£EPA
United States
Environmental Protection
Agency
Municipal Environmental Research EPA-600/2-78-1853
Laboratory September 1978
Cincinnati OH 45268
Research and Development
Short Course
Proceedings
Applications
of Computer Programs
in the Preliminary
Design of Wastewater
Treatment Facilities
Section I
Workshop Lectures
\
\
/
Et< 600/2
78-lF.5a
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U S Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface ir related fields
The nine series are
1 Environmental Health Effects Research
2 Environmental Protect on Technology
3 Ecological Research
4 Environmental Monitoring
5 Socioeconomic Environmental Studies
6 Scientific and Technical Assessment Reports (STAR)
7 ln*eragency Energy-Environment Research and Development
8 ' Special" Reports
9 M scellaneous Reports
This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series This series describes research performed to develop and dem-
onstrate instrumentation ecuipment and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution This work
provides the new or improved technology required for the contrc! ana treatment
of pollution sources to mee! environmental quality standards
This document is available to tne public through the National Tecinical Informa-
tion Service Springfield. Virginia 22161
-------
EPA-600/2-78-185a
September 1978
Short Course Proceedings
APPLICATIONS OF COMPUTER
PROGRAMS IN THE PRELIMINARY DESIGN
OP WASTEWATER TREATMENT FACILITIES
Section I: Workshop Lectures
August 15-19, 1977
Illinois Institute of Technology
Chicago, Illinois 60616
Edited by
James W. Male
University of Massachusetts
Amherst, Massachusetts 01003
and
Stephen P. Graef
Metropolitan Sanitary District of Greater Chicago
Chicago, Illinois 60611
Grant No. R-805134-01
Project Officer
Richard G. Eilers
Wastewater Research Division
Municipal Environmental Research Laboratory
Cincinnati, Ohio 45268
MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
DISCLAIMER
This report has been reviewed by the Municipal
Environmental Research Laboratory, U. S. Environmental
Protection Agency, and approved for publication. Approval does
not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commerical products constitute endorse-
ment or recommendation for use.
11
-------
FOREWORD
Protection
government
The Environmental
increasing public and
pollution to the health and
Noxious air, foul water, and s
monies to the deterioration of
complexity of that environment
components require a
problem.
welfare
concentrated
Agency was created because of
concern about the dangers of
of the American people.
poiled land are tragic testi-
our natural environment. The
and the interplay between its
and integrated attack on the
Research and development
problem solution, and it involves defining the problem, measur-
ing its impact, and searching
Environmental Research Laboratory develops new and improved
technology and systems to prev
water and solid and hazardous
economic, social, health, and
This publication is one of the
is that necessary first step in
for solutions. The Municipal
ent, treat, and manage waste-
waste pollutant discharges from
municipal and community sources, and to minimize the adverse
aesthetic effects of pollution.
products of that research—a most
vital communications link between the research and the user
community.
The purpose of this shor4 course was to introduce and
familiarize participants with
liminary Design of Wastewater
is intended for use in the preliminary sizing and costing of
the various components of a w£
accomplish its intended purpos
during the short course. Con:
of a short lecture describing
the Executive Program for Pre-
Treatment Systems. The program
stewater treatment plant. To best
e, the course was structured to
fully involve the participants; and encourage use of the program
equently, each workshop consisted
some aspect of the Executive
Program, followed by assignment of a problem. The participants
then utilized the program to solve the specified problem. This
hands on approach allowed considerable exposure to the Executive
Program and extensive interaction with the short course faculty.
Francis T. Mayo
Director, Municipal Environmental
Research Laboratory
111
-------
ABSTRACT
This document contains the material used for the Short
Course on the Applications of Computer Programs in the Pre-
liminary Design of Wastewater Treatment Facilities. The users'
manual describes the use of the program and subroutines.
Several examples show appropriate input and expected output for
a variety of applications. In addition, the theoretical back-
ground and computer listing are presented for the main program
and each of the 27 subroutines.
Section I of this report contains the Short Course lectures,
These workshops describe how to use, modify, and/or augment
the Exec Program to meet the user' sspecific needs. Applications
included: (1) the effect of design criteria selection, (2)
multiple flow scheme cost and performance comparison, (3) the
effect of economic parameter selection, (4) subroutine modifi-
cation, (5) cost curve modification, (6) addition of new sub-
routines, (7) subroutine modification for simulation studies,
and (8) use of a stream impact subroutine.
This report was submitted in partial fulfillment of Grant
Number R-805134-01 by the Pritzker Department of Environmental
Engineering at the Illinois Institute of Technology under the
sponsorship of the U. S. Environmental Protection Agency.
This report covers a period from May 23, 1977 to June 22, 1978
and work was completed as of June 22, 1978.
IV
-------
CONTENTS
Page
Foreword ........................
Abstract ........................
Acknowledgement ....................
Short Course Faculty .................. viii
The Role of Computer Programs in Preliminary De-
sign
R. G. Eilers, R. Smith .............. 1
Case I Workshop: Use of the Exec Program to De-
termine the Effect of Design Criteria Selec-
tion on Plant Cost and Performance
T. K. Walsh ................... 18
Case II Workshop: Use of the Exec Program to
Compare the Cost and Performance of Multiple
Flow Schemes
R. G. Eilers ................... 52
Case III Workshop : Use of the Exec Program to
Determine the Effect of Economic Parameters
on Capital and 0/M Costs for a Given Facility
Design
R. J. Avendt ................... 68
Case IV Workshop: Modification of Existing Exec
Program Subroutines
S. P. Graef ................... 86
Case V Workshop: Modification of Existing Exec
Program Cost Relationships
B. F. Winkler .................. 115
Case VI Workshop: Addition of New Subroutines
to the Exec Program
R. D. Letterman ................. 136
v
-------
CONTENTS (Cont.)
Page
Case VII Workshop: Modification of an Existing
Design Subroutine for Process Simulation
Studies
W. J. Maier 161
Case VIII Workshop: Use of the Stream Impact
Program in conjunction with the Exec Program
J. W. Male 178
VI
-------
ACKNOWLEDGEMENTS
Many people contributed to the preparation for the Short
Course on Applications of Computer Programs in the Design of
Wastewater Treatment Facilities. Without their efforts, arrange-
ments would have been incomplete and material unprepared.
Many thanks go to Steve Graef, who has spent considerable
time over the last year refining and documenting the Exec pro-
gram. Dick Eilers and Bob Smith of the USEPA also provided guid-
ance throughout the course of the workshop preparations.
Contributing to the massive typing effort were Margaret
Nolan, Mary Keeley, Pat Woods, Mary Pierce, Janet Peterson and
Dotty Pascoe. In addition, Russ Ritchie helped with local ar-
rangements and everyday details.
A special note of gratitude goes to two IIT students,
Hisashi Ogawa and Phong Nguyen. Hisashi was responsible for main'
taining, updating, and correcting the many computer files and
Phong coordinated the writing and typing of the subroutine users'
guides. For their constant effort and careful attention to
details, I am extremely grateful.
James W. Male
Short Course Chairman
Vii
-------
FACULTY
Raymond J. Avendt, PE
Associate
Consoer, Townsend and Associates
360 East Grand Avenue
Chicago, Illinois 60611
Richard G. Eilers
Systems and Economic Analysis Section
Municipal Environmental Research Laboratory
U. S. Environmental Protection Agency
Cincinnati, Ohio 45268
Stephen P. Graef, PE, PhD
Principal Sanitary Engineer
Metropolitan Sanitary District of Greater Chicago
100 East Erie
Chicago, Illinois 60611
Raymond D. Letterman, PE, PhD
Associate Professor
Department of Civil Engineering
Syracuse University
Syracuse, New York 13210
(formerly of Illinois Institute of Technology)
Walter J. Maier, PhD
Associate Professor
Department of Civil and Mineral Engineering
University of Minnesota
221 Church Street, S.E.
Minneapolis, Minnesota 55455
James W. Male, PE, PhD
Associate Professor of Civil Engineering
University of Massachusetts
Amherst, Massachusetts 01003
(formerly of Illinois Institute of Technology)
Robert Smith, PE
Chief, Systems and Economics Analysis Section
Municipal Environmental Research Laboratory
U. S. Environmetnal Protection Agency
Cincinnati, Ohio 45268
viii
-------
Thomas K. Walsh, PE
Project Engineering
Metcalf and Eddy, Incorporated
50 Staniford Street
Boston, Massachusetts 02114
Barry F. Winkler, PE
Senior Civil Engineer
Engineering Department
Metropolitan Sanitary District of Greater Chicago
100 East Erie
Chicago, Illinois 60611
ix
-------
THE ROLE OF COMPUTER PROGRAMS IN
PRELIMINARY DESIGN
Richard G. Eilers
and
Robert Smith
U. S. Environmental Protection Agency
26 West St. Clair Street
Cincinnati, Ohio 45268
ABSTRACT
Mathematical models in the form of computer programs have
been developed for use in aiding the consulting engineer in
producing cost-effective designs for wastewater treatment systems.
These programs can assist the plant designer by supplementing
his experience and judgment. Some of the programs are capable
of doing both performance and cost analysis, and thereby minimize
the computational work required for examining many alternate
designs for achieving a desired effluent quality at a minimum
cost. Easy access to computing facilities and relatively low
usage cost make the idea of computer-aided design both desirable
and practical.
-------
INTRODUCTION
Since large-scale computers have now been available for
30 years, it is not necessary to discuss whether or not they are
useful for design engineering applications. There are probably
very few remaining areas of the engineering profession in general
that do not make at least some use of computer programs, and the
potential for computerized design techniques in planning waste-
water treatment systems appears to be significant. However,
only a modest effort has been made over the past ten years to
develop practical computer software that can be used for this
particular purpose. The main reason for this seems to be the
more-or-less accepted attitude that no two treatment plant de-
sign situations are the same and, therefore, cannot be solved
by a generalized system of computations. Human judgment based on
experience is often looked upon as being of considerably more
value than systemized computer calculations. Also, there is
frequently strong resistance to changing established design
methods and procedures. Thinking such as this can cause the
planner to totally reject the idea of using preliminary design
software in his activities. The important point that needs to
be made is that computer programs should act as a supplemental
tool to aid the engineer in performing his design work. The
emphasis should be put on assisting and not on replacing the
need for experience and judgment in wastewater treatment plant
design. Since computers are not available to almost everyone
and the cost of computing is extremely cheap (with respect to
how much manual labor is eliminated), the wise engineer will
make use of computerized techniques whenever these methods can
be of assistance in solving design problems.
A computer program is basically a model. The system of pro-
cedure that it represents is described in mathematical form by
means of a computer language, such as FORTRAN. Computer programs
for preliminary design of wastewater treatment systems are models
by which the performance of the system or its cost is studied
by means of adjusting parameters that affect the calculations
being performed. Preliminary design systems usually consist of
a group of individual models that represent the different compo-
nents of the system. These sub-models are then connected to one
another by the flow scheme which joins the components of the
real system. For simulation studies, the input parameters and
the design of a particular treatment system are known, and a
characterization of the system output is sought. The behavior
of the system is observed as input data changes or as the mode of
-------
operation is varied. For design purposes, the inputs and outputs
for the particular system are known, and a system configuration
is sought which will satisfy the established requirements. The
equipment required to do the job and the size of it can be
determined with the design model. Both performance and cost
are calculated for the desired system. In designing wastewater
treatment plants, the problem often becomes one of finding the
sizes, operating conditions, and cost of the unit processes
which make up the system configuration. However, the type of
plant design is often selected on the basis of tradition or the
requirements of some regulatory agency, and not through cost-
effective analysis.
COMPUTER-AIDED PROCESS DESIGN
Much more work in the area of computer-aided process design
has been done in the chemical" industry than in the waste treat-
ment field. Computer-based process design has been commonly
used by all the major oil companies and chemical producers for
several years now. Some examples of various applications would
be: propane recovery from natural gas, methanol synthesis,
and-ammonia production. Since many different design programs
for the chemical industry have been developed and the fact that
they are of little value for waste treatment design, only one
of these programs will be described in order to give an idea
of its structure and capabilities. Also, many of the chemical
design programs are quite similar in various respects.
The CHESS system was developed at the University of
Houston and provides the user with some standard equipment sub-
routines for the most commonly used basic chemical process units
and a thermodynamic properties evaluation routine for some 62
basic chemical components. Additional chemical components may
be added. The system structure is so developed that it allows
individual users to create and add their own equipment or
process module subroutines if needed. Examples of some modules
would be stream divider, distillation, mixing (several types),
heat exchanger, compressor, absorber, etc. Examples of some
chemical components would be hydrogen, methane, water, oxygen,
carbon dioxide, etc. This model does not calculate any cost
information; only performance. The program consists of 6500
source cards, written in FORTRAN, and takes 4OK words on an SDS
Sigma 7 computing system.
There also exists a limited number of executive programs
that can be used for waste treatment studies. The simulation
type models would be PACER(2), SEPSIM^3), GEMCS^4', and
MACSIM (5).These are all essentially general purpose simulation
executive programs which have been adapted for waste treatment
systems by formulating a specialized library of subroutines.
All are very similar in operation. PACER and MACSIM contain a
-------
network analysis routine which locates recycle loops in the flow
scheme and organizes the iterative calculations needed for these
loops. This portion can be bypassed if the user wants to speci-
fy a calculation scheme. GEMCS contains an optional subroutine
to perform network analysis. SEPSIM has no network analysis
capability, because it was developed for waste treatment systems
which have simple networks allowing the order of calculation to
be established by inspection. The design type models would be
ESTHER(6) and ASOP^7! ASOP is a version of ESTHER which includes
a pattern search routine and is capable of choosing a set of
design parameters which will optimize an objective function that
is chosen by the user. ESTHER is a combined design and simula-
tion program which, through a user specified control value,
selects the mode. The design mode calculates equipment sizes
for a given effluent quality and various other outputs for each
process until that is used in the system. ESTHER can handle a
wide variety of waste treatment systems, but the user would
probably have to develop the unit process models that are de-
sired, since most work has centered around characterizing and
optimizing the activated sludge process. The several executive
programs mentioned thus far tend to be more academic than prac-
tical in that they were developed in a university atmosphere
and little or no attention was given to calculating the costs
associated with building and operating the designs which are
produced. A thorough discussion and description of these com-
puter programs is given in a Canadian report(8) on computer-
aided design and simulation of waste treatment systems.
Two other design type models that provide both performance
and cost information to the user are CAPDET(9) and EXEC^10'.
CAPDET allows the user to specify various types of unit processes
for. wastewater treatment. The unit processes together with their
design parameters may then be assembled in sequence to form
various versions of four types of treatment schemes. The pro-
gram processes all combinations of unit processes and evaluates
the treatment cost for each train. The trains are ranked ac-
cording to least average annual cost. The calculated effluent
quality is checked against the desired effluent characteristics,
and those trains not meeting the desired quality are discarded.
Cost data and design criteria are output from the program.
The stream characteristics that are considered differ somewhat
from those of EXEC; pH, C, anions, cations, grease, etc. are
included in CAPDET. This program contains certain unit pro-
cesses not yet developed for EXEC (carbon adsorption, ammonia
stripping, lagoons, etc.) and vice-versa (land disposal, lime
addition to sludge, incineration, and rotating biological con^
tactors). Standard inputs to the program are fixed unless
changed by the user. CAPDET copies much of its content from
EXEC, is not as flexible, is not as detailed, uses no iterative
techniques, and requires a large-scale computing system. Its
major value would be for comparing a large number of treatment
alternatives. The Executive Program (EXEC) is the EPA-developed
-------
computer program for preliminary design of wastewater treatment
systems which will form the basis of this short course-workshop.
For this reason, a detailed discussion of the model background,
development, uses, etc. will follow. The EPA has also
created a number of other specialized design and cost-estimating
programs for wastewater treatment, and these will also be dis-
cussed briefly.
MATHEMATICAL MODELING BY EPA
The Systems and Economic Analysis Section of the Wastewater
Research Division of EPA in Cincinnati, Ohio is concerned with
finding quantitative expressions for calculating the performance
and cost of wastewater treatment processes as a function of the
nature of the wastewater to be treated and the design variables
associated with the individual unit processes. These models are
intended primarily to characterize the treatment of municipal
sewage. Since the procedure for solving all of the quantitative
equations is usually too laborious or complex to be accomplished
by hand calculation, various FORTRAN computer programs have been
developed to perform the task.
BACKGROUND
Mathematical models for wastewater treatment processes are
required to express the performance of the processes over the
full range of operational modes and design criteria. These
models can be steady state, quasi-steady state, or time-
dependent. By quasi-steady state it is meant that a steady
state model is used to simulate a process that is, in reality,
not necessarily steady state. Most sewage treatment systems are
not steady state. The time-dependent or dynamic models are of
interest when the quality of the effluent stream from a process
is important as a function of time, or when the effectiveness of
various kinds of control schemes on a process is being studied.
For a model to be fully effective for design and planning
purposes, it must be based on valid scientific principles, flexi-
ble enough to simulate experimental data from a full-scale pro-
cess (not merely pilot-scale data), and represent the perfor-
mance and cost of the process with adequate precision.
The collection of valid, complete experimental data follow-
ed by adjustment of the model parameters to make the computed
results agree with experimental results within an acceptable
tolerance is also an important phase of model development.
Packaging mathematical models as computer programs not only
provides ease and accuracy of calculation, but also has the
additional advantage of convenience of distribution to interes-
ted individuals, such as consulting engineers and urban planners,
in a readily usable form.
-------
MODELS DEVELOPED
Over the past eight years, a number of computer models have
been developed in-house by the Systems and Economic Analysis
Section and through contracting activity with outside sources.
Each program deals in someway with cost and/or performance of
wastewater treatment systems. All of the computer programs
were written in FORTRAN and designed to run on a 16K IBM 1130
machine, and supporting documentation has been prepared for each.
Table 1 gives a listing of the models which were produced in-
house, and Table 2 shows the models which resulted from extra-
mural sources. A brief description of the most significant of
these computer programs will follow.
EXECUTIVE PROGRAM
The major product of all this effort has been the "Execu-
tive Digital Computer Program for Preliminary Design of Waste-
waster Treatment Systems. It was realized that a tool was
needed which would allow the process designer to select a group
of unit processes, arrange them into a desired configuration,
and then calculate the performance and cost of the system as a
whole. The Executive Program meets this need by simulating
groups of conventional and advanced wastewater treatment unit
processes arranged in any logical manner. Each unit process is
handled as a separate subroutine which makes it possible to add
additional process models to the program as they are developed.
There are presently 24 process subroutines in the program, and
these are listed in Table 3. Additional subroutines are planned
to be included in the future, and a tenative list is shown in
Table 4.
The first step in using the Executive Program is to draw
the desired system diagram showing the unit processes to be
used and the connecting and recycle streams. All streams and
processes are then numbered by the program user. Figure 1 de-
picts a typical, conventional activated sludge treatment system
with incineration for sludge disposal. Volume and characteris-
tics of the influent stream to the system and design varibles
for each process used must be supplied as program input. By an
iterative technique, each process subroutine is called in the
proper sequence and all stream values are recomputed until
the mass balances within the treatment system are satisfied.
Performance, cost, and energy requirements for each unit process
and the system as a whole are included in the final printout.
Detailed cost data applicable for preliminary design esti-
mates is generated by the Executive Program. Construction cost
(in dollars) amortization cost, operation and maintenance cost,
and total treatment cost (all in cents per 1,000 gallons of
wastewater treated) are calculated individually for every unit
process, and a sum total of each cost is given for the entire
-------
Table 1
COMPUTER PROGRAMS PRODUCED BY THE SYSTEMS AND ECONOMIC ANALYSIS
SECTION
1. Preliminary Design and Simulation of Conventional Waste-
water Renovation Using the Digital Computer (1968).
2. Executive Digital Computer Program for Preliminary Design
of Wastewater Treatment Systems (1968).
3. A Mathematical Model for a Trickling Filter (1969).
4. Preliminary Design of Surface Filtration Units-Micro-
screening C1969) .
5. A Generalized Computer Model for Steady State Performance
of the Activated Sludge Process (1969).
6. Fill and Draw Activated'Sludge Model (1969).
7. Mathematical Simulation of Ammonia Stripping Towers for
Wastewater Treatment (1970).
8. Mathematical Simulation of Waste Stabilization Ponds (1970).
9. Simulation of the Time-Dependent Performance of the Acti-
vated Sludge Process Using the Digital Computer (1970).
10. Economics of Consolidating Sewage Treatment Plants by Means
of Interceptor Sewers and Force Mains (1971).
11. Per Capita Cost Estimating Program for Wastewater Treat-
ment (1971).
12. Wastewater Treatment Plant Cost Estimating Program (1971).
13. Design of Concrete and Steel Storage Tanks for Wastewater
Treatment (1971).
14. Water Supply Cost Estimating Program (1972).
15. Cost of Phosphorus Removal in Conventional Wastewater
Treatment Plants by Means of Chemical Addition (1972).
16. A Mathematical Model for Aerobic Digestion (1973).
17. Design and Simulation of Equalization Basins (1973).
18. Mathematical Model for Post Aeration (1973).
19. Optimum Treatment Plant Cost Estimating Program (1974).
20. Waste Stabilization Ponds Cost Estimating Program (1974).
21. Granular Carbon Adsorption Cost Estimating Program (1974).
22. Control Schemes for the Activated Sludge Process (1974) .
23. Cost Estimating Program for Disinfection by Ozonation (1974)
24. Nitrification/Denitrification Cost Estimating Program (1975)
-------
Table 1, Continued
25. Cost Estimating Program for Alternate Oxygen Supply
Systems (1975).
26. Cost Estimating Program for Land Application Systems
(1975).
27. Combustion Model for Energy Recovery from Sludge
Incineration (1975).
28. Energy Consumption by Wastewater Treatment Plants (1975)
29. Stream Model for Calculating BOD and DO Profiles (1976).
-------
Table 2
COMPUTER PROGRAMS PRODUCED AS A RESULT OF CONTRACT ACTIVITY
1. Ammonia Stripping Mathematical Model for Wastewater Treat-
ment (1968) .
2. Mathematical Model for Wastewater Treatment by Ion Exchange
(1969).
3. Mathematical Model of the Electrodialysis Process (1969).
4. Mathematical Model of Tertiary Treatment by Lime Addition
(1969) .
5. Mathematical Model of Sewage Fluidized Bed Incinerator
Capabilities and Costs (1969).
6. Reverse Osmosis Renovation of Municipal Wastewater (1969)
7. Methodology for Economic Evaluation of Municipal Water
Supply/Wastewater Disposal Including Consideration of Sea-
water Distillation and Wastewater Renovation (1970).
8. Mathematical Model of Recalcination of Lime Sludge with
Fluidized Bed Reactors (1970).
9. Computerized Design and Cost Estimation for Multiple Hearth
Incinerators (1971).
10. Cost Program for Desalination Process (1971).
-------
Table 3
UNIT PROCESS MODELS CONTAINED IN THE EXECUTIVE PROGRAM
1. Preliminary Treatment
2. Primary Sedimentation
3. Activated Sludge-Final Settler
4. Stream Mixer
5. Stream Splitter
6. Single Stage Anaerobic Digestion
7. Vacuum Filtration
8. Gravity Thickening
9. Elutriation
10. Sand Drying Beds
11. Trickling Filter-Final Settler
12. Chlorination-Dechlorination
13. Flotation Thickening
14. Multiple Hearth Incineration
15. Raw Wastewater Pumping
16. Sludge Holding Tanks
17. Centrifugation
18. Aerobic Digestion
19. Post Aeration
20. Equalization
21. Second Stage Anaerobic Digestion
22. Land Disposal of Liquid Sludge
23. Lime Addition to Sludge
24. Rotating Biological Contactor - Final Settler
10
-------
Table 4
UNIT PROCESS MODELS TO BE ADDED TO THE EXECUTIVE PROGRAM
1. Ammonia Stripping of Secondary Effluent
2. Granular Carbon Adsorption
3. Ion Exchange
4. Electrodialysis
5. Reverse Osmosis
6. Bar Screening
7. Comminution
8. Grit Removal
9. Flow Measurement
10. Waste Stabilization Ponds
11. Microscreening
12. Rough Filtration
13. Multi-Media Filtration
14. Ozonation
15. Nitrification
16. Denitrification
11
-------
p
•p -p
CO CO
1
O O
•H
>-j &> Q t3 O 4-1
>-i -H OJ rH Id "3
Ql CO CO CD > 6
I I I I I I
P
Q Q
4->
I
CO
1C
&
••d
12
-------
system. Capital cost is also computed by adding onto construc-
tion expenses the costs of yardwork, land, engineering, admin-
istration, and interest during construction. All of the cost
information can be updated or backdated with respect to time
by means of cost indices that are supplied as input to the
program.
The Executive Program cannot be used for extremely detailed
design purposes. However, it can be a valuable preliminary
design tool for the consulting engineer or planner. The per-
formance of existing or proposed wastewater treatment plants
can be simulated along with providing cost estimates for build-
ing and operating these plants. It is also possible to optimize
a particular treatment system by varying design parameters and
noting the effect on performance and cost. Cost-effectiveness
studies can be made by comparing alternate treatment systems.
Initial studies along these lines are becoming of increasing
importance because of the soaring costs of plant construction
that are now being experienced.
A recent application of the Executive Program was an inves-
tigation of the potential economic advantages associated with
261 different methods for treating and disposing of sewage
sludge. Sludge production and the costs of constructing and
operating the various systems were computed. Each system was
either primary or activated sludge treatment followed by some
combination of the following 12 sludge handling processes—lime
stabilization, gravity thickening, air flotation thickening,
single-stage anaerobic digestion, two-stage anaerobic digestion,
aerobic digestion, elutriation, vacuum filtration, centrifuga-
tion, sludge drying beds, multiple hearth incineration, and
land disposal of liquid sludge. The outcome of the study showed
that the cost (in January 1974 dollars per ton of dry solids
processed) for treating and disposing of sewage sludge ranges
from about $30 per ton for anaerobic digestion followed by de-
watering on sand drying beds to over $100 per ton when the
sludge is dewatered by vacuum filtration or centrifugation and
then incinerated. Treatment and disposal of sludges produced
in municipal wastewater treatment plants were shown to account
for as much as 60 percent or as little as 20 percent of the
total cost of treatment. Therefore, careful consideration
should be given to selecting the sludge handling method which
meets the site-specific constraints at a minimum cost. The
Executive Program, which is capable of examining the cost and
performance of a wide variety of alternative sludge handling
schemes, can be used as a management tool to narrow the range
of options when design conditions are known.
The Executive Program has been around for several years now,
beginning with its original development in 1968. The model has
been expanded, modified, and corrected many times since then,
and it will continue to change in the future. The goal will
13
-------
remain the same: to provide the best possible characterization
of the cost and performance of municipal wastewater treatment
systems.
MODELS FOR THE ACTIVATED SLUDGE PROCESS
Considerable effort has been expended in developing more
accurate models for the activated sludge-final settling process.
Previous models that were produced by various researchers cover-
ed a wide range of forms corresponding to differing sets of
assumptions about the hydraulic and biological relationships
believed to be significant in the process. Because of the pro-
blems of measurement and the difficulty of fitting data to
complex models, simplified models were often used which either
omit or make some plausible assumption concerning the role of
various factors in the process.
In all, four different digital computer models for the
activated sludge process have been developed. The first, CSSAS
(Continuous Steady State Activated Sludge), is a steady state
model which is flexible enough to simulate the performance of
any configuration proposed (complete mix, plug flow, multiple
aeration tanks, step aeration, step return flow, contact
stabilization, extended aeration, etc). Two classes of micro-
organisms are considered: heterotrophs which use 5-day BOD as
substrate and Nitrosomonas which use ammonia nitrogen as sub-
strate to produce new cells. The model allows the maximum rate
constant for synthesis to vary with process loading. The second
program, FADAS (Fill and Draw Activated Sludge), attempts to
simulate the biological activity in a fill and draw bench ex-
periment where activated sludge is mixed with substrate in any
proportion. The third program, TDAS (Time-Dependent Activated
Sludge), simulates the dynamic behavior of the biological as-
pects of the activated sludge process. The model numerically
integrates the mass balance and biological rate equations which
are assumed to represent the process. Three classes of micro-
organisms are considered: heterotrophs, Nitrosomonas, and
Nitrobacter. This model can also be used to investigate the
potential advantages associated with the following control
schemes: dissolved oxygen control, sludge wasting control,
and sludge inventory control. The fourth program, CMAS (Com-
pletely Mixed Activated Sludge), is used to simulate the per-
formance of conventional and modified activated sludge,
separate nitrification, or separate denitrification. With an
adjustment of the process parameters, it can also be used to
characterize the pure oxygen activated sludge system.
SPECIALIZED COST ESTIMATING PROGRAMS
When making preliminary cost estimates for buiding and
operating certain wastewater treatment systems, it is often
necessary to have more detailed cost data. For this reason,
14
-------
special economic models were developed for several particular
applications.
A waste stabilization pond cost estimating program computes
the costs of stabilization ponds and aerated lagoons along with
influent pumping, surface mechanical aerators, embankment pro-
tection, and chlorination facilities. The granular carbon
adsorption cost estimating program calculates the costs of in-
fluent pumping, carbon contactors, regeneration facilities, and
initial carbon required. The nitrification/denitrification cost
estimating program predicts the costs of dispersed floe systems
for the removal of nitrogen from wastewater. A cost estimating
program for wastewater treatment by direct land application
computes the costs of preapplication, distribution, renovated
water recovery, and monitoring facilities. All of these econo-
mic models factor in the costs of yardwork, contingencies,
engineering, land, administration, and interest during construc-
tion.
REQUESTS FOR THE MODELS
The real value of all these computer programs can be mea-
sured by their acceptance and use throughout the sanitary en-
gineering field. These models and related work have experienced
wide attention with many requests for descriptive reports and
source card decks coming from consulting engineering firms,
universities, states, municipalities, equipment manufacturers,
other EPA offices, and various organizations interested in the
simulation, design, and costing of wastewater treatment systems.
Over 6000 copies of literature have been distributed during the
last several years in response to requests. Perhaps a better
measure of the applicability and need for this type of informa-
tion is the fact that these requests have come from 47 of the
50 states and 32 different foreign countries. Much of this
interest can be attributed to the fact that there are very few
sources for complete generalized cost and performance estimating
procedures as applied to preliminary design of wastewater treat-
ment processes.
Unfortunately, it is difficult to get good feedback as to
how much use these computer programs are to the people that have
expressed interest in them. However, enough feedback is
obtained to assure that the work is being actively used in many
areas. There are several universities presently using the
models in their coursework. Many consulting engineering firms
have modified some of the programs to fit their own particular
needs. Area planners have used this work in urban development
efforts. Various research and development literaure in the
field cites this work as reference. EPA itself makes extensive
use of the material.
15
-------
Most of this information can be easily obtained through the
National Technical Information Service (NTIS) or from EPA
directly. Some of the computer programs and appropriate docu-
mentation are available through the Civil Engineering Program
Applications (CEPA) organization.
CONCLUSION
The primary goal of this modeling effort is to improve the
rule-of-thumb or hand calculation method of process design which
is still commonly used today. The principal deterrents to better
process design are usually the manual effort required in comput-
ing the cost and performance of alternative designs and the
labor required to accumulate and correlate the large amount of
expeirmental process design performance data which is often
available. The mathematical computer model can minimize the
computational work required for examining alternative designs,
and, if the model has been correctly developed, it will reflect
the best experimental and scientific information obtainable.
Thus, the process designer has within his grasp the tools for
quantitatively selecting the most cost-effective system of pro-
cesses to achieve any desired wastewater treatment goal. The
Systems and Economic Analysis Section within EPA is very much
interested in promoting the use of computerized design tech-
niques in order to achieve better treatment at a minimum cost.
16
-------
REFERENCES
1. Motard, Lee, Barkley, "Chemical Engineering Simulation
System," (1969).
2. Originally developed at Purdue University and Dartmouth
College and later expanded by the Digital Systems Corpora-
tion of Hanover, New Hampshire.
3. Silveston, "Digital Computer Simulation of Waste Treatment
Plants Using the WATCRAP-PACER System," Water Pollution
Control, 69, No. 6, 686-693, (1970).
4. Hoffman, Woods, Murphy, Norman, "The Strategy and an Ex-
ample of Simulation as Applied to a Petroleum Refinery
Waste Treatment Process," (1973).
5. Curry, "Computer Simulation of a Biological Waste Treat-
ment Facility," (1971).
6. Chen, Fan, Erickson, "Computer Software for Waste Water
Treatment Plant Design", J.W.P.C.F., 44, 746-762, (1972).
7. Fan, Erickson, Chen, "Computer Optimization of Biological
Waste Water Treatment Processes," (1973).
8. B & P Silveston Engineers, "Notes—Workshop on Computer-
Aided Design and Simulation of Waste Treatment Systems,"
(1974) .
9. U. S. Army Crops of Engineers, "Computer-Assisted Procedure
for the Design and Evaluation of Wastewater Treatment
Systems," (1975).
10. Smith, Eilers, "Executive Digital Computer Program for
Preliminary Design of Wastewater Treatment Systems," (1973),
17
-------
CASE I WORKSHOP
USE OF THE EXEC PROGRAM TO DETERMINE
THE EFFECT OF DESIGN CRITERIA SELECTION
ON PLANT COST AND PERFORMANCE
Thomas K. Walsh
Metcalf & Eddy, Incorporated
Boston, Massachusetts 02114
ABSTRACT
Use of the Executive Program to evaluate the impact of
design criteria selection on the cost and performance of a par-
ticular wastewater treatment system is presented. A method is
presented for simplifying data assembly and program execution
where more than one case is to be analyzed for a particular
system. An example problem is presented in which the effects of
the mixed liquor suspended solids concentration selected for
design of conventional activated-sludge systems are evaluated
and four analagous example problems are suggested.
18
-------
INTRODUCTION
The purpose of this case study is to present a method for
using the Executive Program to evaluate the effects of changing
design criteria on the cost and performance of a particular
treatment system, and to demonstrate the effect of design criter-
ia selection on the results obtained using the Executive Program.
It is well recognized that the cost and performance of
wastewater treatment systems can be substantially affected by
the values of the parameters used in their design. Due to the
massive amount of detailed computation involved in evaluating
the effects of design criteria, time and financial constraints
often limit the consideration which can be given to criteria
selection. The Exec Program provides a means by which selected
design criteria may be evaluated rapidly, at a relatively low
cost.
