EPA-R2-72-065
November 1972 Environmental Protection Technology Series
Full Scale Parallel Activated
Sludge Process Evaluation
Office of Research and Monitoring
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were'established to facilitate further
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a maximum interface in related
fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
U. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series
describes research performed to develop and
demonstrate instrumentation, equipment and
methodology to repair or prevent environmental
degradation from point and non-point sources of
pollution. This work provides the new or improved
technology required for the control and treatment
of pollution sources to meet environmental quality
standards..
-------
EPA-R2-T2-065
November 1972
FULL SCALE PARALLEL ACTIVATED
SLUDGE PROCESS EVALUATION
By
Erwin D. Toerber
Project ITO^O ENM
Project Officer
Richard G. Eilers
Environmental Protection Agency
National Environmental Research Center
Cincinnati, Ohio ^5268
Prepared for
OFFICE OF RESEARCH AND MONITORING
U.S. ENVIRONMENTAL PROTECTION AGENCY
Washington, D.C. 20^60
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $2.75
-------
Environmental Protection Agency
Review Notice
This report has been reviewed by the Environmental
Protection Agency and approved for publication. App-
roval does not signify that the contents necessarily
reflect the views and policies of the Environmental
Protection Agency, nor does mention of trade names
or commercial products constitute endorsement or re-
commendation for use.
.1 1
-------
ABSTRACT
A comparison was made between parallel activated sludge systems
operating under completely-mixed and plug-flow modes.
Initially, a rhodamine dye tracer study was conducted to determine
conditions necessary to achieve the two operational modes.
The completely-mixed system was operated at 5 constant detention
times ranging from 5 hours to 1 hour. The break in treatment
efficiency (a marked drop below 90% removal of soluble BOD )
occurred between 1 and 2 hours.
No marked difference in treatment efficiency was found between
the two modes during 5 months of parallel operation. Two methods
of solids production analysis were applied to this data. A series
of oxygen uptake tests were performed to evaluate air requirements.
Finally, a set of shock loads were applied in parallel and sepa-
rately to each mode. The complete-mix system did show an advantage
over plug-flow under shock load conditions at a short detention
time (1.5 hours).
A 500 gpd completely-mixed pilot plant was run in parallel with
the full scale system for 4 months. It was most successful in
duplicating the full scale organic removal efficiency.
This report was submitted in fulfillment of Project Number 17050 ENM
under the sponsorship of the Environmental Protection Agency by
the Freeport Illinois Water and Sewer Commission, 230 West Stephenson
Street, Freeport, Illinois 61032. The Project Engineer was Erwin
D. Toerber.
iii
-------
CONTENTS
Section Page
I Conclusions 1.
II Recommendations 3
III Introduction 5
IV Aeration Tank Mixing Study 13
Experimental Procedure 15
Types of Analysis Used 24
Experimental Results 29
V Approach to Evaluating Biological 47
Performance
VI Constant Detention Time Study 55
VII Parallel Operation 63
Operating Data and Comparison of
Parallel Performance 63
Possible Effect of Waste
Characteristics 69
Evaluation of Air Requirements 69
Solids Production 87
Final Settling Tank Performance 96
VIII Shock Load Studies 103
IX Pilot Plant Operation 123
Results 130
Discussion 155
X Improved Design Concepts 177
Economical Evaluations 180
EPA Process Performance Simulations 184
-------
CONTENTS continued
Section Page
XI Acknowledgements 187
XII References 189
XIII Appendix 191
VI
-------
FIGURES
Number pa ge
1 Activated Sludge System 6
2 Aerial View of Plant 8
3 Activated Sludge Air System 10
4 Return Sludge System 11
5 Diffuser Configuration for CM 14
6 Diffuser Configuration for PF 16
7 Dye Analysis System 17
8 Dye Analysis System in Field 18
9 Fluorometer Calibration System 19
10 Fluorometer Temperature Corrections 21
11 Fluorometer Calibrations 23
12 Aeration Tank Flow Patterns 25
13 Temporary Baffling in Aeration Tank 2 26
14 n Versus Sewage Flow 32
15 n Versus Air Flow 32
16 c Curve - Run 2 34
17 c Curve - Run 2 35
18 Tracer Curve - Run 2 33
19 Correlation of Theoretical & Measured Detention 33
Time
20 C Curve - Run 8 39
21 Tracer Curve - Run 8 43
22 C Curve - Run 12 4}
23 Tracer Curve - Run 12 43
Vll
-------
FIGURES (continued)
Number Page
24 Activated Sludge System
Operation in Parallel 45
25 BOD Balance 48
26 Oxygen Balance 59
27 Active Mass Balance 53
28 BOD5 & COD Removal (PEUF to FEF) 59
29 BOD& & COD Removal (PEUF to FEUF) 60
30 % of Days Total FE BOD5 did not Exceed 20 mg/1 60
31 Parallel Operation Removal Efficiencies QQ
32 Parallel Operation - BOD5 & COD Data S7
33 Parallel Operation - Solids Data gg
34 Parallel Operation, Volumetric BOD Loading 7Q
35 Constant T, Soluble BQD$ Removal Efficiencies 71
36 Plug-Flow 02 Uptake Data 78
37 Air Flow & DO for CM System 81
38 Air Flow & DO for PF System 82
39 02 Uptake Comparison - CM o/-
40 02 Uptake Comparison - PF 88
41 02 Uptake Profile Along PF Tanks 8g
42 Activated Sludge Solids Balance q0
43 Eckenfelder Evaluation of Solids Production - Qo
CM 9J
44 Eckenfelder Evaluation of Solids Production - nc
PF J5
45 Effect on Effluent Solids of Solids Loading 97
Vlll
-------
FIGU11F.S (continued)
Number Pa^e
46 Effect on Effluent Solids of Hydraulic 98
Loading
47 Effect on Removal Efficiency of Solids
Leading 99
48 Effect on Removal Efficiency of Hydraulic
Loading ^QQ
49 Tracer Tests on Aeration Tank & Final Settling-
Tank 102
50 24 Hour Plant Evaluation
51 BOD5 Data - Shock Load No. 2
52 COD Data - Shock Load No. 2
53 Solids Data - Shock Load No. 2 1Q8
54 Average DO in AT Systems 2 & 4 -
Shock Load No. 2 109
55 DO Versus Time - CM - Shock Load No. 2
56 BOD Data - Shock Load No. 3
57 COD Data - Shock Load No. 3 113
58 Solids Duta - Shock Load No. 3 114
59 Air & Sewage Flow - Shock Load No. 3 -j 15
60 BOD5 Data - Shock Load No. 5 116
61 COD Data - Shock Load No. 5 117
62 Solids Data - Shock Load No. 5 118
63 Air & Sewage Flow - Shock Load No. 5
34 Pilot Plant Schematic Diagram
65 Typical 24 Hour Flow Pattern During
Parallel Evaluation
66 Pilot Plant Oxygen Transfer Efficiency
IX
-------
FIGURES (continued)
Number Page
67 24 Hour Grab Analysis - Primary Effluent ^49
68 24 Hour Grab Analysis - Final Effluent 150
69 24 Hour Evaluation - DO Profile & Air
Supplied 151
:70 Dye Washout Curve, Tracer Test 152
71 Oxygen Uptake, Run No. 9/20/pp 153
72 % BOD Removal Versus Detention Time, Constant
T Operation
73 K4 Versus MLVSS, 5 & 4 Hours Nominal T
74 K4 Versus MLVSS, 3 Hours Nominal T 161
t ,
75 K4 Versus ?,3LVSS, 2 Hours Nominal T 1S2
76 K4 Versus MLVSS, 1 Hour Nominal T 163
77 Removal Rate Constant K
Eckenfelder AoDroach - BODr Basis nRA
169
78 Removal Rate Constant
Eckenfelder Approach - COD Basis
79 DO Profile, September 8-9, 1371
Full Scale Plant
80 Sludge Production*- Pilot Plant
(Eckenfelder Evaluation)
81 Oxygen Uptake Test Apparatus -.^
82 Oxygen Uptake, Run No. 7/1/pp 195
x
-------
TABLES
Number Page
1 Mixing Study Parameters
Configuration I - Complete-Mix 30
2 Mixing Study Parameters - Plu^'-Flow
Configuration II 31
3 Configuration III 31
4 Summary of Constant Detention Time Studies 57-58
5 Removal Efficiencies, Parallel Operation 64
6 Final Effluent BOD5 Evaluation 64
7 Summary of $2 Uptake Data - CM 73
8 Summary of Q£ Uptake Constants - CM 74
9 Active Mass Balance, Operating Data
(Jan. - May 1971) 75
10 Active Mass Balance, Operating Data
(June - Sept. 1971) 76
11 Summary of 02 Uptake Data, PF 79
12 Summary of $2 Transfer Tests 84
13 Comparisons of Solids Prediction Equations 92
14 Comparison of Shock Loads 3 & 5 121
15 Pilot - Full Scale Comparison.
(June 15-30, 1971) 131
16 Pilot - Full Scale Comparison
(June 1-11 & 26-31, 1971) 132
17 Pilot - Full Scale Comparison
(July 12-25, 1971) 133
18 Pilot - Full Scale Comparison
(August 10-19, 1971) 134
19 Constant Detention Time Operation -
Pilot - Full Scale Comparison - 5 Hour
Nominal Detention Time 135
XI
-------
TABLES (continued)
Number Page
20 Constant Detention Time Operation
Pilot - Full Scale Comparison
4 Hour Nominal Detention Time 136
21 Constant Detention Time Operation 137
Pilot - Full Scale Comparison
3 Hour Nominal Detention Time
22 Constant Detention Time Operation
Pilot - Full Scale Comparison
2 Hour Nominal Detention Time 138
23 Constant Detention Time Operation
Pilot - Full Scale Comparison
1 Hour Nominal Detention Time
24 Summary of Constant Detention Time Studies
Pilot Plant
139
140-141
25 Summary of Constant Detention Time Studies
Full Scale Plant 142-143
26 Summary of Oxygen Transfer Tests
27 Summary of Oxygen Uptake Tests 1 _.
28 BOD Removal Rate Constants
.Li)/
29 Comparison of K4 Values for Pilot Plant
and Full Scale
158
30 Computed Values for KQ
^ 167
31 Sludge Production - Eckenfelder's Evaluation
32 Comparison of Predicted and Measured Values
of Sludge Wasted in Full Scale Plant - Active
Mass Production
173
33 Comparison of Predicted & Measiired Values of
Sludge Wasted in Full Scale Plant -
Eckenfelder's Evaluation
175
34 Comparison of Activated Sludge Measurements
and Model Computations
185
XI 1
-------
SECTION I
CONCLUSIONS
Mixing Study
1. By proper placement of diffusers, use of transverse
flow, and maintenance of the stated minimum air flow
per unit tank volume, it is possible to achieve completely
-mixed conditions with a mean residence time very close
to the nominal.
2. To achieve conditions approaching plug-flow an aeration
tank system using spiral aeration and a length:width
ratio of 18.4:1 was found to be adequate.
Constant Detention Time Study
1. For both completely-mixed and plug-flow under normal con-
ditions (F/M = 0.2 - 1.0 Ib BODs/lb MLVSS), no marked
break (a drop below 90% removal of soluble BODg) in
treatment efficiency was experienced until the nominal
aeration, detention time fell below 1-2 hours.
Parallel Operation
1. Under normal loading conditions (T = 3.1 - 3.6 hrs. F/M -
.27 - .82 Ib BOD5 /Ib MLVSS, 25-55 Ib EOD5/1000 ft3), no
appreciable difference could be found between CM and PF
on the basis of BOD or COD removal efficiency from primary
effluent to final effluent.
2. Two methods of solids production analysis were applied
to approximate solids production from these models, how-
ever, a large amount of scatter was experienced in the
solids data.
3. A series of oxygen uptake tests were performed on the
mixed liquor from each system. From these tests it was
possible to determine rate constants to predict oxygen
requirements using an equation of the form do = K_F +
KeMa. "3T
Average values for the constants were determined to be
Ks = 2.6 hrs"1 and Ke = .026 hrs"1.
4. Final settling performance was evaluated and it was
found that to maintain an effluent SS of 25 mg/1,
loadings should not exceed the following:
-------
Overflow Rate = 700 gpd/ft2
(excluding recirculation)
Solids Loading = 25 lb/day/ft2
(based on total mixed liquor flow)
Shock Load Studies
1. Both CM and PF reacted very similarly to an organic shock
load at a detention time of 2.5 hrs.
2. At a reduced detention time of 1.5 hrs the CM system did
show superiority in mass total BOD5 removal (73% for PF
compared to 84% for CM).
3. During all the shock loads the CM system demonstrated a
less marked drop in aeration tank DO than did PF.
Pilot Plant Studies
1. The pilot plant was found to be reliable as a design tool
(on domestic waste) for predicting the following:
a) BOD removal efficiencies and removal rate constants
b) Oxygen uptake and consequently air requirements
2. The reliability of the plant for predicting solids pro-
duction is very questionable.
Design Analysis
1. A set of reliable design parameters were developed from
this study, and an economic comparison of a CM plant de-
signed on this basis versus one designed using state
standards showed a 2 to 1 factor in favor of the former.
2. The cost analysis outlined in No. 1 was also compared
with the EPA economic computer program. The EPA program
yielded a total cost of approximately 80% of that deter-
mined from costs est.imated for a plant designed according
to the parameters developed from this report.
-------
SECTION II
RECOMMENDATIONS
Though extensive mixing studies were conducted, it was not
until the end of this phase that an optimum design for a
CM aeration tank was determined. Additional tracer tests
to further verify the findings would be advisable.
The shock load studies showed some advantage in CM at high
organic loads and short detention times. It would be good
to apply shock loads of longer duration (3-6 hrs) to see
if a greater difference in system performance would result.
Much difficulty was experienced in developing a reliable
solids prediction equation for either system. More research
is needed in relation to sludge flow and concentration meas-
urements and the analysis procedures for such data.
For other researchers looking for small differences in treat-
ment efficiency of biological waste treatment units, it was
found that the sampling method is all important. The most
reliable samples are grab samples taken on a time dependent
basis, e.g., the shock load studies in this report.
-------
SECTION III
INTRODUCTION
Background
The purpose of this study was to compare the performance
of a completely-mixed (CM) activated sludge system with
a, plug flow (PF) system under parallel operating condi-
tions. The original plan also considered a comparison of
CM with contact stabilization and step aeration if sig-
nificant performance differences could be detected under
the existing loading conditions.
The project originated when the City of Freeport, 111.
(population 30,000) Water & Sewer Commission expanded its
wastewater treatment plant to include secondary treatment
of the activated sludge type in 1968-69. The design was to
be based on the principles of CM developed to date. How-
ever, only sufficient capacity as measured by Illinois EPA
standards would be provided for present flow. It was believed
that the true capability of the plant would be adequate to
handle design future and it was desired to establish that
fact before committing additional funds for capacity not
actually required.
The Illinois Sanitary Water Board (now Illinois Environ-
mental Protection Agency) approved of this design approach
on a trial basis providing that adequate flexibility be
included to allow operation under PF and step aeration con-
ditions also.
A schematic of the activated sludge system as built is
shown in Figure 1. The plant was constructed so that it
could be operated as two entirely separate systems using
different flow patterns, mixing regimes, air rates, and
return sludge and wasting rates.
The demonstration grant (17050 ENM) awarded by the Environ-
mental Protection Agency, Office of Water Programs, was
then to provide for Freeport to utilize the design flexi-
bility of the plant for a full scale activated sludge par-
allel process evaluation.
Since much research has been conducted on pilot plant and
lab-scale concerning CM operation, it was decided to in-
corporate into the program a 500 gpd pilot plant operating
simultaneously with the parallel full scale plants.
Finally, on the basis of the data collected, an analysis
-------
l~
Key
[M| — Magnetic Flow Meter
[p] —Parshall Flume
X —Proportional Sampler
Primary Effluent
Return Sludge
— Final Effluent
1 1 Waste
! 1 A
i\l/ ^— .— .
«- — (-.-I >
. V-x 1 1
I
"vl/~ | |
¥._]
-£^J "^i
1 '
1 i___j~|
fa v^ y i
.
1 >
1 |
| Waste
1
DM)
1
1 /
L
— i
\
'
i
\
y
^ j
._l
/
^
f
I
f-
1
\
/
/
^
s
fc'
>
AT 1
^A^
. J — (--
/
ftT 2
1
/^ 1
FST
x_
v.
>
\
A i A'
J_-|_-l--
<
X\
•
i/
AT 3
t * t
-|-4 — (_
AT4
^
/
FS
\
X
1
*
_i_
i
1
T
.
. V'.tp
T
1
t A t
._[_! 1
-x
*^
•
\
4
/
^
L
?
To C 12 Contact
Tanks
/
/
I/I
1
'
•>
ACTIVATED SLUDGE SYSTEM
Figure 1
From Primary Tanks
-------
of plant design and economic factors would be made. The
ultimate objective of the project then being to provide
significant information to improve future plant designs.
To carry out the study, a technical committee was formed.
The committee met on a quarterly basis to evaluate the
progress of the study and to plan the next quarter's work.
The committee consisted of the following persons:
Technical Committee Members
Erwin D. Toerber
Freeport Water & Sewer Commission
Lawrence M. Madden
Water & Sewer Commission
H.S. Smith
University of Idaho
Wayne L. Paulson
University of Iowa
Charles Swanson &
Richard Eilers
Fred VanKirk
Consoer, Townsend & Associates
Robert Bella
Consoer, Townsend & Associates
The study was divided into five
duration are listed below:
Responsibilities
Project Engineer
Project Manager
Research Consultant
Research Consultant
First Mr. Swanson &
then Mr. Eilers serv-
ed as EPA Project Off-
icer
Design of plant & ana-
lysis of data to apply
to design recommenda-
tions
Analysis of data to
apply to design reco-
mmendations
phases. Each phase and its
1. Aeration Tank Mixing Study May-Sept. 1970
2. CM Constant Detention Time Evaluation Jan-May 1971
3. PF and CM Parallel Operation Oct.-Dec. 1970 &
June-Dec. 1971
4. Shock Load Studies July-Dec. 1971
5. Pilot Plant Operation June-Oct. 1971
Plant Layout
Construction of the plant proper was completed in April 1970,
and air and return sludge piping modifications were finished
in July 1970.
The overall plant layout is shown in an aerial view in Figure
2. The activated sludge system consisting of a blower buildint
-------
AERIAL VIEW OF PLANT
Figure 2
-------
4 rectangular aeration tanks, and 4 circular final settling
tanks is in the upper left hand corner of the picture.
Each aeration tank has dimensions of 57.5' x 25' x 13' water
depth (Tank volume excluding effluent channel is 132,000
gal). There are 5 inlet chambers for primary effluent and
4 inlets for return sludge on 1 side of each tank (See
Figure 1). Mixed liquor leaves the CM tanks over a longi-
tudinal weir extending the full length of the tanks on the
side opposite the inlet chambers. Sluice gates are provid-
ed between tanks and on one end to provide for longitudinal
flow. The flow configurations shown in Figure 1 are for
parallel CM and PF operation. Baffling is shown down the
center of aeration tanks 2 & 4. Details of other flow patt-
erns studied are included in the mixing study portion of
the report.
The final settling tanks are peripheral inlet-peripheral
outlet circular tanks with a diameter of 57' and a water
depth of 14' (Tank volume excluding effluent and influent
channels is 270,000 gal.). Sludge withdrawal is by means
of suction through a moving header at the bottom of the
tank.
Air is supplied by a centrifugal blower with a maximum
capacity of 6,000 cfm. Two identical standby blowers were
included in the design. Air is metered separately to each
system through differential pressure flow tubes. A diagram
of the air piping system is shown in Figure 3. Details of
diffuser placement are included in the mixing study phase
of the report.
The return sludge system is shown schematically in Figure
4. The valving is such that return sludge from each pair
of final tanks can be kept separate. Each waste sludge pump
can be valved to pump only from one system. All waste sludge
is cycled back to the primary settling tanks and from there
is pumped to the anaerobic digestion system. Waste and re-
turn sludge flows are measured using magnetic flow meters.
Automatic proportional samples were installed in the flow
stream at three points. These points are marked with an X
in Figure 1. The unit sampling the primary effluent was
placed at the end of a 200' channel to insure adequate time
for mixing of the effluent from the two primary clarifiers.
Similarly, each final effluent sampler was placed in a man-
hole at the end of a 150' long, 36" diameter concrete pipe
just before entry of the liquid into the chlorine contact
system.
-------
-X*
Air Filters
AT
Centrifugal
Blowers
AT 2
ACTIVATED SLUDGE AIR SYSTEM
Figure 3
AT 3
3
>
!(
W
!XJ
f«XX
^i.
Y'
?4
i > '
v^
.
i ^
X? H
A 4
' j
ti ix
q 1 '-KIT'3*' Kj
1
j
^ i
t i
fry-
.i 4
x y
T
fl
FT
*
i k
XX
rT
X)
Y n
X 1
r i
m
( ^
LS I X
AT 4
-------
-------
The samplers were of the bucket type and sampled approx-
imately 35 ml at a time. The samplers were actuated on a
proportional basis by a signal from a final effluent flow
meter tied into a Parshall flume at the outlet of the ch-
lorine contact tank. Tests showed the units to draw a
sample approximately every 80,000 gallons of flow.
Project Limitations
The results reported herein, of course, pertain uniquely
to the Freeport sewage which is a weak domestic-commercial
waste with limited industrial contribution. This can make
differences in performance between CM and PF systems since
the shock load effect at the head end of the PF aeration
tanks was less pronounced than would be the case with
stronger wastes.
12
-------
SECTION IV
AERATION TANK MIXING STUDY
Introduction
Before any parallel comparisons could be made, it was first
necessary to establish what combination of flow pattern,
diffuser placement, and air application rate would yield a
situation approaching completely-mixed on one side versus
plug flow on the other side. To accomplish this, a rhodamine
dye fluorometer tracer system was developed, tested under lab-
oratory conditions, and used to evaluate the flow regimes in
the aeration tanks under varying conditions.
The basic objective of this phase of the study was then to
determine the effect on the mixing regime of varying the
conditions of:
1. Aeration tank configuration(length:width ratio , number
of tanks)
2. Influent and effluent arrangement
3. Diffuser placement
4. Influent flow rate
5. Air flow rate
Figure 3 shows the arrangement of air piping to the aeration
tanks. Air is metered separately to each side with a venturi-
type flow tube utilizing a differential pressure transmitter
to an indicating meter on the main control panel. An addition-
al flow tube with an inclined manometer was inserted to meas-
ure the air split between aeration tanks 2 & 4. The system of
butterfly valves allows for balancing of air and changing of
diffuser header configurations without draining the tanks.
The flexibility features utilized in the mixing study can then
be summarized as follows:
1. Multiple inlet points for primary effluent and return
sludge
2. Side and end effluent weirs
3. Sluice gates between tanks
4. Individually valved air headers with provision for
creating different diffuser configurations
5. Variable output blowers and individual metering of air
to each pair of tanks
The diffusers were of the nylon sock type with a diameter
of 3" and a length of 2'. Normal arrangement on the headers
were pairs of diffusers on opposite sides of the header
spaced one foot apart. Figure 5 shows the diffuser config-
13
-------
T
25'
1
•57'6'
748 Total Diffusers
22 on each end neader
44 on remaining 15
headers
^Effluent Channel
X
374 Total Diffusers
22 on left end header
44 on remaining 8
headers
Typical Single
Diffuser
Header,
\
198 Total Diffusers
22 on left end header
44 on reiuaiiiin. 4
headers
Ty p i3 a. 1 Do u b\e
Diffuser Leader
Typical
Header
with first 7
of diffusers
V
165 Total Diffv
15 on left end
30 oil remRinin
headers
isers
header
pa:; rs
removed'
DIFFUSKil JOKFIGURATION
AEItATICIT TANK MIXING
Figure 5
FO'l J;
STUDY
-------
urations used for the CM portion of the study. Figure 6
outlines similar information for the PF study.
Experimental Procedure - Tracer System
The fluorometer system is depicted schematically in Figure
7 and in a photograrh in Figure 8. The centrifugal pump draws
a continuous sample (at a rate of lOgpm) from the desired
sample point into the de-aeration chamber. Entrained air es-
capes in the chamber and any excess sample overflows. The
diaphragm pump withdraws a continuous sample from the de-
aeration chamber and transfers it to the fluorometer flow
cell at a rate of 150 ml/min. The relative intensity of
fluorescence of the sample is measured by the fluorometer,
and the output is recorded on the strip chart recorder.
The flow system used was found to be necessary to eliminate
extreme variability in output due to entrained air passing
through the flow cell at a high velocity. This would occur
if the centrifugal sample pump were allowed to discharge
directly to the flow cell.
Two methods were used to determine the lag time of the system.
First, dye was used as a tracer. A twenty-gallon tank was
filled with tap water and dye was added to the water and mixed
thoroughly. The intake of the diaphragm pump was placed in
the liquid and the pump was started. The time until an init-
ial reading was obtained on the fluorometer was measured. In
each case this was 30 seconds.
The additional time until a peak reading was obtained was
measured, and this averaged 1.0 minute for 3 runs with dye
concentrations of 50, 60 and 70 micrograms per liter.
