WATER POLLUTION CONTROL RESEARCH SERIES • 17090 EEM 12/71
INVESTIGATION OF RESPONSE SURFACES
OF THE MICROSCREEN PROCESS
U.S. ENVIRONMENTAL PROTECTION AGENCY
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes the
results and progress in the control and abatement of pollution
in our nation's waters. They provide a central source of
information on the research, development and demonstration
activities in the Environmental Protection Agency, through
inhouse research and grants and contracts with Federal, State,
and local agencies, research institutions, and industrial
organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Chief, Publications Branch
(Water), Research Information Division, R&M, Environmental
Protection Agency, Washington, B.C. 20U60.
-------
INVESTIGATION OF RESPONSE SURFACES
OF THE MICROSGREEN PROCESS
by
Engineering-Science, Inc.
4242 Airport Road
Cincinnati, Ohio 45226
for the
Office of Research and Monitoring
ENVIRONMENTAL PROTECTION AGENCY
Project #17090 EEM
Contract No. 14-12-819
December 1971
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1.25
-------
EPA REVIEW NOTICE
This report has been reviewed by the Environmental Protection
Agency and approved for publication. Approval does not signify
that the contents necessarily reflect the views and policies
of the Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or
recommendation for use.
11
-------
ABSTRACT
Field, laboratory, theoretical, and state-of-the-art studies were
conducted with regard to utilization of microscreens for teritary
treatment applications. Field studies were conducted with two
pilot microscreen units, using a variety of screen sizes and
types, for activated sludge, primary, trickling filter, and ox-
idation pond effluents. Particle distribution of the effluents
(microscreen influents) were found to be the key characterizing
parameter in determination of treatment effectiveness. Overall
effectiveness of solids removal was low, and is ascribed to de-
ficiencies in microscreen design practice for the transfer of
screened solids from the screen to the backwash system and out
of the microscreen unit.
A computer model of the process was developed in a format com-
patible with EPA Executive Program or Optimization of Treatment
Systems. This project was submitted in fulfillment of Project
No. 17090 EEM and Contract No. 14-12-819, under the sponsorship
of the Environmental Protection Agency.
111
-------
CONTENTS
Section Page
I Conclusions I
I I Recommendations 3
III Introduction and Summary 5
IV Characterization of Microscreen Process H
V Experimental Programs 19
VI Results of Field Investigations 27
VII Development of Subprogram Model 65
VIII Acknowledgements 81
IX References 83
X Publications and Patents 85
XI Glossary 87
XII Appendices 89
A Description of Sewage Treatment Facilities 89
B Operating and Analytical Procedures 101
C Field Program Basic Data 111
D Subprogram Model Listing 129
-------
FIGURES
I Flow Diagram for Microscreen System 12
2 Separation, Transfer, and Backwash subprocesses in
Microscreen Process 15
3 Process Flow Sheet 2O
4 Relationship Between Drum Pool and Influent Suspended
Solids Concentration and Drum Speed at Constant Backwash
O(~)
Pressure JU
5 Relationship for Drum Pool and Influent Suspended Solids
Concentration and Backwash Pressure, Drum Speed = 1.8-2.5
sq m/min 3^
6 Relationship between Drum Pool and Influent Suspended
Solids Concentration at Various Drum Speed 33
7 Relationship between Drum Pool and Influent Mean Particle
Sizes and Drum Speed and Backwash Pressure 35
8 Relationship between (a, n~ D - a. _^ .) and Drum Speed and
Backwash Pressure LOG~P LOG~ ' 36
9 Relationship between Overall Suspended Solids Removal
Efficiency and Solids Loading vs. a, nr_p 40
10 Relationship between Overall Suspended_Sol ids Removal
Efficiency and Solids Loading vs. NPS/dp 41
II Relationships between Overall Run Hydraulic Parameter and
Solids Loading - All Wastewater Types 44
12 Correlation Curve-Recovery of Applied Backwash Water as
Throughput Backwash Water for No Influent Flow 47
13 Relationships between Backwash Screen Hydraulic Character and
Run Time and Screen Loading Parameters, Fabric Acclimati-
zation-Run 0 50
14 Response Surface Relationship for Backwash Efficiency and
Fabric Nominal Pore Size -^
15 Response Surface Relationship for Backwash Efficiency and
Drum Speed Stainless Steel Fabrics 5^
16 Response Surface Relationship for Backwash Efficiency and
Drum Speed I Oy Nylon Fabric 5S
v i
-------
Number Page
17 Response Surface Relationship for Backwash Efficiency and
Backwash Pressure 57
18 Relationship between Yield (Effluent Basis) and Specific
Effluent Flow Rate 59
19 Relationship between Purchase Cost/Unit Effective Area of
Microscreens and Effective Area of Microscreen 72
20 Daily Power Requirement per Effective Area, Microscreens
with High-Pressure Spray Backwash Systems 73
21 Predicted vs. Actual Microscreen Suspended Solids Removal
Efficiency across Drum Pool Test Runs 75
22 Calculate Width of Microscreen Fabric Required in Test Runs 76