STUDY APPROACH
The first step in evaluation of design criteria using the
Exec Program should be establishment of a basic set of input
data which contain the user's best approximation of the design
parameter values which will result in the desired cost and per-
formance of the system considered. The basic data file (or deck)
may then be modified by copying it, then editing it to reflect
changes in design criteria. Where card decks are used for input
data, the card or cards containing the design parameter(s) to be
changed would be replaced with a new card or cards containing
the revised value of the parameter(s).
Using this method, a number of input data files may be
assembled. Each file will be exactly the same as all of the
others, except for the value(s) of the parameter(s) whose
effects are being investigated. By copying data files, instead
of retyping or punching them for each case, the chances of in-
advertently changing the value of other parameters is reduced,
and less time is required for data assembly.
After assembly of revised files for each value of the para-
meter (s) being changed, the files may be combined and results
obtained for each case during a single execution of the Exec
Program. When this is done, the combined input deck should be
prefaced with a single card on which the number of cases to be
tried is punched in the first two columns. Such cards should
not be used at the beginning of each case. The above approach
was used in this case study.
BASIC INPUT DATA
The basic input data presented in the following paragraphs
was used in this Case Study. Values assigned to most of the
parameters were selected by the author, but in some cases, it
was necessary for IIT personnel to assign a value to a required
parameter. This is because the version of the program with
which the author has been working is not entirely similar to
19
-------
that being used at IIT. The values selected by IIT were based
on recommended program input values.
Treatment System Considered
The example presented in this case study was based on use
of the secondary treatment system shown schematically in
Figure 1. The liquid treatment portions of the system consist
of preliminary treatment, primary settling, a conventional
activatedr-sludge system and chlorination. The sludge-handling
system consists of gravity thickening of combined primary and
waste-^activated sludges followed by vacuum filtration. Sludge
processing side streams are returned to the primary settling
tanks. The input data used to describe this system are shown in
Table 1.
The system was selected for its simplicity and was not in-
tended to describe any particular treatment plant.
Raw Wastewater Characteristics
The values of the parameters which describe raw wastewater
characteristics used for the purpose of this case study, their
Exec Program variable names, assigned stream matrix (SMATX) lo-
cations, and definitions are shown in Table 2. The values shown
are typical of those associated with medium strength domestic
wastewater, as reported in various literature sources (3,4).
A raw wastewater flowrate of 10 mgd was selected in order to
simplify comparison of results. A value of 200 mg/1 was assumed
for both influent BOD5 and suspended solids. Influent BODg was
assumed to be 30 percent suspended and 70 percent volatile and
the remainder fixed. The ratio of BOD5 to organic carbon (both
suspended and dissolved) was assumed to be 1.87, as suggested by
Smith (5). Total influent phosphorus was assumed to be 10 mg/1
of which 8 mg/1 were assumed to be in the dissolved form. Very
little information is available on the nonbiodegradable carbon
content of domestic wastewaters. Values of 15 mg/1 and 3 mg/1
were assumed for settleable and dissolved alkalinity of 150 mg/1
was assumed. Influent dissolved fixed matter was assumed to be
1,000 mg/1.
Basic Design Criteria
The design criteria for the processes and operations
modeled by the Executive Program are defined by the user as
part of the input data for each case considered. The criteria
are stored on the cfomputer in a decision matrix (DMATX) . The
criteria for a particular process or operation are stored in
a single column of the matrix which is defined by the number
assigned to the process or operation by the program user.
20
-------
Q
GO
W
CO
as
o
•<
S
I
CO
CO
g
i
ft,
O
E
21
-------
TABLE 1 INPUT DATA USED TO
DESCRIBE PROCESS FLOW DIAGRAM
CASE STUDY I
K
0
1
0
0
0
0
0
1
0
9
N
1
0
2
3
0
10
11
0
4
0
IP ROC
1
4
2
3
4
8
7
4
12
0
NAME
PREL
MIX
PRSET
AERFS
MIX
THICK
VACF
MIX
CHLOR
END
IS1
1
2
3
4
10
12
13
14
5
0
IS2
0
20
0
0
11
0
0
15
0
0
OS1
2
3
4
5
12
13
24
20
25
0
OS2
0
0
10
11
0
14
15
0
0
0
III
2
1
0
0
0
0
0
1
2
0
22
-------
TABLE 2 CASE STUDY I
BASIC DATA FILE - RAW WASTEWATER CHARACTERISTICS
SYMBOL
Q
SOC
SNBC
SON
SOP
SF'M
SBOD
VSS
TSS
DOC
DNBC
DN
DP
DFM
MATRIX
LOCATION
SMATX(2,1)
SMATX(3,1)
SMATX(4,1)
SMATX (5,1)
SMATX(6,1)
SMATX(7,1)
SMATX (8,1)
SMATX (9,1)
SMATX (10,1)
SMATX (11,1)
SMATX (12,1)
SMATX (13,1)
SMATX (14,1)
SMATX (15,1)
DESCRIPTION
Flowrate for stream 1, mgd
Solid organic carbon content
of stream 1, mg/1
Solid nonbiodegradable carbon
content of stream 1, mg/1
Solid organic nitrogen contert
of stream 1, mg/1
Solid organic phosphorus
content of stream 1, mg/1
Solid fixed matter content of
stream 1, mg/1
Solid BOD^ content of stream 1,
mg/1
Volatile suspended solids
content of stream 1, mg/1
Total suspended solids content
of stream 1, mg/1
Dissolved organic carbon
content of stream 1, mg/1
Dissolved nonbiodegradable car-
bon content of stream 1, mg/1
Dissolved nitrogen content of
stream 1, mg/1
Dissolved phosphorus content of
stream 1, mg/1
Dissolved fixed matter content
ASSIGNED
VALUE
10.0
33.0
15.0
5.0
2.0
60.0
60.0
140.0
200.0
74.0
3.0
25.0
8.0
of stream 1, mg/1 1000.0
23
-------
TABLE 2 (Cont'd)
MATRIX
SYMBOL LOCATION
ALK SMATX(16,1)
DBOD SMATX(17,1)
DESCRIPTION
Alkalinity of stream 1,
mg/1 as CaCO-
Dissolved BODs content
of stream 1, mg/1
NH3 SMATX(18,1) Ammonia-nitrogen, mg/1
N03 SMATX(19,1) Nitrate-nitrogen, mg/1
ASSIGNED
VALUE
150.0
140.0
15.0
0.0
(1)
(1)
1. Assigned by IIT.
24
-------
Values which were selected for the basic design criteria
for the processes and operations shown in Figure 1, their Exec
Program variable names, DMATX locations, and definitions are
shown in Table 3. Reasons for selection of key parameter values
are discussed in the following paragraphs according to the pro-
cess or operation considered.
Stream Mixers. Subroutine Mix does not provide treatment
or involve cost, thus no DMATX input is involved. The function
of this subroutine is computation of the characteristics of com-
bined wastewater (or sludge) streams.
Preliminary Treatment. The preliminary treatment subrou-
tine (PREL)contains only cost functions at this time. Thus,
the only input parameters which are needed are (a) indication of
the type of treatment, and (b) a value for the excess capacity
factor (ECF) to be used in computing the costs of preliminary
treatment.
Because the major cost item of preliminary treatment sys-
tems (Grit removal facilities) are normally provided in dupli-
cate with each unit having the capacity to handle the full
design flow conditions in plants of this size, a value of 2 was
selected for the ECF of this operation.
Primary Settling. Because it is normally expected that
primary settling will accomplish a 50 percent reduction in
suspended solids, a value of 0.5 was selected for FRPS in sub-
routine PRSET.
Primary settling tanks may be operated to achieve varying
degrees of thickening, with underflow solids (or primary sludge)
commonly having a concentration of between 10,000 and 50,000
mg/1. An underflow solids concentration of approximately 35,000
mg/1 was selected, yielding a value of 175 for URPS.
The excess capacity factor selected for settling tanks
should reflect expected peak flow, the number of tanks which
might be installed at facilities of the size investigated, and
the frequency at which peak flows are expected. Normal practice
indicates that an ECF of 1.2 to 1.3 is acceptable. A value of
1.25 was selected for this study.
Activated Sludge System. Subroutine AERFS models the per-
formance of a conventional activated-sludge system consisting
of an aeration tank(s) and a final settling tank(s). The model
assumes operation at steady-state. The values associated with
the required input parameters were selected on this basis and
are individually discussed below:
25
-------
TABLE 3
BASIC DATA FILE
DECISION MATRIX (DMATX) CONSTANTS
CASE STUDY I
SYMBOL
DMATX
LOCATION
DESCRIPTION
ASSIGNED
VALUE
COST CONSTANTS
CCI
WPI
RI
YRS
DHR
PCT
DA
CCINT
XLAB
CKWH
SUBROUTINE
IPREL
ECF
SUBROUTINE
FRPS
URPS
SUBROUTINE
1,20
2,20
3,20
4,20
5,20
6,20
7,20
8,20
9,20
10,20
PREL
1,1
16,1
PRESET
1,2
2,2
PRESET
EPA STP construction cost index
(1957-59 = 1.0)
Wholesale price index for
industrial commodities
(1957-59 = 1.0)
Fractional interest rate
Amortization period, yrs .
Hourly wage rate, $/hr.
Fractional indirect labor cost
Land cost, $/Acre
Fractional interest during
construction
Laboratory requirements
Electrical energy cost, $/KWH
Type preliminary treatment
Excess capacity factor for
preliminary treatment
Desired suspended solids removal
efficiently (fractional)
TSS of OS2/TSS of IS1
2.6
1.926(1)
0.0575(1)
30.0(1)
5.0(1)
0.15(1)
2500. 0(1)
f1\
0.06(1)
1(1)
0.04
1
2.0
0.5
175.0
HPWK
3,2
PSP ECF 15,2
PST ECF 16,2
SUBROUTINE AERFS
BODS
XMLSS
1,3
2,3
Weekly hours of operation at ,
primary sludge pumps 14.0
Excess capacity factor - ,
primary sludge pumps 1.25
Excess capacity factor for
primary settling tank 1.25
Desired secondary effluent
BODS (SBOD + DBOD) , mg/1 25.0
Design aeration tank mixed
liquor suspended solids
level, mg/l 2.000.0
26
-------
TABLE 3 (Cont'd) BASIC DATA FILE
DECISION MATRIX (DMATX) CONSTANTS -
CASE STUDY I
DMATX
SYMBOL LOCATION
DEGC
CAER20
DO
AEFF20
URSS
GSS
HEAD
ALMD
SUBROUTINE
FST ECF
RSP ECF
BL ECF
AT ECF
SUBROUTINE
TRR
TSS14
GTH
GSTH
ECF
SUBROUTINE
3,3
4,3
5,3
6,3
7,3
8,3
9,3
10,3
AERFS
13,3
14,3
15,3
16,3
THICK
1,10
2,10
3,10
4,10
16,10
VACF
ASSIGNED
DESCRIPTION VALUE
Operating temperature of
activated sludge system, deg . c
Rate constant to be used in
sizing the aeration tank ex-
pressed as a fraction of 0.024
(Ib MLSS-day)-l
Operating aeration tank
dissolved oxygen level, mg/1
Fractional oxygen transfer
efficiency of diffused air
system
TSS of OS2/XMLSS
Design Clarifier overflow
rate, gpd/s.f.
TDK on return sludge pumps,
ft.
Alum dose, mg/1 (for phosphorus
removal)
Excess capacity factor for
secondary settling tank(s)
Excess capacity factor for return
sludge pump(s)
Excess capacity factor for
blower (s)
Excess capacity factor for
aeration tank(s)
Fractional solids capture
TSS content of thickened sludge,
mg/1 50
Thickener overflow rate, gpd/sf .
Solids loading rate on thickener,
Ib/day/s.f .
Thickener excess capacity factor
20.0
1.0
2.0
0.06
3.75
750.0
30.0
0.0
1.2
2.0
1.5
1.25
0.90
,000.0
100.0
25.0
1.75
VFL
1,11
Vacuum filter loading, gph/s.f.
27
7.6
-------
TABLE 3 (Cont'd) BASIC DATA FILE
DECISION MATRIX (DMATX) CONSTANTS -
CASE STUDY I
DMATX
SYMBOL LOCATION
HPWK
TSS15
IVACF
FECL3
CAO
SUBROUTINE
CFECL
CCAO
DPOLY
CPOLY
ECF
SUBROUTINE
DCL2
TCL2
CCL2
DS02
CS02
ECF-S02
ECF-CLF
ECF-CCT
2,11
3,11
4,11
5,11
6,11
VACF
7,11
8,11
9,11
10,11
16,11
CHLOR
1,4
2,4
3,4
4,4
5,4
14,4
15,4
16,4
ASSIGNED
DESCRIPTION VALUE
Weekly hours of operation
Expected filtrate solids
concentration, mg/1
Program control, 0 = land fill,
1 = incineration
Ferric chloride dose, Ib/Ton
dry solids
Lime dose, Ib/Ton dry solids
(cont'd)
Ferric chloride cost, $/lb.
Lime cost, $/lb.
Polymer dose, Ib/Ton dry solids
Polymer cost, $/lb.
Excess capacity factor
Chlorine dose, mg/i
Detention time in CCT, min.
Chlorine cost, $/Ton
Sulfur dioxide dose, mg/1
Cost of sulfur dioxide, $/Ton
Excess capacity factor for
sulfur dioxide feed system
Excess capacity factor for
chlorine feed system
Excess capacity factor for
chlorine contact tank
35.0
2000.0
i.od)
42.0
o.od)
0-05
0.0125(1)
o.od)
0.33(D
1.5
8.0
30.0
300.0
0.0
180.0
1.0
1.0
1.0
1. Assigned by IIT
28
-------
1. Desired secondary effluent BODj..
Present EPA standards require that 7 consecutive day
average effluent BOD- from secondary treatment facil-
ities not exceed 30 mg/1. In order to allow a margin
of safety, an effluent BOD,- of 25 mg/1 was selected.
2. Desired operating MLSS concentration.
A nominal MLSS concentration of 2,000 mg/1 was selec-
ted. This value is fairly typical of those used in
design of conventional activated-sludge systems.
3. System operating temperature.
Normally, a value which reflects some worst-case oper-
ating temperature would be selected (lowest 10 year,
etc.). For simplicity, a temperature of 20 C was
selected.
4. Biological reaction rate constant.
In subroutine AERFS, system kinetics have been defined
by use of a version of the Michaelis-Menton equation.
Based on the authors knowledge of subroutine AERFS,
the value of the rate constant at 20°C (CAER20) which
must be supplied as input should be expressed as a
fraction of the value 0.024 (Ib MLSS-day)~1• This value
(0.024) was evidently emperically determined (7) to be
typical of conventional activated-sludge systems treat-
ing domestic wastewater. A value of 1.0 was selected
for CAER20, as suggested by Smith and Eilers (1,2).
5. Operating aeration tank DO level.
Normal design practice utilizes values of 1 to 2 mg/1
for operating DO. A value of 2 mg/1 was selected.
6. Fractional efficiency of aeration equipment at 20 C.
The normal operating efficiency of diffused air aerat-
ion systems varies from 4 to 7 or 8 percent. An ef-
ficiency of 6 percent (0.06) was selected.
7. Secondary sludge solids content.
In subroutine AERFS, secondary sludge solids content
(URSS) is expressed as a fraction of MLSS. Conven-
tional activated-sludge systems may be expected to
produce sludges with a solids content of between 0.5
and 1.0 percent. A value of 0.75 percent (7,500) mg/1
was selected, resulting in a value of 3.75 for URSS.
29
-------
8. Desired secondary settling tank overflow rate.
Design overflow rates for conventional activated-
sludge systems are typically between 600 and 800 gpd/sf
at average flow. A value of 750 was selected.
9. Expected total dynamic head on return sludge pumps.
For the purpose of this case study, the value of 30
feet recommended by Smith and Eilers was selected (2).
10. Alum dose used for phosphorus removal.
Phosphorus removal was not considered herein, thus a
value of 0.0 was used for ALMD.
11. Final settling tank ECF.
For reasons presented in the section on primary set-
tling, a value of 1.2 was selected.
12. Return sludge pump ECF.
On numerous occasions, return sludge pumping capacity
far in excess of that encountered during normal oper-
ation is required. Extended peak flows, pollutant
loads or plant upsets such as bulking sludge can occa-
sion such use. For this reason, an ECF of 2.0 was
selected for the return sludge pumps.
13. Aeration tank ECF.
The excess capacity factor used for aeration tank
sizing should be based on expected performance under
some peak loading condition. For the purposes of this
case study, a value of 1.2 was selected.
Chlorination. Subroutine CHLOR does not consider treat-
ment. Its purpose is to compute the size and costs of the
desired chlorination system. The selected values of the sizing
parameters were based on values commonly used in conventional
chlorination systems.
Gravity Thickening. The values of the design parameters
selected for subroutine THICK are not typical of those normally
used for design. They were selected to demonstrate that compu-
ted process performance is dependent upon input data.
Vacuum Filtration. Values for the design parameters re-
quired by subroutine VACF are typical of those reported in
current EPA literature (6) .
30
-------
EXAMPLE PROBLEM
A question frequently considered during preliminary design
of activated-sludge systems is whether or not costs can be opti-
mized by adjustment of the value selected for aeration tank
mixed liquor suspended solids concentration (MLSS). Although it
is recognized that higher MLSS concentrations will result in re-
duced volume requirements for the secondary reactor, other
portions of the treatment system such as return sludge pumps,
blowers, clarifiers, and the sludge-handling system may also be
affected. The quantitative effects of alternative MLSS levels
on each portion of the system are difficult to evaluate without
first performing a materials balance on the selected system at
a number of MLSS levels.
This example problem demonstrates use of the Exec Program
to evaluate the effects of various MLSS values.
Input Data
In order to establish general trends associated with
changes in MLSS values, five cases were tried. The first case
consisted of the basic data previously described. In the remain-
ing cases, only the value of MLSS was changed. Values selected
for MLSS were 1,600, 1,800, 2,000 (per basic data), 2,200, and
2,400.
Output Results
Pertinent Exec Program output reflecting overall system
performance and cost are shown in Table 4. Output which reflects
the cost, performance and size of the individual processes and
operations is shown in Table 5.
Minimal statistical analysis of the results was performed.
The average value of pertinent results and the range of calcu-
lated values as a percent of the average were calculated and are
also shown in Tables 4 and 5. The later value was calculated
to indicate the relative variability of each of the parameters
listed in the tables.
Overall System Output
As shown in Table 4, Exec Program output values for system
cost and performance are not substantially affected by changes
in MLSS alone. Sludge production and general characteristics
were the same for each value of MLSS used. Total effluent BOD
is the same for all cases (as would be expected since it was set
as part of program input). The type of effluent BOD is slightly
affected by the selected MLSS value, lower MLSS values being
associated with higher effluent suspended BOD values and lower
effluent dissolved BOD values. This is further reflected in the
31
-------
TABLE 4 EXAMPLE PROBLEM RESULTS - CASE STUDY I
COMPUTED VALUES OF MAJOR SYSTEM PARAMETERS AT VARIOUS MLSS VALUES
. , . Range as
MLSS(mg/i)
f xu\ru*iij j. Jj x\
1600
1800
2000
2200
2400 Average
average (1)
STREAM 24
(sludge production)
TSS ,%
TSS, Ib/day 18
VSS ,% TSS
STREAM 25
(system effluent)
Q, mgd
SBOD, mg/i
DBOD, mg/1
TBOD, mg/1
TSS, mg/1
28
,700
70
9.99
9.2
15.7
24.9
19.9
28
18,700
70
9.99
8.8
16.1
24.9
19.1
28
18,700 18
70
9.99
8.4
16.5
24.9
18.2
28
,700 18
70
9.99
8.1
16.8
24.9
17.6
28 28
,700 18,700
70 70
9.99 9.99
7.8 8.5
17.2 16.5
25.0 24.9
17.0 18.4
0.0
0.0
0.0
0.0
16.0
9.0
0.4
15.8
STREAM 20
(principal recycle)
Q,mgd
TBOD , Ib/day
TSS, Ib/day
SYSTEM COSTS
(C/1000 gal.)
TAMM
TOPER
TOTAL
0.25
1,123
2,712
10.7
9.8
20.5
0.22
1,112
2,692
10.6
9.8
20.4
0.20
1,126
2,732
10.5
9.8
20.3
0.19
1,142
2,775
10.4
9.8
20.2
0.17 0.21
1,100 1,121
2,677 2,718
10.3 10.5
9.8 9.8
20.1 20.3
38.8
3.7
3.6
3.8
0.0
2.0
1. Max.-Min. (100)
Avg.
32
-------
M
Jx
O W
DO:
EH W
CO E-t
w
W £
w 2
rtj 05
U ''C
a,
1 CO
CO W
CO CO D
t-H* U 3
3 O >
CO IX
w o. co
0*4 CO
OJ4 ^
1C O £
W r"3
J «£ CO
03 S 0
O O
a, o os
rtj
W CO >
iJ U
d. 3 EH
3S ^H" ri!
3 "*
w
Q
W
• EH
m a
0.
W 5,
J O
CQ CJ
EH
VI.
C
n
<0 4-
c
01 0
t^ c
C I-
Si
^^
,_(
\
Cn
E
CO
CO
JP
os
w
EH
CU
1
•^4
ft.
1 ".
) r-
11
. O
) n)
v.
1 01
) >
I It
01
O
o r-i r- en o
r*i r~ vo f
o » - o
rH TT O\
rH
fr) ^* CO CO ro C3
O iH CO r-l O •
f> • r- r^ • ^3*
0 - * 0
r-^ ^* O\
i— 1
p") vo ^J* <3\ ro o
O iH CN) O O •
n • (**• r^1 * **j*
0^*0
r-t *f O\
i-H
^ oo in ^* ro o
o rH vo r^ o •
m • i^- ^o • ^*
o ^ ^ o
r—l *T ^
rH
^* cn co r^- ro o
o rH cn ON o •
CO • 00 VO • ^1*
o - *• o
rH TJ» cn
rH
in CM ^r r^- co o
o CM ro cn o •
co • r^ vo • ^
0 • - 0
rH T cn
rH
(A
•d
>l -H
10 >l r-l
T) n) O
\ -o in co
cx rH ja »o ** 2
o> 73 rH rj> w
vr E - <
» g TT 0
S Q •* - -H
O ^3* CQ CO rH CO D
(X O* EH EH O EH CO
O rH vo O O CM
rH rH CM o r~ cn
•o1 m co
rH cn rH o in vo
vo en ^t . on oo
• 1" . VO O rH
rH » CO rH O •
CO . O
o
rH CO
ro in in o rH co
• TT TC . r~ in
rH » . VO O rH
00 CO rH O •
O
O
co in
^ p^
• ^«
rH ^ ^* O O f**
CO TT • 00 VO
VO O rH
CO rH 0
O
0
00 •*
in cn CM o CM co
• *3* ^3* • cn co
rH • • VO O rH
OO CO r-l O •
O
o
in in
f^ CM cn o vo CM
• in co • o o
rH » . VO rH CM
OO CO rH O •
0
0
r- i-
cn ^ vo o in vo
• in co •
• CP rH g
« W g .
PH t3 K to *•
w •-* * cn w I-H
t> 03 O 3 X O
rH
O
^*
in
r~
•
o
o
cn
o
CM
CO
•
o
in
r-~
•
o
p*.
VO
.
o
o
vo
•
o
Ul
T3
•rl
rH
o
Ul
^
^
rH
rH
CO
CO
EH
vo
rH
in
00
CO
*
rH
r-l
in
^t
t
rH
rH
rH
VO
*)•
»
rH
rH
CM
cn
CO
•.
rH
rH
o
vo
CO
^
rH
r-H
VO
VO
CM
rH
rH
>1
TJ
^
v
,—f
,_(
CO
EH
O
O
CO
r~
CO
r*»
00
r~
CO
^-
00
r-
CO
r~
W
CO
EH
^
r-H
f—t
W
*>
33
-------
rH
Q
CD
EH
CO
W
CO
O CO
i a
EH
tn w
EH S
CO <
W 04
04
CO CO
S to W
W W D
^ O 1-3
CQ O *Jl
o 04 >
K Q-
a, to
a; tn
W O J
i-l *~3 S
a< <
X GK O
W O H
04
CO <
• W >
-~ D
TJ J EH
CD < <
3 >
C
•r| Q
-U W
A fc!
0 D
u a,
o
0
3
a
^j
B
H-l— •
0-
01 ~
m -U fl
S t
<0
04 at
<
t
r
<
<;
0
o
£
^^
Oio
B o
x— ' CN
to
o
CO
i-H
O
0
VO
rH
04
u
EH
U
3£
tf
04
^
a.
vo 0
T rH CO Tl- rH O O
ro .....
rH CM CM O O
co vo vo rH r- CM a\
en co in ^ en ro CM
. . ... . CO
o ro o o o rH
r-
CM
ro ^* r^ I-H co CM o>
co co in *r en ro CM
. ... . CO
o ro o o o rH *•
r-
CM
vo rH r~ CM en
en in in ^r en ro CM
00 CO ... . CO
> • O O O rH
o ro r~
CM
vo vo vo rH r- CM en
en oo in TJ* en ro CM
. ... . co
O ro O O O rH »
r-
CM
vo r~ vo o vo CM en
o oo in ^r en ro CM
. . ... . co
iH ro O O O rH -
CM
c~ co vo o vo CM en
rH co in ^> en ro IN
. • ... . CO
rH CO 000 rH
CM
*"•*
•3 rH
4-» CT> OS
C CO W
8 § § i2
~ 0 & EH
rH rViOEH£HWEH .^
0 ^x O4fH Fn £HCOCOCOCOP4l4HCJ
i^OQCOO4LOr^COOO O O »H
14 ^-* PHOUO^^^^r^CJ ^O33
ii ttfrj^MQcJ^EH^EHSH EH
^COKEHOPHr^ Z CJih
EH M ij rH1 M J .
oqco<5 3 c" fc< «ora
DO D > S
to O co pq co
en ro rH o
ro co CM 0
rO rH
CO CM VO rH
vo *3* ^ .
r^ • o in
» o
ro o
I1 in VO rH
at <)> <* •
rH 0 in
o
ro o
in i— vo rH
CM ** *r
•a1 • o in
o
ro o
ro en vo I-H
r~ • o in
0
ro o
CO rH m rH
ro in ^ •
o • o in
- o •
rH Tl- m rH
r- in "31
T . o in
- o •
•<* o
.
rH
01 01
TJ
0 -H
0 rH
0 0
rH 0]
HH \
• "O- 'O ^
W &i
e
En ro
S CO - rH
X O ro CO
EH u rH co
< EH O EH
O
O
]
n
•H
S
1
X
id
E
•
rH
34
-------
1—4
>H
Q
g
W
W
W
U W
a
1 U
CO U
r3 K
O 2
W
•a j E-I
a «
3 >
c
•H Q
•M W
C E-i
0 D
rj d,
— £
O
in o
W
S
E-i
VH ^-~
O i-H
tn — '
IB -P 0>
fl) 0) 10
CJ> O IH
C rl (D
m ai >
« ft 10
(V
n)
Q)
>
O
o
CM
O
o
CN
CN
^
rH
\ O
6 o
x^ CM
OT
CO
O
0
CO
rH
O
O
VO
rH
W
1
<
o TT rn
n o i— <
T
o vo co
r» r» o
rH 00 I-H
• *
O CM
f- VD (N
n I- CN
rH CO rH
• ^
0 CM
O VO CO
in r- rH
rH CO rH
• *
O CM
C^ CA ^
VO r- rH
H 00 H
• ^
O CM
in in co
co r^ c^>
rH CO O
• *
O CM
o vo in
I-H r» en
CM CO 0
• *
O CM
•a
4J
c >,
0 nj >i
U irj Id
> \
8 -o H 3 &
si1* > g
EH ,_!•<>•
», _T Q H
n ^ o w 03
D rH (Q w D
W O EH EH w
O O rH
O rH r^
CM r~ CM
f^, VO VO
^* •
in
I-H
CM Tt »r
r- r- vo
*y •
» in
rH
CM i-H CO
r~ r- vo
^* •
• in
1-1
CM CO CM
r— vo vo
^ •
» in
rH
CN n rH
r~ vo vo
^r •
in
rH
CN a\ o
r- in vo
- in
rH
i-H
n*
10
IT
O
o
o
r- {
VH o
U) •.
dP EH
CO
fc O
P< > W
i < EH
, ,
O
O
rH
>_^
C
iH •
s &>
1 >
• <
s
-1
35
-------
higher effluent suspended solids values calculated for cases in
which lower MLSS values are used.
The calculated flowrate of the principal system recycle
stream is extremely dependent on the selected MLSS value, higher
flowrates being associated with lower MLSS values. However, the
quantity of BOD and suspended solids carried in the stream are
relatively unaffected by the selected value of MLSS, indicating
that a more dilute recycle stream is produced at lower MLSS
values.
The costs calculated for the system are also relatively
unaffected by the value selected for MLSS. Amortization (thus
capital) costs generally decrease with increasing MLSS values-,
reflecting a smaller aeration tank volume (Table 5, VAER).
Operating costs are calculated to be equal for any selected MLSS
value. Due to the slight decrease in amortization costs, total
treatment costs are also computed to decrease slightly with
increasing values of MLSS.
Output for Processes and Operations
The costs of two processes are not affected by changes in
the value of MLSS. These are preliminary treatment and chlorina-
tion. This would be expected since neither process is within
the system recycle loop, thus should not be affected by the
value selected for MLSS.
Primary Settling Tank. The size and performance of the
primary settling tanks is slightly affected by changes in the
valueselected for MLSS, but does not vary in any direct manner
with MLSS. The affects of the value selected for MLSS on its
cost and performance should be minimal as the impact of various
MLSS values is buffered by the sludge-handling system.
Activated Sludge System. The secondary treatment system
is affected in a number of major ways by the selected value of
MLSS. As would be expected, the required aeration tank volume
(VAER) decreases with increasing MLSS values. It is interesting
to note that the computed value of VAER is not directly related
to the selected value of MLSS, all of the computer values equate
to use of a value of approximately 0.56 for F/MLSS, where F is
influent BOD. This indicates that kinetics are not affected by
the value selected for MLSS, as would be expected within the
range of MLSS values selected.
Blower size is related to BOD load (FOOD) and to the mass
of active solids carried in the secondary reactor (MLASS), thus
it varies slightly and inversely with the value of MLSS selected,
The return sludge flow rate (QR) was computed to be rela-
tively unaffected by selected MLSS values. The reasons for this
36
-------
unexpected result are related to the predicted underflow solids
concentration and are discussed in the following paragraphs.
Because the computed value for the area of the final set-
tling tank (AFS) is related to secondary effluent flowrate, the
computed value of AFS is not affected by the selected value of
MLSS.
The parameter XRSS is the ratio of the concentration of
solid material (MLASS, MLBSS, MLDSS, MLISS, MLNBSS) in the set-
tling tank effluent to MLSS, such that:
SS4 = SS * XRSS (1)
where: SS4 = concentration of solids class in
system effluent, mg/1
SS = concentration of solids class in
aeration tank, mg/1
This factor is used in AERFS to model the performance of the
settling tank as a clarifier and is computed using the following
emperical equation:
XRSS = 556.1 * GSS ** 0.49421/
XMLSS ** 1.8165/(24.0 * TA) ** 0.4386 (2)
where: GSS = settling tank overflow rate, gpd/sf
TA = aeration tank detention time, days.
This ratio is applied to all classes of solids carried in the
system to determine the characteristics of the system effluent.
By inspection of equation 2, it can be seen that XRSS is
inversely proportional to MLSS. For this reason, system ef-
fluent solids are computed to decrease with increasing values of
MLSS. This explains why computed SBOD and TSS values for stream
25 are inversely related to the selected MLSS value.
The computed solids concentration of stream 11 (TSS 11,
% solids) is directly proportional to the selected value of
MLSS and its computed flow rate (Qll) is inversely proportional.
However, the solids content (TSS11, Ib/day) of the stream is
relatively unaffected by selected MLSS values. The slightly
increased solids content of the stream at higher MLSS values is
a reflection of predicted enhanced clarification performance of
the secondary settling tank (XRSS) at higher MLSS levels.
The predicted changes in the solids concentration of stream
11 were the result of using a fixed value for URSS (DMATX(7,N)).
37
-------
In subroutine AERFS, the concentration of secondary sludge is
directly related to MLSS according to the following equation:
TSS11 = XMLSS * URSS (3)
where: TSS11 = the suspended solids concentration in
stream 11, mg/1
URSS = ratio of solids concentration in the under-
flow from the final clarifier to aeration
tank MLSS.
Thus, changes to the value selected for MLSS will result in
changes in the flowrate and solids concentration of the waste
sludge stream if the value of URSS is not similarly altered.
The above relationship explains the previously observed
lack of variation in return sludge flowrates. Because secondary
sludge solids concentration is directly proportional to MLSS in
the system as modeled, the return sludge flowrate should be
nearly constant, as computed.
The costs computed for the various secondary treatment sys-
tem components were observed to be related to their computed
sizes, as shown in Table 5.
Gravity Thickener. The computed values for the size and
cost of the gravity thickener are inversely proportional to
selected MLSS values. Because thickener underflow solids
concentration (TSS13) was specified as part of the input data
(DMATX (2,N)), it is the same in all cases. Since the quantity
of solids fed to the thickener (TSS10 and TSS11 in Ib/day) are
nearly the same for all MLSS values and because the underflow
solids concentration is held constant, the flowrate of the
thickener underflow stream (Q13) varies only slightly with
changes in the value selected for MLSS.
The flowrate of the thickener overflow stream (Q14) varies
inversely with selected MLSS values. Since the flowrates of the
primary sludge and thickened sludge streams (Q10 and Q13, res-
pectively) are nearly constant, the major cause of this variation
observed in secondary sludge flowrate (Qll).
The size of the gravity thickener is computed based on
overflow rate (GTH = DMATX (3,N)), or solids loading rate
GSTH = DMATX (4,N)), whichever produces the largest surface area.
In the cases studied, overflow rate governed due to the low in-
put value selected and computed thickener size varied in pro-
portion to the computed flowrate of stream 12 (the thickener
influent stream). This is because the influent (instead of the
effluent) flowrate is used in the program as a basis for thick-
ener sizing.
38
-------
It should be noted that the computed thickener size, cost
and performance are a reflection of the input data which was
used. If the thickener were actually sized using the data pre-
sented in Table 2, it is doubtful that a solids capture of 90
percent and an underflow solids concentration of 5.1 percent
would be achievable.
