EPA-600/2-77-142
September 1977
Environmental Protection Technology Series
TOG, ATP AND RESPIRATION RATE AS
CONTROL PARAMETERS FOR THE
ACTIVATED SLUDGE PROCESS
Municipal Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-77-142
September 1977
TOC, ATP AND RESPIRATION RATE AS CONTROL PARAMETERS
FOR THE ACTIVATED SLUDGE PROCESS
by
Clarence Ortman and Tom Laib
City of Hillsboro
Sewage Treatment Plant
Hillsboro, Oregon 97123
and
C. S. Zickefoose
Stevens, Thompson & Runyan, Inc.
Portland, Oregon 97202
Grant No. R 802983-01-1
Project Officer
Joseph F. Roesler
Wastewater Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental Research Labo-
ratory, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views
and policies of the U.S. Environmental Protection Agency, nor does mention
of trade names or commercial products constitute endorsement or recommenda-
tion for use.
n
-------
FOREWORD
The Environmental Protection Agency was created because of increasing public
and government concern about the dangers of pollution to the health and wel-
fare of the American people. Noxious air, foul water, and spoiled land are
tragic testimony to the deterioration of our natural environment. The com-
plexity of that environment and the interplay between its components require
a concentrated and integrated attack on the problem.
Research and development is that necessary first step in problem solution and
it involves defining theproblem, measuring its impact, and searching for sol-
utions. The Municipal Environmental Research Laboratory develops new and im-
proved technology and systems for the prevention, treatment, and management
of wastewater and solid and hazardous waste pollutant discharges from munici-
pal and community sources, for the preservation and treatment of public drink-
ing water supplies, and to minimize the adverse economic, social, health, and
aesthetic effects of pollution. This publication is one of the products of
that research; a most vital communications link between the researcher and
the user community.
Several control strategies were evaluated during the course of this study to
demonstrate the benefits of automatic control of wastewater treatment plants.
Improved control of wastewater treatment plants increases plant reliability
and thus reduces environmental contamination.
Francis T. Mayo, Director
Municipal Environmental Research
Laboratory
m
-------
ABSTRACT
This research was conducted to determine the feasibility of using TOC (Total
Organic Carbon), ATP (Adenosine Triphosphate), and respiration rates as tools
for controlling a complete mix activated sludge plant handling a significant
amount of industrial waste. Methods were developed for determination of ATP
in sludge with MLSS concentrations up to 10,000 mg/1 using a JRB analyzer.
Control methodology was centered on using a modified F/M ratio. This param-
eter was determined by calculating aerator organic loading based on wastewater
TOC concentration and microorganism concentration and/or activity by TOC and
ATP of the activated sludge. Process control decisions were based on 5 to 7
determinations per day.
Respiration rates were used to indicate the need for increased or decreased
sludge aeration time. When respiration rates were held between 8 to 20 mg
02/g/hr, and other parameters were in an optimum range, regulatory permit
requirements were met.
Process control decision making was aided by the use of a programmable cal-
culator. Process control information was set up so that operators could
input plant data and receive printed out instructions for process settings.
Functional programs included return rates, mode changes, wasting rates, res-
piration rate, and corrected settlometer volume.
This report was suomitted in fulfillment of Grant No. R 802983-01-1 by
the City of Hillsboro under the sponsorship of the U.S. Environmental
Protection Agency. Work was subcontracted to Stevens, Thompson & Runyan,
Inc. This report covers a period from July 25, 1974 to July 24, 1975.
IV
-------
CONTENTS
Foreword ............................ iii
Abstract ................... !!.!!!!!! iv
Figures ................... !!!!!!!!!! vi
Tables ........................... [ j Vii
Acknowledgments ......................... vjii
1. Introduction ...................... 1
Background Information ............... 1
Operation Problems ................. 1
2. Conclusions ....................... 3
3. Recommendations ..................... 5
4. Nature and Scope of the Study .............. 6
Introduction .................... 6
Plant Description .................. 6
Industrial Waste .................. 9
Control Elements .................. 9
Sludge Conditioning Time .............. 11
Phases of the Project ................ 14
Summary of Phases .................. 17
5. Discussion of Results .................. 20
Application of TX ................. 20
Application of ATP ................. 21
Application of Respiration Rates .......... 26
6. Operation Procedures .................. 28
Use of Programmable Calculator During the Study. . . 28
Mode Change Criteria ................ 29
Wasting Criteria .................. 32
RAS Flow Control Criteria .............. 34
7- Experimental Procedures ................. 36
TOC Analysis .................... 36
ATP Analysis .................... 38
Respiration Rate Analysis .............. 46
Respiration Rate Determination ........... 48
Corrected Settlometer Test ............. 48
8. Discussion of Interrelated Parameters .......... 50
Introduction .................... 50
The May-June 1975 Problem .............. 51
Discussion ..................... 55
Summary ....................... 56
Glossary ............................ 57
References ........................... 59
Appendix ............................ 60
-------
FIGURES
Number page
1 Plant layout 7
2 Aerator-clarifier relationship 8
3 Mode change description 12
4 Process control schematic 13
5 Relationship of RAS-TOC to wasting time 22
6 Line of best fit for RAS ATP/TOC correlation 24
7 Relationship of RAS-SS to ATP and TOC 2b
8 Respiration rate - modified F/M relationship 35
9 ATP standardization curves 45
10 Mode 0-2 51
11 Mode 0-4 52
12 Mode 2-2 53
13 Mode 0-2 and 3-1 53
VI
-------
TABLES
Number Page
1 Industrial Waste Load 9
2 Phase Identification 14
3 Sample Collection Times 15
4 Major Information Recorded 15
5 Phase I Range and Average Data 17
6 Effects of Digester Supernatant on Process 18
7 Phase II Range and Average Data 19
8 Phase III Range and Average Data 19
9 TOC Control Comparison 20
10 Mode Number Definition 30
11 Mode Change Instructions 10/21/74 to 7/22/75 31
12 Aeration Mode Change Direction for Calculator Program .... 32
13 Wasting Criteria 33
14 Mode Shift Frequencies 54
vn
-------
ACKNOWLEDGMENTS
We wish to acknowledge the laboratory effort of Mr. Jerry Schulz, Lab Tech-
nician. Mr. Schulz was mainly responsible for the ATP analysis phase of the
project. The data could not have been collected without the support of the
operation personnel at the plant and city administration for the City of
Hillsboro. The assistance given by Mr. Joseph F. Roesler of the U.S. Environ-
mental Protection Agency is gratefully acknowledged.
viii
-------
SECTION 1
INTRODUCTION
BACKGROUND INFORMATION
The Hillsboro, Oregon westside plant is a complete mix activated sludge plant
handling industrial waste from a dog food processing plant, a vegetable and
fruit juice processing plant, and a meat processing plant, in addition to
domestic waste.
Problems at the Westside plant were evident soon after operation began in
that the effluent quality would not meet effluent standards set by the Oregon
Department of Environmental Quality (DEQ). The cause of the early operating
difficulty was traced to large and random variations in plant loading. At
times, the incoming organic loading varied as much as 500 percent due pri-
marily to industrial wastes. As a consequence of this variation, the acti-
vated sludge process was in a nearly constant upset condition.
The difficulties were overcome by a program with three major elements:
1. Financial and regulatory measures were initiated to reduce industrial
waste loading. A stronger industrial waste ordinance was enacted
mandating charges in proportion to the amount of load imposed on the
plant. A concentrated sampling program provided the basis for measur-
ing loading.
2. Operation control was improved by fixing responsibility for plant per-
formance with specific individuals during each shift.
3. A plant process control program using F/M by TOC and respiration rate
to define aerator feed points was initiated. This was done to enable
plant operators to implement changes rapidly enough to improve or
maintain effluent quality with wide variations in waste loading.
OPERATION PROBLEMS
Most of the operation problems that occurred at the Westside plant were re-
lated to the industrial load. The problem was identified as erratic loadings
due to soluble organic material which caused frequent upsets in the secondary
portion of the plant. These shock loads intensified due to a short flow time
between the industry's effluent line and the primary clarifier.
-------
The first method employed was to correct the problem using Primary Clarifier
No. 2 as a holding tank to reduce some of the peaks. Other remedies that were
tried were: utilizing chemical flocculents, variations in DO, dilution, main-
taining a constant F/M by the moving average method and upgrading the pre-
treatment of industrial wastes. None of these schemes solved this problem,
but did indicate that the proper solution required an immediate indication of
organic loading. This led to the use of the TOC (Total Organic Carbon) analy-
sis as a rapid method of determining changes in the organic content of the
primary effluent and final effluent, and the microorganism content of the
mixed liquor and return activated sludge. These data together with results
from a Mailory settlometer, and respiration rates (oxygen uptake) were the
parameters that were used in the control strategy development and evolution.
-------
SECTION 2
CONCLUSIONS
1. The use of TOC information for plant control using samples of primary
effluent and return sludge was particularly effective in plant control.
Loading variations from industrial sources requires the rapid control
response possible using TOC.
2. The ATP procedure allows very little variation to obtain repeatable re-
sults, therefore, compared to TOC is not as useful as a process tool.
3. "Modified" F/M described as the F/M ratio of primary effluent (PE) TOC
entering the aeration basin to return sludge (RAS) TOC entering the aera-
tion basin was found to be an effective process control parameter.
4. Aerator effluent respiration rate (AE-RR) was used to determine at which
point return sludge and/or primary effluent should be added to the aera-
tion basin. Unadjusted high AE-RR (>20) for short periods of time re-
sulted in poor effluent. To reduce AE-RR, the proportion of the aeration
capacity used for sludge reaeration was increased when in the contact
stabilization mode. When in plug flow, the number of quadrants for aerat
ing mixed liquor was increased.
5. AE-RR is controlled in conjunction with mode changes by RAS rate change.
Increased RAS will result in a decrease in AE-RR.
6. The use of the 5-minute corrected settlometer test (CSVs) is a valid
measure of sludge quality. High values (800-1000 ml /I) indicate a "young,"
slow-settling, active biomass. Conversely, low values (200-300 ml /I)
indicate "matured," fast-settling, but less active and responsive, bio-
mass. A CSVs of 400 to 500 ml /I is an optimum value.
7. State standards for effluent quality were usually met when both of the
following conditions existed:
a. The CSVs is maintained between 400 and 500 ml/1.
b. The AE-RR is maintained between 8.0 and 19.0 mg 02/g/hr.
-------
8. The settling rate observed by the CSV5 can be controlled on a long-range
basis by a calculator sludge wasting program which uses RAS TOC and five-
minute CSV5 as inputs. A tendency for the settling rate to increase with
increasing sludge age was observed. Significant changes occur in days to
weeks. That tendency is opposed by one or both of the following:
a. A relative increase in the organic loading rate.
b. A relative increase in the sludge wasting rate.
-------
SECTION 3
RECOMMENDATIONS
1. Investigate transferrability of the developed process control procedures
to another treatment plant.
2. Pursue the possibility of automating existing control loops and relate
this to economic feasibility.
3. Investigate feasibility using tighter process control to meet impending
stringent permit requirements with the goal of forestalling plant
expansion.
4. Investigate potential use of ATP to predict changes in the bacteria which
lead to filamentous bulking.
5. Investigate methods of automating the respiration rate determination at
the Hillsboro plant.
6. Find a means to submit the existing data collected during the course of
this study to computer programming for statistical analysis.
-------
SECTION 4
NATURE AND SCOPE OF THE STUDY
INTRODUCTION
The objective of the study was to investigate, evaluate, and demonstrate a
new practical method of process control of an erratically loaded activated
sludge sewage treatment plant. Three variables, Total Organic Carbon (TOO,
Adenosine Tri-khosphate (ATP), and Respiration Rate (RR), have been suggested
for use in control of a wastewater treatment plant. The use of each of these
three control variables was demonstrated at the City of HillsDoro Westside
Treatment Plant. Simultaneously, the use of an adjustable aerator volume and
a modified F/M control strategy were investigated. All control actions were
manually implemented, although potentially they could be completely automated.
During the course of the study, the basic control strategy used was modified
F/M with a flow pattern of contact-stabilization, return sludge aeration or
plug flow depending on sludge conditioning needs.
PLANT DESCRIPTION
The Westside Treatment Plant is a 2 mgd (0.09 m3/s) plant, treating both
domestic and industrial wastes. Figure 1 shows a plan view of the plant giv-
ing major flow lines and unit process components.