Second, the volume of the system, i.e., diaphragm pump,
connecting hoses and the flow cell was computed and found
to be 225 milliliters. At a flow rate of 150 ml/min the
theoretical detention time would be 1.5 minutes. If the
additional volume of the centrifugal pump, de-aeration
chamber and connecting hose is calculated, this is 1.2
gallons. At a flow rate of lOgpm, the theoretical detention
time in this part of the system would be 0.12 minutes. This
was considered negligible; therefore, the lag time of the
system was established at 1.5 minutes.
To calibrate the fluorometer system, the arrangement shown
in Figure 9 was developed. This arrangement was used to
duplicate the field operation of the dye analysis equipment
to the point of the de-aeration chamber.
The ice water bath was used to control the liquid tempera-
15
-------
A.T. 2
A.T. 4
156 Diffusers
102 Diffusers
Baffling,
153 Diffusers .^^ 102 Diffusers
Diffuser Config'urations For PF
AERATION TANK MIXING STUDY
Figure g
16
-------
Aeration
Tank
Diaphragm
Pump
Channel
Centrifugal
Pump
De~Aeration
Chamber
Thermistor
Readout
Fluorometer Recorder
To
Waste
DYE ANALYSIS SYSTEM
Figure 7
-------
DYE ANALYSIS SYSTEM
IN FIELD
FIGURE 8
18
-------
Ice Bath
Diaphragm
Pump
Thermistor
Readout
Recorder
FLUOROMETER CALIBRATION
SYSTEM
Figure g
Water Bath
-------
ture entering the fluorometer flow cell. This temperature
was measured within the cell itself by inserting a ther-
mistor with connecting leads in the outlet hose, so the
thermistor tip stopped inside the cell just below the glass
viewing window.
The procedure for a typical calibration test is outlined
below:
1. Standard solutions of rhodamine B or WT dye are pre-
pared. The standard solution concentration normally
used was 2 mg/1.
2. The hoses, pumps and flow cell are flushed with test
liquid containing no dye. The system is drained. A
volume of test liquid is added to the solution beaker.
This volume is such that when standard rhodamine dye
solution is added, the total volume in the system will
be two liters.
3. The pump is started and liquid is circulated through
the system. The ice bath is used to adjust the temper-
ature. When the desired temperature is reached, the
fluorometer is zeroed.
4. The desired quantity of dye solution is then added to the
solution beaker using a volumetric pipet.
5. When a constant reading is obtained on the fluorometer
at the desired temperature, this relative intensity
reading is used to establish a calibration point for
the dye concentration in the test solution.
6. If points at other temperatures are desired, the ice
bath is used to change the temperature and additional
relative intensity readings are taken.
A series of runs were made to determine the effect of temp-
erature on fluorescence. Figure 10 shows the curve used for
temperature correction in this study. A similar curve as
reported by the U.S. Geological Survey1 and Butts^ of the
Illinois State Water Survey is shown. This latter curve is
developed from the following relationship presented by
Butts:
Ft = Fs en
-------
CO
.2 1.30
0
0>
t 1.24
o
o
-> 1.20
ff
0.96
19
FLUOROMETER
(Rh
EMPERATURE CORRECTION
damine E
gure 10
Freeport
Dye)
uses a
Butts
2O 21 22 23 24
Fluorometer Liquid Temp. (TF) (ac)
25
26
-------
Aminco fluorometer was used in this study.
Laboratory testing of the Freeport system indicated that
the calibration curve was linear in the range of 0-100
micrograms/liter, but that the slope varied for each run
using rhodamine B dye.
To evaluate the variability in calibration during each
field tracer test, samples of primary effluent were
taken periodically as it entered the aeration test tank.
These samples were composited and used as the test li-
quid for a laboratory calibration test as described
above. Figure 11 shows the results of this calibration
evaluation for rhodamine B and WT dye. The lines shown
were drawn using a least squares technique and the re-
spective correlation co efficients (r) show that rhoda-
mine WT demonstrates less variability in calibration.
This is probably due to its characteristic of less ad-
sorption on suspended solids.
For each field test, the experimental procedure was as
follows:
1. The dye analysis system (Figure 7) was positioned near
the effluent weir of the test aeration tank, and the
suction hose of the sample pump was placed just inside
the effluent weir at a water depth of approximately 1.5.'
2. Prior to the start of a run, the air and primary effluent
flow rates were adjusted to constant levels. The air
flow was controlled with a butterfly valve on the blower
suction and measured through a venturitype flow tube in
the air line. Liquid flow was held constant either by
setting a variable speed raw sewage pump or by ad-
justing stop gates into other tanks to by-pass excess
flow. The liquid flow rate was measured via a Parshall
flume located at the effluent of a final settling tank
in series with the test aeration tank.
3. In each case, dye was added as a pulse input at each
liquid inlet point by releasing it by hand from a one
gallon glass bottle. Zero time was recorded as the
moment the dye was added. In computations, the true
zero time was shifted ahead 1.5 minutes to account for
the lag time of the dye analysis system.
4. During the run, air and sewage flows were recorded
every 15-20 minutes, and the temperature of liquid in
the fluorometer flow cell was checked every 5 minutes.
Samples of the primary effluent entering the tank were
collected periodically to be used for calibration
purposes.
22
-------
50
90
80
O
o TO
O
OJ
50
r40
'35
£
•? 30
4>
20
0) 10
(T
Rhbdamine \A
(r-l-0)
20
FLUOROR1ETER CALIBRATIONS
Figure
Rh >damine
11
40 60 80 100
Dye Concentration (ug/l)
120
-------
Three basic aeration tank configurations were tested. They
are outlined in Figure 12 as:
I. Complete-mix consisting of transverse flow with
single or multiple inlet points, all -over-the-
bottom diffuser placement, and a side effluent
weir (lengthrwidth = 0.43:1).
II. Plug-flow consisting of longitudinal flow through
two tanks in series with one pair of end inlet
points, diffusers placed along one side and an
end effluent weir (length:width = 4.6:1).
III. Plug-flow with temporary baffling changing the
length:width ratio. to 18.4:1, side diffuser place-
ment, one pair of inlet points and an end effluent
weir. Flow pattern III was created by installing a
frame-work covered with neoprene coated nylon in
the center of the same tanks used for pattern II
(see Figure 13) .
Types of Analysis Used
The primary approach used to analyze the data is outlined
by Levenspiel^. if a pulse input of dye is added to the
influent stream of a closed system and the concentration
of dye in the effluent stream is measured with respect to
time, a curve can be plotted which relates these two para-
meters. The curve is called a "C" curve and is defined as
follows:
t - time from moment of dye addition
C - dye concentration at time t
C0- initial concentration of dye if it is
evenly distributed throughout the vessel
T - mean residence time of liquid in vessel
9 - t/T
f(9) - C/Co
From a plot of 9 versus f(9), it is possible to calculate
the mean, ^u , and the varience, (J2. The definitions for a
discrete number of points are as follows:
Zf(0)
If a tanks-in-series model is used to describe the mixing
conditions, n values (n refers theoretically to the number
of completely-mixed tanks in series in the system) can be
computed from g"*2, v;kere n - 1
24
-------
Inlet Points
25.0'
f-H
Water Depth = 13'
I I I 1 11 I It
Complete
Mix
I
575* Effluent Weir
Inlet
Points
57.5'
Plug Flow-n
1 1
1 1
25 O1 »-
»•
&
•»
^
^
Sluice
Gates
Effluent
Weir
Inlet
Points
1 1
t 1
Temoorarv
MM
Plug Flow- HI
Bafflina \
— J
Effluent Weir
AERATION
Sluice
Gates
TANK FLOW PATTERNS
Figure 12
25
-------
TEMPORARY BAFFLING
IN AERATION TANK 2
FIGURE 13
-------
An n value 1 indicates ideal complete mixing and an n
value of oo represents ideal plug flow.
However; Timpany^ states that to accurately determine
0 , a test must be extended to at least ten detention
times. This is impractical due to the length of time in-
volved in full scale testing.
Fair, Geyer; and Okun5 define n in the following manner;
9 Q
n-1 r max, where max = 9 at maximum f(9) for a
n unimodel curve
This was the approach used to calculate n.
The following steps were incorporated into a computer
program and used to compute the mean residence time in
the vessel n, and per cent dye recovery.
1. Compute mean residence time -
T=i =
n
EC
2. From T compute 9^ -
9.: = ti
3. From T compute effective tank volume, V -
V = QAvg T
4. Compute Oo -
Co = Dye weight added
, V
5. Compute Ci/Co = f(9) for equal intervals of 9
6. Find 9 max
7. n = 1
T - 9 max
8. Compute per cent dye recovery -
27
-------
n
weight of dye recovered = ICjAtxQovg.
±=l
% recovery - weight recovered x 100%
weight added
Theta (9)max can also be used to determine the disper-
sion parameter, D , used by Murphy and Boyko6.
Levenspiel3 defined D as outlined below:
dc z D d2c
dt d x2 , defines D where c = concentration at
a point x and a time t in a vessel
L - vessel length
u r mean displacement velocity along vessel
length
For complete mixing conditions JD_ approaches infinity, and
UL 4
for plug flow conditions D approaches zero. Timpany4 pre-
UL * 3
sents curves relating D and 9 max. Levenspiel relates the
uL
parameters n and D by the following equation:
uL
2
2 P f D\ / -uL/D\
1/n =
-------
k =
A In C/Cn
( A VT)T
Thirdly, a comparison of the configurations can be made
utilizing a measure of the extent of dispersion. The
Morril Index is defined by Kothandaraman8 as t90/t!0,
where
*10 = time for 10% of tracer to pass sample point
90 = time for 90% of tracer to pass sampling point
Experimental Results
Table 1 summarizes the data from ten tests made during the
summers of 1970 and 1971 using aeration tank configuration
I as depicted in Figure 12. Table 2 shows similar data
taken during the summer of 1970 for configuration II.
Table 3 presents the data from tracer tests for configuration
III conducted during the summer of 1971.
The first phase of the study consisting of runs 1-4 and
7 evaluated mixing with 5 inlet points and a transverse
flow pattern. Air flow, primary effluent and the number of
diffusers were varied.
The values of n for these runs are very close to 1 indicat-
ing complete mixing. The time-to-peak approach was used to
evaluate n. Figure 14 shows the relationship between n and
sewage flow rate for the multiple inlet configuration.
Figure 15 shows a similar relationship between n and air
flow rate for the same runs. The plots indicate a slightly
increasing n (deviation from ideal complete-mixing) for in-
creasing air and liquid flow rates. However, it should be
noted that for each run both flow rates increased or de-
creased simultaneously. It would appear from the trend of
the data, that increasing liquid flows tended to over -
ride increased back-mixing due to air flows and cause the
system to deviate from complete-mixing. A reduction in the
number of diffusers (run 7) did not appreciably change the
n value.
Runs 5,6,14,15 and 16 were also made using on] y 1 inlet chamber
Again, the time to peak dye concentration was very short
yielding a low 9 max and correspondingly low values of n and
D
uL.
There are factors other than n which must be considered to
completely analyze the mixing pattern. One such factor is k.
Table 1 lists k values comoutfid from the slope of the tracer
29
-------
w
o
Run
1
2
3
4
5
6
7
14
15
16
Qs
mgd
4.35
1.37
5.26
4.20
1.06
1.14
1.21
3.04
2.97
0.99
QA
scfm
3700
mo
6940
3740
925
730
1100
710
II3O
710
No. of
Oif-
fusers
748
748
748
748
748
748
374
216
408
216
Inlet
Points
5
5
5
5
I.End
l,Cntr.
5
l,Cntr.
I.Cntr
l,Cntr.
n
1.05
1.07
1.18
1.13
1.02
1.04
1.03
1.01
1.00
1.04
ft
10.5
7.0
2.5
3.5
31.0
16.0
24.0
75.0
00
26.0
K (min~(;
Meos
.0262
.0098
—
.0268
—
—
.0119
.0144
.0156
0088
Theoret.
0.0216
0.0068
—
0.0208
—
0.0060
0.0160
0.0156
0.0052
Det. Time (min)
Meos
38
102
—
37
—
—
84
69
64
114
Theoret.
46
147
38
48
190
172
167
62
64
192
% DYE
Recovery
186
95
93
102
90
77
53
82
80
65
MIXING STUDY PARAMETERS
CONFIGURATION I - COMPLETE-MIX
Table I
-------
MIXING STUDY PARAMETERS
PLUG FLOW
Run
8
9
10
II
12
13
Q9
mgd
1.97
1.02
1.25
3.06
3.02
1.99
QA
scfm
I960
1110
750
830
1000
930
No. of
Diff users
156*102
156 * 102
102
156*102
156*102
1564(02
Inlet
Points
2, End
2, End
1, End
2, End
2, End
2, End
n
3.50
2.98
1.21
10.3
18.9
*
0.14
0.16
2.0
0.056
0.045
Mean
Residence
Time (miri)
121
145
89
112
133
214
Theoretical
Detention
Time (m\n)
193
372
152
124
126
191
% Dye
Recovery
76
42
136
84
106
100
CONFiG.
n
TABLE 2
CONFIG.
m
TABLE 3
-------
n
vs. SEWAGE FLOW
Figure 14
I.ZU
1.16
1.12
1.08
1.04
1 Of)
/
o
/
/
/
/
y
/
/
O
O/
23456
Qs (MGD)
1.20
1.16
1.12
n
1.08
1.04
1.00
0 IOOO
5OOO
QA 3°°?SCFM)
IT vs. AIR FLOW
Figure is
70OO
32
-------
washout curve. In every case but two (14 & 16), the measur-
ed k value was higher than the theoretical k. This indicates
that the pulse input of tracer was washing out faster than
expected if the entire tank volume were being utilized. The
fact that low values of n were obtained and the slope of the
washout curve was a straight line on a semilog plot indicat-
ed that some tank volume was completely-mixed. However, the
higher measured k indicates that this volume is less than
the actual volume.
An example of the discrepancy in k is shown for run 2 in
Figure 16. The plot was made assuming Co equal to the min-
imum C obtained and T equal to the theoretical detention
time. All the data were initially analyzed using these
assumptions.
A computer program procedure was then developed from
Levenspiel^ to evaluate the mean residence time and effec-
tive volume. The initial concentration, Co is then the
concentration of dye obtained if the total weight added is
distributed throughout this effective volume and T is the
mean residence time. Figure 17 shows the same C curve for
run 2 from the computer data. Theta (9) is expanded be-
cause the mean residence time was calculated to be less
than the theoretical (k is the same for each method).
If the mean residence time is computed from the formula,
n
I C± t±
T = i = 1
n
i = 1
and the run is not extended out_several detention times,
the data curve is truncated and T is calculated to be lower
than it actually is (this is the same problem encountered
in using variance to calculate n).
An example of this for run 2 is shown in Figure 18. The
data points stop at 120 minutes. The curve was extended
along the dotted line by calculating C values from the k
established in Figure 16. Applying the mean residence time
formula to the extended curve results in a. T of 89 minutes.
If the curve is extended to 500 minutes, T calculates to
be 104 minutes.
An easier method of establishing T for a complete-mixing
system is to use the definition of k (k = 1/T or T = 1/k).
For run 2, k = .0098 and t = 102 minutes. Using this
33
-------
(Calculated assuming Co = Cmax S T = T Theoretical)
1.01
0.9
0.8
0.7
0.6
0,5
^0.4
'©n
•• 0.3
o
^0.2
O.I
= .019
K=.OD98
.1 .2 .3
006f
.4 .5 .6
t/T = 6
.7 .8 .9 1.0
C- CURVE- RUN 2
CONFIGURATION 1 - COMPLETE MIX
Figure is
34
-------
/Colculated from Computer Program}
1.0 1.2 1.4 1.6 1.8 2,0
C- CURVE- RUN 2
CONFIGURATION I - COMPLETE MIX
Figure n
FT *J-! »'"• '<(•.-' LJJ WJT. "
35
-------
w
a-.
180
160
140
100
c
o
2 80
c
Q)
O
C
/? 6O
IQ 4Q
Q
^
\ t
X
D
So
\
\
v^
X
f90
ho
•^-^
TRA
CON
: 14,0
:ER
FIGUJ
F
^~>
CURV
tATIO
'igure
»
E-R
N I
18
o
t
— *•
UN 2
- Data
- Point
Exfe
semi
0 V!
)0
^ — .
>MPL
Point!
s front
ision
og pl<
~^— -»
ETE
)f
t
E — .
MIX
— ^,
.
"O 20 40 60 80 100 120 140 160 ISO 200 220 240 260 28O
Time (minutes)
-------
method, the measured detention times are recorded in Table
1. Values of k are not recorded for runs 3, 5,and 6 because
the decay curve exhibited 1 or 2 changes in slope.
Figure 19 presents a comparison of theoretical and measur-
ed detention times for all runs except 3,5, and 6. Scatter
in data points would be expected since conditions other
than liquid flow rate were changing from run to run. By
drawing a line of best fit through the points and forcing
it to go through zero, a relationship can be established.
This line indicates that for an given theoretical detention
time (V/Q), the actual time will be 2/3 of this. A possible
explanation would be the creation of a smaller completely-
mixed zone riding over the aeration pattern result from the
all-over-the-bottom diffuser placement.
It was anticipated that configuration II, a longitudinal
flow pattern with a length-to-width ratio of 4.6:1, would
yield a typical plug flow pattern. Diffusers were spaced
along one side for a 6 foot width to provide spiral flow
aeration. However, Table 2 indicates that with normal
ranges of air and sewage flow, n values of 1 to 3 were ob-
tained. This indicates that the actual physical situation
was one of 2 completely-mixed tanks in series.
Figure 20 is an example (run 8) of the type of C curve
obtained from configuration II. Theta (9) max was 0.715
ind the n value was 3.5. Since a logarithmic decay does
not occur initially when the flow deviates from complete-
mixing, it is not possible to compute the detention time on
the basis of k. Instead, it is necessary to use the mean
residence time approach and extend the test data until C
approaches zero. This_was done for run 8 in Figure 21. The
calculation yields a T of 121 minutes.
Since the prime objective of the overall project was to
compare complete-mixing with plug-flow, it was decided to
modify configuration II by adding baffling to obtain
configuration HI shown in Figure 12. This created a long-
itudinal flow pattern with a maximum length-to-width ratio
of 18.4:1. Table 3 shows that this modification was very
successful, since n values of 10.3 and 18.9 were obtained.
Mean residence time calculations show good agreement between
measured and calculated values. Run 11 exhibited a bimodal
curve. This was caused by some leakage through the baffle
as initially constructed. Therefore values for n or D are
not recorded.
Figure 22 shows the C curve for run 12. Theta (9) max has
a value of 0.903 yielding an n of 10.3. The peak concen-
tration coincided almost exactly with the mean residence
37
-------
zoo
180
160
«
£120
P
-160
w
°«0
1
§60
«
40
20
°(
^_
//
X
P
/
x^
jrfect
'/
X
/
x
> 20 40 60 80 M
Theoretical Detent!
K
s\
/
x
-Best
/
x
X
Fit
0- 5
A- 1
/
/
O
Inlet;
Inlet
x
A
1)0 120 140 I6O 180 200
on Time Cm in,)
CORRELATION OF THEORETICAL a
MEASURED DETENTION TIME
COMPLETE MIX
Figure 19
-------
.8
.3
o>
it
o
o
O5
C-CURVE- RUN 8
CONFIGURATION n - PLUG FLOW
Figure 20
r
t
I
1
f
1
0
S
/*
*~\
\
~~~€^
= i7
\
5
*\
\
\
\
\
\
^
\
\
\
NJ
•)
• 2 4 .6 .8 1.0 1.2 1.4 1.6 1.6 2.0 2.2
t/T = 9
39
-------
40
30
20
a
\
tio
"
4.5
40 80 120 160 200 240 280 320 360 400
Time (minutes)
TRACER CURVE - RUN 8
CONFIGURATION H-PLUG FLOW
Figure 21
-------
2.0
C-CURVE-RUN 12
CONFIGURATION HI-PLUG FLOW
Figure 22
i.o
\
= .903
6
o
r .3
5 .2
.10
.08
•Oft
.6
.8
1.0 1.2
1.4 1.6 1.8 2.0
t/T =
-------
time which also agrees quite well with the theoretical
detention time. Configuration III provides a plug-flow
pattern for process performance comparison with the
completely-mixed pattern of configuration I.
The extent of the longitudinal dispersion for each case
can be characterized by the Morril Index, t^Q/tiQ. The
values are shown for each configuration in Figures 10,
13 and 23. The results are summarized as follows:
Run Configuration Morril Index
2 1 14.0
8 II 4.5
12 III 3.8
As would be expected, the greatest dispersion occurred in
run 2 with an index of 14.0 and the least occurred in run
12 with an index of 3.8.
Conclusions and Recommendations
This study indicates that it is not difficult to achieve
complete-mixing conditions in an aeration tank. The main
problem is designing the tank to achieve maximum utili-
zation of the tank volume for treatment. Recommendations
concerning tank design for complete-mixing include the
following items;
1. Inlet points - number & type
2. Effluent weir design
3. Diffuser density
4. Air flow
5. Improvements in diffuser placement
Table 1 shows similar n values for both multiple and single
inlet arrangements. However, two of the single inlet tests
gave washout curves with changing k values. Also runs 14,
15, and 16 exhibited a short duration peak of dye at the
start of each test. Although this amounted to only about
2% of the total dye recovered, it does represent a
degree of short circuiting. With this in mind, a minimum
of 2 inlet points each for primary effluent and return
sludge for tanks of a size similar to Freeport is re-
commended. The inlet points should be located symmetrically
on one side of the tank.
The transverse flow pattern to a long sharp crested eff-
luent weir performed very well. Diffusers should not be
located directly underneath the effluent channel. The air-
water mixture tends to cause periodic uneven flow over the
42
-------
40 80
120 160 200 240 280 320 380
Time ^minutes)
TRACER CURVE-RUN 12
CONFIGURATION HI - PLUG FLOW
Figure 23
-------
weir.
One final diffuser arrangement was tested in October 1971.
The first 7 pairs of diffusers were removed from the inlet
side of each header. This resulted in total of 165 diffu-
sers ( see the bottom diagram in Figure 5). Flow was as in
CM I shown in Figure 12, i.e., 5 inlet points and trans-
verse flow.
The results of the test are as follows:
n = 1.12
k = .0112
T = 1/k =1.5 hrs
( This compares with a nominal T = 1.2 hrs)
% dye recovery - 96%
This data indicates the most effective use of the tank
volume in a CM situation of all the arrangements tested.
44
-------
Ol
ACTIVATED SLUDGE SYSTEM OPERATING IN CM MODE I ON THE LEFT AND PF MODE III
ON THE RIGHT FIGURE 24
-------
SECTION V
APPROACH TO EVALUATING BIOLOGICAL PERFORMANCE
Introduction
To evaluate the performance data obtained from the study,
several approaches were used. The three basic parameters
considered were treatment efficiency, oxygen uptake, and
solids production. Equations relating these parameters
were taken from analyses developed by W.W. Eckenfelder,
Ross McKinney, and H.S. Smith.
Solids production equations were written using Eckenfelder 's
equations and the active mass balance as described by Smith.
Oxve;en uptake constants were developed from oxygen uotake
tests and expressed in an oxygen balance as outlined by
Smith and an uptake equation presented by McKinney.
Removal efficiencies were related to McKinney 's reaction
rate constant.
The above analyses are outlined briefly in the following
sections.
System BOD Balance
Figure 25 shows a schematic of a CM activated sludge system
outlining the balance of ultimate BOD as presented by Smith9.
The terms used are defined as follows:
Q - influent flow rate
rQ - return sludge flow rate
F0 - PE BODL unfiltered
F - FE BODL filtered
Va - aeration tank volume
K4 - removal rate constant (I/time)
t - nominal detention time in aeration tank
By writing a mass BOD balance around the aeration tank the
following equation results.
QF0 + rQF = K4FVa f QF + rQF
F0 = K4FVa + F - F (K4t + 1)
or F - 1
FQ ( K4t
47
-------
influent
Q,F0
GO
Aeration Tank
Volume,
Va
Mixed Liquor
rQ),F
Metabolism
K4FVQ
Return Sludge
rQ.F
BOD BALANCE
Figure 25
Final Settling Tank
Super -
natant
Zone
rQ,F
Sludge
Zone
Effluent
Q,F
-------
The above analysis makes the assumption that ICj is a con-
stant and not time dependent.
System Oxygen Balance
Figure 26 shows a complete oxygen balance for a CM system
as described by Smith^. The terms used in this figure are
defined as:
Q = influent flow rate
rQ = return sludge flow rate
F0 = PE BODL unfiltered
F = FE BODL filtered
Va = aeration tank volume
Ma = active mass in mixed liquor (oxygen units)
Mar r active mass in return sludge (oxygen units)
Ke = endogenous rate constant (I/time)
Ks = synthesis rate constant (I/time)
AMa = net change in Ma of flow (Q t rQ)
Then, by writing an oxygen mass balance around the aeration
tank:
(1) QF0 4 rQ (F 4 Mar) = (Q + rQ) (F 4 Ma) 4 KeMaVa + KsFV
The active mass balance around the aeration tank can also
be expressed by:
(2) Mass In Mass Out
(Q -f- rQ) AMa 4 rQMar = (Q 4 rQ) Ma
or
Equation(l)can be rewritten:
QF0 4 rQF 4 rQMar - (Q+rQ) F - (Q4rQ)Ma = Va (KsMa + KgF)
substituting equation(2)
QF0 4 rQF 4 rQMar - (Q4rQ) F - (Q4rQ) AMa - rQMar = Va(KsMaf-
QF0 4 rQF - QF - rQF - (Q4rQ) A Ma = Va (KsMa + KeF)
dividing by Q
(F0 - F) - (14r) A Ma = t (KsMa 4 KeF) (3)
Equation (3) expresses concentration of food supplied in
terms of oxygen units (left hand term) to concentration
of oxygen used for synthesis and endogenous respiration
(right hand term) .