Vacuum Filter. The computed size, cost and performance
of the vacuum filter is governed by thickener performance and
was computed to vary only slightly with selected MLSS values
and in proportion to the computed flow rate of thickened sludge
(Q13).
Preliminary Conclusions
It is difficult to evaluate the effect of various MLSS
levels on the cost and performance of the system investigated
using the above output. Because only the value of MLSS was
changed as part of program input, the computed values for secon-
dary sludge solids concentrations varied in direct proportion
to the selected MLSS value. This caused the computed values of
certain key parameters (Q6, TSS11, TSS20, ATHM, etc.) to differ
from what would normally be expected.
Although the thickening performance of the secondary set-
tling tank can be expected to vary with solids loading rate
(thus, MLSS concentration), the expected variation within the
range of MLSS values investigated should be slight and would
most likely be difficult to detect in full-scale operations.
The computed results were obtained because the input value for
URSS was held constant at 3.75 for all input cases.
Revised Input Data
Because of the above relationship, the value selected for
URSS must be reconsidered each time the value selected for MLSS
is changed, otherwise secondary sludge solids content may not
be modeled as desired. This may be done by estimating an accep-
table value for secondary sludge solids concentration (based on
experience, the results of pilot studies, etc.), then calculat-
ing a new value for URSS using the following equation:
URSS = TSS11/XMLSS (4)
TO demonstrate the dependence of program results on the
value selected for URSS, a second set of five cases was tried.
In these cases, both the value of MLSS and the value of URSS
were changed. The values used for MLSS in the second set of
cases were the same as those used in the first. Using equation
4, new values of URSS were calculated for each MLSS level. In
each case, a value of 7,500 mg/1 was assumed for TSS11.
39
-------
The values used for MLSS and URSS in the second set of
five cases are shown in Table 6.
TABLE 6. VALUES USED FOR MLSS AND URSS,
SECOND SET OF CASES
MLSS, mg/1 URSS
1,600 4.69
1,800 4.17
2,000 3.75
2,200 3.41
2,400 3.12
Revised Output
Output obtained using the revised input shown in Table 6
is summarized in Tables 7 and 8. For comparative purposes, the
same parameters listed in Tables 4 and 5 are shown and the same
format is used.
By comparison of the results shown in the four tables,
those parameters whose value is essentially unaffected by
changes in the value selected for URSS were noted. These para-
meters were:
1. The flow rate and characteristics of streams 10, 13,
24, and 25.
2. The size and cost of the preliminary treatment system;
the aeration tank, blowers and final settling tank;
and the chlorination system.
3. The size, cost and performance of the vacuum filters.
4. The quantity and characteristics of secondary sludge
solids and the quantity of BOD and TSS in the principal
recycle stream.
5. The computed values for XRSS.
The major effects of the revised input were in computed
values for:
1. Principal recycle (S20) , return sludge, and waste
sludge (Sll) flowrates (Q20,QR, andQll, respectively).
40
-------
TABLE 7. EXAMPLE PROBLEM RESULTS - CASE STUDY I
COMPUTED VALUES OF MAJOR SYSTEM PARA-
METERS AT VARIOUS MLSS AND URSS VALUES
PARAMETER
MLSS (mg/l) - URSS(n.d.)
1600,
4.69
STREAM 24
(Sludge production)
TSS ,%
28
TSS , Ib/day 18,700
VSS,% TSS
STREAM 25
(System effluent)
0, mgd 9
SHOD, mg/i 9
DBOD, mg/l 15
TBOD, mg/l 25
TSS, mg/ 1 19
STREAM 20
(Principal recycle)
Q, mgd 0
TBOD, Ib/day 1
TSS, Ib/day 2
SYSTEM COSTS
(C/1000 gal.)
TAMM 10
TOPER 9
TOTAL 20
70
.99
.2
.8
.0
.9
.20
,120
,719
.6
.7
.3
1800,
4.17
28
18,700
70
9.99
8.8
16.2
25.0
19.0
0.20
1,124
2,729
10.6
9.8
20.4
2000,
3.75
28
18,700
70
9.99
8.4
16.5
24.9
18.2
0.20
1,126
2,732
10.5
9.8
20.3
2200, 2400,
3.41 3.12 A'
28
18,700 18
70
9.99
8.1
16. 8
24.9
17.6
0.20
1,128
2,736
10.5
9.9
20.4
28
,700 18
70
9.99
7.9
17.1
25.0
17.0
0.20
1,132
2,739
10.5
10.0
20.5
Range as
percent of
verage average (1)
28
,700
70
9.99
8.5
16.5
25.0
18.3
0.20
1,126
2,731
10.5
9.8
20.4
0.0
0.0
0.0
0.0
15.3
7.9
0.4
15.8
0.0
1.1
0.7
0.9
3.0
1.0
1. Max.-Min. (100)
Avg.
41
-------
^M
Q QH
3 W
H H
CO U
U <
to w en
dl < W
CJ 0, D
I en 2
to >
co a
E-i U co
.J O cn
D « eu
en a. o
2 05 Q
O Z
g f~3 i^
>J S CO
D3 CO
O CM t-}
« 0 £
co en
u u a
J D O
0* l-J H
g > <
M O
W f-<
EH
04 a 10
(U
cr
IB
M
V
0 (N
O rH
T •
M m
*->. V
. O rH
•a o •»
• «p*i^»^t
rH CM O - - O
rH »» OI
vo r-'J'r-rHrHroo
n M o rH vo t- o •
• •r^>»r~^D«*3*
•H cn o - - o
rH f CTl
rH
vo r^^*r*r-HrHroo
n H >, -H
0 Itf >, H
o w T) - S " ^
H Ei •* 0
CO . » S Q Tf . rH
pa o ffl co a, - o co o co co
DODftOi'mtorHenD
toE^toSaiOiEHEHaEHco
o o oo o
rH r-H 1-x O
^* in
rH CM VO O
10 CM ^r •
• ^* • vo
rH n rH
00
rH US O O
n in in •
• *f UJ
rH *T rH
CO
d r~ ro o
*r r- 01 •
• *f • \O
H » CO rH
CO
CO T (N O
in 01 T
rH - fO rH
00
n- n TT o
r- m a "<
o
o
rH
.
c
•H •
g Cn
1 >
X
(0
g
rH
42
-------
H
^4
Q
D
EH
en
£t] CO
en K
ft. n
U E-i
1 §
en S en
EH < W
jj 64 O
D " h-3
en tn
2 u
cj en
S O en
W « «
|J f44 D
a
O K Q
« O z
01 5
td S en
iJ en
S O §
0) Q
3 W EH
C EH <
•H a
•P O4
s s
0 0
o u
CO
3
EH
en
0)
tn
c
«
«
—
•
c
•^r-
cn
U3
D
fc
. — „
,_4
x^
CT
a
^"•*
en
en
o
s
<4H -.
OrH
4J 0)
C tr
0) 1)
U M
(H V
0) ^
am
0)
CJ1
<0
l-(
OJ
>
rtj
*
O O4
0 ^t
^* •
CM CO
••
O rH
0 "W
CM •
CM ro
o in
o r-
o •
cs ro
^
o r^
0 rH
CO .
rH ^"
fc
O CTl
O *&
kO •
rH *1*
ff
fr
t
d
o:
04
OOO^**O \£> O OOCA
f~CMOrHO •* rH TCnrH
inrMinooo oo ID r-i-noo
c^ oo f^ oo r*- cj^ oo m ^3* o^
0 rH . PI . . ...
o«o* o ro ooo
0 rH
0 «-H
rH^invooo ro 10 oono^
f** oo i** ^* r"~ oo oo t"* ^* rH
0 rH . 1- . . ...
O • O » O rO OOrH
0 rH
O rH
OcomcftCO cjs vo romoo
oooor-r-Hr~ oo oo vofo
OrH.^T . . ..«
O • O » O rO OOrH
0 rH
0 rH
CMrOinCMCO VO U3 VDrHr^
o>cor-c^r~- cr\ oo insert
o • o • o ro ooo
O rH
0 rH
vorHini'co vo r^ or-r-
ooot— inr- o oo inrooo
rHrH.rO . . ...
O«O- H ro OOO
0 rH
0 rH
inrHinrHoo r-- oo inrooo
rH rH . CM . • ...
O"O* rH CO OOO
. O rH
O rH
^
'O ^
fd
T, en DI
V- ^ en
n -H >, o 04
R rH n) en o g
" O T! en o D
• eo \ EH rH CU
.."a A X.C*EH EH OEHEH
r * fO dP rH jp o Q (/j p^ t/^ jy C-( yj ^n
Jv* ^"* ^ """^ E^ O W O CD CO O O
rj £•*-- ^USCJQOOO
*^CO*-rHHrHgHM J J
— sCOrHu^Ulu^u]^ m en
§p5rHcneneno
g X 0 EH EH > 0
0
o
*_,
c
•H
s
X
(S
rH
43
-------
H
>H
Q
D
EH
CO
W CO
CO PH
< a
U EH
fy
' «
CO 2 CO
EH fti W
iJ d. D
D J
to co <
U CO >
K Ct]
U co
SOto
W Ctt PS
j 0, D
m
O (S a
os o z
CU r} rfj
W Z CO
i-} to
eu c« J
SOS
X CO CO
W W D
•~1 f~\
_-J O
I--4 M
• rfj OH
„ »-> rf*
^fS
a) a
3 W EH
C EH <
•rl O
si
u o
00
M
. 1
t— 1
9
EH
14H .—
OrH
(0 ~-
Id 4J
(U (D id
01 O rl
C M CD
ifl 0> >
os a id
a)
a
id
rl
(1)
>
<
0 CM
0 rH
T •
CM CO
^ O rH
•a o •»
t CM •
C CM CO
CO
to
K »
D o in
o r-
-» o •
*— » CM CO
<;
td
£
o r-
to o rH
CO 00 •
J rH ^J*
s
o en
o VD
t^> •
rH •*
K
M
1
2
04
0 C3
0 0
CM cn
CO CN
00
rH -
r-
CM
CM en
CO CM
. oo
rH
c-
CM
CM
CO CM
. 00
H »
P-
CM
CM en
CO CM
03
rH
r-
CM
CM a\
CO CM
. oo
i-H »
r—
V
JJ «
C w
o 3
0 EH
~ EH
• w « • «
CO CO EH O MH U
P>) tO Hi • H
a I g s \ s
H J
g (n mom
D D > D
to co m w
00 O rH O *** CO O
rH O CM O CN rH H
co o\ vo rH 10 r- o
O «!• T IO r- rH
r- . o m rH oo rH
o
CO O O CM
en en «3 rH co ro o
CO *T *T • VD CO CN
r^ . o in rH co -H
» o •
CO O O CM
r^- en vi> rH r^ en r^
rH <* «3" • v£) P^ rH
r- . o in rH co rH
0
CO O O CN
ro en VD rH r*- en ^*
rH ^T »» VO f-~ rH
r- . o in rH oo rH
^ 0
CO O O CN
CM Cn VD rH *J* rH CO
r~ ^r ^r • VD r^ en
VD • O m rH CO O
« o • -
CO O O CN
'r cn in H in CM rH
r- rr T vo r~ o
VO . o in rH CO rH
- o • • '•>
CO o O CM
rH
id
CP in
•Xj ^,
0 -H Ifl >1
O rH T3 Id
O O X, 73
rH Ot £> ^
VH X rH XI CK
•o- -a *> HO
01 ~ O! » <
e •* - >
» EH CO H T
S CO » rH Q rH
EBOfito'^otom
EHUrHCQrHpgcOD
•JJEHOEHOEHEHCO
0 0 rH
O rH r~
CM r- CM
r~ vo VD
•w
» in
rH
CM T «T
r~ r- VD
*T
^ in
rH
CS rH CO
i — r*~ vo
rr
» in
rH
CM CO CM
r~ VD vo
•q-
- in
i-H
IN T rH
r~ vo VD
•q-
" in
rH
CM en o
r~ in VD
«r
» in
rH
rH
id
tn
o
0
o
rH
X
HH O
ta
dP EH
CO
&4 O
Oi > U
S < EH
0
o
rH
C
•H
S
1
X
id
S
.
rH
44
-------
2. Solids concentration of the waste sludge stream
(TSS11, % solids).
3. Sludge pumping costs.
4. The cost, performance and size of the gravity thickener,
The differences associated with the values of the above para-
meters caused the computed size, performance and cost of primary
settling tank to be leveled out such that they were found to be
unaffected by selected MLSS values.
Overall system costs were affected only slightly by the
revised input. However, considerably different trends in
total system costs were noted.
Sludge Flow Rates and Pumping Costs. As shown in Figures
2 and 3, the value selected for URSS has a significant effect on
the values computed for return and waste sludge flowrates. The
original data indicated that the waste sludge flowrate is depen-
dent on the selected value of MLSS and that the return sludge
flowrate is only slightly affected. Use of the revised input
data produces results which are exactly the opposite. Wasting
rates are nearly constant for any MLSS value, and return rates
are directly proportional to the selected MLSS value. This is
also reflected in sludge pumping costs. Instead of being nearly
the same for any MLSS value (Table 5), the revised output indi-
cates that they are directly proportional to the selected value
of MLSS (Table 8) .
Gravity Thickening. As shown in Figures 4 and 5, results
obtained using the original input indicated both gravity thick-
ener size and the expected rate of flow of its overflow stream
(Stream 14) were inversely related to MLSS. Results using the
revised input indicate that the values of these parameters have
very little dependence on the value selected for MLSS. As with
the original input, thickener size, cost and performance are a
reflection of the input data used.
Total Treatment Costs. Results obtained for total treat-
ment costs using the original and revised data are shown in
Figure 6. The results obtained using the original data indicate
that total treatment cost will be reduced by increasing the
value selected for MLSS, that total operating costs are not
affected by the value selected for MLSS; and that capital costs
(as reflected in the value of TAMM) will be reduced by increas-
ing MLSS.
The results obtained using the revised input indicate
trends which are opposite in all respects to those described
above. These results indicate that total treatment costs are
generally increased for increased MLSS values; that total
45
-------
USING ORIGINAL INPUT
USING REVISED INPUT
1600
1800
2000
MLSS (mg/L)
2200
2400
FIG. 2 WASTE ACTIVATED SLUDGE FLOWRATE VS. MLSS - CASE STUDY I
USING REVISED INPUT
USING ORIGINAL INPUT
1600
1800
2000
MLSS (mg/L)
2200
2400
FIG. 3 RETURN SLUDGE FLOWRATE VS. MLSS - CASE STUDY I
46
-------
0.25
USING ORIGINAL INPUT
USING REVISED INPUT
0.10
1600
1800
2000
MLSS (mg/L)
2200
2400
FIG. 4 FLOWRATE OF THICKENER OVERFLOW (Q14) VS. MLSS -
CASE STUDY I
USING ORIGINAL INPUT
USING REVISED INPUT
1600
1800
2000
MLSS (mg/L)
2200
2400
FIG. 5 THICKENER AREA (ATHM) VS. MLSS -
CASE STUDY I
47
-------
21.0
20.5
20.0
1600
1800
2000
MLSS (mg/L)
2200
2400
(3
8
IT
1-
a- 10.0-
1600
1800
2000
MLSS (mg/L)
2200
2400
11.0
10.0
1600
1800
2000
MLSS (mg/L)
2200
2400
® RESULTS FOR ORIGINAL DATA <•> RESULTS FOR REVISED DATA
FIG. 6 TOTAL TREATMENT COSTS VS. MLSS - CASE STUDY I
48
-------
operating costs react similarly; and that capital costs are
relatively unaffected.
Two major factors associated with the value selected for
URSS combined to cause the opposing results. These factors
were:
1. The operating costs associated with return sludge
pumping are nearly equal using the original input,
using the revised input they increase with increasing
MLSS values.
2. The capital costs for gravity thickening were inversely
related to selected MLSS values using the original
input, using the revised input they are nearly equal.
CONCLUSIONS
A discussion as to the accuracy of the results obtained
using the Exec Program is considered beyond the scope of this
presentation. Metcalf & Eddy has recently started to analyze
the capabilities of the Exec Program and is contemplating a
number of revisions which reflect our experience with process
performance and cost.
It is the author's opinion that the following conclusions
may be drawn from the example problem:
1. That the Exec Program is capable of rapidly performing
the detailed computations necessary to evaluate the
effects of changing design criteria.
2. That a decision to alter the value selected for a par-
ticular design parameter should be checked against its
impact on the values assigned to other input parameters.
3. That the results obtained using the Exec Program should
be checked to assure that they are reasonable and may
be practically achieved. As is the case in using any
computer program, the results obtained are only as good
as the data provided and the individual who must use
them.
RELATED PROBLEMS
Using the basic system and input data described in the
example problem, any number of related problems might be consid-
ered. The following problems would produce results which may
be compared to those obtained as part of the example problem:
1. The effects of correcting thickener sizing and perfor-
mance parameters.
49
-------
2. The effects of changing the value of wastewater tempera-
ture (DEGC = DMAXT (3,N)).
3. The effects of changing the value of the rate constant
used for aeration tank sizing (CAER20 = DMATX (4,N)).
An analagous problem not directly related to the example
problem, but which would produce interesting results would be
varying the expected primary settling tank suspended solids
removal ratio (FRPS = DMATX (1,N)).
50
-------
REFERENCES
1. Smith, R. and Eilers, R. G.; Executive Digital Computer
Program for Preliminary Design of Wastewater Treatment
Systems - Documentation, U. S. Dept. of the Interior,
FWQA, 1970.
2. Smith, R. and Eilers, R. G.; Updates to Executive Digital
Computer Program for Preliminary Design of Wastewater
Treatment Systems, USEPA, 1973.
3. USEPA Technology Transfer; Process Design Manual for
Nitrogen Control, October 1975.
4. Metcalf & Eddy, Inc.; Wastewater Engineering; Collection,
Treatment, Disposal, McGraw-Hill, 1972.
5. Smith, R.; Preliminary Design and Simulation of Conventional
Wastewater Renovation Systems Using the Digital Computer,
U. S. Dept. of the Interior, FWQA, 1960.
6. USEPA Technology Transfer; Process Design Manual for Sludge
Treatment and Disposal, October 1974.
51
-------
CASE II WORKSHOP
USE OF THE EXEC PROGRAM TO COMPARE THE
COST AND PERFORMANCE OF MULTIPLE FLOW SCHEMES
Richard G. Eilers
U. S. Environmental Protection Agency
26 W. St. Clair Street
Cincinnati, Ohio 45268
ABSTRACT
There are many alternative sludge handling methods available
for the treatment and disposal of municipal sewage sludge. The
Exec Program can be used as a tool by the design engineer to
evaluate the cost and performance of alternate sludge handling
schemes in order to determine the most cost-effective system.
Here, the Exec Program is used to simulate four different sludge
handling methods for 1, 10, and 100 mgd plant sizes. The purpose
of this exercise is to determine the most economical design for
each of the plant sizes under consideration.
52
-------
INTRODUCTION
The Exec Program cannot be used for extremely detailed de-
sign purposes, but it can be a valuable preliminary design tool
for the consulting engineer. The performance of many proposed
wastewater treatment systems can be simulated along with pro-
viding cost estimates for building and operating these plants.
The cost data generated by the Exec Program is sufficient for
preliminary design purposes. Construction cost (in dollars),
amortization cost, operation and maintenance cost, and total
treatment cost (all in cents per 1,000 gallons of wastewater
treated) are calculated individually for each unit process, and
a sum total of each of these costs is given for the entire
system. Capital cost is also computed by adding onto construc-
tion expenses the costs of yardwork, land, engineering, adminis-
tration, and interest during construction. All of the cost in-
formation can be updated or backdated with respect to time by
means of cost indices that are supplied as input to the program.
Using the Exec Program, it is possible to optimize a particular
treatment system by varying design parameters and noting the
effect on performance and cost. Cost-effectiveness studies can
be made by comparing alternate treatment systems. Initial
studies along these lines are becoming of increasing importance
because of the soaring costs of plant construction that are now
being experienced.
Several years ago,the Exec Program was used to investigate
the potential economic advantages associated with 261 different
methods for treating and disposing of sewage sludge. This work
is fully described in the article entitled, "Computer Evaluation
of Sludge Handling and Disposal Costs" by Robert Smith and
Richard G. Eilers, which was published in the Proceedings of the
1975 National Conference on Municipal Sludge Management and
Disposal. Sludge production and the costs of constructing and
operating each of the various systems were computed. Each system
was either primary or activated sludge treatment or both followed
by some combination of the following 12 sludge handling
processes--lime stabilization, gravity thickening, air flotation
thickening, single-stage anaerobic digestion, two-stage anaerobic
digestion, aerobic digestion, elutriation, vacuum filtration,
centrifugation, sludge drying beds, multiple hearth incineration,
and land disposal of liquid sludge. The outcome of the study
showed that the cost (in January 1974 dollars per ton of dry
solids processed) for treating and disposing of sewage sludge
ranges from about $30 per ton for anaerobic digestion followed
by dewatering on sand drying beds to over $100 per ton when the
sludge is dewatered by vacuum filtration or centrifugation and
then incinerated. Treatment and disposal of sludges produced in
municipal wastewater treatment plants were shown to account for
as much as 60 percent or as little as 20 percent of the total
cost of treatment. Therefore, careful consideration should be
given to selecting the sludge handling method which meets the
53
-------
site-specific constraints at a minimum cost. The Exec Program,
which is capable of examining the cost and performance of a wide
variety of alternative sludge handling schemes, can be used as a
management tool to narrow the range of options when design con-
ditions are known.
The example problem and the assigned problem for this work-
shop will combine to be a greatly simplified application of the
study just described. Instead of 261 different methods for
sludge handling, only four will be considered. These four sys-
tems will, however, make use of most of the unit processes that
were used in the large study.
This exercise will examine four different sludge handling
schemes for 1, 10, and 100 mgd size plants. The object of the
study will be to determine the most economical design for the
various plant sizes under consideration. The liquid handling
phase of each of the four designs will be the same and consist
of the following unit processes: raw wastewater pumping (RWP),
preliminary treatment (PREL), primary sedimentation (PRSET), and
chlorination (CHLOR). The sludge handling phase of the four
designs will be as follows: System (1) - gravity thickening
(THICK), lime stabilization (LIME), sludge holding tanks (SHT),
and land disposal of liquid sludge (LANDD); System (2) - gravity
thickening (THICK), anaerobic digestion (DIG), sludge holding
tanks (SHT), vacuum filtration (VACF), and multiple hearth in-
cineration; System (3) - gravity thickening (THICK), anaerobic
digestion (DIG), sludge holding tanks (SHT), centrifugation
(CENT), and multiple hearth incineration (MHINC); System (4) -
gravity thickening (THICK), anaerobic digestion (DIG), and sand
drying beds (SEEDS). It will also be necessary to use the stream
mixer (MIX) and stream splitter (SPLIT) processes in drawing up
the system configurations.
In solving the problem it will not be necessary to modify or
augment any of the existing subroutines in the Exec Program. The
user will be required to draw up his own system configuration
and prepare all necessary input data to simulate his design on
the Exec Program.
EXAMPLE PROBLEM
In order to give guidance to those participants that are
relatively unfamiliar with computer applications work or the
Exec Program itself, System (1), (2), and (3) will be solved in
detailed by the lecturer. In doing this, it will not be neces-
sary to change any input or output requirements, nor will any
program modification be necessary.
Figures 1, 2, and 3 give the process diagrams for Systems
(1), (2), and (3). Arbitrary stream numbers and process numbers
have been assigned as indicated on the diagrams. The influent
54
-------
K> OOO«L«~»OI.OOOOO
o »* *+
«o ooaco«-tooocc
CO
o
a
O.
e- e- ^ o a
fc> i-* O t OO
g
a)
4J
CO
I
(0
CO
CO
0)
o
o
J^ OOOOOOOOOOO>
55
-------
CTi
«o
O
cc
O
CM
to
. OO«HOOOITOOO
»-• c* v »*
CO
o
a
a. in
i- it
n O
u a
Z O
^ oooooooooooo«r
(N
-P
(0
>i
CO
I
•H
•O
en
CO
(1)
O
O
M
O4
CN
0)
56
-------
CT>
IG
o
.-i CM
cs
to ooo'oo.ioooirooo
»i CN *^ «^
i- i- *.
U •-» U
a
O
c w c.
UJ
tj
o
a
2 ~+
:»: ooooooooooooi
tn
to
CO
0)
o
o
CO
a)
•H
ELI
57
-------
stream vector, containing the flow and contaminant concentra-
tions that are to be used, is listed in Table 1. The design
variables (DMATX input) for the liquid handling phase of each
system are given in Table 2. These will be the same for each of
the four systems being investigated. Table 3 contains the design
variables that are to be used for the sludge handling phase of
each system. Note that, for simplicity, all excess capacity
factors (ECF) have been set equal to 1.0. Table 4 gives the cost
inputs that are necessary to bring all cost calculations up-to-
date. This will result in all computed cost figures being in
March 1977 dollars, and thereby provide a timely cost comparison
of the four alternative systems. Table 5 lists the punched card
data for Systems (1), (2), and (3) with an influent flow of
1 mgd.
The purpose of this case study is to indicate to the par-
ticipants how they can use the Exec Program to evaluate several
alternate design solutions for a specific treatment goal. This
is the type of problem that frequently confronts the treatment
plant designer, although the problem may take many different
forms and specify different kinds of requirements. Here, the
problem is to determine the lowest cost system, but the same
type of analysis can be used to determine the best performance
system where appropriate. Note that the performance of all four
of these systems under consideration will be the same, because
each one uses the same liquid handling scheme.
ASSIGNED PROBLEM
The participants are to draw up the system configuration of
unit processes along with all connecting streams for System (4).
The process and stream numbers may be arbitrarily assigned. In-
put values for the influent stream characteristics, cost cons-
tants, and process decision variables should be the same as
those used for Systems (1), (2), and (3). Note that values for
the decision variables associated wi'th the sand drying beds
unit process are also given in Table 3. Input data cards to the
program should be prepared based on the system configuration and
the decision variables associated with the unit processes that
are to be used. Once the set of data cards has been prepared
and double checked to eliminate any possible key punching errors,
the program should be run at 1, 10, and 100 mgd flows in order to
generate the desired costs for use in completing the cost com-
parison of the four alternate systems.
After all the test cases have been successfully run on the
Exec Program, the total treatment cost (cents per 1,000 gallons,
March 1977 dollars) of the four systems should be as follows:
58
-------
FORTRAN
Variable Name
Table 1
INFLUENT STREAM VECTOR
Parameter Definition
Influent Value
SMATX (1,1)
SMATX(2,I)
SMATX(3,I)
SMATX (4,1)
SMATX (5, I)
SMATX (6, I)
SMATX (7, I)
SMATX (8,1)
SMATX (9,1)
SMATXUO,!)
SMATX (11, I)
SMATX (12, I)
SMATX (13, I)
SMATX (14, I)
SMATX(15,I)
SMATX (16, I)
SMATX(17,I)
SMATX (18, I)
SMATX (19, I)
I
Q
SOC
SNBC
SON
SOP
SFM
SBOD
VSS
TSS
DOC
DNBC
DN
DP
DFM
ALK
DBOD
NH3
N03
stream number
volume flow, mgd 1.,
solid organic carbon, mg/1
solid nonbiodegradable carbon,
mg/1
solid organic nitrogen, mg/1
solid organic phosphorus, mg/1
solid fixed matter, mg/1
solid 5-day BOD, mg/1
volatile suspended solids, mg/1
total suspended solids, mg/1
dissolved organic carbon, mg/1
dissolved nonbiodegradable
carbon, mg/1
dissolved nitrogen, mg/1
dissolved phosphorus, mg/1
dissolved fixed matter, mg/1
alkalinity, mg/1
dissolved 5-day BOD, mg/1
ammonia nitrogen as N, mg/1
nitrate as N, mg/1
-
10., 100.
105.
30.
10.
2.
30.
140.
224.
254
43.
11.
19.
4.
500.
250.
60.
15.
0.
59
-------
Table 2
LIQUID HANDLING PHASE
DESIGN VARIABLES
Raw Wastewater Pumping (RWP)
DMATXC 1,N) HEAD 30.
DMATX(16,N) ECF 1.
Preliminary Treatment (PREL)
DMATXC 1,N) IPREL 1.
DMATX(16,N) ECF 1.
Primary Sedimentation (PRSET)
DMATXC 1,N) FRPS .5
DMATX( 2,N) URPS 400.
DMATXC 3,N) HPWK 14.
DMATX(15,N) ECF 1.
DMATXC16,N) ECF 1.
Chlorination-Dechlorination CCHLOR)
DMATXC 1,N) DCL2 8.
DMATXC 2,N) TCL2 30.
DMATXC 3,N) CCL2 220.
DMATXC 4,N) DS02 2.5
DMATXC 5,N) CS02 180.
DMATX(14,N) ECF 1.
DMATXC15,N) ECF 1.
DMATXC16,N) ECF 1.
60
-------
Table 3
SLUDGE HANDLING PHASE
DESIGN VARIABLES
Gravity Thickening (THICK)
DMATX( 1,N) TRR .95
Vacuum Filtration (VACF)
DMATXC 1,N) VFL
4.9
DMATXC 2,N) TSS 50,000-
DMATXC 3,N) GTH 700-
DMATXC 4,N) GSTH 8.
DMATX(16,N) ECF 1.
Lime Addition to Sludge (LIME)
DMATXC 1,N) DLIME 200.
DMATXC 2,N) CLIME 25.
DMATXC 16, N) ECF 1.
Sludge Holding Tanks (SHT)
DMATXC 1,N) TD 15.
DMATXC 16, N) ECF 1.
Land Disposal of Liquid Sludge
(LANDD)
DMATXC 1,N) TAYR 15.
DMATXC 2,N) SP .25
DMATXC 3,N) DISP 10.
DMATXC 4,N) TS 1200.
DMATXC 5,N) YRSL 6.
DMATX(15,N) ECF 1.
DMATXC 16, N) ECF 1.
Single Stage Anaerobic Digestion
DMATXC 1,N) TC 15.
DMATXC 2,N) TCIG 30.
DMATXC 16, N) ECF 1.
Sand Drying Beds CSBEDS)
DMATXC 1,N) SOUT .35
DMATXC 2,N) TSS 50.
DMATXC 16, N) ECF 1.
DMATX ( 2 , N )
DMATX C 3 , N )
DMATXC 4,N)
DMATXC 5,N)
DMATXC 6,N)
DMATX ( 7 , N )
DMATXC 8,N)
DMATXC 9,N)
DMATXC 10, N)
DMATXC 16, N)
HPWK
TSS
IVACF
FECL3
CAO
CFECL
CCAO
DPOLY
CPOLY
ECF
35.
200 .
1.
42.
0.
.064
.0125
0.
.33
1.
Multiple Hearth Incineration
(MHINC)
DMATXC 1,N)
DMATXC 2,N)
DMATXC 3,N)
DMATXC 4,N)
DMATXC 5,N)
DMATXC 6,N)
DMATXC 7,N)
DMATXC 8,N)
DMATX ( 9 , N )
DMATXC 16, N)
(DIG)
Centrif ugation
DMATX ( 1,N)
DMATX ( 2,N)
DMATXC 3,N)
DMATXC 4,N)
DMATXC 5,N)
DMATXC 6,N)
DMATXC 7,N)
DMATXC 8,N)
DMATXC 16, N)
ML
NINC
HPWK
SPER
WV
HV 10
TYPE
FC
CNG
ECF
(CENT)
CRR
TSS 200
HPWK
XCEN
POLY
CPOLY
GPMN
CNMIN
ECF
2.
1.
35.
5.
0.
,000 .
1.
. 30
.97
1.
.95
,000.
35.
1.
2.
2.
100.
2.
1.
61
-------
Table "4
DESIGN VARIABLES FOR COSTS
Cost Input
DMATXC 1,20) CCI 2.709
DMATXC 2,20) WPI 1.916
DMATXC 3,20) RI .06
DMATXC 4,20) YRS 25.
DMATXC 5,20) DHR 5.65
DMATX) 6,20) PCT .15
DMATXC 7,20) DA 2000.
DMATXC 8,20) CCINT .06
DMATXC 9,20) XLAB 0.
DMATXC10,20) CKWH .03
62
-------
• • vD
•* O O
C«i «O •
CM
O •
«i IN —i
a o c o
< *»• If) O
SE ~*i CM O
CM
O
O
u
2
1-1 • • un
X c o -*
3t «•> O •
c
o
c
o
e
0)
4J
(Q
c oc
3; • •
«C IN •*
IT,
c c
cv
CV
ir
X • • -
H c cr if!
*-f «-> CV
r> *
fr-: ~* .
2 *"•"
b.
s;
t- ^.
< a ii-
ii- 3 a
fir* • • • O^ • QC • Cl>
(-1 »i •* IT C C C
IT •*" f- \T) >* r*
>* CM • .-
a CM
e
o
o- c
• wH
c
CV
CM
CV
a
a
O
C
o
if
t-i a. ax *- x
5 CL in «;F-lf C» «3»
• cr jr. IT,
Tf' CM ICO&* ^C«-i xC**11
C1 ro o ^* \T> *&
C C O O C C
O
O
c
c
o
2
X »-<
»- X
a.
C
,4
X
CM o-
o o
^
c o
tvj
63
-------
V} • •
v> •C
CS,
in
CNi
o •
I
o - •
oc c c
< •* in
Z. «-i CV
c
o
o
«£>
c
O -i
in c
•-< o ••-> o
c
o
c
o
CM
o
(V
<-* CM
£X in tr> O -4 O
o * c/> •
U-
«c at:
u. 3 a
CC •••C/* • OC • CX
IH —« •* IT O O O
>* (N • •-•
a CM
M oc ax i-f x
xam ic t- m c* 10
• c^ m
^* c^ m oc • ^ «"* vc
CM •» CS
CM • <*\ •* «-.
ITS •».
fi •
u.
u
x M
a.
c
• o
IM
or
o m
o o
o <*
o c
m
o
o o
«K
o o
CM
»H
C
-------
c o
« IN
o c c
I * in o
c» «-• CM c
a 01
c
z
C C ~«
o •
c ac
o o
SO O
IT
g
4J
CO
CO
I
(0
-p
(0
Q
T)
CM
CM
in
CN
I- c i
>• CM • »-i
a CM
•a
z «
a
a c
c
o
o
o
o
IT
a c
CM
CM
.j
a
(u
»- a
(X
c
ax
O\ C IT IT OC
mw« fio *»-' CM-- cv
CM CM »- CV »-t
cc
co
tr
o
o
o
CM
rt
EH
65
-------
Size, mgd System (1) System (2) System (3) System (4)
1 38.1 55.9 62.8 39.8
10 15.0 18.3 17.5 12.9
100 9.7 10.2 8.7 8.1
Complete summary costs for the four systems are listed in Table
6. These costs are taken from the cost summary page (the last
page) of the computer printouts for each system. From this
analysis, it can be seen that System (1) is the lowest in cost
at 1 mgd, and System (4) is the lowest in cost at 10 and 100 mgd.
