.6
Std.
Devi at.
8.9
--
27.7
2.1
0
2.0
26
16
0.57
Mean
12
6
32
2
9
11
10
6
.7
.9
.8
.2
.9
.2
.2
.6
Std.
Devi at
2.5
1.4
3.4
0.79
1.6
1.1
4.3
3.2
83
-------
to 11 mg/1. The bed level decreased considerably on the 26th and remainded
stable until May 13 when it again came to the top of the clarifier.
From April 25-May 12 (18 days) the reactor solids were stable and the
effluent quality was uniform. Results obtained during this period of oper-
ation are summarized in Tables 43 and 44. The F/M ratio was high, and the
SRT was just 1.5 days. The most interesting observation is that the effluent
quality was excellent with a BOD5 of 7 mg/1 and suspended solids of 6 mg/1.
This demonstrates once again that effluent quality is essentially a function
of clarification efficiency over a wide range of operating conditions. The
clarifier bed level rose rapidly on May 14 and was back at the top of the
clarifier within a few hours. The SVI decreased from 1100 ml/gm on May 7
to 250 ml/gm on May 10 and then increased to 1200 ml/gm by May 14.
Regular operation of the system was discontinued on May 16 when the
system was converted to a chemostat (complete mix reactor with no recycle).
The flow was reduced to 76 1/min (20 gpm). This resulted in a reactor de-
tention time of about 6 hours. The air flow rate was maintained at a high
enough level to insure adequate mixing, and the reactor D.O. levels varied
from 3-8 mg/1.
The influent and reactor (effluent) characteristics were monitored
from May 20-June 24 and the results are summarized in Table 45. Samples were
composited over 24-hour periods on Sunday-Thursday only. It can be seen that
about 35% of the influent BOD and COD was oxidized.
On June 4, soluble samples of influent and effluent were taken every
two hours during a 10-hour period and analyzed for TOC. Reactor suspended
solids were measured periodically as were the influent and reactor optical
densities at 540 mp. Results are presented in Figure 6. It can be seen
that the soluble TOC component in the reactor remained unchanged during the
rather sharp rise in influent soluble TOC from 1200 to 1800 hours. Samples
were also withdrawn from the reactor at 1400 and 1800 hours and stirred
vigorously with a magnetic stirrer. This method of aeration was selected
to minimize evaporation losses. The change in soluble TOC in these stirred
samples (Table 46) was so small that it could not be measured.
Since the first examination of soluble components was not done during
the maximum organic loading cycle an additional study was conducted over a
20-hour period on June 17-18. Results are presented in Figure 1. This study
occurred about 2 days after a fairly heavy rain and, as a result, the soluble
TOC was not as high as in the previous study. Once again, the change in
soluble TOC in the reactor was unmeasurable and the two samples which were
withdrawn and stirred (Table 46) also indicated that the soluble TOC in the
reactor during the period of peak loading was essentially the final residual
level.
The chemostat was operated throughout June, July and the first half of
August. Because of heavy and very frequent rains, the influent wastewater
was unusually weak and the dry weather cycle in wastewater strength was
attenuated considerably. From July 11-August 11 the influent BOD only
84
-------
TABLE 43. PROCESS CHARACTERISTICS FOR THE COMPLETE MIX SYSTEM
AT A 1.5 DAY SRT (APRIL 25 - MAY 12, 1975)
Value or Standard
Average Deviation
Aeration Time, hrs. 5.0
Influent Flow, m3/day (gpd) 141.2 (37,300)
Recycle Flow Rate, 1/min (gpm) 30 (8)
Clarifier Overflow Rate, m/day (gpd/ft2) 16.0 (393)
SVI, ml/gm 920 369
MLSS, mg/1 810 52
MLVSS, mg/1 620 36
SRT, days 1.5
F/M, g BOD5 Applied/g MLVSS/day 0.84
Solids Production, g SS/g BOD5 Applied 1.09
Solids Production, g VSS/g BODg Applied 0.80
Solids Production, g SS/g BODs Removed 1.18
Solids Production, g SS/g COD Applied 0.53
Solids Production, g VSS/g COD Applied 0.39
Solids Production, g SS/g COD Removed 0.63
Solids Production, g SS/g BOD5 Removed 1.12
(excluding effluent solids)
Solids Production, g SS/g COD Removed 0.60
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.44
(excluding effluent solids)
Solids Production, mg/1 118
Suspended Solids Production in Effluent, % 4.5
Recycle Suspended Solids, mg/1 2910 272
Recycle Volatile Suspended Solids, mg/1 2135 197
Recycle P04, mg/1 168 21.3
85
-------
TABLE 44. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
COMPLETE MIX SYSTEM AT A 1.5 DAY SRT (APRIL 25 -
MAY 12, 1975).
Influent
Effluent
BOD, mg/1
Inhibited BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
P04, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
108
—
222
19.6
0
15.2
94
69
19.1
Std.
Devi at.
17.0
--
21.3
1.3
0
2.0
13.9
8.9
1.1
Mean
7.3
5.6
34.3
—
0
10.0
5.5
3.6
Std.
Devi at
2.9
2.0
4.6
—
0
1.1
2.2
1.7
86
-------
TABLE 45. INFLUENT, EFFLUENT AND PROCESS
CHARACTERISTICS FOR THE CHEMOSTAT
(MAY 20 - JUNE 24, 1975).
Aeration Time, hrs.
Influent Flow, m3/day
MLSS, mg/1
MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g
Solids Production, g
Solids Production, g
Solids Production, g
Solids Production, g
Solids Production, g
Solids Production, g
(gpd)
MLVSS/day
SS/g BOD5 Applied
VSS/g BOD5 Applied
SS/g BOD5 Removed(1)
SS/g COD Applied
VSS/g COD Applied
SS/g COD Removed^2)
Solids Production, mg/1
BOD, mg/1
BOD, mg/r '
COD, mg/1
COD, mg/1 ^ '
SS, mg/1
SS, mg/1
SS, mg/l(2)
VSS, mg/1
Temperature, °C
Influent
Std.
Mean Devi at.
100 18.5
97 15.3
202 21.9
207 18.2
85 18.1
__
65 13.5
23.7 1.3
Value or
Average
6.2
113.2 (29,900)
81
61
0.26
6.3
0.81
0.61
2.3
0.40
0.30
1.11
81
Mean
--
62
--
134
81
81
81
61
Standard
Deviation
14.3
9.7
Effluent
Std.
Devi at
--
15.1
--
15.0
14.3
16.2
15.4
9.7
(1) May 28 - June 24
(2) May 20 - June 19
87
-------
100 •-
Q 80 f
60--
40--
E
a.
o
^-
in
0.20 •-
< 0.10
u
H 1-
H 1 1 1 1 1 \
INFLUENT FLOW = 74.6j?/min (19.7gpm)
REACTOR DETENTION TIME = 6.55 hours
TEMPERATURE = 24 °C
SUSPENDED SOLIDS
VOLATILE
SUSPENDED SOLIDS
INFLUENT
REACTOR
SOLUBLE EFFLUENT
H h
H - 1 - 1
1
0600 0800 1000 1200 1400 1600
TIME, hour (June 4, 1975)
—I r-
1800
Figure 6. Influent and Reactor TOC and Suspended Solids Concentrations
for the Chemostat on June 4, 1975.
-------
TABLE 46. CHANGE IN SOLUBLE REACTOR TOC
AFTER ADDITIONAL AERATION.
SAMPLE
Date/Hour
HOURS
Stirred
SOLUBLE
TOC*
mg/1
6-4; 1400
0
19
43
12.3
12.8
13.2
6-4; 1800
0
15
39
13.7
13.7
15.2
6-17; 2200
0
35
14.8
16.0
6-18; 0200
0
31
15.7
16.0
* (984H Reeve Angel Filter)
89
-------
0>
E
vt
O
oc
O
100"
80"
60-'
40--
0.2 •-
40-•
30--
PROCESS FLOW = 77.2j!/min (20.4gpm)
REACTOR DETENTION TIME = 6.33 hours
TEMPERATURE = 25 °C
SUSPENDED SOLIDS
O
_ •*•
A
VOLATILE
SUSPENDED SOLIDS
SOLUBLE INFLUENT
SOLUBLE EFFLUENT
-I-
0900 1300 1700 2100 0100
TIME, hour (June 17-18, 1975)
—I—
0500
Figure 7. Influent and Reactor TOC and Suspended Solids Concentrations for
the Chemostat on June 17-18, 1975.
90
-------
averaged 81 mg/1. The plug flow unit also operated during July and the
first half of August.
On August 12-13, 1975 a special study was performed to compare the per-
formance of the chemostat to a "conventional" treatment system operated at a
more conservative organic loading. The chemostat was operated at a steady
flow of 83 1/min (22 gpm). The plug flow unit received an average flow of
129 1/min (34 gpm). The D.O. in both systems was greater than 2 mg/1. The
flow to the plug flow unit was stable except for a 20 minute period around
1930 hours when the influent pump cut off. This brief disruption had no
noticeable impact on the system parameters measured.
Results obtained from the special study are summarized in Figures 8 and
9. The MLVSS in the plug flow system were approximately 20 times higher than
in the chemostat. A soluble influent BODc of 55 mg/1 corresponds to a sol-
uble loading to the chemostat of 4.1 g soluble BODc/g MLVSS/day (MLVSS =
55 mg/1) and a soluble loading to the plug flow unit of 0.35 g soluble BODs/
g MLVSS/day (MLVSS = 1000 mg/1).
As in previous studies the soluble TOC in the chemostat remained unchanged
during the increasing organic loading. The data suggest a very small rise in
COD and BOD as the load increased. The effluent samples from the plug flow
unit were taken directly from the effluent channel and filtered within 5-10
minutes thereafter. This minimized any change in the soluble component which
could occur during clarification. The soluble component from this unit was
also "constant". The difference in soluble BOD, COD and TOC between the two
units averaged 9, 33 and 10 mg/1, respectively.
Results obtained during the 7 steady-state periods of operation with the
complete mix reactor are summarized in Table 47. The solids production values
for operation at the 0.26 day SRT are somewhat distorted when evaluated on
the basis of applied loading. In this case some soluble BOD was escaping
unmetabolized in the effluent. On the basis of the one comparative evalua-
tion between the chemostat and plug flow systems (Figure 9), around 10 mg/1
more soluble BOD5 was probably metabolized in the other steady-state periods
reported in Table 47.
F. Related Complete Mix Systems
As previously indicated in Section 5, there were two other complete mix
reactors operated at the pilot plant during 1974 that were part of another
investigation. The influent wastewater to these systems was pumped from the
same header tank that was used to feed the three reactors described ab9ve.
At several times during the year these systems produced information which was
directly pertinent to the results and observations reported thusfar. Since
this information has not been published elsewhere, those aspects relevant to
the results obtained in the present study will be summarized here.
The 12.72 m3 (3,360 gal) reactor had a considerable amount of filamentous
growth in December 1973 and H202 addition was used to solve this problem on
December 22-24'. On December 27, 1973 automated dissolved oxygen control was
incorporated into the system operation. An automatic air throttling valve,
91
-------
MD
ro
if" 0.2
0.15^-
0.1- -
at
E
Z
60
40
20 ••
Z
o
u
w»
O
_l
O
1500
1000
500
MLSS
MLSS
INFLUENT
CHEMOSTAT
CHEMOSTAT
MLVSS
PLUG FLOW SYSTEM
-Q-
MLVSS
1200 1400 1600 1800 2000 2200
TIME, hour (August 12-13, 1975)
—I
2400
0200
Figure 8. Solids Concentration in the Chemostat and Plug Flow System on
August 12-13, 1975.
-------
70
5--
100- •
25-
H h
H 1 1 1 1 1
CHEMOSTAT
PLUG FLOW EFFLUENT
D
-n
PLUG FLOW EFFLUENT
-GJ-
-Q
PLUG FLOW EFFLUENT
1200
H h
H - 1 - 1 - 1
1
-t-
H 1
1400 1600 1800 2000 2200 2400
TIME,hour(AUGUST 12-13,1975)
0200
Figure 9. BOD, COD and TOC Concentrations in the Influent, Chemostat
and Plug Flow System Effluent on August 12-13, 1975.
93
-------
TABLE 47. SUMMARY OF SYSTEM OPERATION AND PERFORMANCE
WITH THE COMPLETE MIX SYSTEM.