23 Sensitivity of Performance to Drum Pool Characteristics 78
24 Facility Layout, San Leandro, California, Water Pollution
Control Plant 90
25 Set Up of Pilot Microscreen System, San Leandro, California,
Water Pollution Control Plant 94
26 Facility Layout, Concord, California, Water Pollution
Control Plant 95
27 Set-Up of Pilot Microscreen System, Concord, California,
Water Pollution Control Plant "
28 MTA Test Head 106
29 Pi lot-Scale Medium Testing Apparatus (MTA) 107
VII
-------
TABLES
Number Page
I Elements and Information Requirements for Microscreen
Process 16
2 Principal Characteristics and Components of Microscreen
and Chemical Pretreatment Units 21
3 Microscreen Fabrics Available with the Pilot Plants 23
4 Overview of Experimental Program 25
5 Summary of Overall Run Suspended Solids Removal Observations 39
6 Run 0 - Fabric Acclimatization Experiment - Summary of
Observations 49
7 Summary of Cost Data Microscreen Manufacturers 71
8 Program Parameters 79
9 Summary of Weekly Monitoring Data During Pilot Microscreen
Program, San Leandro, California 91
10 Summary of SVI and SDI Data During Pilot Microscreen Program,
Activated Sludge Process, San Leandro, California 92
II Annual Average Characteristics of Wastewater Streams,
Concord, California, Water Pollution Control Plant 97
12 Characteristics of Wastewater Streams Used as Microscreen
Influents, Concord, California, Water Pollution Control
Plant 98
13 Subrun Operating Schedule 105
14 14 through 30 - Basic Data Tables from Field Program HI
VI11
-------
SECTION I
CONCLUSIONS
The major conclusions reached as a result of the field, theoretical,
laboratory, and state-of-the art components of the research are:
Present-day microscreen designs show many characteristics which
mitigate against the effectiveness of microscreen units for
solids separation/tertiary treatment applications.
The microscreen process is not in fact a single process,
but is composed of three essentially independent sub-
processes (solids separation; transfer of screened solids
to the removal zone; removal, generally by backwashing).
The two sub-processes of transfer and backwash ing are the
"weak links" in the overall process. Solids "fall back"
off the screen into the drum pool during transfer, and
splash effects during backwash can lead to concentration
of solids in the drum pool and ineffective ultimate removal,
with the process thus operating in a "liquid-separation"
rather than a "solids-separation" mode.
The "State-of-the-Art", as available in the literature, is
largely descriptive and empirical; few mechanistic or theore-
tical predictive models have been devised or utilized.
As a consequence, it is difficult to compare and transfer
experience from one application to another since it is
not clear what the relevant parameters are.
Insofar as available technology is organized at all, it is
presented in terms of gross parameters, such as the Filtera-
bility Index, which have not been related to, or derived from,
operational or physical models of the process itself.
A mechanical screening model of filtration appears to be an
adequate "first-cut" definition of the solids separation sub-
process.
Fundamental parameters associated with the model which can
be used to transfer, compare, and predict performance of
the separation sub-process are the particle size distribu-
tion and the solids loading rate of the influent to the
microscreen process.
Given the characteristics of current microscreen design
practice, it is difficult to devise theoretical or empirical
models of the transfer and backwash sub-processes.
-------
A steady-state, gross input-output view of the microscreen
process as a whole is totally at odds with the basic nature of
the process, which is fundamentally that of a dynamic feedback
control problem among the three sub-processes.
In the light of the above, the mathematical model of the
microscreen process, which is steady-state, and theoretically-
based in the screening sub-process, and empirical in the
backwash and transfer sub-processes, is not an adequate
predictor on whjch to base the design of microscreen units,
although it can serve to delineate regions of effectiveness.
Accordingly, field pilot studies using small microscreen
units should be conducted prior to final design and, in
fact, prior to deciding definitively to use microscreening
in any given application, prior experience notwithstanding.
(It should be noted that, at present, a designer has essen-
tially a binary (yes-no) decision with regard to installa-
tion of microscreens, said decision of necessity based in
toto on information supplied by manufacturers.)
-------
SECTION II
RECOMMENDATIONS
The principal recommendations for further effort and extension of research
and design practice of the microscreen process are as follows:
Design configurations as now utilized in most commercially
available microscreens should be evaluated and modified to maxi-
mize the efficiency of the transport and backwash sub-processes,
in terms of removing solids from the microscreen unit.
