PB88-102967
Critical Literature Review and
Research Needed on Activated Sludge
Secondary Claritiers
Montgomery (James M.), Inc., Pasadena, CA
Prepared for
Environmental Protection Agency, Cincinnati, OH
Sep 87
c
1
-------
EPA/600/2-87/075
September 1987
PDTTTr* AT T 1TI7D ATT1UT7 DUDTI?©
OKI xA1» jL^Jt JL Ju«!Vn/*.JL vJivJi x£J2» VJiif Vw
and
RESEARCH NEEDED ON ACTIVATED SLUDGE
SECONDARY CLARIFIERS
by
Rudy J. Tekippe
James M. Montgomery, Consulting Engineers, Inc.
Pasadena, California 91109-7009
Contract No. 68-03-1821
Project Officer
Jon H. Bender
Wastewater Research Division
Water Engineering Research Laboratory
Cincinnati, Ohio 45268
WATER ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This material has been funded wholly or in part by the United States
Environmental Protection Agency under Contract No. 68-03-1821 to James M.
Montgomery, Consulting Engineers, Inc. It has been subject to the Agency's
review, and it has been approved for publication as an EPA document. Mention
of trade names or commercial products does not constitute endorsement or
recommendation for use.
Many documents referred to in this report are unpublished but were included to
provide a more complete review of the literature.
ii
-------
FOREWORD
The U.S. Environmental Protection Agency is charged by Congress with protecting the
Nation's land, air, and water systems. Under a mandate of national environmental laws,
the agency strives to formulate and implement actions leading to a compatible balance
between human activities and the ability of natural systems to support and nurture life.
The Clean Water Act, the Safe Drinking Water Act, and the Toxics Substances Control
Act are three of the major congressional laws that provide the framework for restoring
and maintaining the integrity of our Nation's water, for preserving and enhancing the
water we drink, and for protecting the environment from toxic substances. These laws
direct the EPA to perform research to define our environmental problems, measure the
impacts, and search for solutions.
The Water Engineering Research Laboratory is that component of EPA's Research and
Development program concerned with preventing, treating, and managing municipal and
industrial wastewater discharges; establishing practices to control and remove
contaminants from drinking water and to prevent its deterioration during storage and
distribution; and assessing the nature and controllability of releases of toxic substances
to the air, water, and land from manufacturing processes and subsequent product uses.
This publication is one of the products of that research and provides a vital
communication link between the researcher and the user community.
This report presents a critical review of the literature regarding the design and operation
of activated sludge secondary clarifiers. It also presents the findings of a group of
clarifier experts who participated in a symposium that assessed what information was
needed from future clarifier research. These activities were intended to advance the
state-of-the-art of clarifier design and operation and provide direction for future
clarifier research.
Francis T. Mayo, Director
Water Engineering Research Laboratory
iii
-------
ABSTRACT
Secondary clarifiers perform an important dual function in the activated sludge process.
They must provide an overflow low in suspended solids and an underflow sufficiently
concentrated in suspended solids to maintain an activate biomass in the aeration basin.
Since these clarifiers are typically the final process in secondary treatment prior to
disinfection, their performance is essential in achieving final effluent quality-
Secondary clarifier design has been considered by many professionals to be one of the
major causes of inconsistent and unacceptable effluent quality from the activated sludge
process. In view of this, the U.S. EPA authorized a critical review of activated sludge
secondary clarifier literature, an engineering interpretation of this information, and the
preparation of this review document. In addition, the agency sponsored a symposium of
leading experts in clarifier design, research and operations to identify and prioritize
research needs in this field.
This report is organized into five sections. The first presents the background and format
of the symposium. It also explains the organization and preparation of the report text.
Sections 2 and 3 present the conclusions and recommendations, respectively. Section 4 is
the complete text of the literature review. The references used for the report are listed
in alphabetical order in Section 5.
This report was submitted in fulfillment of contract number 68-03-1821 by James M.
Montgomery, Consulting Engineers, Inc., under the sponsorship of the U.S. Environmental
Protection Agency. This report covers a period from October 1984 to September 1986,
and the work was completed as of September 1986,
iv
-------
CONTENTS
Disclaimer ii
Foreword iii
Abstract iv
Contents v
Figures vi
Tables ix
Acknowledgments x
1. Introduction 1
1.1 Background on Symposium 2
1.2 Format of Symposium 2
1.3 Format of Final Report 6
2. Conclusions 7
3. Recommendations 9
4. Literature Review 12
4.1 Sizing Clarifiers 13
4.2 Tank Shape 82
4.3 Inlets 89
4.4 Outlets 130
4.5 Sludge Removal Mechanisms 141
4.6 Operation Factors 163
5. References 175
v
-------
FIGURES
Number Page
1.1-1 Sample rating form for prioritizing research needs. 5
4.1-1 Process cost vs design flow secondary sedimentation. 14
4.1-2 Settling detention time versus temperature for Type III
clarification. 18
4.1-3 Clarification of the calcium carbonate suspension. El
4.1-4 Influence of blanket position and overflow discharge rate
on clarification performance of laboratory continuous pilot
scale sedimentation tank. 23
4.1-5 Relationship of suspended solids in effluent and overflow rate. 24
4.1-6 Effluent TSS vs overflow rate, Benningsen, 25
4.1-7 Suspended solid content in the effluent TSe as a function
of surface loading q^. 26
4.1-8 Suspended solids removal for circular and rectangular primary
clarifiers. 30
4.1-9 Primary clarifier performance - Metro Denver Sewage Disposal
District No. 1 Central Wastewater Treatment Plant. 31
4.1-10 Effect of secondary clarifier overflow rate on final effluent
suspended solids at Corvallis, Oregon. 32
4.1-11 Relationship between detention and overflow rate, "ideal" basin,
"long tube™ test No. 1. 34
4.1-12 Predictions for effluent quality with MLSS constant. 35
4.1-13 SS of effluent from each tank vs inflow variation. 36
4.1-14 Secondary clarifier performance vs hydraulic loading at
Livermore, California. 38
4.1-15 Performance response curves for conventional clarifiers and
flocculator-clarifiers. 38
4.1-16 Effect of overflow rate on effluent quality at Pruszkow. Poland. 39
4.1-17 Relationship between effluent suspended solids and overflow rate. 41
4.1-18 Relationship between solids surface feed and suspended solids
in effluent. 42
4.1-19 Comparison of loading rates required to achieve a 30 mg/L
effluent. 43
4.1-20 Effect of clarifier feed concentration on effluent suspended
solids. 47
4.1-21 Effect of solids surface flux on effluent suspended solids. 47
4.1-22 Effect of surface solids loading on effluent quality. 49
4.1-23 Effect of solids loading on clarification. 50
4.1-24 Example of test duration, resulting into equilibrium. 52
4.1-25 Example of test duration, resulting into sludge discharge. 52
4.1-26 The ATV guideline and test results as a function of surface
loading and sludge volume. 54
4.1-27 Allowable sludge volume loading. 55
4.1-28 Ampfing effluent suspended solids vs sludge volume loading. 56
vi
-------
Number
Page
4.1-29
Greding effluent suspended solids vs sludge volume loading.
57
4.1-30
Ruhleben effluent suspended solids vs sludge volume loading.
58
4.1-31
Limits of surface loading in upflow final settling tanks.
59
4.1-32
Graphical determination of settling tank performance.
61
4.1-33
Clarifier design and operating diagram.
66
4.1-34
Effect of clarifier depth and flocculator center well on
effluent suspended solids.
70
4.1-35
Predictions for effluent quality with sidewater depth constant.
72
4.1-36
Predictions of effluent quality with MLSS constant using
Chapman's model.
73
4.1-37
Turbidity in clear water zone in ZP-units.
74
4.1-38
Thickening and separation zone h^ as a function of surface
loading q^.
76
4.2-1
Typical dispersion curves for tanks.
86
4.3-1
Sketch of one of the original four prototype reaction baffles
constructed for the Holly Hill, Florida, clarifier.
94
4.3-2
Sketch of modified reaction baffle at Holly Hills, Florida.
94
4.3-3
Clarifier inlet diffuser.
96
4.3-4
Improvement in the transparency of the effluent from the
sedimentation stage by stirring flocculation.
96
4.3-5
Various conventional center feed inlet designs.
102
4.3-6
Simple, flocculating, inlet-energy-dissipation center feed
clarifier.
105
4.3-7
Single flocculator performance for Series I.
109
4.3-8
Maximum dimension of peak-sized floe.
109
4.3-9
Typical velocity pattern of center feed tank.
111
4.3-10
Pattern of solids deposition in activated sludge clarifications.
113
4.3-11
Radial distribution of dye concentration in Albuquerque, New
Mexico, secondary clarifier.
115
4.3-12
Radial distribution of dye concentrations in Morganton clarifier
with ring baffle/flocculation chamber.
115
4.3-13
Peripheral feed clarifier flow pattern.
119
4.3-14
Peripheral feed clarifier with "spiral roll" pattern of flow
distribution.
120
4.3-15
Kraus-fall ("downcomer") peripheral feed clarifier.
121
4.3-16
Final clarifier with raised inlet and diffuser modification.
123
4.3-17
Radial distribution of dye concentration in EBMUD secondary
clarifier, 40 min after the start of continuous dye injection.
126
4.3-18
Typical final clarifier velocity profiles, (a) unmodified clarifier,
(b) clarifier with raised inlet and diffuser, (c) clarifier
with raised inlet.
127
4.4-1
Typical circular clarifier weirs.
132
4.4-2
Effects of weir design on effluent solids.
134
4.4-3
Horizontal baffle peripheral weir.
134
4.4—4
Perimeter baffle at 45° slope.
135
-------
Number
Page
4.5-1 Concentric sludge hoppers. 143
4.5-2 Geometry of spiral collector. 144
4.5-3 Typical rectangular sedimentation tank with chain and flight
collector. 146
4.5-4 Typical airlift system. 147
4.5-5 Typical centrifugal system. 147
4.5-6 Dispersion curve of hydraulic (Tow-Bro) mechanism. 153
4.5-7 Akron, Ohio, dye tracer test results. 154
4.5-8 Normalized E curve - sludge tracer - Tow Bro. 155
4.5-9 Normalized E curve - sludge tracer - riser pipe. 156
4.5-10 Sequence of sludge tracer profiles illustrating flow of solids
in the final clarifier using the isoconcentric representation. 157
4.5-11 Typical sludge tracer dispersion curve produced at the return
sludge well. 158
4.5-12 Distribution of total suspended solids marked by La isotope in
cross section Czestochowa radial settling tank. 160
4.5-13 Detention time distribution of returned sludge measured in
run No. 7. 161
4.6-1 Influence of blanket position on clarification performance. 168
4.6-2 Typical configuration of step feed activated sludge plant. 171
4.6-3 Variation of sludge distribution with feed point, in a four-
pass step feed plant. 172
4.6-4 Variation of mixed liquor and return sludge concentrations with
time, following a change in feed point from Pass 1 to Pass 2
with QR/Q = 0.30. 173
viii
-------
TABLES
Number Page
1.1-1 List of Participants of EPA Symposium on Activated Sludge
Secondary Clarifiers. 3
4.1-1 Recommended Design Values for Overflow Rate and Side Water
Depth - Type IE Clarifiers. 16
4.1-2 Solids Loading for Type IE Clarifiers. 17
4.1-3 Empirical Secondary Clarification Models. 28
4.1-4 Summary of Performance Test Results. 46
4.1-5 Data of the Clarifiers and Process Conditions, During the
Test Periods. 53
4.1-6 Results of Solids Flux Tests. 64
4.1-7 Summary of Overflow Rate versus Performance References. 78
4.1-8 Summary of Solids Loading Rate (MLSS) versus Performance
References. 79
4.2-1 Comparison of Rectangular and Circular Clarifiers. 83
4.2-2 Summary of Advantages and Disadvantages for Several Clarifier
Configurations. 84
4.5-1 Summary of Advantages and Disadvantages for Several Circular
Clarifier Mechanisms. 148
4.5-2 Summary of Rectangular Clarifier Mechanism Advantages and
Disadvantages. 149
4.5-3 Comparison of Header and Scraper Mechanisms. 151
ix
-------
ACKNOWLEDGMENTS
This report represents a comprehensive review of what is needed in clarifier design, what
is know from the literature, and what requires additional research. It was written by
personnel from James M. Montgomery, Consulting Engineers, Inc., Pasadena, California.
Direction was provided by U.S. Environmental Protection Agency (EPA) personnel from
the Office of Research and Development in Cincinnati, Ohio. A team of technical
reviewers was formed. The members chosen were university professors, consulting
design engineers, state environmental control agency personnel, wastewater equipment
manufacturers, government official and operational consultants. The reviewers provided
valuable constructive critique of the report and participated in the symposium to
prioritize research needs. The membership of each group is listed below.
TECHNICAL DIRECTION
Project Officer: Jon H. Bender, WERL, EPA, Cincinnati, OH
Technical Directors James Kreissl, WERL, EPA, Cincinnati, OH
MANUAL PREPARATION
James M. Montgomery, Consulting Engineers, Inc.
Author: Rudy J. Tekippe
Production Staff: Frances E. De La Rosa, Donna M. Arcaro, Linda D. Dickson,
Ann J. Mancilla, Jack R. Bencomo
SYMPOSIUM AND TECHNICAL REVIEW
Symposium Participants:
o Mr. Charles S. Applegate, Rexnord, R&D Group, Milwaukee, WI
o Mr. David T. Chapman, Environment Canada, Environmental Protection
Service, Wastewater Technology Center, Burlington, Ontario, Canada
o Mr. Robert M. Crosby, Crosby 8t Associates, Piano, TX
o Mr. John K. Esler, NYSDEC, Albany, NY
o Mr. Walter G. Gilbert, U.S. EPA OMPC (WH-595), Washington, DC
o Dr. Thomas M. Keinath, Clemson University, Department of Environmental
Systems Eng., Clemson, SC
o Dr. Denny S. Parker, Brown 8e Caldwell, Walnut Creek, CA
o Dr. Rudy J. Tekippe, James M. Montgomery, Inc., Pasadena, CA 91109-7009
o Dr. Thomas E. Wilson, Greeley & Hanson, Chicago, IL
o Mr. Charles S. Zickefoose, Brown & Caldwell, Portland, OR
Facilitators:
o Mr. Jon H. Bender, WERL, EPA, Cincinnati, OH
o Dr. Thaddeus (Ted) W. Fowler, College of Education, University of
Cincinnati, OH
x
-------
SECTION 1
INTRODUCTION
Secondary clarifiers are key to the successful performance of the activated sludge
process. They serve to separate out the biological solids and produce a clear effluent
and to concentrate the settled solids for return to the aeration basins. Clarifiers have
served this purpose for decades, but there still remains divergence of opinion as to what
constitutes optimal design. To better understand this process and improve its cost
effectiveness and performance, the U.S. EPA authorized a literature review and
consensus developing research prioritization symposium to better understand the process
and serve as a guide for future research.
The objective of the literature review was to compile knowledge about what is needed
for a clarifier design, what is documented and known from the literature, and what
remains to be learned through further research. It is the intent that this report serve as
a guide for the conduct of and funding of such future research.
The authors of this report reviewed over 300 publications and summarized the findings
herein. The literature review led to identification of numerous areas for future
research. It therefore became necessary to develop a mechanism of prioritizing those
needs. To achieve this, a literature review report rough draft was prepared and
distributed to a number of environmental engineering professionals for review. Sub-
sequently, a symposium was conducted to develop a consensus list of prioritized research
needs. The results of this symposium are likewise contained in this report.
The literature review aspect of this project consisted of conducting a computer search of
pertinent references, collecting copies of the articles and texts, organizing the material
in accordance to this report outline, preparing the rough draft, reviewing and incorporat-
ing comments from peers, and finalizing this report. The symposium used to prioritize
research needs is discussed in the following subsections.
-1-
-------
1.1 BACKGROUND ON SYMPOSIUM
As discussed in this report, activated sludge clarifier research involves a large number of
complex interacting variables. It is therefore necessary to develop some objectives and
priorities for the conduct of future research. Furthermore, the findings of this and other
studies indicate that full-scale units are most valuable for obtaining meaningful
information. Since construction and modification, as well as testing of full-scale units,
is expensive as compared to pilot scale units, it is important that prioritization to
maximize the benefits gained from research investment be achieved.
Clarifier research must also address the issue of operational alternatives and equipment
reliability in addition to design. It is therefore necessary to involve personnel from a
variety of environmental engineering backgrounds to properly establish priorities. Thus,
a symposium attended by such diversified representatives was conducted to review the
findings of the literature review, establish prioritization criteria, and refine research
needs.
1.2 FORMAT OF SYMPOSIUM
Starting on July 15, 1986, a three-day symposium was conducted in Cincinnati, Ohio.
The facilitators and participants are listed in Table 1.1-1. The participants represent
consulting engineering firms, regulatory agencies, equipment manufacturers, operational
consultants and universities.
Prior to the symposium, the participants were given copies of the draft literature
review. Each was assigned approximately one-third of the report to be prepared to lead
discussions on and also asked to read the remainder. The first day of the symposium was
devoted to group discussions of the review on a section-by-section basis. Two tables of
five participants each were set up to encourage participation by all. The members of
each of the two groups changed as the topics of discussion changed from one clarifier
aspect to another. The facilitators consistently visited each of the tables and assisted in
stimulating the discussions.
On the second day, the entire group participated in developing a list of research ideas.
The lists of questions at the end of each of the six subsections of Section 4, a total of 56
-2-
-------
TABLE 1.1-1. LIST OF PARTICIPANTS OF EPA SYMPOSIUM ON ACTIVATED
SLUDGE SECONDARY CLARIFERS.
Participants:
Mr. Charles S. Applegate
Rexnord, R&D Group
P.O. Box 2022
Milwaukee, WI 53201
(414)643-2764
Dr. David T. Chapman
Environment Canada
Environmental Protection Service
Wastewater Technology Center
Box 5050
Burlington, Ontario
Canada L7R4A6
(416)336-4621
Mr. Robert M. Crosby
Crosby & Associates
1610 Avenue J
Piano, TX
(214)423-2735
Mr. John K. Esler
NYSDEC
Rm. 320
Albany, NY 12233
(518)457-7297
Mr. Walter G. Gilber
U.S. EPA OMPC (WH-595)
401 M Street, S.W.
Washington, D.C. 20460
(201)382-7292
Dr. Thomas Keinath
Clemson University
Dept. Environmental System Eng.
Clemson, SC 29631
(803)656-3276
Dr. Denny S. Parker
Brown & Caldwell
P.O. Box 8045
Walnut Creek, CA 94596
(415)937-9010
Dr. Rudy J. Tekippe
James M. Montgomery,
Consulting Engineers, Inc.
P.O. Box 7009
Pasadena, CA 91109—7009
(818)796-9141
Dr. Thomas E. Wilson
Greeley & Hansen
222 S. Riverside Plaza
Chicago,IL 60606
(312)648-1155
Mr. Charles S. Zickefoose
Brown & Caldwell
11625 S.W. 66th Avenue
Portland, OR 97223
(503)639-0626
Facilitators;
Dr. Thaddeus (Ted) W. Fowler
College of Education
University of Cincinnati
Cincinnati, OH 45221-0002
(513)475-4671
Mr, Jon H. Bender
Water Engineering Research Lab
U.S. EPA
Cincinnati, OH 45268
(513)569-7620
-3-
-------
primary questions, were identified. The first basic questions, relating to prediction
equations (or "rating curves") based on numerous independent variables for different tank
geometries, included numerous secondary questions.
Following the listing of research ideas, criteria were developed for their prioritization.
The criteria listed consist of the following:
1. Improvement in effluent suspended solids per dollar invested
a. Research (1.7)
b. Application (4.5)
2. Improvement of existing plants (upgrading) (3.9)
3. Improvement of reliability (process and mechanical) (3.1)
4. Status and depth of existing knowledge (3.8)
5. Feasibility of implementing the research (3.4)
6. General applicability (4.5)
7. Marketability of research (1.9)
8. Institutional constraints (1.0)
Each of the above criteria was rated on a scale of 1 to 5, 5 being most important. The
number in parenthesis behind each of the listed items above represents the average
points scored by the participants.
The prioritization criteria were screened from the list of eight to the following list of
threes
1. Cost effectiveness
2. Status of knowledge
3. Applicability to new and existing plants
After further discussion by the group as a whole, it was decided that the criteria could
further be reduced to Status of Existing Knowledge and Relative Importance. A rating
form was prepared on this basis and used to prioritize research ideas. Figure 1.1-1 gives
a sample of this rating form. A separate sheet was filled out for each participant for
each clarifier aspect. The averages for each of two 5-member tables were then
calculated. Ideas that did not fall in the upper right quadrant, that is, having an
-------
RATING FORM
FOR
SXMPOSIDH OH PRIORITIES FOR RESEARCH FOR
CLARIPIER DESIGN
CLARIFIER ASPECT .<"&f firQ^
GROCJP HO. 2-
GROOP
MEMBERS 1~C ki'fp*-
if
£V in <
DATE 7- ft ~ &C
Cr o$ Joy
Uj\ls c n
RATING MATRIX
Little
DEPTH OF
KNOWLEDGE
Extensive
10
&¦<
&
0-
o
12 3
Mot Xaportant
9 10
Critical
IMPORTANCE
CRITERIA
Depth of Knowledge
WEIGHTIBG
FACTORS
TOTAL
POINTS
Inportance
Figure 1.1-1. Sample rating form for prioritizing research needs.
-------
importance factor of more than 5, and a depth of knowledge less than 6, were dropped
from further consideration.
Based on this screening procedure, the original list of research ideas was reduced to 32 in
number. A full group discussion was then held evaluating the 32 ideas. This group
discussion further reduced the list to 14. This final list is presented subsequently in
Section 2 of this report.
The third day of the symposium was devoted to making assignments of people responsible
for presenting the results to the EPA staff, preparing the presentation materials, and
making the presentations.
1.3 FORMAT OF FINAL. REPORT
This Final Report covers the material of the literature review as well as the proceedings
and findings of the symposium to prioritize research needs. The first section describes
the overall project and explains the activities of the symposium. Section 2 presents the
prioritized list of research needs. The list was prepared by the participants of the
symposium. Section 3 presents recommendations for the conduct of research needs
identified in Section 2. Section 4 consists of the literature review modified from its
rough draft by incorporation of the comments received from the symposium participants
who also served as peer reviewers required by the U.S. EPA. Section 4 is subdivided into
six subsections, each of which relates to a clarifier aspect discussed in the draft
literature review and at the symposium. At the end of each subsection is the list of
initially-proposed research ideas suggested during the first listing at the symposium. The
final section, Section 5, presents the final bibliography. It has been updated to include
references found by the peer reviewers as well as additional references found by the
authors since the symposium.
It is important to understand that the critical literature review is not intended as a
design guide or manual. Its primary purpose is to serve as a guide to prioritize future
research.
-------
SECTION 2
CONCLUSIONS
Based on the findings in the literature and the consensus achieved through the trans-
actions of the prioritization symposium, the following list of 14 key questions are posed
to guide future research about activated sludge clarifiers:
1. What are the rating curves based upon effluent total suspended solids versus
several key significant parameters for each of the following types of clarifiers?
A. Circular-center feed peripheral removal
B. Circular-peripheral feed peripheral removal
C. Circular-peripheral feed inboard removal
D. Rectangular-end sludge removal
E. Rectangular-center sludge removal
F. Rectangular-front sludge removal
Key Significant Parameters (including, but not limited to)
A.
Overflow rate
B.
Depth
C.
Solids loading
D.
Underflow solids concentration
E.
Sludge blanket depth
F.
Settling characteristics
G.
Dispersal solids loading
H.
RAS rate
2. What is the frequency and occurrence of peak flows and short term transients and
how do they affect the stability of clarifiers?
-------
3. What is the rational design variability of flux curves from plant to plant?
4. Are foreign clarification technologies applicable to U.S. plants?
5. What are the variables that best correlate with the performance of the best
clarifiers?
6. What are the flow patterns within various clarifier shapes and configurations which
affect the transport of solids?
7. What is the relationship between inlet design and floe formation and break up?
8. Can an optimally design flocculation structure/inlet cost effectively upgrade an
existing rectangular and/or circular clarifier?
9. What are the shear characteristics of clarifier inlet structures and their effect on
flocculation?
10. What is the effect of aerator flocculation and break up on clarifier performance?
11. How does the performance of in-board launders and peripherally baffled weirs
compare?
12. How do properly designed and configured baffles affect clarifier performance?
13. What is the optimum location of effluent weirs for different clarifier con-
figurations?
14. How do the sludge residence times created by different types of sludge removal
mechanisms affect clarifier performance?
-------
SECTION 3
RECOMMENDATIONS
The literature review and symposium identified a number of basic research needs that
include such areas as data collection protocol, multiple-profession involvement, and
improvement of definitions as well as the list of technical topics presented in Section 2.
These other basic issues are outlined below as recommendations relating to research of
the high priority technical areas.
1. The profession should join to standardize on definitions. The following are
examples to represent terms that lack consistency in the literature:
• Detention time (based on overflow vs inflow)
• Depth (side water vs average)
• Sludge blanket (what defines the top?)
• Blanket depth (measured from surface or floor)
2. Future research should be performed on full-scale tanks. Pilot-size units do not
adequately simulate some variables such as eddy currents, large-scale roll patterns,
wind effects, and density currents,
3. Future research must recognize that there are numerous important variables that
affect performance. A number of these variables exhibit strong interactions with
other variables. Some important variables, such as flow fluctuations, aeration tank
turbulence and flow splitting occur elsewhere in the treatment plant but still have
an effect on clarifier performance.
4. Proper protocol should be established and followed for conducting additional
research on clarifiers. The following points should be considered:
-9-
-------
• Documentation of all conditions is needed. Dimensions, flows, tem-
peratures, wind velocities, SVI and SSVI of tie sludge, MLSS concen-
trations, etc., should be defined even though some may not be the
variables of most interest in a given study.
• Proper experimental design using modern statistical techniques should
be followed in the planning of research. Factorial and fractional-
factorial designs are examples.
• Statistical model building techniques are recommended in the
formulation of equations to relate performance to independent vari-
ables.
• Enough data points should be measured to enable definition of
statistical significance.
¦ Analysis of variance techniques are recommended to help establish
statistical significance of results.
• Research publications should include ideas for further research to help
guide others who may be pursuing parallel research.
• To the extent possible, research results should be subjected to peer
review and published in well-recognized journals or reports, such as
EPA reports to share the knowledge gained and maximize the benefits
to environmental engineering projects.
5. Research on clarifiers should be pursued on several fronts. The following roles are
listed to illustrate some unique advantages of different parties:
• EPA - Serve as a central knowledge bank and potential source of
supplementary funding or knowledge of funding. Provide continuity.
• Universities - Provide guidance and manpower in research planning,
statistics, testing techniques, and data collection.
-10-
-------
• Consultants - Document the performance of tanks they design and
incorporate research findings into new designs so that clarifies design
experiences forward evolution.
• Governmental agencies (owners) - Collect data from operating facilities
and conduct research to optimize the performance of existing plants.
Unbalanced load testing and tracer studies represent examples.
• Equipment manufacturers - Assist in funding and conducting research
efforts regarding mechanisms. Publish as much of the existing in-house
data collected as possible (it is recognized that some data may be
proprietary to gain competitive advantage).
Representatives from all of the above should continue to work together to help further
research needs prioritization and achieve implementation.
-11-
-------
SECTION 4
LITERATURE REVIEW
An extensive review of activated sludge clarifier literature was performed as a major
part of this project. Over 300 articles were reviewed and used to prepare the review
presented in this section. The material was grouped in accordance with the following six
categories:
Sizing
Tank Shape
Inlet
Outlet
Sludge Removal Mechanisms
Operations
A separate subsection is devoted to each of these topics. The references given in the
text are listed in alphabetical order in Section 5.
A number of other aspects were identified that could have been used as section headings.
Skimming devices, drive mechanisms, depth and materials would be examples. These
other aspects were either considered to be of less importance than the six listed above or
incorporated into those listed.
-12-
-------
4.1 SIZING CLARIFIERS
The success of activated sludge wastewater treatment depends to a large degree 011 the
performance of secondary clarifiers to separate and concentrate the system mixed liquor
solids. The objectives in design of the secondary clarifiers include:
1. Remove the suspended solids, producing a clear effluent
2. Thicken the return solids
3. Minimize construction costs
4. Minimize odor release
5. Remove scum
6. Ensure reliability, permitting the processes to remain in service nearly 100
percent of the time
7. Minimize maintenance costs
8. Have an attractive appearance
9. Be structurally sound
Proper sizing of the secondary clarifiers effects the first three of these objectives. It is
obvious that the costs of clarifiers increase with size. This relationship is illustrated in
Figure 4.1-1 in which the construction costs are plotted against design flow rate for a
battery of secondary clarifiers. These data were taken from records of construction
costs maintained by James M. Montgomery, Consulting Engineers, Inc. As indicated by
the equation on the figure, costs increase proportionately to flow rate taken to the 0.7
power. There is therefore some economy of scale, but nevertheless there is a significant
economic advantage in reducing the size of these process units. Obviously excessive
reductions will hinder achievement of the objectives of removing suspended solids and
thickening those solids that settle. The objective of this section is to discuss what needs
to be done to properly size activated sludge clarifiers based on what is known and is
presented in the literature.
The needs of design engineers may be summarized as information required to predict
performance as a function of the given and controllable values of the independent
variables addressed during the design phase. Specifically, design engineers would like to
-13-
-------
15.8
10.0
8.0
6.0
4.0
s
K
< 2.0
-i
o
o
IL 1.0
° 0.8
CD
§ 0.6
~
si 0.4
2
~»
SB
O 0.2
0.1 •
0:08
0.06-
H
* 0.04-
0.02-
0.01
1Q0
DESIGN FL,0W ED
0.1
0.6 1.0 6.0 10.0
DESIGN FLOW (MOD)
60.0 100.0
Figure 4.1-1. Process cost vs design flow secondary sedimentation
(Ref. JMM data, 1983).
-14-
-------
be able to use a curve or mathematical formula to predict the effluent suspended solids
and underflow concentration of a clarifier once they are given information on
temperature, peaking factors, flow rates, mixed liquor suspended solids and other
independent variables. The following paragraphs discuss the amount of information
currently available to help engineers properly size activated sludge clarifiers.
The current predicament of U.S. engineers designing municipal activated sludge clari-
fiers is that they are encouraged to follow regulatory agency guidelines. Such guidelines
can be challenged, but are often not because of the lack of data to demonstrate
inadequacies and because of the lack of time or money available to collect such data.
Regulatory guidelines often do not apply to industrial plant designs and more research is
often done in this area to determine optimal designs and assure adequate performance.
Typical regulatory guidelines for zone settling (Type HI, e.g., activated sludge) are
presented in Tables 4.1-1 and 4.1-2 (308). The sizing procedures involve determining
surface areas based on average or peak hydraulic flow rate and selection of depths which
lie within the 3.7 to 6.1 m (12 to 20-ft) range for most tanks. In addition, surface areas
are sized to accommodate average and peak solids loading (Table 4.1-2). The larger of
the two sizes determined for clarification and solids loading is used for design. These
procedures are obviously very simple and incorporate only a few independent variables,
namely flow rate (average and peak), suspended solids loading and sludge volume index.
When certain design conditions are abnormal, sizing obtained by the above method is
sometimes adjusted by multipliers or factors of safety. An example of this is illustrated
by the multipliers shown in Figure 4.1-2 presented by Reed and Murphy (249).
From reviewing the literature, it was found convenient to discuss the surface area
requirement for clarification and thickening and subsequently depth requirements. This
section, therefore, is organized around these topics and concludes with a summary.
SURFACE AREA - CLARIFICATION
Two parameters that are used to determine surface areas for clarifier design are surface
overflow rate and solids loading. Each is discussed separately in the following
subsections.
-15-
-------
TABLE 4.1-1. RECOMMENDED DESIGN VALUES FOR OVERFLOW RATE AND SIDE WATER DEPTH
TYPE HI CLARIFIERS.
Source
Re commendations and Comments
EPA Design Manual
"Suspended Solids Removal"
"Ten States Standard" (1978)
"Manual of Practice No. 8"
Settling following trickling filtration:
16 to 24 at average Clow, 41 to 49 m3/m^d at peak flow,
side water depth 3 to 3.7 m.
Settling following air activated sludge (excluding extended
aeration): 16 to 33 m^/m^-d at average flow, 41 to 49 at
peak flow, side water depth 3.7 to 4.6 m.
Settling following extended aeration: 8 to 16 at average
flow, 33 m^/m^-d at peak flow, side water depth 3.7 to 4.6 m.
Settling following oxygen activated sludge with primary settling:
16 to 33 m^/m^-d at average flow, 41 to 49 m3/m'*d at peak flow,
side water depth 3.7 to 4.6 a.
Hydraulic loading of final clarifiers shall be based on peak hourly
flow rate and shall not exceed:
Settling following conventional, step aeration, contact
stabilization, and carbonaceous stage of separate stage
nitrification: 49 m^/m^-d.
Settling following extended aeration: 41 m^/m^d
Settling following separate nitrification: 33 m^/tn^'d.
Side water depth: 3.7 m or greater.
{Consideration should be given to flow equalization to reduce peak
hydraulic loadings.)
"An overflow rate of 33 m^/m^-d based on average flow can be
expected to result in good separation of liquid and solids. The
design engineer must also check the peak hydraulic rates that will
be imposed on the settling basin."
Recommended side water depth dependent on tank diameter,
varies from 3 to 4.6 m.
"U.S. Army Technical Manual"
Recommended overflow rate depends on plant flow rate.
Ranges from 4 m^/m^-d (average flow) and 8 (peak flow)
at flow rates between 0 to .0004 m*/s to 24 (average
flow) and 33 m^/m^-d (peak flow) at flow rates above 0.43 m^/s.
Recommended side water depth dependent on tank dimensions,
varies between 2.4 and 4.3 m.
"Naval Facilities Design Manual"
Size units according to criteria requiring largest area.
Maximum 24-hour flow, all units in service: 33 m^/m^-d.
Peak flow with all units in service: 49 m3/m2-d.
Peak flow with one unit out of service: 61
Side water depth for clarifiers following activated sludge: 3.7 m.
Side water depth for clarifiers following other processes: 3.0 m.
Ref. 308
-16-
-------
TABLE 4.1-2. SOLIDS LOADING FOR TYPE HI CLARIFIERS.
Source Recommendations and Comments
"Manual of Practice No. 8"
EPA Design Manual
"Suspended Solids Removal"
"Naval Facilities Design Manual"
"Ten States Standard" (1978)
Recommended solids loading based on sludge volume index, SVL
Values and method of sludge removal from 49 kg/m^-d at an SVI
of 300 to 390 kg/m2,d at an SVI of 100.
Allowable loadings generally governed by sludge thickening
characteristics in cold weather. Settling following air activated
sludge including extended aeration:
98 to 147 kg/m2*d average, 245 kg/m2*d peak.
Settling following oxygen activated sludge with pprimary settling;
123 to 172 kg/m^-d average, 245 kg/m2*d peak.
Determine limiting solids loading on final tanks by methods
indicated on EPA "Suspended Solids Removal." Determine
required final tank area necessary to receive the maximum 24-
hour solids load from aerator with one tank out of service,
without exceeding the linrting solids loading.
Solids loading for clarification following all activated sludge units
shall not exceed 245 kg/m2*d, peak.
Ref. 308
-------
Temperature. 'F
2 00
75
1 50
1.25
6000
^9/1 Of tlm^frr
1.00
0.75
0
5
20
Temperalure. °C
Figure 4.1-2. Settling detention time versus temperature for Type HI
clarification (Ref. 249).
-18-
-------
Surface Overflow Rate
Hie concept of using overflow rate, the flow leaving the clarifier divided by its surface
area, for sizing clarifiers has been attributed by many authors such as Camp (36) and
Fitch (108) to Hazen (133) dating back to 1904. The latter's work concluded that for
discrete particles settling in a liquid flowing horizontally, removal of suspended solids
was inversely proportional to the tank overflow rate and directly proportional to the
surface area of the tank. Camp published three classical papers (36,37,38) on clarifier
design during the time span from 1936 to 1953. In the latest article (38), he states
The settling of the mixed liquor in the activated sludge process is
not a typical free settling process. The process consists primarily
of compaction of the sludge blanket which may occupy a consider-
able portion of the depth of the tank. Even so, final settling tanks
for the activated sludge process should be designed on the basis of
overflow rates rather than detention periods with due allowance
being made for the depth to be occupied by the sludge blanket.
Overflow rates in common use in final settling tanks of the
activated sludge process range from about 1.7 to 5.1 m/h (1,000 to
3,000 gpd/sq ft).
None of the three publications of Camp present full-scale data showing that effluent
suspended solids from activated sludge clarifiers correlate more closely to surface
overflow rate than detention time. The only continuous flow clarifier tested by Camp
and reported on was a small rectangular tank 67,5 cm (20-5/8 in) wide, 14.3 cm (5-5/8 in)
deep and less than 0.9 m (3 ft) long. It is known that Camp visited a large number of
water and wastewater treatment plants in his career; however, his publications do not
contain data that he collected from full-scale plants. He used data published by others
and a rather extensive theoretical development.