STEADY STATE PERIODS
Days at Equilibrium 36 18 30 23 24 24 45
Average Flow, m3/day 113.2 141.2 182.8 205.5 187.8 125.7 115.8
Average Flow, gpd 29,900 37,300 48,300 54,300 49,600 33,200 30,600
Detention Time, hrs. 6.2 5.0 3.9 3.4 3.8 5.6 6.1
Average Temperature, °C 23.7 19.1 23.4 27.0 23.6 16.6 17.6
SVI, ml/gm - 920 694 322 256 92 247
i-D
SRT, days 0.26 1.5 1.8 2.1 2.6 7.1 8.1
F/M, g BOD5 Applied/g MLVSS/day 6.3 0.84 0.76 0.68 0.53 0.26 0.22
Solids Production, g SS/g BOD5 Applied 0.81 1.09 0.94 0.97 0.93 0.77 0.78
Solids Production, g SS/g COD Applied 0.40 0.53 0.44 0.48 0.48 0.37 0.37
Solids Production, g VSS/g BOD5 Applied 0.61 0.80 0.73 0.72 0.71 0.55 0.56
Solids Production, g VSS/g COD Applied 0.30 0.39 0.34 0.36 0.37 0.26 0.26
Effluent BOD,-, mg/1 62 7.3 13.8 12.2 13.5 12.7 25.4
0 "
Effluent Inhibited BODg, mg/1 - 5.6 - - - 6.9
Effluent COD, mg/1 134 34.3 53.4 36.3 43.8 32.8 42.0
Effluent SS, mg/1 81 5.5 13.1 9.5 12.0 10.2 23.0
-------
Effluent VSS, mg/1
Effluent P04, mg/1
Effluent (N02 + N03)-N, mg/1
Waste Sludge, % P
Waste Sludge, % TKN
P Balance, Mass In v Mass Out
01 N Balance, Mass In T Mass Out
Volatile Solids, %
Recycle COD/Recycle SS
Recycle COD/ Recycle VSS
TABLE 47.
(Continued)
STEADY STATE PERIODS
61
-
-
-
-
-
-
75.3
-
_
3.6 9.8
10.0 13.2
0 0
1.9 3.1
10.6
0.94 0.94
0.93
75.0 79.0
1.27
1.62
7.3
11.5
0
2.7
8.0
0.99
1.01
74.8
1.12
1.51
10.7
12.4
0
2.8
8.1
1.02
0.99
76.0
1.14
1.51
6.6
11.2
9.9
2.3
7.7
1.10
-
73.0
1.14
1.56
16.3
14.9
3.2
2.5
7.5
1.10
-
72.1
1.07
1.48
-------
operated by the computer was used to control the dissolved oxygen to a set
point of 1 mg/1 D.O. The submerged D.O. sensor was placed near the mid-point
of the reactor. The digital equation used by the computer was a constant
gain proportional band integral control formula of the form:
Air Flow = G [Kp (SP-D.O.) + Kj z(SP-D.O.)] ' F(Q)
G = constant gain term
Kp = proportional band control constant
SP = desired set point for the D.O.
D.O. = instantaneous D.O. reading from the submerged probe
KT = integral control constant
z(SP-D.O.) = instantaneous D.O. error
F(Q) = feed forward process flow compensations
The equation was recalculated every 36 seconds.
From January through April, 1974, the D.O. was controlled to 1.0 mg/1 ±
0.5 mg/1 including an instantaneous meter flutter of approximately 0,25 mg/1
D.O. From January 15-April 30 the system ran at an average SRT of about
4 days. The flow and volumetric waste during this period were constant.
During February, March and April the SVI stayed between 80-110 ml/gm and
there were no settling problems. Futhermore if Nocardia was present, it
was in such small concentrations that it did not even produce any foam on
the reactor surface. It will be recalled that Nocardia was present in all
three of the other systems at this time. All three of the other systems were
operated at higher SRT's during this period. The steady-state operation pre-
vailing in the 12.72 m3 complete mix system from March 19-April 30, 1974
is summarized in Tables 48 and 49. The effluent suspended solids were higher
than encountered in the other systems in the absence of Nocardia, but this
could simply reflect different hydraulic characteristics in the smaller
clarifier.
The 14.91 m3 complete mix reactor was also operated with D.O. control
to 1.0 ± 0.5 mg/1 at a 4.4 day SRT during February 1974. On March 4, the
D.O. set point was increased to 2.5 ± 0.5 mg/1 and the system was operated
at this D.O. level until April 14. After about 15 days of operation at the
higher D.O. level, the SVI's began to gradually increase from the previously
very steady level of 80-100. ml/gm until they reached 200 ml/gm by* the end
of March. The SVI's then declined very gradually to 160 ml/gm by April 14.
Operation of the 14.91 m3 complete mix system during the period of
March 19-April 14, 1974 is summarized in Tables 50 and 51. This allows for
three turnovers under the higher D.O. levels before considering any data as
representative of steady-state operation. Selected characteristics from the
12.72 m* complete mix system have been summarized in Table 52 for comparative
purposes. It is noteworthy that the cell yield coefficients based on applied
loading only differ by r*> 10%. When the difference in SRT between the two
units is considered in conjunction with the experimental errors of accurately
measuring flows and solids levels, it is apparent that the difference in D.O.
96
-------
TABLE 48. PROCESS CHARACTERISTICS FOR THE 12.72 m3 COMPLETE MIX SYSTEM
AT A 3.9 DAY SRT (MARCH 19 - APRIL 30, 1974)
Value or Standard
Average Deviation
Aeration Time, hrs 2.5
Influent Flow, m3/day (gpd) 123.4 (32,600)
Recycle Flow Rate, 1/min (gpm) 23 (6)
Clarifier Overflow Rate, m/day (gpd/ft2) 20.7 (508)
SVI, ml/gm 96 8.9
MLSS, mg/1 4330 324
MLVSS, mg/1 3100 271
SRT, days 3.9
F/M, g BOD5 Applied/g MLVSS/day 0.37
Solids Production, g SS/g BOD5 Applied 0.95
Solids Production, g VSS/g BOD5 Applied 0.68
Solids Production, g SS/g BOD5 Removed 1.14
Solids Production, g SS/g COD Applied 0.47
Solids Production, g VSS/g COD Applied 0.34
Solids Production, g SS/g COD Removed 0.65
Solids Production, g SS/g BODs Removed 0.84
(excluding effluent solids)
Solids Production, g SS/g COD Removed 0.48
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.34
(excluding effluent solids)
Solids Production, mg/1 113
Suspended Solids Production in Effluent, % 26.4
Recycle Suspended Solids, mg/1 16,590 1579
Recycle Volatile Suspended Solids, mg/1 11,890 1474
Recycle COD, mg/1 18,410 2070
97
-------
TABLE 49. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
12.72 m3 COMPLETE MIX SYSTEM AT A 3.9 DAY SRT
(MARCH 19 - APRIL 30, 1974).
Influent
Effluent
BOD, mg/1
COD, mg/1
TKN, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
119
240
25.1
120
90
18.4
Std.
Devi at.
17.0
26.1
3.0
25.2
16.1
1.4
Mean
20.1
66.0
17.9
30
22
Std.
Devi at
6.6
14.1
2.6
9.8
8.1
98
-------
TABLE 50. PROCESS CHARACTERISTICS FOR THE 14.91 m
COMPLETE MIX SYSTEM AT A 5.3 DAY SRT
(MARCH 19-APRIL 14, 1974).
Value or Standard
Average Deviation
Aeration Time, hrs. 2.8
Influent Flow, m3/day (gpd) 128.3 (33,900)
Recycle Flow Rate, 1/min (gpm) 30 (8)
Clarifier Overflow Rate, m/day (gpd/ft2) 18.5 (454)
SVI, ml/gin 166 33.8
MLSS, mg/1 4630 559
MLVSS, mg/1 3230 369
SRT, days 5.3
F/M, g BOD5 Applied/g MLVSS/day 0.32
Solids Production, g SS/g BOD5 Applied 0.84
Solids Production, g VSS/g BOD,- Applied 0.59
Solids Production, g SS/g BOD5 Removed 0.97
Solids Production, g SS/g COD Applied 0.43
Solids Production, g VSS/g COD Applied 0.30
Solids Production, g SS/g COD Removed 0.57
Solids Production, g SS/g BODg Removed 0.75
(excluding effluent solids)
Solids Production, g SS/g COD Removed 0.44
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.30
(excluding effluent solids)
Solids Production, mg/1 101
Suspended Solids Production in Effluent, % 22.7
Recycle Suspended Solids, mg/1 16,180 1063
Recycle Volatile Suspended Solids, mg/1 11,170 774
Recycle COD, mg/1 17,010 1952
99
-------
TABLE 51 INFLUENT AND EFFLUENT CHARACTERISTICS FOR
THE 14.91 m3 COMPLETE MIX SYSTEM AT A 5.3
DAY SRT (MARCH 19-APRIL 14, 1974).
Influent Effluent
Std.
Mean Devi at.
16.1 5.3
58.7 11.3
19.3 2.2
0 0
BOD,
COD,
TKN,
(N02
PO
A *
ss,
vss,
Temp
mg/1
mg/1
mg/1
+ N03)-N, mg/1
mg/1
mg/1
mg/1
erature, °C
Mean
120
235
24.5
0
20.1
119
87
17.6
Std.
Deviat
20.2
26.9
3.3
0
1.9
29.1
17.3
0.9
23 7.0
17 5.9
100
-------
TABLE 52. SELECTED CHARACTERISTICS FOR THE 12.72 nT COMPLETE
MIX SYSTEM DURING MARCH 19-APRIL 14, 1974.
Value or
Average
Standard
Deviation
Aeration Time, hrs.
Influent Flow, m /day (gpd)
o
Clarifier Overflow Rate, m/day (gpd/ft )
MLSS, mg/1
MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g BOD,- Applied
Solids Production, g VSS/g BODg Applied
Solids Production, g SS/g BOD,- Removed
g SS/g COD Applied
g VSS/g COD Applied
g SS/g COD Removed
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Solids Production
Solids Production
Solids Production
2.4
126.4 (33,400)
21.2 (520)
4480 277
3180 270
4.0
0.38
0.93
0.67
1.14
0.48
0.34
0.67
17,210 1502
12,270 1594
Effluent BOD, mg/1
Effluent COD, mg/1
Effluent TKN, mg/1
Effluent SS, mg/1
Effluent VSS, mg/1
21.6
69.0
18.3
28
21
7.9
16.9
2.1
9.9
8.5
101
-------
levels did not have a substantial (if any) impact on sludge production. The
effluent qualities of the two systems are also quite similar with the small
differences in COD and BOD residuals corresponding to increased suspended
solids in the effluent from the 12.72 m3 complete mix reactor.
During May the Nocardia concentration began to increase considerably in
the 12.72 m3 complete mix system. Throughout June the system was maintained
at an SRT of 4-5 days with automated D.O. control of 1.0 ± 0.5 mg/1. The
Nocardia organisms were competitive in this SRT range under the warmer waste-
water conditions. The increased Nocardia concentrations corresponded to an
increase in the SVI to around 200 ml/gm.
In another effort to ascertain the relationship between D.O., SRT and
Nocardia the 14.91 m3 (3,940 gal) complete mix reactor was operated at con-
stant flow and waste conditions during June and July, 1974. There was no
noticeable Nocardia in this system from June 1-July 6. At the beginning of
June microscopic examination revealed the complete absence of filamentous
growth with the biological solids consisting of very dense discrete particles.
Traces of Nocardia produced a small but noticeable amount of foam on the
reactor by July 7- By July 12 there were large quantities of floating scum
on the clarifier surface. Throughout this period the D.O. was controlled to
1.0 ± 0.5 mg/1. Results obtained during the period of most stable operation
are summarized in Tables 53 and 54. At the 8 day SRT, the system underwent
approximately three turnovers under the constant operating conditions prior
to the steady-state data summarized. The biological characteristics during
this period were very unusual (Average SVI of 37 ml/gm).
Steady-state operating data are also available from the 14.91 m3
(3940 gal) complete mix reactor during the period of August 11-September 3,
1974. During this period the clarifier bed level was very stable and remain-
ed within 0.3 m (1 ft) of the clarifier bottom. Prior to this time the system
had been operated with constant flow and volumetric waste for 12 days (about
five turnovers). These data are summarized in Tables 55 and 56. These data
were collected to compare the performance of two complete mix systems operated
at the same organic loading but with different detention times. This opera-
tion overlaps that in Tables 35 and 36. The effluent qualities from the two
systems were essentially the same. Sludge production values also are not
sufficiently different to see any significant changes as a result of the
different detention times. Of course one would not expect to find such
changes.
Results from the four periods of steady-state operation reported for
the 12.72 and 14.91 m3 complete mix reactors have been summarized in Table 57.
No information was available on either the TKN or P content for the waste
sludge. Hence no materials balances on these components could be performed.