49
-------
Final Settling Tank
VI
Influent
Q,FO
i >•
Volume,
va
Mixed Liquor
(Q+rQ), (F+Ma)
Auto— oxidation
Super —
notont
Zone
t
keMQVa
Energy Oxidation
ksFVa
Affluent f
Q,F
r
Sludge
Zone
Return Sludge
rQ(F
OXYGEN BALANCE
Figure 26
-------
The rate of oxygen consumed can then be related to equa-
tion (3) by the equation as expressed by McKinneyi(J:
(4) ^o = KsMa + KeF
dt
System Active Mass Balance
To complete the analysis of the CM system a complete flow
diagram for active mass is shown in Figur'e 27. Definition
of terms is as follows:
Q - influent flow rate
rQ - return sludge flow rate
wQ - waste sludge flow rate
F - FE BODL filtered
Va - aeration tank volume
Ma - active mass in mixed liquor
Mar - active mass in return sludge
Mae - active mass in final effluent
Ke - endogenous rate constant
Ks - synthesis rate constant
An active mass balance around the aeration tank yields
the following equation:
(5)^ftMal° + rQ Mar * KsFVa = (Q+rQ) Ma + KeMaVa
Tests show Mai to be zero, so this term can be eliminated.
Equation (5) can be rearranged:
rQMar - (Q4rQ)Ma = KeMaVa - KsFVa
or (QfrQ)Ma - rQMar = KsFVa - KeMaVa
Now, a balance around the final settling tank results in the
equation:
(6) (Q-f-rQ)Ma = QMae + rQMar f wQMar
Substituting (6) into (5) yields:
QMae + rQMar + wQMar - rQMar - Va (KSF - KeMa)
dividing through by Q:
(7) Mae + wMar = t(KsF - KeMa)
Eckenfelder Solids Balance
The following terms are presented by Eckenfelder-'--'- to develop
51
-------
a solids production equation.
AX - sludge production per day
V - volume of aeration basin
Q - flow
a - fraction of substrate converted to new cells
Sr - substrate removed
b - fraction per day of VSS oxidized
Xa - MLVSS
X0 - influent SS
Xe - effluent SS
The above terms are related by the following equation:
(8) AX = QX0 -f aSrQ - (bXaV f QXe)
To determine the constants a and b a plot is made of AX/
Xa versus Sr/Xat. The slope of the curve is a and the y -
intercept is b.
52
-------
in
AT
Q.Mai
rQ,Mar
wQ, Mar
ACTIVE MASS BALANCE
Figure 27
-------
SECTION VI
CONSTANT DETENTION TIME STUDY
To determine the performance of the CM system under a
range of detention times, eight different constant in-
fluent flow rates were applied to aeration tanks 1 & 3.
The flow pattern used for all the periods was transverse
flow with one center inlet point. An adjustable stop
gate was installed in each center inlet point, and the
operators adjusted these gates to maintain a constant
flow as measured by the final settling tank Parshall
flumes. The remainder of the plant flow went through
aeration tanks 2 & 4, so these tanks essentially act-
ed as a buffer system.
The following testing procedures were carried out for
this phase of the study and also continued through the
parallel operation phase.
1. Composite samples of primary and final effluent were
taken on a daily basis using the samplers described
in the introduction section and located as shown in
Figure 1. Samples were collected from the automatic
units and refrigerated every two hours. Lab analyses
were run as indicated below:
COD Filtered & Unfiltered
Filtered & Unfiltered
TSS & VSS
NH3
N03
(Details of laboratory analyses are included in the
appendix)
2. Samples of aeration tank mixed liquor and return sludge
were taken every 8 hours and the following determina-
tions were made:
MLSS & MLVSS (solids procedure described in app-
endix)
RS SS & RSVSS
Aeration Tank D.O. & Temperature
Sludge Settled, 30 Minutes
55
-------
3. 02 uptake tests were performed on samples of mixed
liquor, return sludge, and final effluent periodically
during the testing period (These are discussed in
detail in the parallel operation section of the report).
Table 4 is a summary of the operating data for the CM
system for the 8 constant T periods. An attempt was made
to hold the loading in terms of Ib BODs/lb MLVSS constant
throughout the tests. With this in mind, MLSS values were
increased as the constant influent flow rate increased.
However, this procedure was not entirely successful as
evidenced by the increasing F/M ratios with increasing
flow rates. MLVSS values ranged from 1850-3340 mg/1 with
an average of 2760 mg/1.
The parameters used to evaluate the system performance
included the following:
1. % removal BOD5 PE UF to FE F & PEUF - FE UF
2. % removal COD PE UF to FE F & PEUF - FEUF
3. FE BOD5 values
BODs Removal
The upper graphs in Figures 28 & 29 show the removal per-
centages from primary effluent to final effluent for the
total and soluble 5-day BOD. The most significant point
about this data is that the removal percentages did not
decrease appreciably from the 90% range for soluble and
the 80-90% range for the total until the 1 hour nominal
detention time period.
The soluble data indicates that organic removal in the
aeration tanks is almost constant until the mean residence
time drops below 1 hour. Actually, the mean residence time
was even less than 1 hour according to the results of the
dye tracer tests.
Total BODc began to drop off at around 2.5 hours. This
can be attributed to reduced performance of the final
settling tanks at this flow rate (Refer to section on FST
performance). At 2.5 hours nominal T, hydraulic loading
on FSTs would be 500 gpd/ft .
COD Removal
The COD data shown in Figures 28 & 29 corroborate the
conclusions of the preceding section. COD removals are
56
-------
SUMMARY OF CONSTANT DETENTION TIME STUDIES
Table 4
Time
Period
1/1/71
1/31/71
2/1/71
2/13/71
2/17/71
3/12/71
3/17/71
3/26/71
4/5/71
4/16/71
4/22/71
4/24/71
5/9/71
5/10/71
5/31/71
Qs
mgd
1.30
1.24
1.54
1.59
2.08
2.84
2.85*
3.53^
T
(hrs
4.9
5.1
4.1
4.0
3.0
2.2
1.1
0.9
Primary Effluent
BOD
F
82
77
95
39
33
81
104
58
UF
135
127
135
90
78
122
156
121
COD
F
99
98
10;
66
67
72
72
72
UF
287
245
265
208
187
206
226
204
SS
150
115
127
199
107
126
161
104
VSS
126
92
104
124
80
94
120
79
Final Effluent
BOD
F
14
10
13
7
8
10
14
15
UF
19
16
13
17
13
15
33
35
COD
F
44
40
41
44
39
43
44
49
UF
70
59
51
54
57
60
58
73
SS
27
26
16
18
29
36
27
29
rss
22
25
15
14
23
28
22
25
* To only 1 A.T.
j- Avg. for diurnal flow variation
-------
SUMMARY OF CONSTANT DETENTION TIME STUDIES
Table 4 cont'd
oo
Per Gent Removal
Time
Period
1/7/71
1/31/71
2/1/71
2/12/71
2/13/71
2/17/71
3/12/71
3/17/71
3/26/71
4/5/71
4/6/71
4/22/71
4/24/71
5/9/71
5/10/71
5/31/71
PE-BOD (UF)
to FE BOD (F)
89.6
92.1
90.4
92.2
89.7
91.8
91.0
87.6
PE-COD (UF
to FE COD(
84.7
83.7
84.5
78.8
79.1
79.1
80.5
76.0
SS
?)
82.0
77.4
87.4
91.0
73.0
71.4
83.2
72.1
vss
82.5
72.8
85.6
88.7
71.2
70.2
81.7
68.4
F/M
Ib BODs
Ib MLVSS
0.26
0.19
0.24
.21
.33
.51
1.00
1.14
Mixed
Liquor
Temp. (°C)
14
13
13
11
12
14
15
16
A
B
C
D
E
F
G
H
-------
a
o
a>
70
QG
Soluble COD Removal
60-
50.
4
1 Detention Time (hours)
80D5 a COD REMOVAL-PRI. EFF. UNFILT.
TO FINAL EFF. FILT.
Figure 28
59
-------
IGOr
90
80
S 70^
o
0)
c
0)
o
-------
in general lower than BOD5 removals, but again there is
a gradual decrease in efficiency until 1 hour is reached
for soluble values and a more pronounced decrease in
efficiency with decreasing T for total COD values. It
should be noted, however, that the soluble COD efficiency
decrease around 1 hour is not as marked as the correspond
ing one for
Final Effluent BODs Values
Though per cent removal is a good indicator of system
performance for comparison purposes, most regulatory
agencies are basing evaluation of plant operation on
effluent criteria. The data plotted in Figure 30 gives
a good indication of how the CM system would perform if
an effluent BOD5 value of 20 mg/1 is chosen as a lower
limit. It shows the percent of time that the system
would be in compliance.
On this basis, the break point is between 2 and 3 hours.
This is probably the best criteria to use for purposes of
aeration tank design.
Removal Rate Constant
A common ,me thod of characterizing the removal rate is the
formula:
F = 1 _
F0 K4 t + I , where F= Effluent soluble BOD
F0 = Influent soluble BOD
t = Detention time
K4 = Rate constant
Figure 28 shows 3 removal curves corresponding to K4 = 2,
4, and 6 hrs"1 superimposed on the actual measured data.
The actual curve is very flat past 2 hours detention time,
indicating that the apparent K4 value is not a constant
but changes with T.
As an approximation Kq. = 4 hrs ~1 is the best value for
following the measured data. It can be said that the re-
moval curve is bounded by K4 = 2 hrs"1 and K4 = 6 hrs"1.
Other Analysis of Data
Several other analyses were made on the constant detention
data, but it is directly connected with the parallel opera-
tion phase so it is all presented in that section. The
analyses include interpretations of the 62 uptake data and
development of prediction equations for air requirements
and solids production.
61
-------
•J.
to
w
100
80
6O
40
20
0
B
O
A
012345
Detention Time (hours)
OF DAYS TOTAL FE BODg DID NOT EXCEED 20mg/l
FIGURE 30
-------
SECTION VII
PARALLEL OPERATION
Introduction
From the results of the tracer studies, the flow patterns
shown in Figure 1 were adopted for the full scale PF versus
CM parallel operation. Specifically, PF consisted of 2 end
influent points and a baffled aeration tank system 230' long
by 12.5' wide (L/W = 18.4:1) with a 10 foot end effluent
weir. The CM system consisted of two transverse flow tanks
with 5 inlet points and side effluent weirs. Diffuser
placement was as shown in the lower diagram in Figure 6 for
PF and as shown in the 198 diffuser diagram in Figure 5 for
CM. Return sludge entered at the influent end of the PF
system (See Figure 4) . Four side inlet points were used for
return sludge on the CM side. During certain periods of the
parallel study, there occurred some variations in the above
configurations. Significant changes will be described where
they apply in the discussion.
Parallel operation was conducted from June to November 1971.
For the majority of time the flow was split equally between
the two sides. The basic plan was to compare performance
between the sides with equal loads under different con-
centrations of MLSS. Sample collection was continued as in
the constant T phase of the project. I
Discussion of parallel operation is divided into four parts:
a) Operating data and comparison of parallel performance
b) Oxygen uptake studies
c) Solids production studies
d) Final settling tank performance
Operating Data and Comparison of Parallel Performance
The parallel operation phase was divided into nine periods
designated I-Q in Table 5. Periods were chosen to group
numbers of days together during which MLSS values were
similar and flow patterns were identical for each side
of the plant. Thus, for each period the loading (F/M) was
about the same for CM and PF and the nominal detention
time was the same for each (Refer to Table 5) .
To compare performance the following parameters were chosen:
a) Per cent removal of 5 day soluble BOD from primary
effluent to final effluent, i.e., removal based on the
difference between unfiltered BODs inv/the primary
63
-------
REMOVAL EFFICIENCIES
PARALLEL OPERATIONS
Table 5
Dates
j/9-6/16 &
6/20-6/22
7/1-7/11 &
7/26-7/31
7/12-7/25
S/] -S/Q
8/10-8/22
8/23-9/6
9/7-9/12
9/20-9/30
10/1-10/10
Period
I
J
K
L
M
N
0
P
Q
BOD ilemoval
PEUF-FEF
(%)
CM
84.2
89.4
92.9
92.1
88.9
90.5
90.0
90.9
S9.5
PF
84. J
90.1
90.4
92.3
90.0
89.3
88.6
91.9
88.4
COD Removal
PEUF-FEF
(%)
CM
V8.4
75.5
82.0
79.2
82.3
76.0
77.5
76.9
83.7
PF
79.0
78.7
82.0
79.2
80.5
74.3
78.7
78.7
83.2
T
(hrs)
3.1
3.3
3.2
3.1
3.1
3.2
3.1
3.2
3.6
F/M
Ib. BOD5_
Ib.MLVSS day
CM
0.43
0.82
0.57
0.55
3.44
0.37
0.27
0.31
0.27
PF
0.38
0. 75
0.59
0.54
0.53
0.41
0.28
0.35
0.28
FINAL EFFLUENT BOD5 EVALUATION
PARALLEL OPERATION
Table 6
Period
I
J
K
L
M
N
0
P
Q
Avg. FE
BOD5
CM
27
34
28
16
16
24
15
17
16
PF
22
34
19
13
14
24
22
15
13
Days 20 tag/1
Exceeded
CM
5
9
5
1
3
6
0
1
3
PF
5
11
3
0
0
6
2
1
3
Total
Dfl VS
11
17
14
9
13
15
6
11
10
% time
not exc.
CM
54.5
47.1
64.3
88.9
'76 . 9
60.0
100
90 . 9
70.0
PF
54.5
35.3
•<8.6
100
130
60.0
66.7
90 . 9
70.0
Avg.
MLVS3
CM
1714
1100
1272
1023
1375
1560
1990
2380
2380
PF
1953
1198
i «~ o u:
L*-i «_>D
1045
1151
1510
1916
2060
2240
64
-------
effluent and filtered BOD5 in the final effluent. This
essentially evaluates the performance of the aeration
tank system, but does not indicate solids removal effec-
tiveness in the final settling tank.
b) Similarly, percent removal of soluble COD from PE to
FE
c) Total BODs and COD concentration in each final effluent
for a given applied concentration in the primary
effluent
d) Suspended solids concentrations in each FE
e) Finally, as a measure of the ability of each system to
meet a set effluent criteria, an evaluation was made
of the percent of time during which each system ex-
hibited a final effluent concentration below 20 mg/1
BOD5.
Figure 31 shows the comparisons of removal efficiencies of
soluble BODs and COD. The average for each period is plotted
as a single point at the midpoint of the period. This data
does not indicate a significant difference in removal eff-
iciency between the two systems. Generally, efficiencies
remained around 90% for 6005 and 80% for COD.
The ability of each system to meet a 20 mg/1 effluent
standard is outlined in Table 6. Though PF showed some
superiority during the middle periods CM was better during
some others.
The actual effluent averages are shown in Table 6. On the
basis of an overall weighted average, BODs for PF was 21
mg/1 and for CM was 23 mg/1. The most significant point of
this evaluation is that neither system met the 20 mg/1
criteria 100% of the time.
Average values of BODs and COD for PE and FE are shown
graphically in Figure 32. All FE BODs values are very
similar for CM and PF. During period K the FE COD for PF
did come out considerably higher than CM. Other than this
COD values were also all very similar.
The solids data is outlined in Figure 33. There was a trend
towards higher effluent solids in the PF for periods K and
L.
It is difficult to make any definitive statement concerning
removal efficiency as it relates to organic loading. The F/M
ratios are recorded in Table 5. There is a trend towards
slightly higher removal efficiency with a decrease in F/M
(See periods I,J,0, and Q) . One can not clearly distinquish
between the 2 modes of operation.
65
-------
lOOi
95
90
85
_ 80^
o
Q>
70[
BOD
O - CM
A -PF
COD
_L
J_
J_
0 20 40 60 80 100 120 140
Days
PARALLEL OPERATION
REMOVAL EFFICIENCIES (SOLUBLE)
Figure 31
-------
Q
o
o
250-
200
T>
' 150-
100
50
0
150
O —PE
A—FE-CM
Q-FE-PF
PARALLEL OPERATION - TOTAL BOD5 and COD DATA
Figure 32
-------
80
60
40
c/> 20
en
0
100
80
60
40
20-
0
0]
June
1971
-e-
JBr
S.
-B
Q-.
JB
10
20
30f 4O 50 6q 70 80 90f 100 flOf '20 130
July I Aug. I Sept. I Oct. I
PARALLEL OPERATION - SOLIDS DATA
Figure 33
140
-------
Similarlv, Figure 34 shows a plot of removal versus Ib BOD5
/1000 ff3. Here again, there is no marked difference between
CM and PF.
Finally, Figure 35 is a synopsis of soluble BODs removal
efficiencies for varying detention times for both CM and
PF. This plot is referred to in the design analysis section
of the report.
Possible Effect of Waste Characteristics
It must be recognized that the Freeport sewage is a relative-
ly weak domestic-commercial waste with only limited industri-
al contributions. The total BODs of the aeration tank in-
fluent seldom exceeded 150 mg/1 and often was less than 100
mg/1.
An inherent difference in biological environment in CM and
PF systems is the uniform substrate composition throughout
the CM system compared with the variable concentration in
the PF system. At the head end of thfe PF system the influ-
ent is dispersed in only a small part of the tank volume
into which return sludge is added. Not only is the F/M
ratio high in the head end of the PF system but the return
sludge organisms face another shock situation in as much as
they are coming from the region of the system in which sub-
strate concentration is not only the lowest but of different
character than that presented by the influent.
The effect of these differences in the CM and PF systems
could have been masked or minimized in this study due to the
weak waste. This tends to minimize the environmental differ-
ences posed by the two systems. This would be particularly
true when the systems are operated well within their cap-
abilities. It is expected that the effect of differences
might be more pronounced when the systems are operating
near the limit of their capabilities.
Evaluation of Air Requirements
The following evaluations were made to describe the air
requirements of each system. Each evaluation will be dis-
cussed in detail separately. A discussion of hew the over-
all result? apply to desjgn is tl
-------
1
-------
100
90-
o
c
g>
£80
o>
tr
O-Completely Mixed
A-Plug Flow
70
I
I
0
CONSTANT
2 3
Detention Time — Hours
T, SOLUBLE BODK REMOVAL EFFICIENCIES
Figure
'5
35
-------
02 Uptake Tests on CM System
From December 1970 to June 1971, a series of 02 uptake
tests were run on samples of mixed liquor, return sludge
and final effluent from the CM side of the plant.
Each test consisted of removing a given sample from the
flow, placing it in a sealed container under controlled
temperature conditions and measuring the change in DO
versus time with the use of a probe placed in the con-
tainer (A detailed description of the lab and analytical
procedure are included in the Appendix of the report).
Each test was designed to determine the following para-
meters:
1. Initial Q£ uptake rate
2. Energy respiration constant Ks (hrs"1)
3. Endogenous constant Ke (hrs )
4. Active mass Ma(mg/l)
Table 7 is a summary of the results of these tests(All re-
sults of these tests are reported at a temperature of 20°C).
To utilize this data in an active mass (Ma) balance equation,
averages of the pertinent parameters (Ks,Ke, and Ma/Vss were
determined and are presented in Table 8.
The constants Ks and Ke for use in predicting 02 uptake and
solids production were determined in the following manner:
1. On the basis of the average data in Table 8 the
active mass to volatile suspended solids ratios
were set at Ma/VSS = Mar/VSS =0.26
Mae/VSS =0.40
The endogenous respiration constant Ke was set at,
Ke = .026 hrs'1
2. Average operating data for CM was summarized for
8 constant detention time periods (See Table 9)
and 7 normal diurnal-variation periods (See Table 10)-.
3. For each period, the equation Mae -f wMar =T (KSF -
KeMa) was solved for Ks using Ke = .026 and the
appropriate operating data.
4. The average Ks for the set of periods was then
72
-------
Summary of 02 Uptake Data - Completely Mixed
Table 7
Date
L2. 8/70
L/ll/71
L/ 13/71
L/2-5/V1
V 1-3/7 i
2/1^/71
3/16/V1
4/3/71
1/27/71
V2VVI
L/i3/7'l
L/2,5/71
3/1Y/71
3/23/71
5/S/71
Sample
LOG a t-
.ion
AT 1 ML
.-T 1 ML
AT 1 ML
-'•-T 1 ML
h:'£ I ML
/.T 1 ML
AT 1 ML
AT 1 ML
--.T 1 ML
AT 1 ML
AT I RS
AT 1 RS
AT 1 RS
?ST 1FE
AT 1 RS
Ks
,(hrs~1-
(02Unit!
1.03
-
-
-
1.48
0. V2
3.83
2.08
1.16
1.03
-
-
-
-
-
KS 3
(hrs J0
3)02Unv
.03/2
.0214
.0234
.0082
.0221
.0131
.0132
. 0472
.0233
. 0442
. 0159
. .3204
.0193
.0412
.0228
F
mg-/l
• S
22.0
-
-
-
6.1
13.1
7.9
7.2
23.4
12.1
-
-
1.22
-
1.03
Ma
(mg/1)
(02Units)
594
976
853
2411
L1V8
L240
745
571
990
260
3024
2995
1790
6.4
4605
Ma
(mg/1)
(solids units
418
088
oOS
1698
828
8V4
524
402
697
183
2130
2109
1260
4. 5
3243
vss
(mg/1
1310
2480
2560
2530
3000
2570
2160
1820
3230
980
6840
7130
6500
11
11720
I.!a
VSS
J.3
0.2
3.24
).67
).28
3.34
3.24
3.31
3.22
0.19
0.31
0.30
0.19
J.41
3.25
Total )2 UP-
take @ t -0
[mg/l.hr)
; 44.1
L 41.0
35.0
-
35.0
23.4
15.4
41. 3
56.0
24.0
-
-
-
-
CO
Ke when used in
x 1.42 (1.42
equations will be expressed in solids
is the assumed ratio, Q units: solids
units, i.e.,(Kg above)
units)
-------
Summary of 02 Uptake Constants
Completely-Mixed
Table 8
Type of
Sample
Mixed Liquor
Return Sludge
Final Effluent
No. of
Tests
10
4
1
Avg. Ks
(hrs-1)
02 Units
1.19*
1.12ft
-
Avg. Ke
(hrs.-1)
02 Units
0.026
0.020
0.041
Avg.
Ma/VSS
0.26 +
0.26
0.41
* Average of 7 Tests
f Omitting 1 High Value of 0.67
f Average of 2 Tests
-------
ACTIVE MASS BALANCE
OPERATING DATA
January - May 1971
Table 9
Period
Jan. 1-31
Pefo. 1-12
Feb. 13-1 7
.Tar. 12-17
far.2^r-&
\or. -i—
\br.22
5-pr. 24-
Kay 3
ilay 10
lay 31
FE-
vss
(mg/:
22
25
15
14
23
28
22
25
Mae
cmg/i;
)
12.5
14.2
8.5
8.0
13.1
15.9
12.5
14.2
ML
VSS
(ing-/!)
2500
3080
3340
2560
1850
2560
3310
2840
Ma
(rag/
923
L13S
L232
946
683
945
L220
L050
RS
VSS
1) (mg/D
7550
8470
9230
7780
7060
9530
.2750
.2890
Mar
(mg/|
2780
3120
3420
2870
2610
3520
4710
4760
w
.01044
. 00591
.00769
.02086
.00693
.00548
.00520
.00462
t
>rs)
4.9
5.1
4.1
4.0
3.0
2.2
1.1
0.9
FE-
BOD5
(mg/1)
14
10
13
7
8
10
14
15
F
Ung/1)
19.4
13.9
18.1
9. 7
11.1
13.9
19.4
20.8
(wMar)
(mg/1)
29.02
18.44
2-5.30
59.87
18.17
19 29
24.49
21.99
Ks
hrs"1
1.67
2.58
2.24
4.28
2.54
2 92
3.37
3.25
Period
No.
A
B
C
D
£
F
G
H
01
-------
ACTIVE MASS BALANCE
Ooerating Data
June - September 1971
Table 10
Period
o/l *-6/lo .p-.