However, System (3) is very competitive at the 100 mgd level.
In practice, Systems (2) and (3) are usually much more desirable
for larger plants, since Systems (1) and (4) require considerT-
able land areas for plants larger than 10 mgd. Quite often, con-
venient large land parcels are simply not available, especially
in metropolitan areas. For various reasons, such as this one,
it is not always possible to choose the least cost solution even
when it can be accurately determined.
The outcome of this analysis can, of course, be greatly
altered by changing some of the various decision variables asso-
ciated with the sludge handling processes. However, based on
the assumptions that were made, the cost figures that have been
calculated can be assumed to be useful enough for preliminary
design applications. This type of information is of considerable
value to the planner when he is evaluating several options for
solving a specific problem. The principal deterrents to better
system design are usually the manual effort required in computing
the cost and performance of alternative designs and the labor
required to accumulate and correlate the large amount of experi-
mental process design data which is often available. With the
Exec Program, the process designer has within his grasp a tool
for quantitatively selecting the most cost-effective system of
processes to achieve a desired treatment goal. Analysis of
this type leads to obtaining better treatment at a minimum cost.
66
-------
Table 6
TOTAL PLANT COSTS
System (1):
Total Capital Cost, 0
Total Amortization Cost,
<7l,000 gallons
Total OEM Cost,
t/1,000 gallons
Total Treatment Cost,
t/1000 gallons
1 mgd
792,964
17.521
20.611
38.132
Plant Size
10 mgd
2,758,343
6.116
8.851
14.966
100 mgd
14,047,269
3.193
6.496
9.689
System (2):
Total Capital Cost, $ 1,546,321
Total Amortization Cost,
$/l,000 gallons 33.141
Total O&M Cost,
t/1,000 gallons 22.784
Total Treatment Cost,
t/1,000 gallons 55.924
5,113,099
10.958
7.388
18.347
27,346,522
5.861
4. 358
10.219
System (3):
Total Capital Cost, $ 1,566,541
Total Amortization Cost,
t/1,000 gallons 39.668
Total 0£M Cost,
t/1,000 gallons 23.103
Total Treatment Cost,
t/1,000 gallons 62.771
4,258,936
10.196
7.340
17.536
19,163,204
4.371
4.338
8. 708
System (4):
Total Capital Cost, $ 1,024,567
Total Amortization Cost,
$/l,000 gallons 21.959
Total 0£M Cost,
{/I,000 gallons 17.846
Total Treatment Cost,
t/1,000 gallons 39.805
3,175,571
6.806
6.058
12.863
18,360,488
3.935
4.195
8.130
67
-------
CASE III WORKSHOP;
USE OF THE EXEC PROGRAM TO DETERMINE THE
EFFECT OF ECONOMIC PARAMETERS ON CAPITAL
AND 0/M COSTS FOR A GIVEN FACILITY DESIGN
Raymond J. Avendt
Consoer Townsend & Associates,
Consulting Engineers
360 E. Grand Avenue
Chicago, Illinois 60611
ABSTRACT
An integral part of the Executive Digital Computer Program
for Preliminary Design of Wastewater Treatment Systems is the
determination of capital, amortization and operation and main-
tenance costs. The costs calculated for a given facility are
based on various relationships contained in the COST and indivi-
dual process subroutines. The costs data inputs and outputs are
discussed according to explicit, implicit or omitted relation^-
ships within the program. Information is presented to aid in
the selection and evaluation of these costs data. An illustra-
tive problem is included to demonstrate the effects of costs
data on the proper use of the program.
68
-------
INTRODUCTION
The majority of users of the Executive Digital Computer
Program are interested in the program's capabilities to
generate cost data. The cost figures may be used in Facilities
Planning by consulting engineers, planners, and regulatory
agencies to determine the most cost-effective wastewater manage-
ment alternative. The operation and maintenance costs data may
be used by operating personnel and designers to compare either
existing or projected costs with typical values. Individual
process costs may be evaluated in order to determine the most
cost sensitive unit processes and may indicate where redesign
or a change in operating parameters is warranted.
Utilization of the program for costs data requires however,
a knowledge of the costs functions and a judicious selection of
data input values. The cost functions contained in the program
are either explicit or implicit depending on the process or
cost subroutine. The cost functions were developed based on
actual construction or operation and maintenance costs corre-
lated to some design or operation parameter(s). The cost func-
tions were further optimized to reflect variations in local
conditions, cost indices, labor rates and material costs, etc.
EXPLICIT COST CONSIDERATIONS
Primarily, the cost considerations within the Exec Program
are explicit. Therefore, either in the COST subroutine or the
individual process subroutines, data input values are required to
generate costs.
The COST subroutine contains the majority of the explicit
cost considerations. The individual cost data required are made
part of the program as data inputs to a decision matrix (DMATX).
A total of ten (10) inputs parameters to the COST subroutine
are required. These inputs are discussed below:
DMATX(1,20),CCI- In order to account for the variations of
construction cost with time, the sewage treatment plant
construction cost index value is required. The historical
value however is trended to reflect a national average.
Recently the USEPA has started to publish values for twenty
(20) cities in order to correct for variations in regional
construction costs. In the COST subroutine the CCI value is
first ratioed to the base value of 1957-59. Then this ratio
is multiplied times the capital costs calculated for each
process to update the costs to the time reference of the
input value. For printout purposes, the ratio is then
multiplied by the base value and the original input value
is printed in the costs display.
69
-------
DMATX(2,20),WPI- The wholesale price index for industrial
commodities is required to account for the cost variations
of materials and supplies used in operation and maintenance.
Again a base of 1957-59 is used. The value is ratioed to
the base value and the ratio multiplied times the calculated
operation and maintenance costs.
DMATX(3,20), RI- In order to project amortized costs, the
amortization interest rate is fed into the program, ex-
pressed as a fraction. In accordance with the guidelines
for Facilities Planning the rate as issued periodically by
the Water Resources Council in the Federal Register should
be used. The rate is internally used in the COST subroutine
to calculate the amortization factor which is subsequently
used to generate amortized costs for each unit process.
DMATX(4,20),YRS- An amortization period, expressed in years,
is required to generate amortized costs. In Facilities
Planning the period is generally not less than twenty(20)
years.
DMATX(5,20), DHR- The wastewater treatment plant personnel
hourly wage rate is required as an input value. The wage
is expressed as $/hr. This value fluctuates with region
and is typically around $5. This value is used to calcu-
late operation and maintenance costs for each unit process.
DMATX(6,20), PCT- The fraction of direct labor costs that is
charged as indirect labor cost is required to determine
actual total labor cost. Typically this value is approxi-
mately 0.2 to 0.3 depending on the location and labor con-
tracts. This value is used to adjust the operation and
maintenance costs.
DMATX(7,20), DA- The program calculates the amount of land
required for a given treatment facility. The total costs
for the plant include the cost of the required land. The
cost of land represented as $/acre is made a data input.
Typically this value is $l,500/acre. Extreme variations in
this value are quite common depending on the locality. The
cost does not include any improvements.
DMATX(8,20), CCINT- The interest rate during construction,
expressed as a fraction, is required to calculate total
capital costs. This value is used to calculate the cost
of borrowed capital to finance the total capital cost, yard-
work, land, engineering, legal and administrative expenses
during the construction period. This value currently runs
between 0.08 to 0.12.
70
-------
DMATX(9,20), XLAB- The COST subroutine contains an equation
to calculate the cost of maintaining a laboratory facility
as a function of the treatment plant capacity. A value of
1.0 is used for activated sludge plants or a value of zero
is used for primary or trickling filter plants.
DMATX(10,20), CKWH- The cost of electrical power expressed
as $/kilowatt hour is a required data input to calculate
operation and maintenance costs. Typical values run be-
tween $0.01 to $0.04 depending on the locality.
The above ten(10)data inputs to the cost subroutine are
used to calculate the components of total capital, amortized
and operation and maintenance costs. These cost figures are
displayed as a part of the output (OMATX). A total of sixteen(16)
output parameters are generated by the COST subroutines. The
program user should have some knowledge of the means used to
calculate the output values in order to assess the validity of
the output. These outputs are discussed below:
OMATX(1,20), RATIO- The multiplier used to factor into
individual unit processes construction costs for yardwork,
land, engineering, legal and fiscal, and interest during
construction Is printed as a ratio. This value should be
between 1.25 and 1.45.
OMATX(2,20), TCAP- This number is the total capital cost of
the entire treatment system excluding yardwork, land,
engineering, legal, fiscal and interest during construction.
This value is expressed in dollars.
OMATX(3,20), YARD- The total capital cost of yardwork ex-
pressed in dollars. This value is calculated as 14 percent
of the TCAP.
OMATX(4,20), TCC- Subtotal of TCAP and YARD expressed in
dollars.
OMATX(5,20), XLAND- The cost of land required for the
treatment plant. The value is a function of land cost and
plant flow.
OMATX(6,20), ENG- The cost of engineering services for
plant construction is expressed in dollars. It is calcu-
lated as a decreasing function of the TCC.
OMATX(7,20), SUBT1- The subtotal of TCAP + YARD + XLAND +
ENG expressed in dollars.
OMATX(8,20), FISC- The cost of legal, fiscal and adminis-
trative services during construction is expressed in dol-
lars. This cost is calculated as a decreasing function of
71
-------
SUBTl.
OMATX(9,20), SUBT2- The subtotal of SUBTl and FISC is
expressed in dollars.
OMATX(10,20), XINT- The cost of interest during construc-
tion is displayed in dollars. It is calculated as a func-
tion of SUBT2.
OMATX(11,20), ACRE- This value is the total land require-
ment for the plant, in acres.
OMAXTX(12,20), AF- This amortization factor is used in
calculating amortized costs. The value should be between
0.07 and 0.12.
OMATX(17,20), TOT- The total capital cost of the entire
plant is presented in dollars. This represents the sum of
SUBT2 and XINT.
OMATX(18,20), TAMM- The total amortization cost of the en-
tire system, in cents per 1,000 gallons is represented by
this value.
OMATX(19,20), TOPER- The total operation and maintenance
cost of the entire system, in cents per 1000 gallons.
OMATX(20,20), TOTAL- The total treatment cost (TAMM + TOPER)
of the entire system, in cents per 1000 gallons.
The remaining explicit cost considerations within the Exec
Program are required data inputs in the process subroutines.
Examples include:
VACF - DMATX(7,N), CFE- The cost of adding iron, expressed
in dollars per pound. Typical values run approximately 0.1.
VACF - DMATX(8,N), CCAO- The cost of adding alum, expressed
in dollars per pound. An approximate value is 0.2.
TFLOT - DMATX(7,N), CPOLY- The cost of polymer, in dollars
per pound. A typical value is 1.0.
MHINC - DMATX(8,N), FL- The cost of fuel oil in dollars
per gallon. A typical value is 0.45.
MHINC - DMATX(9,N), CNG- The cost of natural gas in dollars
per 1000 cubic feet. A typical value is 2.50.
IMPLICIT COST CONSIDERATIONS
Within the various process subroutines are a few subtle
72
-------
cost considerations which can drastically affect the treatment
facility costs. These implicit cost considerations present the
greatest drawback in using the Exec Program. If the user is un-
aware of their importance and selects the values in a haphazard
manner, the costs data are meaningless.
The excess capacity factor (ECF) is of extreme importance
in using the Exec Program for cost estimating. In most
cases, the cost equation is a function of some process parameter
multiplied by the ECF. An accurate value for the ECF is a re-
quirement for valid output data. Although Eilers and Smith do
not elaborate on the use of the ECF, the following should pro-
vide the program user with sufficient data to more accurately
assign a value to ECF in each individual unit process sub-
routine. It is to be noted that the cost equations for operating
and maintenance do not include the ECF.
RWP, CCOST = f (AP * ECF);
where QP = 1.78 * QISI ** 0.92
PREL, CCOST = f (QIS1 * ECF)
PRSET, CCOST = f (APS); settler
Q , * 1000.
APS = (( ^0 ) * ECF)
PRESET, CCOST = f (PGPM); sludge pumps
Q - * 116,666.7
PGPM = ( ( Ub^ mrjv ) * ECF)
AERFS , CCOST = f (VAER) ; aerator
VAER = f (ECF)
AERFS, CCOST = f (BSIZE) : blower
BSIZE = f (ARCFD * ECF)
AERFS, CCOST = f (QR) ; sludge pumps
QR includes ECF
AERFS, CCOST = f (APS); final settler
QS01 *
AFS - (( GSS - ) * ECF)
73
-------
FACF, CCOST = f (AVF) ;
TSS * Q * 58.13
AVF = ( - • - - • — ~) * ECF
{ FVF * HPWK ' **-*
THICK, CCOST - f (ATHM) ;
Q
ATHM = °S1) * 106 * ECF
(j in
ELUT, CCOST = f (AE) ;
AE includes ECF
TRFS, CCOST = f (VOL); filter
VOL = FAREA * DEPTH * ECF
TRFS, CCOST - f (AFS) ; final settler
Q * 1000
AFS = (-^ - > * ECF
TRFS, CCOST = f (QIgl * 1.5 * ECF); sludge pumps
CHLOR, CCOST = f (QJS1 * DCL2 * 8.33 * ECF); feed system
Q * TCL2
CHLOR, CCOST = f (.. ^ * -, AQ * ECF) '• contact basin
1.44 * /.4b
SHT, CCOST = f (VSHT);
OP * TD * 1000
VSHT = gF ? ^ ±^i * ECF
SLP, CCOST = f (QP * ECF)
The other implicit cost considerations include the use of
QP, PGPM and HPWK. The QP value represents peak flow. This
value is used in various subroutines to calculate cost data. QP
is calculated as 1.78 times Q raised to the 0.92 power. This
value may not represent the actual design condition and generate
erroneous data. The PGPM input is used to establish the firm
pumping capacity in various subroutines. The program user should
be aware of the actual firm pumping capacity required and the
associated costs, both of which are influenced by the ECF. The
HPWK input is the hours per week that the sludge pumps are
operated. If too low a value is selected by the user, the pump-
ing capacity and costs will be inordinately high.
74
-------
OMITTED COST CONSIDERATION
The user of the Exec Program should also be aware of various
cost considerations not included with the program. These
omitted cost considerations include the effects of design con-
servatism, sophistication of instrumentation, fail-safe design,
aesthetics and specific site and soil conditions. Idiosyncrasies
in local conditions, variations in wastewater characteristics
and numerous other variables will significantly affect costs for
specific plants. The costs presented in the program are there-
fore not intended to be precise, but for the purpose of comparing
alternative treatment systems which are capable of achieving com-
parable effluent water quality. The user should compare the
actual costs of existing wastewater treatment plants with Exec
Program costs based on the actual design parameters. This com-
parison will allow the user to determine the ability of the
program to respond to specific design parameters.
The Appendices illustrate application of the previo.us
material to a real world problem.
75
-------
~JL §i of
f\ S^e
t/1
Li-
<-o:
i j i
•a:
O
C3 r—
£ ^-.
_. en
SSE "*
~ fooD
• •"* J
oj *-» r^
r-^ co
n ^
• vo
*
f^.
k
s- A a
/ « \
/ Z \
- Q | T-*
C C3 i— 1
° *^-
-H * £ E m '
-P -S^-
irt • !— '
|" 0 r-<
M CM
•H
C
X O
H O
§ e
»
1
1—
«•>£
a:
Q_
WO) ^
PH i-H
«1
s:
0
o
=O >fl •
0 PJOg
TJ 1 • '
i±
H-
O-—
t3 ~^
^S g>
O E
• O
•-< oo co
OJ CJ
• *r
10
- fJ
^^
M <:
CJ
~z.
2:
"-7
^-i1 \S/
~S8 V
00 Y
"o o 1
ui — < T
fc •« A I
i<33 O / \ I — 1
Q| /£\ I u.
(
0)
fH ro i i
&
X
m
\ «>y
-
01
e o
— -
o i** s:
—i O Ol^
i— 4 o rn *!• •—
3 O
(U
+J
O
76
-------
OQ
X
H
Q
W
ft
ft
rij
Configuration
4-»
c
(^
i
ft
1 1
^j
a
i i
3
O
rti
W
I— ^
o
0
^-J
ft
0)
^^
p-i
y*
B
z
or
o
ae
a.
u
« UJ
»- 1-
^ ^
u a
UJ Z
x o
Ul U
4
V-
•»•
'9
t>4
a
REATMENT SYSTEMS
K
X
Ul
^*
4
Jt
cr ui
o t-
u.
^
ae
u.
o
2
O
»*f
^)
Ul
o
X-
a:
z
1-4
z
UJ
•x.
Q.
ON AGENCY
RCH CENTER
71
IX
UJ
X _J
H- *^
« Ul
5 •
>- 0
mi a
Ul ff
CD <
0 X
(TO
1-4
ae
t-t
13
Ul
i-
o
ce.
a.
i
4
Z
Ul
z
o
a:
^
Z
UJ
•
0
4
S!
UJ
a
^
t-
z
z
o
a:
^
z
uJ
_J
X
o
•_•
§
z
ISION
RT BRANCH
S SECTION
>
1-4
X
u
X
<
UJ
Jj
K
UJ
4
y
Ul
1-
l/l
JB
3
a.
^
i/i
V-
Z
Ul
a
-i
Ul
>
UJ
a
v
0
o
_l
o
z
u
aJ
»-•
(A
J
4
•z.
*
O
y
Z
O
w
Ul
o
z
4
in
z
UJ
l/»
^
l/>
CO
o o r~ o J) o o -o a 3 o a
z
ui
o »-
< »- u ae
U. O U. Z O
UJ
O
o
X
a.
77
-------
e e
o o
o o
CM
in o
» o
CM
ae
•jj
UJ
a o
o o
o o
o o
o o
in o
o o
o o
o o
CM rvt
o o
o o
o o
a* o
in
0»
•o
3 O
o o
o o
M M
O O
o o
0 O
CO O
-o
X]
O 3
O O
O O
CM fM
O 3
O 3
r- o
in a
333
333
333
•M >J M
333
O O 3
3 3 f>
^ 3 r-
o
O
01
30
.n
t>
•o
78
-------
>• o o
< o o
a 10 o ro o
§• o •
m z
Ulkl •*
an
Z1-1
oct
-JO.
x
SS
«2
Z_l
o o
o o
l/)0 l/l O
1/1*1/1*
> CO I- O
-IZ
o o
o o
ao o o
«. o • o •
I/I O O EC O
§ i/i in Q m
^
0 0
o a
no in o
x • o •
Zin 2
o o
o o
in o in o
1/1*1/1*
> oo H a
-o *
-< co t- o
»«
» Q CO
•O •
co o co &
a o
o o
fO 0 ro 0
I • O •
210 2
in r-
1/1 o in o
i/i • t/i •
> *»- o
co CM
* in
rw cO
o a> o a
o »o •
co * co o>
VIN O *
a o
o o
10 o ro o
X . o •
Zin z
CO r- *
cO CM
m
o
ro
§O> Q o
• (
cc m
• O •
o *
•71
U
a ac
uz
zu
_J Z
o o
>o
o
o
TL 0
I/I O
n
0
o
z o
o o
m
o
o
z o
l/l 0
o
o
Z 0
O 0
a
in
0
0
ZO
VI O
o
o
ZO
O 0
a
in
«4 O
*4 O
ZO Z 0
VI O OO
irt 3
in
ro
Z 0
vi in
CM
0
0
z o
a o
3
in
» o
Z >O Z 0
vi m o o
CM o
m
u
ac
x
z
LJ
<£
h-
m
o o
o o
a. o a. o
o • a •
J1 d- aa
o a
o o
ZO ZO
o • a •
i/i o m
O O
O O
O CD •J O
OQ • cD *
ZO Z O
i/i in Q m
o o
o o
a. o a. o
o • o •
i/i *
O O
O O
ZO Z O
O • Q •
i/i o m
.•M M
O O
o o
u o u o
co • co •
Z o H o
1/1 in a m
CM *
0> O
a. o a. o
o • a •
30
00 CO
•o o
za> z o
O • Q •
m cr> in
~H O
* O
(J J> O O
CQ • CO •
za» z o
in * Q in
CO *
a> o
a. a. o
o • Q •
a)
\0 GO
O o
Z 0> Z o
O • Q •
in O1 m
CM
10 *
CO lO
Q. O> 0. O
O • O •
i/1 M ao
CO O
-H r^
Z-< Z O
o • a •
in o m
-« "M
Q
X
H
Q
04
S!
-P (0
3 U
CU-H
-P -P
3 M
O-H
01
rH
-P
O
CM
O
O
O
CM
(O
CD • ZQ •
z in z o
i/i oj Q m
O
O
;j r~ j o
CD • 23 •
Z J1 Z O
in r- o
o
m m
r» r>-
l> 3> (J CO
o • o •
i/i * Q a*
t-- o>
CO o
in in
o 3> -o a<
o • o •
in in a co
r^ a>
O 3
O O
•3 3 i£ O
• J •
3 < 3
CM >n
0 3
0 O
30^0
0
o
o
0
o
: o
r-
<3 o
o
: o
0 < 0
CM in
_
0 < 0
CM n
CM
0 0
in o
o ^ o
• _l •
CJ < 0
.M .n
CM
O < CD
CM
CM
Jt
•J\
l/l
l/l
79
-------
o o
10 o 10 o
X • O •
z
o o
o o
IO 0 fO O
X • O •
z m z
n o 100
x • o •
Zlrt Z
o o
o o
IO O IO O
X • O •
zm z
o o
o o
IO O IO O
X «0 •
z z
o o
o o
io o 10 o
x • o •
Z IT) Z
O O
O O
IO O rO o
X • O •
z
o r>
in a>
in CM in r-
i/i
CM
o
IO
o
m
co o
0) -<
a co
o -o •
tow trr-
o
o
CM
(VI o
IO O
in r- in o
i/i • •
> «o i- o
in o
* o
IO O
it 9
in -•
a to a «o
o • o •
ce
ui • i/i •
> eo i- vo
c\» r-
* to
CM o
O GO
* -<
Q CO QO
O • O •
£Q O CD 0*
-4 * »- 10
CM ^
CO IO
in in
c ^
CM CM
CO >O
* in
OiO OCM
o «o •
CO rf CO CM
o
i-
o a
o o
O o o o
O »O •
33 o
in o
«0 CM
Kl *
i/) o i/) in
1/1 • in •
o>
CM
in
IO
o -•
O O O CO
o • o •
m -* 03 »>
l/» CM Q
vO CM
m f-
in 10 w 10
i/> • in •
> r^ i- o
in ;t
-t o*
r» m
9 m
f. oo
ao» Q as
o • o •
CO tO C O*
W1O Q *
0»
o>
in
t- o
* o
zm z: o
u^ • u. •
in in o o
« 3
a» in
o. -^ o.
O « Q
CM
a>
o •>
O* fO
Z CM Z O
o • a •
in * co
CM
co a
vO O
s: CM s. o
u. • u. •
in ro Q o
in in
-o
m *
t~ >o
a. m Q. *
o • a •
z o zo
o • a •
in «n co
co
CO O
J3 O
z: CM z o
u. • u. •
I/) O OO
r^ a
-H in
•o
CM 2: m
u. • u. •
in «H O in
in CM
•H *
CO CM
IO J>
O» CM
o. to a. o
o • o •
i/i CM *
vO 3-
in
CM
* r>
= o>
z in zo
o • a •
inr> CM
vO
o o
o o
s. o z o
ti. • U. •
in a
a. o a. o
o • a •
o o
o o
z o z o
o • a •
in
O O
z in z o
u. • it •
in r- oo
in o
m
ft *
10 \o
O. CM Q. *
O • Q •
o>
in r>
o ro
z o z o
o • o •
I/I ^ CO
-t CM
m o
•a o
z« z o
u. • u. •
in * o o
o o
o m
o
CM
* *
CM O
o. •a a. o
o • o •
in ^o co
o»
m
* =3
Z-4 Z O
O • O •
in r- in
:0 N
-P
a
o
u
x
H
Q
w
in o
* o
u o o o
as • ia •
Z O O
33 . CO •
Z -O Z O
i/» * Q in
Cf>
vO
IO O
CM O
or- o o
m • 01 •
zm z o
m^ a m
o
ro iO
>o O Z O
in f> Q «n
CM -4
•o a)
(J -H
o • o •
me- o *
M
O • O
I/If- O
«J fM U iO
O • O •
in 10 Q -o
vO vO
CM
03
CM
in co
in ;t
O f- O 43
o • o •
in d- o^
O CM
* IO
r-
CM
O O
o o
o o o o
o • o •
I/) Q
:» =0
O Irt O -^
O • O •
in 10 o *
CM in
CJ 9 O
O • O
in in o
o
ot»
o«
in
.•o o
CO O
* .*: o
• _i •
< o
m
CM
O
O
a£ o
_l •
< o
in
CM
o
o
^ o
_l •
< o
in
CM
O
-O
O
O
O
O
• -I •
CM
(M
o
o
in
CM
•rt o
CM o
3 o ^: o
o
in
.o
-H
en
80
-------
o o o o o
e o o o o
K) O lO O rt O lO O «O
» o o> n o> 10
» O * 01 * 0»
i/iioi/io 1/1 <« m cy> •- o > CM »- m >ojKin
lO O •* •< »H -<
>4 (M
4
0 •< O •« O •-<
oinoo o X If (E kfl 3) N (BlOlX)r>
« Q-i t/to in o
90 * o * o
z o z * z o 1*10
ll*ll.* H*l^« ^«Uk*
I/KM O O I/I (M O O I/I (M O O
.03 O 9
in in in
4-1 (0
3 O
^ CU-H
in-oio*io* «4j+j
o e -- Q) 4J
rH O
Q X> nJ
vOocor»a>r» OM
r-OS>K)aO-l <0
Z^Zf\JZ>CZOZ»OZO )_) p| r]
-< M M ^ 20)
AJ g(U
0*O^O\OO t^i i j
-4O(MO(MO /5 7,i
OM->OJOJO^>a->3 WOJ
Z*Z3 ZtMZO ZCMZO
ifliooin i/i om 1/1
O«D- O»O- O • O •
/> ^ ft -< ft
-M M M
-AJ ."M
v/1 J1
81
-------
o
o
a
Z -I
Ul 4 O
O O O
o
•.0-4
t- O X
I/) O I/)
O -41-
W V Z
I/I Ul
Ul K O
u z
U! Z U •
a: 4 o H-
4 z in
j ui • o
m
u
o z in
O K4 O t—
4 U Z
• Z Ul
h- Z Z
in + o»-
O «-> 4
U «0 t- Ul
2 4 CC
_J !-< I
I- 4 I- _l
« CC CC 4
0. Ul OH
4 0. S O
jO
in •
O-4
u
o .n
K •
in
o
o
I/) ^
3S
o *
u
u •
ui
o
in
i- o
o -«
u
m
o o
«M o
u. « •
in
CD
h- cc
in •
o
u
o
o •
e 10
Q • I/) (t
40 m to
Ul rO 4
I. _l
(M
IO
r-
m
o o e o
(Moo
o m
in
I- fM
in oo
o CM
(J ^^
O 10
rO
;/» •
o
u
IO
O CM
I- •
O
u
0t «H O
Z o >- IO O
o. CM in in r-
o o o CM
ex O «-t •
O O Z v£l
r- >
u
CM
ui in
rs| -«
1-1 CO
in
cc
o
o
tn •
in 10
cc
in
o
o •
CM
Ul
in
o
X *
Ul •
4 *
vn
CM
ftf #•!
o •
ui
u
10
t- • • •
in -4
o
o
10
M II II II II
y-
m
or
ui
u
i/i
yi
ui
u
o
a.
»- o»- »-
I/) t- y» Jl
O l/> O O
o o u o
o o 4 i-
inm
Q. •
4 * 00.
or o ra z
Ul _l -ID
4 D tn Q.
iO
o
vO
rO
o>
iO
GO
CO
or
Ul
J -I
4 I-
z »-
U. i/l
o
o
u •
O O
Ul (M
a
in
f-
in o
ae •
x
in 10
(0
O
•H
-P
cn
•H
0)
u
X fU
L_l (j
D (0
W U
CU
DJ ca
«; ca
0)
o
o
- z
« £
at-
Lj 'jj
4
> J
>-l 4
4 U.
82
-------
o o o
o o o
b. • >- o b. •
O«M JO U •«
UJ O • UJ
0.
u
* 0 *
o o in
H CM >- • K >O
in • _j i/i •
O>4 O O •£>
u o, u
•- a >-
h- in *
H \O O «H f- O
i/> • < o m •
o o • o *
u o u
< <
in £ o o
.* o * >o
>•• om _i \o o (M
J K • UO*->
o in u • m CM
Q. 0 b. 0
o o o o
o • o •
3 * O O»
>- • h- 41 O* tV-0
_j O Ul »•* H UJ
U.
0 0
o •
X • U. «-«
»- CO U
H4
O "H • •
o • o h-
X • «•! t> .no 33*
H-O ZO^ V1CM Qt-
t!> ^) X *O H- t/) h-
-<»--« a *
-> <
• o • *
ino 2t\j <^ 10 u.0
y> o x i > j)
H- O Q. n o a -ft
o* • ^ •
ae» zo _j« a.o
^ 10 U. -f * Is-
!- »- O >
o
ig a
z •
o
o
*
u. •
o
*
UI-H
a
»-
o
•
> 0
X 0
o
o
•H
o
•
>
3K
0
cc m
UJ
0,
^
o
•
* .n
s ro
a.
X
0
O -1
2
»-i
?
3
•
J rvj
S
0
o
u. •
U <-l
bl
1^
1- CM
in •
o *
u
i-
•H
CO
O CM
U
N
O *
1- •
O
•
*
t- in
i/l co
O CO
O 0
o »
CM
•
CO
t- N
l/» -4
O !O
O (O
u.
•
»- 0>
I/) ^
O lO
a) co
0
0
CM •
o o
1/1 aO
0
o
CM •
O
in
a
3
o
•M •
_J O
O (M
U CM
O
CM 0
j *O
(J
H-
0
0
CM •
(_>
000
in CM in o
co 10 o
t- CM CM o
in • • •
o
u
o o o
o o* o
00 0* 0
H- • • •
0
O
• * •
~D t- O
K 10 r-
m •a t-
o o 10
O in o
O CM CM
0 0
t- Ul UJ
O U> Z UJ Z
- o >-
{J £Q (J
X
A
a.
CO
a.
83
-------
I/I
o
CM
o
CO
o
u
o
u
o>
m
*
o
u
o
o
t- o
I/I •
o
u
1/1
o
u
o
o
• o
(V
vfl
o
u
o
o
O O
K- •
I/I
o
(J
m
o
O 'ft
S"
u
I/I
o
o
I- Kl
u) in
or-
(J (7>
010
o co
o •*
o -H
Ul
o
u
I/I
o
u
u
*~+
•
-p
c
0
o
w
X
H
Q
12
H
ft
CM
<
CO
-P 0
3-H
ft-p
.p to
3-H
0 W
(U
g -P
^ 0
0) (0
J3 M
O nl
MX:
PU U
a) u
i-4 CO
Ond)
g o
(tl O
X M
M ft
35 -<
4
X
Ul V
> •r
-« o
at tr
t- o
2 J
a
z
a.
o
U X
.9 /)
<
a: a
< z
-o <
sc z.
o o
O X
3] UJ
< a.
O'
Ot-
* <
a ae
X
o
84
-------
3
E 5
x f-
5 -
o
o»
fm
2
I/)
o
u
1-
z
<
a.
_i
<
H
o
»-
DOLLARS
•
IO
.»
CM
CM
O
K»
^
II
t-
m
o
u
<
H-
f*
a.
<
ij
j
<
H-
0
>"
CENTS/1000 GALLONS
in
«4
II
1-
m
o
u
2
O
-H
>-
<
fsl
«-«
»-
2
O
<
_l
<
>-
O
t-
CENTS/1000 GALLONS
o
*
in
•
o
•H
II
»-
in
o
o
X
+
o
_l
4
»-
o
r—
CENTS/1000 GALLONS
in
>o
-a
•
in
CM
u
»-
in
0
o
t-
z
Ul
K
<
J
(X
K-
_l
<
h*
O
t-
>- •
X
»4
U
u
•
0
< 0
O o
o
•H
O
in
»- *<
o •
a
o
o
X •
x >o
a
o
•
-y> >n
cr CM
>•
(X)
vO
>-• o
a: •
g
u.
•4
5
m
2
Ul
Q
Z
<
-1
X
(J
u
h-
a
oc
<
«
r»
iO
in
•
IO
«-•
CM
r»
in
•H
•H
•
>O
IV
^
-0
CO
r^
•
*
CM
••4
>4
CO
CM
•
*
*
in
*
o
in
o
•
CM
n
o
0
b
X
H
Q
2
W
04
f\.
m
<
4J
3
04
-p -p
3 W
0 0
U
e
0) -P
H C
X) cd
O iH
M P4
A
1
a) n)
rH 4J
p* °
X
M
CM
CJ
•o
(M
X -•
< in
o *
>- -H
-M
O*
rsi
01
(M
ae
o
85
-------
CASE IV WORKSHOP
MODIFICATION OF EXISTING
EXEC PROGRAM SUBROUTINES
Stephen P. Graef
Pritzker Department of Environmental Engineering
Illinois Institute of Technology
Chicago, Illinois 60616
and
The Metropolitan Sanitary District of Greater Chicato
100 East Erie Street
Chicago, Illinois 60611
ABSTRACT
A procedure is presented for modifying the existing sub-
routines in the Exec Program for Wastewater Treatment Design
developed by Smith and Eilers of the USEPA. The user can now
tailor the process models of the various subroutines to meet
his specific design needs. The procedure is demonstrated on
the original trickling filter subroutine, TRFS. The Eckenfelder
model is replaced by the Galler and Gotaas model. An analogous
assigned problem of modifying the second stage anaerobic digester
subroutine, DIG2, is presented as an exercise for the partici-
pants.