G. Soluble Effluent Quality
On a number of occasions grab samples of clarified effluent were taken
from one or more of the systems between 0800-0900 hours. All samples were
filtered immediately through Reeve Angel 984H Filters and taken directly to
102
-------
TABLE 5a PROCESS CHARACTERISTICS FOR THE 14.91 m3 COMPLETE MIX SYSTEM
AT AN 8.4 DAY SRT (JUNE 23 - JULY 11, 1974)
Value or Standard
Average Deviation
Aeration Time, hrs. 2.5
Influent Flow, m3/day (gpd) 143.8 (38,000)
Recycle Flow Rate, 1/min (gpm) 26 (7)
Clarifier Overflow Rate, m/day (gpd/ft2) 20.8 (509)
SVI, ml/gin 36.7 3.8
MLSS, mg/1 7290 426
MLVSS, mg/1 5120 249
SRT, days 8.4
F/M, g BOD5 Applied/g MLVSS/day 0.23
Solids Production, g SS/g BOD5 Applied 0.73
Solids Production, g VSS/g BOD5 Applied 0.52
Solids Production, g SS/g BOD5 Removed 0.86
Solids Production, g SS/g COD Applied 0.39
Solids Production, g VSS/g COD Applied 0.28
Solids Production, g SS/g COD Removed 0.51
Solids Production, g SS/g BOD5 Removed 0.71
(excluding effluent solids)
Solids Production, g SS/g COD Removed 0.42
(excluding effluent solids)
Solids Production, g VSS/g COD Removed 0.29
(excluding effluent solids)
Solids Production, mg/1 88.7
Suspended Solids Production in Effluent, % 17.2
Recycle Suspended Solids, mg/1 33,620 2794
Recycle Volatile Suspended Solids, mg/1 23,520 1946
Recycle COD, mg/1 36,110 3470
103
-------
TABLE 54. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
14.91 m3 COMPLETE MIX SYSTEM AT AN 8.4 DAY SRT
(JUNE 23 - JULY 11, 1974).
Influent Effluent
BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
122
229
23.2
0
103
82
24.1
Std.
Devi at.
18.8
22.6
2.2
0
18.8
16.6
1.0
Mean
18.7
54.7
17.8
0
15.3
12.1
Std.
Devi at
4.2
5.4
2.1
0
4.5
4.4
104
-------
TABLE 55. PROCESS CHARACTERISTICS FOR THE 14.91 m3 COMPLETE MIX SYSTEM
AT A 2.2 DAY SRT (AUG. 11 - SEPT. 3, 1974)
Aeration Time, hrs.
Influent Flow, m3/day (gpd)
Recycle Flow Rate, 1/min (gpm)
Clarifier Overflow Rate, m/day (gpd/ft2)
SVI, ml/gin
MLSS, mg/1
MLVSS, mg/1
SRT, days
F/M, g BOD5 Applied/g MLVSS/day
Solids Production, g SS/g 6005 Applied
Solids Production, g VSS/g BOD5 Applied
Solids Production, g SS/g BOD5 Removed
g SS/g COD Applied
g VSS/g COD Applied
g SS/g COD Removed
Solids Production
Solids Production
Solids Production
Solids Production, g SS/g BOD5 Removed
(excluding effluent solids)
Solids Production, g SS/g COD Removed
(excluding effluent solids)
Solids Production, g VSS/g COD Removed
(excluding effluent solids)
Solids Production, mg/1
Suspended Solids Production in Effluent,
Recycle Suspended Solids, mg/1
Recycle Volatile Suspended Solids, mg/1
Recycle COD, mg/1
Value or
Average
2.4
149.5 (39,500)
23 (6)
21.6 (529)
154
2155
1590
2.2
0.67
0.90
0.66
0.99
0.44
0.32
0.52
0.92
0.49
0.36
96.5
7.3
9760
7160
11,750
Standard
Deviation
39.7
141
99
952
624
1204
105
-------
TABLE 56. INFLUENT AND EFFLUENT CHARACTERISTICS FOR THE
14.91 m3 COMPLETE MIX SYSTEM AT A 2.2 DAY SRT
(AUG. 11 - SEPT. 3, 1974).
Influent Effluent
BOD, mg/1
COD, mg/1
TKN, mg/1
(N02 + N03)-N, mg/1
SS, mg/1
VSS, mg/1
Temperature, °C
Mean
107
218
22.3
0
116
91
27.1
Std.
Devi at.
15.6
20.8
1.1
0
15.8
11.5
0.64
Mean
9.9
33.9
14.2
0
7.1
5.1
Std.
Devi at
3.8
5.8
1.7
0
3.9
2.7
106
-------
TABLE 57. SUMMARY OF SYSTEM OPERATION AND PERFORMANCE
WITH THE 12.72 AND 14.91 m3 COMPLETE MIX
SYSTEMS.
STEADY STATE PERIODS
Days at Equilibrium 24 43 27 19
Average Flow, m3/day 149.5 123.4 128.3 143.8
Average Flow, gpd 39,500 32,600 33,900 38,000
Detention time, hrs. 2.4 2.5 2.8 2.5
Average Temperature, °C 27.1 18.4 17.6 24.1
SVI, ml/qm 154 96 166 37
SRT, days 2.2 3.9 5.3 8.4
F/M, g BOD5 Applied/g MLVSS/day 0.67 0.37 0.32 0.23
Solids Production, g SS/g BOD5 Applied 0.90 0.95 0.84 0.73
Solids Production, g SS/g COD Applied 0.44 0.47 0.43 0.39
Solids Production, g VSS/g BOD5 Applied 0.66 0.68 0.59 0.52
Solids Production, g VSS/g COD Applied 0.32 0.34 0.30 0.28
Effluent BOD5, mg/1 9.9 20.1 16.1 18.7
Effluent COD, mg/1 33.9 66.0 58.7 54.7
Effluent SS, mg/1 7.1 30 23 15.3
Effluent VSS, mg/1 5.1 22 17 12.1
Effluent (N02 + N03)-N, mg/1 0000
Volatile Solids, % 73.6 71.6 69.4 70.1
Recycle COD/Recycle SS 1.20 1.11 1.05 1.07
Recycle COD/Recycle VSS 1.64 1.55 1.52 1.54
107
-------
the laboratory for analysis. Results of these analyses are presented in
Table 58. Data obtained during periods of steady-state operation are pre-
sented in Table 59.
As shown in Table 59, there was no direct correlation between soluble
effluent COD and process SRT. The combination of temperature, SRT and
influent wastewater characteristics all combine to control soluble effluent
quality. Since the influent loading varies considerably over a 24-hour
period, it is difficult to draw conclusions between different systems with
different flow through times-patterns on the basis of 4-5 mg/1 differences
in COD values in the clarified effluent at a given point in time. Certainly
these data indicate that the soluble effluent COD's are insensitive to pro-
cess loading over a wide range. Furthermore there is not enough difference
among the systems to indicate that any one is more desirable than the other
on the basis of soluble effluent quality.
H. Sludge Settling Characteristics
The SVI's and settling characteristics of the fourth pass MLSS from the
step aeration system were previously summarized in Figure 4 and Table 6,
respectively. It will be recalled that the SVI remained reasonably stable
over a large range of process loadings. Furthermore the settling velocities
generally exhibited the type of relationship to suspended solids concentra-
tions and temperature that would be anticipated.
A summary of the settling velocities from the plug flow and complete mix
systems is presented in Tables 60 and 61. Examination of these Tables
reveals a great deal of variability in the rates.
The settling velocities from the step feed, plug flow and complete mix
systems are shown as a function of mixed liquor suspended solids in Figures
10, 11, and 12, respectively. For the complete mix and plug flow systems
the only values shown were those obtained 10 or more days after any HpOo
addition. This arbitrary time limit was used to avoid illustrating the
immediate improvement in settling which followed hLO^ addition since this
obviously results in higher settling velocities than are "normal". The
possible influence of the differing temperature was evaluated by grouping
the data over three narrow temperature ranges. It can be seen that there
was no significant relationship between the settling rates observed at a
given suspended solids level and the process temperature. In other words
although temperature will influence the settling rate for a given sample at
some fixed point in time, it could not be correlated to the overall relation-
ship between settling velocities and suspended solids.
Of the 41 settling tests with the step feed system, the minimum settling
velocity observed was 2.4 m/hr (8 ft/hr). At solids concentrations less
than 3500 mg/1 only 5 of the 49 settling tests shown in Figure 11 for the
plug flow system gave velocities of 2.4 m/hr or less. In contrast, 24 out
of 46 settling tests with the complete mix system produced settling velocities
of 2.4 m/hr or less at suspended solids concentrations less than 3500 mg/1
(Figure 12). The wide scatter in settling velocities with the complete mix
system was anticipated in view of all of the operating difficulties which
108
-------
TABLE 58. COD ANALYSES OF CLARIFIER EFFLUENT GRAB SAMPLES
OBTAINED BETWEEN 0800-0900 HOURS.
COD, mg/1
Date
1974
4-22
4-23
4-24
4-25
4-26
4-30
5-1
5-2
5-7
5-8
5-9
5-13
5-14
5-15
5-16
6-12
6-14
6-18
Step System Plug
Total Filtered Total
36.4
45.1
64.3
45.7
41.5
37.5 22.3
39.1 22.0
41.6 23.4
56.0 22.2
30.4
32.5
System
Filtered
18.6
23.8
21.9
21.0
21.1
26.1
19.5
23.6
Complete Mix Complete Mix*
Total Filtered Total Filtered
62.6
55.3
38.6
48.7
50.0
51.6
47.2
42.2
39.9
32.4
30.6
27.4
31.6
32.1
31.9
34.7
24.7
27.0
-------
Date
1974
6-19
6-20
6-24
7-8
7-10
7-12
7-30
7-31
8-5
8-6
8-12
8-13
8-14
8-15
8-19
8-21
8-22
8-23
8-26
8-29
TABLE 58.
(Continued)
COD,
Step System Plug System
Total Filtered Total Filtered
28.2 24.2
23.6
35.3 16.9
11.9
18.8
, 17.3 17.3
14.0
13.2 16.1
18.9
18.4 18.0
mg/1
Complete Mix
Total Filtered
43.3 28.6
49.0 29.4
39.3 28.1
31.1 27.9
30.9 26.2
27.6 21.1
21.3
11.5
22.8
27.2
24.2
23.5
21.0
19.0
20.4
26.2
Complete Mix*
Total Filtered
54.6 29.0
65.7 27.8
26.3
26.5
24.7
19.6
-------
TABLE 58.
(Continued)
COD, mg/1
Date
1974
8-30
9-10
9-12
9-13
9-18
9-19
9-24
9-25
10-1
10-3
10-8
10-10
11-25
11-27
12-3
12-5
12-10
12-12
12-13
Step System Plug System
Total Filtered Total Filtered
23.7 23.7
26.1
19.8
19.1
29.8
30.8
34.9
29.0
25.1 33.6
28.6
35.6
37.2
32.4 29.9
30.6 26.8
28.7
28.5
37.9 28.6
36.7
30.7
Complete Mix Complete Mix*
Total Filtered Total Filtered
26.5
22.5
24.1
35.4
31.6
33.0
32.3
42.3
44.5
40.5
23.8
23.9
30.2
20.9
* The 14.91 m complete mix reactor
-------
TABLE 59. RELATION OF FILTERED EFFLUENT COD
VALUES TO PROCESS SRT
Sys tern
Step Feed
Plug Flow
Complete
Mix
Period of
Steady State
1974
4/16-5/17
8/11-9/4
9/18-10/19
11/12-12/12
4/7-4/25
6/1-7/11
8/16-9/12
9/24-10/21
11/7-12/14
6/7-6/30
8/9-8/31
9/11-10/10
12/3-1/16/75
6/23-7/11*
8/11-9/3*
SRT
Days
8.0
5.9
4.1
3.7
6.6
4.4
2.9
1.9
4.7
2.6
2.1
1.8
8.1
8.4
2.2
Number of
Grab Samples
4
7
3
4
4
5
7
4
6
6
9
9
4
2
4
FILTERED
Mean
22.5
16.8
27.6
34.4
21.3
23.4
20.0
35.3
28.9
28.8
21.8
34.0
24.7
28.4
24.3
COD, mg/
Standa
Devi at
0.6
4.1
2.1
3.5
2.2
2.4
3.6
1.5
1.3
3.3
4.7
7.6
3.9
0.8
3.2
* 3
The 14.91 m complete mix reactor
112
-------
TABLE 60. 'MIXED LIQUOR SETTLING VELOCITIES
IN THE PLUG FLOW SYSTEM.
DATE
1974
1-15
1-22
2-4
2-11
2-19
3-5
3-18
3-25
4-1
4-8
4-15
4-22
4-29
5-6
5-20
6-3
6-10
6-24
7-1
7-15
7-22
7-29
8-5
8-19
8-26
9-3
9-10
9-16
SUSPENDED
SOLIDS
mg/1
2450
3700
5000
5450
4850
5150
5800
5450
4200
4100
3950
4100
4350
2550
2000
2350
2050
2700
2250
2550
2200
2150
1600
1450
1550
1400
1700
1300
TEMPERATURE
°C
15.0
16.0
16.0
15.5
15.5
18.0
17.0
17.0
17.0
18.5
19.5
20.0
21.0
20.0
23.0
22.5
25.5
24.5
25.5
27.5
27.0
27.5
27.5
27.5
27.0
27.0
25.5
24.5
STIRRING
SPEED
rph
0
0
13.5
13.5
15
15
14
15
15
15
15
15
15
15
15
15
15
20
15
15
15
15
15
15
15
15
15
15
SETTLING
VELOCITY
ft/hr
11.8
5.6
3.4
3.0
3.1
2.7
3.3
6.0
6.9
5.2
7.4
6.9
7.9
9.9
11.5
12.8
14.3
9.4
15.1
8.0
10.0
9.5
10.9
11.7
20.1
25.8
18.0
15.4
m/hr
3.6
1.7
1.0
0.91
0.94
0.82
1.0
1.8
2.1
1.6
2.3
2.1
2.4
3.0
3.5
3.9
4.4
2.9
4.6
2.4
3.0
2.9
3.3
3.6
6.1
7.9
5.4
4.7
113
-------
DATE
1974
9-26
9-30
10-10
10-15
10-22
11-4
11-11
11-18
11-25
12-2
12-9
12-16
12-26
12-30
SUSPENDED
SOLIDS
mg/1
1350
1400
1500
1450
1450
2400
2950
3250
3200
2800
3200
2600
3000
3150
TABLE 60.