The plant commenced operation in March 1971. It has a hydraulic capacity of
2 mgd (0.09 m^/s) dry weather flow and a complete mix aeration tank divided
into four quadrants which provides a great deal of flexibility in terms of
sludge conditioning time. Sparged turbines provide mixing and aeration.
Industrial wastes constitute about 12 percent of the hydraulic flow and 40 to
80 percent of the organic load. An additional food processing industry dis-
charges through vibrating screens at the plant site to an adjacent spray irri-
gation field.
The plant effluent is discharged to the Tualatin River from November 1 to
May 31. From June 1 to October 31 the effluent is irrigated on the city's
280-acre (113 ha) farm and another 200-acre (81 ha) site operated by a local
farmer.
The present NPDES waste discharge permit calls for 20/20, BOD/SS during the
period of the year when the flow is going to the river and zero discharge in
addition to the above requirements when effluent is used for irrigation.
-------
-n
ua
c
n
o>
fti
*<
O
>
'»." va i a
AEflATIOH
0« |
N°.*l!
A
RUT i
om :
..-•Aj
i
|
BASINS
01"
«ai
I wr *"
g| j on
r.,_
-------
Figure 2 depicts the aeration and clarifier relationship with the associated
piping showing the versitility of the major flow lines schematically.
PE
PRIMARY
CLARIFIER
IRAS
1 4 1 rf>
QUAD
o
B
QUAD
O
A
1-1
QUAD QUAD
0 0
C D
t ' *}
1 r RAS
RAS
SECONDARY
CLARIFIER
PE
Figure 2. Aerator-clarifier relationship.
Aeration and mixing is accomplished by four sparged air mixers, one in each
quadrant. Three positive displacement blowers with a maximum combined capa-
city of 1380 cfm supply air to the mixers.
The aeration tanks are separated by a concrete wall and each half of the aera-
tor has a volume of 182,700 gal. (694 m3). The aeration tanks may be vis-
ualized as being divided into four quadrants. Quadrants C and D (on the same
side) have no physical boundary, but do have separate feed lines for primary
effluent and return sludge. A concrete wall separates Quadrants A and B from
Quadrants C and D. There is a 24-inch (0.6 m) sliding gate valve between
Quadrants B and C and a wooden baffle separates Quadrants A and B.
Two variable speed pumps are used for return sludge flow. One pump has a
range of 350 gpm to 1200 gpm (22 to 76 1/s), and the other has a range of 50
gpm to 700 gpm (3 to 44 1/s). Because of discharge line hydraulics, combined
output with both pumps on line yields a maximum pump rate of 1400 gpm (88
1/s). During the plant's operating experience, the demand for return sludge
seldom exceeds 600 gpm (38 1/s).
8
-------
The final clarifier is 60 feet (lb.3 m) in diameter, 1U feet (3.1 m) deep at
the outer wall and 13 feet (4.0 m) deep in the center. The design overflow
rate is 707 gal./sq. ft./day (28.8 m3/m2/day) with a surface area of 2,826
sq. ft. (262 m2).
The plant was placed in operation in March L971. The raw BOD entering the
plant varies from a minimum of 21 to a maximum of 680 mg/1, and daily average
flows vary from 0.6 to 1.5 mgd (0.026 to 0.066 m3/s). Extreme peak winter
flows have exceeded 3 mgd (0.132 m3/s). During the period of January 1972
to June 1973, average monthly BOU removals varied from 38.3 percent to b9.6
percent. In July 1973, when process control using a modified F/M ratio and
varying modes of aerator volume commenced, the plant performance improved and
the average daily BOD removals varied from 90.5 percent to 98.3 percent.
Plant personnel kept aerator effluent respiration rates within the desirable
range by changing the sludge conditioning time (time MLSS has air but no new
food) and contact time (time MLSS in contact with primary effluent). These
changes are accomplished by the valving in the aerator basin.
INDUSTRIAL WASTE
Because industrial waste loading is so relevant to the necessity of this
study, the Table 1 is presented to outline types and quantities.
Table 1.
INDUSTRIAL WASTE LOAD
Source of Waste
Average Flow
gpd
Time of Travel
from Industry
to Plant
Carnation (pet food)
Haley's Food (chili beans)
Kummers (meat packer)
50,000 (190)
50,000 (190)
10,100 (38.4)
12 min.
26 min.
Several Hours
The industrial waste loading on the plant is discontinuous and the flow values
given above are estimated yearly averages. Flow and loading varies on an
hourly, weekly, and seasonal basis making predictable control based on history
essentially impossible.
CONTROL ELEMENTS
As evidence increased that the variable industrial load was contributing to
plant upset, the need for a rapid means of defining loading and process
-------
response became more evident. TOC, ATP, and RR were methods selected to
measure loading and biological response to loading for plant control.
Influent strength was measured by TOC to define instantaneous aerator loading.
By running the test at 3 to 4 hour intervals over a 24-hour period, the change
in organic load was recognized rapidly enough to make process changes. This
value was used as a measure of food or "F" in the F/M ratio.
Both TOC and ATP were used to define the relative quantity of biomass or "M"
for F/M ratios. The calculative procedure is defined later in this report,
however, the essential use of the results was to determine a return rate that
would maintain the instantaneous F/M at the desired value.
The primary effluent and return sludge feed points in the aeration tank allows
variable aeration times for adjusting sludge quality. Respiration rates on
the aerator effluent (AE-RR) are used in conjunction with loading rates to
determine which feed points and aeration tank modes are called for. As a
general rule, as respiration rates trend toward 20 mg 0£/g MLSS/hr. the
amount of sludge aeration time is increased. When the trend is toward 8
mg/g/hr., aeration time is decreased. This concept is discussed in a later
section titled, "Sludge Conditioning Time."
A modified sludge settling test calling for a sludge concentration of 2 g/1 is
used to indicate relative sludge quality and provide information on setting
sludge wasting rates. The test procedure is defined in Section 7.
Loading Ratio Control
Process control by manual feed forward TOC to determine and control a modified
F/M has not been used broadly on a full scale plant. The modified F/M is de-
fined as the ratio of primary effluent TOC on an instantaneous weight basis to
the instantaneous weight of the RAS TOG at the theoretical mixing point. In
Phase II of the study, ATP of the RAS was measured instead of TOC. In the
first case, a control regime was established whereby:
DAC -0+ n*i/m
-------
SLUDGE CONDITIONING TIME
The method whereby the sludge conditioning time is varied is outlined in
Figure 3. The aeration tank is divided into four quadrants. Both the primary
effluent flow (PE) and the return sludge flow (RS) can be directed into any
one of these quadrants. By varying the placement of these flows, the sludge
conditioning time can be varied between contact stabilization and plug flow.
The total aeration time will depend on the existing primary and return flows
and tank volume in use.
To illustrate this process, consider the situation where both the return
sludge (RS) and the primary effluent (PE) are brought into the fourth quad-
rant. This allows for (0) units of reaeration or stabilization time and (1)
unit of contact time. This is defined as mode 0-1. At the other extreme,
consider the situation where the return sludge enters the first quadrant,
while the primary effluent enters the fourth quadrant. In this case, the re-
turn sludge flows through three quadrants before contacting the primary
effluent. Since the return sludge flow is significantly less than the primary
flow, it spends more time in transit, increasing the conditioning or stabili-
zation time. This is defined as 3-1 inasmuch as 3 quadrants are used for
reaeration and 1 for mixed liquor. Similarly, other sludge conditioning times
can be accomplished by appropriate placement of flows. (Reaeration or stabil-
ization is defined as that condition where return sludge is aerated prior to
mixing with primary effluent.)
The purpose of changing modes is to keep the Aeration Effluent Respiration
Rate (AE-RR) between 8 and 20 mg Oz/g/hr. An AE-RR that is too high indi-
cates the need for more aeration time, and conversely, low values require
less aeration time.
An increase in aeration time, in this study, is defined as sludge residence
time in a quadrant receiving only return sludge. The aeration time would
then be decreased by adding primary effluent, increasing the hydraulic
through-put.
Information elsewhere in this report shows the phenomenon of aerator respira-
tion rate changes correlated with increases in return sludge respiration rate.
This increase indicates that unoxidized food recycling from the secondary
clarifier in the return sludge builds up over a period of time and requires
additional stabilization time prior to the addition of "new" food in the pri-
mary effluent. As residence time in the aerator is increased, the respira-
tion rate tends to decrease.
Operators were instructed when to change modes by a program in a programmable
calculator that contains a collection of decision-making rules based on past
response of the process to various respiration rates and mode changes. Inputs
to the calculator program are "Present Mode," "How Many Hours in this Mode,"
and "Present AE-RR." The calculator printed out instructions to stay in the
present mode or change to another mode.
11
-------
MODE RA
NUMBER
3-1
2-2
1-3
0-4
0-1
RAS PE
PE RAS
0-3 [
0-2 [
PE
RAS
AE
AE
AE
j | Primary Effluent only; mixer on, air off.
Return Activated Sludge only; mixer on, air on. The
detention time through these quadrants equals the RAS
detention time and the number of quadrants used is
identified by the first mode number.
Mixed Liquor; mixer on, air on. The detention time
through these quadrants equals the ML detention time
and the number of quadrants used is identified by the
"last mode number."
Figure 3. Mode change description.
12
-------
Process Control Scheme
The relationship of control loops and parameters adjusted is summarized in
Figure 4. Four loops were identified and operated throughout the course of
the study using input to the programmable calculator and the output results
were then implemented manually.
O
VARIABLE MEASURED
AND CONTROLLED.
PARAMETER
TREATMENT
Figure 4.
SLUDGE
Process control schematic.
Control Loops
1. Food to biomass ratio F/M (F = TOC PE M = RAS TOC or RAS ATP)
2. Respiration rate of aeration tank effluent (RR)
3. 5 min. corrected settling volume of aeration tank effluent
(CSV5)
4. Dissolved oxygen in aeration tank (DO)
Key Parameters Adjusted
1. Returned Sludge Flow (RSF)
2. Waste Sludge Flow (WSF)
3. Sludge Conditioning Time (SCT)
4. Aerator Air Flow (AAF)
Other Control Elements
Aerator air was controlled between 1-4 mg/1 in each quadrant containing
sludge in either a contact or reaeration (or stabilization) mode.
13
-------
Waste sludge was normally removed from the bottom of the secondary clarifier
at a fixed rate of 58 gpm (3.6 1/s) and thickened in a flotation thickener.
The number of minutes of wasting per day at the fixed rate was determined on
results of the settling test, loading and RAS-TOC. The specific calculation
is defined in Section 6.
A programmable calculator was used to provide information storage, computation
and process control direction. Many of the control parameters were prepro-
grammed into the calculator and used in conjunction with periodic series of
tests run during the course of a day. These results were input into the cal-
culator and the readout gave the operator specific instructions on the re-
quired process change based on test results.
PHASES OF THE PROJECT
Work under this grant commenced on July 25, 1974, and continued through
July 24, 1975. During this time period, a number of refinements in technique
and control were implemented. The study was divided into three basic phases
preceded by a period of gathering background data. The phases and times of
implementation are shown on Table 2.
Table 2.
PHASE IDENTIFICATION
Phase
Date
RAS and Modified F/M Control
I
II
III
7/25/74 to 12/31/74
1/01/75 to 6/26/75
6/26/75 to 7/24/75
Controlled by TOC, ATP monitored
Controlled by ATP, TOC monitored
Controlled by TOC, ATP monitored
Because one of the objectives of this study was to compare TOC and ATP, con-
trol decisions for RAS rates were made based on results of one parameter while
data was collected simultaneously on the other. Food (F) was measured in all
cases using TOC of the primary effluent.
Chemical and Biological Analysis During Study
The testing program consisted of: a series of tests performed five to seven
times per day, Cl£ residual, DO, pH, SVI, microscopic examination of the bac-
teria in the aerator, depth of final clarifier sludge blanket, and calculation
of the pounds of waste activated sludge.
The TOC test series consisted of a TOC reading on the primary effluent, mixed
liquor, return activated sludge flow and final effluent. In addition, other
tests were used to monitor the process including settlometer, final effluent
turbidity, MLSS concentration, and plant flow. These tests were performed by
the operator on duty.
14
-------
Data Collection
During the course of the study, a series of tests were run on samples col-
lected by plant personnel 5 to 7 times daily. Collection times are shown on
Table 3.
Table 3.
SAMPLE COLLECTION TIMES
Time
(-1 Hr) Sunday
12:00 midnight
4:00 a.m.