The theory of microscreening behavior should be extended on the
basis of:
delineation of the behavior of the three sub-processes of
screening, transport, and backwash;
consideration of the dynamic feedback relationships between
and among these sub-processes.
With regard to the development of a mathematical model of the micro-
screening process:
It appears that the current design practices with regard to the
transport and backwash sub-processes introduce a stochastic
component into the performance of the overall process, making
it difficult to predict process performance on a straightforward
a priori basis. For an adequate mathematical model to be deve-
loped on the basis of physical or operational theories, the
microscreen design itself must be modified to reduce the stochas-
tic component, and make process performance more susceptible to
predictive expressions.
A dynamic formulation is appropriate and necessary.
Particle size distribution (PSD) has been found to be a funda-
mental parameter in determining the effectiveness of the screen-
ing sub-process.
In an "optimal" microscreen design, the overall process performance
should be controlled solely by the effectiveness of the screening sub-
process, for which case the PSD will remain a fundamental parameter.
It is possible to determine empirically the PSD of an effluent in an
existing plant, using direct observations. However, there exists no
predictive model for estimating the PSD of the effluent from a plant
not yet in operation; such a capacity has not yet been developed to
date because PSD has not been considered a parameter of design signifi-
cance. Accordingly, a "boundary condition" for an adequate model of
the microscreen process is the availability of predictive capability
for determining particle size distribution as a function of design and
operation parameters.
-------
SECTION III
INTRODUCTION AND SUMMARY
MICROSCREEN PROCESS PERSPECTIVE
The purpose of the microscreen process is to separate solids from sus-
pensions on a continuous basis, using a separation medium which is con-
tinuously regenerated. The application of the process has evolved in
recent years from that of removing gross solids (solids larger than the
interstices of the separation medium) from process streams to that of
fine solids separation wherein one of the dimensions of the solids is
smaller than the largest dimension of the pores of the separating medium.
The success of the former type of application is dependent on physical
retention, whereas the latter is dependent on some form of filtration.
The latter application of the microscreen, of interest in the present
study, is representative of many new and innovative applications of
processes in wastewater management as a result of the rapid evolution
in water quality objectives in recent years. Because of this circum-
stance, a historical base of performance information from which the
efficacy of the microscreening process can be established does not
exist. This situation is exemplified by comparing the historical data
base available for the microscreen process with that for the activated
sludge process and its variations.
The present report on the microscreening process is a synthesis of the
findings of a two-year investigation, the principal objectives of which
were:
To characterize the microscreen process as to its component sub-
processes, the mechanisms operative in the subprocesses, and the
pertinent solids characteristics in microscreen influents which
affect process performance.
To operate and evaluate the performance of pilot-scale micro-
screen units as a tertiary treatment device for several diffe-
rent types of secondary effluents, using a diversity of micro-
screen fabrics and the full range of operating parameters avail-
able in the pilot units.
To develop a mathematical model of the microscreen process as a
computer subroutine compatible with the EPA Executive Program,
and having the capability of predicting microscreen performance
when the process is used for treatment of secondary effluents
as well as capital, operating, and maintenance costs associated
with use of the process.
The scope of the study was organized into three activity areas to achieve
the above objectives:
(I) A state-of-the-art evaluation of the theoretical and practical
aspects of the study as germane to defining relationships
-------
between performance and design and operating variables for
the microscreen process (response surfaces) valid in any
application context.
(2) Formulation of an operational simulation model of the process
as a tool for indicating probable microscreen parameter beha-
vior to be encountered in the field study and as a basis for
designing the field investigations.
(3) A field-scale pilot plant study, designed to permit validation
or redefinition, of the assumptions and hypotheses of the
operational model, and to provide a basis for calibration of
the final, analytical, mathematical model for use in predic-
ting microscreen performance with a large number of influent
sources, fabrics, and operating modes.
INQUIRY APPROACH AND SUMMARY
The information gained in each of the above steps was used to refine the
predictor mechanisms for each subprocess, and to develop the field
program, as described below.
State-of-Art Evaluation
It was concluded from the state-of-the-art evaluation that:
The conceptualization of the microscreen process is in a primitive
state; efforts to date in terms of describing the operation of
the microscreen as a system of interacting sub-processes, or in
terms of mathematical modelling of the process, have not succeeded
in forwarding the process beyond the phenomenologicaI stage of
evaluation, understanding, and application. Simply stated, there
exists no unifying concept for comparing and transferring state-
of-the-art data from one source to another; nor does there exist
a theory of microscreening explaining why the process responds
as it does in any given application.
Much of the available information on the process, whether it be
at the phenomenologicaI or deterministic stage of process evalua-
tion and application, is proprietary in nature and unavailable to
those individuals responsible for assessing the efficacy of such
microscreen applications.