(Continued)
TEMPERATURE
OC
24.0
23.0
24.0
24.5
23.0
22.0
22.5
21.5
20.5
17.5
19.0
18.0
17.5
18.5
STIRRING
SPEED
rph
15
15
15
15
15
15
15
15
15
15
15
15
15
15
SETTLING
VELOCITY
ft/hr
10.7
20.0
13.5
16.7
22.9
17.7
11.0
9.8
9.8
8.0
11.5
5.5
7.4
4.6
m/hr
3.3
6.1
4.1
5.1
7.0
5.4
3.4
3.0
3.0
2.4
3.5
1.7
2.3
1.4
1975
1-9
1-14
1-20
1-27
2-3
2-10
2-20
3-4
3-10
3-18
3-24
3-31
3700
4300
3900
3600
2850
3650
4750
4050
3300
2500
2550
2100
18.0
16.0
16.5
16.5
17.0
17.5
16.5
17.0
17.0
17.0
18.0
16.5
15
15
15
15
15
15
15
15
15
15
15
15
4.2
5.2
5.7
10.5,
15.3
12.7
10.7
11.2
9.1
16.0
22.8
20.7
1.3
1.6
1.7
3.2
4.7
3.9
3.3
3.4
2.8
4.9
6.9
6.3
114
-------
DATE
1975
4-7
4-14
4-24
4-28
5-5
5-12
5-19
5-27
6-2
6-10
6-17
6-23
6-30
7-8
7-14
7-22
7-31
SUSPENDED
SOLIDS
mg/1
2400
2750
2250
3200
3200
3650
2700
2100
2700
2300
1850
1850
1950
1750
2100
1900
1800
TABLE 60.
(Continued)
TEMPERATURE
OC
17.0
18.0
19.5
18.0
19.5
20.5
21.0
23.5
22.0
23.0
25.0
24.5
23.5
25.5
24.0
26.0
26.5
STIRRING
SPEED
rph
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
SETTLING
VELOCITY
ft/hr
16.5
10.6
13.6
10.0
8.5
8.0
9.4
11.3
8.0
12.2
24.5
m/hr
5.0
3.2
4.1
3.0
2.6
2.4
2.9
3.4
2.4
3.7
7.4
33.0 10.1
36.0 11.0
28.0
13.0
25.2
13.0
8.5
4.0
7.7
4.0
115
-------
TABLE 61. MIXED LIQUOR SETTLING VELOCITIES
IN THE COMPLETE MIX SYSTEM.
DATE
1974
1-9
1-11
1-16
2-11
2-20
2-25
3-5
3-18
3-25
4-1
4-8
4-17
4-22
4-29
5-7
5-13
5-20
6-4
6-10
6-24
7-1
7-8
7-15
7-22
7-29
8-5
8-19
8-26
SUSPENDED
SOLIDS
mg/1
2250
2300
2450
5300
5100
4800
5750
4600
4000
4000
4200
4400
3400
3000
2450
2600
2300
1850
1900
2100
1650
1950
1450
1250
1250
1600
1550
1500
TEMPERATURE
OC
16.0
14.5
16.0
15.5
15.5
15.0
17.5
17.0
17.0
16.5
18.0
19.5
20.5
21.0
20.0
21.0
22.5
23.5
26.0
24.5
25.5
27.0
27.5
27.0
27.0
27.5
27.5
27.5
STIRRING
SPEED
rph
0
0
0
13.5
14
15
15
14
15
15
15
15
15
15
15
15
15
15
15
20
15
15
15
15
15
15
15
15
SETTLING
VELOCITY
ft/hr m/hr
19.1
22.3
7.1
3.8
2.7
2.8
2.9
5.0
6.2
5.6
6.4
2.8
4.3
7.0
7.6
5.7
8.1
9.5
14.3
11.4
13.5
12.9
17.3
6.1
10.1
6.5
7.4
12.6
5.8
6.8
2.2
1.2
0.82
0.85
0.88
1.5
1.9
1.7
2.0
0.85
1.3
2.1
2.3
1.7
2.5
2.9
4.4
3.5
4.1
3.9
5.3
1.9
3.1
2.0
2.3
3.8
116
-------
TABLE 61.
(Continued)
DATE
1974
9-3
9-10
9-16
9-30
10-10
10-22
10-29
11-4
11-11
11-18
11-25
12-2
12-9
12-16
12-26
12-31
1975
1-9
1-17
1-21
1-27
2-3
2-10
2-20
3-4
3-10
3-18
SUSPENDED
SOLIDS
mg/1
1400
1050
1150
1150
1350
2100
2200
2150
2000
2100
2300
2850
3350
3200
3150
2650
2700
2450
2500
2550
2700
2300
2450
3050
2200
1750
TEMPERATURE
OC
26.5
25.5
24.5
23.0
24.0
23.0
22.0
22.0
22.5
21.5
20.5
17.5
19.0
19.5
17.5
18.5
18.0
18.0
16.0
16.5
17.0
17.0
17.0
17.5
16.5
17.0
STIRRING
SPEED
rph
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
15
SETTLING
VELOCITY
ft/hr
18.9
17.1
11.5
13.7
4.0
7.1
5.5
6.5
6.8
7.0
5.4
5.0
5.2
4.1
5.8
7.6
6.5
4.7
5.3
6.1
7.9
20.3
20.2
7.9
13.8
28.8
m/hr
5.8
5.2
3.5
4.2
1.2
2.2
1.7
2.0
2.1
2.1
1.6
1.5
1.6
1.2
1.8
2.3
2.0
1.4
1.6
1.9
2.4
6.2
6.2
2.4
4.2
8.8
117
-------
TABLE 61.
(Continued)
DATE
1975
3-25
4-1
4-8
4-15
4-28
5-6
5-13
SUSPENDED
SOLIDS
mg/1
1650
1500
1400
850
725
750
675
TEMPERATURE
oc
17.0
17.5
17.5
18.0
19.0
19.0
20.0
STIRRING
SPEED
rph
15
15
15
15
15
15
15
SETTLING
VELOCITY
ft/hr m/hr
34.4 10.5
30.0 9.1
20.8 6.3
6.3 1.9
10.0 3.0
8.7 2.7
9.0 2.7
118
-------
4O.U
36.0-
32.0-
28.0-
24.0-
_ 20.0 J
1 1 1 1
• M
A 14-18°C
O 18.1-22.5°C
D 22.6-27.5°C
n
0°
o cP
o
-11.0
-10.0
-9.0
-8.0
-7.0
> ? ^D 14.0-
0
z
£ 12.0-
Ul
"* 10.0-
8.0-
6.0-
4.0-
2.0-
Q O
A
0 [§] ° A
E 0
^ ^
O
O
A
1 1 1 1
-5.0
•4.0
-3.0
• 2.0
• 1.0
Ul
O
Ul
1,000
2,000
3,000
4,000
5,000
Figure 10.
SUSPENDED SOLIDS,mg/j
Mixed Liquor Settling Velocities in the 4th Pass of the Step
Feed System as a Function of Suspended Solids Concentrations.
119
-------
k
_c
\
.*-
>•'
^
u
0
UJ
O
z
•J
^
»—
HI
«/»
tw.u
36.0-
32.0-
28.0-
24.0-
20.0-
1 • 1 1
m A14-18°C
O 18.1-22.5°C
ED D 22.6-27.5°C
Q
B dP
a A
Em A
•11.0
10.0
9.0
8.0
7.0
k
_c
^*
? ? E
18.0
16.0-
14.0-
12.0-
10.0-
8.0-
6.0
4.0-
2.0-
: Q o
L AA
A
Q a A
m
Q 6
D°EIO A
_ Q
Q ra-, O *
OQ * A0 ° A A A
f iS1 °o o1^
^A
EXD^ ° 0 0
°GA
A
0 A
AA A-
A
1 — i 1 __,
>-"
1—
u
5.0 O
UJ
>
O
Z
4.0 3
t—
^.
UJ
(/>
3.0
2.0
1.0
*
1200 2400 3600 4800
SUSPENDED SOLIDS,mg/j?
6000
Figure 11
Mixed Liquor Settling Velocities in the Plug Flow System
as a Function of Suspended Solids Concentrations.
120
-------
36.0-
32.0-
28.0-
24.0-
L.
\ 20.0-
£ :
£ 18.0 ;
U
O
i 16.0-
>
g 14.0-
m 12.0-
10.0-
8.0-
6.0-
4.0-
2.0-
1 1 1 1
A 14-18°C
A O 18.1-22.5°C
D 22.6-27.5°C
.
A AA
: H «
Q Q
Q BA
n n
o
a 2 n
A El O Q^ A
D GDA A0 ^
E 0° A
0 A A A"
1 1 1 1
•11.0
-10.0
9.0
8.0
-7.0
k
_c
—
i E
>•'
5.0 0
LU
>
O
z
4.0 j
^^
»—
HI
3.0
2.0
1.0
1200
2400
3600
4800
6000
Figure 12.
SUSPENDED SOLIDS
Mixed Liquor Settling Velocities in the Complete Mix System
as a Function of Suspended Solids Concentrations.
121
-------
were observed. The step feed system, which posed the least operating diffi-
culty, also produced the strongest correlation between settling velocity and
suspended solids.
The settling velocities obtained during steady-state operation (or within
one day thereof) are shown as a function of suspended solids in Figure 13.
Again there is a very random pattern in the velocity-solids relationships.
The velocities generally decline with increasing solids concentrations but
this is a well known relationship. The steady-state velocities also show
that the settling rates tend to be lowest in the complete mix system.
There was also no fixed relationship between process loading and set-
tling velocities for any of the three systems. On three occassions batch
settling curves were developed for the step feed system from settling tests
using reactor effluent, and recycle solids mixed in various proportions with
clarified effluent. These results are presented in Figure 14. The most
interesting results are the settling velocities obtained on December 11, 1974.
It will be recalled that the system was in steady-state operation from Nov-
ember 12-December 12 at a 3.7 day SRT. The settling velocities were 4.5 m/hr
and 4.4 m/hr on November 25 and December 2 respectively, but increased to
17.1 m/hr by December 9. The validity of the December 9 results were sub-
stantiated by the December 11 settling curves. It will also be recalled that
the plug flow system was operated at a steady 4.7 day SRT from November 7-
December 14, 1974. During this period the settling velocities were a remark-
ably uniform 3.4, 3.0, 3.0, 2.4 and 3.5 m/hr. A batch settling curve was
prepared on December 12 and the results are presented in Figure 15. It can
be seen that the settling velocities were quite adequate at this time. An
additional series of settling tests were run on December 17 prior to ^Op
addition. The substantial deterioration in settling velocities occurred
within 5 days and came after eight turnovers at steady-state operation. Also
it must be noted that the deterioration in settling characteristics in the
plug flow system came at the same time the step feed system was showing ex-
ceptionally high settling velocities. The description of process performance
with the complete mix system was sufficient to show no direct correlation
between process loading and uniform settling rates. The wide variation in
settling curves encountered with this system is shown in Figure 16.
The SVI's in all systems were sensitive to the MLSS concentrations. When
the batch settling curves were run in the 0.15 m (6 in) columns a sample was
also placed in a 1-liter cylinder for a determination of the 30-minute sludge
volume. The results of these determinations for the step feed anfl plug flow
systems are shown in Figure 17 and are presented for the complete mix system
in Figure 18. The SVI's in the complete mix system were normally very much
influenced by the solids concentration. This relationship undoubtedly
played a part in the large variations in sludge volumes which were observed.
There is one additional factor that has not been mentioned thusfar in
the discussion of settling characteristics. That factor is the dissolved
oxygen level and stability. The step feed and plug flow systems were
operated with manual control of the dissolved oxygen levels in each pass.