8:00 a.m.
10:30 a.m.
1:30 p.m.
4:00 p.m.
8:00 p.m.
X
X
X
X
X
Monday
X
X
X
X
X
Tuesday
X
X
X
X
X
X
X
Wednesday
X
X
X
X
X
X
X
Thursday
X
X
X
X
X
X
X
Friday
X
X
X
X
X
X
X
Saturday
X
X
X
X
X
X
X
Determinations were made on samples collected or instantaneous readings were
recorded for the following parameters at each of the times given in Table 4.
During each sample period the following information noted on Table 4 was col-
lected and recorded.
Table 4.
MAJOR INFORMATION RECORDED
Information Categories
Comments
Time of Day
Plant Flow, mgd
Primary Effluent TX, mg/1
RAS TOC, mg/1
Aerator Effluent TOC, mg/1
Sludge Blanket Surface
Depth, ft
Mixed Liquor Aeration Time,
Hrs
Time of sample collection (±1 ^r
from time noted on Table 3)
From plant flowmeter
Dohrmann Envirotech Model 50 TOC
Analyzer
Dohrmann Envirotech Model 50 TX
Analyzer
Dohrmann Envirotech Model 50 TX
Analyzer
Measured with a lighted blanket
finder 2 to 4 times per day
Calculated by the Tektronix Model
31 calculator
15
-------
Table 4 (Continued). MAJOR INFORMATION RECORDED
Information Categories
Comments
RAS Aeration Time, Hrs
Final Effluent Turbidity,
Jackson Units
SVI
5-Minute Corrected Settlo-
meter
Final Effluent TX, mg/1
Mode of Operation
Calculated Modified F/M Ratio
Present RAS flow, gpm
New RAS flow, gpm
Lbs Solids Wasted/Day
Aeration Effluent Respiration
Rate mg 02/g/hr
RAS ATP, mg/1
Aerator Effluent ATP, mg/1
Calculated by the Tektronix Model 31
calculator
Approximated by Spectronic 70 Spec-
trophotometer from 7/25/74 to
2/20/75. A Hach 2100A Turbidometer
was used for the balance of the
study.
Determined 2 to 3 times daily
Determined 3 to 4 times daily
approximately at alternate sample
times
Dohrmann Envirotech Model 50 TX
Analyzer
Aerator mode defining feed points
for PE and RAS
Determined by relationship of PE-TOC
and RAS-TOC or RAS-ATP
Observed reading from RAS propeller
meter
Calculated setting as determined by
program to maintain desired modified
F/M
Calculated amount of sludge to waste.
Readout is in minutes to waste per
program
Determined by YSI DO meter and
concentration of AE
JRB-ATP photometer
JRB-ATP photometer
16
-------
SUMMARY OF PHASES
Phase I - TOC Control While Monitoring ATP
From the initiation of the study until December 31, 1974, RAS rates and mod-
ified F/M were determined by TOC of PE and RAS. The method of calculation is
described in Section 5.
During this period, optimum ranges were defined for several of the basic de-
terminations and parameters. These were first approximations and were refined
later.
The major emphasis was on perfecting technique in gathering background infor-
mation. Considerable time was spent on refining the ATP sample preparation
and testing procedure which had to be revised from that provided by the sup-
plier of the ATP equipment.
Emphasis was placed on finding the best operating ranges for respiration rate,
settleability, RAS-TOC, and confirmed to be bracketed as follows:
1. Settleability measured by the adjusted settlometer at 5 minutes: less
than 600 ml/1 (see Section 7 for description of this test).
2. Aerator effluent respiration rate (AE-RR) between 8 and 19 mg 02/g
MLSS/Hr.
3. RAS-TOC: less than 5,000 mg/1.
The range and average of some of the other key data points is in Table 5.
Table 5. PHASE I RANGE AND AVERAGE DATA
Operation
Parameter
PR I TX mg/1
RAS TOC mg/1
AE TOC mg/1
CSVs ml/1
Flow and Performance
Fin. Eff. JTU
Fin. TOC mg/1
Fin. BOD mg/1
Fin. SS mg/1
Flow mgd
Monthly
Average
Range
118 -
2,589 -
750 -
497 -
Parameter
8.4 -
42
15
19
0.71 -
189
4,010
883
619
17.1
68
40
40
1.43
Monthly
Average
171
3,500
914
575
12.6
61
30
29
0.91
17
-------
Digester supernatant recycling from the anaerobic digesters influenced plant
loading in addition to external loads. Raw suspended solids were recorded as
high as 4,000 mg/1 and 420 mg/1 in the primary effluent in September 1974. A
liquid sludge hauling program was implemented late in 1974 that resolved this
situation.
The following data summarizes the comparison before and after control of
supernatant.
Table 6. EFFECTS OF DIGESTER SUPERNATANT ON PROCESS
Nov.
Dec.
Raw Pri
SS BOD SS BOD TOC
1974 w/super 1,205 586 178 219 184
1974 w/o super 251 229 124 161 118
Fin
SS BOD
29 28
18 15
Phase II - ATP Control While Monitoring TOC
The JRB-ATP photometer was used to analyze for adenosine triphosphate (ATP).
During Phase I, the procedure was perfected and plant operators were in-
structed in correct sample collection, preparation, and determination.
The RAS rate change was made using the ATP determination 89 percent of the
time in January 1975 and 98 percent of the time from February through April
1975 (see discussion later in this section). Simultaneous data was collected
for ATP and TX and a set of confidence limits established allowing a decision
on which parameter was used. These limits are also discussed later in this
section.
Industrial waste loading fluctuated widely during the course of Phase II and
the annual winter high flows necessitated diversion of a portion of the pri-
mary effluent around the secondary units to prevent hydraulic washout. The
decision to divert is based on the amount of the solids in the secondary clar-
ifier under the assumption that it is preferrable to preserve the biomass and
divert chlorinated primary effluent than to spend 10-14 days reestablishing
the growth. Approximately 86 percent of the flow received secondary treatment
between February 1, and April 30, 1975.
During this phase, ATP was used as a control parameter for adjustment of RAS
to maintain a constant F/M. The use of this different parameter combined with
too frequent mode changes led to loss of control of the process. The apparent
reason was the tendency to make corrections too rapidly without allowing the
process to reach a new equilibrium point. The two major changes being made
were adjustment of return rates and trending from plug flow to contact
stabilization.
18
-------
Table 7. PHASE II RANGE AND AVERAGE DATA
Operation Parameter Monthly Average Range Monthly Average
PR I TX mg/1
RAS TOC mg/1
AE TOC mg/1
CSV5 ml/1
84 to 168
3,300 - 3,760
675 - 1,067
408 - 773
121
3,522
834
534
Flow and Performance Parameter
Fin. Eff. JTU
Fin. TOC mg/1
Fin. BOD mg/1
Fin. SS mg/1
AE ATP*
RAS-ATP*
Flow mgd
7.0
40
10
13
5.0
17.2
0.92
- 19.5
- 85
- 45
- 120
- 6.7
- 20.3
- 1.8
10.5
56
21
33
5.8
19.1
1.4
*Four month^ data.
Phase III - TOC Control While Monitoring ATP
This phase was carried out during the last month of the grant period. A re-
vised mode change schedule was implemented that allowed more time for the pro-
cess to stabilize following mode adjustment.
During this period, the primary effluent TOC exceeded 200 mg/1 (17 times in 21
determinations) between July 10 and 12, 1975. A maximum value of 427 was re-
corded at 4:00 p.m. on July 12, 1975. In spite of this shock industrial load,
the process was held in control and plant effluent stayed within acceptable
1 imi ts.
Table 8. PHASE III RANGE AND AVERAGE DATA
Operation Parameter Monthly Average Range Monthly Average
PR I TOC mg/1
RAS TOC mg/1
AE TOC mg/1
CSV5 ml/1
106 to 284
2,844 - 4,593
646 - 1,131
505 - 923
154
3,793
964
700
Flow and Performance Parameter
Fin. Eff. JTU 2.8 - 16.9 5.1
Fin. TOC mg/1 31 - 89 49
Fin. BOD mg/1 6-34 13
Fin. SS mg/1 2-52 13
Flow mgd 0.62 - 1.8 0.83
19
-------
SECTION 5
DISCUSSION OF RESULTS
APPLICATION OF TOC
To meet the objectives of this study, it was necessary to define TOC control
procedures, gather baseline information for comparing TOC to ATP control and
document some of the reasons for success of the application of TOC control
prior to the study period.
Even though successful results were obtained early in the TOC control strategy
application, problems still existed from time to time in maintaining contin-
uous control for a variety of reasons. Three causes for plant upset were
identified: shock loads to the process when sludge condition was not prepared
for rapid increases, lack of a defined wasting program and unexplained influx
of filamentous organisms which seem to be correlated with lowered wastewater
temperatures.
Influent Load Monitoring By TOC
An improvement in the ability of the system to withstand shock loads was evi-
dent between August 1974 and September 1974, when for two five-day periods the
PE-TOC averaged 228 mg/1. In August, the secondary clarifier lost a signifi-
cant amount of solids resulting in a final SS of 275 mg/1 on the 5th day.
Essentially the same loading in September resulted in a peak final SS of 61
mg/1, but the SS was only 22 mg/1 on the 5th day. These data are shown on
Table 9. The ability of the plant to handle high loads in September was cor-
related with the improved sludge quality as measured by both the corrected
settleability and respiration rates. This added credence to the necessity of
holding settling below 600 ml/I in 5 minutes and respiration rates below 20
mg 02/g/hr.
Table 9.
TOC CONTROL COMPARISON
Day of
Week
T
W
T
F
S
Aug. PE-TOC Fin SS
1974 mg/1 mg/1
6
7
8
9
10
192
221
235
210
284
20
24
44
80
275
AE-RR
mg/g/hr
17
28
32
39
57
Sept. PE-TX Fin SS
1974 mg/1 mg/1
10
11
12
13
14
207
200
261
271
199
34
35
61
26
22
AE-RR
mg/g/hr
12
12
15
13
11
20
-------
A typical range for PE-TOC over a 24-hour cycle, when excessive industrial
loading was not evident was, 84 to 148, compared to a typical range of 114
to 219 mg/1 when industrial loading was evident.
Plant upset was most frequently experienced when the PRI-TOC averaged 190 mg/1
or above for more than 48 hours. In July 1975, changes were incorporated into
the wasting and mode adjustment programs in conjunction with modified F/M con-
trol to reduce the possibility of these upsets.
Initially, adjustment of RAS rates to maintain a fixed modified F/M was prac-
ticed. As more experience was gained, the target modified F/M was allowed to
vary with the RAS TOC which resulted in better control during sustained shock
loads. This is covered more thoroughly in Section 6.
RAS-TOC
The TOC of the return sludge was used as a measure of "M" in the modified F/M
ratio. The initial approach in the early phases of the study was to hold F/M
constant by varying the return rate proportionately as the primary effluent
load changed. In July 1975 a new program was put into use whereby the mod-
ified F/M used to control the return rate was varied as the observed AE-RR
changed. This change was made in an effort to reduce the necessity for fre-
quent mode changes after observing apparent process instability when modes
were adjusted during prolonged high loading periods. Early in the study wast-
ing was done at a constant rate from the return sludge line with the variable
being hours of wasting. Experience with the system led to the derivation of
a family of curves shown on Figure 5. These data were programmed into the
printing calculator and wasting was practiced each day starting on day shift.
These conditions held true in the range of RAS-TOC between 4000-6500 mg/1.
Later RAS-TOC, along with CSVs and mode of operation, was used in the deter-
mination of amount of sludge to waste. (More information on this subject is
presented in Section 6 - Operation Procedures.) RAS-TOC was found more sensi-
tive than secondary clarifier sludge blanket level, and more convenient than
RAS-SS as indication of biomass buildup for input to the wasting-time calcu-
lator program. RAS-SS could theoretically be used in place of RAS-TOC since
they were found to correlate at the Westside Plant as follows:
(RAS-SS, g/1) x (371) + 250 = RAS-TX, mg/1
APPLICATION OF ATP
Limited work to date has been done on the application of ATP to process con-
trol; therefore, one of the objectives of this study was to test the capabil-
ity of using ATP values to determine the modified F/M for process control.
ATP has been recognized as one means to identify the viable fraction of the
sludge mass. The premise was made that by varying the proportion of viable
mass as measured by ATP in proportion to the load measured by PE-TOC the mod-
ified F/M would bear a closer relationship to actual "live" mass per unit of
available food.