Largely due to the research and publications by Camp, primary consideration has been
given to overflow rate in activated sludge clarifier design. The guidelines presented
above in Table 4.1-1 are typical. The 1977 issue of WPCF Manual of Practice (MOP) No.
8, "Wastewater Treatment Plant Design," is used as a guide by many practicing engineers
in the United States (307). It gives the following recommended overflow rates for
activated sludge clarifiers:
-19-
-------
1.36 m/ii (800 gpd/sq ft) at average flow
2.38 m/h (1,400 gpd/sq ft) at 3 h sustained peak flow
2.72 m/h (1,600 gpd/sq ft) at 2 h sustained peak flow
MOP No. 8 recommends that the overflow rate be reduced by 0.17 m/h (100 gpd/sq ft)
for each 0.3 m (1 ft) of sidewater depth less than those recommended for a given tank
diameter. No data are presented to show a correlation between effluent suspended solids
and overflow rates. The text states that many documented cases where peak rates of 2.7
to 3.1 m/h (1,600 to 1,800 gpd/sq ft) do not overload the settling basin. It also states
however there are as many or more cases where plant efficiency suffers at much lower
overflow rates.
Other references were sought in this study to establish a better correlation of effluent
quality and overflow rates. One of the most direct correlations was that presented by
Parker (229) for the Livermore, California, full scale circular clarifiers. Effluent TSS
were shown to vary almost linearly from 18 mg/1 at 1.0 m/h (600 gpd/sq ft) to 40 mg/l at
1.23 m/h (725 gpd/sq ft).
In conducting thickening studies, Dick (84) with a suspension of calcium carbonate
experimented with varying overflow rates to determine the effect on effluent turbidity.
The thrust of this study, however, was the examination of thickening properties;
therefore, extensive research on effluent turbidity versus overflow rate was not
conducted. The results of his brief experiments are illustrated in Figure 4.1-3. The
results show a strong correlation between overflow rate and effluent turbidity.
Specifically, as overflow velocity increases from 0.3 m/h to 1.5 m/h (177 gpd/sq ft to
883 gpd/sq ft), effluent turbidity increases from approximately 8 to approximately 45
mg/1. Dick's experiments were conducted in the laboratory using Plexiglas settling and
thickening columns. For the continuous flow experiments, the columns were fed from
the top in the center. The small diameter of the columns precluded the large eddy
currents and velocity patterns known to exist in full-scale activated sludge clarifiers.
Dick's work was published in March 1972.
Similar studies were later conducted by Dick and Johnson (85) from 1975 to 1977. The
results were published in 1982. As with Dick's earlier study, a calcium carbonate
suspension was settled and thickened in the laboratory continuous flow pilot plant
-20-
-------
60
50 -
40 -
i
a
m
DC
3 30 -
H
5
o
Jk
u.
ffi 20
10
A
$
0.5 1.0 1.5 2.0 2.5
OVERFLOW VELOCITY, cm/m&i
3.0
Figure 4.1-3. Clarification of the calcium carbonate suspension (Ref. 84).
-21-
-------
cylinder. The results of overflow concentration plotted against overflow rate are shown
in Figure 4.1-4. Again, a strong correlation of increasing effluent suspended solids with
increased overflow rate is illustrated.
Dick and Johnson {85} also conducted pilot scale studies at the Hatfield, Pennsylvania
Wastewater Treatment Plant. Mixed liquor from the flow equalized Hatfield facility
were metered into the pilot clarifiers at overflow rates between 0.75 to 1.35 m/h (450 to
800 gpd/sq ft). The investigators could not determine any significant correlation
between overflow rate and effluent suspended solids in the pilot scale facility. The
design of the pilot plant's sedimentation tank inlet, however, was found to be a major
lector in influencing clarification. Enlarging the pilot inlet, thereby reducing inlet
velocities, reduced effluent suspended solids concentrations from as high as 60 mg/1 to
approximately 10 mg/1. Thus, at this small scale, inlet geometry was far more
significant to clarifier performance than overflow rate in the ranges cited above.
Variations in the overflow rates of the full-scale clarifiers at Hatfield were not reported
in the reference.
Pflanz (234) reported the results of investigating activated sludge rectangular clarifiers
at Bennigsen and Fallingbostel and a circular tank at Celle, all in West Germany. The
tanks were exceptionally shallow with depths of 1.2 m (4 ft), 1.55 m (5 ft), and 2.27 tn
(7.5 ft), respectively. The correlation between effluent suspended solids and overflow
rate for the Fallingbostel Treatment Plant is illustrated in Figure 4.1-5 for three
different SVI values. In the very low overflow rate range of approximately 0.267 m/h
(150 gpd/sq ft) to 0.535 m/h (300 gpd/sq ft) , the suspended solids in the effluent for a
mixed liquor of SVI equal to 155 increased from 6 to 17 mg/1 for a mixed liquor
suspended solids of approximately 2,000 mg/1. The impact was more dramatic for mixed
liquors with a higher SVI (e.g., SVI = 306). Data from the Bennigsen Treatment Plant
were replotted by the authors of this report and are shown in Figure 4.1-6. These data
indicate that as overflow rates increase from 0.17 to 1.0 m/h (100 to 600 gpd/sq ft),
effluent suspended solids increased from 5 to nearly 35 mg/1. These full-scale data
clearly show a statistically significant correlation between overflow rate and effluent
suspended solids. This may in part be due to blanket interference for such shallow tanks.
Ditsios (88) conducted pilot scale tests on a rectangular tank 18 m (59 ft) long, 1 m
(3.28 ft) wide, and 1.75 m (5.7 ft) deep. His results, shown in Figure 4.1-7, show that
effluent suspended solids concentrations increase with increasing overflow rates.
-22-
-------
80
a
i 70
0
P 60
<
BC
1 50
O
§40
m
to
O
o 30
a
O
| 20
.J
ll-
CC
£ 10
O
1 1
1111
1 - T 1
A
O
m
&
~
m
* 0
-
A
-------
CO ^
Q ?
FALLINGBESTEL
SVI =155
— 3.01
-2.12.
"1.59
0
0
SVI = 201
- 3.01
— 2.12
" 1.59
SVI-306
V
— m
v —
— 2.12
-1.59
0.535 0.401
2.9 3.9
0.267 OVERFLOW RATE(m/h )
5.8 DETENTION TIME (h)
Figure 4-1-5. Relationship of suspended solids in effluent and overflow
rate (Ref. 234).
-24-
-------
OVERFLOW RATE (m/h)
0.1
0.2
0.7
0.9
0.3
0.4
0.5 0.8
0.0
0.8
1.0
TEMP-13-15° C
35
ML8S-3000 MG/L
BENNINGSEN. GERMANY
30
^ 85
u.
15
10
5
0
100
500
0
300
400
600
200
OVERFLOW RATE (gpd/sq.ft.)
Figure 4.1-6. Effluent TSS vs overflow rate, Benningsen (Ref. 234).
-25-
-------
50
0.6
1.0
1.5
qA (m/h)
Figure 4.1-7. Suspended solids content in the effluent TSe as a function
of surface loading (Ref. 88).
-Z6-
-------
Ih a 1972 publication (251) by Rex Chainbelt, Inc., data from the full-size clarifiers at
Racine, Wisconsin, and Hyperion, Los Angeles, California, full-scale plants were used to
develop a regression equation as follows:
Eff. SS = 382 x QrO-12 x (#BOD/dav/#MLSS)°-27
MLSSO.35 x dt!-03
Where: OR = overflow rate in gpd/sq ft (l m/h = 588.8 gpd/sq ft)
This equation has a multiple correlation coefficient of 0.63.
The regression analysis shows a correlation between effluent suspended solids and
overflow rate raised to the 0.12 power. The overflow rate exponent of 0.12 is quite
small and would have the impact of increasing the effluent suspended solids by only 10
percent when overflow rates were increased from 0.68 to 1.35 m/h (400 to 800 gpd/sq ft).
However, the equation also includes detention time raised to the 1.03 power. Since the
full-scale geometry at these plants remained unchanged, the detention time would
decrease as the overflow rate increased. They would not be directly proportional
because the rate of sludge return would perhaps remain approximately constant. This
latter variable was not discussed in the text. If it is assumed that the detention time
decreased by a factor of two as the overflow rate increased by a factor of two, an
increase in overflow rate from 0.68 to 1.36 m/h (400 to 800 gpd/ft) would increase the
suspended solids concentration in the effluent by a factor of 2.2. It must be concluded
therefore that the regression analysis shows a correlation between effluent suspended
solids and overflow rate, even though the latter is quantified in part by detention time.
Tuntoolavest, et al, (294) conducted experiments to study the effects of three oper-
ational variables performed on the Purdue activated sludge pilot plant using the Box-
Behnken method which employs a fractional factorial statistical design approach. The
authors derived a regression analysis equation. It is presented, along with such equations
of other authors, in Table 4.1-3 (143). The overflow rate term is included without an
exponent as a multiplier of the mixed liquor suspended solids concentration. The
equation predicts that the effluent suspended solids is proportional to the overflow rate.
Variations in the overflow rate were tested by the use of a 7.5-cm (3—in) inside diameter
plastic cylinder column. The reference describes no testing conducted to determine if
-27-
-------
TABLE 4.1-3. EMPIRICAL SECONDARY CLARIFICATION MODELS
Model
Effluent TSS Cone, (mg/1)
Source of Data
VilMer
Takamatsu and
Naito
Agnew
Lech
Busby and Andrews
Keinath et al.
Tuntoolavest et al.
Cashion and
Keinath
Chapman
Dietz and
Keinath
12988 OR0-4?* x-0»82 f°*439
2 X0-5 exp(-0.74 t)
{1) 18.2 + 8.01 OR - 0.0033 X
(2) 1937 OR0*12 FM0-21 X"°-35 t-l-03
0.0014 (17.6 - 0.739 T) OR X
0.01088 (Qi/A) X
4.5 + 0.00748 OR X
-7.83 + 7.8 Qa r - 70 r2 +
0.01459 X + 0.013 r X - 0.00138 Qa X
- 0.00248 t X + 0.000162 X OR
48.2 - 4.33 0C + 0.166 0 -
0.352 0C2 - 0.430 02 + 1.19 0C i
-180.6 + 0.004 X + 135.6
(Qi/A) + ( 90.2 - 62.5 (Qi/A) ) D
5.341 + 0.000506 Xd - 1.406 tc
Villier
Takamatsu & Naito
Agnew
Pflanz
Pflanz
Pflanz
Tuntoolavest et al.
Cashion and
Keinath
Chapman
Dietz and Keinath
NOTE:
OR
=
overflow rate from secondary clarifier
(m/h)
X
=
mixed liquor suspended solids (MLSS)
(mg/1)
t
=
hydraulic detention time in
secondary clarifier
(h)
FM
=
food-to-microorganism ratio in the
activated sludge system
(g BOD/g MLSS/h)
T
=
temperature of mixed liquor
(°C)
Qi
=
influent flow rate to secondary clarifier
(cu m/h)
A
=
surface area of secondary clarifier
(sq m)
r
=
sludge recycle ratio
Qa
~
air flow rate in aeration tank
(cu m/h)
Xd
=
dilute blanket solids concentration
(mg/1)
tc
=
detention time in clear zone
(h)
D
=
side water depth of secondary clarifier
(m)
Ac
=
sludge age
(days)
9
=
hydraulic retention time in aeration tank
(hours)
-28-
-------
the inlet and outlet conditions were optimized before the investigation of the various
overflow rates. Because of the small size of the settling columns, large scale eddy
current, wind effects, etc., were not modeled. The correlation between overflow rates
and effluent suspended solids must be judged accordingly.
If one inserts typical values such as an overflow rate of 1 m/h (600 gpd/sq ft) and a MLSS
concentration of 2000 mg/1 into the different equations of Table 4.1-3, a range of
answers is derived. This, in turn, would lead one to question the validity of the
equations. In using the equations, therefore, it is essential to research the ranges and
conditions associated with the sources of data from which the equations were derived.
Extrapolation outside of these ranges was found by the author of this review to produce
different, and sometimes erroneous, answers.
The recently published WPCF Clarifier Design Manual of Practice (308) shows two
figures illustrating correlation between effluent suspended solids and overflow rate for
primary clarifiers treating wastewater. These are shown in Figures 4.1-8 and 4.1-9,
respectively. Figure 4.1-8 shows the results from many treatment plants studied
approximately three decades ago. The figure shows a large scattering of data points.
They represent many different kinds of treatment plants in different locations.
Figure 4.1-9 shows relatively recent data collected at the Metro Denver Sewage Treat-
ment Plant. Whereas strong correlation between effluent suspended solids and overflow
rate are shown in Figure 4.1-8, limited correlation is shown in Figure 4.1-9.
Figure 4.1-10 shows data presented by Matasci, et al (199). There is virtually no
correlation between effluent suspended solids and overflow rate in the range of 0.51 to
2.2 m/h (300 to 1,300 gpd/sq ft). The results were collected at the Corvallis, Oregon
Wastewater Treatment Plant. It employs deep circular centerfeed clarifiers with
flocculating feed wells. The mixed liquor fed to these clarifiers came from an upstream
biological system consisting of a trickling filter followed by a short-term aeration basin.
This TF/SC (trickling filter/solids contact) mixed liquor contains more large dense
particles than typical activated sludge mixed liquor, but nevertheless undergoes zone
settling. The results therefore should be somewhat analogous to activated sludge
clarification, although the values of the effluent quality may be different.
-29-
-------
OVERFLOW RATE, m/h
1 2 3
e
0
•
o
o
0 •
RECTAI
• TANKS
0 CIRCUL
4GULAR
AR TANKS
• c
•
, ° 0
.V
0 ^
• \ J
•
•
o
O •
9
•
•
•
k
\
\
•
MEDIAN LIN
REMOVALS
~ 35% A*
E OMITTING
LESS THAN
ID OVER- -
TES LESS
00 GAL PER
ER SQ FT
o
•
•
e
FLOW RA
¦ THAN 3
\ ° DAY P
K
\
•
•
o
X •
*
•
•
•
i
Figure 4.1-8.
500 1000 1500 2000 2600
OVERFLOW RATE, GAL PER DAY" PER SO FT
Suspended solids removal for circular and rectangular
primary clarifiers (Ref. 308).
-30-
-------
zr
TSS
STANDARD
DEVIATION
69.4%
BOO
36%
6%
201 1 1 L-
I l I I
TSS
60 •
70
60 •
50 ¦
BOD
40.3%
33%
600
BOO
30
201—1 L-
31.8%
1000 1200 1400 600
OVERFLOW RATE, gpdiM) n
600 1000 1200 1400
OVERFLOW RATE.flpd/»q II
1600
20
30
40 50
20
30
40
SO 60
SOUTH COMPLEX <1877-1862>
K»m COMPLEX (1977-1962)
Figure 4.1-9. Ffcimary clarifier performance - Metro Denver Sewage
Disposal District No. 1 Central Wastewater Treatment
Plant (Ref. 308).
-31-
-------
38
OVERFLOW RATE, m/h
0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6
CO
O
—J
o
CO
o
U1
o
z
Ul
a.
CO
3
CO
>-
z
UJ
3
-J
u.
u.
UJ
_l
<
z
u.
36 H
34
32
a
z
o
p
<
GC
I-
z
UJ
o
z
o
o
24
16
14
12
6
4
C0RVALLIS, OREGON :
4/a/83 - m»m
STATISTICAL DATA
.
r -O02
f -0.13
-0J8
KEY
O WET SEASON
o
~ DRY SEASON
o
D O
a*
o ~ oe •+
as ~ * o a
O ++* +-m +4 4D B
OP ~ ~ D1"* »'OK8» lOB-a-O DC3 D OPO
WD O BOMH--H1 D-C+C SB ** ~ O ~ B
DQ DSrKMB +K+4B 0 DO
DO ® 0 as
S ++-H- o a ODD +o B
~ ~ ~ o
0
O 13
o a <* a o
OD ODD
~CD ~ C
~ a
O ~
~
300 500 700 900 1100 1300 1500
SECONDARY CLARIFIER OVERFLOW RATE, gpd/sq ft
Figure 4.1-10. Effect of secondary clarifier overflow rate on final
effluent suspended solids at Corvallis, Oregon (Ref. 199).
-32-
\
-------
Murphy (21?) conducted dye dispersion studies on a model clarifier with a diameter of
1.8 m (6 ft). His primary objective was to determine variations in the hydraulic
characteristics of the basin resulting from different inlet and outlet geometries. He did
nevertheless investigate the effect of varying flow rates from 1.7 to 3.4 m/h (1,000 to
2,000 gpd/sq ft). He found no significant variation in basin performance between these
two levels. No suspended solids removal data were collected. The significance of no
difference in dye dispersion results therefore may or may not relate to possible
differences in full-scale clarifier effluent suspended solids.
In a full-scale tube settler study, Mendis and Benedek (209) were unable to find a
correlation between overflow rate and effluent suspended solids. In the range of 1.0 to
2.0 m/h (600 to 1,200 gpd/sq ft), effluent suspended solids varied from 6 mg to 30 mg/1
for low SVI mixed liquor and ran as high as over 100 mg/1 for SVIs greater than 75. The
authors attribute poor performance data points to be largely due to clarifier thickening
limitations.
Fitch (108) conducted column settling tests and published the data shown in
Figure 4.1-11. By the relative slopes of the lines, Fitch contended that there was a
stronger correlation between effluent solids and detention time than overflow rate. His
definition of overflow rate, however, consisted of sampling depth divided by detention
time. The result is dimension ally correct, but does not truly reflect conditions in a full-
scale clarifier in which overflow rate is defined by the flow rate leaving the basin
divided by surface area. In addition, Fitch used a suspension of calcium carbonate. He
was therefore testing flocculaat settling rather than zone settling typical of activated
sludge mixed liquors. In view of these conditions, it is this author's opinion that Fitch's
results in themselves do not demonstrate that activated sludge clarifier overflow rate
does not correlate as well as detention time.
Cashion and Keinath (39) studied the effect of overflow rates at less than 1.70 m/h
(1,000 gpd/sq ft) on effluent suspended solids using a small scale activated sludge pilot
plant treating primary wastewater. They found that overflow rate had little effect on
effluent suspended solids. They found the characteristics of the mixed liquor to be much
more important and focused on hydraulic retention time in the aeration basin and solids
retention time in the activated sludge process.
-33-
-------
OVERFLOW RATE,m/h
0.2 0.5 1.0 2.0
3.0 r
90
2.5
« 2.0
80
10
0.5
5 6 7 8 9 10
4
2
3
1
OVERFLOW RATE, FT./HR.
Figure 4.1-11. RelationsMp between detention and overflow rate, "ideal"
basin, "long tube" test No. 1 (Ref. 108).
-34-
-------
Chapman, et al, (47) studied mixed liquor settling in a 2.4-m (7.9-ft) pilot scale circular
clarifier to correlate effluent suspended solids with a number of variables including pea'
flow rate. He observed a strong positive correlation between effluent suspended solids
and overflow rate (Figure 4.1-12). At a MLSS concentration of 2,000 mg/1, he found that
increasing the feed flow (overflow plus underflow) from 1.70 to 2.38 m/h (1,000 to 1,400
gpd/sq ft), increased effluent suspended solids from 30 to 47 mg/1. He used regression
techniques to arrive at the following equation to predict effluent suspended solids:
TSS = -180.6 + 4 MLSS + 135.6 (Qa/A)
+ SWD (90.2 - 62.5 Qa/A)
Where: TSS = effluent suspended solids in mg/1
MLSS = mixed liquor suspended solids in g/1
Qa = hydraulic feed flow (Qi + Qr) in m^/h
A = tank surface area in
SWD = sidewater depth in m
70
65
e 55
CO 45
ui
3
40
35
60
MLSS = 2.500mg/l
SWD(m)
u.
u.
id 30
25 -
20 1 1 1 1 1 1 1 1 l
1.5 1.6 1.7 1.8 1.0 2.0 2.1 2.2 2.3 2.4 2.5
FEED FLOW RATE PER UNIT AREA—(Qi+Qr)A, #(m/h)
Figure 4.1-12. Predictions for effluent quality with MLSS constant (Ref. 47).
-35-
-------
The final term in the equation shows interaction between feed flow rate and depth.
Tanks with deeper depths would have less deterioration of effluent quality resulting from
increased overflow rates. For example, if MLSS = 3 g/1, an increase in feed flow from 2
to 3 m/h (1,200 to 1,800 gpd/sq ft) would increase TSS from SO to 92 at D = 1.5 m
(4.9 ft), but only from 33 to 44 at D = 2.0 m (6.56 ft). Although Chapman's results may
be of somewhat limited value because of the small size of his tank, it is this author's
opinion that being larger than most pilot clarifiers reported on in the literature, his data
should suggest valid trends but not necessarily provide quantitative accuracy.
Matsunaga (200) presented data of effluent suspended solids versus overflow rate for
two-story final settling tanks treating activated sludge mixed liquor in Japan. The upper
and lower clarifiers operate in parallel over a range of surface overflow rates from
approximately 0.5 to 2.? m/h (300 gpd/sq ft to 1,600 gpd/sq ft). Effluent suspended
solids increased from 6 to 10 mg/1 for the upper basins and 8 to 16 mg/1 for the lower
tanks. These data show a positive correlation between increasing effluent suspended
solids with increasing overflow rates. The data are shown on Figure 4.1-13.
20
O O A V
O ¦ A ~
UPPER TANK
LOWER TANK
<
m *
r
f
*
w
1
0 2
0 30 4
SURFACE LO
0 50 6
ADIN6 m3/m2.d
1 ' '
0 7
0
o
10
s
Ui
3
-J
U.
U.
141
<0
CO
0.5
FLOW RATIO
Q/0.
Figure 4.1-13. SS of effluent from each tank vs inflow variation (Ref. 200).
-36-
-------
Parker (229) recommended the use of low overflow rates in design of secondary activated
sludge clarifiers. The proposition that effluent suspended solids are more sensitive to
overflow rate when tanks have shallow depths was stated. Full-scale data to correlate
effluent suspended solids versus overflow rates at the Livermore, California, plant were
presented and are reproduced in Figure 4.1-14. la later papers, Parker, et al (227,230),
showed additional correlation plots for full-scale tanks with different inlet
characteristics. The results, reproduced in Figure 4.1-15, also show a strong correlation
for tanks with conventional inlets but little correlation for a deep tank with a flocculator
center well.
The effect of hydraulic loading on effluent suspended solids leaving the full-scale
clarifiers at Pruszkow, Poland, was studied by Mazurczyk and Smith (203). The plant has
circular tanks with a diameter of 32 m (105 ft) and a depth of 2.94 m (9.6 ft). The
results are plotted in Figure 4.1-16. As overflow rates increase from 0.5 to 1.0 m/h (300
to 600 gpd/sq ft), effluent TSS increases from about 20 mg/1 to nearly 80 mg/1.
Solids Loading Rate
Solids loading rate to the secondary clarifier is, by definition, the product of the mixed
liquor solids concentration (MLSS) and the feed flow rate, divided by the clarifier surface
area. As discussed by Parker (229), MLSS concentration has not normally been
considered a factor in secondary clarifier design or removal efficiency. Recent
research, as reviewed in the following section, has investigated the effect of solids
loading on clarifier performance.
Pflanz (234) suggested that effluent suspended solids is directly proportional to MLSS.
As reported above, he conducted a two year study of full scale clarifier performance at
three plants in the Federal Republic of Germany. To approach steady state conditions,
the experiments were carried out at approximately equal wastewater temperatures and
with constant biomass in the aeration tanks. Comparative investigations at low
wastewater temperatures were performed at one of the three plants. Pflanz found the
effluent solids concentration to be proportional to the solids surface feed, which he
defined as the product of overflow rate and mixed liquor suspended solids. Increase in
sludge mass produced a relatively minor increase in effluent suspended solids at low
overflow rates, but caused a significant deterioration in effluent quality at higher
-37-
-------
HI
£ 1°
DC
«
a
2E
O
o
Ul
CO
0
NOTE; (UPLOTTED POINTS REPRESENT
ANNUAL AVERAGES, 19T0-T4
(2)gpd/lt2x0.041
= m3/d&y/m2
JL
S50 600 650 700
OVERFLOW RATE BPd/fl 2
750
Figure 4.1.-14.
Secondary clarifier performance vs hydraulic loading at
Livermore, California (Ref. 229).
80
40
tfi
tn 30
ui
o
<
c
111
>
<
20
10
OVERFLOW RATE,eel/dey/sq ft
800
1,000
1,500
/a.® m DEEP
/CONVENTIONAL CLARIFIER®
/3.T in
CONVENTION*!. CLARIFIER®
/
/ B.Sm DEEP FLOCCUmTOft-CLAMFICT
/ W/O DEHTnuFICATlQH
a.o
o 1.0
OVERFLOW RATE, m/h
Figure 4.1-15. Performance response curves for conventional clarifiers
and flocculator-clarifiers (Ref. 227).
-38-
-------
a
E
Z 180
O
|m*
<160
£"°
O
=• 120
O
o
CO 100
a
3 ¦
o 80
CO
£ 60
ui
3
_i 40
u.
IL
W 20
200
(GPD/SQ FT)
400
600
800
- r -
1
2C
12 is
® ®
i i
23
® ' /
/
/
f
•
i l i
" y'''
.•
tr
i i i i
I l
0.2 0.4 0.8 0.8 1.0 1.2
HYDRAULIC SURFACE LOAD (m3/m2/h)
Figure 4.1-16. Effect of overflow rate on effluent quality at Fruszkow,
Poland (Ref. 203).
-39-
-------
overflow rates. His observations are presented in Figures 4.1-17 and 4.1-18.
Temperature and SVI were also found to affect clarification efficiency. The influence of
sludge recycle was not found to be significant. Lech (187), Busby and Andrews (34), and
Keinath, et al (166), have developed regression models from Pflanz' data base. These
predictive models, as well as others reported in the literature, are presented in
Table 4.1-3 (143) discussed above.
Parker (229) noted that all of Pflanz* data were taken from studies of relatively shallow
clarifiers (1.2 to 2.7 m or 3.9 to 8.9 ft). He proposed that at shallow depths, the sludge
blanket can be expected to interfere with effluent quality. Likewise, Parker (229)
contends that effluent quality is more sensitive to solids loading when sidewater depth is
small. This is logical if the normal rise and fall of the sludge blanket brings the blanket
surface too close to the outlet wr.-ir, thereby permitting some to be swept out of the
tank. Pflanz' data do not show that such high levels were seen.
To investigate design and operating variables in secondary clarification, Chapman (47)
performed experiments to investigate seven potentially important variables using a
factorial experimental design. An activated sludge pilot plant with a 2,4-in (7.9—ft)
diameter circular settler was used in the study. Feed to the plant consisted of degritted
wastewater pumped from a nearby treatment plant. To simulate primary sedimentation,
a fine screen removed approximately 10 to 15 percent of influent suspended solids.
Chapman found MLSS concentration to be a critical factor in solids removal efficiency.
Figure 4.1-12 indicates that at constant sidewater depth and clarifier feed flow rate, an
increase in MLSS degrades effluent quality. Chapman did not find solids loading rate, or
the interaction of MLSS and feed flow rate, to be a statistically significant parameter in
effluent suspended solids removal. Therefore, the lines shown in Figure 4.1-12 are
parallel.
To establish the credibility of his study, Chapman (47) compared his results with those
obtained by other researchers. Figure 4.1-19 compares the solids loading rates required
to achieve a 30 mg/1 effluent as predicted by Chapman (47) and Pflanz (234). Pflanz* full
scale study indicates that higher loading rates are allowable than predicted in the pilot
study. However, the relative change in effluent quality is similar. As shown by
Chapman (47), an increase in overflow rate from 0.85 to 1.15 m/h (500 to 675 gpd/sq ft)
requires that MLSS concentration decrease by 900 and 1100 mg/1 in order to maintain a
30 mg/1 effluent in Pflanz' full-scale and Chapman's pilot scale clarifiers, respectively.
-40-
-------
a} temperature 2 —3 *C
Bennigsen
SS ®-°
0
1 7.0
uj 6.0
x
2 5.0
2
m 4.0
o
O 3.0
to
S 2.0
Q
5 j.o
0.
CO
3 0
to u
fg/M
-
•
(
3.18 0.2-
<
O
-
0
*30 0.40
0.60
0.B9
para
meter*.
u -
overflow rate Im
/hi
0
b)
10 20 30 40 50 60 70
SS IN EFFLUENT Cmfl/I).
temperature 13—15 *C
80
| 8-0
2 7.0
S 6.0
w 4.0
D
a 2.0
0 10 20 30 40 50 60 70 80
SS IN EFFLUENT {mg/l).
Figure 4.1-17. Relationship between effluent suspended solids and overflow
rate (Ref. 234).
-41-
-------
BENNIGSEN
a) temperature 2-3 °C
6.0
5.0
4.0
OJ
2 2.0
O
2 10
o
_i
O
111
X
20
40
50
30
SS IN EFFLUENT(tng/I)
60
70
80
b) temperature 13-15 °C
,2.h J
8.0
9. 7.0
Z 6.0
CO 5.0
4.0
3.0
• I
2.0
1.0
0
10
20
50
30
60
70
40
80
SS IN EFFLUENTCmg/l)
Figure 4.1-18. Relationship between solids surface feed and suspended
solids in effluent (Ref. 234).
-42-
-------
\
V
OpFLANZ DATA FROM CELLE.
SWO-2.3 m
DIA. -33. m
PREDICTED VALUES.
aWITHIN EXPERIMENTAL RANGE
^EXTRAPOLATED
SWD-1. 94 m
Or-~40 L/min
JL.
_1_
JL
0.0 0.2 0.4 0.6 as 1.0
OVERFLOW RATE—(Qi/A) Cm/h)
1.2
1.4
1.6
Figure 4.1-19. Comparison of loading rates required to achieve a 30 mg/L
effluent (Ref. 47).
-43-
-------
Tuntoolavest, et al (294), as discussed in the previous section, studied the effects of
operational variables on clarifier effluent quality. The pilot plant aeration basin utilized
in their study was a 348 1 (92 gal) tank with a liquid depth of 0.84 m (33 in). The test
clarifiers each had a surface area of 0.28 sq m (3 sq ft) and a volume of 227 1 (60 gal).
Synthetic wastewater was prepared for the experiments. Careful control of organic and
biological parameters was maintained throughout the study; steady state operation was
at an organic loading of 0.3 mg COD/mg MLSS/day and a solids retention time (SRT) of
10 days. The major finding of their research was the importance of the effect of MLSS
concentration on effluent suspended solids. A strong direct relationship between the two
variables was found, which is in agreement with the work of Pflanz. The model
developed from the study data shows that a number of interactions between MLSS and
other variables are important in effluent solids prediction. The model is presented in
Table 4.1-3 discussed above.
Agnew at Rex Chainbelt Inc. (251) performed an experimental testing program on full
scale final clarifiers at three treatment plants in order to develop a mathematical model
for clarifier performance. Two models were developed from the collected data. One
model, the first of the two listed in Table 4.1-3 for Agnew, fit data for short
observation periods. Agnew's second model was developed from a larger, more variable
data set. Both models considered MLSS as a solids loading factor. Both showed an
inverse relationship between MLSS concentration and effluent suspended solids, in
contrast to the findings of Pflanz. Data scatter was much greater for the second model
than for the first. The author attributed this to varying biological characteristics over
long time periods. Although the author emphasizes the importance of sludge
characteristics in clarifier performance, control of influential parameters such as sludge
age was not addressed.
Villier (301) and Takamatsu and Naito (281) also developed relationships between MLSS
concentration and effluent suspended solids. Both studies, at bench scale, utilized
suspensions other than activated sludge. These relationships are presented in Table 4.1-3
Villier's equation shows an inverse relationship between MLSS concentration and effluent
TSS. Takamatsu and Naito's equation shows effluent TSS to increase proportionately to
the square root of the MLSS concentrations.
Munch and Fitzpatrick (214) reported results of studies performed on full scale circular
secondary clarifiers at the West-Southwest Sewage Works of Chicago. They found that
-44™
-------
the impact of solids loading rate on clarifier operation is influenced by total hydraulic
loading, that is, the clarifier feed flow rate. Test results axe summarized in Table 4.1-4.
The three lowest solids loadings of the 14 tests, each less than 146 kg/sq m/day (30 lb/sq
ft/day), produced an "acceptable operation." The authors rated general operation of the
clarifiers based primarily on the level of sludge blanket stability. The authors stated
that the clarifiers have consistently attained effluent solids concentrations below
10 mg/1; however, this value was not used as a criterion in evaluating performance. In
seven of the remaining 11 tests, feed flow rates exceeding 2.58 m/h (1,520 gpd/sq ft)
with loading rates varying from 160 to 214 kg/sq m/day (33 to 44 lb/sq ft/day) produced
undesirable operation. Tests 11 and 13 produced satisfactory operation at feed flow
rates of 2.48 to 2.36 m/h (1,460 and 1,390 gpd/sq ft), respectively. The corresponding
solids loading rates were 175 to 199 kg/sq m/day (36 and 41 lb/sq ft/day). Unfortunately,
the authors did not report quantitative effluent data (e.g., mg/1 TSS) from their tests.
Meaningful conclusions based on these data are limited because (1) sludge properties
were not controlled, and (2) the data set is quite small.
Cashion and Keinath (39) studied the effects of MLSS concentration and solids surface
feed's influence on effluent suspended solids (see Figures 4.1-20 and 4.1-21). They
utilized a small, pilot scale, completely mixed activated sludge treatment plant in their
study. The biological reactor had a volume of 0.475 cu m (125 gal), and the two
cylindrical clarifiers had internal diameters of 14 and 38 cm (5.5 and 15 in). The
wastewater used in the investigation consisted of a mixture of primary sludge and
primary overflow obtained from the clarifiers of a small domestic wastewater treatment
plant. As shown in Figure 4.1-20, there appears to be some correlation showing effluent
TSS to increase with MLSS, but the authors report that the correlation was not
statistically significant. When comparing their data to other studies, the range of MLSS
variations is quite low. Many activated sludge plants operate at MLSS values between
2,000 and 3,000 mg/1. Figure 4.1-21 shows a plot of effluent TSS plotted against solids
surface loading (flux). A typical loading value using 1.0 m/h (600 gpd/sq ft), overflow
rate, 50 percent sludge recycle, and 2,000 mg/1 MLSS would be 73 kg/sq m/day (15
lbs/day/sq ft). The data of Figure 4.1-21 range from 5 to 65 kg/sq m/day (1 to 13
lbs/day/sq ft). This range is therefore below typical full scale loadings. This limitation,
coupled with the fact that experiments were conducted on a small pilot scale unit,
-45—
-------
TABLE 4.1-4. SUMMARY OF PERFORMANCE TEST RESULTS.
Test SSR WLR SLR
No. Qr(n»3/s) QeCm^/s) Qt(m3/s) (wP/ofi'W m3/ra*h) Ocg/m'*d) Rating*
1
0.14
0.644
0.78
2.01
12.3
163
X
2
0.14
0.736
0.88
2.30
14.1
142
0
3
0.14
0.705
0.85
2.19
13.5
182
XX
4
0.21
0.696
0.91
2.18
13.4
191
X
5
0.21
0.666
0.88
2.07
12.8
191
XX
6
0.24
0.622
0.86
1.94
11.9
196
XX
7
0.24
0.578
0.82
1.80
11.1
147
X
8
0.26
0.600
0.86
1.87
11.5
199
X
9
0.40
0.644
1.04
2.01
12.3
214
XX
10
0.25
0.727
0.98
2.26
14.0
'160
X
11
0.18
0.613
0.80
1.90
11.7
174
0
12
0.18
0.648
0.83
2.02
12.5
137
0
13
0.18
0.578
0.76
1.80
11.1
200
0
14
0.18
0.653
0.83
2.04
12.5
142
0
NOTE: All numerical values are averages.
* Code for rating:
0: Acceptable operation; high effluent quality with limited rise of sludge blanket,
x: High sludge blanket; corrective measures necessary,
xx: Upset, indicated by discharge or threatened discharge of large quantities of solid to
effluent.
Qr =
recycle flow rate
Qe =
clarifier effluent flow rate
Qt =
clarifier influent flow rate
CCD =
OOXv ~
surface overflow rate
WLR =
weir loading rate
SLR =
solids loading rate
"46-
-------
84 ¦
i
63
42
• •
z
UJ
3 21
• * •
I
I
I
500 750 1000 1250 1500 1750
MIXED LIQUOR SUSPENOEO SOLIDS Ig/m3)
2000
2250
Figure 4.1-20.
Effect of clarifier feed concentration on effluent suspended
solids (Ref. 39).
84
*&
- 63
a
z
IU
8s
D
Z
ui
3
42
21
••
• •
• •
_L
X
10 20 30 40 50 60
SOLIDS SURFACE FLUX
Figure 4.1-21. Effect of solids surface flux on effluent suspended
solids (Ref. 39).
70
-47-
-------
indicates that these correlations may or may not occur in full scale tanks with peaking
factors, large scale eddies, and peak solids loads up to three times higher (up to 243
kg/sq m/day or 50 lb/day/sq ft according to Ten States Standards).