The operators were instructed to maintain the D.O. between 1-2 mg/1. The
measurement of D.O. levels every 4 hours plus adjustment of the air flow rates
122
-------
u
O
O
z
60
50
40-•
30
25
20
15--
10
8
6
5
4
3--
1.0--
A
O
0
A
El
O PLUG FLOW SYSTEM
D COMPLETE MIX SYSTEM
A STEP FEED SYSTEM
El
[
El
4-
+
o
°
4-
1,000 2,000 3,000 4,000
SUSPENDED SOLIDS, mg/ji
5,000
Figure 13. Mixed Liquor Settling Velocities During Steady State Operation
with the Step Feed, Plug Flow and Complete Mix Systems.
123
-------
u
O
O
z
20"
..16
3/12/74 15-16.5°C
A 8/21/74 26-26.75°C
O 12/11/74 18.5-19°C
U
O
O
Z
1,000
2,000
3,000
4,000
5,000
SUSPENDED SOLIDS mg/£
Figure 14. Batch Test Settling Velocities for the Step Feed System.
124
-------
O 6/19/74 23°-24°C
Q 10/3/74 24°-24.5°C
A 12/12/74 17.75°-19°C
V 12/17/74 17.25°-17.5°C
••12
••10
I* e
•6
•4
1,000 2,000 3,000 4,000
SUSPENDED SOLIDS, mg/J?
5,000
Figure 15. Batch Test Settling Velocities for the Plug Flow System.
125
-------
50 ••
40--
30-
O
z
20-
10 ••
0
O 6/26/74 23°-24°C
Q 8/22/74 25.50-26.50C
A10/8/74 23°-23.5°C
Q12/13/74 17.5°-19.5°C
1,000 2,000 3,000 4,000
SUSPENDED SOLIDS, mg/Ji
14
••12
10
O
6 ?
•4
•2
5,000
Figure 16. Batch Test Settling Velocities for the Complete Mix System.
126
-------
300 -•
200 --
100 -•
E
o>
>
(SI
300 --
200 -•
100 ••
STEP FEED SYSTEM
O 3/12/74
A 8/21/74
Q 12/11/74
PLUG FLOW SYSTEM
Q 6/19/74
A 10/3/74
Q 12/12/74
Q 12/17/74
—I 1
1000 2000 3000
SUSPENDED SOLIDS,mg/j2
4000
-t-
5000
Figure 17. Effect of Suspended Solids Concentrations on SVI's for the
Step Feed and Plug Flow Systems.
127
-------
600
500 •
400 ••
E
o>
^300 -f
200 -
100 ••
COMPLETE MIXED SYSTEM
O 6/26/74
A 8/22/74
Q 10/8/74
O 12/13/74
1,000 2,000 3,000 4,000
SUSPENDED SOLIDS,rng/J?
1
5,000
Figure 18. Effect of Suspended Solids Concentrations on the
SVI's for the Complete Mix System.
128
-------
as needed produced acceptable control. However the levels cycled consid-
erably, and it was not uncommon to obtain individual readings between 0.5-
4 mg/1. Since the plug flow system involved eight separate passes and the
step feed system four separate passes all with individual D.O. control, the
end result was that neither of these systems was operated within a narrow
D.O. limit.
When automated D.O. control was added to the 12.72 m3 reactor in late
December 1973, the sludge SVI's and settling characteristics became very
stable by mid-January and remained this way through April. This was the
longest period of stable D.O. control that has ever been achieved with any
complete mix system operated at the pilot plant. D.O. control was added to
the complete mix system at the end of February 1974. When the system was
functioning correctly, the D.O. was maintained at 1.0 ± 0.5 mg/1. However
there were enough failures in the D.O. sensor, the computer hardware, and
the automatic valve/actuator systems that there were no periods of extended
D.O. stability (more than 3 weeks). In some cases the failure was corrected
within a day, and in other cases several days elasped before needed repairs/
recalibrations were made. Whenever a failure occurred, the operators
resumed manual control of the air supply rate. Under manual control the D.O.
levels cycled much like in the step feed and plug flow systems.
The periods of good stable D.O. control in the complete mix system always
corresponded to periods of acceptable settling characteristics. In all cases
the bulking difficulties which were encountered in the complete mix system
throughout the study followed periods where the D.O. control was not stable.
In this case instability means that the D.O. levels fluctuated in the complete
mix system to the same extent as encountered in the plug flow and step feed
systems on a daily basis. There are insufficient data to describe a quanti-
tative relationship between D.O. stability, D.O. level, SVI and settling
characteristics.
129
-------
SECTION 8
DISCUSSION
As indicated in the Introduction, the District of Columbia municipal
wastewater is essentially a relatively weak domestic waste. A combined
sewer system results in large rainwater flows. Hence the results obtained
from this investigation and the discussion which follows are centered on a
wastewater which is relatively easy to treat.
The amount of suspended solids in the effluent was the primary factor
governing variations in carbonaceous effluent quality. This was more im-
portant than either the type of process or the particular loading under
consideration. Even when the complete mix system was operated at a 1.5 day
SRT, the average effluent 8005 was only 7 mg/1 during the period when the
effluent suspended solids were low (5.5 mg/1). This indicates that except
for the chemostat study it is entirely reasonable to compare the three systems
on the basis of applied loading. The F/M vs. SRT relationship is compared
for the step feed, plug flow and complete mix systems on the basis of applied
BODs in Figure 19 and applied COD in Figure 20. Examination of these figures
reveals no noticeable differences among the three systems. This is consist-
ent with the various studies summarized in Section 4.
The data in Figures 19 and 20 have been summarized in Table 62. It will
be recalled that equation (12) in Section 4 indicated a linear relationship
between the F/M ratio (based on substrate removed) and the inverse SRT. The
relationships for applied BOD5 and COD loadings vs. the inverse SRT are shown
in Figure 21. Again there is no discernable difference among the three systems.
A linear regression analysis of the data shown in Figure 21 reveals the fol-
lowing relationships:
where (F/M) = g BOD5 Applied/g MLVSS/day
|- =0.794 (fo - 0.064
c
where (F/M) = g COD Applied/g MLVSS/day
± = 0.381 (£) - 0.059
The linear correlation coefficient based on applied BODs loading was 0.985,
and 0.982 based upon applied COD loading.
Before proceeding further, it is advantageous to remember that the data
in Figures 19 thru 21 and Table 62 cover an 11°C range of wastewater temper-
130
-------
o>
•V.
X
o
•o
1.0 •
0.9-
0.8-
07-
0.6-
g; 0.5 f
§ 0.4 - -
CD
CD
^ 0.3 - -
ik
03--
0.1 -
V STEP FEED SYSTEM
O PLUG FLOW SYSTEM
O COMPLETE MIX SYSTEMS
H h
4 5
SRT,days
Figure 19. F/M Ratio Based on g BOD5 Applied per g MLVSS vs Process SRT.
131
-------
X
o
TJ
0.
Q.
<
O
O
o>
2,0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2 -•
V STEP FEED SYSTEM
D PLUG FLOW SYSTEM
O COMPLETE MIX SYSTEMS
SRT, days
Figure 20. F/M Ratio Based on g COD Applied per g MLVSS vs Process SRT.
132
-------
TABLE 62 . RELATION OF PROCESS LOADING
PER UNIT OF MLVSS TO 1/SRT.
System
STEP
FEED
PLUG
FLOW
COMPLETE
MIX
SRT
days
3.7
4.1
5.9
8.0
9.0
1.9
2.9
3.5
4.4
4.7
5.7
6.6
1.5
1.8
2.1
2.2
2.6
3.9
5.3
7.1
8.1
8.4
1/SRT
days'1
0.270
0.244
0.169
0.125
0.111
0.526
0.345
0.286
0.227
0.213
0.175
0.152
0.667
0.556
0.476
0.455
0.385
0.256
0.189
0.141
0.123
0.119
F/M
g BOD5 Applied/day
g MLVSS
0.46
0.41
0.31
0.20
0.17
0.79
0.51
0.46
0.42
0.36
0.37
0.26
0.84
0.76
0.68
0.67
0.53
0.37
0.32
0.26
0.22
0.23
g COD Applied/day
g MLVSS
1.00
0.85
0.64
0.42
0.35
1.63
1.08
0.94
0.81
0.76
0.74
0.55
1.72
1.64
1.36
1.37
1.03
0.75
0.63
0.54
0.47
0.43
133
-------
V STEP FEED
SYSTEM
' 02 0.4 0.6 0.8 1.0
F/M, gBOD5APPLIED/day
gMLVSS
D PLUG FLOW
SYSTEM
0.8 r
O COMPLETE MIX
SYSTEMS
w>
X
o
TJ
0.2 0.4 0.6 0.8 1.0 1.2 1.4
F/M, gCOD APPLIED/day
gMLVSS
Figure 21. F/M Ratio Based on BOD5 and COD Applied vs 1/SRT.
-------
atures. Gujer and Jenkins (22) observed a 25% increase in the yield coef-
ficient as the temperature was reduced from 21°C to 11°C. Hopwood and
Downing (14) reported that their results suggested that the sludge growth
index passed through a maximum at a temperature of about 9°C. Muck and
Grady (35) also reported variations in the yield and decay constants as the
temperature was changed, and indicated that this has also been observed by
others in pure culture studies. For this reason, it is not terribly helpful
to attempt to estimate substrate removal to the nearest 1 mg/1 in the hope
of refining the yield and decay coefficients and obtaining the "true" values
for the aggregate data.
The relationships between applied loading based on reactor MLSS and the
inverse SRT are summarized in Table 63, and are as follows:
where (F/M) = g BOD5 Applied/g MLSS/day
^ = 0.999 (jjj) - 0.044
c
where (F/M) = g COD Applied/g MLSS/day
i = 0.479 (fo - 0.039
c
where (F/M) = g (BOD5 + SS) Applied/g MLSS/day
|- = 0.504 (£) - 0.045
c
The linear correlation coefficient based on BOD5 was 0.985: based on COD,
0.982; and based on (BOD5 + SS), 0.977. It can be seen that with District
of Columbia primary effluent the sum of the influent BODs and influent SS
is essentially the same as the influent COD. Although the yield and decay
coefficients based on the sum of the influent BOD5 and SS are somewhat dif-
ferent than reported by others (18), it is interesting that this loading
parameter also correlates well with the SRT.
If the inverse of the observed yield is evaluated as a function of SRT
it is also possible to evaluate the system constants b and Y (Equation 10).
Examination of the data in Table 64 reveals the observed yields as a function
of both COD and 6005 applied. The average process temperature is also
indicated. When the inverses of the observed yields are expressed as a func-
tion of SRT the following are obtained:
based on 6005 applied
i Y =
-L r\ f\ r~ r~ f-< i i ^ ~r o '
Y = 0.055 0 + 1.373 ; ,
based on COD applied
—— = 0.106 0 + 2.843; Y =
'obs c -' —' b = 0.037
135
-------
TABLE 63. RELATION OF PROCESS LOADING PER UNIT OF MLSS TO 1/SRT.
System
STEP
FEED
PLUG
FLOW
COMPLETE
MIX
SRT
days
3.7
4.1
5.9
8.0
9.0
1.9
2.9
3.5
4.4
4.7
5.7
1/SRT
days
-1
F/M
6.6
1.5
1.8
2.1
2.2
2.6
3.9
5.3
7.1
8.1
8.4
.270
.244
.169
.125
.111
.526
.345
.286
.227
.213
.175
.152
.667
.556
.476
.455
.385
.256
.189
.141
.123
.119
g BOD5 Applied/day
g MLSS
0.349
0.316
0.220
0.147
0.127
0.614
0.377
0.339
0.318
0.264
0.261
0.186
0.642
0.608
0.511
0.498
0.410
0.267
0.223
0.187
0.160
0.161
g COD Applied/day
g MLSS
0.748
0.655
0.452
0.311
0.257
1.26
0.798
0.692
0.617
0.561
0.519
0.400
1.32
1.31
1.03
1.01
0.792
0.538
0.437
0.393
0.339
0.303
q(SS + BODc;) Applied/day
g MLSS
0.757
0.608
0.465
0.300
0.253
1.18
0.802
0.605
0.609
0.570
0.493
0.376
1.20
1.21
1.06
1.04
0.799
0.536
0.444
0.344
0.337
0.298
-------
TABLE 64. RELATION OF OBSERVED YIELD COEFFICIENTS TO SRT.
System
STEP
FEED
CO
PLUG
FLOW
COMPLETE
MIX
SRT
days
3.7
4.1
5.9
8.0
9.0
6.6
1.5
1.8
2.1
2.2
2.6
3.9
5.3
7.1
8.1
8.4
Avg.
Temp.
19.2
22.9
27.0
20.0
15.8
Observed Yields
1/Y
obs
1.9
2.9
3.5
4.4
4.7
5.7
22.5
26.5
25.3
24.0
19.4
16.5
19.0
19.1
23.4
27.0
27
23
18
17.6
16.6
17.6
24.1
g VSS Prod.
g BOD. Appl.
0.58
0.59
0.55
0.63
0.64
0.66
0.69
0.62
0.54
0.59
0.47
0.60
0.80
0.73
0.72
0.67
0.71
0.68
0.59
0.55
0.56
0.52
g VSS Prod,
g COD Appl,
0.27
0.28
0.27
0.30
0.32
0.32
0.32
0.31
0.28
0.28
0.24
0.28
0.39
0.34
0.36
0.32
0.37
0.34
0.30
0.26
0.26
0.28
g BODc^ Appl .
g VSS Prod.