21
-------
feBOO
€>OOO
55OO
_i
75"
o
e
J) 5OOO
cc
4-SOO
> 5 10 is
HOURS OF WASTING
Figure 5. Relationship of RAS-TOC to wasting time.
22
-------
The return rate using ATP values was established as follows:
RAS purcp rate, „.,/„„ -
where PRI-TOC = Primary effluent TOC
RAS-ATP = Return activated sludge ATP
F/M ratio = Ratio varied from 0.15 to U.35 during
the study
(RAS TOC = RAS-ATP x 187. b)
gpm x O.U63 = 1/s
ATP was correlated with TOC in order to establish the working limits. (The
term 694 was a conversion factor to obtain an expression in gallons per
minute to coincide with the RAS meter readout.)
Determination of Correlation of RAS, ATP and TOC
The basis for establishing correlation for the modified F/M using ATP was done
in November preparatory to beginning actual control.
On December 24, 1974, the return rate control commenced according to the re-
sults of RAS-ATP tests. The printing programmable calculator was programmed
to determine mathematically the return rate from the usual data, plus the RAS-
ATP. Because the laboratory ATP procedure is relatively involved compared to
TOC, a method was instigated to prevent setting the return pump from faulty
ATP determinations. Boundaries were defined on the plot of RAS-ATP, and RAS-
TOC, and when the RAS-ATP for a particular RAS-TOC fell outside of this
boundary, the calculator program assumed that the ATP determination was faulty
and calculated the return rate from the RAS-TOC in the usual way.
The RAS ATP-TOC pairs used were taken between November 5 and November 21,
1974, and the calculator program is based on 83 pairs of data from those
16 days.
The boundaries, as defined, include about 90 percent of the points. This
means that the return rate would be set according to ATP 90 percent of the
time. The calculator was programmed to print out which parameter is being
used each time.
The best-fitting line was drawn by visual inspection after the 83 points were
plotted. These data are presented on Figure 5.
Referring to Figure 5, the procedure used for incorporating the RAS ATP into
the return rate determination is illustrated by the following two cases:
1. RAS TOC = 3,000 2. RAS TOC = 3,000
RAS ATP = 20.00 RAS ATP = 30.00
In case (1), the ATP is within the boundaries. The ATP determination was
assumed to be accurate, and the return rate was calculated according to the
previously described formula.
23
-------
5000
4000
3000
o
o
Jt-
m
oc
2000
1000
10
20
RAS-ATP
30
40
Figure 6. Line of best fit for RAS ATP/TOC correlation,
24
-------
In case (2), the ATP is outside of the boundaries. The ATP laboratory deter-
mination was then considered to be in error, and the return rate was set using
TOC.
Observations on the ATP Testing
The procedure, although difficult, was performed by operation personnel around
the clock. Standards were made up by the chemist or technician. Because the
calculation to obtain the results in mg/1 involved comparing the raw number
from the photometer with the scope of the standard curve, a program was estab-
lished on the programmable calculator.
An explanation of the RAS-SS/TOC/ATP relationships in Figure 7 can be hypothe-
sized as follows: Normally the RAS-SS concentration was maintained around 8
to 9 g/1 corresponding to 3,3UU to 3,5U(J mg/1 TOC. Data for higher points on
Figure 7 were most likely taken soon after a period of increased growth which
led to the higher values. Increased growth means increased proportion of vi-
able biomass, thus the "stretching out" and nonlinearity of the RAS-ATP curve.
Since the degree of "stretching out" of the ATP curve was more than antici-
pated, the RAS rate calculator program was not set up to benefit the process
from this ATH-derived information. Instead, the greater ATP values, than
would be predicted by a linear relationship, led to calculated RAS flow rates
which at times exceeded the hydraulic capacity of the secondary clarifier for
proper operation.
Presumably, a larger range of F/M values could be used with the ATP data in an
RAS rate formula to derive maximum benefit from the viable biomass informa-
tion. But due to the greater difficulty of the ATP laboratory procedure com-
pared to the TOC procedure, and because TOC information seems to be entirely
satisfactory in this type of operational program, ATP shows more promise for
research than as a control tool.
ATP Determination by Operation Personnel
Following the period of perfecting technique for operation of the JRB-ATP
Photometer and establishing a correlation between TOC and ATP, it was neces-
sary to train operation personnel running the routine samples inasmuch as
sampling and determinations were necessary on all three shifts. This was
accomplished by using a cassette tape-slide projector combination and record-
ing the essential steps in sequential fashion. This unique method made the
instructions available on a 24-hour basis to each of the operators involved.
Difficulty was encountered during the early stages of the project in obtaining
repeatable results and this was traced to sample preparation techniques. Much
helpful information was obtained from Harold Bond, EPA National Ecological
Research Lab, Corvallis, Oregon.
25
-------
3000
12
3500
14
4000 4500
RA3- TOC MG/L
16
18
5000
20
RA8-ATPMQ/L
Figure 7. Relationship of RAS-SS to ATP and TOC.
26
-------
APPLICATION OF RESPIRATION RATES
The observation and control of respiration rates proved to be a key factor in
plant operation, being integrally associated with two control loops in the
process control scheme. These two loops tie RR with RAS rate and RR with mode
change.
Historically, it has been noted that the best treatment occurred when RR was
maintained between 8 and 19 mg 02/g solids/Hr. Operation for more than 12
hours at RR > 20 produces poor settling sludge and cloudly effluent.
This measurement is taken on the aerator effluent as flow passes out of the
aerator into the secondary clarifier influent line. The method for determin-
ation is given in Section 7.
Relationship of RR and RAS Rate
Within hydraulic limits, an increase in return rate was observed to cause a
decrease in RR. A program was written for the programmable calculator which
allows the operator to enter the present AE-RR rate. This value is considered
in conjunction with the other input items and a readout is given, providing
the operator with the new setting in gpm. The complete list of input items
are:
1. Influent flow
2. Most recent PE-TOC
3. Present RAS rate
4. Most recent RAS-TOC
5. Present aerator mode
6. Present AE-RR
Relationship of RR and Mode Changes
The configuration of the Hillsboro plant aerator-clarifier relationship was
described in Section 4 and shown in Figure 2. The ability to regulate the
feed points and aeration time is the other major factor in respiration rate
control. Mode change criteria was revised three times during the course of
the study and in July 1975 a scheme was developed which limited the frequency
of changes to prevent undue internally-caused shocks to the process. These
criteria are discussed in the Section 8.
The interrelationship of RAS-TOC has been observed to have a marked effect on
sludge quality and RR. When the RAS-TOC exceeded 5,000 mg/1, the process
trended toward instability. A rapid rise in RAS-TOC and RAS-ATP generally
resulted after industrial shock loads. The most immediate corrective action
was toward longer sludge aeration times which is affected by mode changes that
favor this condition. The RR is used to indicate the progress of stabilizing
excess food that is carried through the aeration-sedimentation cycle and back
to the aerator. There appears to be detrimental effects to the system when
apparently unoxidized food is recycled concurrently with incoming loads.
27
-------
SECTION 6
OPERATION PROCEDURES
USE OF PROGRAMMABLE CALCULATOR DURING THE STUDY
The use of programmable calculator was initiated at Hillsboro in June 1974,
and has been in continuous use since that date. The first piece of equipment
to be used was a Monroe 1775 with a card reader for data input. This unit was
replaced with a Tektronix Model 31 calculator utilizing keyboard program lan-
guage input and alphanumeric printout. The Model 31 can be programmed by plant
personnel and the most difficult program to date was completed in less than
six hours.
The operator used all of the following on a routine basis to carry out the re-
quired operation procedures in connection with the control strategy:
1. Determination of settlometer mixing ratios.
2. AE respiration rate and aeration time.
3. RAS rate.
4. Wasting time required for process control.
5. Cost of polymer for flotation thickener.
6. Aeration mode.
7. AE-ATP in mg/1 from standardization curve.
8. RAS-ATP in mg/1 from standardization curve.
Practical Use of Calculator
There are many advantages to the use of the calculator in its present capac-
ity, not the least of which is uniformity of data transmittal and handling.
The following list summarizes a few of the main features of the concept.
1. Operation instructions are uniform, as an example, RAS pump settings
are established by the preplanned program and after evaluation of the
program parameters printed into the calculator programs, a readout
telling what the new setting should be in gpm is printed out.
2. Arithmetic errors are reduced materially.
3. The time factor for making calculations is essentially eliminated.
4. Information on basic operation instructions are available 24 hours a
day.
5. Elements can be built into the system for eventual plant automation.
28
-------
7.
Complicated procedures, such as determining the slope of the standard-
ization curve to obtain a readout in mg/1 for the ATP test, were ac-
complished quickly, allowing results to be applied immediately for the
process control.
Certain unit cost information is immediately available and useful in
process control. As an example, cost information on polymer use per
dry ton for flotation thickening is included in the sludge wasting
program, keeping the awareness of cost efficiency before the operator.
MODE CHANGE CRITERIA
As discussed earlier, changing aeration modes to maintain the desirable res-
piration rate was an integral part of the control strategy.
Three mode change criteria were used during the course of this study, each one
being a refinement of the previous one and each was instituted following an
evaluation of operation experience.
The reader is referred to Figure 2 for the configuration of quadrants and feed
points.
Mode Change Criteria - 7/25/74 to 10/20/74
Instructions for determining the aerator mode of operation:
1. A respiration rate should be run on the AE two or three times each
shift (a few hours apart) and
if the present respiration
rate comes out:
Greater than 25
Between 20 and 25
Less than 20
2. If yesterday's respiration
rate average was:
Greater than 25
Between 20 and 25
Between 15 and 20
Between 12 and 15
Less than 12
the mode should be (until
the next RR is run):
2-2 unless today's base mode
(see below) is a higher
1-3 mode; in that case, don't
go below the base mode.
Same as the base mode.
today's base mode is:
3-1
2-2
1-3
0-4
0-3
29
-------
Table 10. MODE NUMBER DEFINITION
Mode Number* PE Feed Point RAS Feed Point
3-1
2-2
1-3
0-4
0-3
Quad.
Quad.
Quad.
Quad.
Quad.
D
C
B
A
A
Quad.
Quad.
Quad.
Quad.
Quad.
A
A
A
A
B
*The first digit refers to the number of quadrants in which RAS aeration is
taking place.
The second digit refers to the number of quadrants in which ML aeration is
taking place.
Mode Change Criteria - 10/21/74 to 7/22/75
1. The shift supervisor was responsible for making mode changes. If no shift
supervisor was on duty, the person on duty who had been employed at the
plant the longest was responsible.
2. The objective was to keep the AE-RR within the range 12 to 15 as much as
possible.
3. The principle was:
a. If the AE-RR was less than 12, move in the direction from 3-1 to-
ward 0-3 one or more modes. This tends to decrease the aeration
time and increase the AE-RR or, it provides less RAS stabilization
time and increased AE-RR.
b. If the AE-RR was greater than 15, move toward 3-1 one or more
modes. This increased the aeration time and decreased the AE-RR
or, provided more RAS stabilization and eventually decreased the
AE-RR.
4. In general, when the AE-RR was:
a. Far out of range (less than 8 or more than 19), movement of one
mode once per shift until the AE-RR returned within to 12 to 15.
b. A little out of range (8 to 12 or 15 to 19) the operate consid-
ered the following before making a decision:
30
-------
(1) The trend of the AE-RR.
(2) The PE TOC and its trend.
(3) The usual daily trends, especially for that day of the
week.
Table 11. MODE CHANGE INSTRUCTIONS 10/21/74 - 7/22/75
Mode
Number
3-1
2-2
1-3
0-4
0-3
0-2
0-1
PE Feed
Point
D
C
B
A
A
A
A
RAS Feed
Point
A
A
A
A
B
C
D
Keep the DO
at 1 to 4 mg/1
in Quadrants
All
All
All
All
B, C, D
C, D
C
Keep the Air
Valve Off at
Quadrants
—
—
—
A
A, B
A, B, C
Mode Change Criteria - 7/23/75
Near the end of the study in May and June 1975, a series of upsets occurred
leading plant personnel to review the operation procedure. The sludge was
typically bulky a portion of the time and as is the case with bulking sludge,
good treatment was achieved as long as solids could be kept from overflowing
the secondary clarifier weirs. The system was very unstable during high flow
periods of the day consequently treatment efficiency fluctuated markedly de-
pending on hydraulic loading.