6/23-8/22
7/1-7/11 &
7/23-7/31
7/12-7/25
3/1-V^
8/1-J-8/22
3/23-9/5
9/ 7- 9/1 2
FE-
VSS
Ong/1)
17
18
10
21
23
19
20
ML
vss
(mg/1)
1710
1100
1270
1 32 0
1380
1550
1990
Ma s
fag/1
631
406
469
377
509
576
735
RS
VSS
(mg/tf
7320
3460
4350
3330
5160
6290
9820
Mar
rag/11
2700
1280
1610
1410
1910
2320
3620
v/
. 00862
.01342
.01298
.01477
.01510
. 00886
.0)763
t
hrs^
3.1
3.3
3.2
3.1
3 . 0
3.2
3.1
FE-
BOD =
(mg/D |
15
13
7
5
9
12
7
F
fag/3
22
19
10
9
13
8
10
wMa r
(mg/3)
23.27
17. 18
20 . 90
20.83
28.84
20. 56
27.62
Ks
(hrs
1.23
0. 99
2 . 06
2. 14
2.09
3.10
3.15
Period
. No.
•> (
9
10
11
12
13
14
15
Mae
mg/i:
10
10
8
12
13
11
11
0")
-------
used as the established synthesis constant
On the basis of the above procedure, the average Ks was
determined to be 2.6 hrs"1 for the constant detention time
oeriods.
The following then is a summary of the various constants
that were established for the CM system which are then used
for further evaluations. The ultimate BOD to 5 day BOD ratio
was determined from a series of rate constant tests. A
better approximation was later determined to be 1.46 rather
than 1.4, but 1.4 was used in the preceding calculations.
Ks = 2.6 hrs"1 (02 Units)
Ke = .026 hrs"1 (02 Units)
BODL = 1.4 x BOD5
Ma/Vss = Mar/Vss =0.26
Mae/Vss =0.40
02 Uptake Tests on Plug Flow System
To compare oxygen requirements between the two systems a
series of 02 uptake tests were also run on the plug-flow
side. Since conditions varied along the system, it was
necessary to sample at a number of points to adequately
describe the situation. Table 11 presents the data obtained.
A key to the sample locations is shown at the bottom of
Table 11. Figure 36 shows the relation of the uptake rates
to the sample locations along the length of the baffled PF
system.
Two basic things should affect the 02 demand at any point
in the tanks; BOD concentration and solids concentration.
The average MLVSS for all of the runs shown in Table 11
(except 10/4/71) was 1150 mg/1, so it can be said that soli-
ds were relatively constant. The average ultimate BOD re-
moved for each period is shown in Figure 36 (BODL = 1.46 x
BOD5).
The actual uptake rate should always be something less than
the total ultimate BOD removed as indicated by equation (3),
(F0-F) - (l-fr)AMa - t (KsMa + KeF) ,
developed in Section V. The measured data bears this out. For
example, take the period 7/12 - 7/15:
77
-------
80r-
70-
60-
Avg
O 7/12-7/15/71 132
A 7/26-7/29/71 163
03 8/2 -8/4/71 98
X 8/16-8/19/71 110
10/4/71 182
Removed
50-
oo
40
o>
e
CD
O
Q.
30
20
01
o
10
0
I
I
I
I
20 40
60 80 100 120
Tank Length ( ft )
PLUG FLOW 02 UPTAKE DATA
Figure 36
140 160 180 200 220
-------
Summary of 02 Uptake Data
Plug-Flow
Table 11
Date
7/12/7]
7/13/71
7/14/7]
7/15/7]
7/26/7]
7/27/7]
7/28/7]
7/29/7]
8/2.71
8/3/71
8/4/71
8/16/71
8/17/7]
8/18/71
8/19/71
10/4/71
lO/'Vi
Sample
Locatior
2
3
5
6
2
3
5
6
1
4
7
2
3
5
6
1
1 7
Ks _
(hrs
0.83
1.61
1.39
1.03
1.03
0.83
1.03
1.03
0.83
1.48
0.69
0.92
1.86
1.66
0.59
0.59
1.03
1) *»
(hrs-D
.0311
.0810
.0438
.0252
.0261
.0228
.0265
.0236
.0223
.0491
.0545
.0308
.0399
.0356
.0344
.0393
.0298
F
(mg/i;
21.7
5.0
2.2
3.4
14.6
7.7
3.9
4.5
13.2
2.2
6.1
11.0
3.1
4.2
6.7
34.0
14.0
Ma
:o2
Units
386
185
183
357
383
452
340
314
448
200
143
318
221
309
305
509
554
Ma
(solid
units
)
272
130
129
252
270
318
239
221
316
141
101
224
155
218
215
358
390
VSS
3
tmg/1
1380
1200
710
640
1250
910
1170
870
1180
1050
670
1210
1080
1290
1370
2400
2410
Ma
VSS
)
0.2
0.1
0.1
0.3
0.2
0.3
0..2
0.2
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
Total
3ur=akoe
(mg/l)
) 30
L 23
5 11
)12
I 25
> 17
) 13
5 12
7 21
5 13
5 12
I 20
4 14
18
3 14
5 40
3 31
Key to Sample Locations
At 2
T—
7
1
2
6 ^
At 4
79
-------
F -F 132 mg/1 x 1.71 MOD = 35.5 mg/1
°—=.264 MG x 24 hr/day hr as compared with
a maximum measured rate for that period of 30 mg/1
hr '
Air Flow versus D.O. Tests
To obtain some information about 02 demand from actual full
scale operating conditions, probes were placed in the aera-
tion tanks for 24 hour periods and an attempt was made to
hold the DO constant by varying the air flow.
Figures 37 and 38 show the results of one such test for
CM and PF. It was not possible to hold the DO constant,
but the average DO for each side correlated quite closely
(3.4 mg/1 for CM and 3.6 mg/1 for PF) . During the test
equal sewage flow of equal strength was applied to each side.
Average mixed liquor temperature was 21°C.
To compare the two sides it is necessary to know the trans-
fer efficiency of each system. The next section discusses
the tests performed to determine this. The transfer effic-
iency data has 2 limitations. First, only 1 test was per-
formed on each side and this was at approximately 500 scfm.
As air flow increases, efficiency normally decreases. Sec-
ond, due to slow deoxygenation during the PF test, that
result is questionable.
Some analysis, however, can be made. Four areas should be
considered:
1. Air applied and transfer efficiency
2. Ultimate BOD removal
3. Solids or active mass concentration
4. Application of established reaction rate constants to
predicting 02 required.
Referring to Figure 37, 1,967,000 scf of air was applied
from 8 am 9/9/71 to 8 am 9/10/71. Assuming 10% efficiency
(determined for CM at 500 scfm, the following calculations
can be made:
1,967,000 scf x .10 x .0174 Ib 0?, x 1 _
day scf
,264mg x 8.33 Ib
gal x 24 hrs
day
-64.8 mg/1 (@ 20°C)
hr
80
-------
2500
2000
1500
"1000
o
500-
0
O
a
Q ©1
(39
000
Lrt,
1,967,000 scf
O-scfm ATs 183
-© e 0-
Air Flow 8 DO For cm System
FIGURE 37
8am 10am 12am 2pm 4pm 6pm 8pm 10pm 12pm 2am 4am 6am Sam
Time Of Day
9/9/71
9/10/71
-------
00
to
2500
2000
£ 1500
o
u>
JlOOO
u.
500
0
8
o>
E
O
Q
!—1
e—© ©-
0- scfm ATs2S4
2,135,000 scf
Air Flow a DO For Plug-Flow System Figure 36
8am 10am 12am 2pm 4pm 6pm 8pm 10pm 12pm 2am 4am 6am
Time Of Day
9/9/71
9/IO/71
-------
BOD removal was the same for both systems on this partic-
ular day and the average MLVSS compared very closely (2140
mg/1 for CM and 2020 mg/1 for PF).
BODs values: unfilt. PE filt. FE.CM filt. FE.PF removed
88 me,/iIfl 3j5 7TT
The maximum 02 rate based on the ultimate BOD removed is
equivalent to the following for each system:
.5 x 4.615 MOD x 78 mg/1 x 1.46 x 8.33 Ib = 41.5 mg/1
.264MG x 833 rb x 24 hr/day gal "n7
gal
The following equation is applicable to the CM system:
do = 2.6F + .037 Ma , Ma = MLVSS x .26 = 2140 x .26 = 556
dt
do - 2.6 x 10 4 .037 x 556 = 46.6 mg/1 (@ 20°C)
dt "Kr
This indicates that the rate calculated from the transfer
efficiency of 10% is probably a little high. An efficiency
of 7% would yield 45.4 mg/1 . An efficiency of 8% would
give 51.8 mg/1 .
Now referring to Figure 38, the total air applied to the
PF system for 24 hours was 2,135,000 scf. The transfer
efficiency measured in the field for PF was 3.5%. This
would yield a rate of 24.6 mg/1 . If you assume air
requirements are equal in each system, then an estimate
of the transfer efficiency in the PF can be made by taking
the ratio of the air applied to each times the efficiency
in the CM:
1,967,000 x 8% = 7.3% or .921 x 7% = 6.4%
2,135,000
02 Transfer Efficiency Tests
A series of tests were conducted on the full scale and
pilot plant aeration tanks using the sodium sulfite re-
oxygenation method of measuring Q^ transfer.' 4he results
are shown in Table 12.
ex values were determined from lab scale tests, and (3
values v/ere obtained by aerating the various test liquids
to saturation, i.e., constant DO.
83
-------
Summary of 02 Transfer Tests
Table 12
Test
Conditions
Full Scale
a) AT 1, 748
Diffusers
Pri. Effluent
b) At 1,242 Dif
Prim. Effluent
c)AT 2 156 Dif.
Pri. Effluent
d)AT 4,102 Dif.
Prim. Effl.
Pilot Plant
a) Tap Water
b) Tap Water
c) Tap Water
d) Final Eff.
e) Final Eff.
f) Final Eff.
Lab. Scale
a) 1 liter beak
diffuser stone
Final Eff.
b) Same as a)
But Pri. Eff.
Air
Applie<
scfra
532
531
500
480
.051
.136
.260
.060
.136
.260
00565 .
00565
1
scfm
Diff.
0.71
2.19
3.21
4.71
.051
.136
.260
.060
.136
.260
00565
OOfSfi?;
scfm ,
lOOO-f*""
30.2
30.2
28.3
27.2
15.3
40.7
77.8
18.0
40.7
77.8
160
ifin
KLB tg .
T ZIP. C,
(min ~1
.0823
.0701
.0224
.0190
.0803
.202
.314
.0638
.119
.167
.0423
.0360
Trans.
Eff. ,
>@ 0 f> C
20° c>
10.9%
8.9
3.0
2.7
18.8
17.8
14.6
13.0
10.3
7.6
1.19
1.01
Pro jectec
Eff. in
Mixed
Liquor
12.8%*
10.4%*
3.5*
3.2*
10.7+
10. if
8.3f
13.0
10.3
7.6
_
* based on an
- KFe
. 1.17
based on an average o( - KFe
KH20
= 0.57
-------
Only one major problem was experienced, and this occurred
in aeration tanks 2 & 4. Dispersion of the sodium sulfite
was slow, and it appeared that insufficient cobalt catalyst
was present because it took over 30 minutes for the DO to
drop to the 1-2 mg/1 range. It did not go all the way to zero.
Probably what happened during the re-oxygenation was contin-
ued reaction of some sulfite tending to depress the DO and
reduce the rate of DO increase yielding a lower measured
transfer rate.
Parallel 02 Uptake Tests
Finally an attempt was made to compare measured 02 uptakes
from both systems at the same point in time under similar
loading conditions. The method already described was used
to measure 02 uptake rate except that each run in the seal-
ed flask was only continued for about 1 hour instead of the
usual 24.
During a period from 11 am to 4 pm on November 16, 1971,
the following procedures were carried out:
1. A sample was drawn from the east end of aeration tank 1
(CM) at about 11 am, 1 pm and 2:30 pm.
2. At the same time another sample was drawn from the inlet
area, the half way point, and the outlet area of aera-
tion tanks 2 & 4, the PF system.
Figure 39 shows the results of the three tests conducted
on the CM system. The procedure (outlined in detail in the
appendix) for determining the initial oxygen uptake rate
consisted of determining the slope of the endogenous por-
tion (flat portion) of the curve and subtracting this from
the entire curve to obtain the slope of the synthesis curve.
The two resulting straight lines were then extrapolated
back to t = 0 to find the respective contributions to oxygen
demand (KeMa and KSF) existing in the mixed liquor sample.
This analysis yielded an oxygen uptake rate of 74 mg/l-hr.
This can be compared with the predicted rate using the equa-
tion do = KSF t KeMa.
dt
If this equation is applied to this situation, it would
predict the following for the CM system:
Available Data
Average MLVSS = 3600 mg/1
Average FE BODr = 19 mg/1 ( average of November 14,
15,17, and 18; no data
available for November
16)
85
-------
02 UPTAKE COMPARISON-CM
(Test period 11 am - 4 pm, 11/16/71)
Figure 39
Endogenous -?
X
0- CM 1
A- CM 2
d- CM 3
Average CM - do = 41
dt
33 - 74 mg/l«hr
Synthesis
_L
_L
J_
_L
_L
_L
10
20
30 40 50
Time (min)
60 70 80 9O
100
-------
Ma = .26 x MLVSS = .26 x 3600 = 936
Then:
do - 2.6 x 19 f .037 x 936 = 19 -t- 35 = 84 mg/l-hr
dt
This value of 84 mg/l-hr is in reasonably good agreement
with the measured value of 74 mg/l'hr.
It is of interest from a design standpoint to relate the
above uptake rate to the conventional design parameter
scf air/lb BODs removed. This can be calculated as
follows: assuming an uptake rate of 74 mg/l-hr
74 mg x .264 MG x 8.33 Ib x 24 hr = 3910 Ib 02/day
1-hr gal day
3910 lb/day = 2,809,000 scf/day
.08 x .0174 Ib 02/scf
2 809,000 scf/day = 1060 scf air/lb
127 mg BODs removed x 2.5 MOD x 8.33
1
BODs removed. This value of approximately 1000 scf/lb will
be used in the design analysis section of the report.
Figures 40 and 41 show for PF the uptake analysis describ-
ed previously. Uptake rates ranged from 227 mg/l-hr at the
head end of the PF system to 48 mg/l-hr at the effluent.
These values are shown in a plot uptake versus tank length
in Figure 42. The average rate was calculated by dividing
the Figure 42 plot into 23 length increments, finding the
do at the midpoint of each increment and figuring the aver-
dt
age of these rates. This yielded a value of 107 mg/l-hr.
This value could change appreciably depending on what
occurs between the three points sampled.
Solids Production
Solids wasting data was, of course, collected throughout
the period of the study. For the purpose of the analysis,
data was considered for the period January 1, 1971 to
October 1, 1971. Data covers both constant and variable
flow conditions.
One approach to developing a prediction equation for sol-
ids production involved use of the active mass balance.
This utilizes the equation already developed in Section V:
Mae +wMar = T(KsF-KeMa)
87
-------
800
600-
400-
300-
200-
170
02 UPTAKE COMPARISON - PLUG FLOW
(Test Period 11 am - 4 pra, 11/16/71)
Figure 40
PFI - Influent end of PF system
PF2 - Mdpt of PF system
PFI - do = 170 4 47 = 227 mg/l-hr
"
l\
PF2 - do = 62 -f 32 = 94 mg/l-hr
"3T
0
10
20
30 40 50
Time (min)
60 70 80 90 100
88
-------
100-
80-
60-
40-
30-
PF 3
02 UPTAKE COMPARISON - PLUG FLOW
(Test Period 11 am - 4 pm 11/16/71)
PF3 - Effluent End of PF system
PF3 - do r 25 -f 23 = 48 mg/l-hr
"3T
Figure 41
I
I
20 30 40 50
Time (min)
60
70 80 90
100
89
-------
400
30O
200
W)
S
•oro 100
o
02 UPTAKE PROFILE ALONG PF TANKS
Figure 42
20 40 60 80 100 120 140 160 180 200 220 240
Total Tank Length (230')
-------
To evaluate this approach the following calculations were
made:
First, by considering the equation,
Mae •*• wMar = T(KsF - KeMa) , and using the constants estab-
lished from Tables 7 - 10, it is possible to solve for the
fraction of incoming flow wasted, w:
w = T(KSF - KeMa) - Maf>
Mar
The mass of VSS wasted per day (AX ) is then given by:
AX = (QwMar) 8.33 Ib/gal = QwMar x 22.6 Ib VSS
0726 x 1.42 ~~
where Q = influent flow in MGD
Mar = active mass in return
sludge stream (mg/1, 02 units)
If you define A Ma as the net change in active mass concen-
tration of the entire flow (Q +• rQ) as it moves through the
aeration tank, the following equation can be written:
A Ma (Q + rQ) = KsFVa - KeMaV
In other words the left hand term in the above equation is
equivalent to the active mass produced and can be related
to the mass of VSS produced by the Ma/VSS ratio of 0.26.
The equation can be rewritten:
A Ma (1 + r) = T(KSF - KeMa)
Since the right hand term has already been calculated and
the recirculation ratio r is known for each period, A Ma
can be calculated from:
AMa = T(KSF - KeMa)
(1 + r)
If AX' is defined as the mass of VSS produced per day, then
it can be calculated as follows:
MGD mg/1, 02 units
AX' r (Q -f rQ) x AMa x 1 solid unit
1.42 02 units x
91
-------
Period
A
B
C
D
E
F
G
H
I
J
K
L
M
N
0
Pounds of Volatile Suspended Solids per Day
Active Mass Balance
AX
3460
546
2980
-
950
757
522
790
5340
5174
1725
1390
2290
340
474
AX'
3819
944
2140
88
1562
1779
1326
1919
5803
5614
1998
1948
2903
832
979
Eckenf elder
Approach
AX
940
680
1100
560
900
2420
3060
2590
940
1650
1230
990
940
890
660
Measured
AX
854
690
915
2583
853
1239
1579
1756
1067
750
944
961
1351
928
1267
(AX = mass wasted per day, AX1 = mass produced per day)
COMPARISONS OF SOLIDS PREDICTION EQUATIONS FOR COMPLETE MIX
OPERATION
Table 13
92
-------
CO
CO
.7
.6-
.5-
O
T3
.2
.1
>
a
<
0
-O.I
-0.2
-0.3
ECKENFELDER EVALUATION OF
SOLIDS PRODUCTION - CM
Figure 43
D
0
_L
I
I
I
AX n Sr
a +b
o= 0.8
b=-OJ
H
02 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 I.I I.
Sr (days-*)
Xa t
-------
1 Ib VSS 8.33 Ib = Ib. VSS produced
.26 Ib Ma gal. day
Table 13 is a summary of the results of using the method
of solids production prediction discussed above. Table 13
also includes the results of analyzing the data using the
Eckenfelder approach outlined in Section V.
Figures 43 and 44 show plots of AX/XaV versus Sr/Xat for
CM and PF. The parameters are defined in detail below:
AX = Ib VSS wasted/day
Xa = MLVSS (mg/1) ,
V = tank volume millions of gallons
XaV = Ibs of MLVSS = MLVSS (mg/1) x V(MG) x 8.33(lb/gal)
Sr = ultimate BOD removed = PE unfiltered BODL - FE
unfiltered BODL
t - detention time (days) = Q/V
The lines drawn through the points in Figures 43 and 44 are
merely to indicate the trend. They are not the result of
statistical analysis.
The predicted values of solids wasted under "Eckenfelder
Approach" in Table 13 were determined by using the Sr/Xat
value for each period to find the corresponding AX/XaV
from the trend line in Figure 43. Then by multiplying this
by the known XaV the AX is calculated and reported in
Table 13.
j
Both methods of predicting solids production show consid-
erable scatter. This is one area that still needs further
investigation. Better mathematical models are required that
take into account the fate of all influent solids and con-
sider solids lost in the final settling tank effluent. Also,
improvements are needed in laboratory solids determination
and measurement of waste sludge streams.
94
-------
CD
cn
u>
=*
o
T)
.6
.5-
.4
.3-
.2
-0.1
-.
-0.3
o.i
ECKEIMFELDER EVALUATION OF
SOLIDS PRODUCTION - PF
Figure 44
ol
l I
0.7 0.8 0.9 1.0
-------
Final Settling Tank Performance
Though the primary purpose of the project was the eval-
uation of aeration tank performance under varying con-
ditions, considerable data was also gathered concerning
performance of the final settling tank (FST) system. The
data arose from three basic sources:
a) The CM constant detention time phase of operation
b) Two periods during September and November 1971
when above-normal loads were applied to the
settling tanks
c) A tracer test on a final tank conducted in con-
junction with an aeration tank tracer test
CM Constant Detention Time Period
During this period the flow rate to the CM side utilizing
two settling tanks ranged from 1.3 MGD to 3.5 MOD resulting
in loadings to the clarifiers as shown below:
surface settling rate 200-700 gpd/ft2
solids loading 10-30 Ib/ft2-day
To describe FST performance, four treatment parameters
(final effluent TSS and VSS, and percent removal of TSS and
BODc) were plotted against surface settling rate and solids
loading. The data for the constant T,CM period are shown in
Figures 45,46,47 and 48 as triangular points.
Above Normal FST Loadings
For a period of time in September and again in November 1971,
flow to 2 of the 4 FSTs was cut off resulting in increased
loadings on the remaining two tanks. The loading ranges are
shown below;
surface settling rate 600-1400 gpd/ft2
solids loading 20-55 Ib/ft2day
The data for these two periods are shown as circular points
in Figures 45 through 48.
FST Tracer Test
On October 22, 1971 a tracer test was conducted using the
method described in the tracer study phase of this report.
An additional set of grab samples were taken from the eff-
96
-------
50
*••
c
0>
£40
UJ
£30
x 20
o>
O
0
60
50-
-2 20-
o>
E
O- Daily Values Sept. & Nov. 1971
&—Period Avgs. Jan.—May 1971
A
10-
0
10 20 30 40
Solids Loading (Ib TSS/ft2 day)
EFFECT ON EFFLUENT SOLIDS OF
SOLIDS LOADING
Figure 45
50
97
-------
50
c 40
CD
LU
30
20
~ 10
CO
CO
0
60
50
£ 40
LU
CO
LL.
en
E
CO
CO
30
20
10
0
A
A A
A
A
0 -Daily Values Sept.a Nov. 1971
A -Period Avgs. Jan.—May 1971
200 400 600 800 1000 I2OO 1400
Surface Settling Rate (gpd/ft2)
EFFECT ON EFFLUENT SOLIDS OF
HYDRAULIC LOADING
Figure 46
98
-------
100
90-
tn
a
§ 80
I
1 70
o
60
50
100
O-Daily Values Sept. a Nov. 1971
^-Period Avgs Jon-May 1971
CO
CO
i
8*
90
80
70
60
50
_L
-t-
_L
10 20 30 40 50
Solids Loading (IbTSS/ ft2 day )
EFFECT ON REMOVAL EFFICIENCY OF
SOLIDS LOADING
Figure 47
99
-------
100
90
in
Q
m 80
j_
"5
| 70
tr
* 60
CO
CO
a
o
i
on
50
100
90
80
70
60
50
o- Daily values-Sept.& Nov.
A- Period avgs-Jan-May 1971
O
200 400 600 800 1000 1200 1400 1600
Surface Settling Rate (gpd/ ft2)
EFFECT ON REMOVAL EFFICIENCY
OF HYDRAULIC LOADING
Figure 48
100
-------
luent of FST No. 1 at a point 180° opposed to the inlet
channel. The samples were analyzed for dye concentration
using the fluorometer and the results are shown in Figure
49.
The first curve depicts the change in dye concentration
in the aeration tank effluent. It also represents the in-
put of dye to the FST. The second curve shows the change
in dye concentration in the FST effluent. Since the input
to the tank was not an ideal pulse, it is not possible to
apply the usual tracer mathematics to this situation. How-
ever, by comparing the time between the peak concentration
(inlet to outlet), it is possible to show the general tim-
ing of liquid flow through the system. The time between
peaks was 79 minutes or about 1.33 hrs. (It should be empha-
sized at this point that the discussion is concerned only
with the liquid portion of the flow and not the sludge
portion). During this test the flow rate to the tank was
2.55 mgd. For a tank volume of 270,000 gallons, this yields
a nominal detention time of 2.54 hours or about twice the
peak-to-peak time measured above.
Discussion of FST Performance
Probably the best criterion for activated sludge FST per-
formance is the actual solids concentration in the final
effluent. Primary effluent to final effluent percent removal
is not as meaningful because you are really considering
removal of solids generated in the aeration tanks more than
removal of solids introduced by the primary stream. Figures
45 and 46 show the effects of varying solids and hydraulic
loadings on final effluent solids concentrations.
If a maximum allowable FE TSS level of 25 mg/1 is chosen^
these figures indicate a maximum solids load of 25 lb/ft^
day and a maximum surface settling rate of 700 gpd/ft2.
These loadings are recommended with the realization that
there is interaction between solids and hydraulic loading
which may mask the effect on one or the other.
Figures 47 and 48 show the effect of solids and hydraulic
loading on removal efficiency of BODs and TSS across the
complete activated sludge system. As has been stated, these
relationships depend also on the performance of the aeration
tanks and are not as representative of the FST efficiency
alone. They do show a definite trend towards improved eff-
iciency with reduced loading. The recommended surface
settling rate of 700 gpd/ft2 predicts a BOD5 removal eff-
iciency of 83% and a TSS efficiency of 75%. The recommended
solids loading of 25 lb/ft2 day predicts BODs and TSS re-
moval efficiencies of 84% and 77% respectively.