Insofar as the state-of-the-art data is organized at a I I, it is
organized in terms of gross parameters, such as FiIterabiI ity
Index, which have not been related in any manner to operational
or physical models of the process itself. Thus, the data appears
in a traditional format of statistical correlation, with no
demonstration of causation. (After a review of the potential for
utilization of the gross parameters of fiIterabiIity index,
sludge volume index, and sludge age, it was deemed inappropriate
to pursue this "statistical correlation approach" with these
parameters in the project effort, for the above cited reason,
hence no further investigations on these parameters were carried
out. )
-------
There is little published information on how to design either the
microscreen unit per se or the application of the microscreen unit
for any general set of performance and operational objectives;
each application of the process has been viewed and documented
as unique rather than as an extension of process utility in a
general framework.
Performance of various microscreen installations has been tracked
alternatively by: liquid balances , solids balances, screen and
solids character, and screen loading rates; very few of the
applications have been described with all of the types of infor-
mation defined above; in fact, interpretabIe solids balances
were reported in only two references in the entire body of
literature and the role of PSD was considered in only one
reference (Reference I).
Additionally, based on a review of the solids removal mechanisms poten-
tially operative in the microscreen process, it was concluded that:
No presently available data or theoretical analyses indicate
that one mechanism is universally responsible for particle
removal; nor is it yet possible to show quantitatively which
mechanism may be controlling in the solids separation sub-
process under any given set of physiochemicaI conditions.
In the absence of firm information, and because of the fundamental
development of the microscreen as a screening process, the
mechanism of mechanical screening was selected for use in the
deterministic model developed in the study.
Corollary to the above, the mechanism of mechanical screening can be
documented only by quantitative routing of particle populations onto
and from the screen, PSD data on the influent and effluent streams
being the critical mechanism-1 eve I parameter of concern in the deter-
ministic modelling and field-scale evaluation of the process.
Anticipated Response Surfaces
Operational Mathematical Model
The operational mathematical model was developed to provide a unified
concept in the form of anticipated response surfaces for the pilot
study to provide a framework for examining thi field data, and to serve
as a nucleus for the subprogram model. The development of the model
is described in References 2 and 3. Inasmuch as the development of
the operational model took place prior to the field investigation, the
operational model was calibrated with data developed in laboratory tests
with idealized particle suspensions (Reference 4).
Response surfaces were predicted with the operational model for trick-
ling filter-;,and activated sludge effluents, assuming a log-normal
distritubion of particle sizes in these effluents. Mean particle sizes
were taken as 35y for the trickling filter effluent, and 20y for the
activated sludge effluent, with geometric standard deviations of 2.0
-------
and 1.5 to 4.0 respectively. The nominal pore size of the screen was
assumed to be 30yfor the response surface analysis with the trickling
filter effluent, and 23y with the activated sludge effluent.
The characteristics of the response surfaces developed in these simula-
tions were as follows:
(I) Trapping efficiency:
' The trapping efficiency of the screen at any transverse
section of screen on the screening cycle was found to be a
function of vi or mi, the cumulative solids loading/unit
area on a volumetric or mass basis respectively.
' The trapping efficiency of the screen was found to be a
function of the relative values of initial trapping diameter
of the screen, the mean drum pool particle size (dp), and the
standard deviation of the uni-modal, log-normal pahticle
size distribution (PSD); the rate of increase of trapping
efficiency over the screening cycle was found to increase
with decreasing values of initial trapping diameter (smaller
screen pore size) or increasing values of dp.
Conversely,- the less uniform the PSD of the drum pool sus-
pension (i.e., the lower the standard deviation ai_OG) • the
lesser the rate of increase of trapping efficiency as solids
loading increased over the screening cycle.
The porosity of the cake formed by the retained solids was
found to have little effect on the trapping efficiency.
(2) Hydraulic resistance:
The assumed porosity of the cake formed by the retained
solids was found to have a significant impact on the hydraulic
resistance of the cake; the hydraulic resistance increased
as porosity decreased.
Variation of the standard deviation of the drum pool PSD
was found to have only a nominal effect on the hydraulic
resistance, the effect being one of decreasing hydraulic
resistance as standard deviation increased.
The above indications represented the expected framework and "hypothesis"
for the field investigative program delineation of the physical model of
process behavior.
Observation of Physical Model
The following factors appear to dominate the overall performance of
the pilot units (the physical models of the microscreen process used in
the present study):
-------
(I) A non-uniformity of solids retention occurs spatially across
the screen during the screening cycle as a result of:
Variation of screen pore size distribution as a result of
the manner of construction, handling, and mounting of the
screen.