1.72
1.69
1.82
1.59
1.56
1.52
1.45
1.61
1.85
1.69
2.13
1.67
1.25
1.37
1.39
1.49
1.41
1.47
1.69
1.82
1.79
1.92
g COD Appl .
g VSS Prod.
3.70
3.57
3.70
3.33
3.13
3.13
3.13
3.23
3.57
3.57
4.17
3.57
2.56
2.94
2.78
3.13
2.70
2.94
3.33
3.85
3.85
3.57
-------
The relatively small variation in observed yields over the SRT range examined
results in a poor correlation when using this method. The linear correlation
coefficient based on BODs was 0.625 and based on COD was 0.605. It is also
apparent by visual examination of the data, that considering the average pro-
cess temperature does not account for all of the deviations observed.
All systems were operated at steady-state flow to reduce variations in
reactor and waste suspended solids concentrations and to eliminate the need
for flow proportioning influent and effluent samples. Based upon the results
of Boon and Burgess (19), the sludge production values should not have been
noticeably influenced by the absence of a diurnal flow cycle. Even at steady
state flow, the organic loading to the units cycled considerably over the
course of a day as the strength of the incoming sewage changed. The influent
COD pattern to each of the activated sludge systems over a 5-day period in
January 1975 is presented in Figure 22. Thus even though the results describe
"steady-state" operation., the organic loading cycled more than 2:1 during the
course of most days because of the changing sewage strength.
In general the observed yield and decay coefficients are in good agree-
ment with those values reported in the literature for the past 25 or so years.
In fact if the ratio of influent SS to 6005 in other studies is considered,
there is a relatively good degree of agreement. Data from several of the
investigations described in Section 4 have been summarized in Table 65. It
is not possible to arrive at a uniform basis of comparison for all of the
studies. The results of Boon and Burgess (19) at a F/M ratio of 0.54 agree
closely with those of Wuhrmann (9) at a F/M ratio of <"» 0.5. In both of these
cases, the BODsiSS ratio was ^> 2:1. When the data of Boon and Burgess (19)
and Wuhrmann (9) are compared at a loading of^ 0.85 but with different 6005:
SS ratios, the data of Wuhrmann indicate higher sludge production. On the other
hand, the yield coefficient of Smith and Filers (17) is somewhat higher than
expected for a 1.6:1 BODc^SS ratio. If one assumes a soluble residual COD
of 10% of the average influent COD in the present investigation, the yield
coefficient would be 0.42 g VSS/g COD removed and the decay coefficient would
be 0.06. This agrees well with the results of Gujer and Jenkins (22), but is
somewhat higher than reported by Jenkins and Garrison (6). The highest yield
coefficient reported (11) was for raw unsettled sewage, and this is precisely
what would be anticipated.
It is worth reemphasizing at this point that yield and decay coefficients
based on 6005 or COD removal can be completely misleading when the removals
are based on the residuals in the clarified effluent. These residuals fall
into three broad nonmutually exclusive groups of materials. These are: (a)
the refractory or very slowly degraded residuals which tend to remain no
matter what the loading, (b) the effluent biological suspended solids or
other colloidal material which passes into the clarifier effluent as a func-
tion of solids/liquid separation efficiency in the final clarifier, and (c)
those degradable materials in the influent wastewater or metabolic intermedi-
ates/end products thereof whose transformations can be described (however
imperfectly) by the kinetic equations presented in Section 4. Over a large
range of process loadings, the variations in carbonaceous effluent residuals
are more influenced by the suspended solids carry-over in the effluent than
138
-------
300-r
0)
E
Q'
O
u
t—
Z
ui
=>
u.
u.
UI
200-
100-•
9
3
«r
N
1
0
O
CM
"
1
O
O
^
CM
1
O
O
CM
1
O
O
^
CM
TIME,
1
0
0
CM
hours
— i
o
O
^
CM
1
O
O
CM
1
O
O
CM
1
O
o
CM
1
O
o
^
CM
Figure 22. Variation in COD of D.C. Primary Effluent During a 5-day Period
(January 26-30, 1975).
139
-------
TABLE 65. SUMMARY OF SLUDGE PRODUCTION DATA FROM VARIOUS INVESTIGATIONS.
REFERENCE
Wuhrmann
(9)
F/M
.-'0.5 g BODg
g MLSS • day
•^0.85 g BOD
g MLSS • day
INFLUENT
BOD:SS
2:1
1:1
OBSERVED
SLUDGE
PRODUCTION
.^0.5 g VSS
g BOD5 Appl.
^0.72 g VSS
g BOD5 Appl.
CELL
YIELD
COEFF.
DECAY
COEFF.
DAY'1
REMARKS
Heukelekian
(7)
1.6:1
0.5 g VSS
g BOD5 Appl.
0.055 Determined from
batch studies.
£ Garrett
0 (ID
1.1 g VSS
g BOD5 Removed
0.08 Decay rate from
reference (18).
Data obtained
from raw, unset-
tled sewage.
McCarty
Brodersen
(13)
0.65 g VSS
g BOD5 Removed
Estimate for mix-
ed wastes such as
domestic sewage.
Hopwood
Downing
(14)
0.8 g
g MLSS • day
0.9 g SS
g BOD5 Appl.
Domestic sewage
from a resident-
ial district.
Middlebrooks
et al.
(5)
0.65 g VSS 0.043
g BOD5 Removed
Average BOD re-
moval was 86%.
-------
REFERENCE
Jenkins
Garrison
(6)
F/M
INFLUENT
BOD:SS
TABLE 65
(Continued)
OBSERVED
SLUDGE
PRODUCTION
CELL DECAY
YIELD COEFF.
COEFF. DAY-1
0.33 g VSS 0.04
g COD Removed
REMARKS
COD removal based
on soluble efflu-
ent. Influent
COD:SS = 2.4:1
Smi th
Eilers
(16, 17)
1.6:1
0.79 g VSS
g 8005 Removed
0.071 Effluent BOD's
were normally
<10 % of influ-
ent values.
Eckenfelder
in (18)
0.73 g VSS
g BOD5 Removed
0.075 Results from set-
tled municipal
wastewater.
Boon
Burgess
(19)
0.54 g BOD5 2:1
g MLSS • day
0.85 g BOD5 2:1
g MLSS • day
0.64 g SS
g BOD5 Appl
Q.73 g SS
g BOD5 Appl
Residential
sewage.
Gujer
Jenkins
(22)
Q.48 g VSS 0.07
g COD Removed
0.38 g VSS 0.07
g COD Removed
Yields at 11°C
and 21°C (0.38)
COD Removal based
on Soluble Efflu-
ent.
-------
INFLUENT
BOD:SS
1:1
TABLE 65.
(Continued)
OBSERVED
SLUDGE
PRODUCTION
CELL
YIELD
COEFF.
0.79 g VSS
DECAY
COEFF.
DAY'1
0.064
REFERENCE F/M INFLUENT OBSERVED CELL DECAY REMARKS
This Study
g BOD5 Appl.
1.00 g SS 0.044
q BOD5 Appl.
-------
the other factors mentioned above. While this may be of no importance for
any given study which is designed to establish the relationship between
effluent quality and process loading, it must be evaluated when attempting
to compare the various yield and decay coefficients in Table 65. Even within
a given study one can arrive at erroneous conclusions unless this factor is
considered. This was discussed in Section 4. Since the influent carbonaceous
material to the biological processes in this study was more than 50% colloidal
(Table 4), one cannot focus strictly on the soluble components to overcome
these difficulties. However as noted by Jenkins and Garrison (6) this approach
obviously does have advantages. In general the soluble residual COD's in this
study were roughly 10% of the average influent values and this was the basis
for the 0.42 g VSS/g COD removed yield coefficient mentioned in the preceding
paragraph.
As discussed in Section 4, there have been several reports (23, 24, 25,
26) that the sludge production from so-called similarly operated air and
oxygen systems was significantly different when treating District of Columbia
primary effluent. It was also indicated that examination of the data from
the air system reactor revealed numerous internal inconsistencies. In fact
those data bear no relationship to the results obtained in the present study.
Since the F/M vs. 1/SRT relationship shown in Figure 21 indicates an excellent
correlation for all of the air activated sludge systems, it was felt that it
would be worthwhile to compare the yield and decay coefficients obtained in
the present study with coefficients from the oxygen system.
The data in the oxygen system" report (26) were examined in an attempt to
calculate the yield and decay coefficients for the oxygen activated sludge
system. Examination revealed asterisks which appear beside flow numbers
which have no footnote explanations; different detention times reported for
the same flow; data analyses periods which do not correspond from Table to
Table; incorrect calculations of F/M ratios, eg. May 1971; nitrogen balances
which always show more nitrogen entering the system than leaving the system
even in the absence of nitrification; data in the Figures which do not always
correspond to data in the Tables; and no discussion of what data from the
26 periods in the Tables were used to produce the 15 data points shown in
the F/M vs. SRT relationship or why the other 11 periods were not included.
In short, the data are of such poor quality and so disorganized that there
is no point in attempting to calculate yield or decay coefficients. It is
apparent, however, that previous claims (23, 24, 25, 26) for substantial
differences in sludge production between what were represented as similarly
operated air and oxygen systems are not justified on the basis of the data
presented.
Throughout much of the study reported herein, the effluent (N02 +
values from the complete mix system were less than would have been anticipated
on the basis of process SRT alone. This appears to be largely the result
of the D.O. control to 1.0 ± 0.5 in the complete mix system throughout most
of the study. Other studies on D.C. wastewater not discussed above have
indicated no apparent interference with nitrification in complete mix systems
at high D.O. levels (> 2 mg/1 D.O.). Also it should be noted that H202 is
quite toxic to the nitrifying organisms, and at the dosages periodically
employed for filamentous control ( r^ 250 mg/1 H202) all nitrifying organisms
143
-------
would have been destroyed. Because an attempt was being made to operate all
systems at D.O. levels which can inhibit nitrification (r-'l.O mg/1), it
would be unwise to draw any conclusions about the absence of nitrification
as either a function of process type or process loading.
Because of the different temperatures, D.O. levels, hydraulic detention
times, etc. there are some questions which come to mind when considering the
aggregate data and the relative performance/response of one system as opposed
to another. Figure 23 summarizes the steady-state periods in chronological
order. In several cases a direct comparison between systems can be made in
a way which resolves some of the more obvious questions.
From March 19-April 12, 1974 the 14.91 m3 complete mix reactor and the
12.72 m3 complete mix reactor were operated at D.O. levels of 2.5 and 1.0
mg/1, respectively. The differences in carbonaceous effluent quality appear-
ed to simply reflect differences in the effluent SS levels. The difference
in observed yields was only 0.08 g VSS/g BODc applied of which 0.04 g VSS
would be expected with a yield coefficient of 0.794, a decay coefficient of
0.064 day1 and a 1.3 day difference in process SRT's.
The large complete mix system was operated with an average aeration time
of 3.4 hours at a 2.1 day SRT from August 9-31, 1974. The 14.91 m3 reactor
was operated at a 2.2 day SRT from August ll-9eptember 3, 1974 with an average
aeration time of 2.4 hours. As shown in Table 58, the differences in filtered
COD values were negligible. Differences in average effluent BODjj, COD and SS
were 2.3, 2.4 and 2.4 mg/1, respectively. The observed sludge yields varied
by < 10%. The observed sludge yield in the plug flow system between August
16-September 12, 1974 at a 2.9 day SRT was essentially the same as in the two
complete mix units although this unit was at a slightly lower loading.
The complete mix unit was operated at a 1.8 day SRT between September 11-
October 10, 1974. The plug flow system was operated at a 1.9 day SRT between
September 24-October 21, 1974. A comparison of the filtered COD values in
Table 58 reveals no significant differences. The observed yield in the plug
flow system was about 10% less than in the complete mix system.
The step feed system was operated at a 3.7 day SRT between November 12-
December 12, 1974. The plug flow system was operated at a 4.7 day SRT from
November 7-December 14. The observed solids production values were essenti-
ally the same although the plug flow system should have been slightly less
relative to the step-feed system. The differences in soluble COD values in
Table 58 were quite small.
Although the average F/M ratio in the step feed system at the 3.7 day SRT
was 0.46 g BOD5 Applied/g MLVSS/day, the average process loading to the last
pass of the reactor was 3.6 g 8005 Applied/g MLVSS/day. Even at this except-
ionally high last pass loading the average residual inhibited BOD5 was only
11.1 mg/1 with an effluent solids concentration of 11 mg/1. It is important
to remember in the step-feed system that the average F/M does not represent
the process loading in the final pass prior to clarification. However since
the residual effluent BOD5 was so low even at the highest loading, the use of
average applied loading in the solids production and F/M relationships poses
no significant error.