101
-------
o
to
Aeration Tank Effluent
Final Settling Tank Effluent
20 40 60 80 100 120 I4O 160 180 20O 220 240 260 28O 30O
Time — Minutes
TRACER TESTS ON AERATION TANK 6 FINAL SETTLING TANK
Figure 49
-------
SECTION VIII
SHOCK LOAD STUDIES
introduction
Since no great differences were found in system perfor-
mance under normal operating conditions, it was decided
to apply some shock loads to determine the response of
each side to rapid changes in loading.
To evaluate performance grab sampling was done. During
each test samples were taken periodically at the aera-
tion tank influent and final settling tank effluent. Dur-
ing some tests samples were also taken at the aeration
tank effluent. DO measurements were made periodically in
the aeration tank.
To establish the normal loading variations a 24 hour samp-
ling analysis was performed on the CM system on January 18,
1971. The results of this testing are illustrated in Figure
50.
A source of high strength waste was located at the Kent
Cheese Plant in Kent, Illinois. The cheese plant agreed to
supply the Freeport Waste Treatment facility with 4500
gallon truck loadsof whey from the cheese making process.
The organic strength of the whey was estimated at 65,000
mg/1 BOD5.
During a shock load the flow entering the plant was by-
passed around the grit chamber and primary settling tanks.
The whey was dumped from a semi-truck into a manhole ad-
jacent to the plant. Here it mixed with the incoming flow
from the intercepter system and was then pumped directly to
the aeration tanks.
Three shock load tests will be considered. During the first
one (designated "Shock Load No. 2") the flow was split
equally to each system. During the next test all the flow
was directed into aeration tanks 2 and 4, the PF system
(designated "No. 3"). Finally the last test consisted of
applying all the flow to aeration tanks 1 and 3, the CM
system (designated "No. 5").
Shock Load No. 2
Inlet flow for this test was as shown in Figure 1. The
diffuser pattern for CM was as diagrammed in the " 198
total diffuser" diagram in Figure 5, and for PF as shown
103
-------
10pm 12
Mdnt
Jan. 17
Jan. 18,1971
24 HOUR PLANT EVALUATION - CM SYSTEM
8 10
- Figure 50
-------
2.25
3.50
4.50
4.50
in the lower portion of Figure 6. The average total sewage
flow rate was 5.2 mgd.
The whey was applied over about a one half hour period and
resulted in a sharp increase in loading as reflected by BOD,
COD, and SS values designated as primary effluent in Figures
51,52, and 53 (actually "primary effluent" is a misnomer
since the primaries werebaing by-passed). Final effluent No.
1 refers to the CM system and FE No. 2 to the PF system in
these figures.
As would be expected, the peak BOD and COD concentrations
occurred sooner in the CM system than the PF:
Parameter Time to Peak in Effluent (hrs)
CM PF
BOD
COD
TSS 6.50 9.50
The tracer tests showed that the peak organic concentration
would occur almost immediately at the AT effluent of the CM
tanks, and the peak for PF should occur at the nominal T.
Using the COD values, the detention time in the FSTs would
be 3.5 hours. Nominal detention time would be 4.9 hours.
On the basis of BOD data, T in the FSTs would be 2.25 hrs.
Assuming this to be true, and that T in the PF ATs is 2.4
hours, this would predict time-to-peak for the final eff-
luent to be 4.65 hours. This compares favorably with the
4.5 hours shown above for BOD.
The point is that time-dependent organic tests bear out
the tracer study results. Secondly, the data indicates
that the average detention time of the final effluent port-
ion of the influent flow to the FSTs is something less than
the nominal detention time,
Peak values of final effluent BOD, COD and SS were very
similar for each system. Actual mass removed for each para-
meter for this test was not calculated. Due to the similar-
ity of response, it was decided to do another shock load at
increased loading and decreased detention time.
During the test DOs were also measured in each of the
aeration tanks. For the PF system, measurements were taken
at the .25 and .75 points alone the length. The results for
PF are shown in Figure 54 and for CM in Figure 55. Of sig-
105
-------
§1000
H-
UJ
o
500
°- 0
_ 60"
40
£ 20
c
CM
0
CM
60
40
I
s 20
c
U.
0
o-e
9am
PF
llpm Ipm 3pm 5pm
Time-Hour Of Day
BOD5 DATA-SHOCK LOAD NO. 2
Figure 51
7pm
9pm
-------
o
3OOO
2000-
UJ
£• 1000
o
ol
0>
_g
!t
UJ
o
c
CVJ
6
c
0)
LU
1
L
0
300
200-
100
0
300
200
100
0
9am
C M
PF
_L
llpm
Ipm 3pm
Time- Hour Of Day
COD DATA—SHOCK LOAD NO.2
Figure 52
5pm
7pm
9pm
-------
o
GO
o Total Suspended Solids
A Nfclatile Suspended Solids
Sam
Ipm 3pm
Time-Hour Of Day
SOLIDS DATA-SHOCK LOAD
Figure 53
9pm
NO.2
-------
o
-------
o
Q
c
,o
o
"5 *•
I
0
.E 3000
| £2000
i °tooo
6
ro
*
.*
,o
0
9am
_L
I lam
1,304,400 cm
J_
5pm
Overall
Avg.DO=2.4
Ipm 3pm
Time —Hour Of Day
DO versus TIME-COMPLETE-MIX —SHOCK LOAD NO. 2
Figure 55
7pm
9pm
-------
nificance here is the fact that for the same organic load,
the same sewage flow rate, and the same DO, more air was
required for the PF system than the CM system (1,397,400 ft3
as compared with 1,304,400 ft3).
Shock Load No. 3
During this test the entire flow was directed into the PF
system using the same inlet points and diffuser configura-
tion described for Shock Load No. 2. The Sewage flow rate
averaged 4.0 mgd resulting in a nominal T of 1.58 hours.
To increase the organic load two whey dumps were used. The
load peaks in terms of BOD,COD and SS are shown in the aer-
ation tank influent graphs in Figures 56,57 and 58.
To separate the effects of aeration tank and final settling
tank operation, samples were taken at the effluent of
the PF aeration tank system. Each sample was filtered and
then tested for BOD and COD. These results are shown in the
center graph in Figures 56 and 57.
To carefully evaluate treatment efficiency, the mass of
BOD and COD passing each sampling point was calculated
using the sample data and the average sewage flow rate.
The masses were determined for the period 9:30 am to
4 pm or 6.5 hours. Table 14 contains a summary of these
calculations.
Air flow and DO measurements were taken. Again, the .25
and .75 points were chosen for DO measurement. Figure 59
shows the results of this testing.
Shock Load No. 5
During this test, the entire flow was directed into the
CM system using the same sewage and return sludge inlet
points described for Shock Load No. 2. The diffuser con-
figuration was changed, based on the results of the final
tracer test, to the system using the over balance of dif-
fusers towards the effluent side of the tank as diagrammed
in Figure 5.
Problems with hauling the whey prevented applying a double
shock load as was done in No. 3, so only a single aeration
tank influent peak of BOD,COD and SS is shown in the upper
graph of Figures 60,61 and 62.
The average sewage flow rate was 3.9 mgd as depicted in
the upper graph in Figure 63.
Ill
-------
2000-
c
_D
•E
C
o
1000
0
200
OJ
6
JC
o
o
I
100
200
CM
d
a)
LU
"a
c
u_
100
\
Total Mass
:30am-4
4090 Ib
9am
Ham |pm 3pm
Time-Hour Of Day
BOD DATA-SHOCK LOAD NO. 3 - P F
Figure 56
5pm
112
-------
4000
33000
§ 2000
c
o
B 1000
CM
ti
300
200
£ 100
Total Mass
9:30am-4pm
6791 Ib
0
300
i
| 200
3
^ 100
o
c
il.
9am Mam Ipm 3pm
Time -Hour Of Day
COD DATA-SHOCK LOAD NO. 3 - P F
Figure 57
5pm
.13
-------
0 Total Suspended Solids
A Volatile Suspended Solids
Sam
Ham Ipm
Time-Hour Of Day
SOLIDS DATA-SHOCK LOAD NO.3
Figure 53
5pm
- PF
114
-------
O
LL
a
6
4
4000
3000
2000
a
a
1000
oS
c\j
O
Q
I
3
2
0
9am
3pm
Ham Ipm
Time — Hour Of Day
AIR a SEWAGE FLOW-SHOCK LOAD NO. 3 - P F
Figure 59
5pm
115
-------
2000
§ 1500
g 1000
c
.o
f 500
<3J
<
0
250
200
c
-------
4000
33000
§2000
D
v_
O>
1000
4461 Ib
500
400
c
I 300
LU
c
o
c
o
D
t-
0)
200
100
to
cB
I
I-
co
u.
300-
200-
100-
10am 12am 2pm 4pm
Time Of Day
COD DATA-SHOCK LOAD N0.5-C M
Figure 61
117
-------
£ 100.
0
60-
50
40
30
c 20
0)
E
UJ
0
vss-
TSSs
J_
10am 12am 2pm 4pm
Time — Hour Of The Day
SOLIDS DATA—SHOCK LOAD IM0.5-CM
Figure 62
-------
Treatment efficiency was again evaluated by calculating
masses of BOD and COD entering and leaving the various
points in the system. These calculations are summarized
in Table 14.
Air flow and DO were measured periodically. As in all
the tests, the air flow was adjusted to attempt to main-
tain a DO of 2 mg/1 in each tank. A sharp drop in DO was
noted (refer to Figure 63) during the time the shock load
was applied. Such a drop was also experienced in the pre-
ceding tests. A discussion of system DO response follows
in the next section.
Comparison of Shock Loads No. 3 and No. 5
The comparison data is summarized in Table 14. Each load
was applied at about the same flow rate (4 mgd) yielding
a nominal T of 1.6 hours. F/M loading was essentially
the same (2.4 to 2.5).
These tests indicate that the CM system yields more eff-
ective air utilization. First, comparing Figures 59 & 63,
the drop in DO was not as severe nor of such long duration
in the CM system. Second, Table 14 shows that the scf air
per pound of BODs removed was less for the CM tanks.
Removal percentages were somewhat better for CM. On the
basis of aeration tank BODs removal, CM was 80.4% compared
with 78.3% for PF.
However COD removal data showed an opposite balance of
66.9% for CM and 75.6 for PF.
The CM did show better overall removal for both COD and
6005. It is interesting to note that CM effected consid-
erable removal in the FSTs where the PF did not.
On the basis of the above data it can be concluded that
CM does provide for improved treatment during periods of
extreme shock loads involving a decreased detention time
and an increased organic load. However, it should also be
stated that under more normal detention times; e.g., "Shock
Load No. 2", no significant difference in system performance
between CM and PF could be determined.
119
-------
Q
O
4.0 (
3.0 -
C 2.0|-
8.
Q>
1.0 -
0
o
o>
E
O
Q
4000
3000
2000
1000
0
3.0
2.0
1.0
0
Qavg =3.9 MOD
VAT 3
1,099,500 scf
ATI-
4pm
10am 12am 2pm
Time Of Day
AIR8SEWAGE FLOW—SHOCK LOAD NO.5 - Figure 63
120
-------
COMPARISON SHOCK LOADS NO. 3 & NO. 5
PLUG-FLOW VERSUS COMPLETELY MIXED
TABLE 14
Parameter
Test
Period
Avg. Qs
Detention
Time
SCF
Ib BOD5 remvd
Avg. D.O.
Avg. F/M
Ib BOD^/day
Ib MLVSS
Removal
Percentages
j PE-AT
JUJUc /
D{| PE-FE
|
| AT-FE
1 PE-AT
|
COD /
< PE-FE
\
1 AT-FE
(#3)
Plug-Flow
6. 5 hrs
4 . 0 mgd
1. 58 hrs
431
1.2
2.38
78.3
73.0
-
75.6
76.6
4.1
Completely-
Mixed
(No. 5)
5 hrs
3.9 mgd
1.62 hrs
369
1.5
2.55
80.4
84.4
20.3
66.9
78.5
35.1
Comments
Total air: PF = 1,286,
100 scf
CM - 1,099,
500 scf
Total Ib BOD5:
PF = 4090 Ib
CM = 3526 Ib
DO was depressed to 0.2
for 2 hrs in PF. It
only dropped to a low
of .4 for about .5 hr
in CM. The avg DO in
PF would be lower
except that initial
values prior to the
shock load were quite
high 7-8 mg/1.
Note: removal effect-
ed FSTs for BOD & COD
in CM system.
121
-------
SECTION IX
PILOT PLANT OPERATION
Objective and Scope
The main objective of this investigation was to run a
pilot plant utilizing a complete-mix activated sludge
system in parallel with an existing full scale plant to
determine the value of a model plant as a tool in the
design and operation of future treatment plant. The goals
are enumerated as follows:
1. To compare results of a scale model activated sludge
plant operating under homogeneous mixing conditions
(complete-mix) with a full scale plant operating under
similar conditions.
2. To attempt to vary the operating parameters of the pilot
plant to obtain optimum treatment efficiency and then
compare these established parameter values with parallel
ones in the prototype.
It is intended to obtain design constants from the operation
of both plants, compare results from one plant with the other,
and then see how well the pilot plant constants would have
accurately predicted full scale operation, if such a pilot
plant were to be used in the design of a similar full scale
system. To facilitate the determination of the various design
constants and parameters involved, the following tests were
performed in addition to the routine laboratory analysis:
1. Oxygen uptake tests on samples of mixed liquor from
the pilot plant aeration tank
2. Oxygen transfer efficiency tests using tap water and
final effluent
3. A dye-tracer test to determine the mixing conditions in
the aeration tank.
4. A 24-hour grab sample evaluation
Since similar tests were also run on the full-scale plant,
comparisons of results can be made.
Equipment,Operation arid Testing
Equipment
The pilot plant was installed at the end of the primary
settling tanks, just prior to the point where the primary
effluent enters the tunnel to the aeration tanks. This in-
sured, basically, the same waste to both plants.
123
-------
The pilot plant (see schematic diagram in Figure 64) con-
sists of the following two main units:
1. A 6" x 10" x 8' -0" deep aeration tank column with a
six inch saran wrapped diffuser at the bottom
2. A 13.5" diameter by 7'-ll" average depth final settling
tank equip^d with a small rotary sludge scraper at the
bottom and with glass panels for viewing sludge level.
Primary effluent was pumped to the inlet of the aeration tank,
as was the return sludge from the final settling tank. The
pumps used were of the following specifications:
Randolph positive displacement type Model 500 with
zero-max speed control
catalogue No. 2130
minimum capacity: 130 cc/hr rated against water
maximum capacity: 45 g/hr
tube material: tygon or rubber
tube size: 7/16" OD x 1/4" ID
Air was supplied from the laboratory compressor. A mano-
meter (U-tube type), a thermometer (°F) and a rotameter (to
measure air flow rate) were introduced in the air line. The
rotameter was of the following specifications:
Fisher & Porter, Lab Crest Mark III Tri Flat
flowmeter kit
range: 0.4 cc/min to 23,400 cc/min
Two of the flowmeters were calibrated to give the air flow
rates (cc/min) at prevailing temperature and pressure. The
pilot plant equipment included two automatic sampling units
manufactured by Brailsford and Company, New York and having
the following specifications;
Model EV2 and EV3
Other items included sampling jugs and ice buckets to pre-
serve the samples.
Operation
Primary effluent was pumped into the top of the aeration
tank and into a drop pipe which extended down to a few
inches above the diffuser. The mixed liquor flowed over a
weir through a 1.5" pipe to the settling tank, where it was
discharged at the bottom. The settled solids were collected
by a scraper, and the return sludge was pumped from the tank
124
-------
to
01
Automatic Sampler
for Final Effluent
M-T
^
^
al
P^
1-
t
Efflu
e stor
\.
*
ent "
ed
•ere
UJ:
Auxiliary^
Valve
Overflow
Ice \ Final Effluent
Chest Sampled here
Final
-Settling
Tank
Return
Sludge
Pump
H
c
V
1 1
\
Automatic Sampler
kFor Primary Effluent
.
fe
f
L
r
\
=-*- Plastic
Container
_^- Drop
Pipe
*• Aera-
tion
Tank
Airfror
lab! *•
compres-
sor
T3
0)
0)
tij
c
0)
J
w
§
•e
•H
^
f
, Primary E
Sampled h
Primary
Sample s
her
f fluent
ere
Effluent
tored
i
nice /
Che7
\ ' o _* f.
Saran
Waste Sludg
Line
Rotameter
rffuser
\ Sampling Jug Figure 64 Pilot Plant Schematic Diagram
er
Primar;
Manometer '
Sampling
Jug
Primary
Feed Pump
Tanks
Channel
-------
bottom up to the aeration tank inlet.
Sludge wasting was done by opening a valve located just
beyond the effluent point of the return sludge pump.
The effluent from the final settling tank flowed over a cir-
cular V notched weir into a 3/4" rubber hose.
The primary feed rate was determined by measuring the eff-
luent flow rate. A stopwatch and a graduated container were
used to measure this flow rate as well as the return sludge
rate which was measured just before entry into the aeration
tank.
One important aspect of the whole pilot plant operation was
sampling. It was decided that composite sampling would be
most representative, especially since that was the method
adopted in the full scale operations. During the initial
part of the pilot plant operation, sampling was done man-
ually. Every hour 100 ml of primary effluent was picked up
from the trough in which the influent line of the primary
feed pump was submerged. An equal amount of final effluent
was collected as it came out of the effluent rubber hose.
The composite samples were collected in jugs which were
refrigerated. However, it was felt that this system was
prone to human errors, and it was decided to install two
automatic samplers. One was mounted at the point where the
primary flow discharged into the aeration tank and the other
at the outlet hose of the final settling tank. Plastic con-
tainers were placed at these two points, the liquid level
being maintained by sizing a hole in the bottom of each to
allow flow to go through with a sufficient head. In this
manner the intake of the automatic sampler was always under
the liquid surface. The samplers were set to remove about a
gallon of sample every 24-hour period. They discharged into
plastic jugs which were kept in insulated boxes in which ice
was placed each morning. In this way, the samples were main-
tained at approximately 40° F and the entire sampling system
was automatic.
The pilot plant was essentially operated in two phases:
1. In parallel with the complete mix side of the full
scale
2. At. various periods of constant detention time
There were four periods of parallel operation. During each
of these periods an attempt was made to control hydraulic
flows,(million gallons/day versus 1/hr) in both plants to
yield similar average detention times. The mixed liquor con-
centration was also kept the same so that each plant would
be similarly loaded. It was originally intended to maintain
12-3
-------
the same return sludge to feed ratio, However, due to
resulting operational problems in the pilot plant, the
ratio had to be increased to 0.5 from the original 0.3,
which was maintained in the full scale operations.
Flow equalization was fairly well achieved by matching the
flows exactly in each plant every two hours, starting at
8 am on each day. Return sludge rates were corresoondincly
changed. At 4 pm, or in some cases as late as 10:30 pm, the
flow through the pilot plant was adjusted to a rate which
would correspond to the average flow through the full scale
plant between then and 8 am the next morning. Figure 65
gives an idea of flow variations in both plants. It is a
typical representation.
Five ten-day periods were chosen for the constant detention
operation. In each period the detention time was held con-
stant. The detention times chosen were from a nominal of 5
hours down to 1 hour. It was intended originally to achieve
three levels of MLSS in each period (1,000, 2,000 and
3,000 mg/1). However, owing to limitations of time in each
period, only one or two distinct solids levels were obtained.
In each operational phase, it was intended to maintain a
return sludge to feed ratio of 0.5 and a DO level of 2.0
rag/1.
Testing
Routine laboratory tests were run on pilot plant composite
samples in the same manner as on the full scale composite
samples. The following tests were run daily:
1. Total suspended solids (TSS)
2. Volatile suspended solids (VSS)
3. 5-day BOD on unfiltered samples (BOD5UF)
4. 5-day BOD on filtered samples (BOD5F)
5. Chemical oxygen demand on unfiltered samples (CODUF)
6. Chemical oxygen demand on filtered samples (CODF)
Mixed liquor and return sludge samples were withdrawn at
the same time every day. Besides the solids test on both
samples, the half-hour settling test was also conducted
on the mixed liquor samples.
Other measurements included:
1. Aeration tank DO using a YSI oxygen probe
2. Mixed liquor temperature
3. Air measurements that included flow rate and temp-
erature and pressure readings <-J <
127
-------
50
6
40
A -zn
f-i
f,
"£r
cu
g -p
O «H
H -*
^ -c 20
CD 0>
££
£
CO
2.0
10
n
° 8
Figui
..,,. Pilot Plant
, _».^. ._
/ \ / \
' \ / ^
>/ \
j* ' \ ^^^ rUll bUdle
. J x ,
s
s '
s~. /
~\ /
\ /
\ 1
. , Y /
I
\ y
* •** • — "**
Pilot Plant 37.8 1/hr
Full Scale 4.8 mgd
l 1 1 1 1 1
am 12 N 4pm 8pm 12MN 4am Bam
Time of Day
e 65 Typical 24 Hour Flow Pattern During Parallel Evaluation
-------
Feed and return sludge flow rates were checked periodic-
ally, especially during the parallel operation when it was
necessary to match flows with the full scale plant. During
the constant T operation, flows were checked daily and ad-
justed as required.
Some special tests were conducted, the results of which
were incorporated with the operating data. They were:
1. Oxygen uptake tests
2. Dye tracer test
3. Oxygen transfer efficiency test.
All the laboratory tests were done in accordance with
Standard Methods14 with the following alternate methods
being adopted. In the solids test an asbestos fiber mat
was used to filter composite primary effluent and final
effluent samples, while mixed liquor samples were filter-
ed through 1 inch diameter Gelman Type A glass fiber
filters. All DO measurements were taken with a YSI DO
probe.
Pilot Plant Operational Problems
The only major problem involving malfunctioning of equip-
ment was caused by the sludge scraper mechanism. The orig-
inal motor failed during the initial stages of operation.
This resulted in a very low concentration of return sludge,
since the scraper at the bottom of the settling tank worked
intermittently. Replacement motors were not powerful enough
to work the scraper. Eventually, the whole mechanism was
replaced by a new one which was different from the original
in that the new drive turned at 1 rpm instead of 1/3 rpm.
This caused the solids at the bottom of the final settling
tank to be in continuous agitation. Though return sludge
concentrations were quite normal, it is suspected that the
faster drive did cause a certain amount of solids carry-
over resulting in higher final effluent suspended solids
values.
The primary feed pump and the return sludge pump worked
with reduced efficiency at low flow rates. Also, the hoses
in the flow streams tended to clog because of solids de-
position. Even though the hoses were replaced periodically,
solid deposition was unavoidable. Thus there was a definite
loss of solids in this manner.
From the operational point of view, it was felt that the
sampling program was not ideally representative. Because
of its size, the system was very responsive to slight
129
-------
changes, and a continuously steady-state operation seemed
unlikely. It is felt that a testing program repeated at
least 3 times daily (as in the full scale operation) would
present more meaningful data, expecially during the con-
stant detention time (T) operation. It was found to be ex-
tremely difficult to vary the MLSS concentrations in a
reasonable period of time when operating at long detention
times. It was because of this reason that a study of op-
timum MLSS and detention time combination was not entirely
successful.
Certain improvements in results during the operation were
noticed and the following observation can be made:
1. Automatic sampling was much more representative and
efficient than manual sampling, as far as composite
sampling is concerned. Many errors were avoided by
automation.
2. There was definitely better operation with rubber
hoses than with tygon tubing, especially for lines
leading into and out of the feed and return sludge
pumps. Rubber hosing decreased flow pulsations con-
siderably and had a longer life.
Results
In this chapter the results are presented first. However,
for each period of operation, removal efficiencies, load-
ing and other related parameters have been summarized. The
results of the various special tests conducted are also
reported. Interpretation of the results and operational
data are presented in the second half of this chapter.
Introduction
The results from the parallel operation are given for each
period separately in Tables 15 to 18. Corresponding values
for each parameter for the full scale operation are also
given in these tables.
Tables 19 to 23 summarize each period of constant detention
time individually, giving corresponding full scale values
as before. Table 24 is a summary of the pilot plant const-
ant T operation, and Table 25 is a similar table for the
full scale operation.
For comparing pilot-full scale operation, consider each
period in turn, taking parallel operation first.
130
-------
TABLE 15
Pilot-Full Scale Comparison
June 15-30, 1971
Parameter
BOD Removal %
PEUF to FEF
BOD Removal %
PEUF to FEUF
COD Removal %
PEUF to FEF
COD Removal %
PEUF to FEUF
Solids Removal %
TSS
VSS
Lb BOD
day Ib MLVSS
SCF air
Lb BOD
Lb BOD
1000 ft 3
Detention Time, hrs
Loading on FST
Hydraulic, gpd/ft £
Solids, Ib/day/ft 2
Pilot Plant
84
60
70
58
68
70
0.65
1345
55
2.2
348
7.0
Average: Flow (.000267 mgd) 42.1 lit/hr.