86
-------
INTRODUCTION
One of the interesting facets of engineering design is the
fact that an engineer can establish criteria, size and components
for a project in a variety of ways. An office or commercial
building, for example, could be designed as a steel, concrete,
or timber structure. Moreover, technical procedures for
calculating structural sizes and quantities could originate
with several codes or as specialized design formulae developed
by the designer.
When the authors of the Exec Program first selected the
processes which would be included in the total package, one
of their principal concerns was the choice of process design
equations to use for each subroutine. An examination of current
treatment facility design texts will disclose that cost pro-
cesses can be described by several mathematical models. In
order to meet the needs of the majority of the potential users
the authors selected the most widely used process model when
they formulated the FORTRAN coding for the process subroutines.
The users who have worked with the Exec Program in engineering
practice have found that the subroutines, for the most part,
satisfy their design requirement. Notwithstanding the overall
utility of the Program, many users have found it necessary to
modify at least one subroutine to meet a specific design need.
It is anticipated that future users will also find it necessary
to tailor several subroutines for specific design tasks. There-
fore, it is the purpose of this paper to outline the procedure
for modifying the subroutines and to show by example how the
procedure was applied.
MODIFICATION PROCEDURE
All of the process subroutines can be modified by the user
who has a modest knowledge of FORTRAN and who is familiar with
the Subroutine User's Guides. He should be cautioned, however,
that a change in a subroutine or EXECMAIN could significantly
interfere with other parts of the Exec Program if he is not
careful. Fortunately, most problems can be avoided by referring
to a simple checklist presented in Table 1.
DMATX, Process Design Criteria
The first step in modifying one of the subroutines is to
review the process design criteria of the original version and
decide whether additions, deletions or substitutions are needed.
The user should watch for a change in engineering units or a
need to renumber the DMATX (I,N) sequence.
87
-------
Table 1. Checklist for Modifying an Exec Program Subroutine
1. DMATX, Process Design Criteria
a. Additional Criteria
b. Fewer Criteria
c. Removal or Replacement of Criteria
2. COMMON, Initial Common Statements
a. Changes in the number of Arrays
b. Changes in Array dimensions
c. Additional COMMON/ /... Statements
d. Additional or modified arguments or parameters
3. Algebraic Statements
a. Changes in process sizing equations
b. Changes in stream constituent equations
c. Changes in cost equations
4. OMATX, Process Parameters in Output Array
a. New parameters to be included in output
b. Replacement or deletion of output parameters
5. PRINT Subroutine
a. Changes in labeling format
b. Additional pages or tables of output
c. Format for new subroutines
6. SMATX, Stream Constituents
a. Additional Stream Constituents
b. An additional Input or Output Stream for the process
7. COST Subroutine
a. Are new cost equations compatible with COST Subroutine
8. ENERGY Subroutine
a. New or modified equations for process energy
88
-------
Initial Common Statements
Parameters and arrays which have been defined in the COMMON
statement may need revision. If an additional parameter or
array must be added, which will be passed to the other sub-
routines, then it is necessary to make the changes in the COMMON
statement of all other subroutines and the EXECMAIN. Sometimes
a few new parameters are added which are only passed to several
subroutines. In such cases a labeled COMMON statement, COMMON/
name/..., can be added to the pertinent subroutines. Occasional-
ly array sizes must be enlarged. If so, then all the subroutine
COMMON statements must be adjusted to reflect the new array size,
Algebraic Statments
Nearly all modifications affect the subroutine's alge-
braic statements. A new mathematical process model which re-
places the original one necessitates changes in the process
sizing equations and possibly the stream constituent equations
as well. If a process sizing equation has a cumbersome arrange-
ment of expressions, the original subroutine may have been
simplified by assigning several FORTRAN identifiers or names to
several groups of expressions. These names were then employed
in the process sizing equations. It is important, therefore,
to eliminate the grouped expressions from the original sub-
routine or make them compatible with the new process sizing
equations. A new process model may also affect some of the
stream constituent calculations e.g. all solid constituents
or all dissolved constituents or all carbon related constituents.
Some constituents may be calculated within a loop and may have
to be removed from the loop if it is to be calculated by a new
equation. Modifying the cost equations is easier because they
can be readily identified. For example, a new curve for the
energy consumption for a given process can be added without re-
quiring a change in the equation for operating cost. A change
in the capital cost curve does not require a change in the
amortized cost equations. It is important, however, that each
cost relationship is examined to be certain that all necessary
changes are made.
OMATX, Process Parameters in Output Array
When a subroutine is altered it may be necessary to in-
clude an additional calculated parameter among those printed
in the output. On the other hand, it may be necessary to elim*-
inate from the output one or more of the parameters which were
originally included. The user should check the latter section
of the subroutine FORTRAN statements where parameters are
assigned to OMATX (I,N). As the Exec Program now stands, as
many as 20 parameters may be assigned to OMATX for a give
process.
89
-------
PRINT, Subroutine
All changes in OMATX assignments necessitate changes in
the PRINT subroutine since it specifies the format and labeling
used in the printed output. Each process subroutine has its
format section in the PRINT subroutine labeled for ease of
checking. The user should be careful when examining long
character strings in these format statements.
SMATX, Stream Constituents
Some users may want to enhance a process by adding a second
input and/or output stream. This could affect several sections
of the program. A recheck of the process sizing equations must
be made to determine whether the new streams should be included.
Jn addition, the user must develop equations for calculating the
19 constituents in each of the new process streams. Both of
these tasks create a host of potential error situations. Final-
ly, the user may wish to assign a new constituent to SMATX
(20,1) which is not currently used. If a subroutine modifica-
tion requires several new constituents then the user must expand
the number of rows in SMATX and TMATX which are currently (20,30)
arrays. Such a change affects SMATX and TMATX in all sub-
routines. Moreover, adjustments must be made in lines EXE05600
and EXEO7500 of EXECMAIN. This is a major modification and the
user should be cautious with each change he makes.
COST Subroutine
At present (August 1977), the ENERGY subroutine, which is
the most recent addition to the Exec Program library, includes
simple relationships between equivalent kilowatt hour require-
ments for each process and the raw sewage flow. The energy
relationships are being enhanced as the Exec Program evolves.
Users should examine and modify, if necessary, the pertinent
process energy equations in the ENERGY subroutine when modifying
one of the process subroutines.
ENGINEERING EXAMPLE
The existing trickling filter process model was developed
by Roesler and Smith (1969) from the work of Eckenfelder (1961)
and Rowland (1957) for use with the Exec Program. It is one
of several models used by designers in sizing trickling filters
(Schroeder, 1977). Others include equations by Callers and
Gotaas, National Research Council, and Velz. Some design
offices frequently size trickling filters by several equations
and temper the final design with engineering judgment. Other
offices have a preference for one model when employing synthetic
media for an industrial application and a different model
when sizing a rock media for a domestic wastewater treatment
facility. This paper explains the development of a second
90
-------
trickling filter subroutine for the Exec Program library based
upon the Caller and Gotaas (1964) trickling filter model.
Existing Model
Appendix A presents the derivation of the existing model
Basically Roesler and Smith (1969) start with first order
kinetics (equation 1) and an empirical travel time expression
for a particle of water (equation 2) and develop an equation
for required filter depth as a function of design criteria
specified by the user. These include effluent BOD, hydraulic
loading rate, specific surface area, temperature and recircu-
lation factor. Figure 1 depicts the trickling filter flow dia-
gram used in the derivation. It depicts the recylce stream
being withdrawn and returned to the head of the filter without
passing through the final settling tank. Such an arrangement
leads to an equation for filter effluent BOD in terms of the
known influent wastewater BOD and the specified settler effluent
BOD. Other trickling filter flow diagrams can be described in
terms of the equations of Roesler and Smith (1969). Several
of these are presented in Figure 2. The principal difference
would lie in the materials balance equations across the filters
and final settling tanks. A key assumption in Roesler and
Smith's development is that BOD in both the suspended solid
form and in the dissolved form are removed at the same rate as
the wastewater passes through the filter.
New Model
Derivation—
Appendix B presents the derivation of the trickling filter
process equations based upon the Caller and Gotaas model (1964),
Figure 1 shows the associated schematic flow diagram. Their
model is an empirical formula based upon regression analysis in
which effluent BOD from a trickling filter is correlated wit
influent BOD, hyrdaulic loading rate, recirculation rate,
temperature and depth. Unlike the Eckenfelder equation, the
Caller and Gotaas equation does not include specific surface
area as an independent variable.
When attempting the derivation, it becomes apparent within
a short time that an explicit solution of depth in terms of
known variables is not possible. The effluent BOD is related
to the influent BOD by the 1.19 power and the influent BOD is
not explicitly known. One way of attacking the problem is a
trial and error solution. Basically, values are assumed for the
total and dissolved BOD in the filter influent as well as filter
effluent BOD. The depth is then approximated and used to re-
calculate the filter effluent BOD. The final settler effluent
BOD is calculated and compared with the design specified final
settler BOD. Unless the calculated and specified BOD are
91
-------
1 2
3 4
Figure 1. Schematic Diagram of Trickling Filter
92
-------
Figure 2. Alternative Trickling Filter Schematic Diagrams
93
-------
within the allowable error, the effluent BOD is refined to a
value halfway between the calculated and specified values. The
refined effluent value is then used to recalculate the depth.
This cycle continues until the allowable error between the two
values is attained. Magnitude of allowable error determines
the accuracy of the filter depth required and the number of
iterative loops needed to obtain a satisfactory calculated
effluent BOD. To illustrate this effect, filter depth calcula-
tions were made using errors of 0.6 and 0.4 mg/1. The results,
which are presented in Table 2, indicate that the error in the
depth calculation increases as the required depth becomes larger.
An error of 0.1 mg/1 was also tried; however, the limit on
the number of iterative loops was exceeded before the error
tolerance was met. A value of 0.4 mg/1 was therefore selected.
As in the original subroutine derivation, it is assumed
that the reaction rate constant for both the solid and dissolved
BOD fractions is equal as the wastewater passes through the
filter.
Table 3 lists the original FORTRAN program and Table 4 is
a FORTRAN listing of the modified process equations. A glance
at the line numbers in the right margin indicates which equa-
tions have been substituted. Moreover, the equations on lines
TRF02200, TRF03300 and TRF03500 have been removed. The
statement two lines above line TRF03100 compares the BOD deriva-
tion with the allowable error. Finally, a WRITE statement is
included to observe the convergence of the calculated (CHECK)
and specified (DMATX(1,N)) final settler BOD values. All
of the equations that were based on the Eckenfelder reference
model have been removed or replaced by ones based upon the trial
and error solution of the Caller and Gotaas reference model.
Modification Checklist--
Table 1 serves as a useful checklist for reviewing the
modifications made on the original trickling filter subroutine.
The numbered comments below refer to the numbered points on the
checklist.
1. DMATX (4,N) SAREA was not deleted even though it is
not used as a design criteria for the Caller and Gotaas
model. Had it been removed, the other DMATX criteria
would have needed renumbering. Renumbering would have
affected many of the process equations as well as the
PRINT subroutine which formats the labels for printing
all DMATX values.
2. The COMMON statements were not altered.
94
-------
Table 2.
Effect of Acceptable Error Magnitude on Predicted
Depth
Case 1. DMATX(1,N) BOD Varied from 12 to 30 mg/1
Filter Depths (ft.)
BOD
12 rag/1
15
18
21
24
27
30
Acceptable Error
(mg/1)
0.6 0.4
25.4 ft.
is.4
14.0
11.2
9.2
7.8
6.7
25.0 ft.
18.0
13.8
11.1
9.2
7.8
6.7
Case 2. DMATX(3,N) HQ
HQ
5
10
15
20
25
30
Varied from 5 to 30 mgd/acre
Acceptable Error
(xng/1)
0.6 0.4
9.7
11.2
12.2
12.9
13.6
13.6
9.6
11.1
12.1
12.9
13.5
14.0
95
-------
TABLE 3
Fortran Listing of
Original Process Equations
C ^ TRF00100
c TRICKLING FILTER - FINAL SETTLER TRF00200
c PROCESS IDENTIFICATION NUMBER 11 TRFOOSOO
c TRFOO<*OO
SUBROUTINE. TRFS TRFOOSOO
C TRF00600
C TRF00700
C COMMON INITIAL STATEMENTS TRF00800
C TRF00900
INTEGER osi»os2 TRFOIOOO
COMMON SMATX(20r30)»TMATX<20»30)»DMATX(20»2u> »OMATX(20»20)»IP(20)>TRF01100
UNPf IO»ISl»IS,i.OSl»OS2»N'IAERFfCCOST(20»5) >COSTO<20r5> > ACOST (20» 5) TRF01200
2»TCOST(20»5)»UHR»PCT»wPI•CLAND»DLAND»FLOW 126).POW125)rTKWHDl2b) TRF01300
C TRF01UOO
C TRF01500
C PROCESS RELATIONSHIPS REQO. To CALC. EFFLUENT STREAM TRF01600
C CHARACTERISTICS TRF01700
C TRF01800
HEAD=UMATx(9»N> TRF01900
BOUlN=SMATX(8rISl)-«-SMATX(17rISl) TRF02000
>l)-20.) TRF02100
TRF02200
Qo=DMATX(7»N>*SMATX(2fISl) TRF02300
RHQ=< (DMATX(7rN)-»-l. )*UMATX(3>N) )**XN TRF02tOO
BOD=-»-DMATX(6fN)*SMATX(8>ISl) )/DMATX(l»N) TRF02500
Dh.PTH=RHO*ALOfa< (BOD+DMATX (7»N) ) / (DMATxl7»N)-«-l. ) ) /IBETA*DMATX (^»N) ) THF02600
XPO=ExP(BLTA*uMATX{4»N)*DEPTH/RHQ) TRF02700
BOOO=ciODlN/ (XPO* tDMATx (7 • N) * (1 .-1 . /XPO)+!.)) TRF02800
DtlODO=SMA1X(17»lSl)/(xPO*(DMATX(7»N)*(l.-l./XPO)+l.)) TRF02900
SBOO'+=BODU-DBODO TRF03000
Sb005=SBOU"+*DMATXlb»N) TRF03100
Bt.TAN=;.00-i07*l.l'*l**(DMATX(2>N)-20.) TRF03200
XPON-t.XP(bETAN*DMATX(t»N>*DEPTH/RHO) TRF03300
SONi+=bMATx(S'lSl)*SDOOt/SMATX(8fISl) TRF03400
DNi+=(bMATx(i3»ISl)-»-SMATX(5»ISl)-SON1)/*(l.-DMATX(5»N))/(uMATXI6>N)-DMATX(5fN>) TRF04200
SMATX(2»Ob2)=bMATX(2'ISl>*(l.-DMATX(6»N))/(uMATX<5»N)-DMATX16.N)) TRF04300
SMATX(4»ObI)=SMATX(1f IS1)*DMATX(6»N) TRFO<**tOO
SMATX (H »0b2) =bMATX (11IS1) *DMATX (5»N) TRFO**500
SMATXl5tObl)=bON5 TRFOU600
SMATX(5fOb2)=bONt*DMATX(5»N) TRF04700
SMATX(6fOSl)=bMATX(6»lSl)*DMATx(6»N)*SBODt/bMATXCaflSl) TRFOH800
SMATX(6»OS2)=bMATX(6»iSl)*DMATX(5»N)*bBOD<+/bMATXCa»lSl> TRF04900
SMATX (7»Obl)=bMATx(7»lSD*DMATX(6rN) TRF05000
SMATX (7»Ob2)=bMATX(7»lSU*DMATX(5»N) TRF05100
SMATXl8»Obl)=bBOO'**DMATX(6»N) TRF05200
SMATXl8»Ob2)=bBODH*OMATX(5»N) TRF05300
SMATX (9>Obl)=SBOD<+*DMATX(6»N)-i-SON5+SMATX(<+f ISH *DMATX(6tN) TRF05«*00
SMATXi9»Ob2)=b80Dt*DMATX(5»N)+SON'+*DMATX<5»N)-»-SMATX(t»ISl)*OMATX(6TRF05500
1»N) TRF05600
SHATX(10»OSl)=SMATX(9»OSl)+SMATX(7»OSD-«-SMATX(6»OSl) TRF05700
SMATX(10»OS2)=SMATX(9fOS2)+SMATX(7»OS2)+SMAIX(6»OS2) TRF05800
96
-------
TABLE 4
Fortran Listing of
Modified Process Equations
FILIEK - FINAL bETTutR
lvc.u1iFlC.HTKm NUmbER 11
TKFU010U
TKFUU2UU
TRFUO»*UO
TKFUObOO
IkFUUoOU
TKFU070U
Co ...... 0,
TRr-b
CoMC.Ou lUj
Ojl.Obi; TKFulOOU
SMMTxltU.3u)tlr. ATX(20130)»UMATx(20i2u>»OMATx(2u»20)riP(20>»TKF011UU
• ibi Hbt »OSi »oSt »U» lAt"RF»CCOST(^0»5) »LObTO<«!0»5) »ACobT(cO»5)TKFol20u
120»b) »t/HK«HCl »«P1 »CLANU»DLAND»(-LOWtij) >POWl25) »TKWHDt2b) TRFul30u
IHF0140U
TRFU150U
^b RcLATiOiMSulCS KEQu. Tu CA^t. LFFLuLuT STRt-AM TKFuloOu
KS TnFol70u
TKFU1800
TKFU190U
Ibl) THFU200U
)-20.) THF021UU
Ibl) THFU230U
1 •
(7»NJ
GO To o
) **U.o7
) ** 1 . 5-1
*jMMTX(o»n)-«-L)dOut*( 1 .-DMATX <6»N) )
10"*)
CO' r
bu j>*=DOL)'« + 0 . b» ( DMATX ( A • N ) -CHECK )
/bOu1* T *ubOD«*
iN + boL)t*u
) X (1 / » 1 Si ) +DoOU'**DMATX ( 7 »N) / « 1 .
li- (AUa(CHuCK-uMATx(l»i4) ).bE.O.m bO Ty 2
lo»N)
Ei-FLUutJT
CALCULATIONS
TRFU3100
TKFU3200
TKFU3'*Oo
TRF03bOU
TRF0370U
TRFU3BOO
TRFU3900
TRF01000
TKFOmOU
rN» TRFUH20U
97
-------
3. Extensive changes were made in the algebraic process
sizing equations. No changes were needed in the
stream or cost equations.
4. The OMATX definitions and values were not altered.
5. Since the DMATX numbering was not changed, the PRINT
subroutine remained satisfactory.
6. SMATX values were unaffected.
7. COST was unaffected.
8. No changes were needed in the ENERGY subroutine.
COMPARATIVE RESULTS
Once the modified subroutine was checked and debugged, it
was compared with the original subroutine. To simplify the
comparison, a one process system consisting solely of a trick-
ling filter final settler was evaluated. It was programmed to
treat the typical raw sewage stream which is quantified in the
EXECMAIN User's Guide. Both trickling filter subroutines were
evaluated to characterize their relationships between (1)
hydraulic loading rate and filter depth, (2) effluent BOD re-
quirement and filter depth, and (3) recirculation factor and
filter depth. Table 5 summarizes the input conditions for the
single process calculations.
The results of the Exec Program calculations are presented
in Figures 3a, 3b, and 3c. The Caller and Gotaas model (1964)
predicts a broader range of filter depths as a function of final
settler BOD than does the Eckenfelder model (1961). On the
other hand, a broader range of depths as a function of hydraulic
loading rate and recirculation factor result from the Eckenfelder
model than from the Galler and Gotaas model. The practical
implication is that the Galler and Gotaas model is more sen-
sitive to effluent BOD criteria and the Eckenfelder model
has a greater sensitivity than the Galler and Gotaas model in
terms of hydraulic loading rate and recirculation factor.
In a second comparison of the two subroutines a trickling
filter process was part of a complete wastewater treatment
system as depicted schematically in Figure 4. Table 6 lists
the input conditions used in the Exec Program calculations on
the system in Figure 4. The results are plotted for both filter
models in Figure 5. The figure graphically characterizes the
effect of influent BOD on the required filter depth. The
data indicate that the Galler and Gotaas model is the more sen-
sitive of the two versions to influent BOD concentrations.
98
-------
Table 5.
Input Conditions - Single Process Calculations
SMATX(I,1) I = 2,20
DMATX(I,20) I = 1,10
DMATX(1,N) BOD
DMATX(2,N) DEGC
DMATX(3,N) HQ
DMATX(4,N) SAKEA
DMATX(5,N) URSS
DMATX(6,N) XRSS
DMATX(7,N) RECYCL
DMATX(8,N) GSS
DMATX(9,N) HEAD
DMATX(14,N) ECF
DMATX(15,N)
DMATX(16,N)
Same as typical raw sewage composi-
tion listed in EXECMAIN Users
Guide
Same as typical cost parameters
listed in EXECMAIN Users Guide
Varied with each run (12-30) mg/1
20.0°C
Varied 5-30 mgd/acre
10 ft2/ft3
Varied 2-100
0.6
Varied 0.5-5
2000 gpd/ft2
30.0 ft
1.0
1.0
1.0
99
-------
D
O
CO
w
tl \
o o>
-H E
-H
O
0)
a
w
30
20
10
(a)
RECYCL =1.0
URSS = 2.0
HQ = 10 MGD/acre
Caller and Gotaas
10
15
20
25
Cn
C
•H
•a
D
it)
"O
>i
X
30
20
10
5
(b)
BOD =21 mg/L
URSS =2.0
RECYCL =1.0
10
15
20
25
n
O
4J
O
S,
Recirculation
5 r %
4
3
2
1
(c) \
\ BOD = 21 mg/L
\ URSS =2.0
\ HQ = 10 MGD/acre
: N\
i i v i i
0 5 10 15 20
Required Filter Depth, ft.
25
Figure 3. Process Characteristics of Both Filter Models
100
-------
•H
Q
O
•H
e
o
-P
CO
-------
Table 6.
Input Conditions - Treatment Facility System
Calculations
SMATX(I,1) I = 2,20
SMATX(3,1) SOC
SMATX(8,1) SBOD
SMATX(11,1) DOC
SMATX(17,L) DBOD
DMATX(I,20) I = 1,10
RWP
DMATX(1,N) EHAD
DMATX(16,N) ECF
PRSET
DMATX(1,N) FRPS
DMATX(2,N) RIPS
DMATX(3,N) HPWK
DMATX(15,N) ECF
DMATX(16,N) ECF
TRFS
DMATX
DMATX
DMATX
DMATX
DMATX
DMATX
DMATX
DMATX
DMATX
DMATX
DMATX
d,N)
(2,N)
(3,N)
(4,N)
(5,N)
(6,N)
(7,N)
(8,N)
(14,N)
(15,N)
(16,N)
THICK
DMATX(1,N)
DMATX(2,N)
DMATX(3,N)
DMATX(4,N)
DMATX(16,N)
BOD
DEGC
HQ
SAREA
URSS
XRSS
RECYCL
GSS
ECF
ECF
ECF
TRR
TSS
GTH
GSTH
ECF
DIG
DIG2
DMATX(1,N) TD
DMATX(2,N) TDIG
DMATX(16,N) ECF
>
DMATX(1,N) TRR
DMATX(2,N) TSS
DMATX(3,N) TD
DMATX(16,N) ECF
Same as Table 7 except;
Varied 53,105,158 mg/1
Varied 70,140,210 mg/1
Varied 22,43,65
Varied 30,60,90
Same as Table 7
30.0 ft.
1.0
0.50
200.0
14.0 hrs/week
1.0
1.0
20.0 mg/1
20.0°C
10.0 mgd/acre
10.0 ft2/ft3
2.0 and 50.0
0.6
1.0
2000 gpd/ft2
1.0
1.0
1.0
0.95
50000 mg/1
700 gpd/ft2
8 lb/day/ft2
1.0
15 days
35QC
1.0
0.81
50000 mg/1
1.5 days
1.0
102
-------
200^
-------
Application.
How should the designer decide which trickling filter model
to use? One rational approach is to examine the design and
operating data of a filter installation similar to the type that
the designer may specify, e.g. high rate rock media, shallow
plastic media, tower with plastic media, etc. He should then
compare the results from both models with actual plant data.
Plant data must be examined with a critical eye and it is recom-
mended that the designer discuss the procedures for collecting
and recording the data with the plant manager. With an under-
standing of the quantitative and qualitative performance of.
the filter, the designer can make his selection.
ASSIGNED PROBLEM
Objective
A simple problem has been developed which will lead the
participant through the procedure for modifying a subroutine.
Statement
The second stage anaerobic digester is one of the simplest
subroutines. It requires only three design criteria, one of
which is the total suspended solids concentration in the under-
flow stream, OSl. The user simply picks a constant which he
feels is appropriate. In operating practice however, the actual
underflow sludge solids concentration is not a constant but is
a function of the detention time of the digester.
Modify the subroutine so that the underflow solids concen-
tration is a function of the detention time. From personal ex-
perience it has been my observation that a digested waste acti-
vated and primary sludge mixture will concentrate from a
nominal 2-4 percent sludge to a 6 percent sludge in about 45
days. Assume, therefore, that the underflow solids concentra-
tion will increase according to the curve in Figure 6. Mathe-
matically stated
(60000-TSSIS1)
Approach
Before proceeding, review the checklist in Table 1; then
proceed as follows:
1. Refer to the User's Guide and note all FORTRAN state-
ments that depend upon DMATX (2,N), TSSOS1' especially
those stream constituents which are of a solid nature.
104
-------
70000
-P
s
60000 UXUAfllOM! ±V* ^
* 50000
o
tH
U-f
M
Q)
"S
40000 _
30000
Q)
4J
co
V
&>
•H
Q
CO
•O
O
CO
-------
2. Decide how DMATX (2,N) could be defined by the equation
above and where the FORTRAN statement should be placed
in the subroutine. Make sure that the concentration of
all solids constituents will be proportional to the
change in TSSQS2-
3. Modify the subroutine card deck given you and run the
Exec Program using the data cases listed in Table 7,
4. Prepare a set of curves with (1) TSS ., (2) DN ,,
(3) TSS sl (from the RWP) and (4) COST of the second
stage digester on the Y axis with TD on the X axis.
106
-------
Table 7.
Listing of Data Cards for
Assigned Problem
f\i'.
0
0
0
0
0
'0
0
0
0
Q
*5
0
MO-
1
0
rtb!
1
L,
:>U
0
1
Z
2
0
<*
5
6
0
0
> L(.
G
0
ill
6
u
jut-u sUuKGUiIuE MOD1F
10. 105.
-------
REFERENCES
Eckenfelder, W. W., Jr. 1961. "Trickling Filter Design and
Performance," Journal Sanitary Engineering Division of
A. S. C. E. , 87, 2860.
Caller, W. S. and H. B. Gotaas. 1964. "Analysis of Biological
Filter Variables," Journal Sanitary Engineering Division
of A.S.C.E., 90(SA6);59-79.
Rowland, W. E, 1957. "Flow Over Porous Media as in a Trickling
Filter," Proceedings 12th Purdue Industrial Waste
Conference, Extension Series No. 94, Purdue University,
Lafayette, Indiana, 435.
Roesler, J. F. and R. Smith. 1969. "A Mathematical Model for
a Trickling Filter," U.S. Department of the Interior,
Federal Water Pollution Control Administration, W69-2.
Schroeder, E. D. 1977. Water and Wastewater Treatment, McGraw
Hill Book Co., New York.
108
-------
Appendix A
Derivation of Existing Trickling Filter
Process Equations (1)
Assume BOD removal follows first order relationship with time.
(1)
dC = -k C dt
C = BOD, mg/L
k = rate constant, I/time
Time required for a particle to pass through depth, D, ft. was
estimated by Howland as
k (SAREA)D
t = - - N - (2)
HQfN
k = constant
a
SAREA = specific surface area of
media ft2/ft3
HQf = hydraulic loading rate through
the filter, mgd/ft2
HQf = (1+R)HQ
R = recirculation factor
HQ = hydraulic loading rate
excluding recycle
N = temperature dependent rate
constant
By assuming ka and B do not vary within the filter, equation (1)
can be substituted into equation (2) and integrated to
-K (SAREA) D
HQN _|
C = Ce (3)
C. = BOD applied to filter, mg/L
K = (k) (k )
cl
Referring to Figure 1
Q = flow rate
subscripts refer to stream numbers
in Figure 1
109
-------
Q1(1+R)C2 = Q1C1 + QaRC6 (5)
R = Q-L/CV
C6 = C3 (6)
C, + RC-,
C = -± - i (7)
U+R)
K(SAREA)D
HO^N
Let E = e yf (8)
Substituting equation (9) into (3)
C3 = C2(1/E) (9)
C2 = C3E (10)
C, + RC,
C E « — - - (11)
J d+R)
C = - - (12)
- R
A fundamental assumption made to complete the derivation is that
dissolved BOD and solid BOD are removed at the same rate.
C3 ' C3D * C3S
- R
C1S
±2 - (15)
- R
subscripts
D = dissolved BOD
S = solid BOD
110
-------
From Figure 1
C4 « C3 (16)
C5 = C3D -f- XRSS C3S C17)
XRSS = fraction of influent solids
removed in final settler
in TC;
C = - — - + XRSS - — - (18)
- R E(1+R) - R
C1D + C1S XRSS
C = -±2 - i§ - (19)
° E(1+R) - R
C,n + C,c XRSS
Let X = -i£ - i£ - (20)
C5
E = _ (21)
1 + R
rK(SAREA)D"|
L HQ N J
HQf " X + R
e =
1 + R
Solving for D
(22)
HQfN X + R
_£ in 2L±_£ (23)
K(SAREA) 1 + R
Reconciling the derivation with the TRFS FORTRAN Statements
in Table 3
Replace K with BETA
where BETA is calculated by line TRF02100
Replace N with XN
where XN is calculated by line TRF02200
Replace HQfN with RHQ
where RHQ is calculated by line TRF02400
111
-------
Replace X with BOD
Equation 20 becomes line TRF02500
Replace D with DEPTH
Equation 23 becomes line TRF02600
Replace E with XPO
Equation 8 becomes line TRF02700
Replace 03 with BODO and GI with BODIN
Equation 12 becomes line TRF02800
Replace CSD with DBODO
Equation 14 becomes line TRF02900
Finally, replace C4g with SBOD4 and C5g with SBOD5
Statements TRF03200 through TRF03700 define nitrification which
occurs in the filter. Note that the kinetic equations are
analogous to these for BOD removal with the exception that the
rate constant BETAN is significantly lower than BETA.
112
-------
Appendix B
Derivation of Modified Trickling Filter
Process Equations
The Caller and Gotaas (2) equation for BOD removal in a trickling
filter was selected as the basic model around which the
original TRFS subroutine was modified. Referring to Figure 1
C21'19(0.464)(l+R)°-28(HQf)0'13
(1+D)0.67T0.15
(1)
where the symbols retain the
definitions of Table 2
Solving for depth D
1.19
u,
D =
(0.464) (l-fR)0'28(HQ,)0-13
C3T
0.15
3/2
-1
(2)
Since the empirically developed G and G equation contains
fractional exponents, it was not possible to make the
substitutions utilized in Table 2 for defining the unknown
C3 in terms of the known GI and €5. Therefore, a trial and
error procedure was used.
1. Assume value for C- and C2
2. Calculate GNG1 =
0.464(l+R)0-28(HQf)°'13
m0.15
(3)
3. Calculate depth
' 1.19
D =
GNG1
1.5
-1
(4)
4. Calculate GNG
GNG1
GNG =
d+D)
0.67
(5)
113
-------
5. Calculate C3 and
1*19
= C2 GNG (6)
1.19
c3D =c3
[DBOD2]
|_BOD2J
6. Check whether or not the calculated C5 is within the
allowable error tolerance of the C5 value specified in
the process design criteria DMATX T1,N) .
Note that C4 = C3 (8)
C3S (9)
XRSS C (10)
C3 = C3D + C3S
C5 = C3D + XRS
C5 = C3D + (C3-C3D)XRSS (11)
If equation (11) is less than 0.4 mg/L apart from
DMATX (1,N) , the calculated depth is satisfactory.
If not then let
C
= C3 C12)
3 = C3 + 0.5 (DMATX (1,N)-C5) (13)
C3D = C3D
7. Calculate C~ and C-D and return to Step 3
8. If the calculated C,- is satisfactory continue through
the trickling filter subroutine.
Table 4 lists the FORTRAN process equations for the modified
subroutine. Note that
BOD2 replaces C~ DBOD2 replaces C2D
BOD4 replaces C3 SBOD4 replaces C3_
DEPTH replaces D SBOD5 replaces C5g
DBOD4 replaces C3Q
Since the G and G model does not include an equation for nitrogen
removal analogous to equation (6) , the degree of nitrification
and denitrif ication was assumed proportional to the ratio
BETAN/BETA. Both BETAN and BETA equations are retained from
the original TRFS version.
114
-------
CASE V WORKSHOP
MODIFICATION OF EXISTING EXEC
PROGRAM COST RELATIONSHIPS
Barry F. Winkler
Metropolitan Sanitary District of Chicago
100 East Erie Street
Chicago, Illinois 60611
ABSTRACT
The Exec Program is modified to incorporate concepts
of the cost-effectiveness guidelines of the USEPA in evaluating
the overall monetary worth of a system. The modifications in-
clude staged (or delayed) construction, varying growth rate in
process operating costs as functions of time, incorporation of
varying process lives and salvage values, and interest during
construction on an other than a straight-line basis.
115
-------
INTRODUCTION
The purpose of this case is to introduce the concept of the
time value of capital and operating costs, within the constraints
of USEPA's "Cost-Effectiveness (C-E) Guidelines for Facilities
Planning," in evaluating the monetary costs of alternative sys-
tems, staged construction, or varying growth rates in process
utilization factors.
On the basis of a 25-year planning period, the following
factors were incorporated:
(1) An analysis of maintenance and operating (M&O) costs
based on year-to-year anticipated growth in expendi-
tures instead of design capacity.
(2) An analysis of the effect of time delayed construction
on the inflation-free criteria of the C-E guidelines.
(3) An analysis of the effect of incorporating the salvage
value, as defined in the C-E guidelines.
(4) An analysis of the interest-during construction on a
basis other than straight line cash flow.
The incorporation of these concepts are necessary, because
in evaluating the monetary worth of alternative systems for
which Federal funding is sought under PL 92-500, it is necessary
that the systems be evaluated on the basis of their relative
overall monetary costs. These cost analyses reflect trade-offs
between capital and operating expenditures, within the con-
straints of the planning period.
The changes incorporated in this case also enable the
planner or design engineer to evaluate the marginal effect of
varying the projected growth rate in M&O costs. These costs
usually represent 30 to 80 percent of a project's total estima-
ted cost. Any one of four unique functional relationships can
be evaluated. These include fourth degree straight line, ex-
ponential, sine-squared, or constant, as well as the constants
within these functions.