Solids inputs to the drum pool from fall-back, splashover,
and the influent, and the loss of particles in the effluent,
which serve to create non-uniformity of particle size dis-
tribution and concentration in the drum pool.
(2) Significant losses of solids captured from the drum pool sus-
pensions occur prior to and during the transfer of the concen-
trated sol ids to the washwater col lector.
(3) The key elements of the backwash subprocess appear to be:
The energy contained in the backwash spray system.
The speed of rotation of the drum.
The size of the spray droplets.
The size of the surface indentations in the screen.
The orientation of the pores relative to the trajectory of
the spray-
The manner in which the particles (cake) are attached to the
screen.
The deformation properties of the screen fabric.
(4) A qualitative evaluation of the backwash subprocess indicated
that:
For any given type and size of nozzle and water pressure,
there is a threshold pore size below which cleaning effec-
tiveness declines rapidly, and above which the cleaning
effectiveness is independent of pore size.
For any given combination of type and size of nozzle, water
pressure, and screen size, there is a speed of rotation
which optimizes the synchronization of the intersection of
pores and droplets; above or below this speed the cleaning
effectiveness should decline rapidly.
' For any given combination of nozzle type and size, speed of
rotation, and pore size, there is an optimum water pressure,
above or below which cleaning efficiency decreases.
At the upper and lower ranges of the rotational speeds, sig-
nificant amounts of the retained solids removed by back-
washing fall short or long of the washwater collection trough,
-------
Fabric flexing decreases the capacity of the solids to be
held to the screen during the backwash subprocess.
(5) The above relationships give rise to a multi-dimensional response
surface for the backwash subprocess, with each combination of
dimensions exhibiting either optimal levels or saturation
levels of cleaning efficiency with increasing parameter values.
SUMMARY OBSERVATIONS
As noted above, there was a significant discrepancy between the anticipated
response surfaces derived from the operational model, and the response
surfaces derived from the physical model, i.e., the field program obser-
vations. In keeping with the overall inquiry approach, this discrepancy
indicated that the backwash and transfer subprocess models should be
re-formulated for greater congruence with the "realities" of process
performance. (It should be noted that, in this case, greater congruence
with reality does not imply that the subprocess conceptions were inade-
quate, but rather that the transfer and backwash subprocesses are the major
weak links in the typically-constituted commercially-available microscreen).
The problem with these subprocesses is that solids captured from the drum
pool suspension are not effectively transferred to the washwater collector
from whence they can be removed from the system. The sol ids losses occur-
ring in these subprocesses are recycled directly to the drum pool suspen-
sion, with the result that the drum pool suspended solids concentration
exceeds the influent suspended solids concentration. The result of this
"concentration" effect in the drum pool is that, while the overall micro-
screen process is evaluated and used on the basis of solids reduction
between the influent and effluent, the mechanism effecting solids reduc-
tion is operating against a much greater solids concentration gradient
between the drum pool and effluent than exists between the influent and
effluent. The resolution of this problem will require the development of
microscreen designs to achieve the removal of solids retained during the
separation subprocess cleanly out and away from the drum pool during the
transfer and backwash subprocesses.
The remaining sections present a more detailed description of the investi-
gative procedure, and consequent delineation, of the currently perceived
microscreen process model and response surface. Insofar as possible,
this information was incorporated into the sub-program model of micro-
screen behavior, presented in Appendix D of this report.
10
-------
SECTION IV
CHARACTERIZATION OF MICROSCREEN PROCESS
THE UNIT PROCESS
Microscreening involves the passing of a suspension through a moving medium.
The suspended particles which are removed may be discrete (mineral) or
flocculant (organic or non-organic) and may vary in size from colloidal
(<_ I micron) to coarse suspended particles. Solids separation by micro-
screening is accomplished by a series of complex interactions between
three phases:
A moving solid phase (the screen and cake);
A discontinuous solid phase (the material to be removed);
A liquid phase (usually water).
The two fundamental system aspects of microscreening are:
The screening cycle, during which the suspended solids are removed
and a clarified effluent is produced;
The backwash cycle, which involves flushing the collected solids
from the med ium.
The screening cycle is an operation analogous to the declining rate filter;
as the screen passes from submergence to emergence during the screening
cycle, the headless characteristics of the developing cake increase and
cause a concurrent decrease in the throughput rate. The screen:! ng cycle
is terminated when a segment of the medium emerges from the liquid pool.
During the backwash cycle the accumulated solids are removed from the
medium by the several mechanisms usually associated with hydraulic, air,
or sonic cleaning. Generally, microscreen effluent water is used for
hydraulic cleaning. The yield of the microscreen is the difference be-
tween the quantity of water produced during the filtration cycle and the
quantity of water consumed during the backwash cycle.