144
-------
c
o
E
JAN
FEB
MAR
APR
MAY
JUNE
JULY
AUG
9.0
8.0
/ / / /
5.9
4.1
37
O
_j
u.
O
6.6
4.4
2.9
1.9
47
57
3.5
to
X
O
u
2.6
2.1
1.8
8.1
7.1
1.5
0.26
o
u
5.3
8.4
2.2
a.
s
o
cs
3.9
STEADY STATE SRT,days
Figure 23. Chronological Order of Process SRT's During Steady State Operation
145
-------
The differences of observed yield of ^ 10% between systems mentioned
above are not important even where consideration of the same time periods
eliminates relative differences which may arise owing to the incorrect
measurement of influent waste concentrations. The process flows to each
system were individually measured with magnetic flow meters with the calibra-
tion checked once per week and deviations from the correct values recorded.
These deviations were used to "correct" all totalizer readings. It was not
uncommon to find small deviations of 3.8-7.6 1/min (1-2 gpm) from the metered
values on a week to week basis. By applying the appropriate corrections, it
is felt that the true influent flow was accruately known to within ±1.9 1/min
(±0.5 gpm). At this level of accuracy, the influent flow would be correct to
within ±1.458 at 189.3 m3/day (50,000 gpd) and to within ±2.4% at 113.6 m3/day
(30,000 gpd). The standard deviations of the influent flow measurements
were not presented for any of the steady-state periods because the daily
deviation in the time the totalizer readings were manually recorded (0700
hours ±10 minutes) generally resulted in as much if not more deviation than
resulted from the variation in daily metered flow. The use of the automated
wasting system produced careful control of waste volumes, and it is felt that
the values were measured correctly within ±2.5%. In most cases the standard
deviations of the waste or reactor solids concentrations were 10% or less of
the mean value. As a general point of reference it can be noted that for
25 observations from a normal distribution where the standard deviation is
10% of the mean value, the 95% confidence limits are ±4.1% of the mean value.
A review of the data sumnaries for the step feed, plug flow and complete
mix systems (Tables 17, 32, 47 and 57) reveals no particular relationship
between process loading and effluent quality (except for the chemostat).
In general, the effluent quality deteriorated at the higher SRT values because
of Nocardia and, as has been mentioned several times, the effluent quality
could have been shown to be much worse at the high SRT periods by characteriz-
ing periods of excessive Nocardia growth and high solids carry over in clarifi-
cation. As shown in Table 59, the aggregate data indicate no relationship be-
tween soluble COD residuals and process SRT either within a given system or
among the various systems.
During periods of relatively uniform effluent suspended solids concentra-
tions it is possible to estimate the soluble COD residuals among the systems
without having to deal with excessive data scatter. During August 1974 the
effluent suspended solids concentrations from each unit were multiplied by
1.1 (coefficients estimated from ratios of Recycle COD/Recycle SS)*and this
value was subtracted from the effluent COD concentration. In September the
factor 1.15 was used. Results are presented in Figure 24. It can be seen
that this estimate of the soluble residual COD shows that the two complete
mix systems had the same soluble effluent quality during the period of par-
allel operation at the "same" loading but with different hydraulic detention
times. During September when the plug flow and complete mix systems were
operated at the same loading the soluble effluent qualities were also essen-
tially the same. It should also be noted that the soluble residual COD for
all systems rose during the later part of September. This was a dry month
with essentially no rainfall and the nondegradable residuals rose in all
units as the influent COD also increased. Examination of these data explains
why there is no obvious relationship between the soluble residual COD and the
146
-------
o>
E
O
O
u
50
40
30
20
10
V STEP FEED
D PLUG FLOW
O COMPLETE MIX
O 14.91m3 COMPLETE MIX
I
o~6 r
O)
E
Q
O
12 14 16 18
AUGUST, 1974
14
16
18 20 22 24 26 28 30
Figure 24.
SEPTEMBER, 1974
Estimate of Soluble COD Residuals for Selected Systems
During August and September 1974.
-------
process SRT in the aggregate data in Table 59. Although one can apparently
find small differences in the residual COD values as a function of process
loading at a given point in time, these differences become totally obscured
over long time periods as wet or dry weather plays more of an influence on
the soluble COD residual than does the process loading.
During November and December, 1974 the residual suspended solids were
multiplied by 1.2 and subtracted from the effluent COD in the step feed
system and multiplied by 1.1 and subtracted from the effluent COD's in the
plug flow and complete mix system. Results are presented in Figure 25. It
can be seen that even though the step feed system was operated at a 3.7 day
SRT with an average process loading to the last pass of the system of 3.6
g BOD5 Applied/g MLVSS/day, the soluble COD residual was apparently only
about 6-12 mg/1 higher than in the plug flow system which was being operated
at a F/M ratio of 0.36 g BOD5 Applied/g MLVSS/day.
An attempt was made to compare the 12.72 and 14.91 m3 complete mix systems
by this method during the period that they were operated at different control-
led D.O. levels. Unfortunately the suspended solids concentrations were so
erratic during this time that a comparison of the daily values proved incon-
clusive. Applying this technique to the average values presented in Tables
50, 51 and 52 and using the respective COD:SS ratios from each system yields
a difference in the estimated soluble COD residual of 1.9 mg/1.
One of the more interesting observations when applying this method of
comparison are the results from the 14.91 m complete mix reactor during June
and July 1974. The residual suspended solids in the plug flow, 14.91 m3
complete mix and complete mix system were multiplied by 1.1, 1.1 and 1.15,
respectively and subtracted from the effluent COD values in the appropriate
systems. Results are presented in Figure 26. In contrast to all other
comparisons using this method, the estimated soluble residual COD was
highest in the system with the lowest loading. It will be recalled that
the biological solids present in the 14.91 m3 complete mix reactor at this
time were quite atypical of the normal biomass. The sludge particles were
very dense with no evidence of filamentous growth; the average process SVI
was 37 ml/gm. When the waste rate was increased substantially in July to
eliminate the Nocardia which had developed, the soluble effluent COD values
(as estimated by the subtraction method) became the same as in the similarly
loaded complete mix system (Figure 24). Microscopic examination during August
revealed very similar biota in the two complete mix systems. The tfhly time
the dense discrete floe particles have ever been observed in any system in
more than trace quantities was during June and July 1974.
Because of the changes in the soluble COD component which were observed
during storage (Table 3) it is doubtful that analysis of the soluble compon-
ent of stored composite samples would have been useful. In addition to this,
the results of the brief studies with the chemostat and the results described
above indicate that one would normally be dealing with small changes in
residual COD over large loadings. Attempts to evaluate differences in soluble
BOD over the range of loadings investigated (except for the chemostat) would
largely be futile as shown in Figure 9 and Table 44. As discussed in Section
4, this observation is in accordance with the previous conclusions of many
others.
148
-------
"I
40 +
c*
?'°
§ 20
u
10 +
-4-
-*-
50
40
30
i
Q
20
10 +
10 12 14 16 18 20
NOVEMBER,1974
22 24 26 28
V STH> FEED
D PLUS FLOW
O COMPLETE MIX
30
10
12
14
16 18
20 22
24
26 28
DECEMBER,1974
30
Figure 25,
Estimate of Soluble COO Residuals for Selected Systems
During November and December 1974.
-------
en
O
50
40
30
u 20
10
d
9-
10 12 14 16 18 20 22 24 26 28 30
JUNE,1974
D PLUG FLOW
O COMPLETE MIX
O 14.91 m3 COMPLETE MIX
H-
10
12 14 16 18
JULY,1974
20 22 24 26 28 30
Figure 26,
Estimate of Soluble COD Residuals for Selected Systems
During June and July 1974.
-------
The N, P and COD characteristics of the biomass have been summarized in
Table 66. Because of the sharp rise in the price of Ag?S04, the COD analyses
of the solids were discontinued in early 1975. As mentioned previously, the
results presented for the 12.72 and 14.91 m3 complete mix systems were part
of another investigation and under the direction of another investigator; no
recycle P or TKN measurements were taken. After approximately mid-September
1974, the quality of the TKN and P analyses left something to be desired.
This resulted from a loss of competent laboratory personnel. For this reason
the small deviation of the materials balances from unity for steady state
periods after this time was not considered important. Obviously no material
balances can be made for nitrogen during any period when the system was
nitrifying because of denitrification in the final clarifier. The data
indicate that the COD:SS and COD:VSS concentrations tend to remain relatively
constant and do not deviate from system to system. Similarly the N and P
concentrations show no consistent pattern or deviation from system to system.
Both of these observations are consistent with a number of other literature
reports.
The key to maintaining a good carbonaceous effluent quality was developing
a biomass that gave good solids/liquid separation in the final clarifier. This
was far more important than any of the loadings or systems (except the chemo-
stat) examined. Effluent quality deteriorated either during periods of ex-
cessive Nocardia growth or during any period that the clarifier bed overflowed
because of filamentous growth. The clarifiers for the step feed and plug flow
systems were identical and that for the complete mix system was also virtually
the same size as the other two. On the other hand, the smaller diameters of
the clarifiers for the 12.72 m3 and 14.91 m3 complete mix reactors resulted
in a higher peak overflow rates (calculated by subtracting the surface area
of the center-well skirt and overflow weir inset) than in the other clarifiers.
All overflow rates previously given were based on total clarifier diameter.
For this reason the somewhat higher suspended solids frequently encountered
with the 12.72 or 14.91 m3 complete mix systems could merely reflect some-
what different hydraulic characteristics in the final clarifiers.
The presence or absence of Nocardia could be controlled by selecting a
sufficiently high process loading. On the other hand, the different hydraulic
regimes seemed to have little effect. During February and March 1974, the
plug flow and complete mix systems were operated at similar loadings ( /~» 0.14-
0.18 g BODc/g MLVSS/day) and both systems were plagued with excessive Nocardia.
The step feed system operated at a F/M ratio of 0.17 during February developed
excessive Nocardia growth by early March. During this same time the 12.72
and 14.91 m^ complete mix systems were operated at F/M loadings of 0.3-0.4
with no Nocardia problems.
The relationship of process loading to the presence or absence of
Norcardia during the spring and summer of 1974 is not quite as straight for-
ward^The plug flow system had a small quantity of Nocardia during June at a
loading of 0.42 g BODc/g MLVSS/day. Neither the plug flow or complete mix
systems had any visible traces of Nocardia during August and September at the
high loadings. Nocardia developed in the 12.72 m3 complete mix system during
May as the wastewater temperature increased and the Nocardia was present
151
-------
TABLE 66. SUMMARY OF RECYCLE SOLIDS CHARACTERISTICS.
System
STEP
FEED
PLUG
FLOW
COMPLETE
MIX
SRT
days
3.7
4.1
5.9
8.0
9.0
1.9
2.9
3.5
4.4
4.7
5.7
6.6
1.5
1.8
2.1
2.2
2.6
3.9
5.3
7.1
8.1
8.4
COD
SS
1.29
1.21
1.10
1.10
1.11
1.32
1.12
1.09
1.13
—
1.10
—
1.27
1.12
1.20
1.14
1.11
1.05
1.10*
1.07
1.07
COD
VSS
1.72
1.58
1.53
1.49
1.46
1.69
1.53
1.43
1.54
—
1.49
—
1.62
1.51
1.64
1.51
1.55
1.52
1.52*
1.48
1.54
Suspendi
% P
2.8
2.5
2.1
2.6
2.7
2.5
2.2
2.4
2.2
3.7
2.2
—
1.9
3.1
2.7
—
2.8
___
2.3
2.5
—
ed Solids
% TKN
___
_ _ _
7.7
7.8
9.7
7.6
7.6
7.8
8.8
7.3
—
—
10.6
8.0
—
8.1
—
7.7
7.5
—
Data from Feb. 2-13, 1975 only
152
-------
throughout June at a 4-5 day SRT. The 14.91 m3 reactor also moved into
heavy Nocardia concentrations in early July at a 0.23 F/M ratio but increased
wasting to a loading of/^ 0.7 by August 11 eliminated all traces of Nocardia.
The step feed system was operated at a F/M ratio of only 0.31 during August 11-
September 4 and there were no problems with Nocardia although the loading was
sufficiently low that it would not have been unusual to find this organism.
During the winter of 1975 when the plug flow and complete mix systems
were both back at the lower loadings, the Nocardia returned. However the
concentrations were not as bad as encountered the year before. The rerouting
of the digester elutriation water in January 1975 could have played some role
in the lesser concentrations observed.
Although the Nocardia organism is somewhat filamentous, this organism
was not primarily responsible for all of the bulking problems encountered.
The bulking organisms were morphologically related to filamentous organisms
which may have been Sphaerotilus. Since no bacteriological work was done
other than by microscopic examination, it would not be prudent to say that
this was the causative organism. It is well known that there are a large
number of filamentous organisms that can cause bulking (48).
A simple reading of the operating difficulties with the complete mix
system is sufficient in itself to establish that there was no obvious rela-
tionship between system loading and settling characteristics or process SVI.