MLSS 2427 mg/1
MLVSS 1332 mg/1
PEUF BODc
83.0 mg/1
Full Scale
81
58
78
63
69
68
0.72
908
68
2.1
800
17.3
3.1 mgd
2622 mg/1
1524 mg/1
93.0 mg/1
131
-------
TABLE 16
Pilot-Full Scale Comparison
July 1-11 & 26-31, 1971
Parameter
BOD Removal %
PEUF to FEF
BOD Removal %
PEUF to FEUF
COD Removal %
PEUF to FEF
COD Removal %
PEUF to FEUF
Solids Removal %
TSS
VSS
Lb BOD
day Ib MLVSS
SCF air
Lb BOD
Lb BOD
1000 ft
Detention Time, hrs
Loading on FST
Hydraulic, gpd/ft ^
Solids, Ib/day/ft 2
Pilot Plant
87.1
69.5
72.9
65.3
67.7
67.9
0.63
678
51
2.9
286
4.9
Average: Flow (.000211 mgd) 33.3 lit/hr
MLSS 2066 mg/1
MLVSS 1296 mg/1
PEUF BOD5 98 mg/1
Full Scale
89.4
72.7
75.5
61.9
75.6
75.6
0.82
842
56
3.3
494
6.6
1.94 mgci
1590 mg/1
1100 mg/1
123 mg/1
132
-------
TABLE 17
Pilot-Full Scale Comparison
July 12-25, 1971
Parameter
BOD Removal %
PEUF to FEF
BOD Removal %
PEUF to FEUF
COD Removal %
PEUF to FEF
COD Removal %
PEUF to FEUF
Solids Removal %
TSS
VSS
Lb BOD
day Ib MLVSS
SCF air
Lb BOD
Lb BOD ,
1000 ft
Detention Time, hrs-
Loading on FST 0
Hydraulic, gpd/ft
Solids, Ib/day/ft 2
Average : Flow (000202 mg
MLSS
MLVSS
PEUF BODC
Pilot Plant
88.9
73.4
76.5
67.8
70.5
68.7
0.60
686
38
3.0
268
3.3
31.8 lit/hr
1459 mg/1
982 mg/1
76.0 mg/1
Full
92.
71.
82.
76.
86.
86.
0.
963
45
3.
509
8.
2.
1926
1272
96
Scale
9
2
0
4
2
8
-
57
2
2
0 mgd
mg/1
mg/1
mg/1
133
-------
TABLE ].8
Pilot-Full Scale Comparison
Parameter
BOD Removal %
PEUF to FEF
BOD Removal %
PEUF to FEUF
COD Removal %
PEUF to FEF
COD Removal
PEUF to FEUF
Solids Removal %
TSS
VSS
-Lb-BOD-
day Ib MLVSS
SCF air
Lb BOD
Lb BOD
1000 ft 3
Detention Time, hrs
Loading on FST 2
Hydraulic, gpd/ft
Solids, Ib/day/ft
Average: Flow (.000223 mgd
MLSS
MLVSS
PEUF BODr
August 10-19, 1971
Pilot Plant
91
79.1
75.9
63.8
69.7
69.7
0.41
1750
33
2.7
337
6
) 35.18 lit/hrs
2088 mg/1
1273 mg/1
60 mg/1
Full Scale
89.4
79.1
82.9
72.2
73.9
71.2
0.49
911
43
2.7
580
10.5
2.32
2147 mg/1
1398 mg/1
79 mg/1
134
-------
TABLE 19
Constant Detention Time Operation
Pilot-Full Scale Comparison
5-Hour Nominal Detention Time
Full Scale
Parameter
BOD Removal %
PEUF to FEF
BOD Removal %
PEUF to FEUF
COD Removal %
PEUF to FEF
COD Removal %
PEUF to FEUF
Solids Removal %
TSS
VSS
Lb BOD
day Ib MLVSS
SCF air
Pilot Plant
93.7
74.3
85.3
70.2
79.8
79.6
January
1-31
89.6
85.9
84.7
75.6
82.0
82.5
Februi
1-12
92.1
87.4
83.7
75.9
77.4
72.8
0.29
0.26
2105
1000 ft.
Detention Time, hrs.
Loading on FST
Hydraulic, gpd/ft 2
Solids, Ib/day/ft 2
Average: Flow (.900119 18.7 lit/hrs
MLSS mgd ) 2271 mg/1
MLVSS 1343 mg/1
1070
0.19
1180
24
5.
.80
3.
06
4
41
4.
331
10.
9
5
37
.5.1
316
11.7
PEUF BODC
84 mg/1
1.30 mgd i. 24
3790 mg/1 4450 mg/1
2500 mg/1 308.0 mg/1
135 mg/1 127 mg/1
135
-------
TABLE 20
Constant Detention Time Operation
Pilot-Full Scale Comparison
4-Hour Nominal Detention Time
Parameter
BOD Removal %
PEUF to FEF
BOD Removal %
PEUF to FEUF
COD Removal %
PEUF to FEF
COD Removal %
PEUF to FEUF
Solids Removal %
TSS
VSS
Lb BOD
Pilot Plant
91.9
69.1
85.4
81.1
79.5
78.9
0.35
Full Scale
February March
13-17 12-17
1327
day Ib MLVSS
SCF air
Lb BOD
Lb BOD ,
1000 ft
Detention Time, hrs
Loading on FST
Hydraulic, gpd/ft
Solids, Ib/day/ft 2
Average: Flow (O00148 23.4 lit/hrs
MLSS mgd) 2314 mg/1
MLVSS 1454 mg/1
PEUF BOD 92 mg/1
90.4
90.4
84.5
on p
87.4
85.6
0.24
990
92.2
81.1
78.8
74.0
91.0
88.7
0.21
1060
33
4.04
224
4.3
49
4.1
393
15.1
34
4.0
405
15.8
1.54 mfe-d 1.59 mgd
4620 mg/1 4670 mg/1
3340 mg/1 2560 mg/1
135 mg/1 90 mg/1
136
-------
TABLE 21
Constant Detention Time Operation
Pilot-Full Scale Comparisons
3-Hour Nominal Detention Time
Parameter
BOD Removal %
PEUF to FEF
BOD Removal %
PEUF to FEUF
COD Removal %
PEUF to FEF
COD Removal %
PEUF to FEUF
Solids Removal %
TSS
VSS
Lb BOD
day Ib MLVSS
SCF air
Lb BOD
Lb BOD
1000 ft 3
Detention Time, hrs
Loading on FST
Hydraulic, gpd/ft
Solids, li/day/ft 2
Average: Flow (.000200 mgd)
MLSS
MLVSS
PEUF BODs
Pilot Plant
91.5
74.4
79.5
72.5
70.5
72.7
0.4
960
40
2.99
304
6.3
31.6 lit/hrs
2485 mg/1
1607 mg/1
82 mg/1
Full Scale
89.7
83.3
79.1
69.5
73.0
71.2
0.33
790
38
3.0
530
13.3
2.08 mgd
3000 mg/1
1850 mg/1
78 mg/1
137
-------
TABLE 22
Constant Detention Time Operation
Pilot-Full Scale Comparison
2-Hour Nominal Detention Time
Parameter
BOD Removal %
PEUF to FEF
BOD Removal %
PEUF to FEUF
COD Removal %
PEUF to FEF
COD Removal %
PEUF to FEUF
Solids Removal %
TSS
VSS
Lb BOD
day Ib MLVSS
SCF air
Lb BOD
Lb BOD
1000 ft 3
Detention Time, hrs
Loading on FST
Hydraulic, gpd/ft 2
Solids, Ib/day/ft 2
Average: Flow (000302
MLSS
MLVSS
PEUF BOD5
Pilot Plant
92.6
65.3
79.4
70.0
70.7
66.1
0.69
564
70
1.98
451
8.3
47.67 lit/hr.
2213 mg/1
1610 mg/1
95 mg/1
Full Scale
91.8
87.7
79.1
70.9
71.4
70.2
0.51
430
82
2.2
724
22.5
2.84 mgd
3720 mg/1
2560 mg/1
122 mg/1
138
-------
TABLE 23
Constant Detention Time Operation
Pilot-Full Scale Comparison
1-Hour Nominal Detention Time
Parameter
BOD Removal %
PEUF to FEF
BOD Removal %
PEUF to FEUF
COD Removal %
PEUF to FEF
COD Removal %
PEUF to FEUF
Solids Removal %
TSS
VSS
Lb BOD
day Ib MLVSS
SCF air
Lb BOD
Lb BOD
1000 ft 3
Detention Time, hrs
Loading on FST
Hydraulic, gpd/ft
Solids, Ib/day/ft 2
Average: Flow (000545 mgd)
MLSS
MLVSS
PEUF BOD5
Pilot Plant
89
76
N/A
N/A
69
69
0.74
505
167
1.1
824
38
86. lit/hr
5568 mg/1
3627 mg/1
123 mg/1
Full Scale
91
79
81
74
83
82
1.00
310
210
1.1
727
24
2.85
4850 mg/1
3310 mg/1
156 mg/1
139
-------
TABLE 24
Summary of Constant Detention Time Studies
Pilot Plant
Raw Flow Det. Primary Effluent Final Effluent
1/hr Time BOD COD BOD COD
Period Q hrs T F UF F UF TSS VSS F UF F UF TSS VSS
August 20
to
August 30 18,7 5.06 27 84 51 218 91 66 5 22 32 65 18 14
o September 1
to
September 10 23.4 4.04 38 92 47 204 94 74 7 28 30 39 19 16
September 11
to
September 20 31.6 2.99 27 82 63 171 95 66 7 21 35 47 28 18
September 21
to
September 30 47.7 1.98 28 95 52 170 75 59 7 33 35 51 22 20
April 27 '
to
May 7 86.0 1.1 74 123 - - 124 92 14 29 - - 38 31
-------
TABLE 24(continued)
Period
Per Cent Removal (%)
BOD COD
PEUF- PEUF-
FEF FEF TSS VSS
F/M
Ib BOD
day, Ib MLVSS
Mixed
Liquor
Temp
°C
August 20
to
August 30
93.7 85.3 79.8 79.6
0.29
September 1
to
September 10
91.9 85.4 79.5 78.9
0.35
24
September 11
to
September 20
91.5 79.5 70.5 72.7
0.40
= 21
September 21
to
September 30
92.6 79.4 70.7 66.1
0.69
20
April 27
to
May 7
89
69.0 66.0
0.74
= 16
-------
TABLE 25
to
Period
January 1
to
January 31
February 1
to
February 12
February 13
to
February 17
March 12
to
March 17
March 26
to
April 5
April 6
to
April 22
April 24
to
May 9
Raw
Flow
> mgd
Summary of Constant Detention Time Studies
Full Scale Plant
Det. Primary Effluent Final Effluent
Time BOD5 COD BOD5 COD
T, hrs F UF F UF TSS VSS F UF F UF TSS VSS
1.30 4.9 82 135 99 287 150 126 14 19 44 70 27 22
1.24 5.1 77 127 98 245 115 92 10 16 40 59 26 25
1.54 4.1 95 135 103 265 127 104 13 13 41 51 16 15
1.59 4.0 39 90 66 208 199 124 7 17 44 54 18 14
2.08 3.0 33 78 67 187 107 80 8 13 39 57 29 23
2.84 2.2 81 122 72 206 126 94 10 15 43 60 36 28
2.85 1.1 104 156 72 226 161 120 14 33 44 58 27 22
-------
TABLE 25 (continued)
co
Period
January 1
to
January 31
February 1
to
February 12
February 13
to
February 17
March 12
to
March 17
March 26
to
April 5
April 6
to
April 22
April 24
to
May 9
Per Cent Removal (%)
BOD5 COD
PEUF- PEUF-
FEF FEF TSS VSS
89.6 84.7 82.0 82.5
92.1 83.7 77.4 72.8
90.4 84.5 87.4 85.6
92.2 78.8 91.0 88.7
89.7 79.1 73.0 71.2
91.8 79.1 71.4 70.2
91.0 80.5 83.2 81.7
F/M
= Ib. BOD
day, Ib MLVSS
0.26
0.19
0.24
0.21
0.33
0.51
1.00
Mixed
Liquor
Temp
°C
= 14
- 13
= 13
= 11
= 12
= 14
= 15
-------
Parallel Operation
June 15-30:
Table 15 clearly indicates that the detention time and the
aeration tank solids concentration were maintained relative-
ly the same in both plants. BOD and solids removals were
equally efficient. There was a slightly higher COD removal
in the full scale plant, but there was data for only four
days for the pilot plant. The slightly lower F/M ratio, i.e.,
Ibs BOD/day/lb MLVSS in the pilot plant, is due to a lower
measured BOD influent (sampling points for model and proto-
type are located at different places).
July 1-11 and 26-31:
Referring to Table 16, the comparisons of loadings and
removal efficiencies are fairly good in most cases. The
most notable exception is the suspended and volatile sus-
pended solids removal. This is attributed to operational
problems in the pilot plant arising out of malfunctioning
of the sludge scraper at the bottom of the final settling
tank.
July 12-25:
In this period too, most parameters are well matched. For
reasons mentioned above, the solids removal efficiency in
the pilot plant is considerably lower.
August 10-19:
This period, summarized in Table 18, is undoubtedly the
best period as far as soluble BOD removal efficiency is
concerned. There is slightly lower COD removal in the
pilot plant. Suspended solids and volatile suspended solids
removal is again significantly higher in the full scale
plant. Referring to the aeration tank data for this period,
it is seen that a fairly high DO was maintained during
this period giving an extremely high value of air applied
per Ib of BOD. This might have been one cause of solids
carry-over in the final settling tank.
In summarizing the parallel operation phase, it is seen
that in most cases removal efficiencies and other para-
meters matched quite closely. One parameter that was sig-
nificantly different in each case was the loading on the
final settling tanks; both hydraulic and solids loading-
being much higher in the full scale plant.
144
-------
For the constant T operation, it must be mentioned again
that the pilot plant was operated separately from the full
scale, i.e., at the time of constant T operation in the
pilot plant there was no control over the flow in the full
scale. The comparisons made are with similar full scale
operation at an earlier time. The only common factor is
similar nominal detention time. There was no attempt to
match mixed liquor solids concentrations.
Constant Detention Time Operation
5 Hour Nominal T:
Table 19 shows that there is little difference in average
removal efficiencies, especially for soluble BOD and COD
values. Influent BOD values are quite different, but owing
to different MLSS concentrations, the F/M ratio is compar-
able.
4 Hour Nominal T:
As can be seen in Table 20, most of the removal efficien-
cies and parameters are quite comparable. It is interesting
to note a higher soluble BOD removal in the pilot plant
even though MLSS concentrations is nearly half of that in
the full scale.
3 Hour Nominal T:
This period shows an excellent comparison between the two
systems, with the pilot plant having a slight edge over the
full scale.
2 Hour Nominal T:
Soluble BOD removal efficiency increased in both plants
over the 3 hour period. Similar trends in both the plants
is quite significant.
1 Hour Nominal T:
The pilot plant showed a slight decrease in BOD removal
efficiency. Solids removal efficiency is significantly
lower. Table 24 and 25 summarize constant T operation
for pilot and full scale plants respectively. One marked
difference is that the mixed liquor temperatures were
higher for the pilot plant than for the full scale.
Oxygen Transfer Tests
Two series of oxygen transfer tests (using the re-oxygen-
145
-------
12 '3 ^
ation technique1 Twere run on the pilot plant utilizing tap
water and final effluent from the full scale plant. A
summary is given in Table 26. Transfer efficiency decreases
with increase in air rate as shown in Figure 66.
24 Hour Grab Evaluation
Figures 67-69 show the results of an intensive 24 hour
evaluation where sampling was done every two hours and DO
was maintained constant. Figure 67 is the data for variat-
ion, in primary effluent parameters. Figure 68 is a similar
plot for the final effluent, and Figure 69 gives the aer-
ation tank DO and air supplied over the same period. It is
seen that the peak values for each parameter came at 12
noon. An initial peak in some cases at 12 mid-night is
because a little before that time, supernatant from the
digesters was pumped through the primary settling tanks.
It is interesting to note that the maximum air was supp-
lied immediately after 12 noon. The total air supplied in
the 24 hour period was estimated to be 270 scf, giving
an average DO of 2.1 mg/1.
Tracer Test
The washout curve obtained in the tracer test is shown
in Figure 70. The data points show a logarithmic decay
curve typical of a completely mixed system. The peak con-
centration was reached in just under ten minutes. A K of
0.01007 min"! was measured as the slope of a semi-log plot
of CI/CQ versus ti/T, giving a measured T value of 1.66
hours as compared with a T theoretical of 2.05 hours. This
indicates that the effective tank volume was somewhat less
than the actual, n was calculated to be 1.1, which is
good compared to an ideal complete mix value of 1.0. Dye
recovery was calculated to be 89%.
Oxygen Uptake Tests
As in the full scale studies, oxygen uptake tests were run
on the pilot plant mixed liquor. A typical test result is
plotted in Figure 71. The tests and measured and calculated
values have been summarized in Table 27. An Ma ratio of
VSS
0.26 was obtained; the average values for K2 and Kg being
.029 and 1.44 hrs'1 (Q2 Units) respectively (K2 is equiva-
lent to Kg and Kg is equivalent to Ks).
146
-------
SUMMARY OF OXYGEN TRANSFER TESTS
Table 26
TAP WATER
Run No.
Air Rate
in cc/min
K20 min"1
Transfer
Efficiency %
Liquid Temp.°C
Run No.
Air Rate
in cc/min.
K20, min"1
Transfer
Efficiency
Liquid Temp.°C
0< r KFE
1
1450
.0803
18.8
16.7
FINAL EFFLUENT
1
1700
.0638
13.0
19.0
.79
2 3
3850 7350
.202 .314
17.8 14.6
16.7 16.7
2 3
3850 7350
.119 .167
10.3 7.6
19.0 19.0
.59 .53
K tap water
147
-------
22
20-
16
16
14
12
o
1 01
O O
-------
in
§
CD
CO
CD
CO
Time of the Day
Figure 67 24 Hour Grab Analysis - Primary Effluent
-------
8
CQ
CO
CO
CO
50
40
30
20
10
10
8
6
4
2
40
30
20
10
40
30
20
10
6pm
I
•*•
I
1U JL2MN 2am 4 fa
Time of the Day
br
bpm
Figure 68 24 Hour Grab Analysis - Final Effluent
-------
o
u
o.
D4
3
CO
f-4
•rl
ft.
cn
I
s^
o
Q
yuuu
8000
7000
6000
5000
4000
3000
2000
1000
0
5.0
4.0
3.0
2,0
1.0-
JL
111
Total air = 7,641,600 cc of air
Total air = 270 scf of air/day
Total air = 81 scf of air/cuft of tank
I
_L
J_
_L
1
I
I
6pm 8
9/27/71
2am 4
Mdnt
8
10 12 2pm 4 6pm
Noon 9/28/71
10 12
Mdnt
Time of the Day
Figure 69 24 Hour Evaluation-? DO profile and air supplied
-------
50
A
fi
O
o
o
40
30
20
10
240 260 280 300
j L_
20 40
j L
60 80
Figure 70
I
180 200 220
100 120 140 160
Time in Minutes -»
Dye Wash Out Curve, Tracer Test
-------
lOOr
I
= 0.0511 hours
Ma = 0.24
VSS
-1
Kg = 1.39 hours
F = 12 mg/1
I
I
200
400 600 800 1000
Time in Minutes, t
Figure 71 Oxygen Uptake, Run No. 9/20/99
1200
153
-------
TABLE 27
Summary of Oxygen Uptake Tests
^2» Kg,
-1 -1 F,
Date Sample hrs. hrs. mg/1
Oxygen Uptake in Total Total
Ma, mg/1 VSS, Ma mg/l/hr due to Uptake Uptake
02 Solids mg/1 VSS Synthe- Endoge- mg/l/hr mg/l/day
sis
nous
7/1 M.L. .0296 2.67 3.0 450 317 1190 0.26 7.7 13.5 21.0
504
Ol
8/14 M.L. 0214 1.19 5.0 467 329 1210 0.27 6.0 10.0 16.0 384
8/31 M.L. .0262 1.73 2.0 382 269 1330 0.20 3.0 10.0 13.0 312
9/9 M.L. .0269 0.59 17.0 409 288 2140 0.13 10.0 11.0
21
504
9/20 M.L. .0511 1.39 12.0 489 345 1450 0.24 16.0 25.0 41
984
9/29 M.L. .0199 1.04 12.0 980 690 1570 0.44 12.0 19.5 31.5
756
Average: .029 1.44
Average from
Full Scale
operation: .026 1.19
* M.L. = Mixed Liquor
0.26
0.26
-------
Discussion
The following points should first be noted:
1. The results of the tracer test clearly indicate that
the system operated under nearly ideal complete mix
conditions,
2. For all computations and analyses, except removal
efficiencies, the BOD values used are ultimate values.
These are based on the relation, BODr = 1.46. The
BOD5
BOD rate constant was determined experimentally for full-
scale samples and was calculated to be 0.1 for both the
primary and the final effluent samples. The same value
was accepted for the pilot plant because of very similar
sample properties.
Referring to the parallel and constant detention time op-
erations (Tables 15 to 25) , the following observations can
be made. In the periods of parallel operation, when flows
were duplicated and mixed liquor solids levels were matched,
removal efficiencies and other parameters showed good com-
parison between the pilot and full-scale plants.
The highest removal efficiency (based on soluble BODs) pre-
dicted by the pilot plant was 91% and the highest achieved
in the full scale plant was 93%- the two values falling in
different periods of parallel operation. During the constant
detention operation, the highest removal efficiencies were
94% and 92% respectively for the model and prototype.
It should be born in mind that all removal efficiencies were
based only on primary to final effluent. If they had been
based on raw to final, higher efficiencies would obviously
have resulted. The above discussion on removal efficiencies
thus restricts itself to the activated sludge process only.
In Figures 67-69, the data from 24-hour grab analysis has
been plotted. Referring to Figure 67, the plot for the
primary effluent, it is seen that the strength of the waste
changes considerably during the day; the BOD varying from
a high of 150 rag/1 to a low of 50 mg/1 . Thus, there is an
appreciable fluctuation in oxygen demand as is well defined
in Figure 69. It is therefore quite possible that due to
oxygen becoming limiting, complete biological oxidation is
lacking, resulting in high final effluent BOD values. Figure
68 indicates that if enough air is applied at all times, a
consistently low final effluent BOD can be obtained.
The COD measurements do not seem to show much variation
from period to period, though at high detention times there
155
-------
is positive evidence of better removal.
McKinney's BOD removal rate constant, K4, was determined
using data from each period. The results have been summar-
ized in Table 28 for the pilot plant operation. The values
of K4 range from 2.4 hrs"1 to 7.3 hrs"1. Table 29 compares
removal rate constants for four periods of full scale data.
K4 values from both plants, especially for BODL,_are extreme-
ly comparable with both averaging around 2.9 hrs"1.
Figure 72 is a plot of % BOD removal versus detention time
T. The solid line has been drawn for the constant T opera-
tion. The dashed lines are theoretical curves for K4 vary-
ing from 2 to 8 hrs"1. It is seen that the experimental
curve falls between K = 2 and K = 5. This value of K^ is
considerably lower than those reported by McKinney and
Wallace and others. McKinney reported values in the range
of 15 hrs"1.
10 7
Though McKinney and Smith advocate no dependency of K4
on MLSS, Wallace15 in his work at the University of Iowa
showed definite dependency of K4 on MLSS in some constant
T periods and a. lack of any correlation in others. Figures
73-76 were plotted (for two periods) and indicate a re-
lationship between K4 and MLSS. In the present investigation
at nominal T values of 5 hours and 3 hours, there is every
reason to believe that K4 is solids dependent. There does
not seem to be any correlation in any of the other time
periods. One thing, however, must be born in mind here. In
all periods except one, the maximum level of MLSS concen-
tration is about 3000 mg/1.
That the removal rate constant is solids dependent was
proposed by Eckenfelder1! ±n his model. Data for this
analysis is plotted in Figures 77 and 78. There is a
trend towards a linear relationship, and a straight line
approximated through the data points gives a K of 0.064
days"1 for BOD and .031 days"1 for COD. If we were to
compare the removal rate constants for the two different
theories we would have McKinney K4 = Eckenfelder K x MLSS.
With respect to the BOD analysis then, for the whole pilot
pliant operation Eck K = .064 days"1 and average K.i = 90.6
days"1. For Eck K the equivalent K4 would then be~K x MLSS
(avg) = 103.3 days"1, which gives a 12% difference.