Tests were performed on a 32 m (105 ft) diameter clarifier located at the Pruszkow
Wastewater Treatment Plant near Warsaw, Poland (203). The facility treats 18,000 to
30,000 cu m/day (4.76 to 7.93 mgd) of wastewater composed of equal fractions of
domestic and industrial flow. Mazurczyk and Smith (203) could not find a clear
relationship between solids loading and effluent suspended solids. A plot of their test
results is shown in Figure 4.1-22. Solids loading in the range of 4 to 6.3 kg/sq m/hr (17.6
to 30.9 lb/sq ft/day) produced an effluent TSS between 10 and 80 mg/1, but no correlation
of the two variables is evident. Conclusions based on these results should be made with
caution. Discussing the experimental procedures, the authors state that "Measurements
in the Pruszkow final clarifier were made on chance days and under the conditions
occurring on those days. Because of the routine of plant operation it was impossible to
control specific parameters." Therefore, it is quite possible that uncontrolled variables
were responsible for masking a relationship between solids loading and effluent
suspended solids.
Mendis and Benedek (209) investigated the relationship between solids loading and
effluent quality from full scale tube settlers in Canada. The clarifiers under investi-
gation were 24 m (78.7 ft) square with a sidewater depth of 4.6 m (15 ft). As can be seen
in Figure 4.1-23, no correlation was found. To maintain a control on biological variation,
sludge age was maintained at a value of 10 days during the study. The authors do not
discuss control of other variables. The authors also do not discuss the interaction of
MLSS concentration and feed flow rate when reporting solids loading rates.
Stofkoper and Trentelman (278) conducted studies of full scale circular secondary
clarifiers in the Netherlands. Test data from two plants were used to develop a set of
design guidelines. The guidelines were verified by studying clarifiers at 19 other
installations. They found the interaction of MLSS, surface overflow rate, and sludge
volume index to be critical in clarifier performance. They defined the product of these
three terms as the sludge volume loading.
By interrupting flow to the plants for several hours while maintaining sludge return, each
test began with little c-.- no sludge in the clarifiers. Clarifier performance was evaluated
-------
236
(fl
12 15
® ©
>25
•
1A
•
•
•
0
1
1
•
• «
•
•
1 # •
O
»
•
»
2 3 4 5
SURFACE SOLIDS LOAD (kg/m2/hr)
Figure 4.1-22. Effect of surface solids loading on effluent quality (Ref. 203).
-49-
-------
•;
A svi
LEGEND
75
THICKENING LIMITING
633
BH svi
75
CLARIFICATION LIMITING
© SVI
75
CLARIFICATION LIMITING
ACCEPTED LIMIT
10
SOLIDS LOADING-kfl/h/m5
15
Figure 4.1-23. Effect of solids loading on clarification (Ref. 209).
-50-
-------
by observing the activity of the sludge blanket. Figures 4.1-24 and. 4.1-25 illustrate
conditions when the sludge blanket stabilized and when it continued to rise,
respectively. The test data are summarized in Table 4.1-5. The authors found their
results to agree reasonably well with German design guidelines (ATV), as shown in Figure
4.1-26. However, the test results do indicate that the ATV guidelines may be somewhat
conservative at higher sludge volumes. The sludge volume (VSv) is defined as the ratio of
the volume of sludge in the clarifier and the total volume in the clarifier. Based on their
findings, the authors proposed a new set of guidelines for sludge volume loading; the
relationship is given in Figure 4.1-27.
To investigate the performance of upflow secondary clarifiers (known in Europe as
vertical or Dortmund tanks), Resch (250) varied surface overflow rate, sludge return
rate, and MLSS of these full scale tanks in Germany. Results, as given in Figures 4.1-28,
4.1-29, and 4.1-30, showed a deterioration of effluent quality at high sludge volume
loading. The experiments yielded much higher (50 percent or more) allowable surface
overflow rates for a given sludge volume than given by the ATV guidelines, supporting
the findings of Stofkoper and Trentelman (278). The guidelines were found to be
especially conservative for the extremely deep tank at Ruhleben (14.3 m or 47 ft). The
results of Resch's study, as plotted against ATV guidelines, are shown in Figure 4.1-31.
SURFACE AREA - THICKENING
Satisfactory performance of an activated sludge secondary clarifier requires production
of a concentrated underflow as well as a clarified overflow. The following section
examines research that has been performed to characterize the thickening process in
secondary sedimentation and to develop rational design procedures for thickening.
Solids Flux Theory
Coe and Clevinger (53), the founders of modern flux theory, argued that any layer of
suspension in a thickener has a specific solids handling capacity. The solids handling
capacity is the mass of solids that a unit area of the layer can pass to the layer below.
According to their theory, as published in 1916, the basis for determining the area
required for thickening is to provide sufficient surface to assure that solids axe applied
at a rate less than the solids handling capacity of the limiting layer. To identify this
limiting layer, a series of batch settling tests was advocated.
-51-
-------
— UJ
^ o
-J <
UJ u.
> EC
m 3
-* CO
h ee
UJ ui
* f-
Z <
3 *
© a
uj "J
a 9
o z
3 3
»5
z
0
tr
3
16
H
14
0.4
UJ
o:
0.8
UJ
12
a
*»"¦»
1,2
a
->
I'J
E
10
1.6
CO
*•».
CJ
*
8
«0
*1.
2.0
H*
(3
6
7!
B
2.4
UJ €3
3
4
-1
2.8
U«
z
2
3.2
z
0
• RETURN SLUDGE
~ INFLUENT
~ HEIGHT OF SLUDGE BLANKET
1
1 1 "I- 1.
i'ii*
1
1
1
!
fV
•
j* ~ 1 —
o-y~yo-
bfl
»-Tta 1 l
N
r-|
Jinr
1
•
I t
1
1
«
d
10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00 20.00
T1MECHR)
Figure 4.1-24. Example of test duration, resulting into equilibrium (Ref. 278).
» ~
x
r °
J £
UJ 3
-I CO
I— fr
"J 5
* N
Z <
5 *
CO jx
UJ UJ
S g
o fr
-J «c
CO
z
tr
0 5
16
14
SC
0.4 m
o
0.8 O ^ 12
3™
1.2 5) | 10
13
1.6 * *
„ - 2 5
2*0 H* 43
3 °
2.4 u.
• RETURN SLUDGE
A INFLUENT
_TL EFFLUENT
~ HEIGHT OF SLUDGE BLANKET
2.8
3.2
•
a
ST
_4—
AR1
> 5
TO
>
•
~
• ,
r*~
* i
• 1
' •
j
• 1
•
'I
®1
3*^
n
1 c
J
L i
i
>¦ m
-r|
1600
1400
1200
1000
800
600
400
200
0
9.00 10.00 11.00 12.00 13.00 14.00 15.00 16.00 17.00 18.00 19.00
TIME(HR)
Figure 4.1-25. Example of test duration, resulting into sludge discharge
(Ref. 278).
DJ
G
w
©
O
H
Z
UJ
3
UJ
Z
d
Q
-52-
-------
TABLE 4.1-5. DATA OF THE CLAHIFXERS AND PROCESS CONDITIONS, DURING THE TEST PERIODS.
Sludge
D.S.
Sludge
Surface
Return
in
Sludge
D.S.
Volume
Loading
Ratio
Influent**
Index
Loading***
Loading
Clarifier
vA
ra^/tm^'h)
R
G
kg/m^
Isv
ml/g
SSVI3>5
ml/g
SA,
kg/(m^-h)
1/lm^-h)
diameter
m
awd
m
Almelo
-•
0.92
0.28
1.9
100
70
2.2
170
40
Apeldoorn NBT1
-
1.02
0.83
0.4
440
-
0.7
180
43.3
2.5
Apeldoorn NBT5
X
1.02
0.93
0.3
550
-
0.6
170
43.3
2.5
Bevcrwijk
-
0.85
0.68
2.8
130
60
4.0
310
30
2.0
Deventer NBT3
-
0.49
0.5
3.0
160
100
2.2
240
36
2.5
Deventer NBT4
-
0.49
0.5
2.3
200
(200)
1.7
230
36
2.5
Echten
-
0.79
0.91
2.6
100
75
4.2
220
45
2.0
Gieten
-
0.50
0.67
5.4
130
80
4.5
250
39
1,5
Goor
+
0.49
1.02
2.2
440
260
2.2
480
33.8
2.0
Haarletn/Schalkwijk
-
0.36
0.54
0.6
250
(150)
0.4
40
45
1.5
Hapert
-
1.50
0.33
3.5
50
40
6.4
260
30.3
1.5
Harderwijk groot
X
0.75
0.66
5.2
120
70
6.8
500
46
2.0
Harderwijk klein
X
0.68
0.70
5.2
120
70
6.4
450
38
1.5
Helmond 1
X
0.44
0.39
3.1
-
(180)
1.9
-
47,9
1.5
Helmond 2
X
0.68
0.36
4.0
-
(180)
3.7
-
47.9
1.5
Hoensbrock
X
1;0
1.0
4.6
70
50
10.0
35C
44
1.5
Huizen
-
0.64
0.81
0,6
300
(300)
0.8
130
41
2.0
Hulst
-
0.90
0.33
2.0
90
100
2.4
160
42
2.0
Jourc
X
0.48
1.27
4.9
120
90
5.5
290
37.3
1.5
Oss****
0.46-0.82
0.48-1.36
1.9-5.9
130-200
90-130
2.5-7.4
220-630
41.8
2.0
Raalte
-
0.74
0.51
3.7
60
65
3.9
160
41.5
2.5
Rijen*****
0.52-1.09
0.2-1.2
2.5-3.8
110-200
80-130
2.6-5.8
210-500
45.5
2.25
Rijssen
+
0.73
0.69
2.3
170
100
2.8
290
41
1.5
Uden-Vcghel
+
1.17
0.46
3.4
90
90
5.8
360
41
2.0
Wijk bijj Duurstede
+
0.90
0.39
3.3
75
80
4.1
240
35.6
1.5
* Signs have following meanings
+ the test resulted In sludge discbarge
- the test resulted in balance
x no balance (equilibrium) was achieved
** At end of the test.
* * * gA = tg + Br/A). Ga to compare WRC - guideline and solid flux theory.
**•*10 different measurements (4 times + and 6 times -).
*»*»«14 different measurements {6 times +, 3 times x, and 5 times -).
-------
AT
V CURV
'E
X
m
X
X SLUDGE DISCHARGE
® EQUILIBRIUM
A NOT YET BALANCED
%
v
\x
Mr
XX
X
X
#
©
®
St A)
A
(
i
X
mi
A
)
XX
> <8
%
©A
©
w
A
A
©
Figure 4.1-26.
200 400 600 800 1000
SLUDGE VOLUME, VSv(mg/l)
The AT\r guideline and test results as a function of surface
loading and sludge volume (Ref. 278).
-54-
-------
1.5
1,4
VSA- 300l/(m*h)
1.2
VSy\ = 1 /3VSv »200iCm2h)
1.0
0.8
VS. = 400l/(m2h)
5 0.67
0.6
0.4
0.2
0
0
200
400
eoo
800
1000
SLUDGE VOLUME VSvCml/I)
Figure 4.1-27. Allowable sludge volume loading (Ref. 278).
-55-
-------
t t
I-
z
UJ
3
-I
Ul
U.
UJ
UJ
x
h
Z -C
— o
« E
5 uj
-J 05
O »-
co
o
UJ
o
z
UJ
Q.
U>
3
m
50
40
30
20
10
fo o CN
(D in w
CM [Nl f"-
AMPFING
Y /
%
V
y/
x/ /
'/A
YA
p
I
•
•
•
i
1
TSg= Wmg/t ¦
s = 7,1 mp/I •
• • •
• • *
•
•
• •
[/?
i J/y
V/
A
100 200 300 400 500 600 700 800 900 1000
SLUDGE VOLUME LOADING, VSp (l/m2h>
Figure 4.1-28. Ampfing effluent suspended solids vs sludge volume loading
(Ref. 250).
-56-
V
-------
Iff t
N (0 «
« « 10
no ®
o
o
50
40 <
30 «
UJ
0)
20
10 <
GREDING
TSe=9,7 mg/l
• =7,6 mg/l
• •«
Y~/
100 200 300 400 S00 800 700 800 800 1000
SLUDGE VOLUME LOADING VSp (l/nr*h)
Figure 4.1-29. Greding effluent suspended solids vs sludge volume loading
(Ref. 250).
-57-
-------
SUSPENDED SOLIDS IN THE EFFLUENT
TSe mfl/I
44 44 ^
TSe >500 mfl/I
RUHLEBEN
TSE=7.4 mfl/l
S = 2.8 mfl/l
e
o
o
100 200 300 400 500 800 700
SLUDGE VOLUME LOADING VSF l/m2h
800 800 1000
Figure 4.1-30. Ruhleben effluent suspended solids vs sludge volume loading
(Ref. 250).
-58-
-------
HIGH LOAD
WITH CLEAR
EFFLUENT
OVERLOAD
WITH SLUDGE
IN THE EFFLUENT
AMPFING
GREDING
RUHLEBEN
ATV - DIMENSIONING
VERTICAL TANKS
HORIZONTAL TANKS
100 200 300 400 500 800 700
CORRECTED SLUDGE VOLUME VS* «/m3)
SCO
800
Figure 4.1-31. Limits of surface loading in upflow final settling tanks
(Ref. 250).
-59-
-------
Several years later (1952), Kynch (181) .formulated the mechanism of batch thickening
lor ideal suspensions. In his work, Kynch assumed that the velocity of any particle in a
layer was a function only of local concentration. Three years later, Talmage and Fitch
(283) showed that according to the Kynch analysis, multiple batch settling tests for
determining the limiting flow were unnecessary.
Dick and Ewing (S3) have contended that the Kynch theory is valid for "ideal"
suspensions, but not for activated sludge. They concluded in their 1967 publication that
the design approach of Talmage and Fitch was unsound for activated sludge clarifiers;
they supported the use of batch settling tests to determine limiting flux.
Flux Plot Analysis
Flux analysis had first been applied to analyses of secondary clarifiers by Dick (62). To
assure that the clarifier accomplishes its thickening function according to Dick's
analysis, solids must be applied at a rate that is not in excess of the rate at which they
are able to reach the bottom of the tank. Alternative final settling tank design and
operating conditions can be analyzed by a procedure developed by Yoshioka, et al (319).
As shown in Figure 4.1-32 (87), a tangent to the batch flux curve intercepts the ordinate
axis at the value of the limiting flux Gl- The tangent intercepts the abscissa at the
corresponding underflow concentration Cy.
To simplify the graphical procedure developed by Yoshioka, et al, Vesilind (300) and Dick
and Young (87) have developed empirical relationships between the settling velocity and
sludge concentration. A statistical analysis of 159 sets of experimental data (95)
indicates that Vesilind's model gives a better prediction.
Keinath (163,166,171) extended the settling flux design approach developed by Dick for
the analysis of clarifier operation by introducing the state point concept. The state
point is defined by the intersection of a sludge recycle operating line and an overflow
rate operating line on the batch flux plot. Operation strategies based on Keinath's state
point concept are discussed later in Section 4.6.
-60-
-------
1
c0 CL
SUSPENDED SOLIDS CONCENTRATION, c1
Figure 4-1-32. Graphical determination of settling tank performance
(Ref. 87).
-61-
-------
ISV Analysis
Wilson and Lee (315) compared clarifier design approaches of the limiting flux theory and
the initial settling velocity (ISV) technique. To characterize the thickening area
requirement, the ISV approach assumes that the overflow? rate equals the initial settling
velocity at the design MLSS divided by a safety factor. The form of the equation is:
Qe/A = aXfn/SF
where: Qe/A= overflow rate
Xp MLSS concentration
SF= safety factor
a,n= sludge settling constants
By mathematical derivation, Wilson and Lee showed that the single clarifier area
predicted by the ISV technique (at SF = 1) is equivalent to the minimum area predicted
by the limiting flux theory.
In a later paper (314), Wilson describes the ISV test procedure and offers a new
interpretation of the results, the Rated Capacity concept. He defines the Rated
Capacity as the "mixed liquor flow (exclusive of the return sludge flow) which can be fed
to a clarifier such that upflow velocity of the clarifier equals the ISV of the mixed
liquor." Mathematically, the Rated Capacity is defined as the product of ISV and
clarifier surface area. If the Rated Capacity is greater than the clarifier influent flow
rate, excluding return sludge flow, Wilson suggests that the clarifier is not overloaded.
If the Rated Capacity is low, remedial actions are provided in the text to reduce the
solids loading. Riddell, et al (254) showed that the rating equation developed by Wilson
can be used with either the power settling model of Dick and Young (87) or the
exponential settling model of Vesilind (300). To account for non-ideal conditions of
settling basins, he modified the rated capacity equation by including a safety factor
term.
Severin and Poduska (266) have proposed a similar simplification of flux analysis. Using
dimensionless parameters that are functions of initial settling velocity, they developed
an operating line that was used to accurately predict thickening failures.
-62-
-------
Verification of Solids Flux Theory
Dick and Javaheri (84) found the predictions from flux analysis to be strongly dependent
upon the type of suspension. They found suspensions of calcium carbonate, water
softening sludge, and glass beads to conform to the theory more closely than suspensions
of activated sludge. Munch and Fitzpatrick (214) reported studies on the performance of
full scale clarifiers. They found flux theory to overestimate the solids handling capacity
of the facilities. T .ey suggested that non-ideal conditions, such as density currents and
the inhibition of sludge concentration (occurring at excessive hydraulic flows) are not
considered by the theory.
Hibbard and Jones (135) also found solids flux analysis to overestimate solids handling
capacity of clarifiers. The investigators carried out continuous thickening tests in a
0.38-m (1-2—ft) diameter settling column. Activated sludges used in the tests were
collected from two different domestic wastewater treatment plants with SVI values of
200 and 250 ml/g, respectively. Under different feed and underflow rates, the authors
consistently found analysis of the batch settling tests to overestimate the system
capacity. Sample results are presented in Table 4.1-6.
To verify a dynamic mathematical model developed from solids flux theory, Pizarro (236)
collected laboratory data using settling columns (diameters of 0.15 and 0.06 m or 0.5 and
0.2 ft) and a calcium carbonate suspension. Results presented by the author indicate
that theoretical solids flux predictions overestimate thickener capacity. The general
dynamics of the thickening process, however, were found to be in reasonable agreement
with theory.
Anderson (4) introduced the concept of "batch flux efficiency" to analyze deviation from
solids flux theory. He defined the batch flux efficiency as the ratio of the actual batch
flux to the limiting batch flux. He also introduced the "operating point," which is similar
to the state point, but is based on underflow flux (the mass of solids transmitted in the
underflow per unit time and clarifier surface area) rather than applied flux. Using data
from different investigations on limiting flux conditions, batch flux efficiencies were
computed. He found the full scale tanks to perform at efficiencies substantially less
than 100 percent. (36 to 57 percent), whereas he found the pilot plant to perform at 98
percent. He attributed non-ideal performance to conditions such as density currents,
-63-
-------
TAPLE 4.1-6. RESULTS OF SOLIDS FLUX TESTS
Conditions: Feed rate - 1.8 m/h; Underflow rate - 0.9 m/h
Sludge Total
Bed wt. of
Depth Sludge
Test Concentration Flux Capacity (kg/m2h) In Bed
No. Feed Underflow Overflow Pred. Actual Difference (m) (kg/m^)
1/1
3680
6145
1210
6.72
5.76
0.96
1.33
5.140
1/2
3095
5375
765
II
5.16
1.56
1.06
4.014
1/3
3030
5770
265
II
4.86
1.86
0.68
1.474
1/4
2520
4490
320
5.10
4.02
1.08
0.69
1.807
1/5
2677
4770
520
1
4.32
0.78
1.13
3.094
1/6
2385
4440
285
1
3.96
1.14
0.60
1.425
1/7
2180
3930
260
II
3,54
1.56
0.12
0.160
1/8
2480
4480
355
II
4.02
1.08
0.64
1.519
1/9
2430
4545
330
11
4.08
1.02
0.82
2.174
1/10
2500
4600
365
t!
3.96
1.14
0.38
0.815
Ref. 135
-------
temperature changes, and wind effects. Suggesting that design based on perfect batch-
flux efficiency as unsatisfactory, he suggested an efficiency of 50 percent he used for
design. As Anderson pointed out, this is equivalent to a safety factor of 2 on the batch-
flux loading.
Recognizing that actual solids flux capacity may fall below theoretical values for full-
scale clarifiers, Parker (229) has suggested that secondary clarifiers should be designed
for solids flux limitation at peak hydraulic flow rather than average flow conditions. He
argues that such a design allows for the highest effluent quality at average flow, while
also providing reserve sludge storage capacity for peak hydraulic operation. Typically,
treatment plants have a peak to average flow ratio of 1.5 to 2.5 or more. A rough safety
factor of 2 would often satisfy the conditions requested by Parker and meet the safety
factor of 2 proposed by Anderson (4).
Solids flux theory has also been used to predict underflow concentrations. For two pilot
plants, Dick and Young {87} found that predicted underflow suspended solids
concentrations were within -10 percent to +20 percent and -20 percent to +20 percent of
observed values. White (312) was able to predict underflow concentrations to within ±20
percent of observed values at 8 out of 10 full-scale tanks, overloaded with respect to
thickening.
Based upon tests recently performed at the Jones Island and South Shore wastewater
treatment plants in Milwaukee, Wisconsin, Daigger and Roper (75) developed and
published a graphical technique to determine permissible clarifier loadings. Numerous
batch settling tests were conducted to determine a correlation between initial settling
velocity (ISV), SVI and the initial solids concentration, Ci. Regression analyses were
effective, provided that the SVI values were separated into four data groups. The results
led to tlie following equation:
„ -(0.148 + 0.0021 (SVB)Ci
Vj = V0 e
where: V = m/h
C = g/1
SVI = ml/g
The data were then used to develop the clarifier design and operating diagram shown in
Figure 4.1-33. With this diagram, it is hypothesized by the authors that SVI tests could
be run and solids loading limits be readily determined for any overflow rate selected.
-65-
-------
ao
24.4 m/d
(600 gpd/sq 10
20.4 m/d
(BOO flPd/»Q ft)
80 ml/gL
limiting nui iiNit fo* y*mous t*u
ohoatixo iixn won v«moui
uMOinnowmtTts —
350
TO
16.3 mid
(400 gpd/«q ft)
300
100 ml/fl *
12.2 m/d
J,y (300 Cpd/tq il)
250 -
60
160 ml/g
TOO ml/g /"
z ^
a g 200
>.40
-a
160
30
4.1 m/d "
[(100 gpd/
*q tl) —
100
20
]}50 tnl/g
^ 2.0 m/d "
(60 gpd/sq ft)"
60
10
—I 1 _JL„
S
0
0
10
30
IS
20
25
UNDERFLOW SUSPENDED SOLIDS CONCENTRATION Cu (gll)
Figure 4.1-33. Clarifier design and operating diagram {Ref. 75).
-66-
-------
DEPTH
Adequate tank depth in clarification was discussed as far back as 1904 by Hazen (133).
He considered settling of discrete particles and showed that in theory removal was not a
function of depth. Camp (36) recognized that depth played a significant role in settling
of flocculant particles. He conducted some bench scale testing on iron floe to support
his hypotheses. In a 1953 publication, Camp (37) further discussed depth effects on
settling of flocculant suspensions. He stated ". . .the rate of flocculation in deep tanks
with low velocities is independent of the velocity or depth. If it is assumed that
flocculation continues throughout the settling period (in other words, that the limiting
size of the particle is not reached), the amount of flocculation will be directly
proportional to the detention time. Hence, an increase in depth for the same overflow
rate will produce a proportional increase in the amount of flocculation. An increase in
depth will thus increase removal, but it is not possible to state without settling tests
whether the same increase in removal might not be obtained more economically by a
decrease in overflow rate," Camp presented extensive theoretical development to
justify his position, but present-*-! no data from full-scale clarifiers with different tank
depths.
hi the same time frame, Anderson (5) in 1945 studied a number of variables affecting the
full-scale activated sludge clarifiers serving the Chicago Southwest Treatment Plant.
He conducted experiments to demonstrate the effects of different tank depths by
varying the sludge blanket thickness in a circular tank with a sidewall depth of 3.55 m
(12 ft). He therefore did not actually study different tank depths, but rather different
depths to the sludge blanket. His data showed that a tank operating with a sludge
blanket of about 3.05 m (10 ft) below the surface produced a maximum bottom velocity
of 1.7 m/min (5.6 ft/in in) and the maximum return velocity of about 0.85 m/min (2.8
ft/min). By raising the sludge blanket to almost 1.2 m (4 ft) below the surface, the
bottom velocity increased to 4.45 m/min (14.6 ft/min) and the return velocity to about
1.52 m/min (5 ft/min). He therefore concluded that the velocities were in approximate
inverse ratio to flow depth. He stated the need for a design criteria relating tank depth
to the distance between the tank's inlet and outlet. Although he cited a need for
additional research, he stated the following opinion:
1. For circular tanks with center inlet, the radius should not exceed about 5
times the sidewater depth.
-67-
-------
2. For rectangular tanks where there is a better opportunity to reduce the inlet
velocity, the flow length should not exceed about 7 times the depth.
3. The depth below the effluent weirs should not be less than about 3.05 m
(10 ft) in order to avoid any disturbance caused by the mechanism and density
current nor less than 3.66 m (12 ft) if the weirs are located at the upturn of
the density current.
McKinney and O'Brien (207) discussed secondary clarifier depth for activated sludge
treatment along with other basic design concepts in 1968. They stated that secondary
clarifiers need sufficient depth to permit good sludge separation and less disturbance due
to inlet currents. They stated that more depth is required for higher concentrations of
mixed liquor suspended solids. A minimum sidewater depth of 3.05 m (10 ft) was
recommended. No data were presented to quantify the effects of different depths.
hi 1982, Davis (78), discussing activated sludge plant designs in South Africa, proposed
that theory is unable to determine sidewall depth requirements for an activated sludge
clarifier. He recommended the depth of 3 m (9.8 ft) as a safe minimum and contended
that shallow tanks axe prone to failure under shock loads. No data showing effluent
solids concentrations versus tank depth were presented.
In 1984, Sackellares, et al (258) conducted research on three activated sludge clarifiers
with a diameter of 64 m (210 ft) and a sidewater depth of 4.1 m (13.5 ft). The authors
followed the techniques of dye injection and suspended solids measurement analogous to
that of Crosby (70). The tanks settled differentially taking the weirs out of level, and
the research was conducted to determine the effects of uneven weir loading around the
perimeter of the tank. The authors concluded that activated sludge clarifiers that are
relatively deep with numerous inlet ports perform better than others. Extensive data are
presented, but these data do not show a correlation between varying depths and effluent
suspended solids. Visual correlations between blanket level and effluent quality were
observed and it was inferred that deeper tanks would produce a more stable, higher
quality tank effluent.
The above references in this section represent a sampling of articles that qualitatively
relate clarifier performance to tank depth. Most such articles contain little, if any, data
-68-
-------
to explicitly describe the effects of depth 011 clarifier performance. Some recent
research, however, has attempted to look at clarifier depth as an independent variable.
Parker (229) presented data of monthly average effluent suspended solids concentration
versus clarifier depth for a number of treatment plants in the western states (see
Figure 4.1-34). Whereas the data do show a correlation of performance versus depth, the
following three points must be acknowledged:
a. The average overflow rates, as well as the depths, vary among clarifiers.
b. MLSS concentrations are not reported. Differences in MLSS concentration
likely influence the effluent SS concentration.
c. It is not clear if the differences in effluent suspended solids concentrations
from one unit to another are statistically significant.
Nevertheless, the data show average values decreasing from approximately 1? mg/1 to as
low as 5 mg/1 as tanks were deepened from 3.7 to 5.5 m (12 to 18 ft). Effluent quality
variabilities were also shown about each median. The values show a narrower range of
effluent solids for the deeper tanks implying that the greater depth provides for more
stable operation. In view of these findings, Parker recommended depths of 5 to 6 m (16
to 20 ft) for tank diameters of 27 m (90 ft) or more, hi looking at the data for the
Renton, Washington, old and new clarifiers, operating side by side at equal overflow
rates, increasing the depth from 4.3 to 5.5 m (14 to 18 ft) reduced the effluent suspended
solids from approximately 9 to 7 mg/1. This may be judged as a relatively small
improvement in performance for the increased cost of an additional 1,3 m (4 ft) of depth
although such improvement may be important at certain installations. The range of
values appeared approximately equivalent for the different depths at Renton. It should
be noted, however, that the old Renton clarifiers were peripheral feed whereas the new
ones were centerfeed with flocculating center wells.
The data by Parker, et al (230) also show that effects of depth are not totally
independent. Higher overflow rates were possible to achieve a given effluent quality
from deeper tanks,
Boyle (19) proposed a method of correlating SVI and MLSS to rationally size clarifier
depth. A graph was presented to determine depth, given a value of SVI and MLSS. For
-69-
-------
CLARIFIER DEPTH Cm)
3.3 3.7 4.3 4.9 5.5 6.1 6.7
35
30
O
P
<
K
H
SC
Ui
0
z
o
o
CO
s 25
_J
o
0)
a
UJ
s»
IB
3
CO
»-
z
Ul
z>
_l
u.
ll
UJ
Ul
a
<
ac
u
>
<
>•
z
»-
sc
o
15
10
179
UVERMORE, CA
LIG6ND:
•• 90-«RC£NT1LE VALUE
II 50—PERCENT1 LI VALUE
«• 10-KRCENT1LE VALUE
AVERAGE OVERFLOW
RATf,tpdm2.EXCLUD.
ING RAS FLOW ANQ
CENTER WELL AREA
MOTE
(UftaOJOM-m
(21 gpd/ft2 x 0.0*1
«m3/ci»v/m2
600
MEDFORD.OR
669
(OLD)
RENTON.WA
512
6eg
(NEW)
RENTON.WA
(WITHOUT
FLOCCUUSTOR
CENTER WELL),
748
SIMt. CA
1
CORVALLIS. OR
(WITH FLOCCULATOR
CENTER WELL)
I
10
12
14 16 18
CLARIFIER DEPTH, ft
20
22
Figure 4.1-34. Effect of clarifier depth and flocculator center well on
effluent suspended solids (Ref. 229).
-70-
-------
an SVI of 100, the required depths varied linearly from 3.0 to 4-3 m (10 to 14 ft) as mixed
liquor solids increased from 2,000 to 6,000 mg/1. Curves for other SVI values were not
presented nor was a data base presented from which the curve was developed. No
correlation between effluent suspended solids and depth was presented.
Stofkoper and Trentelman (278) discuss a design procedure which incorporates a settling
index to size clarifier depth. The procedure is presented later in this section. In pilot
scale studies, Chapman (47) found the effect of clarifier sidewater depth to be
interactive with the influent flow to the clarifier. He performed tests with a 2.4-m
(7.9-ft) diameter clarifier and an activated sludge pilot plant that treated degritted
wastewater. For an increase in flow to the secondary clarifier, he found the effluent
quality to deteriorate uiore at lower depths than at higher depths. Figure 4.1-35
illustrates the jrelationship Chapman observed. The clarifier equation developed by
Chapman illustrates the need for caution when using a regression model for prediction.
Chapman observed poorer solids removals when the feed flow rate (overflow and
underflow) increased. However, at sidewater depths greater than 2.2 m (7.2 ft), the
model predicts the opposite trend (143). This trend is shown in Figure 4.1-36.
Rather than explicitly investigating depth, research has often been conducted to
determine the impact of detention time on clarifier performance. The definition of
secondary clarifier detention time used by investigators is somewhat unclear. A
commonly accepted definition is based on clarifier overflow; however, it has also been
defined based on clarifier feed flow (52).
Fitch (108) performed column settling tests with a calcium carbonate suspension.
Plotting detention time versus overflow rate, he concluded that solids removal was
governed more by detention time than by overflow rate for flocculant suspensions. His
results are shown above in Figure 4.1-11. Scale and definition limitations of Fitch's work
are discussed above in the surface overflow rate portion of this section.
Full-scale studies of 20 activated sludge plants in Sweden were conducted by
Fischerstrom, et al (146). The turbidity of the clarifier effluent decreased with
volumetric sludge concentration and with increase in true detention time (as by dye
curves) up to one hour; further increases in detention time have no significant effect.
Figure 4.1-37 shows the correlation. The ratio between actual and theoretical detention
-71-
-------
SWD = 1.78m
MLSS{q/l)
1.5 1.6 1.7 1.8 1.9 2.0 2.1 2.2 2.3 2.4
FEED FLOW RATE PER UNIT AREA-(Qi/Qr)A(m/h)
2.5
Figure 4.1-35. Predictions for effluent quality with sidewater depth
constant (Ref. 47).
-72-
-------
X-2000 mg/l
1.50
1.75
2.00
ui 30
2.50
CLAR1F1ER INFLUENT FLOW DATA PER UNIT AREA (M/hr)
Figure 4.1-36. Predictions of effluent quality with MLSS constant using
Chapman's model (Ref. 143).
-73-
-------
2P
5000
2500
—effluent
•
0.S M/h
0.25 M/H
30
SO
90
120 MIN
Figure 4.1-37 Turbidity in clear water zone in ZP-units (Ref. 146).
time in the clarifiers varied between 0.2 and 0,5. Tests were also compared to settling
in columns. The effluents from the sedimentation tanks were less turbid than from the
liquid settled in a long tube for the same period. They attributed the difference to the
greater turbulence in the clarifiers causing agglomeration of the suspended solids.
Density currents observed in the full-scale tanks were found to have an effective depth
of about 0.6 m (2 ft) regardless of the actual depth of the tank. The tanks examined had
depths ranging from 1.4 to 33.4 m {4.6 to 11 ft}.
In another study by Rex Chainbelt (251} discussed above, a regression equation developed
on the basis of full-scale tank performance tests showed detention time to be
statistically more significant than overflow rate in predicting performance. Likewise,
Dietz and Keinath (169) presented a model developed from a laboratory scale facility
using a calcium carbonate suspension. Clarification efficiency was found to be strongly
-74-
-------
correlated to the detention time in the clear zone above the influent feed well, while the
effects of overflow rate were found to be insignificant. As discussed above in the
effects of overflow rate section, Tuntoolavest, et al (294) and Villier (301) found the
effects of overflow rate and detention time to be approximately equivalent. Lee, et al
(104) conducted model simulation and optimization of multi-stage final activated sludge
clarifiers. The authors used this model to predict that unless scouring occurs,, the
volume or detention time is not the major factor governing the removal efficiency in
either single-stage or multi-stage clarifiers. Their theoretical investigations covered
tank depths from 1 to 7 m (3 to 22 ft). The optimum theoretical depth was
approximately 2 m (6.5 ft). This value is much less than the deep tanks proposed by
Parker (229) suggesting that the theoretical developments were not simulating full-scale
performance; although the work by Lee, et al was intended for rectangular clarifiers and
Parker was discussing circular tanks.
Based on tracer studies of full-scale plants in Poland, Mazurczyk and Smith (203157)
questioned the importance of detention time on clarifier performance. They did not
conduct the full-scale tests on effluent quality versus depth or detention time directly.
German clarifier design guidelines (ATV), as discussed by Stofkoper and Trentelman
(278), includes a rational approach to sizing clarifier depth. To size tank depth, four
zones are considered in the ATV guidelines:
• Thickening zone
• Separation zone
• Water zone
• Buffer (storage) zone
The size of the thickening zone is determined from the estimated sludge volume. The
separation zone is set at 0.8 to 1.0 m (2.62 to 3.28 ft), with a lower bound of 0.5 m
(1.64 ft) at peak flow. The water zone is designed for a minimum of 0.5 m (1.64 ft). The
required buffer zone is a function of the decrease in dry solids concentration in the
aeration basin, which in turn can be determined by two methods:
• Perform a solids balance using a peak flow sludge return rate.
-75-
-------
• Select the maximum allowable surface loading at peak flow to determine the
peak sludge volume.
The ATV guidelines place a constraint on the design procedure such that the depth is
always greater than 2 m (6.56 ft). Another design constraint is that the dry solids
concentration in the aeration basin should not be allowed to decrease by more than 30
percent during peak flows.
Ditsios (88) conducted pilot scale investigations at the Graz, West Germany, wastewater
treatment plant. A rectangular tank 18 m (59 ft) long, 1 (3.28 ft) wide, and 1.75 m
(5.84 ft) deep was tested at flow rates of 0.5, 1.0 and 1.5 m/h (294, 589 and
883 gpd/sq ft). The results at 100 percent RAS rate (shown as RV = 100%) are presented
in Figure 4.1-38. The depth values plotted (h2,4) represent the depth from the top of
sludge blanket to the tank bottom. The curves indicate an exponential increase in
blanket thickness with an increase in overflow rate. The conclusion suggested is that
deeper tanks may be loaded at higher overflow rates so long as other variables are within
reasonable ranges and do not become limiting.
1.75
~ 1.60-
I 1-°°-
~
T*
ISV = 80 - 150ml/g
VSA = 200 ml/I
VSA = 100 ml/l
u>
z
o
0.5
1.0
1.5
SURFACE LOADING,qA(m/h)
Figure 4.1-38. Thickening and separation zone nEj4 as a function of surface
loading (Ref. 88).