Process Components
The microscreen process, as constituted in the pilot plants used in the
present study, has six interrelated components as shown in Figure I:
(I) The Drum Pool, serving as a reservoir for the solids suspension
to be microscreened. Based on observations of the pilot plant
operation, the solids in the drum pool are derived from three sources:
(a) The microscreen influent suspension (primary source).
(b) The fall-back of solids from the screen-solids-water
complex during the portion of the drum cycle between
emergence of a transverse segment of screen and movement of
the segment into the zone of influence of the backwash
sprays.
11
-------
FIGURE
INFLUENT
SPLASHOVER
WASHWATER
COLLECTOR
SPLASHBACK
J
THROUGHPUT
WASHWATER
DRUM POOL
FALL-BACK
SCREEN (ON DRUM)
APPLIED'
WASHWATER
CLEAR WELL
(UNDER DRUM)
I
EFFLUENT
COLLECTOR
BACKWASH
T
TREATED
EFFLUENT
ALTERNATIVE EXTERNAL
SOURCE OF BACKWASH
WATER
FLOW DIAGRAM FOR MICROSCREEN SYSTEM
12
-------
(c) The splashover of so I ids-laden backwash spray water that
falls short or long of the washwater collector rather than
into the collector as intended.
As a result of the above, the characteristics (concentration, particle
size distribution, etc.) of the suspension in the drum pool are generally
dissimilar to those of the influent stream with drum pool suspended
solids concentrations generally exceeding influent suspended solids
concentration.
(2) The Screen (or microscreen fabric), serving as the matrix on
which the screen-sol Ids^water medium effecting particle removal is
formed.
(3) The Backwash System, serving the dual function of:
(a) Applying energy (in the form of a pressurized spray of
washwater) to the screen to dislodge retained particles;
(b) Effecting the collection and transport of solids-laden
washwater away from the microscreen in the Washwater
Col lector.
Because of the fall-back of retained solids from the screen after emergence
from the drum pool, not all particles captured in the screening cycle
are actually transferred to the zone of influence of the backwash system.
Splashover (defined above) occurs as a result of the size, shape, and
location of the washwater collector relative to the trajectory of the
"shower" of backwash water passing through the moving screen. Splashback
occurs as a result of the capture of water from the backwash spray on the
outer surface of the drum, from where it is conveyed (with the rotation of
the drum) directly to the clear well without passing through the screen.
Because of splashback and splashover, the throughput washwater flow rate
(as measured in the washwater collector) can be significantly less than the
applied washwater flow rate, depending upon operating conditions.
(4) The Clear Wei I, containing not only screened process effluent
but also splashback from the backwash system.
(5) The Effluent Collector, conveying effluent from the clear well to
ultimate disposal, and serving as an intake pool for backwash system
water.
" BPROCESS
The microscreen system as described above has three distinct subprocesses
operative over the drum cycle, viz:
(I) Separation Subprocess, in which solids are captured on, or passed
through, a transverse segment of screen as the screen is transferred
through the drum pool from submergence to emergence.
(2) Transfer Subprocess, in which the captured solids on the screen
segment are transferred from the point of emergence of the segment
13
-------
from the drum pool to the zone of influence of the backwash sprays.
(3) Backwash Subprocess, during which energy is transferred at some
level of efficiency from the applied backwash stream to the screen
segment in order to remove retained solids and to rejuvenate the
solids retention and hydraulic capacities of the screen segment.
The impact of each subprocess on overall process behavior can be envi-
sioned in terms of a profile of the retained solids on the screen segment
during a drum cycle. Such a profile, developed on the basis of observation
of the microscreen pilot units, is shown in Figure 2.^ The key features of
the profiIe are:
(I) The accumulation of retained solids on the segment in the
separation subprocess at a decreasing rate over the screening cycle
(submergence ->• emergence), corresponding to a decreasing rate of
liquid throughput as the hydraulic resistance of the screening medium
increased at constant hydraulic head.
(2) The fall-back of solids from the screen-solid-water complex
during the transfer subprocess, due to:
(a) Erosion of sol ids from the complex at emergence as a
result of surface turbulence in the drum pool.
(b) Drainage of water from the complex.
(3) The reduction of retained solids to a residual level during the
backwash subprocess, the residual level representing the initial
mass of retained solids on the segment as it is submerged and enters
the next screening cycle.
Subprocess Elements
The microscreen process, when viewed at the level of a transverse segment
of screen passing through the drum cycle, is analogous to a declining
rate filter and the liquid and solids transfers and the physical state
of the system occurring therein can be described in terms of the elements
operative in a granular bed filter system. A listing of these elements,
and the information required to describe each is presented in Table I.
The elements can be categorized as associated with operation (liquid
throughput rate, solids input and emission rate, solids retention),
control (headloss, backwash, surface renewal), or system state (solids
character, filter medium).