Persistent upset conditions generally lead to frequent adjustments in process
control which, in turn, generate further swings of the pendulum perpetrating
further upset. Plant personnel reviewed the trend which was developing in an
effort to pinpoint internal causes of upset inasmuch as similar situations had
been handled successfully in the past.
Process instability appeared to be related to something associated with the
way changes were being made in response to directions given operators by the
existing calculator program. It was determined that personnel were carrying
out the directives as printed out for a given situation.
Following the review of both data and operation procedures, it was decided to
revise the mode change criteria to allow less frequent changes. The rationale
behind this move is discussed in more depth in Section 8 titled "Discussion of
Interrelated Parameters."
Table 12 outlines the calculator program elements involved in the change
criteria.
31
-------
Table 12. AERATION MODE-CHANGE DIRECTION FOR CALCULATOR PROGRAM
(Instituted 7/23/75)
If the present and if the present and if the process has
mode is: AE R.R. is: been in the present mode:
0-2
0-2
0-2
0-2
0-3
0-3
0-3
0-3
0-3
0-3
0-3
0-3
0-4
0-4
0-4
0-4
0-4
0-4
0-4
1-3
1-3
1-3
1-3
2-2
2-2
2-2
2-2
3-1
3-1
<16.0
16.0
16.0
>26.0
26.0
26.0
< 8.0
< 8.0
8.0
19.0
19.0
>26.0
>26.0
>26.0
19.0
26.0
26.0
< 8.0
< 8.0
8.0
22.0
22.0
>30.0
>30.0
<18.0
18.0
>40.0
>40.0
<20.0
20.0
>40.0
>40.0
<20.0
>20.0
22.0
30.0
30.0
- 40.0
- 40.0
(all cases)
<8.0 hours
>8.0 hours
(all cases)
>72.0 hours
<72.0 hours
(all cases)
<12.0 hours
>12.0 hours
<8.0 hours
8.0 - 12.0 hours
>12.0 hours
<24.0 hours
>24.0 hours
(all cases)
<24.0 hours
>24.0 hours
<8.0 hours
>8.0 hours
(all cases)
(all cases)
<12.0 hours
>12.0 hours
(all cases)
(all cases)
<12.0 hours
>12.0 hours
(all
(all
cases)
cases)
then:
stay in 0-2
stay in 0-2
move to 0-3
move to 0-4
move to 0-2*
stay in 0-3
stay in 0-3
stay in 0-3
move to 0-4
stay in 0-3
move to 0-4
move to 1-3
stay in 0-4
move to 0-3
stay in 0-4
stay in 0-4
move to 1-3
stay in 0-4
move to 1-3
move to 0-4
stay in 1-3
stay in 1-3
move to 2-2
move to 1-3
stay in 2-2
stay in 2-2
move to 3-1
move to 2-2
stay in 3-1
*Program will not allow a move to mode 0-2 unless AE-RR daily average has
been below 8.0 for three consecutive days.
WASTING CRITERIA
The secondary clarification at the Hillsboro plant allows waste sludge to
be drawn either from a "sludge pocket" under the sludge collection arms or
from the RAS line.
32
-------
Sludge is wasted to a flotation thickener at a constant rate of 58 gpm (3.6
1/s), through a positive displacement pump. The amount to be wasted was found
to be a function of settleability and RAS TOC. The settleability factor was
determined from the CSVs results and RAS TOC was previous days average.
The criteria for wasting time (WT) for various daily averages of RAS TOC and
CSV 5 is based on an assumed RAS SS of 10.0 g/1. Actual wasting time (WT) is
calculated by modifying "X" from Table 13 as follows:
WT = (X)
10
) Hrs
6 + 0.4(RAS SS)
Table 13.
WASTING CRITERIA
RAS-TOC mg/1
7000
6000
5000
4000
3000
2000
Mode 1-3 to 0-4
X = 15 Hrs
for CSV5<500
X = 10 Hrs
for CSV5>500
X = 10 Hrs
for CSV5<500
X = 5 Hrs
for CSV5>500
X = 5 Hrs
for CSV5<500
X = 0 Hrs
for CSV5>500
X = 0 Hrs
all cases
Mode 0-3
X = 15 Hrs
for CSV5<500
X = 10 Hrs
for CSV5>500
X = 10 Hrs
for CSV5<500
X = 5 Hrs
for CSV5>500
X = 5 Hrs
for CSV5<500
X = 0 Hrs
for CSV5>500
X = 0 Hrs
all cases
Mode 0-2
X = 10 Hrs
for CSVt,<500
X = 5 Hrs
for CSV5>500
X = 5 Hrs
for CSV5<500
X = 0 Hrs
for CSV5>500
X = 0 Hrs
all cases
33
-------
Although the wasting program was in a continuous state of refinement prior to
and during the study, the same principles, discussed in Section 8, were
applied. At the beginning of the study only RAS-TOC and CSV5 were used as
input parameters to the wasting-time calculator program. During the course of
the study it became evident that the aeration mode must also be input to pre-
vent loss of long-range process control, indicated by increasing CSVs values
or a slower sludge settling rate.
The final, very successful, wasting program consisted of a calculator program
containing the criteria of Table 14. Sludge was wasted to a flotation thick-
ener at a constant rate through a positive displacement pump from the "sludge
pocket" at the bottom center of the secondary clarifier.
The sludge concentration from this location was related to the RAS concentra-
tion by a formula based on actual data. The amount to be wasted is a func-
tion of three parameters, aid averages of the previous day: RAS-TOC, aeration
mode, and CSVs. The daily wasting time (WT) was found by modifying the X
hours from Table 14 by the previous formula which corrected for the concentra-
tion and the clarifier exit location of the material wasted.
RAS FLOW CONTROL CRITERIA
As noted earlier, the objective in varying RAS was to hold the modified F/M
within a predetermined range of values which in turn affects the AE-RR and
thus the condition of the biomass and quality of the effluent.
Initially the goal was to maintain a constant value as determined by the ratio
between PE TOC as a measure of F and RAS TOC for M. As more data was col-
lected and reviewed it became evident that the system could not be controlled
when high loading forced respiration rates to corresponding high values. Con-
sequently, an empirical equation was written to allow the modified F/M ratio
used in RAS rate control to be varied in order to help hold respiration rates
within the desired range. Holding RR below 20 by altering the mode has im-
proved effluent quality at the Hillsboro plant.
As shown in Figures 2 and 3, the aerator configuration can be approximated as
four quadrants with the capability of introducing return sludge and/or primary
effluent to each quadrant. By trending toward contact stabilization when it
is desired to reduce RR and trending toward plug flow and less quadrants in
use for mixed liquor when an increase in RR is indicated, the process can be
held in balance.
This control relationship is displayed in Figure 8.
34
-------
20
DC
X
o
i
cc
cc
I
111
.5
z
OT 10
Ul
CC
Q.
0.15 O.20 O.2S
MODIFIED F/M RATIO
O.30
0.35
Figure 8. Respiration rate—modified F/M relationship.
35
-------
SECTION 7
EXPERIMENTAL PROCEDURES
The major instruments used during this study were the Dohrmann-Envirotech
Model 51 TOC Analyser and the JRB ATP Photometer. In order to apply these
units to the goals of this research specific procedures were developed. Par-
ticularly in the case of the ATP Photometer, considerable time was spent per-
fecting a procedure that would yield repeatable results and be adaptable to
use by operators on 24-hour shifts.
The following outlines procedures that were used by shift personnel as fre-
quently as seven times a day in order to obtain the required operation infor-
mation for process control.
TOC ANALYSIS
TOC Analyser As A Process Tool
As a process tool, the Dohrmann-Envirotech TOC Analyser has proven to be a re-
liable instrument and capable of being used by as many as 10 different indi-
viduals in a week performing as many as 135 determinations. The unit is cal-
ibrated daily requiring approximately 20 minutes. The only major expenses
have been associated with repair or replacement of the pyrolysis tube.
Gas and chemicals cost about $3.00 per day.
Equipment and Procedures
Samples
1. Aerator Effluent (AE), Return Activated Sludge (RAS), Primay Effluent
(PE), and Final Effluent (FE).
Sample Preparation
1. AE - Dilute 250 ml of AE to 500 ml with distilled water. Place the
sample in a blender and blend it at top speed (liquify) for 90 seconds.
2. RAS - Dilute 50 ml of RAS to 500 ml with distilled water. Place the
sample into a blender and blent it at top speed (liquify) for 90 seconds.
3. PE - Run full strength. No blending necessary.
36
-------
4. FE - Run full strength. No blending necessary.
5. None of the samples were acidified prior to analysis.
Equipment and Reagents
1. Envirotech Model DC-50, Total Organic Carbon Analyzer.
2. Compressed Gasses
a. Ultra pure helium
b. Hydrogen
C. Compressed air
3. Osterizer Blender
4. 5U Microliter Syringe
5. Manganese dioxide
6. Potassium Biphthalate MW 204.23
Reagent Instructions
1. Mn02 - Changed daily.
2. Potassium Biphthalate - Dissolve 0.6375 g in distilled water and dilute
to 1 liter. (Standard stock solution.)
Calibration
1. Sample Size - 30 microliters.
2. Sample - 300 mg/1 potassium biphthalate TOC standard.
3. If necessary, after analyzing the standard with the TOC analyzer, set the
span so the display reads 300.
Sample Analysis - Procedure
1. Set TOC analyzer in Organic Carbon Mode for all samples.
2. Sample Size - 30 microliters.
3. Push the "Baseline" button.
4. Move the boat into the pyrolysis zone.
5. When the baseline has stabilized, somewhere less than a reading of 20,
move the boat to the heat sink.
37
-------
6. Push the "Operate" button.
7. Using a well mixed sample, withdraw at least 50 microliters of sample
being careful not to trap air bubbles in the barrel.
8. Invert the syringe and tap it gently to remove any trapped air, if
present.
9. Eject all but 30 microliters of sample and wipe off the syringe.
10. After the boat has been in the heat sink for at least 45 seconds, remove
it to the injection septum.
11. Inject the 30 microliters sample carefully into the Mn02 in the boat.
12. Push the "Start" button.
13. Move the boat immediately into the vaporize zone.
14. After approximately 2 1/2 minutes, a signal will sound. Move the boat
into the pyrolysis zone at this time.
15. After approximately another 2 1/2 minutes, the signal will sound again in-
dicating the end of the integration period. The number in the display is
the concentration of TOC in mg/1. (Except AE and RAS samples, multiply
the AE by 2 and the RAS by 10 to determine TOC in mg/1.)
16. If more samples are being run at this time, push the "Baseline" button.
If the display is stabilized at less than 20, withdraw the boat to the
heat sink and repeat the procedure starting with No. 5. If the "baseline"
is greater than 20 or has not stabilized, repeat the procedure starting
with No. 4.
17. If no more samples are to be run, withdraw the boat to the heat sink and
leave in this position until the next use.
ATP ANALYSIS
Inasmuch as very little information was available on sample preparation and
procedural technique for use of the JRB ATP analyser in conjunction with plant
process control, it was necessary to develop this as part of the study.
Procedural Observations
1. The temperature of the Tris buffer must be maintained at 100 degrees
Celsius before and after the sample has been injected into the buffer.
2. The best boiling time for the samples must be determined to insure com-
plete lysing of the cells. If the samples are boiled for too great a time
period, the ATP will begin to be destroyed.
38
-------
3. After the samples have been boiled, distilled water must be added to the
system to bring the volume back to 10 ml.
4. When storing the samples for later analysis, they must be cooled and
frozen immediately after being boiled to prevent loss of ATP from the
system.
5. The samples must be thawed slowly and in an ice bath. They must also be
maintained in an ice bath until analyzed.
Conclusions Drawn from Observations
1. The greatest temperature reached in the buffer in the test tube was 98
degrees Celsius. If a higher temperature is necessary, as has been indi-
cated, a different solution needs to be used for heating the buffer. This
temperature did not change after 30 minutes of being in a boiled water
bath.
2. Contrary to JRB's instruction manual, the boiling time of the sample does
appear to be important. The ATP concentration of samples analyzed after
having boiled for 5 to 10 minutes was fairly constant. However, after 15
minutes, the ATP concentration was found to be 10 to 15 percent greater.
More testing indicated 15 minutes to be the best boiling time.
3. The thawing rate and the temperature the samples are thawed at also affect
the system. Samples thawed in an ice bath had 2 to 5 percent greater ATP
concentrations than samples thawed in a warm water bath. Furthermore,
samples kept in an ice bath lost 10 to 15 percent of their ATP after 1 1/2
hours while samples kept in the open at room temperature lost 20 to 30
percent of their initial ATP concentration during the same time period.