-76-
-------
SUMMARY
The above text presents extensive data from tests on pilot and full-scale tanks, treating
mixed liquors from activated sludge, suspensions of chemical floes and raw sewage.
These tests have been performed to help determine important parameters in sizing
clarifier surface area and depth.
As summarized in Table 4.1-7, correlations between effluent suspended solids and
overflow rate is clear in some references, but essentially non-existent in others. In some
references, there are insufficient details regarding all of the variables that effect
effluent quality. For example, blanket levels, MLSS, temperature, depth, peaking
factors, SVI, and other such variables are not all accounted for in all references.
Although it is obvious that extremely high overflow rates would lead to an increased
level of turbulence and prevent sedimentation, the sensitivity of effluent quality in the
normal ranges of overflow rates tested and reported upon in the above references do not
consistently show direct correlation. It is the author's opinion that if activated sludge
clarifiers are equipped with optimized inlet geometry, adequate tank depth and adequate
weir lengths, some positive correlation between effluent suspended solids and overflow
rates would be established. Because of the statistical variations of effluent suspended
solids with time, it appears to be difficult to quantify the effects of variations in
overflow rate (over the range of 0.85 to 1.7 m/h or 500 to 1,000 gpd/sq ft) on effluent
suspended solids. The interaction found by Chapman (47) and the comment by Parker
(229) suggest that effluent suspended solids and overflow rate have a stronger positive
correlation at shallower depths. In. this author's opinion, the surprising lack of
correlation found by Matasci, et al (197) is probably due to the rapid settling
characteristics of the TF/SC sludge solids and the uncommon, large tank depths at the
Corvallis, Oregon, plant. In a recent paper by Parker and Stenquist (227) data are
presented to show that effluent suspended solids concentrations remain low over a broad
range of overflow rates from clarifiers with flocculating center feed wells for activated
sludge as well as for TF/SC sludge. The authors attribute the phenomenon to be largely
due to the flocculating feed well design.
The above references relating to the correlation of effluent quality and solids loading are
summarized in Table 4.1-8. Similar to the comparison between effluent quality and
overflow rate discussed above, the correlations are not always positive and strong. In
-77-
-------
TABLE 4,1-7
SUMMARY OF OVERFLOW RATS VERSOS PERFORMANCE REFERENCES
Author
Eef.Ho.
Suspension Tested
Scale of Tank
Degree of Correlation
Comments
Camp
36,37,38
Dye dispersion
Pilot) small
N/A
Long narrow tanks have better dye
dispersion curves.
Parker
229
Mixed liquor
Full scale
Positive, strong
Cordoba-Molina
59
Diatomaceous earth
Pilot, 4 ft diameter
Positive, strong
Dick
84
Calcium carbonate
Plcxiglas column
Positive, strong
Dick and Johnson
85
Calcium carbonate
Plexiglas column
Positive, strong
Pflanx
234
Mixed liquor
Full Urcct., 1 circ.)
Positivet strong
Three German plants
Rex Chainbelt
251
Mixed liquor
Full
Positive, strong
Data from operating plants. Correlator
strength includes detention time change.
Tuntoolavcst, ert ai
294
Mixed liquor
Pilot, 3-la column
Positive, strong
WPCF
3D8
Raw sewage, primary clar.
Full
Positive, moderate
Figure 4.1-8
WPCF
308
Raw sewage, primary clar*
Full
None until 1Q00 gpd/
sq ft, then positive,
moderate
Figure 4.1-9
Matasci, ejt al
199
Raw sewage, primary clar.
Full
Positive, moderate
Matasci, et al
199
TF/SC mixed liquor
Full
None
Mendis And Benedek
209
Mixed liquor
Tube settler
None
Fitch
108
Calcium carbonate
Pilot column
Positive, moderate
Flocculant settling
Cashion and Keinath
39
Mixed liquor
Pilot
None
Chapman
47
Mixed liquor
Pilot, 2.4 m diameter
Positive, moderate
Strong Interaction with depth.
Matsunaga
200
Mixed liquor
Full
Positive, moderate
Two-story tanks
-------
TABLE 4.1-8
SUMMARY OF SOLIDS LOADING RATE (MLSS) VERSUS PERFORMANCE REFERENCES
Author
Ref. No.
Suspension Tested
Scale of Tank
Degree of Correlation
Comments
Pflanz
234
Mixed liquor
Full (2 reel., 1 circ.)
Positive, strong
Three German plants
Chapman
47
Mixed liquor
Pilot, 2.4 m diameter
Positive, strong
Strong interaction with depth.
Tuntoolavest, ct al
zn
Mixed liquor
Pilot, 3-la column
Positive, strong
MLSS roost statistically significant
parameter.
Rex Chalnbelt
251
Mixed liquor
Full
Negative, moderate
Data from operating plants.
Villi er
301
Calcium carbonate
Bench
Negative, strong
Takamatsu and Naito
2B1
Calcium carbonate
Pilot
Positive, strong
Munch and Fitzpatrick
214
Mixed liquor
Full
Positive, moderate
Data from operating plants, limited data.
Cashlon and Kelnath
39
Mixed liquor
Pilot
None
Experimental loadings below typical
range.
Maxurczyk and Smith
203
Mixed liquor
Full
None
Data from operating plant.
Mendls and Benedek
209
Mixed liquor
Full
None
Tube settlers
Stofkoper and Trentelraan
278
Mixed liquor
Full
Positive, moderate
Looked at sludge volume loading.
Reach
250
Mixed liquor
Full
Positive, moderate
Looked at sludge volume loading.
-------
general, however, the literature associated with lull scale tanks supports a positive
correlation between solids loading and effluent TSS. It is not possible to prepare a single
curve or equation of effluent TSS versus solids loading to fulfill the objective of
providing designers with a new tool* It is the author's opinion that there are too many
other variables that interact. Such a relationship may be prepared for a given plant
when considering its expansion but a universal one does not appear to be feasible.
Data correlating effluent quality and MLSS were found to be quite scattered in MLSS
values between 1,000 and £,000 mg/1. Over a broader range of MLSS values (e.g., 1,000
to 10,000 mg/1) a positive correlation of effluent TSS with MLSS would be reasonably
expected.
The desire for having a straightforward method to determine area requirements for
thickening is largely satisfied by the flux analyses discussed above, except for the safety
factors that are needed to translate quiescent settling data to full-scale applications-
Additional research needs in this area therefore lie in measuring and predicting the
safety factors for different tank configurations and loadings.
The objective of correlating tank depth to effluent quality alone such that design
engineers could objectively determine an optimum value is overly simplistic. In most
publications, the correlation involves other variables such as sludge blanket levels, flux
variations and mixed liquor settling properties.
SUMMARY
On the basis of materials presented in the literature review and discussions at the
symposium, the participants jointly listed the following questions to be addressed in
research of activated sludge clarifier sizing:
1. What is the full-scale tank rating curve (effluent TSS vs significant parameters,
such as overflow rate, depth, solids loading, underflow solids concentration, sludge
blanket depth, settling characteristics, dispersed solids loading and RAS rate) for
each type of circular and rectangular clarifier:
-80-
-------
A. Circular types
• Peripheral feed/peripheral removal (PFPR)
• Peripheral feed/center removal (PFCR)
• Center feed/peripheral removal (CFPR)
• Center feed/inboard removal (CFIR)
B. Rectangular types
• Conventional (front end sludge removal)
• Gould (central removal)
• L.A. Co. (rear end removal)
2. What is the frequency and occurrence of peak flows and short term transients, and
how do they affect the stability of clarifiers?
3. What is the rational design variability of flux curves from plant to plant?
4. How do solids and liquids measures of hydraulic efficiency relate to clarifier
performance?
5. Are the foreign technologies applicable to U.S. plants?
6. What are the variables that best correlate with the performance of the best
clarifiers?
7. What is the maximum size of tank; especially considering wind effects?
-81-
-------
4.2 TANK SHAPE
A variety oi shapes are available for activated sludge clarifiers. This section describes
shapes listed in the literature and gives relative advantages and disadvantages of the
most popular ones.
SHAPES AVAILABLE
Secondary clarifiers are designed to utilize either radial or plug flow. Radial flow tanks
are built in a circular shape or in a modification of that shape (square, hexagonal, and
octagonal). Plug flow clarifiers are rectangular in shape.
Tekippe (289) conducted a survey of twenty major United States consulting firms
specializing in wastewater treatment plant design. Out of these, nineteen generally
preferred circular tanks over rectangular tanks for plants smaller than 2 cu m/sec (50
mgd). A number of consulting firms design rectangular tanks for plants larger than this
because the configuration lends itself to modular design and more efficient use of land
space. Economics, quick sludge removal, and better mechanism reliability were the
major reasons cited for using circular clarifiers.
ADVANTAGES AND DISADVANTAGES
A number of advantages and disadvantages of different shapes have been stated in
Tables 4.2-1 and 4.2-2, which were taken from the recently published WPCF manual
"Clarifier Design" (308). Numerous reviewers of the WPCF manual have criticized the
validity of some of the Table 4.2-1 entries to the authors of this report. One economic
analysis (210) showed rectangular clarifiers to be more economical to construct because
of their common walls between tanks which resulted in a major item of savings.
However, another author suggests (152) that common wall arrangements have drawbacks
in that they are difficult to bypass and modify should there be a need. Relative cost
comparisons between rectangular and circular tanks must be reviewed carefully because
there are often differences in redundancy and associated costs. Specifically, a smaller
number of circular tanks are often compared to a larger number of rectangular tanks
-82-
-------
TABLE 4.2-1. COMPARISON OF RECTANGULAR AND CIRCULAR CLARIFEERS
Rectangular Clarifiers
Circular Clarifiers
Advantages
Disadvantages
Less land and construction
cost in a multiple unit
design*
Longer flow path and less
chance for short-circuiting.
Higher limits for effluent
weir loading rates.
More even distribution of
sludge loads on collectors.
Longer detention time for
settled sludge.
Less effective for high solids
loading.
Shorter detention time for
settled sludge.
Better effect of dynamic fil-
tration.
Simple sludge collecting sys-
tem.
Low maintenance require-
ments.
Higher chance for short-cir-
cuiting.
Lower limits for effluent weir
loading rates.
Uneven distribution of sludge
loads on the collecting
devices.
Ref. 308
-83-
-------
TABLE 4.2-2. SUMMARY OF ADVANTAGES AND DISADVANTAGES FOR SEVERAL
CLARIE3ER CONFIGURATIONS
Parameter
Advantages
Disadvantages
Shape
Circular
Minimal construction
materials for single unit.
Simplest mechanical equip-
ment lor solids and scum
removal.
Requires most plant surface
area.
No common-wall economy of
construction.
Square
Low maintenance.
Common-wall economy of
construction.
Potential for common pipe
galleries.
Discrete inlet ports may cause
jets.
Corner sweep mechanism with
high maintenance.
More flow in directions of the
corners.
Hexagonal and Octago- Common-wall economy of
nal construction.
More construction materials
required for single units.
Skimming problems.
Filleted corners sloping to cir-
cular scraper mechanism.
More construction material
for single units than for cir-
cular tanks.
Ref. 308
-84-
-------
when economic studies are performed (289). A typical large circular tank is 40 m (130
ft) or so in diameter whereas a large typical rectangular tank is 6 m (20 ft) wide by
100 m (330 ft) long. The circular tank has twice as much area and therefore one half as
many tanks would be needed. On this basis, circular tanks are often found to be more
economical for large plants. If both shapes were to have the same area per tank, an
equivalent circular tank would be about 26.5 m (87 ft)in diameter. Equal numbers of
such small circular tanks would increase their relative cost and probably favor
rectangular units. Circular tanks are generally considered more reliable (289); therefore
an argument can be made that the reduced flexibility of the smaller numbers can be
tolerated. Pumping station costs must also be considered. In rectangular tank
construction, pumps may be located in common wall galleries whereas with circular
tanks a separate structure and longer feed and withdrawal piping are needed.
Camp (37,38) contended that long, rectangular tanks are superior to radial flow tanks for
two reasons: (a) minimization of inlet and outlet effects, and (b) minimization of short-
circuiting. His recommendations are based on experience and laboratory data that are
not presented in his papers.
Camp (37,38) stated that the inlet and outlet zones will each occupy a portion of the
flow length at least equal to the depth of the tank. Based on this premise, it follows that
radial flow tanks have only a very small settling zone after allowances are made for inlet
and outlet zones. For this reason, he recommended that settling tanks be designed such
that the length is 20 or more times greater than the depth and width (3S).
Typical dispersion curves for various tank configurations were presented by Camp (37,
38) and reproduced in Figure 4.2-1. These curves were used by Camp to support his
belief that short-circuiting can be minimized by the use of long, rectangular claxifiers.
A mathematical hydraulic analysis of radial and horizontal flow tanks (50) supports the
conclusions of Camp.
Fitch (107) questioned the recommendations of Camp for long, rectangular tanks. He
argued that, theoretically, distortion of the flow net at the inlet and outlet ends do not
reduce clarifier efficiency. He also argued that dye tests cannot be used to determine
the theoretical superiority of one tank shape over another. In addition, Fitch indicated
-85-
-------
5.6
5.2
4.8
4.4
3.2
2.4
2.0
—I-
0.8
0.4
t
Values of
A. theoretical sludge dispersion curve
B. radial flow circular tank
C. wide rectangular tank
D. long, narrow rectangular tank
E. round-the-end baffled mixing chambers
F. ideal tank
Figure 4.2.1. Typical Dispersion Curves for Tanks (Ref. 37}
-86-
-------
that long sedimentation basins may even be inferior, in that streaming introduced as a
result of additional side-wall drag may outweigh the benefit of reduced short-circuiting.
Although a number of model and full scale tests have been performed on clarifiers
(12,146,159j223,234,243), little data has been generated to objectively compare the
performance of circular and rectangular basins. Figure 4.1-8 indicates that tank shape
has little to do with suspended solids removal in primary settling tanks. Although
unpublished, the author of this report has observed rectangular and circular primary
clarifiers at Orange County Sanitation District's plant in Southern California in which
the overflow rates of the rectangular tanks had to be reduced to about 75 percent of that
of the parallel circular tanks. Since both shapes of clarifiers received the same
wastewater with the same flow variations and quality, it serves as a case history showing
large 46 m or {150 ft) diameter circular tanks to be better than rectangular tanks. This
contradicts the prediction that would be made based on the work of Camp (37,38).
A full scale study of parallel activated sludge systems operating at similar overflow
rates indicated that Gould tanks perform better than either circular or conventional
rectangular tanks (308). Gould tanks are rectangular clarifiers with relatively high
length-to-width ratios and mid-length or effluent-end sludge hoppers. The average
effluent suspended solids measured from parallel circular, rectangular, and Gould tanks
over a twelve month period was 11.4, 10.3, and 8.3 mg/1, respectively. Both rectangular
tanks produce a better effluent than the circular tank. Geometries were not presented
but the clarifiers were located in Phoenix, Arizona, and could be further researched.
Another aspect of tank shape is the bottom. Hopper bottom circular tanks are used in a
number of plants in Europe (250) but rarely, if ever, in the U.S. The ATV guidelines (278)
show that overflow rates may be increased by as much as 30 percent for certain types of
mixed liquors if such a bottom shape is used. Data by Resch (250) show that several full-
scale, hopper bottom (also called vertical flow) tanks perform even better than the
design guidelines indicate. Because such tanks are not in significant use in the U.S.,
further review of the European tank shape literature was not pursued.
-87-
\
-------
SUMMARY
On the basis of materials presented in the literature and discussions at the symposium,
the participants jointly listed the following questions to be addressed in research of
activated sludge clarifier shape:
1. What are the flow patterns within various clarifier shapes and configurations which
affect the transport of solids?
2. Are there any redeeming features of other than circular or rectangular clarifiers?
3. What is the relationship between clarifier shape and sludge removal mechanisms on
performance, reliability, O&M costs, etc.?
4. What is the relationship between bottom slope and clarifier performance?
5. How do the rating curves change as a function of shape?
6. What parameters are best correlated with process performance?
-88-
-------
SECTION 4.Z INLETS
Inlet design has long been recognized as a very important element in performance of
secondary clarifiers treating activated sludge (5, 36, 38,158, 78). The following roles for
clarifier inlets have been identified:
• Dissipate inlet energy
• Distribute the influent evenly in vertical and horizontal directions
• Reduce the impact of density currents
• Minimize disturbance of the sludge blanket
• Promote flocculation
The dissipation of inlet energy is generally considered to be a very important role of
inlet structures. Studies conducted in recent years have indicated that enlarging the size
of the inlet zone can promote flocculation of the suspended solids and improve removals
(229, 230, 279). They found the energy inherent in flow of mixed liquor is sufficient to
promote flocculation. Mechanical flocculators were provided in some tanks, but it was
found that the same degree of solids removal occurred with them turned off.
Comparative testing by Mau (202) and Fitch and Lutz (109) show that breaking the
influent flow to a clarifier into parallel streams and subsequently impinging those
streams against each other is one of most effective ways of dissipating inlet energy.
Specifically, Mau showed that use of a single row of slotted baffle boards was effective
in distributing the flow across the end of a rectangular tank, but had limitations in inlet
energy dissipation. He observed that construction of a second row of baffle boards
causing impingement considerably improved inlet energy dissipation. Fitch and Lutz
(109) likewise found that impingement was very effective in inlet energy dissipation in
center-feed circular tanks.
The distribution of influent evenly in vertical and horizontal directions is generally
considered desirable by design engineers ( 289). Data by Camp (38), Dague (74), Wallace
(304), and others that have studied hydraulics flow through clarifiers by the use of
tracers show that poor distribution or jetting of the incoming flow results in short
circuiting as evidenced by early initial trace and peak concentration values.
-89-
-------
Density currents are commonly observed in activated sludge secondary clarifiers and are
considered detrimental to clarifier performance (5). In an attempt to control density
currents through inlet design, Crosby (68) installed a full depth cylindrical center-feed
inlet baffle containing numerous ports with deflector plates in a full scale activated
sludge secondary clarifier. The inlet device failed to prevent the formation of density
currents or improve performance. Some inlets are designed to reduce the impact of
these density currents by feeding the mixed liquor into the lower portions of the tank or
spreading the mixed liquor flow over a large area (208).
Another function of clarifier inlets is to avoid disturbance of the sludge blanket to the
point of degrading effluent quality. In full scale tank research by Crosby (70), it was
shown that inlet baffles with bottom elevations near that of the sludge blanket surface
resulting in resuspension of settled solids and a deterioration of effluent quality.
In the United States it is commonplace for mixed liquor leaving the aeration basins to be
transported to clarifiers by means of open channels or enclosed pipes. In most instances
(289), designers in the United States insert flow-splitting overflow weir devices to obtain
positive equal distribution of mixed liquor to parallel secondary clarifiers. The divided
mixed liquor is then fed through pipes or gates before it enters the inlet of the clarifier.
Such mixed liquor transport does not necessarily destroy floe particles but does not
promote optimal growth. The concept of providing integral flocculation between the
aeration basin and the quiescent portion of the sedimentation tank could reasonably be
assumed to have merit. In fact, after years of research in clarifier design, Camp (38)
suggested several alternatives in design of rectangular clarifiers in which the first one
third or more of the tank volume was dedicated to flocculation using reel-type
(horizontal shaft), mechanically driven flocculators. Full scale data to prove an
advantage of such designs were not given, however.
Research by Chapman (47) also showed the value of flocculation to enhance removal of
suspended solids in pilot scale clarifiers. Such findings promote the classical contention
of Camp (38) that integral flocculation and sedimentation improve suspended solids
removal.
An increasing amount of full-scale plant data is being presented in the literature to
support the concept of inlet flocculation zones improving the performance of activated
-90-
-------
sludge clarifiers. Papers by Kalbskopl and Herter (153) and Parker and Stenquist (227),
for example, present such data for rectangular and circular tanks, respectively. The
latter shows large differences in effluent suspended solids, particularly at overflow rates
above 1.0 m/h (600 gpd/sq ft). These data are summarized in the following subsections
of this review.
The above describes a number of concerns that must be addressed in design of inlet
structures for all activated sludge clarifiers. hi addition to these general concerns, there
are many design features that are best described in reference to clarifier shape. For this
purpose, the following categories are identified:
o Rectangular clarifiers
o Circular, center-feed clarifiers (including square, hexagonal, and octagonal
basins)
o Circular, peripheral-feed clarifiers
Inlets for each of these clarifiers are discussed in the following subsections.
RECTANGULAR CLARIFIERS
The concerns of rectangular clarifier Inlet design can be conveniently categorized by
inlet configurations, dimensions, and flow patterns created.
Configurations
Although rectangular activated sludge clarifiers have been used successfully for decades,
the authors have seen very little standardization of inlet designs in observations at
dozens of activated sludge plants. Most tanks observed contain two or more openings,
carefully spaced across the end of the tank at elevations that range from at the surface
to mid-depth or deeper. There is usually some type of baffle immediately downstream of
these openings to prevent jetting of the flow into the tank.
Nearly all treatment plants employing rectangular secondary clarifiers have two or more
such clarifiers constructed in parallel with one or more common longitudinal walls and
fed by a channel connecting head ends of the tanks. The distribution of flow to the
-91-
-------
multiple inlets in a bank of such rectangular tanks is normally accomplished by a
manifold conduit system or by keeping the inlet ports relatively small, thus creating a
significant headloss between the feed channel or conduit and the contents of the tank. A
rule of thumb mentioned by design engineers from several firms to the authors is to keep
the loss of head through the port into the tank 10 to 20 times the magnitude of the loss
of head from one end of the feed channel to the other. More sophisticated hydraulic
methods to achieve good distribution have been published by Chao and Trussell (45).
For clarifiers fed from an open channel it is desirable to have a free surface into the
tank to allow the passage of floatable material. Met ports with a free surface have an
advantage in this regard. An alternative, however, is to use submerged ports and have a
downward opening gate located in one or more of the tanks to allow the passage of
floatables.
A number of authors have described a variety of inlet baffle facilities to achieve
satisfactory performance. Mau (202) conducted a number of model studies on the use of
vertical slotted inlet baffles. He compared a single row of such baffles to double rows in
which the openings of the upstream row were matched with baffle boards of the
downstream row. His results indicated that best performance could be obtained from the
double slotted baffles.
The design suggested by Mau (202) has been used successfully in the Randolph Park
Wastewater Treatment Plant in Tucson, Arizona, and, as per discussions with that
author, in other locations. The upstream row of baffle boards necessarily involves close
spacing of the baffle boards (on the order of 2.5-cm or 1-in). It is essential, therefore,
that primary clarifiers or fine screens be used upstream to ensure that debris does not
pass into the secondary clarifier and plug these rather narrow openings.
Mau (202) evaluated the effectiveness of various inlet designs by comparing dye tracer
curves. Full scale plants were later used to verify that satisfactory performance could
be obtained.
Rohlich (2546) conducted a number of dye study experiments using a simple inlet weir
followed by a straight downstream underflow baffle and various reaction jet baffles in an
attempt to identify an optimum system for use in American Petroleum Institute (API)
-92-
-------
separators. The resulting recommended design consisted of circular target baffles that
were dish shaped with an opening in the middle. Flow entered the tank by means of a
submerged orifice inlet. The flow then struck the target baffle and was deflected back
against the head end of the tank. Some of the flow, however, passed on through the
opening in the center of the target baffle. Hie flow that passed was effective in
minimizing the back current along the surface of the tank, normally resulting from
density currents. The remainder of the flow, deflecting against the head end of the tank,
induced flocculation and dissipated much of the inlet energy. This design has become
quite standard by API for separating oil and water mixtures. It has not, however, been
used extensively in activated sludge treatment.
A design somewhat similar to that discussed by Rohlich was recently investigated by
Collins and Crosby (54) and Crosby (70). These tests were conducted at a full scale
facility located at Holly Hill, Florida. The first set of experiments was performed with
reaction baffles as illustrated in Figure 4.3-1. The results showed that this baffle did
little to improve the performance above that of simple port openings. It was observed
that flow hitting the reaction baffle tended to plunge toward the bottom and scoured
some of the previously settled sludge from the sludge hopper located at the influent end.
The inlet reaction baffle was then modified to that shown in Figure 4.3-2. A bottom
plate was provided in an attempt to mitigate the density current induced scouring of the
hopper. In addition, the modified baffle eliminated the five holes of the original design.
It was found that rags tended to plug these openings and thus reduce their effectiveness.
Test results showed that the bottom plate did not significantly reduce the density
currents.
The problem of density currents at the inlet end of the lectangular tank was also
addressed by Imam, et al (144). These authors recommended inlet baffles of about 30
percent at the tank depth submergence to allow the incoming flow to spread across much
of the end of the tank and yet prevent scouring of the settled sludge. Data obtained
from a full-scale tank settling a discrete-particle suspension supported their contention.
The problem of density currents disturbing the sludge in the hoppers convinced Gould
(118) to locate the sludge removal hoppers at the far end of the rectangular tanks. Gould
therefore proposed feeding the tanks at a relatively deep elevation to minimize the
-93-
-------
Figure 4.3-1.
Sketch of one of the original four prototype reaction baffles
constructed for the Holly Hill, Florida, clarifier (Ref 70).
Sketch of modified reaction baffle at Holly Hills, Florida.
(Ref. 70).
-94-
-------
effect o£ density currents, but allow the density currents to move the sludge along the
bottom into the distant hoppers. Flight speeds were paced to match the velocity o£ the
density current moving along the floor. He determined an optimum flight speed to be
approximately 1 m/min (3 fpm) but did not show data to verify this value.
The same basic principle of removing sludge at the far end and using most of the inlet
end for flocculation and energy dissipation has been followed for approximately 25 years
by the County Sanitation Districts of Los Angeles County (CSDLAC). The Districts
maintain numerous rectangular secondary clarifiers at eight different activated sludge
treatment plants. Although side by side tank data have not been published, plant
operating records show that these tanks can be loaded to peak overflow rates of above
1.7 m/h (1,000 gpd/sq ft), while maintaining effluent suspended solids in the vicinity of
10 mg/1.
The standard secondary clarifier inlet structure design of the CSDLAC is illustrated in
Figure 4.3-3. It has been designed to induce flocculation and dissipate inlet energy by
impingement.
Camp (38} favored the use of mechanical flocculation at the inlet end of sedimentation
basins. He recognized the density current phenomenon in activated sludge plants and
favored integral flocculation to improve performance.
Kalbskopf and Herter (153) conducted full scale tests with paddle mixers used to
flocculate mixed liquor entering rectangular secondaries at the Emscher Mouth Treat-
ment Plant, West Germany. The facility treats wastewater from a municipal region of 5
million people. It has 72 clarifiers, 58 m (190 ft) long and 18 m (60 ft) wide.
Experiments involving (a) no flocculation chambers, (b) a separate flocculation chamber
containing two paddle (or horizontal shaft reel) mixers ahead of the clarifier, and (c)
inlet zone paddle mixers, produced the data shown in Figure 4.3-4. The data show that
flocculation in a separate tank improves clarification as measured by transparency of the
effluent. Inlet zone flocculation further improves it. These improvements were
observed for all flow rates ranging from 1 to 4 hr detention time. At 3 hr, for example,
clarification improved from about 17.5 cm (6.9 in) to 21 and 22.5 cm (6.4 and 8.8 in),
respectively. Thus, improvements may be significant but are not large.
-95-
-------
m:
PLAN
ELEVATION
Figure 4.3-3. Clarifier Met Diffuse* (Ref. 308).
2 2.5
INLET ZONE
FLOG.
SEPARATE
CHAMBER FLOC.
52.0
WITHOUT FLOC.
1.5
ui 1.0
1.0
2.0
3.0
4.0
CLARIFIER DETENTION TIME (HRS)
Figure 4.3-4. Improvement in the transparency of the effluent from the
sedimentation stage by stirring flocculation (Ref. 1S3).
-96-
-------
A number of Inlet designs were investigated by Price, et al (243). These tanks, all
located in Europe, had sludge removal hoppers at the influent end. All but one of the
tanks tested were treating primary sewage. The vertical and horizontal flow patterns
established in the tanks were correlated with influent design characteristics. The
authors concluded that a lack of symmetry in feed is to be avoided. Complicated inlets
were not found to be necessarily better than simple ones. They recommended that care
should be taken to eliminate scouring of the deposited sludge. They also observed that a
submerged weir uniformly fed can give reasonably good results without a baffle, but did
not show performance data to quantify differences.
A number of authors have reported on features of rectangular clarifier design practiced
by engineers in Japan (159, 160, 200, 223). In pilot studies of water treatment plant
clarifiers, Kawamura (159) recommended the installation of three sets of full cross
section perforated baffle plates. It was recommended that 2 to 3 ram (0.08 to 0.12 in) of
headloss be observed across each plate and velocity of flow through the orifice should be
on the order of 15 cm/s (0.5 ft/sec). Relatively higher removal efficiencies were
observed for activated sludge final clarifiers by Ohuno and Fukada (223) from tanks
containing perforated baffle plates. It was recommended that approximately 5 percent
open area be provided in these baffle plates. Hydraulic loading up to 2.1 m/h
(1,235 gpd/sq ft) could be achieved if three baffles were constructed in clarifiers with
the 5 percent opening criteria. Field analysis and dye tests were presented to verify the
author's design recommendations.
Matsunaga (200) presented design criteria for multi-story (two and three-story) rec-
tangular clarifiers used in activated sludge treatment. Inlet details were not studied in
depth, however, the designs showed means of eliminating short-circuiting that has been a
problem with a number of multi-story tanks constructed in the United States. This
reference is of value for construction of facilities in areas with severe space limitations,
but would generally not apply to most clarifiers designed in the United States.
Dimensions
Guidelines on sizing inlet feed ports are presented in a number of environmental
engineering textbooks and design manuals.
-97-
-------
The following listing of criteria is typical:
Design Data
Reference
hf = kfl (1 - r
hf
1,
3 £o
where,
hf = frictional head loss in pipe
from first to last port.
hfi= frictional head loss across
first inlet port.
hf0 = frictional head loss across
pipe when entire entrance
flow is to flow through the
pipe.
r = ratio of flow through the last
inlet port to that through
the first inlet port,
n = 1/2 for orifices
3/2 for straight edge weirs
5/2 for V-notch weirs
Channels designed for a velocity of
0.3 m/sec (1 fps) at one-half design flow.
Inlet port velocities 4.5 to 9 m/min
(15 to 30 fpin).
Max spacing between ports 3 m (10 ft).
"Unit Operations" by L. Rich (253)
Inlet port loss
Ten States Standards (122)
Joint ASCE-WPCF Design Manual (307)
EPA Manual on "SS Removal" (297)
The references cited in the above listing generally present the design data based on
observations or readings of their authors without secondary references to original
research or side-by-side comparison studies to quantify the effects of variations from
the dimensions suggested.
Figure 4.3-3 shows the configuration of a typical inlet diffuser of the CSDLAC (308).
The specific sizes of inlet pipe and diffuser components are not the same for all tank
sizes and flow rates. The relative geometry remains approximately the same. Most of
-98-
-------
the CSDLAC rectangular secondaries are 6.1 m (20 ft) wide and contain three diffusers
spaced across the inlet end at approximately mid-depth.
The research of Mau (202) led to his recommendation of the following dimensions for
double vertical-slotted inlet baffles:
Criteria
Values
Minimum velocity through the first 0.03 m/sec {0.1 fps)
baffle.
• Maximum slot spacings.
• Typical first baffle slot widths.
• Width of solid portion of the
second baffle.
0.61 m (2 ft}
1 to 10 cm (0.4 to 4 in)
Slightly wider than the
opening centered im-
mediately upstream in
the first baffle.
Mau (202) also states that a distance of from 9 to 15 cm (0.3 to 0.5 ft) between baffles is
satisfactory for velocities in the slots of the first baffle between 0.03 to 0.16 m/sec (0.1
and 0.52 fps). For distances between the baffles of leas than about 0.1 m (0.3 ft) at high
flow rates, the sideward velocity between the baffles was found to result in excessive
turbulence and poorer hydraulic performance.
Flow Patterns Created
Variations in the details of inlet structure design could reasonably be expected to
produce some variations in the flow patterns established within the tank. It has been the
author's observation that until the work of Crosby (70), such flow patterns were normally
measured by the injection of a single dose of concentrated dye solution and manual or
automatic recording of the dye concentration versus time in the tank effluent. Visual
observations of the dye movement and dispersion within the tank also provide insight to
the patterns of hydraulic movement (73, 287). Nevertheless, comments about the flow
patterns created were found in several references for rectangular tanks.
Specifically, as far back as 1943, Gould (118) commented that flow patterns of a
rectangular tank with the sludge hoppers at the outlet end exhibited a positive current
-99-
-------
along the bottom of the tank with a velocity of 1.5 to 3 m/min (5 to 10 fpm). Return
flows along the top of the tank were also observed. The flow pattern, though largely
attributed to the sludge hopper location, may also be attributed in part to the
introduction of the mixed liquor at the inlet end in the lower one-half of the tank depth.
Gould reported that such a low elevation was feasible because there were no sludge
hoppers that may otherwise have had problems of blanket scour.
Mau (202) used methylene blue tracers for visual observations of flow patterns created
by his experimental vertical slotted inlet baffles. Sodium phosphate was added to the
influent and used as a tracer for analytical measurements because recovery of the
chemical was better than methylene blue. The relative merits of the various inlet
configurations were determined by visual observations and such ratios as initial tracer
time to theoretical determine time (ti/T), modal time to theoretical detention time
(tm/T), and peak time to theoretical detention time (tp/T). He observed that the visual
short circuiting was minimal when double slotted baffles were used, and his analytical
results supported his visual findings.
By far the most informative literature findings illustrating flow patterns in rectangular
tanks were those reported by Crosby (70). Rectangular secondaries at the Holly Hill,
Florida, and the SandLake Road plant of Orange County, Florida, were studied using a
sophisticated multi-sample technique involving a continuous feed of dye tracer solution
to full scale tank influent. Numerous dye concentrations recorded at various time
intervals from the start of the dye flow were plotted on the longitudinal cross section
schematic drawing of the clarifier to show how the dye front moved through the tank.
At the Holly Hill plant, Crosby conducted side-by-side tests with the original inlet
structure, consisting of four 0.3-m (12-in) square ports and a 1-m (3—ft) deep full width
baffle located 1 a (3 ft) downstream. The flow pattern showed that the influent was
directed toward the tank bottom and moved rapidly toward the outlet end. A back
current at the surface was also observed.
The reaction baffles shown in Figure 4.3-1 were then installed and tested. The flow
patterns shifted somewhat with less severe velocities along the bottom of the tank.
Suspended solids removals were improved after the change in inlet design; however,
other changes in feed channel geometry and flow rate variations were also believed by
-100-
-------
Crosby to have influenced performance. Additional tests on the clarifier flow patterns
were conducted at Holly Hill after the installation of the modified baffles shown in
Figure 4.3-2. Less solids scouring were observed, but the basic flow pattern was not
affected.
At the Sand Lake Road plant in Orange County, Florida, Crosby (70) also tested the
hydraulic flow patterns in an existing rectangular clarifier with a travelling bridge sludge
removal mechanism. He observed that some bottom sludge near the inlet end was
degassifying and provided some vertical mixing of the influent. A resulting plug flow
hydraulic pattern was established as evidenced by Crosby's dye-front testing technique.
He considered the vertical mixing effect to complement the tank performance.
The author, along with staff from the CSDLAC, has observed the flow pattern and
settling characteristics of mixed liquor in the rectangular tanks with diffusers as
illustrated in Figure 4.3-3. Influent suspended solids were observed to have flocculated
within the first 10 ft or so of the inlet end. The solids then settle and move en masse
toward the hopper at the far end. Extensive flow pattern studies such as those of Crosby
(70) were not performed on these tanks.
CENTER-FEED CIRCULAR
Tekippe (289) found that thirteen out of twenty of the major consulting engineering firms
in the United States surveyed prefer circular center-feed clarifiers for activated sludge.
Of these, eleven prefer the standard center-feed design, and two prefer its variation, the
flocculating center-feed configuration. This section presents a critical literature review
of the process configuration, dimensions and resulting flow patterns.
Configurations
The basic center-feed circular tank inlet configuration consists of an inlet pipe
discharging into a cylindrical, bottomless baffle which directs the influent downward and
radially outward. As shown in Figure 4.3-5, the feed pipes may be either horizontal or
vertical; the latter preferred for large tanks because of economics (289). Based on the
author's experience in design and cost estimation, as tanks increase in size, the size of
the feed pipe also increases and becomes relatively more expensive to support within the
tank than burying it below the floor slab and bringing it up into the feedwell center.
-101-
-------
-------
According to Fitch and Lutz (109) and the opinions expressed by several equipment
suppliers to the author of this review, a serious disadvantage of many side feed designs is
the asymmetrical velocity pattern created by the inability of a center feedwell to
dissipate the influent velocity head. The asymmetry distorts the normal radial flow
pattern throughout the tank and can produce an overall reduction in clarifier efficiency.
A design by Fitch and Lutz (109) divides the horizontal inflow into two and impinges the
two halves against each other in a unique center feedwell design to reduce asymmetry.
They claim that use of a central vertical feed pipe eliminates this problem.