Boon and Burgess (19) also observed that changes in settleability measured
during periods of constant flow rates could not be related to any measured
parameter or condition of operation nor could the same settling rates be
obtained from similarly operated units. Hopwood and Downing (14) also noted
that the results of SVI measurements were not very informative with no clear
trend in the data. In general the poorest settling characteristics were
encountered with the complete mix system, and this observation is consistent
with previous reports of Chudoba et al. (37, 38) and Rensink (39). In con-
trast to these two investigations, however, there was no consistent rela-
tionship between settling characteristics and process loading.
In summary, the results of this investigation indicate that a step feed
system constructed with sufficient flexibility to divert or split the flow to
any segments of the reactor offers the best physical arrangement for secondary
treatment of District of Columbia wastewater. This conclusion would extend
to most wastewaters where one is not confronted with quantitative, qualitative
or toxic shock loadings. Even under these conditions, however, the step feed
configuration offers the possibility of rapid adjustment. The obvious advan-
tages of such a system have been enumerated at some length by Busby (49), and
demonstrated in full scale operation by Joyce et al. (50).
153
-------
SECTION 9
REFERENCES
1. Metea If and Eddy, Inc., Hastewater Engineering, McGraw-Hill Book Co.,
New York, p. 5, 1972.
2. Lawrence, A. W. and McCarty, P. L., "Unified Basis for Biological
Treatment Design and Operation," Jour. San. Eng. Piv., Proc. Amer. Soc.
Civil Eng., 96_, SA 3, 757, 1970.
3. Sherrard, J. H. and Schroeder, E. D., "Relationship Between the
Observed Cell Yield Coefficient and Mean Cell Residence Time in the
Completely Mixed Activated Sludge Process," Water Res., 6., 1039, 1972.
4. O'Melia, C. R., "Coagulation and Flocculation," in Physicochemical
Processes by Weber, W. J. Jr., Wiley-Interscience, New York, 1972.
5. Middlebrooks, E. J. and Garland, C. F., "Kinetics of Model and Field
Extended-Aeration Wastewater Treatment Units," Jour. Hater Poll.
Control Fed.. 40_, 586, 1968.
6. Jenkins, D. and Garrison, W. E., "Control of Activated Sludge by Mean
Cell Residence Time," Jour. Water Poll. Control Fed., 40^1905, 1968.
7. Heukelekian, H. E., et al., "Factors Affecting the Quantity of Sludge
Production in the Activated Sludge Process," Sewage Works Jour., 23,
945, 1951.
8. Sawyer, C. N., "Activated Sludge Oxidation. VI. Results of Feeding
Experiments to Determine the Effect of Variables Temperature and Sludge
Concentration," Sewage Works Jour., 12_, 244, 1940.
9. Wuhrmann, K., "High-Rate Activated Sludge Treatment and its Relation
to Stream Sanitation, I. Pilot-Plant Studies," Sewage Works Jour.,
26_, I, 1954.
10. Torpey, W. N. and Chasick, A. H., "Principles of Activated Sludge
Operation," in Biological Treatment of Sewage and Industrial Hastes,
edited by McCabe, J. and Eckenfelder, W. W., Reinhold Publishing
Corp., New York, 1956.
11. Garrett, M. T., "Hydraulic Control of Activated Sludge Growth Rate,"
Sew, and Ind. Wastes., 30, 253, 1958.
154
-------
12. Helmers, E. N., et al., "Nutritional Requirements in the Biological
Stabilization of Industrial Wastes. II. Treatment with Domestic
Sewage," Sewage Works Jour.. 2_3, 884, 1951.
13. McCarty, P. L. and Brodersen, C. F., "Theory of Extended Aeration
Activated Sludge," Jour. Hater Poll. Control Fed., 34_, 1095, 1962.
14. Hopwood, A. P. and Downing, A. L., "Factors Affecting the Rate of
Production and Properties of Activated Sludge in Plants Treating
Domestic Sewage," Jour. Inst. Sew. Purif. (G.B.), 435, 1965.
15. Benedek, P. and Horvath, I., "A Practical Approach to Activated Sludge
Kinetics," Hater Res., _!_, 663, 1967.
16. Smith, R. and Eilers, R. G., "A Generalized Computer Model for Steady-
State Performance of the Activated Sludge Process," U.S. Dept. of the
Interior, Cincinnati, Ohio, 1969.
17. Smith, R., "Estimating the Rate of Sludge Production in the Activated
Sludge Process," Internal EPA Memorandum, Oct. 10, 1972.
18. Design Guides for Biological Wastewater Treatment Processes, prepared
by the City of Austin, Texas and Center for Research in Water Resources,
Univ. of Texas, Water Poll. Control Res. Series, Environ. Prot. Agency,
11010 ESQ 08/71, August 1971.
19. Boon, A. G. and Burgess, D. R., "Effects of Diurnal Variations in Flow
of Settled Sewage on the Performance of High-Rate Activated-Sludge
Plants, Water Poll. Control (G.B.), 493, 1972.
20. Toerber, E. D., "Full Scale Parallel Activated Sludge Process Evalu-
ation," Environ. Prot. Tech. Series, Environ. Prot. Agency, EPA-R2-72-
065, November 1972.
21. Toerber, E. D., et al., "Comparison of Completely Mixed and Plug Flow
Biological Systems," Jour. Water Poll. Control Fed.. 46_, 1995, 1974.
22. Gujer, W. and Jenkins, D., "The Contact Stabilization Activated Sludge
Process-Oxygen Utilization, Sludge Production and Efficiency," Water
Res., 9_, 553, 1975.
23. Stamberg, J. B., et al., "Activated Sludge Treatment with Oxygen,"
Paper presented at 68th National Meeting AIChE, Houston, Texas, 1971.
24. Stamberg, J. B., et al., "System Alternatives in Oxygen Activated
Sludge," Paper presented at 45th Annual Water Pollution Control Feder-
ation Conference, Atlanta, Georgia, 1972.
25. Stamberg, J. B., et al., "Activated Sludge Treatment Systems with
Oxygen," Environ. Prot. Tech. Series, Environ. Prot. Agency, EPA-670/
2-73-073, September 1973.
155
-------
26. Stamberg, J.B., et al., "System Alternatives in Oxygen Activated
Sludge," Environ, Prot. Technol. Series, Environ. Prot. Agency,
EPA-670/2-75-008, April 1975.
27. Parker, D. S., "A Discussion of Air or Oxygen Activated Sludge,"
Presented at the 48th Annual Conference of the California Water
Pollution Control Association, South Lake Tahoe, Calif., April 1976.
28. Kalinske, A. A., "Comparison of Air and Oxygen Activated Sludge Systems,"
Paper presented at 48th Annual Water Pollution Control Federation
Conference, Miami Beach, Florida, October 1975.
29 Drnevich, R. F. and Gay, D. W., "Sludge Production Rates in Activated
Sludge Systems," Proc. 28th Purdue Ind. Haste Conf., 504, 1973.
30. Drnevich, R. F. and Stuck, J. D., "Error Sources in the Operation of
Bench and Pilot Scale Systems Used to Evaluate the Activated Sludge
Process," Paper presented at 30th Purdue Ind. Waste Conf., Lafayette,
Indiana, 1975.
31. Heidman, J. A., "Experimental Evaluation of Oxygen and Air Activated
Sludge Nitrification Systems With and Without pH Control," Environ.
Prot. Technol. Series, Environ. Prot. Agency, EPA 600/2-76-180, 1976.
32. Lawrence. A. W. and Brown, C. 6., "Biokinetic Approach to Optimal
Design and Control of Nitrifying Activated Sludge Systems," Paper
presented at the Annual Meeting of the New York Water Pollution Control
Association, New York City, 1973.
33. Poduska, R. A., "A Dynamic Model of Nitrification for the Activated
Sludge Process," Ph.D. Thesis, Clemson Univ., Clemson, South Carolina,
1973.
34. Sherrard, J. H. and Lawrence, A.W., "Design and Operation Model of
Activated Sludge," Jour. Env. Eng. Piv., Proc. Amer. Soc. Civil Eng.,
99_, 773, 1973.
35. Schroeder, E. D., "The Relationship Between Process Configuration and
Process Performance," Jour. Water Poll. Control Fed., 47_, 1005, 1975.
*
36. Muck, R. E. and Grady, C. P. L., Jr., "Temperature Effects on Microbial
Growth in CSTR's," Jour. Env. Eng. Div., Proc. Amer. Soc. Civil Eng.,
100, 1147, 1974.
37. Chudoba, J., et al., "Control of Activated Sludge Filamentous Bulking-
I. Effect of the Hydraulic Regime or Degree of Mixing in an Aeration
Tank," Water Res., 7_, 1163, 1973.
38. Chudoba, J., et al., "Control of Activated Sludge Filamentous Bulking-
III. Effect of Sludge Loading," Water Res., 8, 231, 1974.
156
-------
39. Rensink, J. H., "New Approach to Preventing Bulking Sludge," Jour.
Water Poll. Control Fed., 46_, 1888, 1974.
40. Bisogni, J. J., Jr. and Lawrence, A. W., "Relationships Between Bio-
logical Solids Retention Time and Settling Characteristics of Activated
Sludge," Hater Res., 5., 753, 1971.
41. Standard Methods for the Examination of Hater and Wastewater, 13th Edition,
American Public Health Association, Inc., Washington, D.C., 1971.
42. Young, J. C., "Chemical Methods for Nitrification Control," Jour. Water
Poll. Control Fed., 45, 637, 1973.
43. Methods for Chemical Analysis of Water and Waste, Technology Transfer,
Environ. Prot. Agency, EPA-625-6-74-003, 1974.
44. Gales, M., Jr., et al., "Methods for Quantitative Determination of
Total Phosphorus in Water," Jour. Amer. Water Works Assn., 58, 1363,
1966.
45. Lechevalier, H. A., "Actinomycetes of Sewage Treatment Plants," Interim
Progress Report No. 2, Rutgers, The State University of New Jersey,
April 1974.
46. Cole, C. A., et al., "Hydrogen Peroxide Cures Filamentous Growth in
Activated Sludge," Environ. Prot. Technol. Series, Environ. Prot.
Agency, EPA-670/2-73-033, October 1973.
47. Heidman, J. A., "Sequential Nitrification-Denitrification in a Plug Flow
Activated Sludge System," Environ. Prot. Technol. Series, Environ. Prot.
Agency, (In Preparation).
48. James, H. L., et al., "Activated Sludge Bulking-A Literature Review,"
Dept. of Civil Eng., Univ. of Minnesota, Minneapolis, Minnesota,
October 1974.
49. Busby, J. B., "Dynamic Modeling and Control Strategies for the Activated
Sludge Process," Ph.D. Thesis, Clemson Univ., Clemson, South Carolina,
1973.
50. Joyce, R. J., et al., "Optimization of an Activated Sludge Plant Using
TOC, Respiration Rate and Sludge Settling Volume Data," Paper presented
at WWEMA Industrial Water and Pollution Conference, Detroit, Michigan,
April 1974.
157
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-77-108
3. RECIPIENT'S ACCESSIOl*NO.
4. TITLE AND SUBTITLE
5. REPORT DATE
PILOT PLANT EVALUATION
OF
ALTERNATIVE ACTIVATED SLUDGE SYSTEMS
.ncruniu/-\ic
August 1977 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
James A. Heidman
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Government of the District of Columbia
Dept. of Environmental Services
EPA-DC Pilot Plant, 5000 Overlook Ave., S.W.
Washington, D.C. 20032
10. PROGRAM ELEMENT NO.
1BC611
11. CONTRACT/eRANTNe.
68-03-0349
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory--Cin., OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final Report - 10/73 to 9/75
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer: Irwin J. Kugelman (513) 684-7633
16. ABSTRACT
Step feed, plug flow and complete mix activated sludge systems were compared on a
pilot plant scale under similar operating conditions with the same municipal waste-
water. The process loading to each system was varied over a wide range during the
course of the investigation. Extended periods of steady state operation at constant
flow provided extensive data on effluent quality, sludge yield, settling character-
istics, etc. at several fixed F/M loadings for each of the system configurations.
All systems demonstrated that the variability in carbonaceous effluent quality was
mostly influenced by the suspended solids concentrations in the effluent over a wide
range of process loadings. Sludge production was the same within experimental error
in all systems at comparable SRT's. Analysis of the aggregate data from all systems
produced a yield coefficient of 0.79 g VSS/g BODs applied and a decay coefficient of
0.064 day~'. The sludge from the complete mix system exhibited the poorest settling
characteristics. A step feed system was found to offer the best physical arrangement
for secondary treatment of District of Columbia wastewater.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI ' dd/GrOUp
Activated Sludge Process
Waste Treatment
Settling
Complete Mix System
Plug Flow System
Step Feed System
System Comparison
13B
13. DISTRIBUTION STATEMENT
Release To Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
170
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
158
U. S. GOVERNMENT PRINTING OFFICE: 1977-757-056/6^88 Region No. 5-1
------- |