4. When samples were analyzed without freezing prior to analysis but rather
immediately after boiling them, AE ATP concentrations of about 10 ppm and
RAS ATP concentrations of about 35 ppm were obtained.
Procedural Problems
Certain errors were identified during the early stages of procedural develop-
ment and these are identified below.
1. Not injecting the samples directly into the boiling Tris-buffer but rather
getting it onto the sides of the test tubes.
2. Not allowing the buffer to become hot enough before injecting the samples.
This results in a slow killing of the bacteria and considerably low ATP
readings.
3. Boiling the samples unnecessarily long which might result in incorrect
readings.
4. Mislabeling the samples for storage.
39
-------
5. Using incorrect volumes and dilutions for the samples.
6. Letting the samples sit too long before fixing them for analysis.
7. Leaving the sample out to cool allowing possible sample contamination.
Air-borne bacteria may then consume a significant portion of the ATP be-
fore the sample analysis is performed.
Equipment and Procedures
Reagents
1. Tris Buffer (Tris [hydroxymethyl] amino methane hydrochloride)
Avg. M.W. = 147.7
Dissolve 2.954 g in distilled water and bring the volume up to 1 liter.
Adjust the pH to 7.75 with HC1. Sterilize in an autoclave at 121 degrees
Celsius for 15 minutes.
2.954 g/1 = 0.02m
2. Adenosine 5' - Triphosphate
1 mg - preweighed vial (Sigma Chemical Co.)
Dissolve the contents of the vial in sterilized Tris buffer and bring the
volume up to 10 ml. Keep the sample frozen.
ATP cone. = 100 mg/1
3. Firefly Extract
FLE-50 from Sigma Chemical Co. - premeasured vials.
Add approximately 15 ml sterile Tris buffer to the vial of firefly ex-
tract. Cap the vial, shake to mix the contents and refrigerate the sample
for at least 12 hours before use. Do not shake up the vial after it has
been in the refrigerator and is ready for use. The solids which have
settled out may cause unreliable readings in the ATP determination. Use
only the clear supernatant.
Preparation of Glassware
1. All glassware was washed with soap and water and rinsed at least 5 times
after the last trace of soap was noticed.
2. The glassware was then soaked in a 10% solution of HC1.
3. This was followed by 5 rinses in tap water followed by 5 rinses in dis-
tilled water.
40
-------
4. All glassware was heat dried in a drying oven and then cooled to room
temperature before use.
Equi pment
1. Autoclave
2. Heating mantle
3. Sand
4. Scintillation vials
5. ATP photometer - JRB Inc.
6. Ice bath
7. Automatic pi pet with disposable tips
Procedure
1. Samples
a. Aerator effluent diluted 1:2 with distilled water.
b. Return activated sludge diluted 1:10 with distilled water.
2. ATP Standards
a. 100 mg/1.
b. 9.9 ml sterile Tris buffer + 0.1 ml "A" = 1 mg/1.
c. 18.0 ml Tris buffer + 20 ml "B" = 0.1 mg/1.
d. 5 ml Tris buffer + 5 ml "C" = 0.05 mg/1.
e. 9 ml Tris buffer + 1 ml "C" = 0.01 mg/1.
Keep Standard "A" frozen until needed. Thaw slowly in an ice bath. Re-
tain Standards "C", "D" and "E" for the calibration of a standard curve.
Make them fresh daily or more frequently if needed.
3. Sample Preparation
a. Fill the heat mantle with sand.
b. Place 9.5 ml sterile Tris buffer into one scintillation vial and 9.8
ml sterile Tris buffer into another vial. Each vial should be pre-
marked (etched) at the 10 ml mark.
c. Place the vials into the sand and turn the heat mantle on high.
41
-------
d. Place a boiling stone into each vial and heat to boiling.
e. When the buffer in each vial is boiling, add 0.5 ml of RAS to the vial
with 9.5 ml buffer, and add 0.2 ml of AE to the vial with 9.8 ml
buffer.
f. Let the samples continue to boil for 6 minutes.
g. Remove the vials from the sand and place them into an ice bath
immediately.
h. When the samples have been cooled, bring the volume back to 10 ml in
each vial with distilled water.
4. Sample dilution.
AE = 1:2 initial dilution
= 1:50 sample dilution
1:100 final dilution
RAS = 1:10 initial dilution
= 1:20 sample dilution
1:200 final dilution
Sample Analysis
1. ATP Photometer
a. The photometer was allowed to warm up for at least 30 minutes prior
to use.
b. The attenuator control was set at 0.0.
c. The high voltage control was set at 6.u.
d. The zero control was set at 5.09. This gave an average of 18 CPM
"noise" when the photometer was run without a sample.
2. Sample Preservation
a. All samples, including the standards, were left in an ice bath during
the analysis procedure.
b. AE and RAS samples were usually discarded after analysis. However,
occasionally these samples were frozen so they could be analyzed at a
later date.
c. Standards were kept in the refrigerator until needed. They were pre-
pared fresh daily except the "A" standard which was frozen and thawed
as needed to prepare new standards.
42
-------
d. Sterile Tris buffer was kept In small volumes. After the buffer flask
had been opened and used, any remaining buffer in that flask was dis-
carded if not needed.
e. Firefly extract was kept in the refrigerator until needed.
3. Sample Analysis
a. Using the automatic pi pet, inject 0.5 ml of firefly extract into a
scintillation vial. Swirl the vial to disperse the liquid over the
bottom of the vi al.
b.
(1) Place the sample in the light chamber and place the cap on the
chamber.
(2) Put the timer on 6 seconds.
(3) Pull out the slide and push the start button.
(4) At the end of the integration period (6 seconds), record the num-
ber in the display as the background reading.
(5) Push in the slide and remove the vial from the chamber.
(6) Push the reset button.
(7) Push the 60 second timer.
(1) Using the automatic pipet with a clean pipet tip, withdraw 0.5 ml
of one of the samples (well mixed - either AE or RAS) or standards.
(2) Inject this sample in the vial containing firefly extract. At the
same time the sample is injected into the vial, push the foot
starting switch.
(3) Swirl the vial to mix the contents.
(4) Place the vial into the light chamber and place on the cap.
(5) Pull out the slide.
(6) At the end of the integration period, record the number in the
display as counts per minute for the sample.
(7) Push in the slide and remove the vial.
(8) Push the reset button and repeat this procedure for all samples.
43
-------
4. Determination of ATP Concentration*
a. For each series of samples, a calibration curve was determined from
the three ATP standards, U.I mg/1, 0.05 mg/1 and 0.01 mg/1.
b. If the background readings were extremely high or differed consider-
ably (higher) from readings obtained for other samples analyzed with
the same firefly extract, the vials were discarded (rewashed).
c. The reference points for all samples were determined by subtracting
the background reading from the counts per minute for that sample.
d. The slope of the standard curve was determined.
e. Using the 0.1 mg/1 standard as a reference point and the slope of the
curve, the counts per minute for the unknown samples were converted to
concentration (mg/1) and multiplied by the appropriate dilution factor
to determine actual ATP concentration in the sample.
f. As the ATP standard curve changes slowly with time for the same fire-
fly extract, the standards should be run several times at various in-
tervals if there are many unknown samples to be tested at the same
time.
The calibration method is as follows:
Slope (1) was found by running determinations on the known (K) concen-
tration of the three standards. The unknown (U) was analyzed and
Slope (2) compared with Slope (1) using the programmable calculation.
See Figure 9.
cinno I9\ - CPM(K) - CPM(U)
ilope ^' ~ ATP(K) - ATP(U)
ATP(K) - ATP(U) = CPM(K) - CPM(U)
AIKIM AIKIUJ slope (2)
ATP(U) - ATP(K) - CPM(K) - CPM(U)
ATP(U) - ATPIK) slope (2)
*For this study, a program was written for the programmable calculator which
determined the ATP concentration of the samples. By entering the background
and counts per minute for each sample and the standards, the ATP was read out
in mg/1.
44
-------
The assumption is made that Slope (1) equals Slope (2), then:
ATP(U)
= ATP(K) -
CPM
CPM
0.1
SLOPE cn
KNOWN(K) CONCENTRATIONS
05
APT
1.0
ATP(K),CPM(U)
ATP(U), CPM(U)
SLOPE (2\
UNKNOWN(U) CONCENTRATION
ATP
Figure 9. ATP standardization curves.
45
-------
RESPIRATION RATE ANALYSIS
Respiration Rate (RR) Procedure Use
The procedure used for this study was perfected over a period of time prior to
commencement of this study and has been found to yield repeatable results.
As described earlier, plant operators made these determinations 4 to 7 times
daily using the following equipment and procedures.
Equipment and Procedures
Sample
1. Aerator Effluent
Equipment
1. International Centrifuge Model HM-S - Use the same speed for all samples.
(Set on maximum speed.)
2. Gooch Crucibles
3. Whatman GF/A Filters 2.1 cm
4. YSI Model 51A with self-stirring BOD probe and Meter
5. Analytical Balance
Necessary Determinations
1. AE Suspended Solids Concentration (g/1)
a. Direct weight determinations made daily and a daily "factor"
determined.
b. Determined indirectly by centrifuge for every respiration rate using
the "factor."
2. Oxygen uptake during a 3-minute period.
AE Concentration (g/1)
1. Factor
a. Centrifuge two 40 ml samples of AE for 15 minutes at a set speed.
b. Determine the average ml of sediment in the two tubes and record this
number.
c. Determine the weight of a crucible and filter.
46
-------
d. Place the crucible on a suction apparatus, wet down the filter and
apply suction.
e. Filter 5 ml of AE through the filter.
f. When all the liquid has passed through the filter, remove the crucible
and dry it in a drying oven for 1 hour.
g. Remove the crucible and cool to room temperature in a dessicator.
h. Determine the weight of the crucible and dry residue.
i. To determine the daily factor:
(1) Wt. of residue in grams = (H)-(C).
{2) Wt. of residue (g) 1000 ml = wt f residue
U) ml filtered 5 ml x 1 wt* OT res1due
Wt. of residue (g/1) = F,rtnr (g/1)
ml of sediment in 40 ml tube rdctor ml
2. Indirect Determination
a. Centrifuge two 40 ml samples of the AE being used for the respiration
rate for 15 minutes at the set speed.
b. Determine the average ml of sediment in the tubes.
c. The AE concentration is approximately equal to the factor (m1
times the (ml sediment) = g/1
Oxygen Uptake
1. Aerate a well mixed sample of AE by pouring some AE (about 500 ml) into a
1 liter jar, capping the jar and shaking it for about 1 minute.
2. Pour the aerated AE into a 300 ml BOD bottle.
3. Insert the BOD probe, adjust the temperature, set the scale on DO and turn
on the probe stirrer.
4. When the DO concentration is declining at a constant rate (approximately
2 minutes after pouring sample into BOD bottle), begin the readings.
5. Take an initial reading of DO and then once per minute for the next 3 minutes,
6. Run the data (AE concentration g/1) and the DO readings through the calcu-
lator which is programmed to determine the respiration rate.
47
-------
RESPIRATION RATE DETERMINATION
1. The calculator program is set up to calculate the respiration rate in mg/1
Og per AE per hour. This is done by determining the Q£ depletion over
a 3-minute period, extrapolating this depletion for an hour period and
dividing the 02 depletion by the AE concentration in g/1.
2. Calculation:
AE RR = Oa + b - c - 3d)(6)
e
where a = first DO reading
b = second DO reading
c = third DO reading
d = fourth DO reading
e = AE suspended solids concentration (g/1)
CORRECTED SETTLOMETER TEST
Use of Settlometer
A 2 1 Mailory Settlometer with a procedure developed by plant lab personnel
was used to determine sludge settleability. Most applicable results were
obtained when the 5-minute value was used based on a sample containing 2 g/1
of solids. The solids were blended from Aerator Effluent (AE) and Return
Activated Sludge (RAS) according to the following procedure.
Equipment and Procedures
Samples
1. AE
2. RAS
Equipment
1. 21 Mailory Settlometer
2. Stop Watch
Procedure
1. Determine AE and RAS concentrations in g/1 using lab centrifuge and apply-
ing gravimetric/centfifuge calibration factor.