For designs with a central vertical feed pipe the use of an open ended pipe would
obviously minimize entrance port plugging. It is the author's opinion that the design
would unlikely retain any horizontal velocity vectors, but would inherently create a
rather sharp cascade of influent at the center of the tank. This type of influent
structure often forms a boil at the surface in the center of the tank creating a degree of
radial non-uniformity.
As evidenced by the equipment offered in catalogs of most U.S. clarifier suppliers (e.g.,
Eiraco, Dorr-Oliver, FMC), most manufacturers have gone to the vertical pipe with ports
or slots to project the flow radially outward into the influent well. The number and size
of inlet ports feeding the center feedwell has not been standardized. Very little
information has been found in the literature to provide guidance on the number and size
of these ports for circular tanks of different sizes. Sackeliares, et al (258) compared the
performance of three full-scale clarifiers, two with six inlet ports and the other with
four. The authors presented data showing that suspended solids removal efficiency was
significantly higher in the clarifier with the two additional ports. The fact that
differences were observed indicates that small standard inlet feed wells do not fully
dissipate the momentum jets of the feed ports.
The port hole design is amenable to any type of sludge collector mechanism. The most
complex collector is the hydraulic suction unit with riser-pipe, conveyance tubes. This
design is illustrated and discussed in Section 4.5. Interferences may result when the
vertical segment of the collector pipes passes in front of a port hole. The normal inlet
velocity pattern may then be disturbed and a jetting effect developed inducing
turbulence. The phenomenon was observed by Crosby (70).
-103-
-------
Recent papers by Parker (229, 230), Stukenberg, et al (279), and McKinney (208) show
that improved performance in the clarification of activated sludge mixed liquor can be
achieved by replacing conventional center feedwells with flocculating feedwells. The
latter are identified as being larger and equipped with some means of stirring the
contents of the well such as by mechanical mixing or controlled inflow velocities. These
larger feedwells provide more energy dissipation volume and also reduce the cascading
effect. Even though Tuntoolavest, et al (294) have cited biological floe bond strength to
be greater than that of some chemical floes (for example, alum or calcium phosphate)
research by Parker, et al (231) indicates that activated sludge floe particles are very
delicate. They may be broken up In part during transport from the aeration basins to the
clarifiers. Provision of flocculation facilities in the center feedwell enables some of
these particles to again grow, thus promoting sedimentation. If chemicals are added to
promote floe growth, flocculating feedwells optimize their success.
There are various types of flocculating center-feed clarifiers. A very simple, non-
mechanical design is illustrated in Figure 4.3-6 (232). The influent enters through feed
pipe ports into an inner feedwell with a bottom. The sidewalls contain a number of small
door openings that impart tangential velocity to the water leaving the well. The flow
then spreads and drops down into the tank through a large concentric outer feedwell
which extends down to 40 or 50 percent of the tank depth. The tangential stirring
imparted by the small gates induces flocculation with input energy derived from the
influent flow. The effect is advertised (232) as being beneficial in two w? rtz first, it
induces flocculation; and second, it dissipates the inlet energy over a larger volume.
The directional deflectors or doors attached to the inner chamber can be controlled by
chain linkage or automatically by a motor or hydraulic activator (232). The automatic
systems cure level controlled, thus imparting a controlled degree of energy dissipation
across the gates. Without such a device, the velocity through the doors would vary in
accordance with diurnal flow fluctuations. The effectiveness of such automatic control
versus the manual control offered by chain linkage has not to date been found in the
literature.
Some designs provide for mechanical mixing of the contents within the flocculation
chamber. For example, four slow-speed, pitch-blade impeller mixers or picket-fence
type paddle mixers rotating about vertical shafts are available (101, 232). This type of
-104-
-------
HM_r
L..—*- *
»;
%
w %\ i
i *
..... i
IF*—
^ n .
I , ( J 1 i
i:
\\
tj
*
!
(C^"
:
//
-rr^} rr^
| M
ry ¦—
r -
c\
/ \
M
)
ii
i
i
i
i
Figure 4.3-6. Simple, flocculating, ialet-energy-dissipation center feed
clarifier (Ref. 308).
-105-
-------
feedwell was recommended by McKinney (208) in 1977, tried on a full-scale basis, and
later reported on by Stukenberg, et al (279). The latter report good performance but
could detect no significant difference in clarification efficiency when the mixers were
turned off at one installation and set of operating conditions.
There are, of course, other flocculation chamber designs available from various
equipment manufacturers (232). For example, conically shaped flocculation zones sure
common in upflow clarifiers used for water softening and purification. The authors to
date have not seen them applied in activated sludge treatment of wastewater. It is
commonly known in the water plant design profession that the diurnal flow variations
make it difficult to form a stable blanket.
Dimensions
Based on the authors' observations of many activated sludge plants and discussions with
several U.S. equipment suppliers, a typical feedwell size for a conventional center-feed
activated sludge clarifier is 20 to 25 percent of the tank diameter. This size range has
generally been based on the criteria that downward velocities of the tank inflow do not
exceed peak velocities of 1.5 to 1.8 m/min (5 to 6 fpm) and an average downflow velocity
of 0.6 to 0.8 m/min (2.0 to 2.5 fpm). Some of the suppliers, however, propose feedwell
diameters that do not exceed 11 to 14 m (35 to 45 ft) regardless of tank size. A search
of the literature, however, has not found published data that correlate full-scale tank
inlet well size versus effluent suspended solids removal to define 'an optimum ratio of
inlet well size to tank size or such a similar relationship. If detention time is used as a
criterion, the work of Parker, et al (228) using bench scale facilities, may be of value.
The degree of primary particle removal at different energy gradients is compared at
detention times from 5 to 40 min. Longer times were found to improve removals.
According to suppliers, the submergence of the center feedwell has been varied from as
little as 30 percent to as much as 75 percent of the tank depth. Some of the
manufacturers have verbally recommended that submergence be two-thirds of the
sidewater depth. Allowing for a common 1:12 bottom floor slope, this would place the
bottom of the well near mid-depth at the center of large diameter tanks.
-106-
-------
Other manufacturers have verbally recommended that the bottom of the center feedwell
be located one foot below the bottom of the center-feed pipe ports. The submergence
would therefore not be directly related to the sidewater depth of the tank.
Another major U.S. manufacturer, in a letter to the author of this review, recommends
that the circumferential area below the feedwell should be about equal to the feedwell
cross sectional area to discourage any increase in velocity as the liquid enters the lower
portion of the clarifier. In this case, the opening under the feedwell would be measured
as the sidewater depth minus the feedwell depth. This requirement may conflict with
the clarifier feedwell velocity criteria and the sidewater depth criteria discussed above.
In large clarifiers with hydraulic suction sludge pickup, the peripheral area available
under the feedwell is more than that described by the sidewater depth minus the
feedwell depth. No published performance data were found to test these hypotheses for
optimal inlet geometry.
On some designs the inlet ports may be too close to the bottom of the feedwell baffle
resulting in high velocity jets that are carried into the settling zone. Most manu-
facturers recommend that the inlet baffle be extended at least one foot below the
bottom of such inlet ports. Again, such criteria are verbal recommendations for design.
Most manufacturers report to have or know about comparative research data but publish
very little of it to retain a competitive edge over other suppliers.
For the flocculating center feedwell design, Parker (229) recommends that the radius of
the outer feedwell be sized to obtain a detention time of 20 to 30 minutes. For a typical
tank and loading rate, the radius of the outer feedwell could be calculated as about 25 to
35 percent of the tank radius. As observed by the author of this review, a typical inlet
feedwell supporting the deflector gates has a radius of approximately 10 to 13 percent of
the tank radius. Most tanks contain eight or so diffuser gates.
Mechanical mixers, although not found necessary by Stukenberg, et al (279) for some
activated sludge applications, typically are located at a radius midway between the inner
and outer baffle walls (232). Upward pumping rotation with blades at about 20 percent
of the tank depth below the surface is provided to minimize the impact of downdraft on
the sedimentation of solids near the bottom of the tank. Variable speed drives permit
optimal degrees of energy input. It is the opinion of some engineers that the author has
-107-
-------
discussed this with that the use of turbine flocculators may create lloc breakup due to
localized high energy gradient at the tips.
Bradley and Krone (28), using a bench-scale, laminar flow apparatus, found an energy
gradient of 15 per second to be an optimum value for biological floe aggregation.
Research by Hui, et al (142) showed that energy gradients as low as 10 sec-1 may be
most appropriate floe formation of biological solids produced in activated sludge or
trickling filter/solids contact processes (trickling filters followed by short term aeration
with sludge recycle). The gradients were compared in jar-test beakers with mixed
liquors containing floes that were deficient in filamentous organisms and therefore
considered weaker than normal.
An extensive study of floe strength and factors contributing to it was reported on by
Magara, et al (194). These authors reported that activated sludge floe is three to six
times as strong as alum floe. Settleability of the floe was improved by increases in floe
size and density, with density being the "most significant physical property affecting the
settling."
Extensive studies of activated sludge flocculation were conducted by Parker, et al (228).
Results correlated primary particle removals as a function of energy gradient, G (sec-*),
and flocculator detention time. A typical plot is given in Figure 4.3-7 which shows
optimal energy gradients in the 40 to 60 sec"* range. Other experiments showed the
optimum to be in the 30 to 40 sec"! range. Maximum dimensions of peak sized floe
particles were also measured and plotted against energy gradient, as shown in
Figure 4.3-8. The results show the largest dimensions achieved in the 10 to 40 sec-*
range.
In some conventional tanks, the feedwell rotates with the sludge scraper mechanism,
whereas in others it remains stationary. No evidence was found that would favor one
design over another in terms of suspended solids removal performance. The feedwell can
be supported from the bridge or from the sludge collector mechanism.
Most of the feedwell designs have scum ports or small openings near the surface to allow
floatables to pass into the tank. Observations by Perkins and Wood (233) support the
premise that openings in the center-feed baffle at the surface to allow scum to enter
-108-
-------
r "U"
n* I *0.030
¦Bl ^
«j !~ wilder
G.mc"1
T«30 mill
Figure 4.3-7. Single flocculator performance for Series I (Re£. 228).
C SERIES M
2000
<00
•0
40
60
Figure 4.3-8. Maximum dimension of peak-sized floe (Ref. 2Z8).
-109-
-------
into the settling zone lead to short-circuiting. They recommend putting doors on the
slots that can occasionally be opened to allow the scum to leave the center feedwell, but
normally be in the closed position. Units without such openings often require manual
skimming if floatables become a problem. Data were not presented to show an effect of
floatables on effluent quality.
Flow Patterns Created
Studies as far back as Anderson (5) have shown that the flow pattern of a center-feed
circular clarifier involves inflow that moves downward and radially outward from the
center feedwell. As it reaches the wall, it rises and approaches the weirs. This
phenomenon was observed by such researchers as Dague (74) and Robinson (255) on a
pilot scale and more recently by Crosby (70) on a full-scale basis.
To be most meaningful, the various design configurations and dimensions discussed above
should be related to the resulting flow patterns. The side-feed, center feedwell discussed
above has an obvious horizontal velocity vector to be dissipated if a true radial flow
pattern is to be established. According to Fitch and Lutz (109), the distorted flow
pattern was serious enough to warrant design of a special feedwell to create
impingement and maximize dissipation. Dye curves were used to evaluate the
differences of various designs, but the techniques were not related to three dimensional
liquid movement to help quantify affects on performance.
For the conventional, vertical pipe, center feed design, model studies by Robinson (255)
demonstrated that this type of influent feedwell produces a velocity pattern typical of
that shown in Figure 4.3-9. Similar patterns were observed by Crosby (70) and Murphy
(217), the former at full-scale.
The work of all the authors cited in the above three paragraphs have indicated that small
center feedwells typically produce a density current waterfall effect that can be quite
pronounced. The pattern results in a solids deposition profile which is higher in the
center of the tank and gradually decreases with distance from the center. McKinney
(208) contends that the waterfall effect could be considered advantageous in creating
some degree of sludge compaction in center-feed tanks with scrapers and a central
collection hopper. Data by other researchers such as Anderson (5) and Crosby (70) show
-110-
-------
V
1
D
fn
JU
q
n -
lL>
Stagnant zone
Blanket
filtration
zone
f
Compacted sludge
Figure 4.3-9- Typical velocity pattern of center feed tank (Ref 255).
-------
an increase in blanket elevation near the wall. The former had clarifiers with scraper
mechanisms whereas the latter studied clarifiers with hydraulic sludge removal. Such
results suggest preferential transport or inadequate solids pickup at the tank periphery.
A typical pattern of deposition is presented by Anderson (5) and is shown in
Figure 4.3-10.
The most valuable, comprehensive work on measuring How patterns in circular clarifiers
was presented by Crosby (65 through 71). His report to EPA (70) contains 220 figures to
illustrate physical components of clarifiers, most of which were circular, and flow
patterns measured by tracers and samples of suspended solids. This work is considered
by the authors of this review to be of particular value because the research was
performed using full scale tanks. Flow patterns were developed for the following center
feed tanks:
Design
Overflow
Plant
Diam. (m)
Depth (m)
Rate (m/h)
Albuquerque, New Mexico
41
4.0
1.16
Dallas Co., Texas, Ten Mile Creek
32
3.7
1.34
Morganton, North Carolina
24
3.0
1.36
Oakland, California, EBMUD
14
3.0
1.07
Stamford, Connecticut
40
4.0
1.28
At Albuquerque, the center feedwell was about 7.9 m (26 ft) or approximately 20 percent
of the tank diameter and extended to mid-depth. Dye tracer tests showed that the flow
pattern was typical in that the inflow dropped to the bottom of the inlet baffle and then
continued in a near-horizontal, radially outward pattern (see Figure 4.3-11). The sludge
blanket surface was slightly above the feedwell bottom. Crosby (70) considered the tank
design to be adequate to the point of not recommending structural changes and further
experimentation.
At the Trinity River Authority's Ten Mile Creek Regional Wastewater Treatment Plant
in Dallas County, Texas, Crosby (70) studied a single 32 in (105 ft) diameter, 3.66 m
(12 ft) sidewater depth circular center feed activated sludge clarifier. Its maximum
depth at the center was 4.9 m (16 ft). Its inlet feedwell was 4.7 m (15.3 ft) in diameter
-112-
-------
WFLUEWT THRU CENTER FEED
19.2M RADIUS
RADIAL FLOW DlFFUSER
PERIPHERAL EFFLUENT WEIR
AFPRQX. WS EL ~4 80
EFFLUENT 15 MG L
INFLUENT WELl
m
FEED
2290
.MG'L
1M
2M
.57,
764
to.ooo
-0 i
DRMV-OFF 18.900 MG L
NOTE;
TAIL OF ARROW INDICATES START OF VELOCITY READING
VELOCITIES IN «M S
Figure 4.3-10. Pattern of solids deposition in activated sludge clarification
(Ref. 5).
-113-
-------
and was submerged to a depth of 2.1 m (6.8 ft). Radial dye distribution test analogous to
that plotted in Figure 4.3-11 were likewise conducted. The results showed that the
influent dropped deep into the center of the tank, flowed radially outward, and upward
along the peripheral wall toward the weirs. Near-surface recirculation back toward the
center feedwell was also observed. At an overflow rate of 1.33 m/h (784 gpd/sq ft),
Crosby observed an uneven sludge blanket surface just below the level of the inlet baffle
bottom. He suggested that this condition represented a turbulent flow pattern, but
observed relatively low concentration solids distributed across the entire surface. The
inlet feedwell diameter is approximately 15 percent of the tank diameter. This is
smaller than the 20 to 25 percent range often used. However, Crosby did not comment
on its adequacy based on resulting flow patterns.
At Morgan ton, North Carolina, Crosby (70) studied two 24.4-m (80—ft) diameter center-
feed clarifiers with sidewater depths of 3 m (10 ft). The inlet feed wells were 16 ft in
diameter and submerged to a depth of 5 ft. Similar to the Ten Mile Creek clarifier
discussed above, the influent was found to plunge to the bottom of the tank, flow radially
outward along the bottom and upward along the wall toward the effluent weirs. The
phenomenon at Morganton was judged by Crosby to be quite severe. In an attempt to
counter it, Crosby directed the installation of a mid-radius ring baffle as shown in Figure
4.3-12. He subsequently conducted a dye and suspended solids measurements within the
contents of the tank. The resulting dye concentrations after 38 minutes of continuous
dye injection are shown in Figure 4.3-12. The baffle broke up the high radial velocities
along the floor by forcing flow up over the mid-radius baffle. A wall rebound effect
remained, but was less severe.
The mid-radius baffle when tested on a side-by-side comparison to the original design
resulted in a reduction of suspended solids from 89.1 mg/1 for the unmodified tank to
52.3 mg/1 for the modified tank—a 41.3 percent improvement. Crosby felt that this was
due in part to flocculation within the mid-radius baffle. He therefore referred to the
chamber formed by the baffle as the ring baffle/flocculation chamber.
The author of this review finds the Morganton results to be of major significance. As
contended by Parker (229) and McKinney (208), the inlet energy can apparently be used
to induce flocculation in large center well structures and result in improved removal of
suspended solids.
-114-
-------
too
Figure 4.3-11. Radial distribution of dye concentration in Albuquerque,
New Mexico, secondary clarifier (Ref. 70).
20
.12
60
80
,100
48'
Figure 4.3-12. Radial distribution of dye concentrations in Morganton
clarifier with ring baffle/flocculation chamber (Ref. 70).
-115-
-------
Brief experiments were performed by Crosby (70) at the East Bay MUD process water
plant clarifier located at Oakland, California. This single secondary tank had a diameter
of 13.7 m (45 ft) with a sidewater depth of 3.0 m (10 ft). The feed well had a diameter
of 3.7 m (12 ft). The depth of the feedwell was not noted. The experiments provided
relatively little information. In the multi-point dispersion tests (consisting of taking
samples of dye at three points around the perimeter of the tank) showed several peaks in
the curve of dye concentration versus time at one of the three sample points, hut not at
the other two. Crosby suspected that this was due to vertical riser pipes of the sludge
removal mechanism passing in front of a porthole in the center feed pipe. No additional
testing was performed to verify this hypothesis.
The fifth and final center feed circular tank examined and reported by Crosby (70) was
located in Stamford, Connecticut. The Stamford Water Pollution Control Facility
included two circular elarifiers, 40 m (130 ft) in diameter with sidewater depths of 4 m
(13 ft). The bottom slope produced a center depth of 5.6 m (18.4 ft). The inlet feed
wells were 5.5 m (18 ft) in diameter and submerged to a depth of 1.8 m (5.8 ft) below the
surface. Inside the feedwells were four mixed liquor inlet ports positioned at 90 degrees
from one another. The center column had a diameter of 1.2 m (4 ft). Sludge was
removed from the tank by ten riser pipes, distributed along two scraper arms which
removed sludge by hydraulic suction.
Numerous tests on hydraulics and suspended solids patterns were conducted by Crosby
(70). Of significance in center feedwell design was the finding that the riser pipes when
passing in front of one of the four mixed liquor ports partially blocked the port, causing a
jetting effect of the mixed liquor into the tank.
Crosby (68) subsequently modified the inlet structure of one of the two Stamford
elarifiers. He did this by providing a full-depth, center feed inlet baffle containing 50
discrete inlet ports uniformly dispersed over the baffle surface with small baffle plates
extending outward several inches from the bottom of each port. The intent was to
evenly distribute flow at all depths and in,a radially outward pattern. The small baffle
plates were intended to minimize the influent density current.
Dye and suspended solids testing showed this modified baffle to be inferior to the
standard center-feed baffle that was previously in place at the Stamford plant. The in-
-116-
-------
tank measurement data showed that the mixed liquor inflow followed a density current
pattern to the bottom of the tank where it flowed radially outward. The baffle plates
were not effective in breaking up this fall.
Based on knowledge gained from his analysis of circular tanks, Crosby (70) considered the
relative level of the sludge blanket surface to that of the center feedwell bottom to be
important. He stated his belief that better performance could be obtained with a center
feedwell bottom that is either well above the sludge blanket surface or somewhat below
it. A shallow blanket thus separated from the bottom of the well was considered optimal
for good settling sludges, but not always so for sludges that settle poorly. In the latter
case, he claimed that it may be possible to improve performance by carrying a relatively
thick blanket that provides some degree of solids filtration, as well as settling which
would also provide more depth for compaction of the solids at the bottom. The concept
of filtration through a blanket to obtain good removals is supported by data from full-
scale Dortmund tanks in Germany (250).
Crosby (70) stated that a condition he felt to be avoided is when the bottom of the
feedwell is very near the surface of the sludge blanket; the latter was felt, in effect, to
form an artificial bottom to the tank. The feedwell bottom at that elevation creates a
relatively high radial velocity and turbulence of flow across the surface of the sludge
blanket that impeded settling and even resuspended solids.
In studies of full-scale field units employing mechanically flocculated center feedwells in
which four vertical shaft surface pitch-blade turbine mixers were used, Stukenberg, et al
(279) and Parker (229) did not show significant performance benefits from mechanical
mixing. The dissipation of inlet energy was considered by the investigators to be
adequate to achieve floe growth.
PERIPHERAL-FEED CIRCULAR
A common design alternative to the conventional center-feed circular tank is the
peripheral-feed tank in which the influent enters the tank at the perimeter. There are
many peripheral feed tanks in service across the United States. In the survey conducted
by Tekippe (289), six of twenty major consulting engineering firms in the United States
prefer peripheral-feed circular secondary clarifiers to any other type for activated
sludge.
-117-
-------
The basic principle of peripheral feed creates a radial inward flow pattern using ports or
pipes around the perimeter to distribute the flow evenly. In another design without Met
ports or pipes a tangential velocity of the inflow is created which produces a spiral flow
that moves toward the center of the tank and subsequently rises to the effluent launders
of the surface. The configurations, design criteria, and resulting flow patterns are
discussed below.
Configurations
A popular peripheral-feed secondary clarifier design developed by Envirex, Inc. (98) is
illustrated in Figure 4.3-13. The flow enters the tank bottom through a number of inlet
ports and flows downward under the skirt and then radially toward the center where it
rises and either leaves by way of an effluent launder from a central area or returns
radially outward to a peripheral weir.
An alternative style is the peripheral feed tank marketed in the United States by
Lakeside Equipment Corporation (183) illustrated in Figure 4.3-14. hi this design, no
orifices are used to distribute the flow, but a spiral pattern is established by introducing
flow into the tank body near the bottom as the flow passes below the peripheral baffle or
skirt.
Another form of peripheral-feed inlet involves the use of "downcomer" pipes. One
particular type known as the Kraus-fall peripheral feed clarifier is marketed in the
United States by Smith and Loveless, Kansas City, Missouri. A sectional schematic view
of this tank is illustrated in Figure 4.3-15. The downcomer concept is analogous to
multiple-port type except that downcomer pipes are used instead of the peripheral skirts.
The downward velocities in the pipe are higher than the downward velocity behind the
skirts; however, they are directed toward the center of the tank to minimize disturbance
of the sedimentation process.
A design that may appear somewhat similar to the downcomer concept was used at the
Detroit, Michigan Wastewater Treatment Plant (242). Diffuser pipes located around the
perimeter of the tank were constructed as shown in Figure 4.3-16. Early performance of
the tank was not satisfactory and modifications of the original inlet design were
performed to raise the elevation of the release point and lower the inlet velocity by
-118-
-------
D
u.
u.
. . Ui
Figure 4.3-13. Peripheral feed clarifier flow pattern (Ref. 308).
-119-
-------
SCUM PlPf
*N?M MANDl i
SECONDARY SLUOGE
A DIGESTER
SUPERNATANT
INI IT TROUGH
PRIMARY ClARlFiER
RECIRCULATION
TOSCUM
LINE
.SLUDGE LINE
m
Figure 4.3-14- Peripheral feed clarifier with "spiral roll" pattern of flow distribution.
(Ref. 308).
-------
Effluent Channel
Perif. Feed Channel
_ Kraus-fall
Downcomer
Pipe (typ.)
Figure 4.3-15. Kraus-fall ("downcomer") peripheral feed clarifier (Ref. 308).
-121-
-------
increasing the cross sectional area of the inlet diffuser. The design shown in
Figure 4.3-16 is the latest modification.
Dimensions
The basic overall dimensions of peripheral-feed clarifiers such as diameter and depth are
generally sized on the basis of overflow rate, depth, and detention time as discussed
earlier in this review (289). No published references were found to provide detailed
guidance to design engineers as to the sizing, spacing, and other such features of the
inlet skirts, feed ports, or downcomer pipes. The practice in the United States for design
engineers providing such clarifiers is to contact engineers from the equipment supply
companies, obtain these dimensions from the suppliers, and base their designs on them.
Based on discussions between the author of this review and an equipment supply engineer
of the peripheral-feed tank involving orifices, it was learned that early designs were
based on a headloss across the orifices of about 2.5 cm (1 in) at average flow. The
engineer stated that for many tanks this was satisfactory; however, for those with
peaking factors (peak flow divided by average flow) exceeding 3:1, some maldistribution
of inflow and solids deposition was observed. Design criteria were then changed to
provide more headloss (about 6 cm or 2.5 in) and equal distribution at average flow.
Peaks of more than 3:1 have been accommodated with the higher loss. When
exceptionally high peaking factors are encountered, a second row of orifices set at a
higher elevation in the walls of the feed channel are used to keep the orifice headloss
down and still obtain reasonable distribution. At the other extreme, that of diurnal
minimal flows, channel velocities were reported to be low and orifices losses minimal.
Some inequality of distribution was reported likely; however, under such conditions, the
overflow rates are low and clarifier performance was reported to be satisfactory.
Minimum flow maldistribution therefore was not generally considered to be a significant
problem.
Another element of concern in design obviously would be the size of the feed channel. A
manufacturer's representative engineer reported that a minimum velocity of 0.3 m/sec
(1 fps) should be maintained in the channel. The cross sectional area of the channel
therefore is decreased as the flow progresses around the tank, and the flow in the
channel is reduced by a fractional take off at each port or downcomer pipe. The design
-122-
-------
VENT MH
SCUM PIT
TROUGHS
GRATING
VENT MH
SCUM «
baffle
'OUTLET
OUTLET
PLAN
EL 107 0
EL 106 5
SCUM BAFFLE
outlets
45' BEND'
EL 89 34
SECTION A-A
Figure 4-3-16. Final clarifier with raised Met and diffuser modification
(Ref. 308).
-123-
-------
by Lakeside Engineering Corp. likewise provides for the skirt-to-wall width to decrease
as the incoming flow makes its way around the tank periphery.
Flow Patterns Created
A number of models and full-scale diagrams done on circular clarifiers have indicated
that peripheral-feed circular clarifiers have higher hydraulic efficiencies than center-
feed models (73, 287). Side-by-side full-scale tests performed in Sioux Falls, South
Dakota, showed that in addition to better hydraulic efficiency, peripheral-feed tanks are
also found to have achieved higher percent suspended solids removal than the existing
center-feed design at that plant (62, 178). The findings at South Dakota, however, have
not led to a universal preference of vertical feed tanks (289). This may be partially
attributed to the fact the center-feed clarifier in South Dakota was fed with a horizontal
pipe imparting some asymmetrical velocity vectors that hindered its performance com-
pared to other center-feed tanks.
Resulting flow patterns of peripheral-fe^'l circular clarifiers were studied in the mid-
1950s. A paper by Katz and Geinopolos (158) showed that dye tracer studies in which dye
concentration from a single injection plotted against time compared favorably with that
achieved of well designed rectangular clarifiers studied by Rohlich (256). These early
tests were conducted using the Envirex (98) multi-port peripheral design.
Crosby (70) conducted tracer studies on the large full-scale peripheral-feed equipment
clarifiers installed in the East Bay MUD Wastewater Treatment Plant located at
Oakland, California. The plant has twelve 43-m (140—ft) peripheral-feed, peripheral
overflow secondary clarifiers with a sidewater depth of 4.3 m (14 ft). The design
effluent flow rate to each tank is 0.44 cu m/sec (10 mgd). Flow to the peripheral inlet
channel splits just downstream of a servo-controlled butterfly valve to flow bi-
directionally around the tank periphery. Mixed liquor is introduced to the main part of
the clarifier from the peripheral channel downward through multiple, individually baffled
inlet ports. Tests were performed with the clarifier operating at 65 percent, 100
percent, and 150 percent of the design surface overflow rate.
hi Crosby's multi-point dispersion test, he observed that dye injected in the influent
arrived at the effluent weir in a fairly short period of time (40 percent of the theoretical
-124-
-------
detention time at 100 percent design loading), but was dispersed uniformly around the
perimeter of the tank. The radial distribution of flow across the cross section of the
tank was also extensively studied by Crosby. A typical plot of the results after 40
minutes of dye injection period at 100 percent of design capacity is illustrated in
Figure 4.3-17. Crosby reported that the incoming flow dropped to the level of the sludge
layer surface, progressed inwardly toward the center of the tank, began to rise and fall
outwardly toward the peripheral weirs. As flows increased from 65 to 150 percent of the
tank's design capacity, the dye front moved closer to the center of the tank before
making its ascent and outward movement. Crosby considered this change with increased
loading to be a stabilizing factor inherent in the design of the peripheral-feed tank. The
arrival time at the weir confirmed this, by being longer at higher loading and higher
overflow rates.
At 150 percent of the design flow rate, the East Bay MUD test clarifier experienced a
serious loss of solids to the effluent weir. In the basic clarifier design, the flow fed to
each tank was regulated by a large motor-operated butterfly gate which in turn was
controlled by a feedback loop involving measurement of a flow from each clarifier and a
computer to properly divide the total by the number of tanks in service. The tests
indicated that when the control gates were closed far enough to create a significant loss
of head, velocities entering the peripheral distribution channel caused uneven distribu-
tion among the orifices transmitting flow into the tank. Crosby attributed this uneven
distribution to be a problem with the design of the feed gate control system rather than
the tank itself.
In another study by Crosby, Young and Associates (71), the design of 20 peripheral-feed
clarifiers of the port inlet type was examined. Tracer and solids distribution analyses
were performed at different flow rates to determine characteristics of uneven
turbulence or maldistribution. It was concluded by those authors that the basins were
achieving good performance and no modification of the clarifier design was
recommended.
Data by Porta, et al (242) show that the modified inlet which was made larger and located
at a higher elevation than the initial design had reduced inlet velocities from as high as 6
m/s (12 fpm) to between 2.5 and 5.0 m/s (5 and 10 fpm) and reduce the velocity of the
return flow. The flow patterns before and after modification are shown in Figure 4.3-18.
Suspended solids removal was progressively improved by the modifications.
-125-
-------
I
I
Figure 4.3-17. Radial distribution of dye concentration in EBMUD secondary
clarifier, 40 min after the start of continuous dye injection
{Ref. 70).
-126-
-------
EFFLUENT WEIRS
SCUM BAFFLE
WATEn level
0-2 ft min
EFFLUENT
CHANNEL
3-8 It nun
INFLUENT
CHANNEL
12 i! mm
5-11 It mm
0-1 ft mm
INLET PIPE
UNMODIFIED CLARIFIER
0-3 ft mm
*>-10 ft mm
Tott
INLET PIPE
' & oiFFusen
4-7 ft min
0-1 K min
min
CLARIFIER WITH RAISED INLET & DIFFUSER
1-4 n min
ii
MODIFIED
10-15 ft mm
INLET PIPE
0-3 ft mm
3-6 ft mm
0-2 ft mm
CLARIFIER WITH RAISED INLET
Figure 4.3-18. Typical final clarifier velocity profiles, (a) unmodified
clarifier, (b) clarifies with raised inlet and diffuser,
(c) clarifier with raised Met (Ref. Z42).
-127-
-------
The clarifiers at Detroit were studied extensively by Anderson and Edwards (3).
Contrary to many other studies, mixed liquor added to the clarifiers through peripheral-
feed diffusers at about mid-depth did not appear to plunge to the bottom as many
predicted by density differences. Anderson and Edwards (3) studied the clarifier
performance using advanced modeling techniques of entropy equations and finite
difference algorithms to predict effluent quality. They measured and used the models to
predict the effects of varying solids flux into the tanks and the effect on such blanket
levels. The models presented were also shown to successfully model clarifier upset
conditions and predict underflow solids concentrations.
BOTTOM FEED
A novel concept of distributing influent to a circular clarifier with two rotating arms
located at 90 degrees to two vacuum sludge collectors was presented by USSR engineers
Koltsova, et al (177). The bottom moving distributor which releases mixed liquor
influent near the bottom eliminates the density current problem of cascading inflow.
The effluent is removed at the surface by withdrawal pipes with slots or orifices. These
pipes are attached to the same central rotating chamber and, therefore, rotate in the
same direction as the bottom distributor and sludge collector mechanism.
The authors report that "radial setting tanks have unsteady turbulent conditions
subjected to the influence of density and convective flows." They present data which
show that tanks equipped with the rotary feeder can handle up to twice the solids loading
rates with comparable performance. A claim of an 80 to 90 percent increase in tank
volume efficiency based on dye concentration peak travel times is also made. In the
United States, Dorr-Oliver has marketing rights for the concept and has been conducting
full scale tests in the metropolitan treatment plant at Denver. No data from the tests
have been published at the time of this writing.
SUMMARY
On the basis of materials presented in the literature and discussions at the symposium,
the participants jointly listed the following questions to be addressed in research of
activated sludge clarifier inlet designs
-128-
-------
1. For rectangular clarifiers, what is the effect of inlet design on flow distribution
between parallel clarifiers?
2. For rectangular clarifiers, what is the effect of inlet design on solids distribution
between parallel clarifiers?
3. What is the relationship betwen inlet design and floe formation and break up?
4. What is the optimum size of the flocculation zone of a circular and rectangular
clarifier? Also, consider no flocculation zone.
5. How can inlet energy be best used to promote flocculation in a rectangular tank?
6. What is the affect of inlet design on horizontal and vertical flow distribution within
a clarifier?
7. What benefits can be shown by increasing inlet baffling within a clarifier?
8. What are the shear characteristics of clarifier inlet structures and their effect on
flocculation?
9* What are the costs and benefits of retrofitting existing clarifiers with flocculation
structures/inlets?
10. What are the flow and performance characteristics of a Dortmond clarifier?
11. What is the effect of inlet on performance?
12. What is the effect of transport and distribution of mixed liquor on clarifier
performance?
13. What is the effect of transport and distribution of mixed liquor on aerator
flocculation and break up?
14. Do we need to skim inlet flocculation feedwells?
-129-
-------
4.4 OUTLETS
Outlet structrjes for activated sludge clarifiers are intended to remove effluent from
the clarifier without disturbing the sludge blanket or distorting the flow net that would
induce turbulence and impede sedimentation. Outlets normally consist of overflow weirs
or submerged orifices located to optimize a clarifier's hydraulic detention time and
minimize short-circuiting. A variety of outlet designs have been used over the years. As
discussed below, the designs vary considerably depending upon the shape of the clarifier;
however, a few criteria are common (122):
• Weirs should be adjustable for leveling.
• Weir loading rates, expressed in terms of cu m/m/day (gpd/lineal ft), are used
by many state regulatory agencies to guide the design of engineers.
• Weirs are to be designed so that they are not submerged at maximum design
flow.
• Weir troughs are to be designed to maintain a velocity of at least 0.3 m/sec
(1.0 fps) at one-half design flow.
Based on observations of many plants and discussions with many design engineers, it is
current practice to design V-notch, overflow weirs with notches located on 15- to 30-cm
(6- to 12-in) centers. Data to support these observations were not found in the
literature.
CIRCULAR TANK OUTLETS
Type
Effluent structures for circular tanks consist of effluent weirs, launders, or submerged
orifices. Outlet geometry choices consist of:
• Simple, peripheral weir
• Cantilevered, double or multiple launders
• Suspended launders (suspended from the bridge)
• Baffled, peripheral single weir
• Submerged orifices
-130-
-------
Simple peripheral weirs often consist of steel or fiberglass weir plates bolted on to the
tank walls or troughs as shown in Figure 4.4-1 (a). Launders can be cantilevered, such as
shown in Figure 4.4-1 (b) or suspended from an over-water bridge or other structure.
Submerged orifices are likewise supported structurally but consist o£ pipes with
numerous orifice openings located below the water surface such that the pipe flows full.
A downstream control valve may be used to control water level in the tank (192).
Placement
Activated sludge circular clarifiers often experience a phenomenon of solids rising along
the outer wall and approaching the outlet structure. This was observed as far back as
1945 by Anderson (5) and more recently by Crosby (70). Anderson (5) found better
suspended solids removal by locating the weir structure away from the wall similar to
the sketch of Figure 4.4-1 (b).
There is no consensus among practicing engineers today as to the best effluent weir
placement and design. In a survey conducted by Tekippe (289), the majority of major
consulting engineering firms in the United States prefer simple, single peripheral weirs.
The others preferred the cantilevered, double launder for most circular tank designs.