The distinction made between the screening medium and solids retention
in the elements of the microscreen is fundamental to the inquiry approach
used in the present study. The screening medium is defined as the physi-
cal-chemical-biological medium on which the solids are captured and
retained. The components of the medium are water, suspended solids,
colloidal solids, dissolved solids, and the screen and the mechanisms
operative in this environment are those related to inertial, electrical,
concentration, and chemical forces effecting solids capture and imparting
14
-------
SEPARATION, TRANSFER. AND BACKWASH SUBPROCESSES
IN MICROSCREEN PROCESS
Ul
Q
LU
•Z.
-------
TABLE 1
ELEMENTS AND INFORMATION REQUIREMENTS FOR
MICROSCREEN PROCESS
Element
Information Requirements
Liquid Throughput Rate
Solids Input and Emission Rate
Solids Character
Screen (microscreen fabric)
Solids Retention
Filter Medium
Surface Renewal
Backwash
Headless
Influent liquid flow rate; liquid physical properties
Mass rates for dissolved, colloidal, and suspended solids
Type of pretreatment; particle description; particle size
distribution; particle shear strength
Nominal aperture; weave; material; clean screen headloss
factor; effective pore (or particle trapping) diameter;
and rate of presentation
Residual retention of dissolved, colloidal, and suspended
solids on filter after backwash; retention of dissolved,
colloidal, and suspended solids on filter during screening
cycle (both retentions analogous to a surface density
mass/area)
Effective pore diameter; cake porosity, depth, and density;
composite headloss character; solids separation mechanism
Rate of drum rotation
Applied rate of backwash; net rate of backwash, pressure;
cleaning efficiency
Pressure drop
-------
a hydraulic resistance to the captured solids. The interaction of the
components and dominant mechanisms in the medium serves to establish the
behavior of the separation subprocess. Solids retention represents,
at any point in the drum cycle, the inventory of suspended, colloidal,
and dissolved solids stored on the screen (Figure 2). Solid retention
is characterized in the present study by a surface density parameter
(MC, mass of solids retainer per unit area).
The elements listed in Table I can be interrelated and summarized in the
form of a feedback model having the following cause -> effect relationships
(Reference I):
(I) The liquid throughput rate and headless on a screen segment
being a function of the ava i lable head (drum pool -»• clear we I I ) and
the screening medium hydraulic resistance.
(2) Solids retention being a function of the solids character,
solids input rate (drum pool -> screen), liquid throughput rate
(drum pool -> clear well ), and screening medium.
(3) Rejuvenation of the medium being a function of the efficiency
of the backwash subprocess and the rate of drum rotation.
17
-------
SECTION V
EXPERIMENTAL PROGRAM
The experimental program was organized around an equipment system con-
sisting of two microscreen units, a head tank unit, and appurtenances.
With this basic equipment an experimental program was developed using
combinations of different fabrics and five different wastewaters at two
different sewage treatment plants as described below.
MICROSCREEN AND CHEMICAL PRETREATMENT UNITS
Process Flow Sheet
The basic equipment used in the study consisted of two microscreen units
and one chemical pretreatment unit, each of which was mounted individually
on trailers to expedite mobility. A process flow sheet for the chemical
pretreatment unit and one microscreen is shown in Figure 3. The prin-
cipal characteristics and components on the units are listed in Table 2.
The piping system (Figure 3) is designed to allow the transfer, by either
gravity or pumping of:
(I) Process influent to the head tank unit (HTU) or, bypassing
the HTU, to the microscreen unit.
(2) HTU effluent to the microscreen unit.
(3) Microscreen effluent and throughput washwater from the micro"
screen unit.
The piping system also permits applied washwater to be transferred via
the backwash pump from either of two sources: the microscreen effluent,
or an external washwater source (e.g., tap water).
A manometer system was installed to permit measurement of liquid levels
in each microscreen unit and simultaneously in the drum pool (upstream of
the screen), in the clear well under the screen, and in the backwater of
the weir box.
Control Variables
The principal control variables with the experimental equipment are:
(I) Source and character of process influent.
(2) Type and character of fabric.
(3) Headless across the screen from submergence to emergence.
(4) Backwash pressure (and flow rate).
(5) Drum rotational speed.
Five types of influent streams were used in the present study:
(I) Clarified standard rate activated sludge process effluent.
(2) Clarified high rate trickling filter effluent.