Another attempt was made to show the extent of correlation
between K and K4. If we assume an unfiltered PEBODr of 100
mg/1 and a FEEDER of 10 mg/1, using an average K4 of 3.8
hrs"1 we have t as 2.4 hrs =0.1 days. If we now go to
Figure 77, for Se = 10 mg/1 we have So_~Se - 0.55 from
Xa t
153
-------
TABLE 28
BOD Removal Rate Constants
Period
1
2
3
4
5
6
7
8
9
* F 1
Fo K4 t + I
days
58.7
57.9
66.0
90.0
78.4
47.2
88.0
155.0
174
FQ = BOD L PEUF
-1
hours
2.4
2.4
2.8
3.8
3.3
1.9
3.7
6. 5
7.3
-------
TABLE 29
Comparison of fy Values for
Pilot Plant and Full Scale
Period
1
2
3
4
Average
Period
1
2
3
4
Average
BODL Basis
P
58.7
57.9
66.0
90.0
68.2
F
47.4
60.8
98.5
74.2
70.2
BODL Basis
P F
2.4 1.9
2.4 2.5
2.8 4.1
3.8 3.1
2.85 2.90
K
4,
-1
days
COD
P
25.4
22.9
26.9
27.9
25.8
-1
hours
COD
P
1.1
0.9
1.1
1.2
1.1
Basis
F
39.8
22.2
34.3
43.2
34.9
Basis
F
1.7
0.9
1.4
1.8
1.5
P: Pilot
F: Full-Scale
158
-------
100-
90
80
H
(8
70
60
50
Detention Time, T
Figure 72 % BOD Removal Versus Detention
Time, Constant T Operation
- K=2
159
-------
W
1
H
Nominal T = 5 hrs
I I | I I
[DUO"1100 1200 1300 1400 ibOO
Nominal T = 4 hrs
J I I I I I I (
1000 1100 1200 1300140015001600 1700 1800 1900
MLVSS, mg/1
Figure 73 K4 versus MLVSS, 5 and 4 hrs. Nominal T
-------
8
7
6
5
3
2
Nominal T = 3 hrs
I I I I
I I 1 I I I
1000 1200
1400 1600 1800
2000 2200
MLVSS, mg/1
Figure 74 K4 versus MLVSS, 3 hrs Nominal T
-------
11
10
9
8
7
' 2
f, '6
X
5
4
2
1
1
0 Nominal T = 2 hrs
^™
-
O
0
—
o
-
o o 0
t 1 1 1 1 1
.000 1200 1400 1600 1800 2000 2200
MLVSS, mg/1
Figure 75 K/j. versus MLVSS, 2 hrs Nominal T
-------
I— 1
7>
co
16
14
12
en 10
^t- 8
X
6
4
2
0
Nominal T = 1 hour
0
O
o
0 0
o
1 1 1 1 1
0 1000 2000 3000 4000 5000
MLVSS, mg/1
Figure 76 ]<4 versus MLVSS, 1 hr Nominal T
-------
1.3
1.2
1.1
1.0
0.9
0.8
H W
>,
0!
01
CO
o
CO
ITS
X
0.7
0.6
0.5
0.4
-1
K = 0.064 days
0.3
0.2
0.1
i
8 1012 14
6 18 20
Se, ng/1
Figure 77 Removal Rate Constant, K,
Eckenfelder Approach - BODi Basis
164
-------
1.1
1.0
0.9
0.8
0.7
Hw
'£
•8 0.6
(U
CO
CO
0.5
0.4
0.3
0.2
0.1
10 20 30
Se, mg/1
Figure 78 Removal Rate Constant
Eckenfelder Approach - COD Basis
40
50
165
-------
which value an Xa of 1640 mg/1 can be determined. Then
K4 = KxMLSS = .064 x 1640 - 4.3 hrs"1 which compares
24"
well with a measured K^ of 3.8 hrs. K and K4 do not nec-
essarily yield the same detention time for a given set of
S0j Se and Xa values. Data available is inadequate to make
definite conclusions about the relative preference of one
constant over the other.
In continuing the analysis of the operating data, consider
Table 27 which summarizes 0^ uptake data. K2(Ke) and Kg(Ks)
were averaged out to be 0.029 hrs-1, 02 units (.041, solid
units) and 1.44 hrs"1, 02 units, respectively, and average
Ma was 0.26, which compares extemely well with an Ma of
VSS VSS
0.26 on the full scale and a value of 0.27 predicted by
Frazier1 .
In the oxygen uptake analysis, determination of K2 is
more reliable than the corresponding Kg value which de-
pends on initial uptake rates which change quite rapidly.
Thus, accepting an average value for K2, values for Kg were
determined from the operating data (using the equation Mae
t wMar = T(KsF-KeMa) for each period of pilot plant operation
These have been given in Table 30. Since the oxygen uptake
tests for the pilot plant were conducted using only mixed
liquor samples, values for Mar and Maf» were assumed as 0.26
VSS VSS
and 0.4 respectively, equal to those from full scale data.
Kg was averaged to be 2.1 hrs , neglecting values greater
than 3 hrs"1.
Values of 0.029 hrs"1 for K2 and 2.1 hrs"1 for Kg were then
used in predicting oxygen uptakes from the mass balance
equation and consequently used to predict air requirements.
Referring back to Figure 69 for the 24 hour period when an
average DO of 2.1 mg/1 was maintained, about 270 scf of air
was supplied. The following is a theoretical air requirement
analysis based on K2, Kg and the active mass equation.
From the oxygen uptake data and the operational data, we
have K2 and Kg as 0.041 hrs"1, solids units (.029 hrs"1,
02 units) and 2.1 hrs"1, 02 units, respectively, and M^
VSS
is calculated to be 0.26. For this period of the 24 hour
grab evaluation, the MLVSS was averaged out to be 2655
mg/1, and the final effluent, filtered BO^ was estimated
to be 12 mg/1. The total uptake is then given by do - KaF +
dt ~
166
-------
TABLE »9
Computed Values for Kg
*
Period Kg in hours
1 1.53
2 1.25
3 1.86
4 3.46
5 '2.84
6 1.32
7 2.78
8 2.84
9 4.13
Neglecting values greater than 3, Avg. Kg = 2.1 hours.
Average from full scale operation Kg = 2.4 hours.
*Based on:
Mae + w Mar = T(KgF -
16'
-------
K2Ma> which gives in terms of oxygen units a total re-
quirement of 1287 mg/l/day. The 270 scf/day of air supp-
lied can be converted into similar units to facilitate a
comparison. An oxygen transfer efficiency of 9.5% is esti-
mated as follows. The total air supplied was 7,641,600 cc/
day (Figure 69) . This can be converted to 5307 cc/min. Know-
ing the air rate, the transfer efficiency is obtained from
the final effluent curve in Figure 66. The oxygen transfer
rate can then be calculated as follows: do = 270 scf x
"dT day
.0174 Ib 02 x .095 x 453.6 x 10Jmg - 2140 mg/l/day
scf 94.62 liters IF"
This indicates that more air was supplied than predicted by
the uptake equation above.
To study the capacity of the pilot plant to predict air re-
quirements for the full scale plant, a similar analysis was
performed using the full scale operating data. Figure 79 is
a plot of a 24 hour DO monitoring of the full scale plant.
The total air supplied was estimated at about 1.7 million
scf on September 8-9. This gives an air rate of about 600
scfm per tank. In the same period, the following values
were obtained from the full scale operation:
MLVSS - 2113 mg/1
= 10 mg/1
The total uptake is then do/dt = KgF 4 K2Ma= 1043 mg/l/day.
Air supplied to the full scale tank can be converted to the
same units as before. From oxygen transfer efficiency tests
on the full scale aeration tank, the transfer efficiency was
estimated to be 10% at an air rate of 500 scfm per tank (Note
the similarity in the pilot and full scale oxygen transfer
efficiencies) . Using a transfer efficiency of 10%, the rate
at which air was supplied is calculated to be 1370 mg/l/day,
which does agree closely with the 1046 mg/l/day required,
keeping in mind that DO's did fluctuate above 2 mg/1 dur-
ing the 24 hour period.
16:
-------
25OO
20OO
1500-
g. IOOO
en
3 500
o
o
Q
0
J
3465 square X 500 scf =1,732,500 scf
square
\
J_
AVG. DO =2.1
j_
lOam I2n 2 4 6 8
Time-Hour of the Day
DO PROFILE , SEPT. 8-9 1971,
Figure 79
10
I2mn
8
FULL SCALE PLANT
-------
Solids Production
The two methods of solids production analysis (active
mass balance and Eckenfelder solids balance) outlined
in Section V were applied to the pilot plant data as
they were to the full scale data.
The following is an analysis of solids production using
the Eckenfelder approach.
Figure 80 is a plot of AX versus Sr from the data in
v ^7 V - T
XaV ^a'1
Table 31. The constants a and b are calculated as 1.0
(dimensionless) and ~0.22 (days'1) respectively. The
dashed line is a similar plot for full scale operations
which gives a and b as 0.8 and -0.10 respectively. The
final prediction equation for the pilot plant is then,
A X r itQ g
-°-22
It is interesting to note that the a and b values for
the pilot plant and the full scale PF solids analysis
agree exactly (a = 1, b = -.22)
Solids production was also estimated using the basic active
mass equation, Mae + wMar = T(KsF - KeMa) from which w can
be determined since all other constants have already been
established.
Then sludge wasted is given by AX - wQ x RSVSS.
To study the applicability of each model and to see how
well the pilot plant constants could predict sludge accum-
ulation, values of A X were calculated from each model
using the data from the full scale CM operation. These
predicted values of AX were then compared with the values
of sludge wasted actually recorded during full scale op-
erations. Tables 32 and 33 respectively show measured and pre-
dicted values for sludge wasted using the active mass eval-
uation and the Eckenfelder evaluation.
To evaluate the Eckenfelder approach, Sr was calculat-
Xa-t
ed for each full scale operational period. The correspond-
ing A X was estimated from Figure 80 and A X calculated
XaT""
from operational values of Xa and V. For the active mass
balance model, actual values were used in the active mass
equation to calculate w and finally AX. In this analysis
170
-------
No.
TABLE 31
Sludge Production - Eckenfelder's Evaluation
Period
A X (days"1)
(days-1)
1
2
3
4
5
6
7
June 15-30, 1971 0.50
July 1-11 & 26-31 0.29
July 12-25 0.53
August 10-19 0.33
August 20-31 0.19
Sept. 1-10 0.09
Sept. 11-20 0.29
Sept. 21-30 0.22
April 27 - 0.49
May 7
0.60
0.63
0.67
0..48
0.32
0.38
0.45
0.71
0.82
171
-------
0.6
0.5
OA
0.3
0.2
7 O.I
w
o
•o
* K o
SOLIDS PRODUCTION - PILOT PLANT
(ECKENFELDER Q3
EVALUATION }
Figure 80
i /
O
Full Scale-/
CM '
©8
Pilot Plant-CM
I I
o.r
2 03 0.4 05 0.6 0.7 OB 09
(days-1)
-0.1
-0.2
172
-------
TABLE 32
Comparison of Predicted and Measured Values
of Sludge Wasted in Full Scale Plant-
Active Mass Evaluation
Period
A X
Predicted, in
Ibs/day
AX
Measured; in
Ibs/day
A
B
C
D
E
F
G
H
I
J
K
L
M
N
O
1514
854
3507
3614
815
512
1182
745
1067
750
944
961
1351
928
173
-------
for periods 2 to 8, all values of AX predicted were nega-
tive, meaning no sludge wasting, when actually sludge was
wasted in the routine operation.
The result of the pilot plant solids analysis is parallel
to the full scale work. Namely, that more study is needed
to accurately predict solids production under field con-
ditions.
174
-------
TABLE 33
Comparison of Predicted and Measured Values
of Sludge Wasted in Full Scale Plant -
Eckenfelder Evaluation
Period
AX
Predicted
(Ib VSS/day)
AX
Measured
(Ibs VSS/day)
A
E
r*
\s
D
E
F
G
H
I
J
K
L
M
N
0
606
203
661
169
773
2480
3498
2970
792
1839
1288
1012
876
789
394
854
690
915
2583
853
1239
1579
1756
1067
750
944
961
1351
928
1267
175
-------
SECTION X
IMPROVED DESIGN CONCEPTS
General
The intent of some research programs is to determine with
data the reliability of design concepts on full scale op-
erations. The homogeneous mixed system has been tested ex-
tensively in the laboratory, pilot plants, and full scale
industrial waste operations. The data from these different
operations have indicated the potential of this activated
sludge system to operate at lower detention periods and
higher loading conditions (F/M and lb/BOD/1000 ft3) than
normally allowed by the state standards and still achieve
the desired removal efficiencies.
An effort was made in this research program to gather data
relative to the application of the homogeneous mixed system
to a municipal sewage wastewater flow. The results given in
other sections of this report show that the homogeneous mix-
ed system proved to be effective in treating the Freeport
wastewater flow at lower detention periods and higher load-
ing conditions than the design criteria established by the
State Agency.
The application of test results to the design of full scale
installations is the desire of the design engineer. The
initial design of the Freeport Plant was performed with very
little previous design criteria available for a homogeneous
mixed system. The completed plant design proved to be very
operable but, as in any initial endeavor,some modifications
in future designs are desirable. The following items describe
some modifications which would result in a more optimum
design.
Number of Air Piffusers Required
The original design for the air diffusion system varied for
each tank due to the flexibility required for the different
modes of operation. One tank had 833 diffusers, another 633
diffusers, both due to special test requirements, and the
remaining two tanks each had 408 diffusers. After careful
examination of the oxygen uptake data, air flow data and
transfer efficiency tests given in other sections of this
report, it appears that about 1,000 ft3 of air per pound
of 5-day BOD applied is needed to supply sufficient oxygen
in the system/Based upon the design value of 7,900 pounds
per day BOD applied to the aeration tanks, a total of 5,480
C'fm is the average air demand of the system. Each fine
bubble diffuser was designed to supply 6 cfm and therefore
177
-------
228 diffusers required in each tank.
The above design would satisfy the requirement for oxygen
demand, but in the paper entitled "Evaluation of Mixing
in Aeration Tanks" by E.D. Toerber, W.L. Paulson, S.C.
Mehta, and H.S. Smith, the following statement was made:
"A recommended absolute minimum density for good mixing
should be 20 diffusers per 100 square feet." This cri-
teria would result in 287 diffusers per tank.
To meet both the oxygen demand and mixing requirements of
the system, each 25 foot wide by 57.5 foot long tank should
be supplied with 290 diffusers resulting in a flow rate of
about 5 cfm each. This would reduce the total number of
diffusers in the four tanks from 2,282 to 1,160.
Another modification of the original diffuser design is
the location of the diffusers in the aeration tanks. It
was indicated in the dye tests (described in Section IV),
that more diffusers placed on the effluent half of the
aeration tank bottom resulted in more complete mixing of
the tank contents. The results of this diffuser arrange-
ment seems logical because more backmixing could improve
utilization of the tank volume. Further research should
be performed to confirm this point for other tank configura-
tions .
Blower Capacity
The blower range was not adequate during some periods of
plant operation. The blowers, with a range of about 3,500
to 6,000 cfm each, could not be operated at a low enough
output for periods of low oxygen demand, thereby wasting
some of the air produced during these periods. This problem
resulted from the much lower initial flows and 5-day BOD
loads that normally occur during early stages of design
periods. As previously explained, only about 5,480 cfm of
air is required for the aeration system. In a new design
three 2,750 cfm blowers would be installed with one of the
blowers as a stand-by. This arrangement would provide a
range of blower output between 1,500 and 8.250 cfin, which
would result in enough flexibility to cope with the diff-
erent loading conditions.
Instrumentation to Regulate Blowers
One of the major deficiencies of many sewage treatment
plants is lack of adequate instrumentation for optimum
operation. Part of this over-all instrumentation defi-
ciency is in the lack of automatic adjustment of blower
output to meet oxygen demand. Figure 50 illustrates a
typical diurnal BOD load to the Freeport. sewage treatment
178
-------
plant.
As indicated in Figure 50, the BOD load to the treatment
system will have large fluctuations during the course of
the day. Therefore, in a biological treatment system the
microorganisms will exert similar fluctuations in their
demand for oxygen while degrading the organic material.
To properly regulate the air supply, it is customary at
most plants to periodically measure the dissolved oxygen
at the effluent from the aeration tanks and to adjust
the blower output in an attempt to maintain the desired
dissolved oxygen concentration. Recent technological
advances have produced dissolved oxygen probes which
can, with proper maintenance,give continuous and instan-
taneous readings which can be converted to control signals
for automatic adjustment of blower output. This system
enables the blower output to be optimized by maintaining
a set dissolved oxygen level in the aeration tanks and to
prevent excessive air output. Further research in this
area should be pursued to provide the operators with a
tool for improved plant operation.
Spacing of Inlets to Aeration Tanks
Considerable discussion has occurred concerning the proper
spacing of wastewater and return sludge inlets into an aer-
ation tank which would result in a homogeneous mixed system.
Part of the present study dealt with mixing regimes existing
in the aeration tanks. The mixing regimes were studied util-
izing a rhodamine B dye test. The results, which are summar-
ized in another section of this report, indicate that two
inlet ports each for primary effluent and return activated
sludge for each fifty feet of tank length are sufficient.
The inlet ports should be located symmetrically on one side
of the tank.
The inlet ports for the primary effluent exhibited a solids
settling problem. Future designs should maintain enough
velocity in the inlet ports to prevent excessive settling
of the solids.
Optimum Detention Period in Aeration Tanks
The "Ten State Standards" state that a conventional bio-
logical treatment system should be designed for a loading
of 40 pounds of 5-day BOD per 1,000 cubic feet of aeration
tank volume, which would result in a detention period of
5.25 hours for the Freeport design conditions. The existing
Freeport plant has exhibited good removals of soluble BOD
in as little time as one hour, as shown in Figure 35.
179
-------
After careful examination of the data and taking into con-
sideration the minimum detention time required in the aer-
ation tanks during high flow conditions, it can be stated
that a two hour detention period (based upon Q) for the
average design flow conditions (6.75 mgd) would be adequate
for design purposes.
Economical Evaluation
General
One of the major governing factors in the design of treat-
ment facilities is the construction cost. It is the responsi-
bility of the engineer to provide a design which will produce
the results required at the lowest possible construction
cost, at the same time considering the financing capability
of the client. If the engineer is able to design an optimum
system without being compelled to fully comply with conserv-
ative state standards, the client would definitely profit.
The following paragraphs describe: (a) an EPA cost estimat-
ing program; (b) the estimated construction cost of the
Freeport facility if designed in accordance with the cri-
teria developed during the test program; and (c) an esti-
mated construction cost of a treatment facility based upon
the Freeport design conditions and designed in accordance
with conventional design criteria.
Following the explanation of the basis for each final cost,
a comparison of the three cost results is given. This com-
parison takes into consideration the cost of the aeration
tanks, blowers, blower building, final sedimentation tanks,
return sludge pumps, associated piping, plumbing, heating,
ventilation, air conditioning, equipment, excavation, elec-
trical work, roadways and landscaping. The remainder of the
treatment facility units (primary treatment, sludge disposal,
disinfection) were not included because the units would be
similar for all designs.
EPA Cost Estimating Program
The U.S. Environmental Protection Agency has devised a
computer program (Wastewater Treatment Plant Cost Esti-
mating Program, by R. Eilers and R. Smith) which estimates
the construction costs of treatment plants. The basic data
utilized by the EPA in creating the computer program was
based upon past construction cost records kept by a major
consulting engineering firm. These costs were broken down
into the cost for each major process of the over-all treat-
ment facility, namely, aeration tanks, final sedimentation
tanks, etc. The construction cost for each orocess is
180
-------
related to flow, surface overflow rate, liquid volume, BOD5
concentration, etc., depending upon the major process under
consideration. For example, the cost of the final sedimen-
tation tanks is related to the surface overflow rate.
The final tank loading deserves special explanation. The
surface overflow rate chosen for purposes of comparison was
800 gpd/ft2. This was based on an engineering judgement by
Consoer, Townsend and Associates who did the design analy-
sis. This differs from the loadings recommended in Section
VII (Surface Overflow Rate = 700 gpd/ft2, Solids Loading =
25 Ib/ft2-day). The comparison does not take into account
the relatively low solids loading recommended. The type of
final settling tank will also affect the choice of allowable
loadings.
Based upon the above criteria, the following parameters were
inserted into the program:
BOD5 Removal in Aeration Tanks
MLVSS
Influent Dissolved Nitrogen Concentration
RSVSS/MLVSS Ratio
Aeration Detention Time
Diffuser Efficiency
Final Sedimentation Tank Overflow Rate
Average Daily Flow
Peak Flow
125 mg/1
2,000 mg/1
15 mg/1
4.2
2.0 hours
8%
800 gpd/ft*2
6.75 mgd
16.6 mgd
The computer program listed the following costs for each
unit:
Aeration Basin Structure $ 99,126
Aeration Diffused Air 332,208
Final Sedimentation (Multiple Basin) 207,759
Recirculation Pumping 94,339
Yardwork 102,683
Total Estimated Construction Cost...$ 836,115
The total construction cost is based upon an ENR Index of
1308.61.
Estimated Construction Cost, Homogeneous Plant
The actual construction costs of the following listed
existing units at the Freeport Plant were determined when
the Engineering News-Record Construction Cost Index was
about 1140.0:
181
-------
Tunnel 3 51,000
Blower Building 508,000
Aeration Tanks 328,000
Final Tanks 334,000
Outside and Miscellaneous Work 100,000
Total Construction Cost $1,321,000
Taking into consideration the design modifications pre-
viously described in this section of the report, namely,
size of the blowers, number of inlets, number of diffusers
and the size of the final tanks, the cost of the following
listed units would be modified as shown:
Blower Building $ 340,000
Aeration Tanks 258,000
Final Tanks 300,000
To the above values add the previous tunnel and miscell-
aneous work cost, and the resulting adjusted construction
cost is $1,049,000 at an ENR Index of 1140.0.
To have the same basis for comparison with the other con-
struction costs, the January, 1970 ENR Index of 1308.61
was chosen and the resulting construction cost is $1,200,000.
Estimated Construction Cost, Conventional
Design Plant
The "Ten State Standards" have established criteria for
design of sewage treatment facilities. Utilizing part of the
above criteria in designing a conventional biological system,
a construction cost estimate of the final design was made.
The criteria which affect the biological system are as
follows:
(a) 40 pounds of 5-day BOD per 1,000 cubic feet of
aeration capacity
(b) 1,500 cubic feet of air per pound of 5-day BOD
to the aeration tank
(c) 75% return rate capacity of the average design
flow
(d) 800 gallons per day per square foot of final
tank surface area
The average design flow is 6.75 mgd and the BOD load to
the aeration tanks is 7,900 pounds per day, as previously
stated in this section.
182
-------
Based upon the above criteria and design conditions, the
following capacities would be required:
(a) Aeration Tanks 197,000 cu ft
(b) Blower Capacity 8,200 cfm
Use 3-4, 120 cfm blowers with
one as stand-by
(c) Diffusers, @ 6cfm/diffuser 1,370
(d) Return Activated Sludge Pump Capacity 5.1 mgd
Use 4 - 1.3 mgd variable speed pumps
(e) Final Tanks 8,430 sq ft
The comparative cost associated with the equipment and
tankage for the above items based upon an ENR Index of
1308.61 is as follows;
Tunnel $ 60,000
Blower Building 520,000
Aeration Tanks 950,000
Final Tanks 450,000
Outside and Miscellaneous Work 300,OOP
Total Construction Cost $2,280,000
Comparison of Different Construction Costs
The following table lists the construction costs assoc-
iated with the three different economical evaluations of
the treatment facilities based upon the previously de-
scribed design criteria. All the costs are based upon an
ENR Index of 1308.61.
Economical Evaluation Method
EPA Cost Estimating Program $ 836,115
Estimated Construction Cost,
Homogeneous Plant 1,200,000
Estimated Construction Cost,
Conventional Design Plant 2,280,000
The standard to be used in the comparison is the esti-
mated construction cost of the facilities if designed
in accordance with the criteria developed during the
test program. The EPA cost estimating program was con-
siderably lower than the estimated construction cost.
It should be noted that the EPA program estimate can vary
by as much as ± 40% of the actual construction cost of a
particular plant, due to the many special considerations
that may arise in building a new plant. Cost variations
are due to differences of design, local construction site
183
-------
conditions, current and projected labor costs, climate and
seasonal factors, the economic condition at the time con-
struction bids are taken, and many other factors which are
not readily subject to evaluation on an average basis. The
EPA program is only intended to provide a preliminary con-
struction cost estimate for an average plant (strictly
speaking, no average plant really exists). Therefore, the
comparison between the two costs is within the intended
range of the computer program estimate.
This research program has proved the Freeport homogeneous
activated sludge plant to be an efficient wastewater treat-
ment facility which has cost the Freeport residents only
about one-half as much as a conventional plant as indicated
in the comparison previously given. The Freeport demonstra-
tion also strongly indicates that more latitude should be
allowed the designer in the application of design standards
established by state agencies or that the standards for ac-
tivated sludge treatment should be modified.
It is the intent of the engineer to design a facility that
satisfies the conditions at the lowest cost. This approach
will challenge the engineer's innovative ability because
many aspects must be considered in arriving at the final
design. If the engineer is hampered by too conservative
criteria, part of the incentive and challenge is removed.
Not only is the engineer affected adversely, but the overall
process efficiency suffers and the taxpayer bears the added
cost. Treatment efficiencies have suffered in some cases be-
cause over-designed facilities do not promote maximum per-
formance. Also the taxpayer is required to pay for extra
facilities which are not needed to obtain the desired re-
sults.
It is hoped that the results of this research program will
lead to more efficiently designed biological wastewater
treatment plants.
EPA Process Performance Simulations
The EPA has developed another computer program ( A general-
ized Computer Model for Steady-State Performance of the
Activated Sludge Process, by R. Smith and R. Eilers). This
program is a steady-state model of the conventional acti-
vated sludge process which can be used to simulate all
performance aspects of the process. This model has been
shown to fit measurement data from a wide range of process
models, including complete-mix, plug-flow, step aeration,
contact stabilization, extended aeration, etc.