These changes are incorporated into the Exec Program by:
(1) adding a present worth subroutine for M&O Costs (PREWO),
(2) requiring the user to add some or all of the design
input data, and
(3) changing the cost subroutine.
The input data includes DMATX (21 to 27, 1 to 20) for selecting
116
-------
construction time scheduling and M&O growth factors in evalua-
ting capital costs (PWCP) , salvage value (SALVG) , and M&O costs
(POW) . The cost subroutine modifications are made to evaluate
interest during construction as a non-linear distribution for
projects requiring in excess of three years to construct.
Appendix A contains a glossary of new terms added for this
analysis .
For comparative purposes this case is designed to run paral-
lel to the base Exec Program. The effect of the C-E guidelines
can be evaluated by comparing the appropriate summaries, both of
which appear in the printout.
EXAMPLE PROBLEM
The user is required to input up to seven pieces of data
DMATX positions 21 to 27 for each of up to 12 process units
included. The user has the option to input up to five additional
pieces of data in DMATX positions 28 to 32 for any of the in-
cluded process units. Special considerations for process 20
(i.e. Admin., lab., etc.) which has no required input data, but
eight optional data points to modify preset values.
The data must be added in F-format for each of N processes
used, (except N=20) , as shown in Appendix B. The values are
punched onto the data cards following DMATX (20, N) as follows:
(a)
DMATX (21, N) - DBS (N) - the decision route for deter-
mining the rate of growth in percent utilization of
M&O costs; Cols. 41-50 on card #12 (see Appendix B) .
If DBS (N) = 1.0, then M&O = A (1 ,N) [l.+A (2 ,N) * T -
A(3,N) *T + A(4,N) *T]
A ( 9
V '
If DBS (N) = 2.0, then M&O = A(1,N) * e
If DBS (N) = 3.0, then M&O = A(1,N) + A(2,N) *
2
Sin
Max
(b)
(c)
If DBS (N) = 4.0, then M&O = constant = COSTO (N,l+5)
DMATX (22, N) = A(1,N), Do Not Use, calculated from
FUT(N) = COSTO (N) and T = TMAX(N).
DMATX (23, N) = A(2,N) * 1000, required if DBS (N) =
1.0, 2.0, or 3.0; Cols. 61-70 on card #13.
117
-------
(d) DMATX (24,N) = A(3,N) * 1000, required if DES(N) =
1.0, Cols. 71-80 on card #13.
(e) DMATX(25,N) = A(4,N) * 1000, required if DES(N) =
1.0, Cols. 1-10 on card #14.
A(2,N), A(3,N), A(4,N) are automatically set at 0.
unless data is otherwise provided.
(f) DMATX(26,N) = TIME(N) = Year process N is placed in
service, Time = 1 to 25., Cols. 11-20 on
(g) DMATX(27,N) = TMAX(N) = Year process reaches full
capacity, Cols. 21-30 on card #14.
(h) DMATX(28,N) = AIFE(N,1) = Life of each No. 1 sub-
process in years, Cols. 31-40 on card #14.
(i) DMATX(29,N) = AIFE(N,2) = Life of each No. 2 sub-
process in years, Cols. 41-50 on card #14.
(j) DMATX(30,N) = AIFE(N,3) = Life of each No. 3 sub-
process in years, Cols. 51-60 on card #14.
(k) DMATX(31,N) = AIFE(N,4) = Life of each No. 4 sub-
process in years, Cols. 61-70 on card #14.
(1) DMATX(32,N) = AIFE(N,5) = Life of each No. 5 sub-
process in years, Cols. 71-80 on card #14.
AIFE values are preset at 25.0 years unless data is
otherwise provided.
If an entire subroutine process life is to be changed
from 25.0 years each of the required number of sub-
process (equal to the i-value of the size of the CCOST
(N,i) matrix), must be changed.
If the process (or sub-process) life is set so that
the units useful life expires before the end of the
design period, the program assumes that a duplicate
process facility, (or multiple process facilities),
will be constructed at the time-adjusted capital cost
and placed in service at the time the facility expires,
i.e. Assume $1,000 worth of centrifugation at AIFE =
10 years is required in year TIME = 0., and DESIG =
25.0 years.
Thus, if RI = 5%, then the capital cost is calculated
as follows:
118
-------
CCOST = CAPTAIL COST - SALVAGE VALUE
CCOST = $1000(1+0.614 + 0.377) - $1000 (0.5) (0.295)
= $1,991 - $147 = $1,844
Similarly, the percent utilization and M&O growth
curve is assumed to be continuous from the point the
initial process life expires.
For the 20th process, (i.e. N=20), Administrative, Labora-
tory and associated overhead costs, the provisions are made for
the inclusion of the following eight pieces of data in positions
DMATX(21+26 and 28+29, 20) on cards #8 and 9, APPENDIX B. Each
of these design matrix data has a preset internal default value
as indicated, if none is provided.
(a) DMATX(21,20) = AMATX = number of different processes
used, Cols. 41-50 on card #8, present at 19, (i.e.
N=19+l = 20).
(b) DMATX(22,20) = Inflation rate for capital expenditures
in percent per year, Cols. 51-60 on card #8, preset
at 0%.
(c) DMATX(23,20) = Average inflation rate for M&O expen-
ditures in percent per year, Cols. 61-70 on card #8,
preset at 0%.
(d) DMATX(24,20) = DESIG = Design period in years, Cols.
71-80 on card #8, preset at 25.0 years.
(e) DMATX(25,20) = YER = Maximum number of years for which
interest during construction can be assumed to be
projected to be expended on a straight line basis,
Cols. 1-10 on card #9, preset at 3.0 years.
(f) DMATX(26,20) = TIME(20) = Year process N=20 is placed
in service. Cols. 11-20 on card #9, preset at 0 years.
(g) DMATX(27,20) = Blank not used, Cols. 21-30 on card #9,
preset at 0.
(h) DMATX(28,20) = AIFE(20,1) = Life of Number 1 sub-
process for N=20, Cols. 31-40 on card #9, preset at
25.0 years.
(i) DMATX(29,20) = AIFE(20,2) = Life of Number 2 sub-
process for N=20, Cols. 41-50 on card #9, preset at
25.0 years.
119
-------
INTERPRETATION OF OUTPUT
The modifications to the Exec Program included in this case
are designed to provide the following printouts:
(1) A list of the processes and stream designations used
in the run.
(2) A listing of the stream characteristics as used in
previous case analyses.
(3) A listing of process characteristics for the basic
Exec Program.
(4) A listing of Total Plant Costs for the basic Exec
Program
(5) A listing of Total Cost-Effectiveness Plant costs
to be compared to the four totals (i.e. TOT, TAMM,
TOPER and TOTAL) listed for the total Plant Costs.
(6) A comparative listing of the Case Five C-E Analysis
by process, where;
CE M&O represents the present worth M&O costs
from this analysis,
COSTO represents the alternative present worth
M&O costs at design capacity, from the basic
Exec analysis,
CE CAP represents the present worth capital costs
allowing for delayed construction,
CCOST represents the present worth capital expen-
ditures not adjusted for delayed construction
SALVG represents salvage value to be subtracted
from capital costs when equipment life exceeds
the design period,
CE TOT represents total present worth of CE M&O,
CE CAP and SALVG for the process,
TOTAL represents comparable total present worth
of basic Exec analysis,
TIME represents the year the process was placed in
service,
TMAX represents the year process reaches full
capacity,
-------
(7)
DES represents the functional M&O relationship
chosen 1, 2, 3, or 4, and
IF represents the N value corresponding to the
process designation previously printed.
Listing of input data DMATX(21+32,N) for each of 20
processes.
By modifying the design matrix data, either in the selection
of the empirical equation used for the M&O analyses or in the
construction scheduling (or life), the effect on the overall
relative costs are readily apparent.
From the input data present in Appendix C, the following
comparative results, summarized in Appendix D, are obtained:
Total Capital, $ x 10
Amortized Capital, C/1000 gals
M&O, C/1000 gals.
Total Treatment, C/1000 gals.
(*including salvage credit)
Exec Program
$6473.00
13.87
9.10
22.97
CE Analysis
$4800.00
10.29
7.10
16.84*
The difference in total capital between the Exec program and
the CE analysis is due to two factors;
(1) delayed construction of various segments of the
system up to as much as ten years at 0% inflation,
(2) salvage value credits for processes whose useful lives
are projected to extend beyond the end of the period.
The aeration tanks (AERFS), based on a 35 year salvage credit,
or $93,000 (12.5% of capital), was the most significant. The
overall difference being nearly $1.7-million. A comparable
difference is also evident in the amortized capital items. The
itemized Exec capital items on Appendix D are the same as those
which appear as CCOST in Process Characterises. If the process
in question is placed in service at time 0, the CE Capital
expenditures will be nearly equal to the "Exec Cap" value,
differening only by the respective ratio values which pro-rate
the overhead capital expenditures.
The M&O costs by process in Appendix D are equal to, or less
than the Exec M&O costs (COSTO) for each analysis. The degree of
difference is a function of the process parameters chosen.
121
-------
Internally the progarm calculates the A(1,N) value for each
process, having one fixed point COSTO (converted to $/year) at
100% utilization (i.e. TMAX year). In addition, values for
A(2,N) through A(4,N) will affect the calculation, depending on
the processes included.
The overall M&O unit costs cited above reflect this selec-
tion of parameters. Care must be exerted to avoid choosing
parameters which will generate negative CE - M&O values.
Appendix D will alert the user to this unique process or pro-
cesses that are improperly defined.
Lower M&O unit costs (on a CE basis) plus the amortized
capital costs result in a 22.4% decrease in the total treatment
cost. If this system were being evaluated against others,
changes of this magnitude could effect the "best solution".
Similarly these results can be used to evaluate the sensitivity
of the M&O costs on the system by varying the parameters used
to calculate these present worth values (and subsequently unit
cost).
122
-------
FUT(N)
O
U
O
08
s
(100% UTILIZATION)
M&O, $/yr - PUT(N)
M&O, $/yr =
[l.+A(2)*T-A(3)*T2+A(4)*T3]
0 TIME(N)
TIME T, yrs
DESIG TMAX(N)
PRESENT WORTH OF M&O COSTS FOR PROCESS N:
T=DESIG
PW(N) , $= Z [(A(1,N)) (1+A(2,N)*T-A(3,N)*T2+A(4,N)*T3)*(SPWF(T))]
T=TIME(N)
WHERE :
,N) = FUT(N)/[1.+A(2,N)*TMAX(N)-A(3,N)*TMAX(N)
+A(4,N) *TMAX(N) 3]
SPWF(T) = l./d.+RI)
RI = RATE OF INTEREST
Figure 1. Maintenance and operating cost curve for DES(N)=1.
123
-------
FUT(N)
O
O
O
•a
g
(100% UTILIZATION)
M&O, $/yr =
FUT(N)
X
I M&O, $/yr = A(l)*eA(2)*T
0 TIME(N) DESIG TMAX(N)
TIME Tf yrs
PRESENT WORTH OF M&O COST FOR PROCESS N:
T=DESIG
PW(N), $ = I [(A(l,N)*eA(2'N)*T)*(SPWF(T))]
T=TIME(N)
WHERE:
A(1,N) = FUT(N)/eA(2'N)*TMAX(N)
SPWF(T) = l.(
T
Figure 2. Maintenance and operating cost curve for DES(N)=2
124
-------
FUT(N)
CO
O
(100% UTILIZATION)
M&0=FUT(N)
TIME(N)
TIME T, yrs
PRESENT WORTH OF M&O COSTS FOR PROCESS N:
T=DESIG
,2
DESIG TMAX(N)
PW
(N)= Z [rA(l,N)+A(2,N)*SIN
.
(SPWF(T))
•]
WHERE:
SPWF(T)
FUT(N)-A(2,N)
= l./(l.+RDT
Figure 3. Maintenance and operating cost curve for DES(N)=3.
-------
FUT(N)
i-i
>i
cn
o
u
o
ca
g
M&O, $/yr = A(l) = FUT(N)
0 TIME(N)
DESIG TMAX(N)
TIME T, yrs
PRESENT WORTH OF M&O COSTS FOR PROCESS N:
T=DESIG
PW(N), $ = I A(1,N)*SPWF(T)
T=TIME(N)
WHERE:
A(1,N) = FUT(N)/eA(3'N) X TMAX(N)
SPWF(T) = l./(l.+RDT
Figure 4. Maintenance and operating cost curve for DES(N)=4
126
-------
Table 1
TOTAL PLANT COST
TOTAL CAPITAL, $ x 1000
TOTAL AMMORT. CAP., C/1000 GAL.
TOTAL M&O, C/1000 GAL.
SALVAGE (CREDIT), C/1000 GAL.
TOTAL TREATMENT, C/1000 GAL.
CCI
WPI
RI
YRS
DHR
PCT
DA
CCINT
XLAB
CKWH
RATIO
TCAP,
YARD,
TCC, $
XLAND,
ENG, $
SUBT1,
FISC,
SUBT2,
XINT,
ACRE
AF
$ x 1000
$ X 1000
x 1000
$ x 1000
x 1000
$ x 1000
$ x 1000
$ x 1000
$ x 1000
EXEC
PROGRAM
$6473.0
13.87
9.10
(N.A.)
22.97
2.257
1.675
0.060
25.0
4.73
0.150
1000.0
0.060
1.0
0.02
1.331
4863.0
681.0
5544.0
20.0
459.0
6023.0
39.0
6062.0
411.0
20.0
0.078
C-E
PROGRAM
$4800.0
10.29
8.09
(0.55)
17.83
2.257
1.675
0.060
25.0
4.73
0.150
1000.0
0.060
1.0
0.02
1.335
3596.0
504.0
4100.0
20.0
358.0
4478 0
34.0
4512.0
289.0
20.0
0.078
127
-------
EH O
O
U o
S3
W to-
CN
n
r^
co
co
W
U
O
o
O o
> o
1-5 -H
CO X
to-
ioo»rM
in
CO
CO
I X
o
o
o
00
•s
u
w
a o
o
U o
«* «3 00
CN «N r-t
o r-~ CN oo oo r~
CN .H i-l CN
•
CO
O
CJ
O
-------
APPENDIX A
GLOSSARY OF ADDITIONAL TERMS USED FOR CASE FIVE ANALYSIS
1. System life;
(a) AIFE (20,5) = life of each subsystem in each of the
20 subroutines.
2. Capital costs;
(a) FUCAP (20) = Capital cost adjusted by CCI index,
$ x 10 , working value FUCP.
(b) PWCP (20) = Present Worth of future capital expenditures
delayed in time, $ x 10, working value PCP.
(c) TOTX = Sum of Capital costs without salvage value
adjustment.
3. Maintenance and Operating Costs;
(a) FUT (20) = Annual M&O cost, $/yr with system at 100%
capacity, $ x 10 , working value, FT
(b) PW (20) = Sum of PMOX's, where PMOX represents the calc.
M&O cost (in present worth terms) for each year of
operation, $ x 10 ,
(c) PWMO = Sum of PW(N) for N processes, $ x 10 ,
(d) TOPRX = Total Operating Costs, <=/thousand gal.
4. Salvage Values;
(a) SALVG (20) = Salvage Value of capital at end of design
period, $ x 10-3, (Present worth basis),
(b) TSALV * Sum of SALVG (N) for N processes, C/thousand gal.
5. Total Costs;
(a) TAMMX = Ammortized Capital, ^/thousand gal.,
(b) TOTLX = Total Unit Treatment Cost,
-------
6. Time Constraints:
(a) TIME (20) = Year process placed in service, T=l -*• DESIG
(b) TMAX(20) = Year process reaches full capacity.
(c) DESIGN = Design period, 25 years.
(d) Tl = Working "AT" in PREWO subroutine.
(e) T = Working TMAX(N) in PREWO subroutine.
7. Interest Rates:
(a) RI = Amortized interest rate, fraction.
(b) CCINT = Interest rate for the cost of interest during
plant construction, fraction.
(c) DMATX(22,20) = Projected annual inflation rate for
capital expenditures, percent.
(d) DMATX(23,20) = Projected annual increase in M&O rate
scales, percent.
(e) CNT=CNTCP = Difference between RI, amortized interest
rate, and DNATX(22,20), inflation factor, fraction.
(f) CNTMO = Difference between RI, and DMATX(23,20), the
projected M&O wage growth, fraction.
8. Miscellaneous
(a) DES(N) = Decision variable (input) in choosing one
of four M&O formats.
(b) A(1,N) to A(4,N) = Parameters required for M&O functions,
(c) AMATX = N, number of processes used in the analysis.
130
-------
CO
H
en
>H
sl
JZ
J3*
en
cn
§
?
n
£4
CJ
a
CM
fri
Ed
p_,
en
a o
CJ
X
H
Q ••
Z W
W >
& rH
^ ^
cn
rj
O
a
M
a
S
flj
£H
*J«
a
tB
4->
IB
Q
§
(U
-U
H
4fc
•0
^^,
CN ~
O
• 00
O rH
CJ
to
ta -i
G O
3 H
HH »
O <0
• 4J
0 -H
Z EH
0)
en rH
3 -H
« EH
rH CN
o
• 00
O rH
CJ P-
o
• r-
rH 1
O H
O &
o
• to
H 1
O rH
cj m
o
. in
rH 1
O rH
CJ •*
O
rH 1
O H
CJ ro
O
• ro
—4 |
0 H
U CM
0
• CN
rH 1
0 -H
O rH
• O
H rH
O 1
CJ rH
H
Q 1 Z 1 1
CO O 1 M 1 1
en to i cj i i
> a i u i i
Q 1 II
O « 1 II
« j i < i i
cn «!i a i i
i I i
S S 1 H 1 1
PL, fc. 1 CJ 1 1 '
en a i cu i i
X
1 1 EH
O fit 1 B 1 S
en a i Q i «2
i ii
z i cn i i
O Z 1 P4 1 I
en Q i >H i i
CJ CJ 1 II
Z Z 1 H 1 |
en Q i ai i i
Z
1 X 1
8CJ CN H S 1
O O O4 « 1
CO Q Z 3 CJ 1
Z
i m i
en ro w ft \
en 33 cj ,4 i
a EH z u x i
^-* ^-*
^ rH --^ 0
—, -H - O CN ~
rH » 00 CN - O
» VO rH •. O CM
00 i-H. t 00 rH
t t r- t t rH
rH Cr, rH rH CTI CM
X X X X X X
EH EH EH EH 6H EH
i^ tfj rtj ri, i
» in CN | rH *-~
r~ rH ^- | G *>
— — < 1 0 g
g
'H >0 rH
--^. 2 ^ CT\ ^ O
^. Z — 0) IT) U
Z - 1 X CO to O
- rH | ft 3 M
ro H 1 S d) r( O
~^ ~- I EH 01 > O >H
en n) IH
a> x: -tJ
1 — rj 4-> U(
— . Z O T3 D< H H eu O
Z - 1 W P, 3 0 X
« 0 I g O X n)
U IT) 6
M CD Ej w
^ ^ ^ O TD >H O
Z Z 1 Z >H O tn
» « | v IT) MH
rH cn i n- 73 4j H
— -~- \ — CD td bj
-------
Card
APPLNDIX c
DATA USED IN CASE FIVE EXAMPLE
9 PROCESS, N=l
10 DMATX{i-»8,N)
11 DMATX{a*16,N)
12 DMATX(21-v24,N)
13 DMATX(25-*-27,N)
14 PROCESS, N=2
15 DMATX{1*8,N)
16 DMATX(9+16,N)
17 DMATX(21-24,N)
18 DMATX(25+27,N)
19 PROCESS, MIX
20 PROCESS, N=3
21 DMATX(l-*-8,N)
22 DMATX(9-16,N)
23 DMATX(21-*24,N)
24 DMATX(25*27,N)
25 PROCESS, N=4
26 DMATXU+8.N)
27 DMATX(9>16,N)
28 DMATX(21-*24,N)
29 DMATX(25*27,N)
Col.
1-10 J
10.0
254.0
15 0
2.25:
1
Col.
1
0
Col.
1-10
in n
Col.
1
0
Col.
1-10
l n
Col.
1
0
0
'Col.
1-10
0.05
Col.
1
0
Col.
1-10
13.0
30.0
nc
Col.
11-20
105.0
43.0
0 0
1.675
.02
Col.
3-4
1
Col.
11-20
Col.
3-4
2
Col.
11-20
Col.
3-4
0
3
Col.
11-20
5.0
coi.
3-4
4
Col.
11-20
2000.0
0
in n
Col.
21-30
3U.O
11.0
.06
1.0
Col.
7-8
15
Col.
21-30
Col.
7-8
1
Col.
21-30
Col.
7-8
4
2
Col.
21-30
Col.
7-8
3
Col.
21-30
20.0
en n
Col.
31-40
10.0
19.0
25.0
Col.
10-15
RWP
Col.
31-40
Col.
10-15
•PREL
Col.
31-40
Col.
10-15
MIX
PRSET
Col.
31-40
Coll
10-15
AERFS
Col.
31-40
1.0
Col.
41-50
2.0
4.0
4.73
Col.
20-21
1
Col.
41-50
Col.
20-21
2
COI.
41-50
1 n
Col.
20-21
3
4
Col.
41-50
Col.
20-21
5
Col.
41-50
1.0
1.2
1.0
Col.
51-60
30.0
500.0
.15
Col.
30-31
0
Col.
51-60
Col.
30-31
0
Col.
51-60
Col.
30-31
19
0
Col.
51-60
Col.
30-31
0
Col.
51-60
0.5
1.0
Col.
61-70
140.0
250.0
1000.0
Col.
40-41
2
Col.
61-70
Col.
40-41
3
Col.
61-70
500 0
Col.
40-41
4
5
Col.
61-70
In
Col.
40-41
6
Col.
61-70
3.0
1.0
5.0
Col.
71-80
224.0
60.0
.06
Col.
50-51
0
Col.
71-80
1.0
Col.
50-51
0
Col.
71-80
40 0
Col.
50-51
0
7
Col.
71-80
1 2
i n
Col.
50-51
8
Coa.
71-80
700.0
1.2
1.0
* DHARX(22-24,20) optional.
** Card no. 9 with DMATX(25-29,20) optional data omitted.
Card must be included.
132
-------
APPENDIX C
-------
APPENDIX C
(Cont.)
59 PROCESS, N = 10
60
61
62
63
DMATX(1-»8,N)
DMATX(9-16,N)
DMATX(21-24,N)
DMATXl25*27,N)
64 PROCESS, N=ll
65
66
67
68
£9
DMATX(1-»8,N)
DMATX(9-»16,N)
DMATX(21-»24,N)
DMATX(25-»27,N)
Col.
1
0
Col.
1-10
2.0
Col.
0
Col.
1-10
8.0
Col.
1
9
Col.
3-4
10
Col.
11-20
1.0
Col.
3-4
11
Col.
11-20
30.0
10.0
0
Col.
7-8
14
Col.
21-30
35.0
Col.
7-8
12
Col.
21-30
220.0
30.0
0
Col.
10-15
MHINC
Col.
31-40
5.0
Col.
10-15
CHLOR
Col.
31-40
2.5
0
Col.
20-21
14
Col.
41-50
0.
Col.
20-21
25
Col.
41-50
180.0
4.0
0
Col.
30-31
0
Col.
51-60
10000.0'
Col.
30-31
. 0
Col.
51-60
1.2
0
Col.
40-41
15
Col.
61-70
1.0
Col.
40-41
26
Col.
61-70
1.2
0
Col.
50-51
0
Col.
71-80
0.3
Col.
50-51
0
Col.
71-80
1.5
0
134
-------
a
E-
£
Cf
[1
PC
b
Q
a
/a
5
2
c?
0 ^ o,
•9 EH in
W 2 ^
o a
o M < m
o ri EH H
rH ^ O 00
EH 4 O
(0 M > 0
CO ^ 3 <«•
) H CO
I CO
J
*5
^ rj ^* rH
] V>
i
1 r i
pq O CM
1 H X CM
t/>
«
)
.
&1 O CM
U "* °
i u>
1
CO
W H
Q
X
< oa in
S >< CM
EH
|g -
CO ~»
co m
H H
O ~
^
t~
^f
00
CM
^"
to
r-t
CM
CM
^*
O
CM
rH
•O
CM
in
CM
CM
rH
to
CM
-
^^
rH
s
a.
o
en
m
in
to
^*
r~
o
^*
^t
rH
m
0
CO
CO
00
rH
•*
r-
rH
rH
in
in
^
CM
EH
H
a.
^,
CO
m
rH
in
CO
00
ao
CM
ro
rH
CO
en
^«
r-
to
o
CM
m
00
rH
rH
O
m
o
rH
_
CO
CO
to
m
CM
to
o
CM
VO
ao
rH
rH
to
H
O
r*.
CM
CO
CM
0
ro
rH
^^
00
«
g
ro o
m r-t
t** r^
in co
r-t tO
in to
in CM
CM CM'
to in
CM ^*
01
00 •*
£ 5
00 00
CM 00
rH rH
OO CO
to r-
rH
CM CO
0 0
co ro
in rH
^_^
rH
— CM
^o ^
— CM
o o
H H
a Q
co in
r*^ ro
CM CM
rH
r** co
rH ^*
CM r-
00 rH
00 to
rH en
r*. r^
to
rH 00
V CO
rH in
m •*
oo r~
CM
CO CO
oo r~
CM
ro ro
O o
CO CO
in o
rH
— ^ ^-.
to r~
EH" U
£C
to
to
o
•-H
in
^*
ao
CO
r-
c—
r~
CM
en
in
CM
en
CM
CM
en
CM
*J*
o
CO
in
?
rH
CJ
•z
^
rH
in
en
t—
ro
to
CO
ao
r-t
r-t
r-t
to
u>
en
^o
CM
m
OO
rH
to
en
H
to
in
CM
rH
in
i-H
CO
r-t
m
^*
r- o
00 O
CM 00
^j*
in-
f r-t
CO rH
r~ en
CM
•* 00
ro oo
r~ in
^
vs-
^*
o
0
£t
nj
i-J
•
• O
C -P _5
•HO) «;
'O O
i4« EH
135
-------
CASE VI WORKSHOP
ADDITION OF A GRANULAR BED FILTRATION
SUBROUTINE TO THE EXEC PROGRAM
S. C. Chay
Argonne National Laboratory
Argonne, Illinois 60439
and
Raymond D. Letterman
Department of Civil Engineering
Syracuse University
Syracuse, New York 13210
ABSTRACT
A subroutine is described which can be included in the Exec
Program and used to estimate the performance and cost of a
wastewater treatment system which includes a granular bed fil-
tration process. The subroutine uses filter design and opera-
ting conditions such as the filtration rate, media size distri-
butions and influent SS concentration to calculate the filter
plan area requirement when the filter run length is limited by
the headless constraint. The plan area requirement is the basis
for determining the costs associated with the filtration process.
Details of how to incorporate the subroutine into the Exec
Program and sample results are described.
136
-------
INTRODUCTION
The importance of granular media filtration in the treat-
ment of wastewater has risen dramatically with the implementa-
tion of the Federal Water Pollution Control Act Amendments of
1972. Lykins and Smith (1) have reported that over 1500 treat-
ment plants will apply tertiary filtration in order to meet
current water quality standards. An equivalent number of
plants will be required to meet anticipated standards by 1985.
The Exec Program in its present form does not contain a
unit process subroutine for granular media filtration. The
purpose of this case study is to outline the derivation of such
a subroutine and to describe how it is incorporated into the
existing Exec Program.
FILTRATION SUBROUTINE
Flow Diagram
Figure 1 shows the general configuration of the filtration
system used in this analysis. Note that while this is a typical
system, there are a number of variations of this general scheme
in use. For example, the backwash water holding tank is some-
times omitted or replaced by a clarifier. Equalization tanks
are used in some installations prior to the filters. In some
cases, a separate wet well may be used in place of the chlorine
contact unit as a source of backwash water.
Design Equations
The approach used in this analysis was to base the capital
cost of the filters on the total plan area of the filter beds
(A). Ives (2) has evaluated this approximation and reported
that it is reasonable. Huang and Baumann (3) have also des-
cribed its use.
The magnitude of A can be determined using the following
equation,
A = NWP' (1)
where Q is the design raw wastewater flow rate, and NWP is the
net filtered water production per unit plan area per unit time.
The net filtered water production per day is calculated
by subtracting the backwash volume per run from the filtered
water production per run (using a per unit plan area basis)
and then multiplying this result by the total number of filter
runs per day. In equation form this is,
137
-------
01
C w
•H O
n ca
O 4-1 4J
i-H fS -H
,n o e
u u :=>
1
to oo
O i-l C
id o n)
PP EC H
CO
1
o
<0
O
CO
p.
0
P-.
o
§
CO
R)
O
cd
CO
n
cu
CO.
S m
§• 0
I u
o
l-l (U
« -H
T3 M-l
C -H
O M
o n)
(1) rH
w o
C 3 DK
o 4J pa
30) +
a
H
§ c
S o
y -H
ca 43
PQ
(0
M
^J
iH
•H
En
0)
X!
-P
M-i
O
rt
•H
a
o
•tH
4J
«0
e
CD
X!
O
CU
cn
•H
138
-------
9
NWP (gal/ftVday) = WP = (QB) (TB) - , (2)
— + TB
QF
where ,
WP is the filtered water production per filter run
(gal/ft2) ,
2
QB is the average backwash flowrate (gpm/ft )
TB is the filter down-time per backwash (min) , and
2
QF is the filtration rate (gpm/ft ) .
The filtered water production per filter run (WP) can be
determined using an expression which is based on a simple mass
balance across the filter bed. This equation is,
where,
FK is an experimentally determined coefficient proportional
to the mass density of the deposit within the filter
(gal mg/ft3!)
F is a fraction between 0 and 1, the magnitude of which
depends on the distribution of deposit within the filter
bed,
E is the fractional removal efficiency of SS across the
filter bed,
CO is the steady state influent suspended solids, concen-
tration (mg/1) ,
NF is the number of equal depth layers used in analyzing
the filter bed,
D is the overall depth of the filter bed (ft) ,
AH is the overall terminal headloss (ft of water) , and
AK(I) is the clean bed headloss across layer I per unit
layer depth and per unit filtration rate (ft2/gpm) .
139
-------
The derivation of Eg (3) and methods to determine FK and
F have been described by Letterman (4).
The total backwash volume per day, BW, is calculated by
multiplying the backwash volume per filter run by the number of
filter runs per filter per day, i.e.,
i 44n —fi
BW (mgd) = [A] [ (QB) (TB) ] wp (10 b) (4)
The mean concentration of suspended solids (SS) in the recycled
backwash water is given by,
(WP) (.E) (CO) . .
bb (QB)(TB) (*}
The volume of the backwash water holding tank, V, (see
Figure 1) can be determined by assuming that its volume should
be equal to the volume of water produced by the backwashing of
all the filter beds in rapid succession. This is given by,
V = [A]((QB)(TB)] (6)
The flowrate capacity, BP, of the backwash pumps is given
by,
BP = (AB) [|] , (7)
where M is the number of individual equal-sized filter beds in
the system.
The flowrate capacity of backwash recycle pumps is simply
the backwash recycle flowrate, BW, as given by Eq. (4).
Design Equations - Assumptions Used
The following assumptions were made in deriving the design
equations.
1. The filtered water production per filter run is deter-
mined by the overall headless constraint and not by
effluent quality.
2. The suspended solids removal efficiency is constant
during the filter run. Extensive field studies by
FitzPatrick and Swanson (5) support both of the above
assumptions.
3. The filtration rate is constant during the filter run.
This type of operation is common in wastewater filtra-
tion. However, the equations can be modified and used
140
-------
to evaluate a system with declining rate of filters.
4. The system is operating under steady state conditions,
i.e., the concentration, physical/chemical characteris-
tics, etc. of the filter influent SS are constant with
time.
Filtration takes place within the media, i.e., there
is negligible cake formation on top of the bed. This
is also supported by FitzPatrick and Swanson (5) who
have observed that under most field conditions, where
dual or multi media beds are used, the suspended solids
penetrate the top surface of the bed.
Cost Equations
The cost equations used in this analysis are based on a
set of expressions developed by Van Note et al (6) for a dual
media filtration system receiving secondary effluent. The
system used in Van Note's et al. cost analysis is essentially
the same as the one shown in Figure 1. The equations developed
lump together the individual units in the system (pumps, filters,
holding tank) and express their overall cost as a function of
the flowrate. These equations have been converted from a flow^
rate to a unit filter plan area basis using the filtration rate
which Van Note et al assumed in their analysis. An additional
equation has been included for the backwash electrical costs.
These equations are listed below.
1. Capita cost, C (in January, 1971 dollars) of the filter
system including backwash water storage and all pumps
and piping
C = 6378.1 A°*66; (8)
2. Base man-hour requirement, BMH (in man-hours/year),
p^
BMH = 0.1224 + 0.00058A ; (9)
3. Base material costs, BMC (in January, 1971 dollars/
years), 451.33 AO-68 (10)
4. Variable O & M costs (excl. backwashing electrical
costs), COMV (in C/1000 gal.),
= BMC (— ) (— ) (11)
where WPI is the wholesale price index of industrial
commodities for the year to be used as a basis for
costs.
141
-------
5. Fixed 0 & M costs, COMF (in C/1000 gal f 1
COMF = (BMH) (MHR) Q (12)
where MHR is the labor rate in $/man-hour.
6. Electricity cost for backwashing, ECBW (in C/IOOO gal.)
FPRW - 114fi (BW) (HP) (CKWH) , 1 . n_.
ECBW - 1146 - - (3650~Q} (13)
where HD is the total dynamic pumping head in feet
(including backwash and recycle)
efficiency (decimal) and CKWH is the per unit KW hour
electrical power costs ($/Kw-hr) .
Cost Equations - Assumptions Used
The following assumptions were used in adapting and apply
ing Van Note's et al. (6) cost equations.
1. It was assumed that the cost of the overall filtration
system can be determiend using the filter plan area as
the critical design parameter. In most cases this
assumption is made reasonable by the fact that the
filter beds are the dominant cost item in the filtra-
tion system. As shown by Equations (4) , (6) and (7) ,
the sizes of other components in the system (backwash
holding tank, backwash and recycle pumps) are propor^
tional to the filter plan area, however, they are also
a function of design parameters such as the backwash
rate and duration, and the number of filter beds in
the system. Therefore, for example, if the objective
is to analyze the effect of the backwash rate on the
treatment system performance and cost, it may be
necessary to use individual cost equations for the
system components rather than the more comprehensive
equations shown.