2. Determine the mixture of AE and RAS to be used in the Settlometer as
follows:
48
-------
a. Dilution Method A
(1) ml of RAS = j^
(2) ml of AE = Mf'° 9 x
ML g/I i
(3) If the sum of (1) and (2) is equal to or less than 2,000,
add these volumes and make up to 2,000 ml with tap water if
necessary.
(4) If the sum of (1) and (2) is greater than 2,000 use dilution
method B.
b. Dilution Method B
(1) Determine ml of RAS and AE to be used by the following
formula:
ml of RAS = (2.0QO)(AE 9/1 - ?)
ml ot RAb (AE g/1 -RAS g/1)
and ml of AE = 2,000 - ml of RAS
(2) After settlometer has been filled, stir contents gently and
reverse direction of stirring paddle to stop swirling motion.
(3) Time for 5 minutes and read sludge/water interface.
49
-------
SECTION 8
DISCUSSION OF INTERRELATED PARAMETERS
INTRODUCTION
In May and June 1975, 60 to 90 days prior to the end of the study period, pro-
cess trouble developed that required an analysis of several basic premises
considered as determing factors in process control. This section summarizes
the operating conditions during this period and presents the rationale for
altering earlier procedures based on conclusions drawn from evaluating process
response to conditions.existing during the period under consideration.
The principle cause of plant problems was believed to be attributed to a too
young, slow settling sludge. This resulted in poor quality effluent on a num-
ber of days when sludge bulking occurred in the secondary clarifier.
Prior to implementing a defined sludge wasting program, the plant frequently
had process failure due to improper sludge maturity. However, between July
1974 and May 1975 the program had been successful in preventing sludge bulking
for reason of a too "young" biomass. As an initial premise the use of RAS ATP
in place of RAS TOC in the modified F/M ratio control program was suspected
to be the principle cause of the failure.
Examination of previous data by plant personnel led to the theory that there
appears to be a natural maturation process which causes changes to take place
in the biomass during periods of limited food supply. These changes counter-
act the changes which take place during periods of abundant food supply. A
biomass which is trending toward greater "maturity" has been observed at
Hillsboro to have a faster settling rate as measured by the 5-minute corrected
settlometer.
The biomass, if allowed to become very "mature", is also observed to be less
active - its ability to quickly absorb organic material is reduced. Just the
opposite is observed to happen during periods of heavy wasting or heavy or-
ganic loading. The "maturity" of the biomass, and therefore the settling rate
of the floe, can be controlled by controlling the wasting rate. The degree of
control is limited because the wasting rate is also used to control the quan-
tity of biomass in the system. Furthermore, there is no easy way to predict
which way the settling rate will trend except by experience, much less calcu-
late the exact value, from the other process parameters. The way in which the
food input, growth rates, wasting rates, quantity of biomass in the system,
temperature, kinds of microorganisms present, etc., are related are obviously
going to be infinitely complex.
50
-------
Personnel at the Hillsboro plant therefore, use an information-feedback con-
trol method in which parameter values involved in the above relationships were
chosen, based on experience and intuition, to derive the equations from which
the actual wasting time is calculated. Control relationships are thus estab-
lished whereby the parameters (RAS TOC, 5-minute corrected settlometer) are
used as feedback information to calculate the control parameter (wasting rate)
which, in turn, controls the feedback parameters. The equations were refined
or adapted and after they were used for a time, refined again. A calculator
program thus evolved which proved to be successful in controlling both the
settling rate of the floe and the quantity of biomass in the system on a long-
range basis, without knowing the complex relationships between the process
parameters involved.
THE MAY-JUNE 1975 PROBLEM
It was assumed that a "too-young" biomass developed by too frequent and too
many mode changes. The course of events was hypothesized as follows:
Assume the aeration process is in 0-2. (The first digit refers to the
number of quadrants being used for sludge conditioning in a contact sta-
bilization mode of operation while the second digit refers to the number
of quadrants in which ML is being aerated.) See Figure 10.
RAS
Figure 10. Mode 0-2.
Primary effluent is entering at Quadrant A. RAS at Quadrant C. Quadrants A
and B contain only primary effluent, and Quadrants C and D contain mixed liquor.
51
-------
Consider the transition when the mode is changed to 0-4:
Figure 11. Mode 0-4
All quadrants are filled with mixed liquor. When the RAS feed point is moved
from Quadrant C to Quadrant A for several hours or more until the new desir-
able equilibrium is reached, an undesirable situation is created. Until the
mixed liquor is built up in Quadrants A and B, the F/M ratio in those quad-
rants is extremely high. This mode change appears to slow down the natural
maturation process of the biomass, replacing the balance (between that process
and a controlled growth) with wild new growth. This situation normally causes
little harm to the activated-sludge process because of its short duration.
However, during May, there were sufficient number of these situations to be in
the principle cause of the "too-young" sludge problem.
The mode change from 0-2 to 0-4 was only an illustration. Actually, 0-2 to
0-3 and 0-3 to 0-4 would result in the same situation but to a lesser extent.
The foregoing is not the only problem with mode-change transitions. An addi-
tional problem occurs when moving from a noncontact stabilization mode to a
contact stabilization mode. As an illustration, suppose the process was in
mode 0-4 and the AE RR indicated insufficient aeration time. Other modes
trending toward contact-stabilization can handle a much higher organic loading
but only after the solids concentrations have stabilized. During the transi-
tion, for about 24 hours, it is actually equivalent to moving to a lower mode.
Moving from 0-4 to 2-2 is similar to moving from 0-4 to 0-2 approximately the
first 24 hours. See Figure 12.
52
-------
AE
Figure 12. Mode 2-2.
By definition of the mode numbering and considering the transition, the con-
tact time is reduced from the "sludge detention time through four quadrants"
to "the sludge detention time through two quadrants." The RAS stabilization
time, which is increased from zero to "the RAS detention time through two
quadrants," eventually overcomes the problem by the overall ability to absorb
and metabolize organic material. This is conjectured to be due to the pro-
longed period of RAS aeration, during which time the bacteria have metabolized
the food sorbed during the time in the aeration basin and are ready to be fed
again. In other words, a large pool of hungry bacteria is created which has
the ability to metabolize a great quantity of food. But this is true only
after the new equilibrium concentrations are reached in the aeration basin.
It was observed that the AE RR rose even higher than the previous high value
indicating to the plant operator that a "new" mode was needed. This problem
seemed to be the root of the May/June difficulty because it caused plant oper-
ators, following the existing mode-change instructions, to make too-frequent
multiple changes to new modes thus creating even more instability in the
process.
In fact, moving from 0-2 to 3-1 in a short period of time combines both of
these undesirable situations resulting in double trouble until the new equil-
ibrium concentrations are reached.
PE
RAS
RAS
A/
JAE
PE
MODE 0-2
Figure 13. Mode 0-2 and 3-1.
MODE 3-1
53
-------
Table 14 shows the effects of excessive "mode shifting" during May 1975.
Table 14. MODE SHIFT FREQUENCIES
Date
5/12
5/12
5/13
5/13
5/14
5/14
5/15
5/16
5/17
5/18
5/18
5/19
5/19
5/20
5/20
5/20
5/21
5/22
5/22
5/23
5/23
5/24
5/25
5/25
Time
1:30 p.m.
5:30 p.m.
5:30 p.m.
9:00 p.m.
10:30 a.m.
1:30 p.m.
5:30 p.m.
5:30 p.m.
7:30 a.m.
10:30 a.m.
4:30 p.m.
11:00 a.m.
5:30 p.m.
1:30 p.m.
5:30 p.m.
8:30 p.m.
1:30 p.m.
1:30 p.m.
8:30 p.m.
12:00 a.m.
8:30 p.m.
8:00 p.m.
1:30 p.m.
4:30 p.m.
RAS
TOC
2980
3930
2730
1770
3250
2990
3200
—
—
3870
3280
2970
3150
3090
3630
4200
3370
3180
2930
3590
2330
3600
3350
3590
AE
TOC
460
708
772
728
550
858
756
—
—
866
706
830
722
656
830
662
644
984
804
826
962
1036
958
1000
Mode CSV5
0-2
0-4
1-3
2-2
1-3
0-4
0-3
0-4
1-3
0-4
0-3
0-2
0-3
0-2
0-3
0-4
0-3
0-2
0-3
0-4
1-3
0-4
0-3
0-2
530
510
500
500
540
480
430
480
550
550
900
510
610
600
610
660
670
750
670
650
670
600
850
850
Fin Eff. Sludge
JTU Surface
13.5
10.3
14.0
20.0
10.0
13.0
19.0
24.0
17.0
17.0
22.0
23.0
25.0
19.0
19.0
20.0
14.0
14.5
21.0
21.0
10.0
9.2
13.0
14.0
6.5
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
7.0
—
7.0
5.5
7.0
7.0
7.0
7.0
7.0
7.0
6.0
7.0
6.0
2.5
2.0
RR
20
23
23
27
9
11
9
19
19
8
8
6
16
12
26
22
10
13
18
13
15
8
6
4
Effective 7/23/75, the mode change rules were replaced by a calculator program
which enabled plant operators to accurately make mode changes from a new, more
elaborate set of rules discussed in Section 6 and displayed on Table 12. This
set of rules involves "the number of hours in the present mode," and is de-
signed to prevent too-frequent mode changes in one direction caused by not
allowing sufficient time in the contact stabilization modes for solids concen-
trations to stabilize. Furthermore, it was designed to prevent moving oppo-
sitely too fast, as this is almost always observed to cause higher turbidity
in the final.effluent (see Table 14). This was thought to be due to a large
net move of sludge from the aeration basin to the secondary clarifier during
mode shifts.
54
-------
DISCUSSION
Conditions Prior to May 1975
With the foregoing presentation in mind it is possible to expand on some of
the initial premises. In this regard two questions might be asked:
1. Why did the operation program fail to keep the biomass at the right
"maturity" for good settling?
2. Why didn't the condition occur during a similar situation in 1974?
These two questions are related. During the period in May and June 1974 an
entirely different set of mode-change instructions were being used. In fact,
the activated-sludge process was never put into a mode less than 0-4, and mode
changes were much less frequent than in May of 1975. That set of instruc-
tions was abandoned during October of 1974 because the plant was producing a
poor quality effluent of high turbidity, but no evidence of a young, bulky
sludge was observed. The main cause was, more probably, related to digester
supernatant return due to insufficient hauling of sludge. The poor quality
sludge resulted apparently because the process was in a "higher" mode than de-
sirable to obtain the best quantity of biomass in the system. The mode-change
program shown on Table 13 is a compromise between the two previous selections.
Another point to be considered is the wasting program which considers RAS-TX
concentrations before June 20, 1974, the wasting program maintained the RAS
TOC below about 4,000 with the exception of May 1974, when the RAS TOC aver-
aged between 4,500 and 5,000. A lower RAS TOC was maintained after that time
because higher RAS TOC's were correlated with higher final effluent turbid-
ities. The higher modes and higher RAS TOC's also result in more biomass in
the system which may lead to a poor quality effluent.
Considering the capacitance of the system, increased biomass tends to prevent
changes in the "maturity" of the biomass. Consequently, an increased biomass
will tolerate a larger quantity of food entering the aerator without effecting
the maturation rate significantly. On the other hand with a small amount of
biomass in the system, the natural maturation process will be easily disturbed
by a large quantity of food in the aeration basin and as a result new growth
will begin to predominate.
It has been observed that when the settleabi1ity of the floe slowly decreases
over a period of many days to weeks, conditions are normal and may be reversed
by either temporarily decreasing the organic loading (which can seldom be
done) or temporarily decreasing the wasting rate. But when settleability de-
creases rapidly, in a few days or less, conditions are extreme indicating that
the natural maturation process is apparently being slowed, or even stopped.
This seems to indicate that the types of new growth have been changed from the
normal spectrum of types of bacteria present in the "matured" biomass to the
quick-growing type which are undesirable and cause difficulty in efficient
sedimentation in the secondary clarifier.
55
-------
Conditions Existing in May 1975
The effects of the RAS ATP control coupled with modified F/M ratio decisions
for setting RAS flow rates appear to have influenced instability when the bio-
mass was in a "too-young" condition.
Control decisions caused high RAS flow rates which, when combined with the
problem of slow-settling floe, resulted in some serious, undesirable loss of
biomass from the system due to sludge bulking in the secondary clarifier.
The process of replacing the partly-matured biomass lost from the system with
new growth tended to perpetuate the problem.