Consulting firm design specialists contacted in the survey gave a number of reasons for
their preferences, but did not cite side-by-side performance test results. Their
arguments in favor of the single weir included simplicity, low cost, and the ability to
skim the entire surface with simple, reliable equipment. Arguments favoring the double
launder were based on the premise that their location away from the wall allows solids
caught in the wall-effect updraft to resettle before the flow reaches the weir. Lower
weir loads were also cited as advantages. Evidence to support the claim of better
performance from double launders was given by Anderson (5) in his parallel tank study at
Chicago in 1945. He states that "a higher rate may be used for weirs located away from
the upturn of the density current than for weirs located at or near the upturn." Data
from 38.4-m (126—ft) diameter tank tests by Anderson (Figure 4.4-2) show that effluent
suspended solids varied from 10.6 to 18.2 mg/1, depending upon the placement and length
of the weirs.
Anderson stated that it was his opinion that for effluent weirs located away from the
upturn of the density current, the overflow rate should not exceed 250 cu m/m/day
-131-
-------
/TANK WALL
V
S4P
WEIR PLATE
V
SCUM BAFFLE
(NOT SHOWN)
(A) SIMPLE, PERIPHERAL WEIR
TANK WALL
WEIR PLATE
SCUM BAFFLE
LAUNDER
CANTILEVER SUPPORT
CB) CANTILEVERED WEIR
Figure 4.4-1. Typical circular clarifier weirs.
-132-
\
-------
(20,000 gpd/ft) of weir; and for weirs located within the upturn zone, the rate should not
exceed 186 cu m/m/day (15,000 gpd/ft). His tests were conducted with a mixed liquor
SVT of 52. It was his opinion that poorer settling sludges would show even more
difference between the two lines drawn through test points in Figure 4.4-2.
An extensive critical review of weir loading literature was performed by Graber (120).
He was unable to find actual plant data that showed any improvement in performance
from primary clarifiers when weirs were located inboard on cantilever supports as
opposed to at the tank perimeter. He presented the results of computer model
calculations to support his contention that reduced weir loadings should not improve
primary clarifier performance, but he did not conduct full scale or pilot scale tests to
verify his model results. He stated that he supported the contention of Anderson (5) that
activated sludge clarifier weirs should be placed on inboard cantilevers but, again,
showed no new supporting data. Likewise, Parker (229) has published support of this
concept but did not cite any additional parallel test results.
Parker (229) has recommended use of inboard launders, placed at about 75 percent of the
tank radius. For large tanks, he recommends use of two double-sided launders, placed at
60 and 80 percent of the tank radius.
In the survey by Tekippe (289), 70 percent of the 20 consultant firms contacted normally
designed simple peripheral weirs for circular activated sludge clarifiers as well as
primary clarifiers. Reasons cited for this position against cantilevered double launders
included:
o Higher costs
o More difficulty in hosing down the weirs
o More difficulty in keeping the launders level
o Inability to skim the outer surface {between the outer weir and the wall)
o Marginal, if any improvement in reducing the wall effect updraft
o Occasional lack of structural integrity
o Questionable improvement in clarification efficiency
-133-
-------
*>-<>•
-M-lr
ffl
TANK 124-0 01A.
II -o"».w.o.
W
*
CONCENTRIC CIRCULAR
WEIR TROUGHS. I2*WIDE,
WElR ON EACH SIDE.
20
2
Ql
ol
i
en
8'
O
in
o
UJ
§'
UJ
Q.
10
3
-------
Tekippe (289) reports visiting plants in which the outer weir was boarded off by the
operators to improve performance and the escape of solids from the wall effect. Tanks
in which these cantilevered weirs are located sufficiently far (say 20 percent of the tank
radius) from the wall reportedly work well (229).
The provision of a horizontal baffle placed below a single peripheral weir was studied by
Robinson (255). In pilot tests, it was demonstrated that this baffle deflected the flow
rising along the wall back toward the center of a center-feed tank, and suggested that
such deflection could eliminate much of the updraft wall effect and improve suspended
solids removal. This concept, illustrated in Figure 4.4-3, was carried to full scale by
Stukenberg, et al (279). Crosby (70) also experimented with a full scale perimeter baffle
sloped at 45° and located below the perimeter weir. The baffle addition achieved a 38
percent improvement in suspended solids removal. An example of the sloped baffle is
illustrated in Figure 4.4-4. This compares to the 42 percent maximum improvement
shown by the tests of Anderson (5) in which cantilevered inboard launders were used to
enhance suspended solids removals.
•0
1»
•0
20
40
CO
94*
60
•40
-20
Figure 4.4-4. Perimeter baffle at 45o siope (Ref. 70).
-135-
-------
Suspended launders (suspended from the bridge) have been used for years in the design of
peripheral feed secondary clarifiers. Envirex (formerly Rex Chain Belt) changed its
standard design (as shown in Ref. 98) from a suspended launder to a single peripheral
weir following research and a paper presented by Boyle (25) in the early 1970s. Data
published by Boyle shows that the single peripheral effluent weir performed as well or
better than the suspended launders. The peripheral weir also permitted skimming of the
surface by conventional skimmers and simplified construction and leveling of the weirs.
Another peripheral feed design, named Spiraflo by Lakeside Engineering Corporation
(183), retains the use of suspended launders. Data has not been found in the literature to
compare performance of effluent structure alternatives for the Spiraflo peripheral feed
tanks.
The authors of this report have observed that some engineers have resorted to the use of
serpentine weir structures in an effort to increase the weir length without resorting to
cantilevered double launders or suspended launders. This design practice has largely
been discontinued (289). No data could be found in the literature to support the premise
that the increased weir length of a serpentine design resulted in an improvement in
performance above that of the simple peripheral weir.
The use of submerged orifices in sedimentation tanks was discussed by Lutge (192). The
concept involves using an effluent control valve to determine the overall rate of tank
outflow and submerged orifices to remove that flow uniformly. Some design criteria
were presented for future design, but no data of suspended solids removal were
presented. Subsequent studies at the same plant were reported by Parker (229). Data
for the Renton, Washington, (old) clarifiers shown in Figure 4.1-32 were from tanks with
submerged orifices. The results compare favorably to those of other tanks, but the other
tanks were at other treatment plants or of different inlet design. The concept of
submerged orifices has not caught on, and very few final clarifiers are designed on the
principle today (289).
Dimensions
The dimensions of outlet structures for circular tanks are obviously influenced by weir
loading rates, tank size, velocity requirements, and placement criteria discussed above.
-136-
-------
Ten State Standards (122) specify that weir loadings should not exceed 124 cu m/m/day
(10,000 gpd/ft) for plants treating less than 0.044 cu m/sec (1 trtgd), nor 186 cu m/m/day
(15,000 gpd/ft) for larger plants. These criteria are based on peak flow including pump
delivery rates if pumping to the clarifier is necessary.
The issue of weir loading limits has remained controversial for years. The Ten State
Standards value of 186 cu m/m/day (15,000 gpd/ft) is not generally accepted by many
professional engineers. Specifically, the WPCF Manual of Practice No. 8 (307) states on
page 261 that "circular settling basins with diameters between 38 to 46 m (125 to 150 ft)
have performed well at weir loadings in the range of 370 to 500 cu m/m/day (30,000 to
40,000 gpd/ft)." It also states that "double launders on tanks less than 46 m (150 ft) in
diameter seem of questionable value."
These statements were made without supporting data or references to such data in other
publications. The Manual was, however, reviewed by 27 joint committee members
selected by WPCF for their expertise in wastewater treatment.
Parker (229) has recommended the use of weir loading limits of 100 to 150 cu m/m/day
(8,000 to 12,000 gpd/ft). No supporting data were given, however.
Cross sectional dimensions of the trough are normally made large enough to prevent weir
submergence and maintain the 0.3 m/sec (1.0 fps) velocity at one-half of design flow as
specified by Ten States Standards (122). It is commonly believed (215) that this velocity
is adequate to prevent solids deposition, but no data were found in the literature to test
this hypothesis.
RECTANGULAR TANK OUTLETS
Types
Most rectangular final clarifiers used in activated sludge treatment today are con-
structed with weir troughs that cover the final one-third or so of the tank length.
Alternatives include end weirs and submerged orifice effluent collectors. The latter
consist of pipes located near the surface which contain orifices to remove effluent from
a zone comparable to that of weir troughs. Downstream control valves or weirs are
normally provided to regulate flow.
-137-
-------
Placement
The simplest outlet design is an overflow weir located at the downstream end of a
rectangular tank# For long, narrow tanks one might speculate that this design would
provide an adequate hydraulic flow net, produce a reasonably good dye dispersion curve,
but remains vulnerable to the wall-effect updraft of suspended solids from the sludge
blanket. Pilot studies by Kawamura (160) show that simple overflow weirs establish
hydraulic patterns that are comparable to those of multiple weir outlets. Giles (116)
preferred a straight baffle at the end wall for activated sludge treatment.
In discussions with engineers of other firms (289) and engineers with the County
Sanitation Districts of Los Angeles County and the City of New York (which design
rectangular secondary tanks), the authors have learned that most design engineers in the
United States show a preference for the construction of multiple launders to remove
effluent over a broader portion of the clarifier outlet end. The use of multiple launders
permits a reduction in the weir loading rates to levels much lower than possible with end
weirs.
r
The ujidraft wall effect discussed for circular tanks is also a problem for rectangular
tanks. Designers have addressed the concern by locating the weirs several feet away
from the outlet end wall or by blocking off the V-notches along the last ten or so feet of
launders that run parallel to the tank length.
The orientation of the weir troughs is important for a number of practical reasons. For
clarifiers with chain and flight removal mechanisms, troughs may be oriented at 90° to
the length of flow. For collectors with bridge mechanisms, however, it is necessary to
orient the troughs parallel to the tank length to allow passage of the mechanism arms
between the troughs. No data has been found to favor one orientation of troughs over
another.
The effluent troughs in older rectangular tanks were often constructed of concrete with
weir plates bolted to the concrete with hole tolerances for leveling. Many modern
clarifiers, however, use fiberglass launders with metal supports.
-138-
-------
Dimensions
As for circular tanks, dimensions of outlet structures of rectangular tanks are influenced
by weir loading rates, velocity requirements, and placement. The Ten State Standards'
(122) stipulation of 124 cu m/m/day (10,000 gpd/ft) for small plants and 186 cu m/m/day
(15,000 gpd/ft) also applies to rectangular tanks. These values are, however, judged to
be too low by some engineers. Specifically, Gould (118) recommended that activated
sludge clarifiers with launders located away from the upturn of density current should
not exceed 248 cu m/m/day {20,000 gpd/ft), whereas weirs located within the upturn
zone should not exceed 186 cu m/m/day (15,000 gpd/ft). These values were also
recommended by Anderson (5). As stated above, the WPCF Manual of Practice No. 8
(307A) suggests that weir loadings can be increased to 372 cu m/m/day (30,000 gpd/ft)
without adverse effect. No data, however, are presented in the Manual to show that
these higher loadings do not have some effect on the updraft of solids and effluent
suspended solids.
Excessive lengths of weir to reduce the loading rate not only increase the cost of
construction, but cbviously also increase operational maintenance such as cleaning,
painting, and leveling.
Side by side tests of rectangular activated sludge clarifiers with different weir loadings
would be interesting to determine the relationship between effluent suspended solids and
outlet weir length.
SUMMARY
On the basis of materials presented in the literature and discussions at the symposium,
the participants jointly listed the following questions to be addressed in research of
activated sludge clarifier outlet designs:
1. What is the importance of weir loading on clarifier performance?
2. How do the performances of inboard launders and peripherally baffled weirs
compare?
3. What is the performance of radial launders?
-139-
-------
4. How do in-tank baffles affect performance?
5. What is the best configuration of an in-tank baffle?
6. What is the optimum location of effluent weirs for different tank configurations?
?. What techniques can be used to upgrade existing launders?
8. What is the role of tube settlers in activated sludge clarifiers?
9. What is needed for algae control?
-140-
-------
4.5 SLUDGE REMOVAL MECHANISMS
Settled activated sludge is collected and removed from the tank floor by a number of
different mechanisms. To be effective, these sludge removal mechanisms should collect
solids quickly to prevent denitrification and floating, or oxygen depletion and odor
formation while maintaining a concentrated sludge for feed back to the aeration tank.
Available sludge removal mechanisms for circular and rectangular clarifiers are dis-
cussed in this section. Advantages and disadvantages of each, as discussed in the
literature, are also presented. The objective of this section is to show what is needed to
select and design optimal removal mechanisms and what is available in the literature.
The needs of a design engineer attempting to select a sludge removal mechanism include
relative costs, knowledge of how quickly the mechanism can remove solids from the tank
and what effect the mechanism may have on effluent quality and the concentration of
the return sludge. Relative mechanism costs are readily available from equipment
suppliers and not the subject of a literature review. Long term O&M costs, however,
depend upon equipment reliability and repair or maintenance frequency. Such data
should be published but none have been found in this study.
The speed at which sludge can be removed from a tank is often stated as a reason to
favor one design over another (308). The need to do this has been discussed by Dick (81).
He summarizes by saying that it may very well be important to remove the sludge
quickly enough to prevent denitrification (and resulting floating sludge solids) and release
of phosphorous to solution; however, it may be desirable to leave the solids in the tank
long enough to obtain adequate compaction of the settled sludge. Excessive detention
times in some instances favor the growth of filamentous bacteria that subsequently may
impede settling.
This author is aware of tracer studies to measure solids residence times at certain
municipalities (e.g. County Sanitation Districts of Los Angeles County) but such data
found in the literature are quite limited.
-141-
-------
Circular Task Sludge Removal Types
Sludge that settles in circular tanks is removed by plows which transport the sludge to a
central collector hopper or by vacuum devices which pick the solids off the floor.
Primary and fixed film biological process secondary clarifiers use almost exclusively
plow mechanisms. Hydraulic suction types are generally limited to activated sludge
plant applications, where plow type mechanisms are also common.
Plow Scrapers. Scraper type elarifier mechanisms consist of severed staggered plows or
scrapers which progressively move the sludge toward a central hopper (see Figure 4.5-1).
The plows or scrapers are typically mounted at an angle to a double arm truss structure
which rotates about a vertical axis. The movement of activated sludge is predominantly
by hydraulic flow with the mechanism moving the heavy fraction of solids that do not
flow (208). The sludge hopper can be offset (Figure 4.5-la) or annular (Figure 4.5-lb) in
shape. Some plants, such as the Phoenix, Arizona, 23rd Avenue plant, have a
combination of central and annular hoppers.
Spiral Scrapers. In an attempt to move the solids to the center faster, European
engineers have developed a spiral shaped scraper mechanism capable of transporting the
solids from the outside wall of the tank to the center in one revolution of the
mechanism. Figure 4.5-2 shows a sketch of this design. Mathematics to serve as a basis
for design of spirals is given by Warden (305).
Hydraulic Suction Systems. Clarifier solids can be removed quickly from the bottom of
the circular tanks by using hydraulic mechanisms that lift the solids from the floor along
the entire tank radius (308). Because hydraulic head* differential is established by use of
pumps or adjustable weirs or gates, the head of liquid in the tank is used to transport
solids into the collector arm. Suction type mechanisms are used extensively to remove
the relatively light weight solids of activated sludge secondary clarifiers.
There are fundamentally two different types of suction removal mechanisms. The first
type, commonly called an "organ-pipe" or "riser pipe" type, has a separate collector pipe
for each inlet orifice. Each riser pipe has an adjustable telescoping weir, movable
sleeve, or ring arrangement which allows the operator to adjust the sludge flow
independently for each pipe. The other type of hydraulic suction collector has a single or
-142-
-------
Plan
-Tank diameter
Access walkway
¦Primary drive
Spur gear drive
Overflow
Scum
'dflflector Skimmi
Freeboard Scum outlet port-
Inlet ports
tnNuenl well
Scum trough
Truss arm
!?:
tScym
outlet
p— Center column
Adjustable squeegees
Inlet line
Sludge draw off
Sectional elevation
<«>
.Scum ramp and trough
Intluent feedweli
[ffluent weir and scum battle
.Turntable gear
Influent
Skimmer,
'Drive cage
Scum
Long arms
Underflow (sludge)
(b)
Figure 4.5-1. Concentric sludge hoppers (Ref. 308).
-143-
-------
Figure 4.5-2. Geometry of spiral collector (Ref. 289).
-144-
-------
double arm, called a header, extending across the full radius of the tank. The arm is
tubular and has a number of orifice openings. The sizing and spacing of the orifices are
predetermined by the manufacturer and the operator is not given the flexibility to vary
the relative amount of sludge removed from any selected tank radius.
A specification parameter discussed in the literature is the rotational rake speed. In a
pilot study, Chapman (47) found the effect of rake speed on effluent quality to be
negligible. However, Bender and Crosby {14) report a 10.5 percent reduction in effluent
TSS resulting from a reduction in rake speed to 56 percent of the design level at a full
scale facility.
Rectangular Tank Sludge Removal Types
Settled sludge in rectangular tanks is often collected by a chain and flight mechanism.
Traveling bridge collectors have been developed and used extensively in Europe for
rectangular basins handling flows greater than about 0.044 cu m/sec (1 mgd). They have
been used in the United States but to a minor degree (289).
Chain and Flight. The flights are attached by their ends to two parallel chains which
pass over sprockets (Figure 4.5-3). The flights move slowly along the clarifier floor and
scrape the settled sludge to the collecting hoppers at the end of the clarifier. At the
same time, the partially submerged flights serve as skimming devices in their return path
to skim any floating solids, grease, and oils.
Traveling Bridge. The traveling bridge mechanism equipped with a scraper is seldom
used for activated sludge secondary clarifiers. However, suction mechanisms are
relatively common. Two types of systems used with the traveling bridge design are the
airlift and centrifugal collectors (Figures 4.5-4 and 4.5-5), Other suction systems with
different traveling mechanisms, such as Clari-Vac and Trac-Vac collectors, have also
been used (308).
ADVANTAGES/DISADVANTAGES OF SLUDGE REMOVAL TYPES
Summaries of advantages and disadvantages of the different sludge removal types are
given in Tables 4.5-1 and 4.5-2 for circular and rectangular tanks, respectively (308).
They are further discussed in the following paragraphs.
-145-
-------
Scum Collector
Effluent Launders
Inlet Baffle
PLAN
Collector
drive
|nlluent-
—CfiX drive
PJ |"U I 1 [«¦¦¦ w m
jjjj r_ -iT
9115F
Sludge pipel
MT
Water level
I8YS1
Effluent
adjustable
Scum trough^ I ^Baffle yy weirs
Flow-
im trough^ I ^Baffle /\
Travel
Flights-. T
-V——
"0.
•>^.Emuem h
ELEVATION
Figure 4.5-3. Typical rectangular sedimentation tank with chain and
flight collector (Ref. 308).
-146-
-------
blower -Drive Discharge trough
Control panel
Skimmer Eduetor pipe
Oiffuser
Sludge guide
/
Air line
Intake port S
3
Figure 4.5-4. Typical airlift system (Ref. 308).
Control panel
Drive
*=¦
F
t
¦rJ
M
V
V
n
JI J
t.i
-T-
fV "T"
1
1' taVCrn
¦tier
SWm
Sludge guide
"Sludge pickup header
Figure 4.5-5. Typical centrifugal system (Ref. 308).
-147-
-------
TABLE 4.5-1. SUMMARY OF ADVANTAGES AND DISADVANTAGES FOR SEVERAL
CIRCULAR CLARIFIER MECHANISMS (REF. 308).
Parameter
Advantages
Disadvantages
Sludge Collectors
Plow scrapers
Spiral scrapers
Low cost.
Commonly available.
Proven track record of
performance.
Does not clog with raw waste-
water sludge.
Capable of having heavy
sludges.
Compatible with any scum re-
moval mechanism or feeding
system.
Low cost.
Maintenance from tank sur-
faces without dewatering the
tank for access.
Rapid sludge removal.
Long-time requirements to
remove sludge from tank (de-
nitrification potential).
Incompatible with some scum
removal mechanisms.
Commonly used with circular
sludge hopper that costs
more.
Suction mechanisms
Orifice (or manifold
type)
Flat floor accommodation.
Adjustable relatiave sludge
withdrawal rates.
Reduces denitrification
Reduces denitrification
flotation.
Less stirring than organ pipe
alternative.
Simplicity in design.
Often requires higher return
activated sludge flowratesj
therefore, more electrical
energy.
Narrow tubes clog easily.
High cost.
Potential for safety problem.
More stirring of tank.
Difficult to obtain uniform
withdrawal of sludge if dense.
Requires frequent adjustment
to tube flowrate.
Inability to make field ad-
justments.
Orifice clogging is not
easily detected.
Inability to make field ob-
servations.
Ref. 308
-148-
-------
TABLE 4.5-2. SUMMARY OF RECTANGULAR CLARIFIER MECHANISM
ADVANTAGES AND DISADVANTAGES (Ref. 308).
Parameter
Advantages
Disadvantages
Sludge Collectors
Chain and flight
Traveling bridge with
scraper system.
Traveling bridge with
suction system.
Serves dual function by re-
moving both sludge and. skim-
mings to collecting hopper
and trough.
Provides continuous scraping
to prevent excessive sludge
build-up.
Helps sludge thickening.
Is less expensive and requires
less maintenance compared to
chain and flight system.
Is not necessary to drain the
clarifier for any maintenance
of sludge collecting system.
Less likely to have sludge
accumulation problems.
Provides direct return of
sludge to aeration tanks.
"Is relatively effective for
high solids loading.
Is rather ineffective for high
solids loading.
is higher in maintenance cost.
Causes excessive accumu-
lation of settled sludge
which causes floating prob-
lems with nitrifying acti-
vated sludge.
Provides no steady flow of
sludge.
Causes flow surge in aeration
tanks or clarifiers.
Has higher wter content in
sludge solids.
Ref. 308
-149-
-------
Circular Tanks
la a survey conducted by Tekippe (289) most major XJ.S. consulting firms prefer hydraulic
over mechanical sludge removal for circular tanks. The reason most often cited was the
belief that sludge could be removed more rapidly by the former mechanism.
Parker (229) has recommended the use of the hydraulic suction, header-type mechanisms.
Sludge removal is obtained by pumping from a small wet well piped directly to the
header.
Suction type mechanisms also have some disadvantages when compared to scraper
mechanisms. Ih practice most treatment plants equipped with suction removal devices
return sludge quickly to the aeration basin buy may achieve less compaction of solids
than mechanisms with plows (279). If sludges are thin and/or deposition is uneven, large
amounts of tank water can be collected by the mechanisms. The "rat holing" effect for
scraper mechanisms also may permit this phenomenon. General comparative obser-
vations by McKinney {208) have suggested that dilute sludges are more common with
suction removal systems.
Data reported by Katz and Geinopolos (157) and also presented elsewhere (29) indicate
that higher return sludge concentrations can be obtained with mechanical sludge
collection systems than with hydraulic systems. Since the data were taken from
different plants treating different wastewaters (as seen by substantially different SVI
values), variables other than sludge removal type may have been more critical in return
sludge concentration.
The literature was found to contain relatively little published data on the effectiveness
of different removal mechanisms relating to sludge residence times or underflow
concentrations. In 1930, when activated sludge circular clarifiers were typically
equipped with scraper mechanisms, Townsend and Brower (291, 292) conducted research
on the first hydraulic mechanism (called "Tow-Bro") in the full scale 29 m (95 ft)
diameter clarifiers of Milwaukee. Depths of sludge blanket and percent solids in the
RAS were measured for different discharge pipe valve settings and resulting RAS flow
rates. The results are listed in Table 4.5-3. With the hydraulic suction device, it was
possible to maintain shallower blankets and still remove sludge at equal or larger sludge
solids concentrations.
-150-
-------
TABLE 4.5-3. COMPARISON OF HEADER AND SCRAPER MECHANISMS.
Hydraulic
Suction
Scraper
Position of discharge pipe
Percent solids in sludge
Depth (ft) of sludge blanket
Closed Position
1.26
2.5
Closed Position
1.36
6.00
Position of discharge pipe
Percent solids in sludge
Depth (ft) of sludge blanket
0.5 ft head
1.32
4.5
0.5 ft head
1.12
4.5
Position of discharge pipe
Percent solids in sludge
Depth (ft) of sludge blanket
1.0 ft head
1.38
6.5
1.0 ft head
1.00
6.00
Position of discharge pipe
Percent solids in sludge
Depth (ft) of sludge blanket
1.5 ft head
1.25
6.3
1.5 ft head
1.00
6.5
Position of discharge pipe
Percent solids in sludge
Depth (ft) of sludge blanket
2.0 ft head
1.10
6.2
2.0 ft head
1.0
7.0
Position of discharge pipe
Percent solids in sludge
Depth (ft) of sludge blanket
2.5 ft head
0.95
4.5
2.5 ft head
1.00
7.3
Mixed liquor solids approximately
.33%
.33%
Note: Divide ft of head by 3.28 to obtain m.
(Ref. 292)
-151-
-------
In 1956, Katz (156) reported that the conversion of a plow mechanism to hydraulic
suction solved a problem of rising sludge solids in the tanks at Sioux Falls, South Dakota.
He attributed the difference to the ability of the suction mechanism to remove solids
more quickly. Data quantifying sludge conditions, flows, concentrations or detention
times were not presented. Soon thereafter, Katz and Gianopolis (157) presented a paper
on solids retention times in circular tanks. An Oil Red T.A.X. dye tracer was added to
the mixed liquor entering the test tanks and concentrations in the HAS were measured
and plotted against time. Fifteen tests were conducted in tanks ranging from 7 to 38 m
(23 to 126 ft) in diameter. Basin depths ranged from 2,4 m to 4 m (8 to 13 ft). Hydraulic
"Tow-Bro" collection mechanisms were used to remove the sludge. The results of a
typical test are shown in Figure 4.5-6. Residence times of approximately 20 minutes
were measured.
Similar unpublished data from tests at the full scale Akron, Ohio, clarifiers were supplied
to the authors of this review by the manufacturers of the "Tow-Bro" hydraulic collection
mechanism (Figure 4.5-7). The peak dye concentration of the RAS was found to be about
10 minutes, whereas the centroid was about 25 minutes. The peak hydraulic residence
time of the overflow was about 25 minutes. These tests were run at a relatively high 2.5
m/h (1,500 gpd/sq ft). Additional unpublished tracer curves from the same supplies are
shown in Figures 4.5-8 and 4.5-9. The first is from a Tow-Bro, header-type hydraulic
mechanism, whereas the latter is from a riser-pipe hydraulic unit. The Tow-Bro
achieved a residence time of 23.4 minutes, and the riser pipe design obtained a time of
38.1 minutes. The differences in shapes of the curves probably relate to the larger
number of ports in the Tow-Bro mechanism.
Mazurczyk, et al (204) likewise conducted sludge tracer dispersion curve testing of full-
scale circular tanks in the Pruszkow, Poland, wastewater treatment plant. A radioactive
isotope was added to the mixed liquor and samples were collected from locations across
a section of the tank. The RAS tracer concentration was also monitored and recorded.
Tracer profiles taken at 10 to 20, 20 to 40, and 60 to 70 minutes after the dosage was
added are shown in Figure 4.5-10. The sludge moved rapidly along the sludge blanket
surface to the wall and then moved downward and back toward the center feed well. It
spread rather uniformly across the tank floor and, after an hour, tended to concentrate
in an area of about one-third of the radius from the center. Sludge removal was by riser-
pipe, hydraulic suction. Figure 4.5-11 shows the tracer concentration in the RAS versus
-152-
-------
l
•
*
*>
1/
f
-JfC
vjV*
g" —
—r1
*
•
•
* 1
180
TIME - minutes
Figure 4.5-6 Typical dispersion curve of hydraulic (Tow-Bro) mechanism
(Ref. 157).
-153-
-------
1 1 1 1 1
DIA. = 102 FT(31M) | | |
OVERFLOW RATE - 1500 QPD/SQ FTC2.65M/H)
1
\ D
I
\
YE
IN F
ETt
IRN
3LI
JDG
E
HI
M
;1UHN SLU
XED LIQUO
uafc hiuw
R FLOW -
- 3.! MUl
15.8 MQO (
i lU.IOM-/
3.70M3/S)
1
1
1
\
\
I
\ I
YE
IN 1
FFI
-UE
1
1
1
[
\
/
\
* /
y
1
1
1
<
_ •»
V
1
1
1
»
"V
1
f
T
*
0 5 10 15 20 25 30 35 40 45 50 55 60 75 80 105 120
TIME, MIN.
Figure 4.5-7. Akron, Ohio, dye tracer test results.
-154-
-------
-
SLUDGE TRACER TEST-TOW BRO
JL
AREA UNDER CURVE IS 77.10
MEAN TIME (Tg) =23.41 MIN
, VARIANCE =.4841
/
\ T10 = 8.77 MIN
— /
\ T90 = 41.8 MIN
\ T90/T10 = 4.77
/
\ T50 =19.23 MIN
— /
1 1 1 1 1 1 t
i i I i i ¦ i i i i i ¦ I i i i i i i i i i
0 12 3
t/Tfl
Figure 4.5-8. Normalized E curve - sludge tracer- Tow-Bro.
-155-
-------
1.5
1.0
o
o
¦*.
o
UJ
.5
SLUDGE TRACER TEST - RISER PIPE
AREA UNDER CURVE IS 78.24
MEAN TIME (Tg) =38.14 MIN
VARIANCE-.5488
T10 = 8.6 MIN
T90 = 86.65 MIN
T90/T10 =10.07
T50-30.41 MIN
0.0
-i L.
t/Tg
Figure 4.5-9- Normalized E curve - sludge tracer - riser pipe.
-156-
-------
14.0 b 12.0 m 10.0 m 8.0 a 6.0 n 4.0 m 2.0 n 0.4 a 0
10-20 mlnut
30-40 ninutit ifttr itirt of trietr riltiit
60-70 nlnutii «f:«r itirt sf trietr r»leit»
Figure 4.5-10. Sequence of sludge tracer profiles illustrating flow of solids in the final
clarifier using the isoconentration representation (Ref. 204).
-157-
-------
¦eroii-jeetfon soundings
m
700
* 600
200
haekaround from return sludge
100 ¦
tin*/ sludge collection aechanlsa position
Figure 4.5-11. Typical sludge tracer dispersion curve produced at the return
sludge well (Ref. 204).
-158-
-------
time. The tracer first appeared to leave the tank after 30 minutes of settling. The peak
concentration occurred after about two hours. The numerous small peaks were
attributed to solids distribution imperfections caused by 24 vertical slots in the inlet
diffuser baffle. The repetitive series of gradual peaks at approximately two-hour
intervals was attributed to non-uniform solids pickup because the time interval cor-
responded to that of the collection mechanism. The above tests were run with a blanket
thickness between 1 and 2 m (3.3 and 6.6 ft) in a tank with a sidewater depth of about 3
m (10 ft).
Similar studies were conducted at the Czestochowa, Poland, activated sludge plant (203).
The circular units there were 34 m (112 ft) in diameter and 3.5 m (10 ft) in sidewater
depth. As shown in Figure 4.5-12, the center feed well was located at the tank surface
and sludge was plowed into a central RAS funnel, or hopper. A relatively deep sludge
blanket was formed. Subsequent tracer profiles and the RAS time distribution diagram
(Figure 4.5-13) showed that much of the sludge short circuited from the inlet directly
into the central hopper.
The above data show some differences in the sludge movement through tanks with
hydraulic and scraper mechanisms. It is this author's opinion that such results are not
necessarily representative of differences in mechanisms alone. Because the tests were
at different plants, the mixed liquor characteristics, blanket levels, and temperatures
were different in addition to variations in tank geometry.
Rectangular Tanks
Advantages and disadvantages of mechanical versus hydraulic sludge removal are similar
for rectangular tanks as with circular tanks. Sludge hopper location is an issue that has
been investigated specifically for rectangular tanks (118). Often primary tanks are
designed with hoppers at the head end of the tank. Since scraping of the settled sludge
to the influent end hopper is performed against the normal density current flow
direction, biological floes can be easily broken and re-suspended (118). Therefore, the
influent end hopper design is not ideal for applications to the secondary clarifiers of
activated sludge plants. Clarifiers with effluent end or mid-length sludge hoppers are
sometimes called Gould tanks (308). Gould tanks avoid many of the hydraulic problems
of conventional clarifiers. At Phoenix, Arizona, the plant personnel report verbally that
-159-
-------
LENGTH FROM BASIN CENTERS IN METER8
[77777
7.0
100
110
¦U
WOO
x>
-300:
¦ft
£ -returned sludge lunnel
Figure 4.5-12. Distribution of total suspended solids marked by La isotope
in cross section Czestochowa radial settling tank (Ref. 204).
-------
450
*BD.
200,
ISO
100,
so.
13"
Figure 4.5-13. Detention time distribution of returned sludge measured
in run No. 7 (Ref. 204).
-------
recent tests of parallel tanks under similar overflow rates indicate the Gould tanks are
superior to conventional rectangular clarifiers, Effluent TSS concentrations below 10
mg/1 were reported but such data remain unpublished.
SUMMARY
On the basis of materials presented in the literature review and discussions at the
symposium, the participants jointly listed the following questions to be addressed in
research of activated sludge clarifier sludge removal mechanisms:
1. What is the relationship between sludge removal mechanisms and sludge residence
times?
2. What is the relationship between sludge residence times and effluent quality?
3. What is the effect of design and speed specifications of sludge removal mechanisms
on effluent quality?
4. What is the effect of sludge rheology on sludge flow within a tank?
5. What are the comparative O&M costs of the different sludge removal mechanisms?
6. What is the role of fixed bottom removal systems?
7. What are the relative merits of scum removal in the clarifier versus the Clg
contact basin?
8. What is the optimal system for scum removal?
9- Does the method of sludge removal effect the performance of the biological
process?
-162-
-------
4.6 OPERATION FACTORS
Effective operation of a secondary clarifier is critical in producing a high quality
effluent. The plant operator must make sure that the clarifier returns sludge in
sufficient quantity and concentration to the aeration basin, and at the same trie does
not lose large amounts of solids in the effluent. His primary variable is adjustment of
the return sludge flow rate (RAS).
Different operation strategies have been proposed in the literature to guide an operator
in determining sludge return rates {163, 164, 165, 166, 171, 237, 238). The effect of
recycle rates on clarifier performance is discussed in the literature {47, 234, 250, 250,
294, 301). Sludge blanket level, which is dependent on sludge return rate, is also
discussed with respect to clarifier performance {70, 85, 115, 203, 214, 229, 235, 258).
Control of sludge inventory by using a step feed procedure is discussed (29, 164, 165, 167,
170). Contact stabilization, an extreme case of the step feed modification, is not
discussed in the context of secondary clarifier performance.
The principal needs of a design engineer and a plant operator include knowledge about
the correlation between RAS rate, underflow TSS, overflow TSS, blanket level and sludge
characteristics. The designer needs to know what range of RAS capacity to provide and
what degree of control automation that an operator needs. The operator needs to know
what blanket level is optimal for effluent TSS and underflow TSS, how to control the
blanket level by RAS rate adjustments, and how to compensate for changes in sludge
characteristics and flow rates. Findings in the literature help meet some of these needs.
Another operational variable at some plants Is the flow rate of plant influent pumps.
Wide variations can create hydraulic transients which degrade effluent quality (54, 198,
242). The on/off operation of large influent pumps or recycle pumps must therefore be
avoided and aeration tank weirs designed to dampen hydraulic transients (224).
-163-
-------
SLUDGE RETURN
Literature describes several strategies proposed to optimize clarifier performance. They
and their affects on performance are discussed below.
Strategies
The common design strategy in the U.S. is to provide 25 percent to 100 percent RAS
capacity for conventional and 25 percent or so to 150 percent for extended aeration
activated sludge plants (307). This author is unaware of any plants provided with such
capacity that have reported a need for more.
McKinney (207) has recommended the use of a constant recycle rate as a clarifier
control strategy. This strategy allows the sludge blanket to rise and fall in accordance
with transient flows. His logic is based on accommodating diurnal peaks and the
resulting changes in solids flux. According to McKinney, in the morning hours when
flows are low, constant pumping moves solids from the clarifier into the aeration basin.
The aeration basin has much more volume than the clarifiers so the MLSS concentration
does not increase by an undue amount. When the morning and mid-day peaks arrive, they
enter the aeration basin displacing a large flow of mixed liquor from the basin into the
clarifiers. This increase of flux stresses the clarifiers but they have been previously
emptied of settled solids and have maximum blanket storage capacity. Tank depth must
be adequate for storing this peak flux. When the peak flow subsides, the constant RAS
again exceeds the incoming solids flux and the blanket level drops. The cycle is repeated
on a daily basis. In this manner, the blanket level rises and falls to accommodate the
peaks and the aeration tank is provided with maximum early morning solids to keep
relatively low F/M values when the organic loading occurs. No data were presented by
McKinney (207) to compare this strategy to others but this author is aware that
McKinney does extensive consulting work and has found the above strategy to be
successful in many cases.
Maintaining instantaneous F/M control in the aeration basins requires the sludge return
to be regulated. Three F/M strategies have been described by Keinath, et al (40, 168),
one of which involves sludge return control. In this strategy, the clarifier is used to
store solids during low organic loadings. Conversely, during high organic loadings solids
-164-
-------
are transferred to the aeration tank by an increase in recycle flow. Based on pilot
studies, they concluded that improved organics removal did not justify the decrease in
solids removal resulting from hydraulic transients imposed by near constant F/M control.