This program was used to simulate some of the measurement
184
-------
COMPARISON OF ACTIVATED SLUDGE MEASUREMENTS AND MODEL COMPUTATIONS
TABLE 34
Parallel Operation
Complete-Mix Mode Complete-Mix Mode 7-1/7-31, 1971
4-24/5-9, 1971 1-1/1-31, 1971 Complete-Mix Mode Plug-Flow Mode
M
MLVSS, mg/1 3301
TBOD, ng/1
DBOB, mg/1
NH3, mg/1
Q15, mgd
Q17, mgd
AIR, scfm
M
M
01
33
14
16
02
90
-
3309
30
20
16
.03
.96
-
2501
19
14
20
.01
.70
-
2490
18
12
20
.03
.60
-
1166
31
16
14
.03
.62
1083
1192
16
12
14
.06
.52
830
1143
28
15
14
.02
.61
1539
1177
13
6
14
.05
.37
885
00
Parallel Operation
8-1/8-31, 1971
Complete-Mix Mode Plug-Flow Mode
Parallel Operation
9-1/9-30, 1971
Complete-Mix Mode Plug Flow Mode
MLVSS, mg/1
TBOD, mg/1
DBOD, mg/1
NH3, mg/1
Q15, mgd
Q17, mgd
AIR, scfm
M
1338
17
8
14
.03
.65
996
C
1321
12
7
14
.05
.65
625
M
1230
16
8
14
.02
.64
1551
C
1244
9
4
14
.05
.57
657
M
2111
19
9
14
.02
.74
1343
C
2125
13
8
14
.03
.57
772
M
1795
17
11
14
.01
.66
1477
C
1799
8
3
14
.03
.62
792
Note: Refer to page 186 for Key to all abbreviations
-------
data which was gathered on this project. A summary of this
effort appears in Table 34. Based on raw data averages over
a selected simulation period, several measured system para-
meters are chosen for comparison to predicted values from
the computer program. Note that several simulations of the
parallel operation data are given in Table 34. Based on
these results, it can be concluded that the model can
produce a fairly good simulation of the complete-mix mode
of operation, but the program consistently predicts that
better performance should have been obtained from the plug-
flow mode; which may indicate a deficiency in the activated
sludge computer model.
Below is a tabulation of the abbreviations used in Table 34:
MLVSS = mixed liquor volatile suspended solids in the
aerator, mg/1
TBOD = final effluent unfiltered 5-day BOD, mg/1
DBOD = final effluent filtered 5-day BOD, mg/1
NH3 = final effluent ammonia nitrogen, mg/1
Q15 s waste stream flow from the activated sludge
process, mgd
Q17 = return stream flow to the activated sludge
process, mgd
AIR = total minimum air requirement for the activated
sludge process, scfm
M - average measured value from the raw data
C = computed value from the program
186
-------
SECTION XI
A CKNOWLEDGEMENTS
Thanks must first go to Mr. Lawrence M. Madden who served
as project manager. His enthusiastic support of the project
engineer should be recognized. Along with Mr. Madden, the
Freeport Water and Sewer Commission are to be commended for
their farsightedness in authorizing construction of a plant
with enough flexibility to be adapted to the research study.
i
Mr. William Boll, Superintendent, provided invaluable
technical assistance, especially during the initial stages
of the project.
Mr. Charles Swanson, who served as project officer for the
first part of the project, made several key recommendations
as the work progressed and was instrumental in initiating
the program. Mr. Richard Eilers, who succeeded Mr, Swanson
as project officer, was very helpful in doing a number of
computer evaluations of the operational data.
The research consultants, Dr. H.S. Smith and Dr. Wayne L.
Paulson, provided the majority of the ideas for the plan of
study conducted, reviewed the final report and made numerous
revisions to improve this report.
The operating personnel at the Freeport Wastewater Treatment
Plant should be acknowledged for the analytical and operation-
al work that they performed. The following were the key per-
sons :
Mr. Joel McCulloch, chemist
Mr. Glenn Kelly, chief operator
Mr. Michael Ross, lab technician
Mr. Ralph Sager, technician
Mr. Cy Crcthers, maintenance operator
Mr. Donald Maher, maintenance operator
Mr. Suman Mehta was in charge of the mixing study phase of
the project.
Mr. Bharat Mathur was in charge of the pilot plant study
and wrote this section of the final report.
The engineers from Consoer, Townsend and Associates should
be recognized. Mr. Fred Van Kirk served as senior member on
the project technical committee. Mr. Ed Davel was the process
design engineer for the plant construction. Mr. Robert Bella
wrote the design analysis section of the final report.
187
-------
Finally, much thanks should go to Mrs. Lila Bruce, secretary,
and Mr. John Johnson, draftsman, for the large volume of
quality work which they did to insure that the report was
finished on schedule.
188
-------
SECTION XII
REFERENCES
1. Wilson, James, "Fluororaetric Procedures for Dye Tracing",
Department of Interior Techniques of Water Resources
Investigations of the United States Geological Survey,
Book 3, Chapter A-12.
2. Butts, Thomas A.. "Fluorometer Calibration Curves and
Nomographs", Journal Sanitary Engineering Division,
ASCE, Vol. 95, SA-4, page 705, August 1969.
3. Levenspiel, O., "Nonideal Flow", Chemical Reaction
Engineering, John Wiley and Sons Inc., New York, 1962, Ch 9
4. Timpany, Peter, "Variations of Axial Mixing in an
Aeration Tank", Master's Degree Thesis, Department
of Civil Engineering, McMaster University, 1966.
5. Fair, G.M., Geyer, J.C. and Okun, D.A., "Treatment
Kinetics", Water and Wastewater Engineering: Vol. 1.
Water Supply and Wastewater Disposal, Ch. 22, John Wiley
and Sons,Inc.,New York,1968.
6. Murphy, Keith L., and Boyko, Boris, Longitudinal Mixing
in Spiral Flow Aeration Tanks, paper presented at the
Second National Symposium on Sanitary Engineering Re-
search, Developement and Design, Sanitary Engineering
Division ASCE, Cornell University, 1969.
7. Smith, H.S., "Oxygen Uptake in a Completely Mixed Acti-
vated Sludge System", PhD Dissertation, Iowa State Un-
iversity, Ames Iowa, 1963.
8. Kothandaraman, V., and Evans, Ralph L., "Hydraulic
Model Studies of Chlorine Contact Tanks", paper at
44th Annual Meeting, Central States Water Pollution
Control Association, June 1971.
9. Smith, H.S., personal correspondence to E.D. Toerber,
May 1970.
10. McKinney, Ross E. , "Mathematics of Complete-Mixing
Activated Sludge", ASCE J SEP No. SA 3, pp 87-113,
May 1962.
11. Eckenfelder, W.W., and Ford, D.L., Water Pollution
Control, Pemberton Press, New York, 1970.
189
-------
12. Conway, R.A., and Kumke, G.W., "Field Techniques for
Evaluating Aerators", Journal Sanitary Engineering
Division, ASCE, pp 21-41, April 1966.
13. Morgan, Philip, andBewtra, J.K.,"Air Diffuser Effic-
iencies", Journal Water Pollution Control Federation,
pp 1047-1059, October 1960.
14. i Standard Methods .for the Examination of Water & Waste-
water, 12th Edition, 1965. APHA, AWWA, WPCF.
15. Wallace, D.A., "Effect of Mixed Liquor Suspended Solids
Concentration on Performance of a Completely Mixed
Activated Sludge System with Variable Reaction Times",
Master's Thesis University of Iowa, Iowa City, Iowa,
1968.
16. Frazier, G. J. , "Study of Active Mass in a Homogeineous
Activated Sludre System", Master's Thesis, University
of Iowa, lowas City, Iowa, 19SS.
190
-------
SECTION XIII
APPENDIX
Oxygen Uptake Tests
A schematic diagram of the oxygen uptake test apparatus
shown in Figure 81 (This figure is appropriately labelled
and indicates the apparatus involved). Initially, the YSI
oxygen probe is calibrated. In this investigation it was
calibrated against final effluent. The following steps are
carried out:
1. Pour 1100 cc tap water in a utilometer and aerate.
2. Attach separatory funnel to utilometer.
3. When sufficiently high DO is read on the recorder;
start stop watch and collect sample (mixed liquor in
this case). This is stop watch zero (ti).
4. The time at which the sample is withdrawn from the
aeration tank is noted (t£)• This is zero for the sample
(mixed liquor in this case).
5. Pour 1500 ml sample into the utilometer. Save a portion
for solids determination.
6. Seal the utilometer and note the temperature.
7. Allow oxygen consumption until DO drops below 1 mg/1.
8. Reaerate to a sufficiently high DO and reseal.
9. Repeat 7 and 8 and run experiment for 24 hours.
191
-------
Automatic
Temperature
Control
V
To YSI Oxygen Meter
Heater
',0
CO
Water Bath
Thermometer
BODProbe-
Magnetic •
Stirrers
OXYGEN UPTAKE TEST APPARATUS
Figure 81
YSI
Oz Analyser
Strip Chart
Recorder
-------
Calculation of Uptake
From the strip chart the following parameters are recorded:
1. Slope of line between time interval t' - t"
2. Time interval t1 - t"
3. Time t = t' + t' - t"
2
Temperature and calibration corrections are applied to give
the uptake in mg/l/hr at 20° C.
A curve is plotted between oxygen uptake do/dt in mg/l/hr
and t. Refer to Figure 82 for a typical curve. The straight
line portion of the curve is extended to meet t - 0, in
this case at do/dt = 13.5 mg/l/hr. The slope of the straight
line is determined as follows;
t = 0, for do/dt =13.5
t = 10 hours, for do/dt - 9.9
then K2 = log 9.9 - log 13.5 = 0.029hrs~1
10
then Ma in oxygen units = 13. 5 = 450 mg/1
.0296
or Ma in mass units = 450 = 317 mg/1
Since VSS was determined to be 1190 mg/1
Ma = 0.26
VSS
The difference between the curved portion of the plot and
the straight line is then plotted for a convenient time in-
terval. The slope of this line is determined as before to
be - 2.67 hrs"1, which is K9. Then 7.7 = F in mg/1 . F is
ultimate BOD. Kg
Oxygen Transfer Efficiency Test
This test was conducted by using the pilot plant aeration
tank. The following steps detail the experimental procedure:
1. Fill tank with sample in which transfer efficiency
is desired and choose air flow rate.
2. Calibrate DO probe and immerse in sample to get
continuous DO monitoring.
3. Record the DO concentration at saturation, Cs and
also the liquid temperature.
4. Introduce a pre-determined amount of sodium sulphite
mixed with cobalt chloride catalyst ( to give a COC12
concentration in the tank of 1.5 mg/1). Start the
stop watch. This is time zero. If appropriate
193
-------
quantities of sulphite have been added the DO should
go to zero.
5. When the DO begins to increase, record the time in
minutes and Ct, the DO corresponding to that time,
intil DO reaches 80% of saturation value.
The following steos are involved in the calculation of the
transfer efficiency:
1. Plot CS/(CS - Ct) versus time on a semi-log paper.
2. Find slope of curve = K at temperature of liquid,
say KT-
3. Correct KT for 20° C, where K20 = KT(1.024) ^U~T
Then rate of oxygen consumption dc - Cs x K20 x ®0
my/l/hr. "cTt
4. Convert air supply rate to in^/l/hr. Then transfer
efficiency = rate of oxygen consumption _
rate of air supply
oc , the oxygen transfer coefficient = K(waste water)
K(tap water)
194
-------
5C
A2
Ma
VSS
-1
= 0.0296 hours
= 0.26
-1
JC = 2.67 hours
F = 3 mg/1
1
1
200 400 600 8CO 1000 1200 1400
Time in Minutes, t
Figure 82. Qxygvn u;.take, Run Mo. 7/1/PP
-------
Biochemical Oxygen Demand Test
1. Apparatus
1.1 Incubation bottles, 300 ml capacity with ground glass
stoppers. Bottles should be cleaned with a good detergent
and thoroughly rinsed and drained before use. The BOD
bottles should have flared mouths so that an air tight
water seal can be provided during incubation.
1.2 Air incubator thermostatically controlled at 20° C±
1° C. All light should be excluded to prevent formation
of DO by algae in the sample.
1.3 Yellow Springs Instrument Company, Model 54 oxygen
meter.
1.4 Y.S.I. Model S420A self stirring BOD bottle oxygen
probe. The probe is ^especially designed for use wit'i
standard BOD bottles. The probe is provided with an agi-
tator for stirring of the sample solution at a constant
rate.
2. Reagents
2.1 Distilled water for preparation of dilution water.
This water must contain less than 0.01 mg/1 copper and
be free of chlorine, choloramines, caustic alkalinity,
organic materials or acids.
2.2 Phosphate buffer solution: Standard Methods - 12th
Edition
2.3 Magnesium sulfate solution: Standard Methods - 12th
Edition.
2.4 Calcium chloride solution: Standard Methods - 12th
Edition.
2.5 Ferric chloride solution: Standard Methods - 12th
Edition.
2.6 Seeding material: 2 liters of mixed liquor from the
aeration tank should be placed into a glass container
and continually aerated. Daily 500 ml of mixed liquor
should be drawn from the aeration container. Place mixed
liquor into 1 liter graduate cylinder and allow solids
to settle. 10 ml of clear liquor should then be drawn
off this solution and diluted with distilled water to
400 ml. The 500 ml of mixed liquor should be replaced
196
-------
by adding 500 ml of primary- effluent to the aeration
container. A solid level of no less than 100 ml/1
should be maintained in the aeration container. This
level may be maintained by adding return sludge or
mixed liquor to replace the 500 ml portion that was
drawn from the aeration container.
3. Procedure
3.1 Preparation of dilution water: The distilled water
that is used is aerated with clean compressed air for a
sufficient length of time to become saturated with DO.
The distilled water should be as near 20° C as possible.
The distilled water is placed in a 2.5 gallon bottle and
10 ml each of phosphate buffer, magnesium sulfate, cal-
cium chloride and ferric chloride solutions are added .
3.2 Seeding: 2 ml of seed as prepared in paragraph 2.6
should be added to each BOD bottle.
3.3 Pre-treatment of sample:
a) Pasteurization: samples should be heated to 55°
C for 15 minutes and allowed to cool to room
temperature before use (only final effluent
samples are pasteurized) .
b) Filtering: filtered samples should be filtered
through a glass fiber filter (Gelamn Type A,
part 61593). For comparison, both filtered and
unfiltered samples are run simultaneously.
3.4 Dilution Technique: raw sewage is diluted by placing
10 ml of sample by pipette into a BOD bottle half filled
with dilution water. The same dilution procedure is used
for primary effluent (filtered and unfiltered) . Final
effluent unfiltered is diluted by placing 25 ml of sample
by pipette into a BOD bottle half filled with dilution
water and then filling the BOD bottle to the top. 50 ml
of sample is used for final effluent filtered.
4. Calibration of Oxygen Meter and Probe
4.1 Two BOD bottles should be filled with dilution water.
The water should be added by a siphon with it's outlet
held below water level to prevent further aeration.
4.2 The DO of the first bottle should then be determined
by use of the Winkler test for dissolved oxygen.
4.3 Place the YSI probe into the second BOD bottle and
calibrate meter to the DO reading found by the results
197
-------
of the Winkler test of the first BOD bottle.
4.4 A Winkler test should then be run on the second BOD
bottle. If and only if the results of the two Winkler
tests are the same, are the YSI probe and meter accur-
ately calibrated.
5. Use of Oxygen Meter and Probe
5.1 Initial DO: BOD bottle should be prepared with seed
and sample as described in paragraphs 3.2,3.3, and 3.4.
The probe should be inserted into bottle and a DO read-
ing obtained. The sample is then incubated in the BOD
bottle for five days at 20°C.
5.2 Final DO: After five days of incubation the BOD
bottle should be removed from the incubator, and the
DO determined by the YSI meter and probe.
6. Calculation of BOD
BOD = (I-F) x 300
I = Initial DO
F = Final DO
P - Sample size in ml
A blank is run with each bottle of dilution water to
determine any depletion in DO of the dilution water.
198
-------
Chemical Oxygen Demand Test
1. Apparatus
1.1 Refluxing apparatus, 250 ml erlenmeyer flasks with
ground glass 24/40 neck and 300 mm jacket, liebeg con-
densers with 24/40 ground glass joint, electric hot
plate extraction heater (6 burner) producing 10.3 watts/
sq in of heating surface to provide adequate boiling of
contents of the refluxing flask.
2. Reagents
2.1 Stock potassium dichromate 0.250 N with 0.129 of
sulfamic acid added to prevent nitrite interference.
2.2 Standard potassium dichromate solution 0.10N
made by diluting 400 ml of stock to 1000 ml.
2.3 Sulfuric acid reagent, cone. KoS04 containing 22 g
silver sulfate Ap-2 ^04, per 9-lb bottle.
2.4 Stock ferrous ammonium sulfate solution 0.10N.
Dissolve 39g Fe (NH4) 2 (804) 2 • BI^O in distilled water.
Add 20 ml cone. E^SO^ cool, and dilute to 1000 ml.
2.5 Standard ferrous ammonium sulfate titrant 0.04N.
Made by diluting 400 ml of stock solution to 1000 ml.
This solution must be standarized daily against the
standard potassium dichromatic solution.
Standardization; Dilute 10.0 ml standard potassium di-
chromate solution to about 100 ml. Add 30 ml cone. H2S04
and allow to cool. Titrate with the ferrous ammonium
sulfate titrant, using 3 drops of ferroin indicator.
N = ml K9 Cr2 °7 x 0.25
ml Fe (NH4) 2 (S04) 2
2.6 Ferroin indicator solution: Fisher Scientific Co.
# P-69, 1, 10 Phenanthroline ferrous sulfate complex.
2.7 Silver sulfate reagent powder.
2.8 Mercuric sulfate, analytical grade crystals.
2.9 Sulfamic acid, analytical grade.
Procedure
3.1 To each ;,lass refluxing flask (blank and sample
199
-------
flasks) add 0.4g mercuric sulfate. This is measured with
a Hach Company #638 measuring spoon,
3.2 Add 20.0 ml boiled distilled water to the blank and
20.0 ml of sample to the sample flasks. Filtered and
unfiltered sample are used. Filtered samples are fil-
tered through a Gelman Type A glass fiber filter, Part
#61693.
3.3 Add exactly 10.0 ml standard potassium dichromate
0.10 N solution to each flask.
3.4 Carefully add to each flask 30 ml of the concen-
trated sulfuric acid containing the dissolved silver
sulfate and thoroughly mix by swirling.
3.5 Add 10 to 15 3mm glass boiling beads.
4. Attach condenser and reflux both blank and sample for
2 hours.
5. Cool and wash down the condenser with 80 ml of distilled
water. This should give a total volume of about 140 ml.
This volume is necessary to allow the indicator to func-
tion properly.
6. After sample is cooled to room temperature and diluted,
add 2 or 3 drops of ferrion indicator to each flask. Use
the same number of drops for each flask.
7. Titrate each flask with the standarized ferrous ammonium
sulfate. The color change is sharp, going from blue-green
to reddish-brown to indicate the end point.
8. Calculation:
COD, mg/1 = (a-b)c x 8,000
ml of sample
a = ml ferrous ammonium sulfate solution used for
blank
b = ml ferrous ammonium sulfate solution used for
blank
c - Normality of ferrous ammonium sulfate solution
200
-------
Mixed Liquor Suspended Solids & Volatile Suspended Solids Test
1. Apparatus
1.1 Gooch crucible
1.2 Gelman #61693 glass fiber filter, type A, 1 inch
1.3 Vacuum filter flask
2. Procedure
2.1 A sample of 1 liter volume should be taken
2.2 The crucible and glass fiber filter should be
oven dried at 103° C to 105°c for one hour. Allow
to cool in desicator and weigh.
2.3 Sample should be thoroughly mixed and 10 ml of
sample placed in tared crucible under vacuum. Leaving
vacuum applied, rinse the measuring pipette with 10 ml
distilled water and filter through glass fiber filter.
2.4 Dry crucible and solids for two hours at 105°C.
Allow to cool in desiccator and weigh.
2.5 For volatile solids, ignite the crucible for 15
minutes at 600°C in a muffle furnace. Cool to room
temperature, partially in air and finally in a desiccator,
and weigh.
3. Calculation:
rag/liter total suspended solids =
mg weight of dried solids x 1000
ml of sample
mg/liter volatile suspended solids =
mg weight of volatile solids x 1000
ml of sample
The volatile solids are equal to the dried total solids
minus the residue upon ignition.
4. Reporting
Report as mg/1 suspended solids and mg/1 volatile
suspended solids.
201
-------
Ammonia Nitrogen Test
1. Apparatus
1.1 1000 ml Florence flask, used for distillation
1.2 Vertical condenser (glass)
1.3 Glass tubing, stoppers
1.4 Electric heater
1.5 500 ml Erlenmeyer flask, used for sample collection
1.6 Spectrophotoraeter for use at 425 nm and providing
a lightpatch of 1 cm or more
1.7 Glass boiling beads
1.8 Nessler tubes
1.9 5 ml and 20 ml volumetric pipettes
2. Reagents
2.1 Ammonia free distilled water
2.2 Rochelle salt solution
2.3 Stock ammonium solution, 1.00 ml - 1.00 mgN
2.4 Standard ammonium solution, 1.00 ml - 0.01 mgN
2.5 Nessler reagent (Fisher SO-N-16)
2.6 Phosphate buffer solution
3. Procedure
3.1 400 ml of sample should be placed in Florence flask
(1000 ml) and 25 ml of phosphate buffer solution added
to the sample.
3.2 Distillation: distill 200 ml at the rate of 3-5
ml/min into 500 ml Erlenmeyer flask.
3.3 Pipette 5 ml of distillate into 50 ml Nessler tube.
Dilute sample to 50 ml with ammonia free distilled water,
add two drops of Rochelle salt and 1 ml of prepared
Nessler reagent.
202
-------
3.4 Color Developement: Allow 10 minutes for color
developeinent. Read the percent transmittance at 425 nm
against the blank. From values obtained, find concentra-
tion (mg/1) on standard curve, transmittance vs. concen-
tration.
3.4-1 Standard curve preparation. Prepare a series of
standards in the following concentrations: 0.0, 0.1,
0.3, 0.5, 0.8, 1.0 mg/1. Nesslerize standards and
allow 10 minutes for color development. Read trans-
mittance at 425 nm. From values obtained, plot percent
transmittance against concentration for the standard
curve.
4. Calculation
Ammonia N, mg/1 = (50) A x B
nl sample "C"
A - rag/1 N found colorimetrically
B - ml total distillate collected
C - ml distillate taken for nesslerization
OU.S. GOVERNMENT PRINTING OFFICE: 1972 514-149/102
-------
SELECTED WATER i. Report No.
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
3. Accession No.
w
Full Scale Parallel Activated Sludge Process Si ReP°rtDat<> Jan., 1972
Evaluation 6-
. *• Performing Organization
7. Autho.r(sl , . Report No.
Erwin a. Toerber, H. S. Smith, W. L. Paulson,
Bharat Mathur, Robert Bella
9. Organization ' 1705° ENM
Freeport Water and Sewer Commission
10. Project No.
11. Contract/Grant No.
230 West Stephenson Street
Freeport, Illinois 61032
r > 13. Type of Report and
Period Covered
12. Sponsoring Organization Environmental Protection Agency, Office of Research and
15. Supplementary Notes Monitoring
Environmental Protection Agency report
number EPA-R2-72-065, November 1972.
16. Abstract
A comparison was made between parallel activated sludge systems operating
under completely-mixed and plug-flow modes.
Initially, a rhodamine dye tracer study was conducted to determine conditions
necessary to achieve the two operational modes.
The completely-mixed system was operated at 5 constant detention times ranginc
from 5 hours to 1 hour. The break in treatment efficiency (a marked drop be-
low 90% removal of soluble BOD5) occurred between 1 and 2 hours.
No marked difference in treatment efficiency was found between_the two modes
during 5 months of the parallel operation. Two methods of solids production
analysis were applied to this data. A series of oxygen uptake tests were
performed to evaluate air requirements.
Finally, a set of shock loads were applied in parallel and separately to
each mode. The complete-mix system did show an advantage over plug-flow
under shock load conditions at a short detention time (1.5 hours).
A 500 gpd completely-mixed pilot plant was run in parallel with the full
scale system for 4 months. It was most successful in duplicating the
full scale organic removal efficiency.
17a. Descriptors
*Activated Sludge, *Biological Treatment, *Secondary Treatment, Capital Costs.
Water Pollution Control, Wastewater Treatment, Sewage Treatment, Computer
Simulation
17b. Identifiers
*Complete-mix, *Plug-Flow, *Tracer Tests, *Shock Loads, *Freeport Illinois,
Treatment Plant Design
Hc.COWRR Field & Group 05D
*• Availability 19. Security Class.
(Report}
20. Security Class.
(Page)
21. No. of Send To:
Pages
PriVa WATER RESOURCES SCIENTIFIC INFORMATION CENTER
flice ,j s DEPARTMENT OF THE INTERIOR
WASHINGTON, D. C. 20240
^Abstractor Erwin D. Toerber |Institution Freeport Water & Sewer Commission
WRSIC 102 (REV. JUNE 1971) GPO 913.281
------- |