As the settling rate of the floe became poorer, the biomass on the bottom of
the secondary clarifier became less dense. This caused both the RAS TOC and
the RAS ATP to decrease, thus creating a demand for higher return rates. The
problem stemmed from the fact that the RAS ATP values decreased more than the
RAS TOC. Using the RAS ATP parameter as a control, higher return rates were
demanded compared to using RAS TOC for control. Consequently, it appears that
decisions based on RAS TOC might have alleviated the sludge bulking problem.
Other Considerations in Process Control
On April 17, 197b, alum was added to the aeration basin for control of fila-
mentous organisms. Following this addition, on April 23rd, wasting of sludge
was much higher than desirable. During this period plant operation deviated
from the wasting program for several days. It is believed that the alum addi-
tion interfered with the wasting program by giving the erroneous indication
that there was more biomass present in the system than actually measured by
RAS-TOC tests. This resulted in excessive wasting of "matured" biomass from
the system and increased the 5-minute corrected settlometer readings. Curing
the first part of June 1975, heavy organic loading occurred after the "young"
sludge was already present. This caused another upward trend in the 5-minute
corrected settlometer further perpetuating the problem.
SUMMARY
The foregoing was presented to illustrate how the results of this study were
applied to plant operation and also how plant problems might be diagnosed by
the case of process information.
56
-------
GLOSSARY
AE - Aerator Effluent. This term is distinguished from ML (mixed liquor) by
the point of collection which is in the aeration tank immediately prior to
passing to the secondary clarifier. ML samples may be collected at various
defined points in the aerator after PE and RAS combine.
ATP - Adenosine Triphosphate. As determined by a JRB-ATP Photometer. ATP is
a substance found in all living cells and therefore provides a means of mea-
suring the portion of the activated sludge that is viable. See Section 7 for
the procedure used in this study.
Aerator Mode. A term expressing numerically the number of quadrants the RAS
and ML are receiving aeration. The first digit defines the number of quad-
rants in which RAS is reaerated or stabilized before mixing with PE to form
ML. The second digit is the number of quadrants ML receives aeration after
PE and RAS mix to form ML.
CSVt;. Corrected settling volume after 5 minutes settling in a 2000 ml set-
tlometer and adjusted to a concentration of 2 g/1 following the procedure out-
lined in Section 7.
ML - Mixed Liquor. Refers to any point in the aerator after PE and RAS are
combined.
Modified F/M. The relationship of food to microorganisms at the juncture
point of the PE and RAS. Values were defined either by PE-TOC for "F" and
RAS-TOC for "M" or RAS-ATP for "M".
PE - Primary Effluent. Sample collected at the distribution box prior to
entry into the aerator. The TOC of the PE sample is used as a measure of "F"
in the modified F/M determination.
PEF - Primary Effluent Flow. Measured in mgd and capable of being fed to any
or all of the aerator quadrants.
RAS - Return Activated Sludge. The context of the text delineates usage with
respect to flow, concentration, etc.
Reaeration (Stabilization). The condition whereby return sludge is aerated in
a quadrant or quadrants prior to mixing with primary effluent.
RR - Repiration Rate. Also, oxygen uptake rate as determined by a YSI DO
meter according to the method described in Section 7. Most frequently used
for finding the rate in the aerator effluent (AE).
57
-------
RSF - Return Sludge Flow. Measured in gpm in the modified F/M ratio or ex-
pressed as mgd when relating to ratio of RSF/PEF.
SCT - Solids Conditioning Time. A term expressing the relative time solids
remain in the aerator during one cycle from entry to exit.
$¥5. Settling Volume reading after 5 minutes settling on an undiluted sample
of mixed liquor or AE. May either be determined in a 2 liter Mailory settl-
ometer or 1000 ml graduated cylinder.
TOG - Total Organic Carbon. As measured by a Dohrmann Envirotech Model 50 TOC
Analyzer by the method defined in Section 7 of this report.
58
-------
REFERENCES
1. Biospherics, Incorporated, "Biomass Determination - A New Technique for
Activated Sludge Control," Water Pollution Control Research Series,
17050, EOY, January 1972.
2. JRB Incorporated, San Diego California, "JRB ATP-Photometer Instruction
Manual," August 1972.
3. JRB Incorporated, San Diego, California, "Methods for Data Analysis Using
the JRB ATP-Photometer," January 1973.
4. Petersack, J.R. and Richard G. Smith, "Advanced Automatic Control
Strategies for the Activated Sludge Treatment Process," Environmental
Protection Technology Series, EPA-670/2-75-039, Way 1975.
5. Rickert, David A. and Joseph V. Hunter, "Effects of Aeration Time on
Soluble Organics During Activated Sludge Treatment," Journal of Water
Pollution Control Federation 43, !_ 134.
6. Emery, Richard, Eugene Welch and Russell Christman, "The Total Organic
Carson Analyzer and Its Application to Water Research," Journal of Water
Pollution Control Federation 42, 9 1834.
7. Joyce, R.J., C. Ortman and C. Zickefoose, "How te Optimize An Activated
Sludge Plant," Water and Sewage Works. October 1974, pg. 96.
59
-------
APPENDIX
METRIC EQUIVALENTS
METRIC CONVERSION TABLES
Recommended Units
Onciipiion
Ltnglh
Art*
Volumi
Mm
Timi
Fore*
Unil
meter
kilomttir
millimeter
centimeter
micromftif
tquirt meter
square kilometer
square unlimettr
squvt milltmetir
hectare
cubic mtitr
cubic centimeter
liter
kilognm
gram
milligram
tonne
sicond
day
ytir
ntwton
Symbol
m
km
mm
cm
Htn
m'
km2
cm'
mm2
hi
m3
cm3
1
kg
mg
I
s
dly
yror
N
Comments
Basit SI unit
The hectare (10,000
m2) u i recognized
multiple unit and
will remain in inter
nali o nil use.
The lite* is now
recoo.nized es the
special name for
the cubic decimeter
Bane SI unil
1 tonne * 1.000 kg
Basic SI unil
Neither the day nor
the year is en SI unit
but both are impor-
tant.
The newlon is that
force that produces
an acceleration of
1 m/s2 in e mass
of 1 kg.
English
Equivalents
39.37 in • 3.28 fl •
1.09yd
0.62 mi
0.03937 in.
0.3937 in.
3537 X 103-103A
10.744 sq ft
• l.196sqyd
6.384 sq mi -
247 acres
0.155 sq in.
0.00155 sq in.
2.471 ecres
35.314 cu fl *
1.3079cuyd
0.061 cu in.
1. 057 qf 0.264 pi
- 0.91 X 10-* icre
ft
2.205 Ib
0.035 01 * 15.43 gr
0 .01543 jr
0.984 ton (long) -
1.1023 ton (short)
0.22481 Ib Iwiight)
• 7.5 poundils
Description
Velocity
linear
engulir
Flow (volumetric)
Viscosity
Pressure
Temueralure
Work, energy.
quantity of heel
Power
Application of Units
Description
Prtcipitition.
run -of I,
evaporation
Rivtrflow
Flow in pipes,
conduits, cnan-
ntls.avef weirs.
pumping
Otschirgtsor
abstractions,
yields
Usaoe of water
Density
Unit
millimeter
cubic mtttr
per ncond
cubic mtttr per
•tcond
liter ptr second
cubic meter
Ptr day
cubic meter
per year
liter ptr person
per day
kilogram per
cubic meter
Symbol
mm
it|3/s
m*
l/s
m3/day
m3/yr
I/person
day
kg/m3
Comments
For meteorological
purposes it may be
convenient to meas-
ure precipitation in
terms of mus/unit
ana (kg/m 3).
1 mm of rein •
Ikg/sqm
Commonly called
the cumec
1 l/s - 86.4 m3/diy
The density of
water under stand-
ard conditions is
1.000kg/m3or
1.000 j/l
English
Equivilents
35.314 cts
15.85 gpm
1.83 X I0-3,pm
0.264 gcpd
0.0624 Ib/cu ft
Description
Concentration
BOD loading
Hydraulic load
oaf unit area:
l.g. filtration
rates
Hydraulic load
per unit volume;
e.g. biological
Inters, lagoons
Air supply
Pipn
diameter
length
Optical units
Recommended Units
Unit
meter per
second
millimeter
per second
kilometers
per second
radians per
second
cubic meter
per second
liter per second
poise
newton per
iquere meter
kilonewton per
square meter
kilognm (lorce)
per square
centimeter
degree Kelvin
degree Celsius
joule
kilojoule
wan
kilowslt
joule per second
Symbol
m/s
mm/s
km/s
rad/s
m3/,
l/s
poise
Him7
kN/m'
kgl/cm'
K
C
J
U
W
kW
J/s
Comments
Commonly celled
the cumec
The newlon is not
yet well -known es
the unit ol lorce
and kg! cm? will
deeriy be used lot
sometime. In this
field the hydraulic
heed expressed in
meters is en iccept
eble elternelive.
Basif SI unit
The Kelvin end
Celsius degrees
era identical.
The use of the
Celsius scele is
recommended es
it is the former
centigrade scale.
1 goule • 1 N-m
1 Witt • 1 J/S
English
Equivalent)
3.28 IPS
0.00328 Ips
2 230 mph
15.850 gpm
•2 120 dm
15.85 gpm
0.0672'lb
lie II
0 00014 psi
0.145 psi
14 223 psi
5F
9
2.778 X 10''
kwhr-
3.725X10"'
hp-hr • 0.737S6
Mb- 9.48 X
10-4 Btu
2.778 kwhr
Application of Units
Unit
milligram per
liter
kilogram per
cubic meter
perdiy
cubic mnei
per squire meter
per day
cubic meter
per cubic meter
per day
cubic meter or
liter of free eir
per second
millimeter
meter
lumen per
square meter
Symbol
mg/l
kg/m3 day
m3/m? dey
m3/m3dey
m3/s
l/s
mm
m
lumen/m?
Comments
If this is con-
verted to a
velocity.!!
should be e>-
pressed in mm/s
(1 mm/s * 8E.4
m3/m2 dey).
English
Equivilents
1 ppm
0.0624 Ib/cu-ft
dey
3.28 cu ft/sq It
0.03937 in.
39.37 in. «
3.28ft
0.092 ft
cendle/sq ft
60
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-77-142
2.
4. TITLE AND SUBTITLE
TOC, ATP AND RESPIRATION RATE AS CONTROL P
FOR THE ACTIVATED SLUDGE PROCESS
7. AUTHOR(S)
Clarence Ortman, Tom Laib,
3. RECIPIENT'S ACCESSION- NO.
5. REPORT DATE
ARAMFTFRS September 1977 ( Issuing Date)
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
and C.S. Zickefoose
9. PERFORMING ORGANIZATION NAME AND ADDRESS
City of Hillsboro
Sewage Treatment Plant
Hillsboro, Oregon 97123
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laborator
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
15. SUPPLEMENTARY NOTES
Project Officer: Joseph F.
10. PROGRAM ELEMENT NO.
1BC611
11.-e©AWRA6TVGRANT NO.
R 802983-01-1
13. TYPE OF REPORT AND PERIOD COVERED
v— Cin.. OH Flnal
14. SPONSORING AGENCY CODE
EPA/600/14
Roesler (513-684-7617)
16. ABSTRACT
This research was conducted to determine the feasibility of using TOC
ATP and respiration rates as tools for controlling a complete mix activated
sludge plant handling a significant amount of industrial waste. Control
methodology was centered on using F/M ratio which was determined by the TOC
of the influent to the aerator and the TOC (or ATP) of the return sludge
Process control was affected manually and based on 5 to 7 determinations per
day. Respiration rates were used to indicate the need for increased or
decreased sludge aeration time. Process control decision making was aided
by the use of a programmable calculator. Process control information was
set up so that operators could input plant data and receive printed instruc-
tions for process settings. Functional programs included return rates,
mode changes, wasting rates, respiration rate and corrected settlometer vol-
ume.
17.
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
Waste treatment
Activated sludge process
Microorganism control (sewage)
Sewage treatment
Control
Process control
Control theory
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
EPA Form 2220-1 (9-73)
b.lDENTIFIERS/OPEN ENDED TERMS
Respiration rate
TOC
ATP
19. SECURITY CLASS (This Report)
UNCLASSIFIED
20. SECURITY CLASS (This page)
UNCLASSIFIED
c. COSATl Field/Group
13B
21. NO. OF PAGES
69
22. PRICE
51 •& U.S. GOVERNMENT PRINTING OFFICE !977-757-05b/b51l Region No. 5-11
------- |