That is, the extra hydraulic load placed on the clarifier by increasing the solids under
aeration (via higher HAS) when organic and hydraulic loads peaked, was more detri-
mental to clarifier performance than warranted by improved BOD removals. For this
same reason, Chapman (48, 49) also opposes the use of near-constant F/M strategies.
Extending the settling flux approach developed by Dick for thickener design, Keinath
(163, 166, 171) introduced the "state point" concept (defined in Section 4.1) for secondary
clarifier operation. The state point can be used to analyze clarifier response to changes
in recycle and overflow rates. In a series of papers, Keinath (164, 165) and Keinath and
Laquidara (170) describe solids inventory control strategies using the state point concept.
A recycle rate control strategy was recommended for thickening overloads and a certain
class of thickening and clarification overloads (when the state point has an ordinate
value less than the maximum on the batch flux curve).
Dixon (89) has recently questioned the validity of Keinath's state point analysis. Dixon
presents an extensive review of the literature and bases his arguments on theories that
the original work of Hazen (133) and Fitch (107) are basically correct and were
incorrectly modified by Laquidara and Keinath (184) and Dick (81). No new data are
presented. He points out that the operator may adjust the wasting rate in addition to the
return rate to help in controlling clarifier performance. Of course, downstream solids
handling capacity must be able to handle the changes in such wasting. Dixon presents a
number of practical considerations that help explain why batch column settling and
thickening data differ from that of full-scale tanks. Specifically, he discusses the effect
of tunneling which represents non-uniform removal of solids across the tank bottom due
to the location of the sump or the narrow band occupied by one or two hydraulic suction
arms. He states that a goal of operation should be to produce a solids underflow
concentration as high as possible consistent with satisfactory overflow clarity.
SVI is often used by plant operators to control the activated sludge process. Dick (81,
86) has questioned the use of SVI to estimate return sludge concentration (which is used
to determine recycle rate). Pipes (237, 238) found that using SVI for sludge return
calculations usually predicts solids concentrations on the low side, the result being higher
-165-
-------
recycle rates than necessary. However, he contends that from a practical standpoint the
SVI is still a better approach for an operation than the use of solids flux methods. White
(313) proposes the use of SSVI in which the SVI test is modified by first adjusting the
initial MLSS concentration to a standard value. It is this author's opinion that the SSVI
test would be a definite improvement when comparing different plants or widely
different MLSS levels at the same plant.
Another strategy that the author of this report has observed is the adjustment and
optimization of air flow rates in the channels conducting mixed liquors to the final
clarifiers. Plants equipped with mechanical aeration often tear apart floe particles and
hinder settleability (229). If the mixed liquor is transported from the aeration tanks to
the clarifiers by aerated channels, optimizing the rate of aeration can induce floe
formation and improve settleability before it reaches the clarifier. This author has
observed such improvement in simple beaker tests but has not prepared or seen published
data to quantify the improvements.
The Rated Capacity concept of Wilson (314) has been used as a strategic method of
making adjustments to the return rate and, more recently, to the aeration basin feed
point (29). This technique is discussed above in Section 4.1 of this report.
Effects on Clarifier Performance
In pilot scale studies of a hopper-bottom circular tank, Chapman (47) found increases in
clarifier feed flow to degrade effluent quality. He recommended the use of constant
sludge return rates. To minimize the effects of transient flows on turbulence levels in
the secondary clarifier, he further recommended the use of flow equalization for smaller
plants (48, 49). Others (250, 258) have also shown that increased recycle flow degrades
effluent quality but may keep the blanket level below the weirs and thereby prevent a
serious process upset.
Tuntoolavest, et al (294) studied the effects of sludge recycle ratio using a laboratory
scale pilot plant. They observed that turbulence produced by sludge recycle induced floe
formation at low return rates (30 percent); however, higher return rates (55 to 80
percent) seemed to induce floe breakup. This implies that operators should experiment
with different return rates to determine an optimal range for effluent clarity. The range
-166-
-------
may vary with MLSS and settleability. The data resulting from their study suggested
that the effect of sludge recycle on effluent suspended solids was also dependent on the
air flow rate in the aeration basin. At a high air flow rate (6.8 cu m/h or 240 cu ft/h),
they found recycle ratio to be directly proportioned to effluent suspended solids. At a
low air flow rate (4.5 cu m/h or 160 cu ft/h) an inverse relationship was found to exist.
At an intermediate air flow rate (5.7 cu m/h or 200 cu ft/h) effluent quality was found to
be independent of sludge return. Tuntoolavest, et al suggested that these findings explain
why Villier (301) found a direct relationship between effluent TSS and recycle rate,
whereas Pflanz (234) saw none. In contrast, Chapman (47) did not find air flow rate to be
a significant parameter in predicting effluent quality.
SLUDGE BLANKET LEVEL
Both positive and negative aspects of maintaining a sludge blanket have been discussed in
the literature. The trade-offs between these aspects, however, have not been well
defined. A good qualitative description of the mechanisms and probable effects of
different blanket levels is given by Parker (229).
The traditional argument for maintaining a high sludge blanket is to provide for solids
flocculation and entrapment. In a pilot scale study using a calcium carbonate suspension,
Dick and Johnson (85) found effluent quality to be better when the sludge blanket was
maintained below the influent feed level. The highest effluent suspended solids
concentration was obtained when the blanket was maintained at the feed level. Data
from the study are presented in Figure 4.6-1.
Based on full scale tests, Crosby (70) concluded that either high or low sludge blanket
levels are acceptable under certain conditions; blanket depths near the bottom of the
center feedwell should be avoided. If sludge settleability is poor, flow rates are steady,
and the "half-of-tank-depth" feedwell configuration is available, a high blanket should be
maintained. With well-settling sludge, Crosby recommends use of a lower blanket height.
In another study (84), Dick and Javahari found no relationship between thickening per-
formance and blanket depth. Mazurczyk and Smith (203) found no relationship between
clarification performance and blanket depth. The argument against providing a high
sludge blanket is that the clarifier becomes susceptible to solids carryover during
-167-
-------
0 20 40 60 80 100 120 140 160 180 200
OVERFLOW DISCHARGE RATE, cm3/MIN.
A BLANKET AT FEED ELEVATION, GA = CONSTANT = 251.97 kg/d-m2
~ BLANKET BELOW FEED ELEVATION, GA = CONSTANT « 232.26 kg/d-m2
O BLANKET ABOVE FEED ELEVATION, GA = CONSTANT = 232.65 kg/d-m2
Figure 4-6-1. Influence of blanket position on clarification performance.
-168-
-------
transient flows (214, 229, 258). It has also been suggested that poor performance at high
sludge blanket levels is associated with the increase in clarifier dead space and a reduced
volume for true settling (235). Anderson (5) actually conducted velocity measurements
of the density currents and return currents in full-scale tanks at high and low blanket
levels. A strong correlation of increased velocities with increased blanket thickness was
observed. Modifying the empirical relationship proposed by Pflanz (234), Ghobrial (115)
included a term Vs/V describing the sludge blanket height and settling characteristics to
predict effluent TSS. The proposed equation is of the form:
xc.k(QimixJMSL
where: Xe = effluent TSS concentration
K = proportionality coefficient
Qi/A= influent flow rate to clarifier per unit area
X = MLSS concentration
Vs = accumulated sludge volume
V = clarifier volume
He contends that the measurement and correlation of the Vs/V ratio is a useful
substitute for such settling tests as the SVI and can be used to assist in process control.
The South-Pest treatment plant in the vicinity of Budapest, Hungary, was reportedly
operated in this manner. He further proposes that sludge blanket levels can be
electronically sensed and used to create a feedback signal to trigger automatic or
manual process control.
The author uses data published by EPA to verify his model and obtained excellent
correlation. He also conducted laboratory scale tests to relate blanket levels and loading
to effluent TSS and underflow TSS concentrations. He pointed out that the ratio is also
affected by MLSS concentration changes and recycle flow rate changes. He defines
(Vs/V)cr as the critical sludge blanket level and found it to vary from one tank geometry
to another. Although not discussed in his paper, the author also modified Pflanz'
equation by including recycle flow.
-169-
-------
SLUDGE INVENTORY MANAGEMENT
To lower the solids flux to the secondary clarifiers, some have used step feed control in
the aeration basins (29, 164, 165, 16?! 170). Step feed is a modification to the activated
sludge process by which aeration tank inflow can be fed at different locations. As
discussed by Buhr, et al (29), Figure 4.6-2 shows a typical step feed configuration. The
feed flow may be fed to one of the four passes; all sludge return flow is added at the
head of the tank.
The solids load in the aeration tank varies with the point of feed. Figure 4.6-3 uses
examples to show the variation. Example A shows the expected solids concentration
when the influent is fed to the first pass. Examples B, C, and D illustrate steady-state
conditions when the feed point is moved to other locations. Example D approximates the
contact-stabilization process. By maintaining a majority of the solids load at the inlet
end of the tank, the clarifier solids loading can be reduced.
The effectiveness of this concept was illustrated by Buhr, et al (29) by the curves in
Figure 4.6-4. By shifting the feed point from the first stage to the second in the
aeration basin train at a constant 30 percent RAS flow ratio, the MLSS leaving the
fourth stage and entering the clarifier was reduced by over 40 percent. The RAS
concentration of TSS dropped from about 4.3 to 2.5 times the average MLSS
concentration in all four aeration tanks. In effect, the concept permits a strategic
shifting in solids inventory by reducing flux to the clarifier at the expense of storing
more in the first stage of the aeration basins. The changes occur quite rapidly with
near-equilibrium conditions being established in about 50 percent of the theoretical
aeration basin detention time.
State point concepts have also been used by Keinath and co-authors (164, 165, 167, 170)
to analyze the step feed strategy. They recommend step feed control for heavily
overloaded conditions with respect to both thickening and clarification.
Garrett, et al (114) conducted full-scale operational strategy variation tests at Houston's
Southwest Wastewater Treatment Plant. They confirmed the concept of improving
performance during high hydraulic loading, storm flow periods by conversion to sludge
reaeration, a form of step-feed aeration. They also found that a reduction in the RAS
-170-
-------
WASTE
PASS I
FINAL
EFFLUENT
pass z
SECONDARY
CLARIFIER
PASS 3
Q*Q.
Figure 4.6-2. Typical configuration of step feed activated sludge plant
(Eef 29).
-171-
-------
Q./Q > 0.30
INFLUENT
(a )
4330
AVQ. ML88 • 1000
1000
1000
1000
RETURN 8LUDQE
C b )
(c)
2380
AVQ. ML38 > 1000
1620
AVQ. ML88 • 1000
1620
660
360
1620
660
660
380
(
1240 AVQ. ML83 ¦ 1000
Variation of sludge distribution with feed point, in
four-pass step feed plant (Ref. 29).
Figure 4.6-3.
280
1240
1240
1240
-172-
-------
n
1
2
a
A
Xn
X
\
\
•RETUF
SLUDG
EN
iE
c
u
a
c
\
i
»
>
2
1"
X
o
&
e
u
b
ti
3
" j
II /
L /
/
•
PASS 1
/
j
%A
Figure 4.6-4.
Variation of mixed liquor and return sludge concentrations
with time, following a change in feed point from Pass 1 to
Pass 2 with Qr/Q = 0.30 {Ref. 29).
-173-
-------
rate from 100 percent to 50 percent of the average influent flow rate reduced the
effluent TSS from 10.2 tng/1 to 6.9 tog/1.
SUMMARY
On the basis of materials presented in the literature review and discussions at the
symposium, the participants jointly listed the following questions to be addressed in
research of activated sludge clarifier operations:
1. What is the cost effectiveness of chemical addition to clarifiers?
a. Continuously
b. Intermittently
2. What impact does the level and type of aeration have on clarifier performance?
3. What is the affect of flow transients on clarifier performance?
4. What is the impact of various strategies of sludge blanket level and/or RAS
control?
5. What instrumentation is needed?
6. What is the effect of mixed liquor settleability on effluent TSS?
?. How do we define and measure sludge blankets?
8. How do we best measure settleability?
9. How do we best use settleability results to optimize operation?
10. How can step feed strategies be optimized?
11. How much RAS capacity is needed for various modes?
-174-
-------
SECTION 5
REFERENCES
1. Aibrecht, A. E., R. E. Wullschleger and W. J. Katz. In Situ Measurement of
Solids in Final Clarifiers. J. Sanit. Eng. Div., ASCE, (SA):183-198, 1966.
2. Anderson, D.T., and M.J. Hammer. Effects of Alum Addition on Activated
Sludge Biota. Water and Sewage Works, 12Q(l}:63-67, 1973.
3. Anderson, H.M., and R.V. Edwards. A Finite Differencing Scheme for The
Dynamic Simulation of Continuous Sedimentation. AIChE Symposium Series,
77(209):227-238, 1981.
4. Anderson, J.A. Thickening Performance of Activated-Sludge Settlement Tanks.
Water Pollution Control (U.K.), 80(4):521-528, 1981.
5. Anderson, N.E. Design of Final Settling Tanks for Activated Sludge. Sewerage
Works J., 17(1): 50, 1945.
6. Attir, U., and M.M. Denn. Dynamics and Control of the Activated Sludge
Wastewater Process. AIChE J., 24(4):693-698, 1978.
7. Baker, K. D. and G. D. Reed. Evaluation of Mechanical Flocculation of
Activated Sludge for Improvement of Secondary Clarification. An unpublished
paper presented at the 57th Annual Conference of the Water Pollution Control
Federation, New Orleans, Louisiana. 1984, 9 pp.
8. Banks, C.J., M. Davies, I. Walker, and R.D. Ward. Biological and Physical
Characterization of Activated Sludge: Comparative Experimental Study at Ten
Treatment Plants. Water Pollution Control (U.K.), 75(4):492-508, 1976.
9. Banks, D.H. Upward-flow Clarifier for Use in Treating Sewage Effluents.
Surveyor (GB), No. 3745, 21, March 14, 1964.
10. Bargman, R.D., and J. Bargerding. Characterization of the Activated Sludge
Process. National Technical Information Serivce, 63 pp., 1973.
11. Barr, D.LH. Density Currents and Public Health Engineering. Water and Water
Engineering. 508, December 1966.
12. Barran, H.A. Field Observations of Density Current Phenomena in Rectangular
Final Settling Tanks. An unpublished report prepared by Gore & Storrie Ltd.,
Cons. Engrs. for Mississauga, Ontario, Canada, 1981, 7 pp.
-175-
-------
13. Barth, E.F. Device to Aid Pilot-Plant Final Settlement. Environmental Science
and Technology, 2:139, 1968.
14. Bender, J. and R. M. Crosby. Hydraulic Characteristics o£ Activated Sludge
Secondary Clarifiers. Project Summary Report, U.S. Environmental Protection
Agency (No. EPA-600/S2-84-131), 1984.
15. Benedek, A., J.J. Bancsi, and J.W.G. Rupke. Use of Polyelectrolytes for Over-
loaded Clarifiers. J. Water Pollut. Control Fed., 47(10):2447-2460, 1975.
16. Bisogni, J.J., and A.W. Lawrence. Relationships Between Biological Solids
Retention Time and Settling Characteristics of Activated Sludge. Water
Research (U.K.), 5:753-763, 1970.
17. Bohlier, Paul V. Discussion of Making Full Use of Step Feed Capability by H.Q.
Buhr, M.F. Goddard, T.E. Wilson, and V/.A. Ambrose. J. Water PoUut. Control
Fed., 57(9):973, 1985.
18. Boyle, W.H. Mass Loading Made Clear. Water and Wastes Eng., 11(7):49-5Q,
1974.
19- Boyle, W.H. Don't Forget Side Water Depth. Water and Wastes Eng., 12(6):32-
33, 1975.
20. Boyle, W. H. Peripheral Clarifier. Unpublished paper presented at the 1982
Water Pollution Control Association Annual Conference, Springfield,
Massachusetts, 1982, 19 pp.
21. Boyle, W.H. Ensuring Clarity and Accuracy in Torque Determinations. Water
and Sewage Works, 125(3);76-77, 1978.
22. Boyle, W.H. Get the Most From the Final Clarifiers. Water and Wastes Eng.,
12{10):53-55, 1975.
23. Boyle, W.H. Improved Secondary Clarifier Performance: Peripheral vs Center
Effluent Arrangement. Water and Sewage Works, 123{10):9Q-94, 1976.
24. Boyle, W.H. Ranking Sludge Removal Methods. Water and Sewage Works,
127{6):70,72, 1980.
25. Boyle, W.H. Peripheral Influent Peripheral Effluent Final Clarifier. Unpublished
paper presented at Water Pollution Control Federation Conference, Miami
Beach, Florida, 1975, 7 pp.
26. Boyle, W.H. Sludge Blanket in Activated Final Clarifiers, Water/Engineering and
Management, (H13)R-124, R-130, 1981.
27. Boyle, W.H. Final Clarifier Sludge Removal Hydraulic Verification. Unpublished
paper presented at the ASCE National Conference of Environmental Engineering,
Boulder, Colorado, 1983, 6 pp.
28. Bradley, R.A., and R.B. Krone. Shearing Effects on Settling of Activated Sludge.
J. Sanit. Eng. Div., ASCE, 97(SAL):59-79, 1971.
-176-
-------
29. Buhr, H.O., M.F. Goddard, T.E. Wilson and W.A. Ambrose. Making Full Use of
Step Feed Capability. J. Water Pollution Control Fed., 56(4):325-330, 1984.
30. Buhr, H.O., M.F. Goddard, T.E. Wilson and W.A. Ambrose. Response to
discussion by W.L. Jones, and E.D. Schroeder of Making Full Use of Step Feed
Capability, J. Water Pollut. Control Fed., 57(9):975-976, 1985.
31. Buhr, H.O., M.F. Goddard, T.E. Wilson and W.A. Ambrose. Response to
discussion by Paul V. Bohller of Making Full Use of Step Feed Capability, J.
Water Pollut. Control Fed., 57(91:973-974, 1985.
32. Burde, C.S., and P.N. Cheremisinoff. Sewage Sludge Concentration: Removal
Equipment and Methods. Water and Sewage Works, November 1974.
33. Busby, P.R.A. Problems in the Operation of Final Sedimentation Tanks of
Activated-Sludge Plants in the Greater London Area. Progress in Water Tech-
nology (U.K), 8(6):461-464, 1977.
34. Busby, J.B., and J.F. Andrews. Dynamic Modeling and Control Strategies for the
Activated Sludge Process. J. Water Pollut. Control Fed., 4?(5):1975.
35. Busch, A.W., and R.L. Irvine. Solids Separation in Activated Sludge System
Design and Operation. J. Water Pollut. Control Fed., 52(4):804, 1980.
36. Camp, T.R. A Study of the Rational Design of Settling Tanks. Sewage Works J.
8(5):742, 1936.
37. Camp, T.R. Sedimentation and the Design of Settling Tanks. ASCE Trans-
actions. Paper No. 2285, pp. 895-958, 1945.
38. Camp, T.R. Studies of Sedimentation Basin Design. Sewage mad Industrial
Wastes, 25(1):1, 1953.
39. Cashion, B.S., and T.M. Keinath. Influence of Three Factors on Clarification in
the Activated Sludge Process. J. Water Pollut. Control Fed., 55(10):1331-1337,
1983.
40. Cashion, B.S., T.M. Keinath and W.W. Schuk, Control Strategies for the Acti-
vated Sludge Process. Unpublished paper presented at the 50th Annual Con-
ference of the Water Pollution Control Federation, 1977, 27 pp.
41. Cerwick, J.A. Innovative Treatment Scheme Offers Cost Advantages and
Operating Simplicity. A paper presented at the 1981 Water Pollution Control
Federation Conference, Detroit, Michigan, 1981, 17 pp.
42. Chain Belt Company. New Style Tow-Bro Header (Unitube Tests, Ann Arbor
Sewage Treatment Plant, Ann Arbor, Michigan), Research and Development,
Project No. 3W-54 (J-60558), Rpt. No. 1, 1956, 11 pp.
43. Chain Belt Company. Extract from Nineteenth Annual Report of the Sewerage
Commission of the City of Milwaukee for the year 1932.
44. Chao, A.C., and T.M. Keinath. Influence of Process Loading Intensity on Sludge
Clarification and Thickening Characteristics. Water Research, 13:1213, 1979.
-177-
-------
45. Chao, J.L., and E.E. Trussell. Hydraulic Design of Flow Distribution Channels.
3. Environmental Eng. Div., ASCE, 106(EE2):321-334, 1980.
46. Chapman, D.T., et al. Effect of Dissolved Oxygen in Activated Sludge Systems.
J. Water Follut. Contol Fed., 48:2486, 1976.
47. Chapman, D.T. The Influence of Process Variables on Secondary Clarification.
J. Water Pollut. Control Fed., 55{11):1425-1434, 1983.
48. Chapman, D.T. Final Settler Performance During Transient Loading.
Unpublished paper presented at the 57th Annual Conference of the Water
Pollution Control Federation. New Orleans, Louisiana, 1984, pp. 227-234.
49. Chapman, D.T. Final Settler Performance During Transient Loading. J. Water
PoUut. Control Fed., 57(3):227-234, 1985.
50. Chiu, Y. Boundary Layer Separation Concept of Settling Tanks. J. Environ-
mental Eng. Div., ASCE, 100(EE6):1213, 1974.
51. Christoper, S., and R. Titus. New Wastewater Process Cuts Plans Costs 50
Percent. Civil Engineering (ASCE), 53(5}:39~42, 1983.
52. Clark, J.W., W. Viessman, Jr., and M.J. Hammer. Water Supply and Pollution
Control, Third Edition, New York, Harper and Row Publishers, 1977, p. 578.
53. Coe, H.S., and G.H. Clevenger. Methods for Determining the Capacities of
Slime Settling Tanks. Transactions, American Society of Mining Engineers, 55,
356-384, 1916.
54. Collins, M.A. and R.M. Crosby. Impact of Flow Variation on Secondary Clarifier
Performance. Unpublished paper presented at the Water Pollution Control
Federation Conference, Las Vegas, Nevada, 1980, 28 pp.
55. Comings, E.W., C.E. Pruss, and C. DeBord. Continuous Settling and Thickening.
Industrial and Engineering Chemistry, 46(6):1164, 1954.
56. Coneway, C.R. and J.V. Matson. Dynamics and Control of Suspended Solids in a
Two-Step Activated Sludge Plant. AIChE Symposium Series, 74(178):75-83, 1978.
57. Conway, Richard A. and Victor H. Edwards. How to Design Sedimentation
Systems from Laboratory Data. Chemical Engineering, 167-170, 1961.
58. Coppock, E.R. UL Evaluation of Settling Basin Performance. J. of the New
England Water Works Association, 91(2): 143-164, 1977.
59. Cordoba-Molina, J.F., R.R. Hudgens, and P.L. Silveston. Settling in Continuous
Sedimentation Tank. J. Environmental Eng. Div., ASCE, (EE6):1263, 1978.
60. Cordoba-Molina, J.F., R.R. Hudgins, and P.L. Silveston. The Gravity Clarifier as
a Stratified Flow Phenomenon. Can. J. Chem. Eng., 57(6); 249, 1979.
61. Crabtree, H.E. Some Observations on Denitrification in Activated Sludge Final
Settlement Tanks. Water Pollution Control (U.K.), 82(3):315—329, 1983.
-178-
-------
62. Cranston, J.A. Circular Radial-Flow Sedimentation - A New Design Proposal.
Effluent and Water Treatment J. (U.K.), 5: 412-415, 1965.
63. Creagh, J.P., G.W. Kumke, and R.A. Conway. Performance of Internally
Clarified Activated-Sludge Process Treating Combined Petrochemical-Municipal
Waste. In: Proc. 23rd lad. Waste Conf., Purdue University, West Lafayette,
Indiana, 1968, pp. 567-582.
64. Crits, G.J. A Packaged Module Clarifier Features Floe Centrifugal Separation,
Tube Settling and Filtration. In: Proc. Water Industry *81; International
Conference, Brighton, U.K. 1981, 11 pp.
65. Crosby, R.M. Understanding Secondary Clarifier Behavior. Unpublished paper to
be presented at the 1985 National Conferene of ASCE, 1985, 6 pp.
66. Crosby, R.M. The Flow Pattern Test and Other Tricks. Unpublished paper. 1984,
5 pp.
67. Crosby, R.M. and J.H. Bender. Hydraulic Considerations that Affect Secondary
Clarifier Performance. U.S. Environmental Protection Agency Technology
Transfer Report, 1980, 15 pp.
68. Crosby, R.M. Clarifier Newsletter. Unpublished newsletter by Crosby, Young
and Assoc., Piano, Texas, 1983, 2 pp.
69. Crosby, R.M. Making Clarifiers Work Better: The "Bubble Gum" Approach.
Southwest and Texas Water Works Journal, 1981, 4 pp.
70. Crosby, R.M.. Evaluation of the Hydraulic Characteristics of Activated Sludge
Secondary Clarifiers. U.S. Environmental Protection Agency Project Report.
Contract. No. 68-03-2782, 1984, 230 pp.
71. Crosby, Young and Assoc., tic. Secondary Clarifier Hydraulic Analysis Morris
Forman Water Treatment Plant. A report prepared for Camp, Dresser and
McKee, Inc., 1982, 57 pp.
72. Crowther, J.M., J.F. Dalrymple, T. Woodhead, P. Coackley, and LM. Hamilton.
A Correlation Method for the Estimation of Retention Times at Full-Scale
Sewage Treatment Plants. Water Research, 14:567, 1980.
73. Culp, G.L., K. Hsiung, and W.R. Conley. Tube Clarification Process, Operating
Experiences. JSED, ASCE, 95, SA5, 829, October 1969.
74. Dague, R.R. Hydraulics of Circular Settling Tanks Determined by Model-
Prototype Comparisons. Master's Thesis, Iowa State University, Ames, Iowa,
1960.
75. Dalgger, G.T. and R.E. Roper, Jr. The Relationship between SVI and Activated
Sludge Settling Characteristics. J. Water Pollut. Control Fed., 57(8):8 59-866,
1985.
-179-
-------
76. Daugherty, James L. and Robert A. Davis. Suspended Solids Removal of a
Clarifier and Sand Filter - Viewed as a System. An unpublished paper presented
at the 55th Annual Water Pollution Control Federation Conference, St. Louis,
Missouri, 1982, 11 pp.
77. Davis, C. and R.F. Wills. Flow Patterns in a Rectangular Sewage Sedimentation
Tank. JWPCF, 34(3};241-242, 1962.
78. Davis, C. Aspects of the Performance of Final Clarifiers in the Activated
Sludge Process. Water Science and Technology (South Africa), 14(9-11): 151-4—
1518, 1982.
79. Di Bernardo, L. and M.F. Giorgetti. The Use of a Perforated Baffle at the Inlet
of Rectangular Settling Basins. J. American Water Works Association, 72(9) :528-
533, 1980.
80. Dick, R.I. Analysis of the Performance of Final Settling Tanks. Tribune du
Cebedeau, 33(441-442):359-367, 1980.
81. Dick, R.I. Folklore In the Design of Final Settling Tanks. J. Water Pollut.
Control Fed., 48(4):G33-644, 1976.
82. Dick, R.I. Role of Activated Sludge Final Settling Tanks. J. Sanit. Eng. Div.,
ASCE, 96(SA2):423-436, 1970.
83. Dick, R.L, and B.B. Ewing. Evaluation of Activated Sludge Thickening Theories.
J. Sanit. Eng. Div., ASCE, 93(SA4):9, 1967.
84. Dick, R.I., and A.R. Javaheri. Continuous Thickening of Non-Ideal Suspensions.
PB-209 S31, National Technical Information Service, Springfield, Virginia, 1972,
74 pp.
85. Dick, R.I., and J.H. Johnson, Jr. Cause of Inefficient Solids Suspension in the
Activated Sludge Process. PB83-108381, National Technical Information Serv-
ice, Springfield, Virginia, 1982, 32pp.
86. Dick, R.I., and P.A. Vesilind. The Sludge Volume Index - mat Is It? J. Water
Pollut. Control Fed., 41(7):1285, 1969.
87. Dick, R.I. and K.W. Young. Analysis of Thickening Performance of Final Settling
Tanks. A paper presented at the 27th Annual Purdue Industrial Waste
Conference, Purdue University, Lafayette, Indiana, 1972, 34 pp.
88. Ditsios, M. Relationship between Surface Loading, Depth and Effluent
Suspended Solids at a Rectangular, Horizontal Pilot-Scale Secondary Clarifier.
Water Science and Technology (Japan), 16:293-302, 1984.
89. Dixon, D.C. Capacity and Control of Clarifiers and Thickeners. J. Water Pollut.
Control Fed., 57(l):46-52f 1985.
90. Dobbins, W.E. Effect of Turbulence on Sedimentation. Transactions, ASCE,
109:629, 1944.
-180-
-------
91. Dodson, Gary. S. Secondary Sedimentation Tank Configuration Design
Considerations. An unpublished paper presented at the 48th Annual Conference
of the California Water Pollution Control Association, South Lake Tahoe,
California, 1978, 10 pp.
92. Doedens, H. The Optimal Content of Dry Solids of Sludge in Activated-Sludge
Plants. Wasserwirtschaft, Stuttg (West Germany), 59:1-4, 1969.
93. Eckenfelder, Jr., WW., and Melbinger, N. Settling and Compaction Character-
istics of Biological Sludges. Sewage and Industrial Wastes, 29(10):1114, 1957.
94. Effluent and Water Treatment J. A New Activated Sludge Treatment System.
18(7): 347, 349, 1978.
95. Ekama, G.A., and M. Smollen. Comparison of Empirical Settling-Velocity
Equations in Flux Theory for Secondary Settling Tanks. Water (South Africa),
10(4):175-184, 1984.
96. El-Baroudi, H.M. Characterization of Settling Tanks by Eddy Diffusion.
J. Sanit. Eng. Div., ASCE, SA3.-527, 1969.
97. El-Baroudi, H.M., and D.R. Fuller. Tracer Dispersion of High Rate Settling
Tanks. J. Environmental Eng. Div., ASCE, 99(3):347, 1973.
98. Envirex, Inc. Rex Rim-Flo Clarifers. Bulletin 315-131 (1982).
99. Erickson, L.E., J.C. Heydweiller, and L.T. Fan. Consideration of Suspended
Solids Transport in Activted Sludge System Synthesis: L Simultaneous Use of
Clarification and Sludge Thickening Models. AIChE Symposium Series,
73(167):1Q5-111, 1977.
100. Esler, J.K. Optimizing Clarifier Performance. Unpublished paper presented at
the 57th Water Pollution Control Federation Conference, New Orleans,
Louisiana, 1984, 39 pp.
101. Esler, J.K. and T.J. Miller. Full-Scale Clarifier Modifications at the Oak
Orchard Wastewater Treatment Plant. Unpublished paper presented at the 58th
Annual Water Pollution Control Federation Conference, Kansas City, Missouri,
1985, 29pp.
102. Esler, J.K. and T.J. Miller. Full-Scale Clarifier Modifications in New York
State. Unpublished paper presented at the Annual Winter Meeting, New England
Water Pollution Control Association, Boston, Mass., January 8, 1986.
103. FMC - Link Belt. Clarifiers for Clarification Solids Removal. Bulletin 19000
(1976).
104. Fan, L.T., C.R. Lee, T. Takamatsu. Optimization of Multi-State Secondary
Clarifier. 3. Water Pollut. Control Fed., 48(ll);2578-2589, 1976.
105. Feeley, G.R., and S. Krakauer. Clarifier Forms and Removes Sludge
Hydraulically. Water and Wastes Eng., 5(12);39-41, 1968.
-181-
-------
106. Finger, R.E. Solids Control in Activated Sludge Plants with Alum. J. Water
Pollut. Control Fed., 45:1654-1662, 1973.
107. Fitch, E.B. Flow Path Effect on Sedimentation. Sewage and Industrial Wastes,
28(l):l-9, 1956.
108. Fitch, E.B. The Significance of Detention on Sedimentation. Sewage and
Industrial Wastes, 29(10):1123, 1957.
109. Fitch, E.B. and W.A. Lutz. Feedwells for Density Stabilization. J. Water Pollut.
Control Fed., 32(2):C147-156, I960.
110. Flanagan, M.J. Control Strategies for the Activated Sludge Process. Instru-
mentation Technology, 24(7):35-43, 1977.
111. Ford, D.L., and W.W. Eckenfelder, Jr. Effect of Process Variables on Sludge
Floe Formation and Settling Characteristics. Jr. Water Pollut. Control Fed.,
39il850-1859, 1967.
112. Forster, C.F., and J. Dallas-Newton. Activated Sludge Settlement - Some
Suppositions and Suggestions. Water Pollution Control (U.K.), 79(3):338-351,
1980.
113. Fujiinoto, E., and K. Goto. Evaluation of the Resident Mass of Sludge in Final
Sedimentation Tanks for Waste Sludge Control. Water Science and Technology
(Japan), 13(9): F153-158, 1981.
114. Garrett, M.T., Jr., Josten Ma, William Yang, Gurdip Hyare, Teresa Norman, and
Zaki Ahmad. Improving the Performance of Houston's Southwest Wastewater
Treatment Plant, USA Water Science and Technology (Vienna), 16:317-329, 1984.
115. Ghobrial, F.H. Importance of the Clarification Phase in Biological Process
Control. Water Research, 12(ll):1009-10l6, 1978.
116. Giles, J.H.L. Inlet and Outlet Design for Sedimentation Tan Its. Sewage Works J.
15(4):609, 1943.
117. Girling, R.M. Experiences with High Rate Settlers Applied to an Existing Final
Clarifier, Winnipeg, Canada. Progress in Water Technology, 8(6):521-53G, 1977.
118. Gould, R.H. Final Settling Tanks of Novel Design. Water Works and Sewerage,
90:133-136, 1943.
119. Gould, R.H. Continuously Operating Settling and Clarifier Tank with Rotating
Feed Arrangement. U.S. Patent Office, No. 3,036,715.
120. Graber, S.D. Outlet Weir Loadings for Settling Tanks. J. Water Pollut. Control
Fed., 46(10):2355-2362, 1974.
121. Grady, C.P.L., and M. Tuntoolavest. Effect of Activated Sludge Operational
Conditions on Sludge Thickening Characteristics. J. Water Pollut. Control Fed.,
(54(7):1112-1117, 1982.
-182-
-------
122. Great Lakes-Upper Mississippi River Board of State Sanitary Engineers. Recom-
mended Standards for Sewage Works, 1978 Edition.
123. Guo, P.H.M., D. Thirumurthi, and B.E. Jank. Evaluation of Extended Aeration
Activated Sludge Package Plants. J. Water Pollut. Control Fed., 53(l):33-42,
1981.
124. Hale, F.D. and S.R. Garver. Viscous Bulking of Activated Sludge. A paper
presented at the 56th Annual Water Pollution Control Federation Conference,
Atlanta, Georgia, 1983, 8 pp.
125. Hamlin, M.J., and A.H. Abdul Wahab. Settling Characteristics of Sewage in
Density Currents. Water Research, 4:609-626, 1970.
126. Hammam, S. and C.A. Sastry. Design of Settling Tanks with Special Reference
to Inlet Arrangements. Enviromental Health (India), 8: 46-49, 1966.
127. Hanbury, M.J., A.J. Rachwal, D.J. Critchard, and D.W.M. Johnstone. Design and
Operation of Final Settlement Tanks: Use of Stirred Specific Volume Index and
Mass Flux Theory. Tribune du Cebedeau (U.K.), 33 (443): 411-425, 1980.
128. Hanbury, M.J., A.J. Rachwal, and D.W.M. Johnstone. Settlement Characteristics
and Settlement Tank Performance in the Carrousel Activated-Sludge System.
Water Pollution Control (U.K.), 78(3):337-356, 1979.
129. Handley, J. Sedimentation: An Introduction to Solids Flux Theory. Water
Pollution Control (U.K.), 73(2):230, 1974.
130. Hartley, K.J. Outlet Weirs on Water Treatment Sedimentation Tanks.
Unpublished report for in-house use. Engineering and Water Supply Dept.,
Education Centre, Adelaide, So. Australia, 1980, 18 pp.
131. Hartley, K.J. Ruminations on Wind Effects in Sedimentation Tanks, 1980, 10 pp.
132. Hassett, N.J. Concentration of a Continuous Thickener. Industrial Chemist, 40,
29-33, 1964.
133. Hazen, A. On Sedimentation. Transaction ASCE, 53(45): 1904.
134. Helmer, E.R. Field Report Comparison of Rim-Flos. Unpublished Envirex, Inc.
report, 1969, 3 pp.
135. Hibberd, R.L. and W.F. Jones. The Design and Operation of Activated-Sludge
Final Settling Tanks. Water Pollution Control (U.K.), 78(l):14-32, 1974.
136. Hirsch, A.A. Functional Effluent Baffling - A Key to Settling Basin Efficiency.
Wat. & Sew. Wrks. Ref. No. 1966, R-141, 1966.r
137. Heinke, G.W., Qazi, MA. and A. Tay. Design and Performance Criteria for
Settling Tanks for the Removed of Physical-Chemical Floes - Volume II.
Research Report No. 56, Environment, Canada, Ontario Ministry of the Environ-
ment, 1977, 115 pp.