2. The use of the filter plan area as the critical design
parameter implies that the cost per unit area of filter
is a function only of the size of the plant. Therefore,
although it is possible to analyze the effect of the
terminal headless and overall depth of the filter
bed on system cost, caution should be used in varying
these parameters as it is likely they determine to some
extent the cost per unit plan area of filter bed A
more detailed cost breakdown for the filter beds would
be necessary to correct this shortcoming.
142
-------
PROGRAMMING
Incorporating the Design and Cost Equations in the Executive
Program
The design and cost equations described in the previous
sections were combined in a filtration(FILT) subroutine (see
Appendix A). A computational flow chart for the FILT subrou-
tine is given in Figure 2 . A symbol for the process with
input and output stream designations is shown in Figure 3
Listings of the contents of DMATX and OMATX for the FILT sub-
routine are given in Tables 1 and 2. Modifications were
necessary in the EXECMAIN program and the PRINT subroutine in
order to call the filtration subroutine and to print the new
input and output quantities. The specifics of these modifica-
tions are described in the following section.
Modifications of the original program.
1. EXECMAIN - A listing of the modified portions of EXEC-
MAIN is given in Appendix B. Two statements have been
added to the original program so that the FILT sub-
routine is called by EXECMAIN. In addition the GO TO
statement was modified. Both changes are shown in
Appendix B.
2. Subroutine PRINT -A listing of the modified portions of
subroutine PRINT is given in Appendix C. Five state-
ments, numbered 230 to 240 have been added to the
original subroutine so that the decision and output
matrix parameters listed in Tables 1 and 2 are printed.
EXAMPLE RESULTS
The treatment system diagrammed in Figure 4 was used to
illustrate the application of the FILT subroutine. The config-
uration shown is a typical activated sludge system with
tertiary granular bed filtration. The backwash water in this
case is drawn from the chlorine contact unit and recycled after
use to a point just before the preliminary operations. Also
shown in Figure 4 are the recycle loop numbers (K) and the
process stream numbers.
In these examples it was assumed that the proportionality
constant, FK, in Eq. (3) is equal to 5.7 x 10 gal-mg/ft^-1.
This value was determined by Letterman (7) in a pilot plant
study of the filtration of clay suspensions treated with cationic
polyelectrolytes. It is possible that the value of FK for
secondary effluent particulate matter is significantly different.
However, a rough test of the above value using data on biological
solids capture per unit increase in headloss compiled by Baumann
and Cleasby (8) suggests that it is of the correct order of
143
-------
Transfer values of stream characteristics and
process design variables from main program to the
subroutine
Compute concentration of SS and associated pollu-
tants in the effluent stream using removal
percentage = E 100
Estimate the water production and net water
production using equations 2 and 3
i
i
Calculate the filter area using equation 1
\
Calculate the capital cost using equation 8
Calculate the 0 & M costs (including the electrical
costs for backwashing using equations 8, 10, 11, 12
& 13*
* see footnote Table 1
i
Define the output parameters and return to the
main program.
C ^"Vcufc-Ht. tonal 'Fl
r'h?,r4~ for
*
i
-------
Granular Bed Filtration
(FILT)
OS 2
(Recycled
Backwash)
IS1
OS1
(Filtered Water)
IS2 (Backwash)
Figure 3
Granular Bed Filtration Process
Symbol with Stream Designations
145
-------
Table 1. Listing of the contents of DMATX for the FILT
subroutine.
DMATX (1,N)
DMATX (2,N) =
DMATX (3,N) =
DMATX (4,N)
DMATX (5,N) =
DMATX (6,N) =
DMATX (7,N) =
DMATX (8,N) =
DMATX (9,N) =
DMATX (10,N) =
DMATX (11,N) =
DMATX (12,N) =
DMATX (13,N) =
DMATX (14,N) =
DMATX (16,N) =
*DMATX (17,N) =
*DMATX (18,N) =
Fractional suspended solids removal efficiency,E,
Downtime per backwash, TB (min).
Filtration rate, QF (gpm/ft2).
Overall terminal headloss, &E (ft of water).
Overall depth of the filter bed, D (ft)
Deposit density coefficient, FK (gal/ft^/mg/l).
Fraction of the maximum filtered water
production per filter run, F.
Clean bed headloss across layer 6, per unit
depth and per unit filtration rate, K(6)
(ft2/gpm).
Clean bed headloss across layer 5, per unit
depth and per unit filtration rate, K(5)
(ft2/gpm).
Clean bed headloss across layer 4, per unit
depth and per unit filtration rate, K(4)
(ft2/gpm).
Clean bed headloss across layer 3, per unit
depth and per unit filtration rate, K(3)
(ft2/gpm).
Clean bed headloss across layer 2, per unit
depth and per unit filtration rate, K(2)
(ft2/gpm).
Clean bed headloss across layer 1, per unit
depth and per unit filtration rate, K(l)
(ft2/gpm).
Backwash rate, QB (gpm/ft^).
Excess capacity factor, ECF.
Total dynamic pumping head for backwash and
recycle, HD (ft of water)
Fractional overall pump efficiency for backwash
and recycle, EFF.
146
-------
Table 1. Continued
* Note: The inclusion of DMATX (17,N) and DMATX (18,N) in the
FILT subroutine would have exceeded the 16 row capacity of
the DMATX as provided in the main program. Changes could
have been made in the program to increase the DMATX capacity,
however, since hand calculations showed that the ECBW is
insignificant compared to the other 0 & M costs, the ECBW
calculation (Eq. (13)} was omitted from the subroutine. Any-
time a new subroutine is added to the EXEC program care
should be taken not to exceed the capacity of the common
statements such as DMATX, OMATX or SMATX. However, the
capacity of these statements can be increased by further
modifications to EXECMAIN.
147
-------
Table 2. Listing of the contents of OMATX for the FILT
subroutine.*
OMATX (1,N)
Filtered water production per filter run, WP
(gal/ft2)
OMATX (2,N)
OMATX (3fN)
Filter plan area, A (ft2)
Net filtered water production per filter
run, NWP (gal/ft2)
OMATX (4,N)
Fractional suspended solids removal
efficiency, E.
* See Footnote Table 1
148
-------
41
U
o
4J
ID
C
O
U
0)
O
2
04
D
tn
-------
magnitude. Research to derive and confirm appropriate values
of FK is underway (9).
The values of AK(I), the clean bed headloss per unit depth
and unit filtration rate, used in these examples were those
determined by DiDomenico (10). DiDomenico conducted pilot
plant experiments using a dual media bed consisting of a layer
of anthracite coal three times as deep as the underlying sand
layer. The effective size (in mm) and uniformity coefficient
of the coal and sand layers were 1.2, 1.6 and 0.5, 1.4 respec-
tively. The values of AK(I) are listed below:
Layer I = AK(I) (ft2/gpm)
1 (top) 0.030
2 0.030
3 0.028
4 0.038
5 0.140
6 0.180
The deposit distribution factor, F, was assumed to be 0.5.
A method for estimating the magnitude of F using the layer by
layer headloss distribution at run termination has been des-
cribed by Letterman (4). In general, F is equal to 1, its
maximum value, when the terminal headloss is distributed evenly
across all the equal depth layers of the bed and is equal to
its minimum value when the headloss is localized in one stratum
of the bed. FitzPatrick and Swanson (5) have observed that
in the filtration of activated sludge effluent using dual media
filters most of the deposition takes place in the top several
inches of the coal. This suggests that in this type of system
the magnitude of F is in the range 0.2 to 0.5.
It was assumed that the suspended solids removal efficiency
across the filter bed is 70 percent. This value is near the
middle of the range of efficiencies (50 to 90 percent) reported
by Kriessl (11) and by FitzPatrick and Swanson (5) for the fil-
tration of secondary effluent using dual media filters. It was
also assumed that this removal efficiency applies to the removal
of the particulate forms of BOD, phosphorous and nitrogen. It is
notable that a number of investigators (3,12) have reported
that the SS removal efficiency of granular bed filters treating
secondary effluent is essentially independent of the magnitudes
of design and operational parameters such as the filtration
rate, media size distributions and influent SS concentration.
150
-------
Typical values were assumed for the other operational
parameters needed in the filter design equations. These include
1. Terminal headless, AH = 7.2 ft of water
2
2. Backwash rate, QB = 20 gpm/ft
3. Downtime per backwash, TB = 10 min.
4. Filter bed depth, D = 2 ft.
The influent stream characteristics used in the analysis
are given below:
Flow-rate, mgd 10
Solid organic carbon, mg/1 105
Solid nonbiodegradable carbon, mg/1 30
Solid organic nitrogen, mg/1 10
Solid organic phosphorous, mg/1 2
Solid fixed matter, mg/1 30
Solid 5-day BOD, mg/1 140
Volatile suspended solids, mg/1 224
Total suspended solids, mg/1 254
Dissolved organic carbon, mg/1 43
Dissolved nonbiodegradable carbon, mg/1 11
Dissolved nitrogen, mg/1 19
Dissolved phosphorous, mg/1 4
Dissolved fixed matter, mg/1 500
Alkalinity, mg/1 250
Dissolved 5-day BOD, mg/1 60
Ammonia nitrogen as N, mg/1 15
Nitrate as N, mg/1 0
Pertinent input design parameters for processes in the
treatment system are listed with the tables of results.
Filtration Rate
The effect of the filtration rate on system cost and per-
formance was determined using filtration rates from 2 to 10
gpm/ft^. The results are listed in Table 3. Note that increas-
ing the filtrationrate in this range decreases the filter and
total system costs appreciably. According to Eq. (4) increas-
ing the filtration rate increases the backwash recycle rate.
This increases the flowrate through the primary and secondary
units. According to Table 3, the effect on system performance,
in this case, is negligible.
Mixed Liquor Suspended Solids (MLSS)
The effect of the MLSS concentration on system cost and
performance is listed in Table 4. In this case it appears that
151
-------
Table 3. Effect of the Filtration Rate on Cost and
Performance.
(GSS = 700 gpd/ft2, MLSS = 2000 mg/1, E = 0.7)
Effluent
Filtration Rate, Total Cost Filter Cost TSS BOD Total P Total N
QF (gpm/ft2) CO/1000 gal) (c/1000 gal) (mg/1) (mg/1) (mg/1) (mg/1)
2 35.7 11.1 4.4 8.7 5.4 22.1
4 31.7 7.1 4.4 8.7 5.4 22.1
6 30.1 5.5 4.5 8.7 5.4 22.1
8 29.3 4.6 4.5 8.7 5.4 22.1
10 28.7 4.0 4.5 8.7 5.4 22.1
152
-------
Table 4. Effect of MLSS on Cost and Performance
(QF = 4 gpm/ft2, GSS = 800 gpd/ft2, E = 0.5)
MLSS
(mg/1)
Total Cost
(C/1000 gal)
Filter Cost
(0/1000 gal)
Effluent
TSS
(mg/1)
BOD
(mg/D
Total P
(mg/D
Total N
(mg/1)
1000
32.63
7.09
10.3
8.9
5.4
22.3
1500
31.82
7.06
8.8
9.4
5.4
22.2
2000
31.40
7.04
7.9 9.7
5.4
22.2
3000
30.97
7.02
6.8 10.2
5.4
22.2
4000
30.95
7.00
6.1 9.7
5.4
21.9
153
-------
the MLSS concentration has only a slight effect on the filtra-
tion process cost. However, increasing the MLSS concentration
from 1000 to 4000 mg/1 decreases the effluent TSS by approxi-
mately 40 percent, from 10.3 to 6.1 mg/1. A review of the de-
sign equations in the process subroutines suggests that this
is primarily a result of increased secondary clarifier perfor-
mance.
Secondary Clarifier Overflow Rate
An interesting trade-off exists between the secondary
clarifier and the filters. As the clarifier overflow rate is
increased the cost decreases and the effluent SS concentration
increases. This increases the loading on the filters, which
decreases the filtered water production per filter run, and in-
creases the plan area requirement and cost. Table 5 shows the -
effect of increasing the overflow rate from 400 to 1200 gpd/ft
on cost and performance. Note that although the filtration
process costs increase as expected, the overall system cost
decreases with increasing overflow rate. The effluent TSS in-
creases from 3.4 to 5.8 mg/1. For some undetermined reason
the effluent BOD decreases slightly as the overflow rate is
increased from 600 to 1200 gpd/ft^. The overflow rate appears
to have little effect on the total P and total N concentrations.
154
-------
0)
•p
&
^
o
rH
(U
O
^1
i *H s-\
>4H _ rH
WM ^^
*— V ^**
O 00
N-^
/_>>
,_|
t»J *^^
M **"
S
>M^
/_N
4-) rH
CO CO
O 60
u
o
rl O
Q> O
4J H
rH ~~-
•H O
PT[ >^/
^^
jj rH
cio
Q
o o
o
rH O
CO rH
0 " w
6 o
Oi rH rH rH rH
H CM CM CM CM
CM CM CM CM CM
•* oo oo r*«
o rH r^ co oo
o • • • •
co «a~ «* in >n
-d-
CM i-H rH rH rH
CO CO CO CO CO
0 O O O O
o o o o o
•a- vo oo o CM
•—1 w-4
rn r*l
155
-------
REFERENCES
1. Lykins, B.W. and Smith, J.M., "Interim Report on the Impact
of the Public Law 92-500 on Municipal Pollution Control
Technology," Environmental Protection Agency, EPA 600/2-76-
018, Jan (1976).
2. Ives, K. J., "Optimization of Deep Bed Filtration," First
Pacific Chemical Engineering Congress, Part I, Section 2.
Separation Techniques, 99-107, Society of Chemical Engineers,
Japan and A.I.Ch.E., Oct. 10-14 (1972).
3. Huang, J.Y.C. and Baumann, E.R., "Lease Cost Sand Filter
Design for Iron Removal," J. San. Eng. Div. , ASCE, SA2, 92^
171, April (1971).
4. Letterman, R. D., "Fundamental Considerations-Filtration,"
paper presented at the Nineteenth Annual Public Water Supply
Engineers Conference, April 5-7, Champaign, Illinois (1977).
5. FitzPatrick, J. A. and Swanson, C. L., Performance Tests on
Full-Scale Tertiary Granular Media Filters, paper presented
at the US/USSR Symposium on Physical-Mechanical Methods of
Wastewater Treatment, EPA, Cincinnati, Ohio, April 5-6,1977.
6. Van Note, R. H. et al., A Guide to the Selection of Cost-
Effective Wastewater Treatment Systems, EPA-430/9-75-002,
(1975).
7. Letterman, R. D., "Optimizing Deep Bed Water Filters Using
a Deposit Distribution Concept," Filtration and Separation,
13, 4 (1976).
8. Baumann, E. R. and Cleasby, J. L., "Wastewater Filtration-
Design Considerations," EPA Technology Transfer Seminar
Publication, July (1974).
9. Ahad, M., M.S. Thesis in preparation, Pritzker Department of
Environmental Engineering, Illinois Institute of Technology,
Chicago.
10. DiDomenico, E. J., The Effect of the Media Size Distri-
butions on Pretreatment for Direct Filtration, M.S. Thesis,
Pritzker Department of Environmental Engineering, Illinois
Institute of Technology, Chicago (1975).
11. Kriessl, J. F. ,- "Granular Media Filtration of Secondary
Filtration," article in News of Environmental Research in
Cincinnati. EPA (1974).
156
-------
12. Dahab, M. F. and Young, J. C., "Unstratifled Bed Filtration
of Wastewater," J. Environmental Engineering Division,
Am. Soc. Civil Eng., February (1977).
157
-------
APPENDIX A
Program Listing for the FILT Subroutine
UUAL MtoiA FILTRATION
SUbHOuTINt FILT
JO
oO
70
00
DIMENSION AK)
COMMON SM«TX<«;0.30) .Ti«lATX(20.30) »OMATA (20.2u ) .OMATX120
. IS.i»OSl»oS<;,N.lAERF»CCOST(i!0»5) .eObTOUO .5)
> .bHK.HCT.wPI.CLANU.DLAND
Do 10 1=11.17
bWHTX I l.Osl )=bMATx< I » 1S1>
FK=L)MATX(o»N)
CU=:>MMTX< lot ibi)
D-UMA IX(5»N)
Oh=UM«TX( JrN)
DtLfH-UMATXC+.N)
AK(3)=DMAfX(ll»N)
AKC+)=DMAIX(1U»N)
AK(b)-UMAIX(9»N)
DO ^0 1=3.10
SMATX( I.Obl)=bMATX(I»
Ci.06=0.0
Du bO I=lf6
IF (DtLTH-AK(l)*D*QF)
Bi.0(i( l)=ALOfalU(DE:LTH/(AK(I)*D*OF)
CLOt.=CUOG+bLOt> ( I >
Gu TO 50
CONTINUE
WH=F*KK*0/ (E*CO*NF ) *Ci_Ofc>
WHN=lH«»0 . O/ ( WH/QF-iTu > * C *P-TB*QB )
CCObT (N.I) =6378. 1»A**0.66
BhH=A/ ( 0 . 1224+ 0 . OoObfl* A )
COMF=bMH*uHR/ ( 36bO
COMV=oMC*wPI / ( 3650
CObTO(Nfl)=COMF+COMV
SMATX ( 2 . I b2 ) =^B*Tt>/ ( WH/QF+TB ) * A*0 .
00 60 1=3.17
SMATX(I»IS2)=bMATX(I»OSl>
SMATX(2.0b2)=uB*Tb/l*H/QF+TB)*A*O.OOl4t
UO 70 1=3.10
bK.ATX ( i »0b2 ) =»IP*SMATX ( i . ibi ) *E/ ( QB*TBJ +SMATX < i »
00 80 1=11.17
SMATX(I.Ob2)=bMATX(IHS2>
OMATX(l.N)=wP
OMATX(2.N)=A
OMATX(3.N)=»»PlM
RtTURu
FIL00100
FIL00200
FIL00300
FIL00100
.20).IP(?0).FIL00500
»ACOST(2n.5)FIL00600
FIL00700
FIL00800
FIL00900
FIL01000
FIL01100
FIL01200
FIL01300
FIL01UOO
FIL01500
FIL01600
FIL01700
FIL01800
FIL01900
FIL02000
FIL02100
FIL02200
FIL02300
FIL02400
FIL02500
FIL02600
FIL02700
FIL02800
FIL02900
FIL03000
FIL03100
FIL0320U
FIL03300
FIL03UOO
FIL03500
FIL03600
FIL03700
FIL03800
FIL03900
FILOUOOO
FILOU100
FILOU200
FIL04300
FILU4400
FILOUbOO
FIL04600
FILOU700
FILOl+800
FILOU900
FIL05000
FIL05100
FIL05200
FIL05300
FIL05UOO
FIL05500
FIL05600
FIL05700
FIL05800
FILOS900
FIL06000
FIL06100
158
-------
c
c
c
c
c
c
c
APPENDIX B
Modifications to EXECMAIN
410 IF (IFAIL) 760,760,360
420 GO TO (430,440»450»460»470»480,490»500»510»520»530»540,550»560.
1,560'590,600,610,620,630,640,650,660,665), IPKOC
430 CALL PREL
GO TO 670
CALL PRSET
— —^-V4
G-*- -*.L
nt
GO TO 670
<*50 CALL ALRFS
stateme
IF THE REQUIRED MLASS, BOD5 OR MLSS CAN NOT BE ATTAINED)
IN THE AERFS SUBROUTINE, IAERF WILL BE RETURNER FROM
AEKFS WITH A VALUE OF 1 (ONE) - THIS TRANSFER CONTROL
TO STATEMENT 760 WnlCH WILL TERMINATE THE DESIGN CASE
460
500
510
520
530
540
550
560
570
580
590
600
610
620
630
640
650
660
665
IF (IAERF)
CALL MIX
GO TO 670
CALL SPLIT
GO TO 670
CALL DIG
GO TO 670
CALL VACF
GO TO 670
CALL THICK
«0 TO 670
CALL ELUT
GO TO 670
CALL SBEOS
GO TO 670
CALL TRFS
GO TO 670
CALL CHLOR
GO TO 670
CALL TFLOT
GO TO 670
CALL MHINC
Gp TO 670
CALL RWP
GO TO 670
CALL SHT
GO TO 670
CALL CENT
GO TO 670
CALL AtROb
GO TO 670
CALL POSTA
GO TO 670
CALL EQUAL
GO TO 670
CALL OIG2
GO TO 670
CALL LANDD
GO TO 670
CALL LIME
GO TO 670
CALL R8C
GO TO 670
CALL FILT
670,670,760
C
C
EXE2H600
57QEXE2U700
EXE24900
EXE25000
EXE25100
EXE25200
EXE25300
EXE2SUOO
EXE25500
EXE25600
EXE25700
EXE25800
EXE25POO
EXE26000
EXE26100
EXE26200
EXE26300
EXE26UOO
EXE26500
EXE26600
EXE26700
EXE26800
EXE26900
EXE27000
EXE27100
EXE27200
EXE27300
EXE27UOO
EXE27500
EXE27600
EXE27700
EXE27800
EXE27POO
EXE28000
EXE28100
EXE28200
EXE28300
EXE28<*00
EXE28500
EXE28600
EXE28700
EXE28800
EXE28900
EXE29000
EXE29100
EXE29200
EXE29300
EXE29i»00
EXE29500
EXE29600
EXE29700
EXE29800
EXE29900
EXE30000
EXE30100
EXE30?00
EXE30300
EXE30UOO
159
-------
APPENDIX C
Modifications to PRINT
OuTPUl FORMAT FOR PROCESS CHAKACTtRiST ICS AND PAPAMtlTtKS
.VKiTE (I0>50>
FUKMAI < ini(////(*4tx' 'PROCESS CHARACTERISTICS' »//(b8X( «ccosr =
iiiAu i.osT( LOLLARD* ./fbttX( »cobTo = OPERATIC + MAIUTENANCE cObT(
GAL. • »/(58X( • ACOST = AMORTISATION C05T( CtNTS/1000 b«L
'TcosT - TOTAL TREATMENT COST* CENIS/IOOO GAL.'>//)
UO olU 1=1(^0
IH (IH(J); bOtblO(6U
Gu TO
i!70( 300 ( 32(1
KK
PRT03000
PRTU3100
PRTu320U
PRT03300
PKT03<+00
tPKTu3e,oo
'PHTU3700
PHTU390U
PRT04000
PKTU4100
'3PRTU420G
PRTU4400
PHTU4bOU
PRT04600
KBC
PRT32700
PRT32800
PRT32900
PRT33000
bdO WKI1E (10(590) I( (DMAIX(Jd) (J=l»9) ( (OMATx(jd) ( J=ldO) PRT33100
byo FORMAT dxdHPd2(2X('ROTATING*.HX( »BOD'»«»X( »XNSTG» (5X( 'UEGC«.«*X(PRT33200
I'tfPABI' (<*X( 'QPANI' (bX( 'GSS' (5X( 'BODN»(6X( 'TSS«(5X( »CPUY'(/»bX( 'RIOPRT33300
2LOGICAL' f bX»2V-9.1(3F9.2.2F9.1(2F9.2(/(6X( 'CONTACTOR-' (/(6X( 'FINAL PRT33<*00
3SETTLLR' (bX( 'UPA8'(5X( 'IjPAN' (<*X( »APSTG»r5X( 'AREA' (<*X( 'FNSTG' (4X( 'RPRT33500
HNbTG'(tX('RATIO'(bX(«PREM»(5X('QPAT'(bX('AFb'(/(21X(2F9.2(2F9.0(2FPRT33600
b9.2(F9.3(F9.2(F9.3(F9.1(/) PRT33700
WKITE (10(600) (OMATX(J(i)(J=ll(17)(CCOST(I(l)(COSTo(Id)(ACOST(I(PRT33SOO
11)(TCOST(I»1).DMATX(IDd)(CCOST(I(2)»COSTo(I(2)» ACOST(1(2)(TCOST(IPRT33900
2(2) (DMATXdSd) PRT3UOOO
boo FORMAT (2<*X('HDSD'(bX(»uRss»(5X('NTRN*(<*X(«NSHFT'(4X('COSTM*(<*x»'CpRT3<»ioo
10STE'(4X('COSTL'(/(21X(F9.1(F9.3»2F9.1(3F9.U»//(6aX('CCOST'(<*X.'COPRT3U200
2STO' «*X('ACOST'( I(OMATX(i*d)((DMATX(J(I)(J=2d'»4X('AK(*»)'(/(6X( 'MEDIA FI
2LTER'(3X(bF9.2(F9.0(«»F9.2(//»2tX»'AK(3)»»;iX('AK(2)«.5X('AK(l)«.5X»
3'UB'>bX»'wP((7X('A'(7Xt«WPN'»6X(/»21X.itF9.2(3F9.1«//)
WKITE(IO(606> CC.OST(Id) »COSTO(I»1) (ACOST(I.l) (TCoSTdd) (DMATX (16
Id)
606 FORMAT (68X('CCOST'(«*X('COSTO'»<+X»'ACOST* •4X(«TCOST«»6X('ECF'»/»66X
1(F9.0(3F9.3(F9.2(//)
610 CONTINUE
J
PR
V
•tl
CO
C
C
C
C
OUTPUT FORMAT FOR COSTS OF MISCELLANEOUS FACILITIES
RTSHSOO
PRT3H600
PRT3H700
PRT3U800
PRT3H900
160
-------
CASE VII WORKSHOP
MODIFICATIONS OF EXISTING DESIGN SUBROUTINES
FOR PROCESS SIMULATION STUDIES
Walter J. Maier
Civil and Mineral Engineering Department
University of Minnesota
Minneapolis, Minnesota 55455
ABSTRACT
The original process subroutines of the Executive Program
were developed for use as a design tool for sizing and cost
estimating new facilities. The programs can be used for
process simulations by calculating a series of cases with
different input values for flow rate, effluent concentrations
and other major process variables. However, process simulation
calculations for a fixed size processing unit can be facilitated
by making minor changes in the subroutines. The size of the
process units are given as data input and effluent characteris-
tics are calculated as output.
161
-------
PROCESS SIMULATION
Process simulation studies are widely used in the chemical
and metallurgical industries:
a) to establish process variable effects
b) to establish pseudo optimum conditions
c) to provide a framework for process performance
analysis
Process design and process simulationare closely related.
Design calculations make use of available correlations to cal-
culate residence time, chemical addition rates, and recycle
rates to achieve a desired effluent. Formulation of operating
strategies are another example where process simulation is a
prerequisite. The objective is to define process variable
control points to achieve a desired effluent quality or perfor-
mance level; for example, in activated sludge treatment con-
trollable process variables such as recycle rate, sludge draw-
off and air supply rates must be specified as a function of
raw sewage flow rates, incoming BOD and temperature.
Computerized process control can be considered as an
ongoing process simulation in which actual performance is
compared to the process simulation model results in order to
evaluate the need for changing the set points of the controls.
It is obvious that process simulations are no better than
the mathematical models correlations, and data that go into them.
Mathematical models of fluid flow are well defined so that flow
systems, e.g. pipe networks, sewer systems are susceptible to
precise simulation. The physical separation processes such
as sedimentation can also be modeled with good success provided
the size, shape and density characteristics of the solid parti-
cles can be described. However, modelling biological processes
are still in their infancy and the available process models
are not as precise as one could wish for. Process simulation
of activated sludge treatment has been only marginally success-
ful. This is largely due to the fact that the process simula-
tions are based on over simplified mathematical models which
treat waste materials as a single constituent when it actually
consists of many different constituents and treats the active
biomass as though it were a single species of bacteria rather
than a mixture of microorganisms. More sophisticated models are
being developed that rectify some of these shortcomings. These
newer models will incorporate variable microorganism and enzyme
concentrations as well as variable waste composition and flow
rate. The point is that they will be far too complicated for
hand calculation and will have to be programmed for computer
applications.
162
-------
The Executive Digital Computer Program is a first genera-
tion process model. It was intended to be used for preliminary
process design, e.g. to calculate equipment size (detention
time) and investment-operating costs for a specified flow rate,
raw waste characteristic and effluent characteristics. In
this form it is a very useful tool for comparing alternate pro-
cessing sequences and for comparing the cost effectiveness of
alternatives. In its present form the program calculates the
complete process flow and mass balances on each of the major
constituents for any specified effluent characteristic. In
order to apply the program to existing facilities where deten-
tion time is fixed and effluent characteristics are variable,
the program needs to be modified. Size of equipment (detention
time) is specified as input and effluent concentration is trea-
ted as the dependent variable calculated using the same process
correlations. Two examples are described; the primary sedimenta-
tion subroutine is modified to allow calculating effluent con-
centration using a sedimentation tank of a fixed size and allow-
ing the flow rate and/or the raw waste water characteristics
to change. The second example illustrates a modificat-ion of
the activated sludge subroutine to allow calculating effluent
BODj. concentrations for different flow rates but using an aera-
tion basin of fixed size.
PRIMARY SEDIMENTATION SUBROUTINE (PRSET)
The existing subroutine calculates the overflow rate and
hence the tank surface area required to achieve a specified
degree of solids capture. The revised subroutine specifies
the size of the sedimentation tank as data input and calculates
the fraction of solids removed. The revised subroutine there-
fore allows calculating solids removal for different raw sewage
flow rates (variable Q) and for different raw sewage suspended
solids concentrations. The subroutine for primary sedimenta-
tion relates solids removal to overflow rate using a modified
form of the correlation from "ASCE Manual of Practice, #36,
1959".
FRPS = 0.82 e-(GPS/2780)
where FRPS = fraction of incoming suspended solids removed
in the settler
2
GPS = overflow rate gal/day-ft
The overflow rate (GPS) determines the required surface area
of tank for any given flow rate. The degree of thickening of
the underflow is specified as input; URPS is the ratio of
suspended solids in the incoming sewage. All suspended
solids (organic carbon, nitrogen and phosphorus) are assumed
to follow the same distribution. Input is required for FRPS,
URPS, HPWK and ECF (excess capacity factor).
163
-------
The revised program deletes the input value for FRPS and
substitutes data input for the tank surface area (APS); the
process variable correlations are rewritten in order to calcu-
late the value of GPS from the given tank surface area and the
design flow rate (A). This allows calculating FRPS and the con-
centration of solids in the overflow and sludge stream as in
the original program. The pertinent Fortran program statements
are listed in Table 1.
Table 2 lists the program changes and the required Input/
Output changes. The proposed change in the second data card
which specifies a value for KEEY allows using either the origin-
al design program or the modified program. If the original
program is used (KEEY=0), a value for FRPS must be specified as
input; if the modified program is used (KEEY=1), APS must be
specified as input on the first data card of PRSET.
Use of the modified PRSET program is illustrated below
using the common treatment scheme outlined in Table 3. The
parameter variations and calculated effluent suspended solids
concentrations are listed in Table 4 and shown graphically in
Figures 1-2.
Using a fixed value of APS 93,7000 ft ) and influent sus-
pended solids concentration of 260 mg/1, the supernatant sus-
pended solids concentration is shown to increase with flow rate
(Figure 1). The advantage in using the computer program for
this type of analysis is that it automatically material balances
the whole plant, that is, it includes the effects of recycle of
supernatant from downstream process units back to the primary
clarifier.
ACTIVATED SLUDGE-FINAL SEDIMENTATION SUBROUTINE (AERFS)
The combined activated sludge-final sedimentation process
subroutine is designed to calculate the aeration tank volume,
reciruclation rate, surface area of the final sedimentation
basin, and the air requirement for specified input values of
raw waste flows, effluent BOD, mixed liquor suspended solids
concentration, biochemical rate coefficient, temperature, oxygen
transfer efficiency, minimum dissolved oxygen concentration in
the aerator, and the overflow rate and thickening capacity of
the final clarifier. The revised program specifies the volume
of the aeration tank (VAER) and allows calculating either
effluent BOD for a specified value of mixed liquor suspended
solids (MLSS) or it calculates the required MLSS to achieve a
desired effluent BOD.
Process performance is described by a simple first order
growth rate equation which relates the six major process vari-
ables of the aerator,
164
-------
TABLE 1
Modified Subroutine PRSET
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
PK1NAKY
PROCESS
SUUKGUTINE PRbET
SEDIMENTATION
IDENTIFICATION
PRS00100
PRS00200
PRS00300
PRS00100
PRSOOSOO
PRS00600
PRS00700
PRSOOflOO
PRS00900
OSltOS2 PRSOlOOO
COMNON sN'ArX<20»30) , TMATX(20»30) »DMATX ( 20 , 20 > »OyATX(20»20)»IP<2n) .PRS01100
llNf>,IO,ISI.IS2,OSlr052,N»IAERF,CCOST(20»b),COSTO(20»5),ACOST(2D.5)PPSQ1?00
INITIAL STATEMENTS
2.TtOST(20,b),DHK»PCT,l*PI,CLAND,DLAND»FLOW<2'3>,POW<25),TKWHD<25)
COMMON/KEEY/KEEY
ASSIGNMENT OF DESIGN VALUES TO PROCESS PARAMETERS
HPWK=DMATX(3.N)
iF(KttY.EQ.O) GO TO <*0
APS=DMATX(i*,N)
GPS=SMATX(2fISI)*1000./APS
FRP<,=EXP( (-GPS-551. 71/2760. )
OMATX(1,N)=FRPS
PROCESS RELATIONSHIPS
CHARACTERISTICS
KEOD. TO CALC. EFFLUENT STREAM
SMATX(2fOS2)=DMATX(l,N)*SMATX(2,ISl)/nMATX(2»N)
SMATX(2.0S1)=SMATX(2,ISD-SMATX(2»OS2)
TEMP1=(1.-UMATX(1.N))*SMATX(2.ISI)/SMATX(2.OS1)
TEMP2=OMATX(l.N)*SMATX<2rISl)/SN'ATX<2,OS2)
EFFLUENT STKEAM CALCULATIONS
DO 10 1=3.10
SMATX(I,OS1)=TEMP1*SMATX(I.IS1)
10 SMATX(I,OS2)=TEMP2*bMATX(I.ISI)
DO 20 1=11,20
SMATX(I,OS2)=SMATX(I,1S1)
20 SMATX(I,OS1)=SMATXU,OS2>
CALC. OF OUTPUT SIZES AND QUANTITIES
PGPM=SMATX ( 2 • OS2 ) *1 16666 . 7/HPWK*DMATX
IF(KEEY.EQ.l) GO TO 50
GPS=-2780.*ALOG«OMATX(1,N))-551.7
APS=SMATX (2,IS1)*1000. /GPS
50 APS=APS*DMATX(16.N)
CALC. OF CAPITAL COSTS FOR PRIMARY SETTLER BASIN BASED
PRS01300
*
ppsomoo
PPS01SOO
PRS01600
PRS01700
PRS01800
*
*
*
*
*
PRS01900
PRS02000
PPSO?100
PRSO?200
PRS02300
**
PRSOZ'SOO
PPSOP600
PRS02700
PRS02800
PPS02900
PRS03000
PRS03100
PRS03200
PRS03300
PRS03UOO
PRS03500
PPS03600
PRS03700
PRS03SOO
PRS03900
PRSOUOOO
PRS04100
PRSOU200
*
PRSOU300
**
**
PRSOU500
PRSOU600
PRSO<»700
Note: The subroutine was modified by adding 7 statements; the
additional statements are identified by asterisks. Changes in
Input/Output are listed in Table 2.