-183-
-------
138. Hubbell, G.E. Experiments with Inlet Devices for Sedimentation Tanks. Sewage
Works 3. ft(4):774-783, 1934.
139. Hubbeli, G.E. Hydraulic Characteristics of Various Circular Settling Tanks.
American Water Works Association., 30(2):335-353, 1938.
140. Hudgins, R.R., P.L. Silveston, and M.C. Murphy. Suppression of Density Waves in
Clarifiers. Water and Science Technology (Canada), 14(9-11):1510-1513, 1982.
141. Hudson, H.E., Jr. Density Considerations in Sedimentation. J. American Water
Works Association, 64(6):382-386, 1972.
142. Hui, A.M., B.L. Michalczyk, R.J. Tekippe, D.A. Coe, and R.B. Stallings. Pilot
Study Investigation of TF/SC and RBC/SC Processes. A paper presented at the
55th Annual Water Pollution Control Federation Conference in Atlanta, Ga.,
October 4, 1983.
143. Hutton, P.H. An Evaluation of Secondary Clarifier Equations Using an
Optimization Model. Master's Thesis, University of Illinois, 1985.
144. Imam, E., J.A. McCorquodale, and J.K. Bewtra. Numerical Modeling of
Sedimentation Tanks. J. Hydraulic Engineering, ASCE, 109{12):1740-1754, 1983.
145. Ingersoll, A.C., J.E. McKee, and N.H. Brooks. Fundamental Concepts of
Rectangular Settling Tanks. In: Proc. ASCE 81(590); 590-1-590-28, 1955.
146. Isgard, E., C.N.H. Fischerstrom, and I. Larsen. Settling of Activated Sludge in
Horizontal Tanks. J. Sanit. Eng. Div. ASCE, 93(SA3):73-83, 5302, 1967.
147. Jennings, S., and C. Grady, Jr. The Use of Final Clarifier Model in Under-
standing and Anticipating Performance Under Operational Extremes. In: Proc.
27th Ind. Waste Conf., Purdue University, West Lafayette, Indiana, pp. 221-241,
1971.
148. Johnstone, D.W.M., A.J. Rachwal, and M.J. Hanbury. Settlement-Tank
Performance in the Carrousel Activated-Sludge System. Water Pollut. Control,
337-356, 1979.
149. Jones and Henry Engineers, Inc. Dye Tracer Studies on Final Settling Tanks at
Bay View Wastewater Treatment Plant. Engineering report prepared for City of
Toledo, Ohio, 1982, 10 pp.
150. Jones, W.L. and E.D. Schroeder. Discussion of Making Full Use of Step Feed
Capability by H.O. Buhr, M.F. Goddard, T.E. Wilson and W.A. Ambrose. J. Water
Pollut. Control Fed., 57(9):974-975, 1985.
151. Joyce, M., W. Schultz, and A. Strom. Replacement of Activated Sludge
Secondary Clarifiers by Dynamic Straining. EPA-670/2-75-045, U.S. Environ-
mental Protection Agency, Cincinnati, 1975.
152. J. Environmental Eng. Div., ASCE. Final Clarifiers for Activated Sludge Plants.
105(EE5):803-817, 1979.
-184-
-------
153. Kalbskopf, K-H and. H. Herter. Operational Experiences with the Sedimentation
Tanks o£ the Mehanical and Biological Stages of the Emscher Mouth Treatment
Plant. GWF Wasser/Zbwasser, 125(4): 200, 1984.
154. Kalbskop£, K-H. Discussion of the Design Parameter for Secondary
Sedimentation Tanks. Water Research, 6:429-431, 1972.
155. Karr, P.R., R.B. Smith, and M. Masingale. Determine Sludge Age with Clarifier
Solids. Water and Wastes Eng., 17(4):30,35,47, 1980.
156. Katz, William J. In-house memorandum. Rex-Chainbelt Corp., Milwaukee,
Wisconsin. April 3, 1956.
157. Katz, W.J. and A. Geinopolos. A Study of the Solids Retention Characteristics
of a Final Clarifier in the Activated Sludge Process. A paper presented at the
First International Water Pollution Conference, London, England, 1962,
11 pp. 1-11.
158. Katz, W.J. and A. Geinopolos. A Comparative Study of the Hydraulic Character-
istics of Two Types of Circular Solids Separation Basins. Unpublished paper
presented at the Manhattan College Waste Treatment Conference, 1957, 5 pp.
159. Kawamura, Susumu. Hydraulic Scale-Model Simulation of the Sedimentation
Process. J. American Water Works Association, 372-379, 1981.
160. Kawamura, Susumu. Pilot Studies of Mixing and Settling. Unpublished paper
prepared for in-house seminar at James M. Montgomery, Consulting Engineers,
Inc., Pasadena, California, 1981, 23 pp.
161. Keefer, C.E. Relationship of Sludge Density Index to the Activated Sludge
Process. J. Water Pollut. Control Fed., 35:1166-1173, 1963.
162. Keinath, Thomas M. Discussion of Capacity and Control of Clarifiers and
Thickeners by D. Christopher Dixon, J. Water Pollut. Control Fed., 57, 46, 1985.
163. Keinath, T.M. Design and Operational Criteria for Thickening of Biological
Sludges. PB-262 967, National Technical Information Service,
Springfield, Virginia, 1976, 63 pp.
164. Keinath, T.M. Operational Dynamics and Control of Secondary Clarifiers. A
paper presented at the 57th Annual Conference of the Water Pollution Control
Federation, 1985.
165. Keinath, T.M. Solids Inventory Control in the Activated Sludge Process. Water
Science and Technology, 13(10):413-419, 1981.
166. Keinath, T.M., et al. A Unified Approach to the Design and Operation of the
Activated Sludge System. In: Proc. 31st Ihd. Waste Conf., Purdue University,
West Lafayette, Indiana, 1976, 914-939 pp.
167. Keinath, T.M., and B.S. Cashion. Control Strategies for the Activated Sludge
Process. EPA-600/2-80-131, U.S. Environmental Protection Agency Report,
Cincinnati, Ohio, 1980, 173 pp.
-185-
-------
168. Keinath, T.M., B.S. Cashion, and W.W.. Schuk. Control Strategies for the
Activated-Sludge Process. J. Water Pollut. Control Fed.. 51(4):815-833, 1979-
169. Keinath, T.M., and J.D, Dietz. Experimental Evaluation of the Significance of
Overflow Rate and Detention Period. J. Water Pollut. Control Fed., 56(4) :344-
350, 1984.
170. Keinath, T.M., and V.D. Laquidara. Clarification Failure Control Strategies.
J. Environmental Eng. Div.» ASCE, 108(EE6):1121-1127, 1982.
171. Keinath, T.M., M.D. Ryckman, C.H. Dana, D.A. Hofer. Activated Sludge-Unified
System Design and Operation. J. Environmental Eng. Div., ASCE, 103(EE5):829-
849, 1977.
172. Kermode, R.I., and R.W.J. Brett. Process Control of Activated Sludge Treat-
ment. PB-227 238, National Technical Information Service, Springfield, Virginia,
1973, 88 pp.
173. Kleffner, L Two-Storeyed Final Sedimentation Tanks. Water Research, 6:433-
435, 1972.
174. Klein, H, Sedimentation Effect of High-Rate Settling Tanks. Wasserwirtschaft
(West Germany), 73(10):33?-J43, 1983.
175. Klein, J.O. Maintenance of Rectangular Settling Tanks. A paper presented at
the 54th Annual Conference of the Water Pollution Control Federation, Detroit,
Michigan, 1981, 21 pp.
176. Knocke, W.R. Activated Sludge Thickening: Floe Size and Volume Consider-
ations, submitted for publication in the J. Water Pollut. Control Fed., July 1983.
177. Koltsova, S.I., LV. Skirdov, and K.M. Morozova. Methods of Improvement of
Secondary Settling Tanks Operating Efficiency. Industrial Water Engineering,
15(5):24-27, 1978.
178. Konrad, W.N. Studies on a Peripheral-Feed Circular Basin. Unpublished paper
presented at the Florida Water Pollution Control Federation Conference,
Orlando, Florida, 1961, 8 pp.
179. Koopman, B., and K. Cadee. Prediction of Thickening Capacity Using Diluted
Sludge Volume Index. Water Research, 17(10):1427-1431, 1983.
180. Kovalcik, Robert N. Single Waste-Treatment Vessel Both Flocculates and
Clarifies. Chemical Engineering, 117-120, 1978.
181. Kynch, G.J. A Theory of Sedimentation. Transactions of the Faraday Society,
48:166, 1952.
182. La Clair, L.M., J.R. McWhirter, and W.L. Ross. Integral Circular Wastewater
Treatment Process. Union Carbide Corp,. U.S. Patent 3,963,609, 947(3),:1261,
1976.
183. Lakeside Equipment Corporation. Spiraflo Clarifier. Bulletin. 1210 (1979).
-186-
-------
184. Laquidara, V.D., and T.M. Keinath. Mechanism of Clarification Failure.
3. Water PoUut. Control Fed., 55(l):54-57, 1983.
185. Larock, B.E., W.K.C. Chun, and D.R. Schamber. Computation of Sedimentation
Basin Behavior. Water Research, 17(8):86l, 1983.
186. Lauria, D.T., J.B. Uunk, and J.K. Schaefer. Activated Sludge Process Design.
J. Environmental Eng. Div., ASCE, 103 (EE4) :625-645, 1977.
187. Lech, R.F. Application of Control Strategies to the Activated Sludge Process.
Ph.D. Thesis, Purdue University, 1973.
188. Lech, R.F., H.C. Lira, C.P.L. Grady, Jr. and L.B. Koppel. Automatic Control of
the Activated Sludge Process-I, Development of A Simplified Dynamic Model.
Water Research, 12(2):81-90, 1978.
189- Lee, C.R., L.T. Fan, M.C. Kuo, T. Takamatsu. Two-Stage Settling Improves
Sludge Removal Efficiency. Water and Sewage Works, 124(S):41-44, 1977.
190. Lee, S.E., B. Koopman, H. Bode, andD. Jenkins. Evalution of Alternative Sludge
Settleability Indices. Water Research, 17(10):1421-1426, 1983.
191. Lodholz, J.C., and H. Pentz. Equipment Applications in Wastewater Treatment
Clarifiers. Public Works, 107(6);9?-100, 1976.
192. Lutge, T.V. Hydraulic Control Utilizing Submerged Effluent Collectors.
J. Water Pollut. Control Fed., 41(8):1453, 1969.
193. Maeda, M., J. Hirotsuji, K. Maeda, and S. Osada. A Method for the Evaluation of
the Effectivenss of a Quality Control Operation in Activated Sludge Process.
Water Science and Technology, 13(10)5471-476, 1981.
194. Magara, Yasumoto, Shiochi Nambu, and Kohken Utosawa. Biochemical and
Physical Properties of an Activated Sludge on Settling Characteristics. Water
Research, 10(1), 1976.
195. Mahmud, Z., and Y.C. Wu. Performance of Final Clarifier as a Function of
Waste-Water Characteristics. Ins Proc. 13th Mid-Atlantic lid. Waste Conf.
1981, pp. 370-377.
196. Mancini, J.L. Gravity Clarifier and Thickener Design. In; Proc. 17th Ihd. Waste
Conf., Purdue University, West Lafayette, Indiana, 1962, pp. 267-277.
197. March, R.P. and M.J. Hamlin. An Investigation into the Performance of a Full-
Scale Sedimentation Tank. J. Proc. list. Sew. Pur if. 1966, pg. 118.
198. Maskell, A.D. and J.P. Lumbers. Automatic Flow Balancing. Progress in Water
Technology, 6, 1974.
199. Matasci, R.N., Kaempfer, C. and Heidman, J.A. Full-Scale Studies on the
Trickling Filter, Solids Contact Process. A paper presented at the 58th Annual
Conference of the Water Pollution Control Federation, Kansas City, MO,
October 1985.
-187-
-------
200. Matsunaga, K. Design of Multistory Settling Tanks in Osaka. J. Water Pollut.
Control Fed. (Japan), 52(5):950-954, 1980.
201. Matsui, S,, and N. Furuya. Activated Sludge Control in the Secondary Settling
Basin Using the Sludge Settling Analyzer. Progress In Water Technology,
12(3):213-220, 1980.
202. Mau, G.E. A Study of Vertical-Slotted Inlet Baffles. Sewage and Industrial
Wastes, 31(12):1349, 1959.
203. Mazurczyk, A.L. and J.E. Smith, Jr. Hydraulic Efficiency of Radial Final
Settling Tank. A report on Project JB-5-532-26 prepared by the Institute of
Meteorology and Water Management, Water Engineering Division, Warsaw,
Poland, 1983, 106 pp.
204. Mauurczyk, A.L., J.E. Smith, C.S. Applegate, and M. Metier. Methodology for
Assessing Clarifier Operation as Demonstrated at Pruszkow, Poland. An
unpublished paper presented at the 53rd Annual Conference of the Water
Pollution Control Federation, Las Vegas, Nevada, 1980.
205. McCarthy, R.L. and R. L. Sokol. Control Systems for Return and Waste Sludge.
Waste Engineering and Management, pp R-152, R-154, 1981.
206. McKinney, R.E. Level Detection System. U.S. Patent Office, No. 3,923,655,
941(1):315, 1975.
207. McKinney, R.E., and W.J. O'Brien. J. Water Pollut. Control Fed.,
4Q(11):1831-1843, 1968.
208. McKinney, R.E. Sedimentation Tank Design and Operation. Proceedings from a
Short Course, James M. Montgomery, Cons. Engrs, Inc., January 8, 1977.
209. Mendis, J.B., and A. Benedek. Tube Settlers in Secondary Clarification of
Domestic Wastewaters. J. Water Pollut. Control Fed., 22(7):1893-1897, 1980.
210. Mercer, R.H. Rectangular vs. Circular Settling Tanks. The American City.,
1973, 3 pp.
211. Miller, Michael A. and Geoffrey Q. Miller. Activated Sludge Settling in High
Purity Oxygen Systems - A Full Scale Operating Data Correlation, An
unpublished paper presented at the 51st Annual Conference of the Water
Pollution Control Federation, Anaheim, California, 1978.
212. Murch, F. Labelling of Particulate Matter for Tracer Studies in Incinerators,
Sedimentation Tanks and Mixers. International J. Applied Radiation and Iso-
topes, 28(10/11):889-894, 1977.
213. Mulbarger, M.C., K.L. Zacharias, F. Nazir and D. Patrick. Activated Sludge
Reactor/Final Clarifier Linkages Success Demands Fundamental Understanding.
Unpublished paper presented at the 57th Annual Conference, Water Pollution
Control Federation, New Orleans, Louisiana, 1984, 20 pp.
214. Munch, W.L., and J.A. Fitzpatrick. Peformance of Circular Final Clarifiers at
an Activated Sludge Plant. J. Water PoUut. Control Fed., 50(2): 265-276, 1978.
-188-
-------
215. Murk, John S. and Andrew L. Gram. Abstract of unpublished paper entitled
"Sludge Movement in Rectangular Final Sedimentation Tanks" presented at the
43rd Annual Conference of the Water Pollution Control Federation, Boston,
Massachusetts, 1970, 3 pp.
216. Murk, John S. Factors Affecting Sludge Movement in Final Sedimentation Tanks.
Report prepared by County Sanitation Districts of Los Angeles County, 1969, 8
pp.
217. Murphy, K.L. Tracer Studies in Circular Sedimentation Basins. In: Proc. 18th
Ind. Waste Conf., Purdue University, West Lafayette, Indiana, 1963, pp. 374-390.
218. Muszkalay, L. Hydraulic Field Tests on a Settling Basin. In: Proc. Fifth
International Conference, Advances in Water Pollution Research. (U.K.), 1970,
7 pp.
219. Mynheir, M.D. and C.P.L. Grady, Jr. Design Graphs for Activated Sludge
Process. J. Environmental Eng. Div., ASCE, 101(EE5):829-846, 1975.
220. Nambu, Yasumoto Magara Shoichi, and Kohken Utosawa. Biochemical and
Physical Properties of an Activated Sludge on Settling Characteristics. Water
Research, 10(l):71-77, 1976.
221. Nogita, S., S. Watanabe, and S. Mori. Experiences of MLSS Control of the
Activated Sludge Process. Water Science and Technology (Germany), 13(10):421-
426, 1981.
222. Norris, Dan P., Denny S. Parker, Marvin L, Daniels, Eben L. Owens. High-
Quality Trickling Filter Effluent without Tertiary Treatment. J. Water Pollut.
- Control Fed., 54{7):1087-1098, 1982.
223. Okuno, N., and H. Fukuda. Analysis of Existing Final Clarifiers and Design
Consideration for Better Performance. Tokyo Metropolitan Government,
Department of Sewage Works, 20 pp.
224. Olsson, Gustaf, -Toe Stephenson, and David Chapman. Hydraulic Shocks and Their
Impact on Secondary Settler Performance. An unpublished paper presented at
the 57th Annual Conference of the Water Pollution Control Federation, New
Orleans, 1984, 16 pp.
225. Onuma, K. and Y. Funakoshi. Tentative Guidelines for Troubleshooting of
Existing Wastewater Treatment Plants. Progress Report, U.S./Japan Joint
Research Project (Japanese Side), UJJRP-JS-17, Chapter 7, March 1983.
226. Orcutt, J.C. Fluidized Bed Clarification as Applied to Wastewater Treatment.
PB-216 604, National Technical Information Service, Springfield, Virginia, 1972.
227. Parker, Denny and Richard Stenquist. Flocculator-Clarifier Performance. J.
Water Pollut. Control Fed., 58(3):214-219, 1986.
228. Parker, Denny S., Warren J. Kaufman, and David Jenkins. Physical Conditioning
of Activated Sludge Floe. J. Water Pollut. Control Fed., 43(9):1817-1833, 1971.
-189-
-------
229. Parker, D.S. Assessment of Secondary Clarification Design Concepts. J. Water
Pollut. Control Fed., 55(4):349-359, 1983.
230. Parker, D.S., R. Stenquist and T. Holstrom. Performance of Flocculator-
Clarifiers for Meeting Polished Secondary Requirements. Unpublished paper
presented at Utah Water Pollution Control Federation Conference, 1985, 28 pp.
231. Parker, D.S., et al. Floe Breakup in Turbulent Flocculation Processes. J. Sanit.
Eng. Dw., ASCE, 79, 1972.
232. Peabody Welles. Clarification Equipment. Technical Bulletin PW 9000 (1976).
233. Perkins, R., and T.R. Wood. Some Thoughts on the Design of Secondary-
Settlement Tanks in the Light of Operational Experience in the West Midlands.
Water Pollution Control (U.K.), 78(1):72-81, 1979.
234. Pflanz, P. and F.S. Bregas. Performance of (Activated-Sludge) Secondary
Sedimentation Basins. la: Proc. 4th International Conf. Water Pollution
Research, Prague, 1969, pp 569-581, 583-593 .
235. Pintenich, J.L., and R.A. Bell. Hydraulic Characterization of Wastewater Treat-
ment Basins. In: Proc* Bid. Waste Conf., Purdue University, West Lafayette,
Indiana, 1976, pp. 441-448.
236. Pizarro, D.R. Dynamic Response of Final Settling Tanks to Transient Loading
Conditions. PB-262-359, National Technical Information Services,
Springfield, Virginia, 1976, 141 pp.
237. Pipes, W.O. Activated Sludge Bulking and Settling Tank Performance. J. Water
Pollut. Control Fed., 51(10):2534-2537, 1979.
238. Pipes, W.O., and M.W. Schaffner. Underflow Rate and Control of an Activated
Sludge Process. J. Water Pollut. Control Fed., 50(l):20-30, 1978.
239. Pitman, A.R. Settling Properties of Extended Aeration Sludge. J. Water Pollut.
Control Fed. (South Africa), 52(3):524-536, 1980.
240. Placek, T.D. Phase and Compression Behavior in Sedimentation and Thickening
of Liquid-Solid Suspensions. National Technical Information Service, PB3-
249664, 1983, 12 pp.
241. Pollock, R.J., W.A. Barkley, and L.A. Glasgow. Particle Size Reduction by
Breakage in Biological Wastewater Treatment. Biotechnology and Bioengineer-
ing, 25(4):901-918, 1983.
242. Porta, F.R., S.E. Williams, and A.R. Posthuma. Plant Scale Final Clarifier
Studies at Detroit. A paper presented at the 53rd Annual Conference of the
Water Pollution Control Federation, Las Vegas, Nevada, 1980, 25 pp.
243. Price, G.A. and M.S. Clements and M.I. Mun. Some Lessons from Model and
Full-Scale Tests in Rectangular Sedimentation Tanks. Paper No. 1, Papers
submitted for publication, in Water Pollution Control, 1974, pp. 102-113.
-190-
-------
244. Rachwal, A.J., M.J. Hanbury, D.J. Critchard, and D.W.M. Johnstone. The
Application o£ Settleability Tests for the Control of Activated-Sludge Plants. Ins
Bulking of Activated Sludge, Chambers and Tomlinson, eds.(U.K.), 1982, pp. 224-
244.
245. Randall, C.W., and L.D.Benefield. Verdict Still Undecided on Best System for
Settleability, Part 2. Water and Sewage Works, 126(6):7Q-72, 1979.
246. Rankin, R.S. Weirs on Sedimentation Tanks. Public Works, (12):66, 1959.
247. Rebhun, M., and Y. Argaman. Evaluation of Hydraulic Efficiency of Sediment-
ation Basins. J. Sanit Eng. Div., ASCE, 91(SA5):37-45, 1965.
248. Reed, S.C., T. Buz2ell, and S. Buda. A Floating Settler for Low cost
Clarification. In; Proc. 1st Int. Meet. Soc. Eng. Sci., 1973, 12 pp.
249. Reed, S.C. and Murphy, R.S. Low Temperature Activated Sludge Settling, Jour.
Sanitary Engineering Division, Proc. Am. Soc. of Civil Engr. 95, 747 (1969),
1263.187.Resch, H. Investigations on the Performance of Upflow Final Settling
Tanks in Activated Sludge Plants. Progress in Water Technology, 12(3):2G1-211,
1980.
250. Resch, H. Investigations on the Performance of Upflow Final Settling Tanks in
Activated Sludge Plants. Prog. Wat. Tech., 12(3):201-211, 1980.
251. Rex Chainbelt, lie. A Mathematical Model of a Final Clarifier. U.S. Environ-
mental Protection Agency, Office of Research and Monitoring, Project No.
17090 FJW 02/72, Contract No. 14-12-194, U.S. Government Printing Office,
Washington, D.C., 1972, 106 pp.
252. Rexnord, Inc. High Rate Hydraulic Study of Rim-Flo, Peripheral Takeoff
Clarifier, Oakland East Bay Municipal Utility District, Corporate Research and
Development, Environmental Research Center, Project No. 3W54 (J-61772), Rpt.
No. 11, 1979, 2 pp.
253. Rich, Linvil G. Unit Operation of Sanitary Engineering. John Wiley & Sons, Inc.,
New York, New York (1961).
254. Riddell, M.D.R., J.S. Lee, and T.E. Wilson. Method for Estimating the Capacity
of an Activated Sludge Plant. J. Water Pollut. Control Fed., 55(4):360-368,
1983.
255. Robinsion, J.H. A Study of Density Currents in Final Sedimentation Basins. M.S.
Thesis, University of Kansas, 1974.
256. Rohlich, G.A. Investigation of the Behavior of Oil-Water Mixtures in Separators,
Final Report, University of Wisconsin, August 1951,
257. Ross, R.D. The Influence of Waste Activated Sludge on Primary Clarifier
Operation. An unpublished paper presented at the 57th Annual Conference of
the Water Pollution Control Federation, New Orleans, Louisiana, 1984, 13 pp.
-191-
-------
258. Sackellares, R.W., W.A. Barkley, and R.D. Simmons.. Comparison of Parallel
Clarifiers with Level and Nonlevel Weirs. In: . Proc. 39th Lid. Waste Conf.,
Purdue University, West Lafayette, Indiana, 1984, pp. 651-644.
259. Saffran, E.P. What Makes Sludge Collector Drive Tick? Water/Engineering and
Management, 130(l):24-25,28,45, 1983.
260. Seiko, F.F., and S.S. Chandler. Method and Apparatus for Separation of Sludge.
U.S. Patent Office No. 3,822,204, 924(1):107, 1974.
261. Schaffner, M.W. and W.O. Pipes. Underflow Rate and Control of an Activated
Sludge Process. J. Water Pollut. Control Fed., 50(l):20-30, 1978.
262. Schamber, D.R., and B.E. Larock. Numerical Analysis of Flow in Sedimentation
Basins. J. Hydraulics Division, ASCE, 107(HY5):575-591, 1981.
263. Scherer, C.H., and V.J. Jordan, Jr. Gravity Thickening Techniques at a Water
Reclamation Plant. J. Water Pollut. Control Fed., 42:180-189, 1970.
264. Schibelka, Frank E. Factors Affecting the Performance of Final Clarifiers in the
Activated Sludge Process. An unpublished paper presented at the 48th Annual
Conference of the California Water Pollution Control Association, South Lake
Tahoe, California, 1976, 15 pp.
265. Semon, J.A. Secondary Clarifier Distributed Inlet and Transient Flow Study. A
report prepared under agreement between Stamford, Ct. and Municipal Environ-
mental Research Laboratory, Office of Research and Development, U.S.
Environmental Protection Agency, Cincinnati, Ohio, 1985, 42 pp.
266. Severin, B.F., and R.A. Poduska. Prediction of Clarifier Sludge Blanket Failure.
J. Water Pollut. Control Fed., 57(4):285-290, 1985.
267. Shiba, S., Y. Inoue, and Y. Ueda. Dynamic Estimation of Settling Basin
Performance. J. Environmental Eng. Div., ASCE, 1113-1129, 1979.
268. Sherman, R.W. Judging the Sludge Blanket. J. Water Pollut. Control Fed.,
Highlights Deeds and Data, 17(5):9-10, 1980.
269. Sidwick, J.M. Rationalisation of Dimensions and Shapes for Sewage Treatment
Works Construction. Effluent and Water Treatment J. (U.K.), 20(6):285-288,291,
1980.
270. Silveston, P., W. Stepko, N. Sakata, J. Cordoba-Molina, K. Rietema, and R.
Hudgins. On the Use of Residence Time Distributions for the Design of
Clarifiers. The Canadian J. of Chemical Engineering. 57:83-87, 1979.
271. Silveston, P.L., J.F. Cordoba-Molina, and R.R. Hudgins. The Use of Flow
Contractions to Improve Clarifier Performance. Prog. Wat. Tech., 12:385-394,
1980.
272. Slechta, A.F., and W. R. Conley. Recent Experiences in Plant-Scale Application
of the Settling Tube Concept. J. Water Pollut. Control Fed., 43(8):1724-1738,
1971.
-192-
-------
273. Smollen, M. Settling Characteristics of Activated Sludges. Water Report (South
Africa), (14): 6-7, 1981.
274. Spar ham, V.R. Improved Settling Tank Efficiency by Upward Plow Clarification.
Water and Wastes Treatment, 12:801-811, 1969.
27 5. Spar ham, V.R. Upward Flow Clarification, Industrial Water Engineering,
p. 18-20, 1970.
276. Speece, R.E., M.J. Humenick. Solids Thickening in Oxygen Activated Sludge.
J. of the Water Pollut. Control Fed., 46(l);43-52, 1974.
277. Stevenson, D.G. The Definition of Surface Rates in Clarifiers - A Review of
Current and Earlier Practices. J. of the Institution of Water Engineers and
Scientists (U.K.), 35(2):179-185, 1981.
278. Stofkoper, J.A. and C.C. Trentelman. Guidelines for Round Final Clarifiers of
Activated Sludge Processes. H2O Magazine. (Netherlands), 12 pp., 1983.
279. Stukenberg, J.R., L.C. Rodman, and J.E. Touslee. Activated Sludge Clarifier
Design Improvements. J. Water Pollut. Control Fed., 55:(4):341-348, 1983.
280. Sundaram, T.R., J.E. Santo and N.L Shapiro. An In-Depth, Cross-Flow
Separation Technique for the Removal of Suspended Solids from Wastewaters.
Industrial Water Engineering, pp. 9-18,, 1978.
281. Takamatsu, T. and M. Naito. Effects of Flow Conditions on the Efficiency of a
Sedimentation Vessel. Water Research, 1, 1967.
282. Takamatsu, T., M. Naito, L.T. Fan. Optimization of the Activated-Sludge
Process -Optimum Volume Ratio of Aeration and Sedimentation Vessels. Water
Research, 3:433-443, 1969.
283. Talmage, W.P. and E.B. Fitch. Determining Thickener Unit Areas. Industrial
and Engineering Chemistry, 47,38-41, 1955.
284. Tang, N. Using Settling Flux Approach for Dynamic-State Control of Activated
Sludge Processes. Unpublished paper presented at the 55th Annual Conference
of the Water Pollution Control Federation, St. Louis, Missouri, 1982, 31 pp.
285. Tanthapanichakoon, W., and D.M. Himmelblau. Simulation of a Time Dependent
Activated Sludge Wastewater Treatment Plant. Water Research, 15(10)il 185—
1195, 1981.
286. Tchobanoglous, G., and M.E. Burchett. Facilities for Controlling the Activated
Sludge Process by Mean Cell Residence Time. J. Water Pollut. Control Fed.,
46(5):973-979, 1974.
287. Tekippe, R..T., and J.L. Cleasby. Model Studies of Peripheral Feed Settling Tank.
J. Sanit. Eng. Div., ASCE, 94(SAD:85-102, 1968.
288. Tekippe, R.J. and Coe, D.A, WPCF Text, p IV-69- Innovation in Design of a
Large Advanced Secondary Treatment Plant. A paper presented at the 54th
Annual WPCF Conference in Detroit, Michigan (October 6, 1981).
-193-
-------
289. Tekippe, R.J. Activated Sludge Circular Clarifier Design Consideration. A
paper presented at the 57 th Annual Water Pollution Control Federation
Conference in New Orleans, Louisiana, 1984, 23 pp.
290. Thibault, G.T., and K.D. Tracy. Controlling and Monitoring Activated-Sludge
Units. Chemical Engineering, 85{20):155-160, 1978.
291. Thirumurthi, D. A Breakthrough in the Tracer Studies of Sedimentation Tanks.
J. Water Pollut. Control Fed., 41(11):R405, 1969-
292. Townsend, Darwin W. Summary Report upon Sedimentation Tank Capacity and
Sludge Concentration Tests, 1931, 14 pp.
293. Townsend, Darwin W. and James Brower. New Sludge-Removal Apparatus
Developed at Milwaukee. Report from Engineering News-Record, 1930, 4 pp.
294. Tuntoolavest, M., E. Miller, and C.P.L. Grady. Factors Affecting the
Clarification Performance of Activated Sludge Final Settlers. J. Water Pollut.
Control Fed., 55(3):234-248, 1983.
295. Turner, M.T., and G.L. Walton. Shallow-Depth Settling. Process Biochem.,
5(l):27-29, 1970.
296. U.S. Environmental Protection Agency, Technical Assistance Project at the
Thomas P. Smith Wastewater Treatment Plant, Tallahassee, Florida, Engineering
report prepared by E.P.A., Region IV, Surveillance and Analysis Division, 1976,
18 pp.
297. U.S. Environmental Protection Agency. Process Design Manual for Suspended
Solids Removal. EPA 625/l-75-003a (January 1975).
298. Vasiliev, V.B., and V.A. Vavilin. Activated Sludge System Design for Large
Variations of Organic Loading. Biotechnology and Bioengineering (U.S.S.R.),
24(ll):2337-2355, 1982.
299* Vaughn, C.B. Settling Behavior of Waste Suspensions. J. Sanit. Eng. Div., ASCE,
83(SA5):1423-1, 1957.
300. Vesilind, P.A. The Influence of Stirring in the Thickening of Biological Sludge.
Ph.D. Thesis, University of North Carolina, 1968.
301. Villier, R.V. Removal of Organic Carbon from Domestic Wastewater. Master's
Thesis, University of Ohio, 1967.
302. Vosloo, P.B.B. Some Factors Relating to the Design of Activated-Sludge Plants.
Water Pollution Control, 69(5):486-494, 1970.
303. Wagner, F. The Separation of Activated Sludge from Treated Sewage with Jet
Separation. Wasserwirtschaft (West Germany), 70(6):232-235, 1980.
304. Wallace, A.T. Design and Analysis of Sedimentation Basins. Water and Sewage
Works, 1,R-219, 1967.
305. Warden. J.H. The Design of Rakes for Continuous Thickeners. Filtration and
Separation., 113-115, 1981.
-194-
-------
306. Water Pollution Control (Canada). Clarifier Designed for Use with Surface
Aerator. 111(11):28,30-31, 1973.
307. WPCF-ASCE Joint Committee, Wastewater Treatment Plant Design, WPCF
Manual of Practice 8, 1977.
308. Water Pollution Control Federation. Clarifier Design. MOP No. FD-8, Lancaster
Press, Lancaster, PA, 1985.
309. Wear, M.R., J.A. Grantz, and R.J. Thompson. Biological Treatment of Coke
Plant Waste Utilizing an Integral Clarification Concept. In: Proc. 35th lhd.
Waste Conf., Purdue University, West Lafayette, Indiana, 1980, pp 342-353.
310. Weber, C.L., and C.D. Jacobson. New System Puts the Wood to Wastewater.
Water and Wastes Eng., 12(12):51-52,64, 1975.
311. Wheale, G., D.J. Williamson. Unusual Behaviour of Ciliated Protozoa in a
Secondary Settlement Tank. Water Pollution Control (U.K.), 79(4):496-500, 1980.
312. White, M.J.D. Design and Control of Secondary Settlement Tanks. Water
Pollution Control (U.K.), 75(4):459-467, 1976.
313. White, M.J.D. Experience of the Use of a Simple Parameter to Predict the
Maximum Solids Loading on Sedimentation Tanks in Large Activated-Sludge
Plants. Progress in Water Technology, 8(6):515-519, 1977.
314. Wilson, T.E. Application of ISV Test to the Operation of Activated Sludge
Plants. Water Research, 17, 707, 1983.
315. Wilson, T.E., and J.S. Lee. Comparison of Final Clarifier Design Techniques.
J. Water Pollut. Control Fed., 54(10):1376-1381, 1982.
316. Wonda, T.W.M., K. Rietema, S.P.P. Ottengraf. Effects of Density Gradients and
Velocity Profiles on Sedimentation Efficiency. Chemical Engineering Science,
32:351, 1977.
317. World Water. Simplified Clarifier Axes Treatment Stages., 1982, pp. 22-25.
318. Wullschleger, R.E., W.J. Katz, and A.E. Albrecht. Solids Sensing Apparatus for
Clarifier Sludge Blankets. Water Works Wastes Eng., 2(9):100-102, 1965.
319* Yoshioka, N., Y. Hotta, S. Tanaka, S. Naito, and S. Tsugani. Continuous
Thickening of Homogeneous Flocculated Slurries. Chemical Engineering, Tokyo,
21,66-74, 1957.
320. Young, J.C., J.L. Cleasby, and E.R. Baumann. Flow and Load Variations in
Treatment Plant Design. J. Environmental Eng. Div.,ASCE, 104(EE2):289-303,
1978.
-------
TECHNICAL REPORT DATA
(Please read /mtrucltom on the reverse be/ore compltiingl
1. REPORT NO. 2.
EPA/600/2-87/07 5
3. RECIPIENT'S ACCESSION NO.
PB8Q. 1 02967BS
4. TITLE ANO SUBTITLE
Critical Literature Review arid Research Needed on
Activated Sludge Secondary Clarifiers
S. REPORT DATE
September 1987
S. PERFORMING ORGANIZATION CODE
7. AUTWOR(S)
Rudy J. Tekippe
8. PERFORMING ORGANIZATION REPORT NO.
9, PERFORMING ORGANIZATION NAME AND AOORESS
James M. Montgomery, Consulting Engineers, Inc.
P.O. Box 7009
Pasadena, CA 91109-7009
ID. PROGRAM ELEMENT NO.
11. CONTRACT/RMXXJW-
68-03-1821
12. SPONSORING AGENCY NAME ANO AOORESS
Water Engineering Research Laboratory, Cinti, OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
13. TYPE OF REPORT ANO PERIOO COVERED
Final Report
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer: Jon H. Bender {513) 569-7620 . MTTC
(FTS) 684-7620 Submitted to NTIS
IS. ABSTRACT
This report presents a critical review of the literature regarding the
design and operation of activated sludge secondary clarifiers. It also presents
the findings of a group of clarifier experts who participated in a symposium that
assessed what information was needed from future clarifier research. These
activities were intended to advance the state-of-the-art of clarifier design and
operation and provide direction for future clarifier research.
17. KEY WOROS ANO DOCUMENT ANALYSIS
J. DESCRIPTORS
b.IDENTIFIERS/OPEN endeo terms
e. cosati I'idd,'Group
18. DISTRIBUTION STATEMENT
t9. SECURITY CLASS /Tilts tt,-porn
Unclassified
21. NO. OP PAGES
206
He I ease to Public po. security class tnispj^-i
1 Unclassified
22. PRICE
SPA Form 2220-1 (R>«. 4-77) previous edition i* onioitrc
i
------- |