165
-------
TABLE 2
Modification of PRSET Subroutine Program
Computer program changes
(a) Added statements (*):
Statements between PRS01800 and PRS01900
Statements between PRS04200 and PRS04300
(b) Replaced or modified statements (**)
Statements between PRS02300 and PRS02500
Statements between PRS04300 and PRS04500
Input/output change
(a) Second data card: Between column 1-2, add KEEY=I2
KEEY=0: Original program
KEEY=1: Modified program
(b) Add DMATX(4,N) as required input value to the first
data card of PRSET.
(c) Add common statement at beginning of the PRSET sub-
routine
COMMON/KEEY/KEEY
(d) Add common statement at beginning of executive program:
COMMON/KEEY/KEEY
166
-------
Table 3
Common Treatment Flow Diagram
12
1J
2
<
i
3
<
5
7
\
.1
10
13
K
0
1
0
0
1
0
0
1
0
N
1
0
2
3
0
4
5
0
6
IPROC
1
4
2
3
4
8
7
4
12
PROCESS
PREL
MIX
PRSET
AERFS
MIX
THICK
VACF
MIX
CHLOR
IS1
1
2
3
4
5
8
9
10
6
IS2
0
12
0
0
7
0
0
11
0
OS1
2
3
4
6
8
9
14
12
13
OS2
0
0
5
1
0
10
11
0
0
167
-------
Table 4
PRSET - Parameter Variations and Output Results
flow rate
Q (mgd)
0.5
1.0
2.5
5.0
7.5
10.0
15.0
5
5
5
5
5
5
5
5
10
influent
SS (mg/H)
260
260
260
260
260
260
260
26
78
130
260
390
520
780
1300
1300
effluent
SS (mg/£)
60
70
97
135
164
186
217
17
44
70
135
200
266
397
658
904
168
-------
240
in
V)
c
1)
3
w
200
160
120
Influent bS = 260 mg/1
id
•H
H
80
40
8 10
Flow Rate (mgd)
12
14
Fig. 1
Primary affluent SS for Varying Flow Rate
169
-------
1000
cr<
a
W
M
-p
c
0)
3
t-l
w
800
600
400
200
5 mgd
200 400 60C 800 1000
Primary Influent SS (mg/1)
1200
1400
Fig. 2
Primary Effluent SS for Varying Primary Influent SS
170
-------
AtBODin-BODout) = CBODout) CCAJ3R1 CMLSSl CVAERl
where; Q = flow rate
BOD. = inlet BOD,
in 5
BOD = outlet BOD5 which is equivalent to the BOD5 in the
0 aerator for a well mixed reactor
CAER = rate coefficient corrected for temperature
MLSS = mixed liquor suspended solids concentration
VAER = aeration tank volume.
By specifying any combination of 5 variables, the equation can
be solved for the 6th unspecified variable.
The existing program specifies the first five variables
as input and calculates VAER. The modified program uses the
same equation; it has two options that allow calculating BOD ,
or MLSS concentration. The pertinent Fortran program sections
of the AERFS subroutine are listed in Table 5; the program
changes are listed in Table 6 along with the changes in Input/
Output. Program changes are keyed in the second data card by
specifying a value for KEY. The original design program is
used for KEY =0. By setting KEY = 1, the program reads in a
value for VAER and calculates the effluent BOD concentration.
For KEY = 2, the program reads in VAER and BOD and calculates
the required MLSS concentration in the aerator?
Use of the modified AERFS program is illustrated using the
common treatment scheme outlined in Table 3, The parameter
variations and calculated values are listed in Table 7 and
illustrated graphically in Figures 3-5.
ASSIGNED PROBLEM
Participants may choose to use the prepared program modi-
fications to carry out a short process variable study or to
modify one of the other subroutines as an exercise.
a) Process variable studies
Using the common data input from previous problems
modify the appropriate data input cards and use the
revised program to calculate effluent characteristics
for a series of flow rates ranging from 25 percent to
300 percent of the base case.
b) Modification of other subroutines, e.g., thickener or
trickling filters-final sedimentation.
171
-------
Table 5
c
t
c,
c
Modified Subroutine AERFS
ACTIVATED SLUDGE - FINAL SETTLER
PROCESS IDENTIFICATION NUMBER 3
C
C
c
c
c
SUBROUTINE AERFS
INITIAL STATEMENTS
INTEGER OS1.0S2
COMMON SMATX(20r30) » TMATX ( 20 > 30 ) »DMATX (20»20> ' OI/ATX ( 20 r 20 ) r IP(?(1
HNH»IO,ISl,lS2»OSlrOS2fNr lAERFrCCOST«20»5).COSTO(?0»5) »ACOST(20r
2»TCOST(20»b) »DHR.PCT«WPI»CLANDrDLAND»FLOW(25) »POW(25) . TKWHDC25)
COIv.MON/KEY/KEY
PROCESS RELATIONSHIPS REQD. TO CALC. EFFLUENT STREAM
CHARACTERISTICS
HEAD=LlMATX(9»N)
t)OD2=SMATX(8»ISl)-«-SMATX(17» I SI)
DB002=SMATX ( 17 , IS1 >
CEDR=.18*1.0t7**(DMATX(3»N)-2e. )
IF(KEY.EQ.O) GO TO 1000
VAEH=UMATX(11»U)
TAzVAER/SMATX(2»ISl>
IF(KEY.EQ.l) GO TO 1000
SA=(bOU2-DMATX(l»N) ) / (DMATX( 1 ,N) *CAER*TA*2<*. )
XMLSS=SA*1000.
DMATX(2»N)=XMLSS
GO TO 3000
1000 SA=DMATX(2,N)/1000.
IFfKEY.EQ.O) GO TO 2000
BOO=BOD2/ ( 1 . +TA*CAEK*SA*24 . )
DMATXdrNlrBOD
GO TO 3000
2000 TA=(BOD2-DMATX(l»N) ) / (DMATX ( 1 , N) *CAER*SA*21 . )
VA£R=SMATX(2»IS1)*TA
3000 XRsS=bb6.1*DMATX(BrN)**.'*942/DMATX(2»N)**1.8165/AEF01ino
5) AFF01200
AEF01300
AEFOUOO
AEFOlbOO
AFF01600
AFF01700
AEFOIBOO
AEFOlPOO
AFF02QOO
AEF02100
AEF02200
AEFo?300
*
*
*
*
*
*
**
*
*
AEF02hOO
AFF02*800
AEF02POO
AEF03000
AEF03100
AEF03?00
AEF03300
AFF03UOO
AEFQ3500
AEF03600
AEFQ3700
AEF03800
AFF03900
AEFOUOOO
AEFO<*100
AEFOU200
Note: The program was modified by adding 13 statements; the ad-
ditions are identified by an asterisk. The changes in input/
output are described in Table 6.
172
-------
Table 6
Modification of AERFS Subroutine Program
Program changes
1. To find BOD from given VAER
VAER=DMATX(11,N)
TA=VAER/SMATX ( 2 , ISl )
II_III r
4 statements
1000 SA=DMATX (2,N)/1000.
IF(KEY.EQ.O) GO TO 2000
BOD=BOD2/(1.+TA*CAER*SA*24.
DMATX (1,N)=BOD
GO TO 3000
2. To find MLSS from given VAER and BOD
VAER=DMATX(11,N)
TA=VAER/SMATX(2,ISl)
IF(KEY.EQ.l) GO TO 1000
SA=(BOD2-DMATX(1,N))/(DMATX(1,N)*CAER*TA*24.
XMLSS=SA*1000.
DMATX(2,N)=XMLSS
GO TO 3000
3. Input/Output Change
a) 2nd data card: between column 3-4, add KEY=I2
K=0: Original program
K=l: Find BOD from given VAER
K=2: Find MLSS from given VAER and BOD
b) Input VAER=DMATX(11,N) on the second data card of
AERFS. Calculations for BOD and MLSS are not affected
by the XMLSS=DMATX(2,N) as required in the original
program.
c) Add common statement at beginning of AERFS subroutine
COMMON/KEY/KEY
d) Add common statement at beginning of executive program
COMMON/KEY/KEY
173
-------
Table 7
AERFS-Parameter Variations and Output Results
Flow Rate MLSS
Q(mg/A) (mg/£)
0.5 2000
1.0 2000
1.5 2000
2.0 2000
3.0 2000
4.0 2000
5.0 2000
6.0 2000
7.5 2000
10.0 2000
15.0 2000
BOD
(mg/£)
' 2.3 S
4.5
6.6
8.7
12.7
16.5
20.1
23.6
28.3
35.6
47.8
X j*
5.0 7357 6
5.0 4306 10
5.0 2015 20
5.0 849 40
5.0 455 60
v /
5.0 400
5.0 600
5.0 1000
5.0 2000
5.0 4000
5.0 6000
f ^^
64.5
50.8
35.5
20.1
10.7
7.3
J
Calculated values are enclosed by brackets.
174
-------
§
+J
c
0)
W
50
4°
30
tJ
O
O
(U
CO
20
10
6 8
Flow Rate (mgd)
10
12
14
Fig* 3
Secondary Effluent BODg for Varying Flow Rate (mg/1)
175
-------
8000
w 6000
4000
2000
Q = 5 mgd
V = 1 mg
10
20 30 40 50
Secondary Effluent BOD, (mg/1)
60
Fig. 4
MLSS For Varying Secondary Effluent SS (mg/1)
176
-------
Q
O
80
-------
CASE VIII WORKSHOP
USE OF A STREAM IMPACT PROGRAM IN
CONJUNCTION WITH THE EXEC PROGRAM
James W. Male
Department of Civil Engineering
University of Massachusetts
Amherst, Massachusetts 01003
ABSTRACT
A simple subroutine is described which will determine the
dissolved oxygen concentration in a stream receiving waste from
a treatment plant analyzed by the Exec program. The subroutine
can be attached to, or incorporated into the Exec program to
determine the downstream effect of wastewater treatment plant
configurations. An example is given and suggestions are made for
extensions and refinements of the' subroutine.
178
-------
INTRODUCTION
Planners and designers often require knowledge of the
effects of a waste treatment plant downstream of the waste dis-
charge, as well as the effluent quality itself. The Executive
Program for Preliminary Design of Wastewater Treatment Systems
provides characteristics of the effluent, but currently it is
not capable of predicting downstream water quality as a function
of treatment process parameters. However, several programs exist
which can predict downstream water quality given the characteris-
tics of the waste discharge. (Norton et al., 1974; Hydrocomp
International, Incorporated, 1976).
This paper will briefly describe the basic theory behind
receiving stream models in general, and also details on how to
augment the Exec program with a simple receiving stream model.
The resulting combination will not only allow the prediction of
effluent water quality but also resulting downstream characteris-
tics. This will permit comparison of the predicted pollutant
concentration with surface water quality standards.
The receiving stream model requires input from the last
treatment process in the system and external input describing the
water quality upstream, and stream characteristics downstream of
the treatment plant effluent.
The stream subroutine will model the dissolved oxygen con-
centration in the receiving stream. Dissolved oxygen is only one
parameter of water quality, however, it is the most widely used
indicator because:
(1) the mathematical relationship between DO and BOD is
well understood.
(2) the effect on DO of other oxygen demanding matter can
be predicted with reasonable accuracy, and
(3) DO itself affects the quality of a stream.
It should be noted that the DO concentration is not recorded
in the stream matrix (SMATX) within the Exec program. In addition,
not all processes use dissolved oxygen as a design parameter.
Therefore, care must be taken in selecting process configurations
that allow a reasonable estimate of the plant effluent dissolved
oxygen concentration. If this cannot be accomplished, a reason-
able value can be assumed and input as a subroutine modification.
The process symbol shown in Figure 1 will be used to repre^-
sent the receiving stream subroutine.
179
-------
IS2
(Upstream flow)
IS1
(Treatment Plant
Effluent)
OS1
(Downstream)
Figure 1. Process symbol for subroutine STREAM
180
-------
RECEIVING STREAM MODEL
The subroutine developed for this case study is a simpli-
fied dissolved oxygen sag curve. The basic calculations are
similar to many existing programs for calculating the DO concen-
tration of a stream.
Basic Theory
The determination of the DO concentration is based on the
assumption of a first order decay rate for BOD and a rate of
change of dissolved oxygen proportional to the DO deficit. The
resulting equation, originally developed by Streeter and Phelps
(1925), is shown below:
k L -k, t -k9t -k-t
Dt • 4°kl (e - e ' + Dod (1)
where:
D = the dissolved oxygen deficit at time t, mg/1,
k, = deoxygenation coefficient, days ,
k_ = reaeration coefficient, days ,
L = initial BOD, mg/1, and
D = initial dissolved oxygen deficit at time 0, mg/1.
The reaeration coefficient, k~, is calculated knowing certain
characteristics of the receiving stream (0 "Conner and Dobbins,
1956) :
k2 = -
where:
D = molecular diffusivity of oxygen in water at 20 C,
III »•)
ft: /day,
V = stream velocity, ft/day, and
H = stream depth, ft.
The saturation concentration of oxygen is calculated using:
C = 14.62 = 0.3898T + 0.00696T2 = 0.00005897T3 (3)
s
181
-------
where:
T = temperature.
Using equation (1) the DO deficit, and therefore the DO concen-
tration can be predicted as a function of distance in the down-
stream direction. A curve representing equation (1) is shown in
Figure 2.
The minimum DO concentration occurs at time t . This
critical time can be calculated by:
k( ]r ^ "\f ^ r^
f\ \ J\. ^\ JV -i I LJ
£•11 £ J. O \ I t A \
k7 (1 k^ir >' (4)
- k-
^2 Kl
By substituting the value of t into equation (1), the maximum
deficit (minimum concentration^ can be determined. Likewise,
knowing the travel time for a certain stream reach, we can also
calculate the DO concentration at the end of the reach. Often
the end of the stream reach comes before the critical time, mak-
ing the concentration at the end of the reach the minimum concen-
tration.
Equation (1) is valid for a stream reach with essentially
constant characteristics. If the stream values change with
distance downstream, then another calculation must be made.
Typical changes include changes in the river cross-sectional
area, changes in the value of the deoxygenation coefficient, k,,
and the inflow of a tributary or sewage treatment plant, which
will change both the flow and water quality.
To handle such changes a new reach is defined and equation
(1) is used to calculate new values for the second reach.
Initial conditions for the second reach now correspond to a
mixture of the end conditions for reach one and the tributary
inflow (if a tributary exists). To determine the initial BOD,
L , for the second reach, it is also necessary to calculate the
BOD at the end of reach one. This is done using a simple first
order decay equation:
-k t
Lt = Lo e (5)
where:
L. = BOD at time t.
A mass balance is calculated to determine the initial con-
ditions for reach two given characteristics of both the upstream
reach and influent tributary. Initial conditions include tempera-
ture, DO, BOD, and flow.
182
-------
c
o
•H
•u
•U
c
<1)
u
c
o
u
o
Q
Saturation Concentration
DO Deficit
Time (or distance) downstream
Figure 2. Typical dissolved oxygen concentration profile
183
-------
Similar calculations can be made as we move downstream
for any number of subsequent reaches. Figure 3 shows a typical
curve for five reaches.
Subroutine STREAM
The previous theory has been incorporated into a simple
computer program to illustrate one way of augmenting the Exec
program with a stream impact model. Other possibilities and
refinements will be discussed in a later section.
The Fortran listing for subroutine STREAM is shown in
Appendix A. The theory equations and Fortran statement numbers
correspond in the following way:
Equation Number Fortran Statement
(1) STR08700,08800,09200,09300
(2) STR07200
(3) STR07800
(4) STR08600
(5) STR09400
The mass balance is calculated in statements STR06100-06400.
Each iteration of the problem corresponds to calculations for
one reach. The program is set up to read one input card (in
315 format) to determine:
(1) the number of reaches to be analyzed, NREACH,[no units]
(2) fho flow stream number corresponding to the effluent
of the sewage treatment plant, ISTREM, [no units] and
(3) the last process in the treatment plant listing the
effluent DO concentration, JPROC [no units]
The next input card (in 4F10.2 format) lists the river
characteristics just upstream of the treatment plant discharge.
They include:
(1) dissolved oxygen, DOUP, [mg/1]
(2) BOD, BODUP, [mg/1]
(3) flow, QUP, [ft3/sec]
(4) temperature, TEMPUP [°C]
The remaining data cards correspond to each reach of the
receiving stream. The subroutine is set up so that each reach
has a tributary at its head. If, in fact, no tributary exists,
a tributary flow of zero is input. Each input card (in 8F10.2
format) will have:
184
-------
c
o
•H
-U
m
o
G
O
U
O
Q
Saturation Concentration
Time (or distance) downstream
Figure 3. Dissolved oxygen concentration profile for a
five reach stream
185
-------
(1) the length of the reach, RLNGTH, [miles]
(2) the depth of the reach, DEPTH, [feet]
(3) the deoxygenation coefficient at 20°C, XKl, [days ]
2
(4) the cross-sectional area of the reach, XAREA, [ft ]
(5) the tributary flow, QTRIB, [ft3/sec]
(6) the tributary dissolved oxygen concentration, DOTRIB,
[mg/1]
(7) the tributary BOD, BODTRB, [rng/1] and
(8) the tributary temperature, TEMTRB [°C].
The data card for the first reach does not require informa-
tion on the tributary characteristics since EXECMAIN will pro-
vice effluent characteristics. Therefore, the last four entries
on this card may be left blank.
Output from the subroutine includes the input data, the BOD,
DO, flow, and temperature at the end of each reach, and the mini-
mum DO in the reach. An example of the input and output for
the subroutine is shown in Appendix B.
Example
The subroutine was appended to the Exec program to analyze
the downstream effect of the treatment plant configuration shown
in Figure 4. To use the subroutine in conjunction with the Exec
program, one Fortran statement must be added to EXECMAIN. The
statement:
CALL STREAM
must be added immediately after the CALL PRINT statement at the
end of EXECMAIN.
To illustrate the effect of process variations, the effluent
DO design parameters DMATX(4,6) for the post aeration process
were varied to determine the effect on the stream. The results
are shown in Table 1.
EXTENSIONS AND REFINEMENTS
Several refinements and extensions can, and should, be made
before the subroutine can be applied effectively to a real world
problem.
186
-------
-------
Table 1
Instream Minimum DO
Concentrations Resulting from
Design Parameter Variations
Run
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Plant
Flow
(MGD)
10
10
10
10
10
50
50
50
50
50
100
100
100
100
100
Post Ae
Effluer
Cmg/1)
2.0
3.0
4.0
5.0
6.0
2.0
3.0
4.0
5.0
6.0
2.0
2.0
2.0
2.0
2.0
Activated Sludge
Effluent BOD
(mg/1)
13.0
13.0
13,0
13.0
13.0
13.0
13.0
13.0
13.0
13.0
10.0
15.0
20.0
25.0
30.0
Instream
Minimum DO
(mg/D
6.59
6.61
6.63
6.63
6.64
6.06
6.11
6.17
6.23
6.28
5.74
5.35
4.96
4.55
4.16
Note; Upstream characteristics (1) Q = 500 cfs, (2) DO = 9.0 mg/1,
C3I BOD =3.0 mg/1, (4) Temp = 15°C
188
-------
Improved DO Equation
Equation (1) includes only two effects on the DO concen-
tration. These are the BOD and the reaeration capability of the
stream. Several other factors also effect the DO concentration
and can be incorporated into the DO sag equation. These include:
(1) oxygen demand by benthic organisms,
(2) continuous input from overland flow or groundwater,
(3) reaeration due to photosynthesis, and
(4) scour of bottom deposits.
These effects are further discussed by Nemerow (JL9741 a^nd
Thomann (1972).
Other Parameters
As mentioned earlier, dissolved oxygen is only one measure
of stream quality. The concentration of other pollutants can
be as important in determining the quality of a stream and
compliance with water quality standards. The stream subroutine
could be expanded to calculate the concentrations of both conser-
vative and non-conservative pollutants at various points down-
stream (Thomann, 1972).
Internalizing Subroutine STREAM
The subroutine described in this case study was developed to
be independent of the Exec program and involve minimal modifica-
tions to the Exec program. It was also developed in this manner
to provide a contrast to the approach taken in Case VI by
Letterman. In his presentation, an additional subroutine was
added to the Exec program as another process in the treatment
plant. Input was accomplished using the existing DMATX and out-
put was included in the PRINT subroutine. In this manner the
subroutine was included in the iterations of the Exec program
and sizing of the process was accomplished to meet prespecified
design criteria.
The STREAM subroutine could also be included in such a way.
This approach would allow the iterative process to use the DO
concentration in the stream as a determining factor in the ulti-
mate size of the treatment process. In conjunction with this
approach, it is possible to include instream aeration in the
STREAM subroutine. Work is in progress (Tabatabaie, 1977) to
include a stream/instream aeration subroutine as a "process"
in the wastewater treatment plant. Obviously such an analysis
will depend heavily on streamflow characteristics, especially
during low flow months. It is interesting to note, however, that
189
-------
during such periods even advanced waste treatment may not provide
sufficient removal to maintain the surface water quality stan-
dards (Whipple et al., 1970).
ASSIGNED PROBLEM
As an assigned problem the short course participants may
choose to pursue one of the following three exercises:
(1) Using the same process described in the write-up,
evaluate the effect on a stream with the following
characteristics:
Treatment Plant
Discharge
length = 2.7 mi
depth = 7.9 ft
area = 1700 ft2
k = 0.25 day"1
length = 4.1 mi
depth = 9.2 ft 2
area = 2,000 ft
k = 0.25 day"1
DO = 8.3 mg/1
BOD =2.5 mg/1
Flow = 300 cfs
Temperature = 18°C
DO = 7.2 mg/1
BOD =5.1 mg/1
Flow = 100 cfs
Temperature = 19 C
(2) Using the same receiving stream, evaluate the effect
of a different process configuration, or
(3) Add statements to the subroutine to determine the
concentration of dissolved fixed matter (DFM) at
the end of each stream reach.
190
-------
REFERENCES
Hydrocomp International, Inc. 1976. Hydrocomp Simulation
Programming Operations Manual, fourth ed., Palo Alto,
California.
Nemerow, N. L. 1974. Scientific Stream Pollution Analysis/
McGraw-Hill Book Co., New York.
Norton, W. R., L. A. Roesner, D. E. Evenson and J. R. Monser.
1974. "Computer Program Documentation for the Stream
Quality Model QUAL-II," Prepared for the U.S.E.P.A. Systems
Development Branch, Washington, D. C.
O'Conner, D. and W. Dobbins. 1956. "The Mechanism of Reaeration
in Natural Streams," Journal of the Sanitary Engineering
Division ASCE, SA6, 1115-1-1115-30.
Streeter, H. and E. Phelps. 1925. "A Study of the Purification
of the Ohio River," U. S. Public Health Service Bull. No.
146, Washington, D. C.
Tabatabaie, M. 1977. Unpublished rough draft of M.S. thesis,
Dept. of Environmental Engineering, Illinois Institute of
Technology.
Thomann, R. V. 1972. Systems Analysis and Water Quality Manage-
ment, McGraw-Hill Book Co., New York.
Velz, C. J. 1970. Applied Stream Sanitation, John Wiley & Sons,
Inc., New York.
Whipple, W., F. P. Coughlan and S. L. Yu. 1970. "Instream
Aerators for Polluted Rivers," Journal of the Sanitary
Engineering Division ASCE, 96(SA5):1153-1165.
191
-------
APPENDIX A
Subroutine STREAM
c STROOIOO
C RECEIVING STREAM STR00200
C STR00300
SUBROUTINE STREAM STROOUOO
C STR00500
C COMMON INITIAL STATEMENTS STROObOO
C STR00700
COMMON SMATX<20.30) .TMATX (20 '30 ) .UMATX (20.20 > »OMATX (20 .20) . IP (20) .STR00800
llNP.IU»ISl.IS2»OSl.nS2»N»IAERF»CCOST(20.5) .COSTO(20»5) . ACOST(20 . 5)STR00900
2»TCOST(20.5) »DHR»PCT»WPI »CLAND»DLAND»PROCNO( 10 ) .FLOW<25) »POW(25) .TSTR01000
3KWHD(2b> STROllOO
WRITE (10.10) STR01200
1U FORMAT UH1.48X. ' IN^TREAM DISSOLVED OXYGEN ') STR01300
WRITE (10.20) STROluOO
2U FORMAT C/////i+X»'REACH' rlOXr 'PHYSICAL CHARACTERISTICS' » 16X »' VARISTR01500
1ABLE CHARACTERlSTlCS'f 15Xr 'TRIBUTARY CHARACTERISTICS') STR01600
WRITE (I0e30> STR01700
30 FORMAT DOUP»RODUP»QUP»TEMPUP STR03800
80 FORMAT (HF10.2) STR03900
C STR04000
C READ DOWNSTREAM CHARACTERISTICS FOR NREACH REACHES (LENGHTSTROUlOO
C DEPTH. DEOXYfiENATION COEF. .CROSS SECTIONAL AREA.TRIBUTARY STRO<*200
C FLOW .DO .BOD AND TEMPERATURE STR01300
C STR04400
00 160 1=1. NREACH STROH500
READ (INP.90) RLNGTH.UEPTH.XK1.XAREA.QTRIB.DOTRIB.BODTRB.TEMTRR STRO<*600
9U FORMAT 18F10.2) STROU700
C STRO<*800
C CONVERT FLOw TO MGD STR04900
C STR05000
QUP=OUP/1. 54723 STR05100
«TRIB=QTRIB/1. 54723 STR05200
IF (I.GT.l) GO TO InO STR05300
QTR1B=SMATX(2.ISTKEM) STR05UOO
DOTRIb=UMATX«+. JPROC) STR05500
BODTRb=SMATX(a.ISTRFM)+SMATX(17.ISTREM) STR05600
TEMTRb=DMATX(b.JPROc) STR05700
C STR05800
C MASS BALANCF AT BEGINNING OF KEACH STRC5900
C STR06000
10U Q=OUP+QTRIB STR06100
192
-------
APPENDIX A (Cont.)
00=(OuUP*UUP+bOTRlH«QTRI8>/G STR06200
BOD=(bODUP*GUP+ROOTpB*OTRlB)/Q STR06300
TLMP=(TEMPUP*GUP+TEMTRB*OTRIB)/0 STR06UOO
C STR06500
C CALCULATE WpLOClTY IN FT/DAY STR06600
C STR06700
Vt.LOC=< <0»1.5<+723)/xAREA)*86(*00. STR06ftOO
C STR06900
C CALCULATE RpAERATlON COEF. STR07000
C STR07100
XK2=(10.0019U»VELOC)**0.5)/(DEPTH**1.5) STR07200
C STR07300
c TEMPERATUKE ADJUSTMENTS sTR07*oo
C STR07500
XK1 = XM*(1.0H7**(T£MP-20.) ) STR07600
XK2=XK2*<1.0241**(TFMP-20.)) STR07700
SATDO=(l<*.62-lU.389p»TEMP)-M0.006y6*TEMP**2)-(0.00005897»TEMP**3))STR07800
DEFDO=SATOO-DO STR07900
C STR08000
C CALCULATE MINIMUM DO STR08100
C STR08200
TTlME=RLNGTH*S280./vELOC STR08300
TC=TTiME+l. STR08HOO
IF «XK.2*DEFDO-XK1*ROD) .GT.O. ) GO TO 110 STR08500
TC=(1./(XK2-XK1) )*AI.OG( (XK2/XM>*(1.-(XK2-XK1)*DEFDO/00
AMXDEF=IXK.1*60U)/(XK2-XK1)*(EXP(-XK1*TC)-EXP(-XK2*TC))+DEFDO*EXP(-STR08700
1XK2*TC) STR08800
C STR08900
c CALCULATE oo AND boo AT END OF REACH STRogooo
C STR09100
110 ENODEF = (XK1*80D)/(XK2-XK1)*(EXP(-XK1*TTIME)-EXP(-XK2»TTIMF) )-«-DeFDOSTR09200
1ȣXP(-AK2*TTIME) STR09300
tNDBOU=BOO*EXP(-XKl»TTIME) STR09i*00
IF (TC.LT.TT1ME) GO TO 130 STR09500
IF (Dc.FL)O.LT.t.NDDEF) GO To 120 STR09600
DOM1N=UO STR09700
GO TO 1UO STR09800
120 OOMIN=SATUO-ENODEF STRoggoo
60 TO 1UO STR10000
13U DOMIN=SATDO-AMXDEF STR10100
140 IF (DOMIN.LT.O.) DOwlN=0. STR10200
C STR10300
C CONVERT FLO- TO CFS STR10400
C STR10500
QTRIB=uTRIB*l.b^723 STR10600
0=0*1.5U723 STR10700
C STR10800
C OUTPUT STR10900
C STR11000
ENDDO=SATDO-ENDDEF STRlllOO
IF (ENODO.LT.O.) ENnDO=0. STR11200
WRITE (10.150) I.RLNGTH.DEPTH.XAREA.XKlrDOMIN.ENDDO.ENDHOD.Q^FMPiSTRl1300
IDOTRIB.BODTRB.OTRIB.TEMTRb STRlluOO
150 FORMAT (/1X.13.2X.Fq.2.tX.F6.2.IX»F10.2.3X.FH.2.10X.F5.2.2X.Ffa.2.FSTR11SOO
17.2»lX,F8.2»l+X.F5.2.5x.F6.2.3X.F6.2»F6.2.2X.Ffa.2) STRllhOO
C STR11700
C INTIALI2E FoR NEXT REACH STR11800
C STR11900
DOUP=LNDDO STR12000
BODUP-ENDBOD STR12100
QUP=0 STR12200
TEMPUP=TEMP STR12300
160 CONTINUE STR12UOO
RETURI^i STR12500
END STR12600
193
-------
-p
o
o
in
rH
r-
•
CN
VO
13
S-l
as
O
for EXECMA
d)
O
C
(1)
3
O1
<1>
CO
(0
4J
(0
•0
rH
e
^
o
c
0)
J5
-p
^
o
rH
rH
0
-------
z
Ul
^
X
o
^
UJ
^
^J
o
si
»*4
Q
z
UJ
*—
Ul
z
\f\
u
p
o:
u
u
<
X
X
u
^
K
O
to
CC
4
4
4
4
I/I 4
O 4
>-> 4
1- 4
l/> 4
*-4 4
Ct 4
Ul 4
t- 4
U 4
< 4
at ^
4 4
X 4
0 4
4
4
4
UJ 4
J
^X <
>-< 4
at 4
4 4
> 4
4
4
4
4
4
41
41
41
41
1/1 41
U «
1-1 4
Ul 41
•-! 41
Q£ 4
J »
t- #
O 4
< 4t
T; *
< *
i «
«
i
-i *
•^ 41
U 41
!/? *
>- «
X »
a. #
*
4>
*
*
»
X *
< *
J »
a: »
^
y
1*1
i-
^
o
_j
u.
Q
O
co
0
o
0.
z
t Ul
» t-
t
^
t
t
t
t •»
• O
li.
t
(
»
t
t
» a
> z
t UJ
f
»
»
t O
f Z
» UJ
>
>
t
» o
t Z
>• )-(
> z
f r<
z
h
h
•H
^
u
Ul
Jl
Jl
tj)
^
at
u
X
a.
UJ
Q
^-
I
3
Z
-*
Q
0
CQ
Q
O
O
•
^
0
-^
—*
•
r~
u.
•
3
-0
^»
•
H-
u_
-^
•^
z
o
o
•
0
•
\£
*^
^O
^
•
*
in
*Q
r-
h*
r-
"t
r^
h-
cg
^
o
CM
•
o
o
•
•H
f^
f».
O
3*
•
*0
0
-H
•
KJ
—4
e
0
•
0
«H
O
o
•
o
-0
o
o
•
f-
o
o
•
10
^
in
•
\o
^
vO
^
•
*
•
(M
O
o
•
»
>0
vO
•
vC
^4
^O
*
*
CO
O
vO
fO
r-
•^
^
f*.
U>
^
O
O
0
-
-o
o
o
•
in
.H
O
o
•
o
0
•
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing]
1. REPORT NO.
EPA-600/2-78-185a
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Short Course Proceedings; APPLICATIONS OF COMPUTER
PROGRAMS IN THE PRELIMINARY DESIGN OF WASTEWATER
TREATMENT FACILITIES; Section I: Workshop Lectures
5. REPORT DATE
September 1978 (Issuing Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
James W. Male and Stephen P. Graef (Editors)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Pritzker Department of Environmental Engineering
Illinois Institute of Technology
Chicago, Illinois 60616
10. PROGRAM ELEMENT NO.
1BC611
11. CONTRACT/GRANT NO.
R-805134-01
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research Laboratory—Gin.,OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
EPA Project Officer: Richard G. Eilers (513) 684-7618
16. ABSTRACT
This document consists of the notebook supplied to each participant in the short
course. It is divided into two main sections. Section I, contained herein, contains
the nine workshop lectures. The lecture writeups provide information on how to
utilize the Executive Program to meet specific user needs. Such needs may call for
modification to or addition of a subroutine to the program. Section II contains the
users' guide and program listing, and it describes how to use the main program and
each of the 27 subroutines. This document describes the most recent version of the
Executive Program. However, the continuing nature of the work in this area means that
revision and additions are likely. These modifications will not change the basic
structure of the program. Care should be taken to verify that the users' guide
corresponds to the correct version of the Executive Program.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Waste treatment
*Models
Sewage treatment
Design
*Cost estimates
*Performance
*Cost effectiveness
Mathematical models
Sewage treatment
Water pollution
Executive program
Preliminary design
Computer program
Design engineering
Sanitary engineering
13B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
206
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
196
1USGPO: 1978-657-060/1470 Region 5-11
-------
|