19
-------
FIGURE
PROCESS FLOW SHEET
HEAD TANK UNIT
1 1
L
j^
Z
ID
UJ
LU
-_ - *
CXL
0
OO
CtL
O
i
NOTE:
1
6ea 400 liter tanks I
•^^
™
_J
UNIT -^ <
r~*"
1
1
THROUGHPUT
WASHWATER
J
X
I
1
v r
. PROTFSc;
1 EFFLUENT-*
V ^EFFLUENT CHANNEL
PROCESS PUMP CAN ALSO BE USED TO
PROCESS
INFLUENT
\ \
\
1
1
1
1
|
1
^ , JL PROCESS
/ N PUMP
~" "~ \ /
\ y
>^^ ^s
^APPLIED WASHWATER
EXTERNAI
IER BOX WASH-
WATER
SOURCE
~~ V / ~~
BACK~T^ASH
PUMP
PUMP EFFLUENT FROM
EFFLUENT CHANNEL AND/OR THROUGHPUT WASHWATER
20
-------
TABLE 2
PRINCIPAL CHARACTERISTICS AND COMPONENTS OF
MICROSCREEN AND CHEMICAL PRETREATMENT UNITS
UNIT
ITEM
DESCRIPTION
Microscreen
Head Tank
Drum size
Drum submergence
Drum rotation rate
Maximum headless through screen
Process pump capacity
Backwash system capacity
Individual tanks
4-ft diameter x I-ft wide; 0.8 sq m screen area
50 to 65 percent of drum diameter
Variable from one to 12 rpm, or 12 to 150 ft/min, or 0.8
to 9.6 sq m screen area/min
20 cm (8 inches)
Variable to 240 Ipm (liters/minute)
Variable to 35 psig and.to a maximum backwash rate of
35 Ipm
Tank volume, 400 I (6 each, interconnected)
-------
(3) Unclarified high rate trickling filter effluent.
(4) Primary effluent.
(5) Oxidation pond effluent.
The characteristics of these individual streams are described in Appendix A.
The types of fabrics used in the experimental program are listed in
Table 3. The selection used included a total of eight stainless steel
fabrics having a nominal pore size (NPS) of 12 to 40y. The other fabrics
used in the selection were made of nylon, polyethylene, and polyester,
having a range of NPS from 10 to 25y. The detailed characteristics of all
fabrics are discussed in Reference 4.
System Operations
The chemical pretreatment and microscreen units were used consistently in
one of two operating modes (Mode A or B) for the runs made in the experi-
mental program. The common aspects of the two modes were:
(I) The HTU (head tank unit) was used to provide for gravity
transfer of flow into the microscreen units.
(2) Two one-HP, 1-1/2 inch diameter intake/discharge Marlow pumps
were used (one per microscreen unit) to transfer process influent
from the source to the HTU (these pumps were not supplied with the
mi croscreens).
(3) The microscreen process pumps (one per unit) were used to
transfer the microscreen effluent flow from the effluent channel to
the point of ultimate liquid disposal (this latter flow option is not
illustrated in Figure 3).
(4) Tap water was used for screen backwash ing.
In the Mode A operations, both microscreen units were supplied from a
common influent source (the CPU), although each unit was operated inde-
pendently. Mode A was used for most of the runs in the field program
and the experimental protocol for these runs is described subsequently.
In the Mode B operation, one of the microscreen units (Unit A) was used
solely to produce a throughput backwash stream containing selectively
larger particles as a function of the screen size used. The throughput
backwash stream from Unit A was mixed with the HTU effluent to form a
composite process influent to Unit B containing a controlled range of
particle sizes and suspended solids concentrations. With appropriate
screen selection in Unit A and proportioning of the throughput backwash
stream from Unit A with HTU effluent, it was possible in Mode B to selec-
tively control the particle size distribution and concentration of sus-
pended solids in the drum pool of Unit B.
EXPERIMENTAL PROGRAM
The general approach taken in the field program was as outlined in
Reference 4. This approach was comprised of three phases of activity:
22
-------
TABLE 3
MICROSCREEN FABRICS AVAILABLE WITH THE PILOT PLANTS
Manufacturers
Nominal Pore
Rating (microns)
40
30
23
21
18-22
13 - 18
12 - 15
25
25
20
10
Material
Stainless Steel
Stainless Steel
Stainless Steel
Stainless Steel
Stainless Steel
Stainless Steel
Stainless Steel
Nylon
Polyethylene
Pol yester
Nylon
Weave
Dutch Twi 1 1
Dutch Twi 1 1
Dutch Double Warp Twill
Dutch Twil 1
Reverse Dutch Twi 1 1
Single Dutch Twi 1 1
Reverse Dutch Twi 1 1
Square
Recta ngu lar
Square
Calendared Square
Mesh
Count
120/400
1 20/600
1 20/600
200/830
600/125
165/800
720/140
462/462
1 1 2/48
200/400
462/462
Pore Census
(No./sq.in.)
96,000
144,000
144,000
330,000
150,000
264,000
200,000
420,000
10,500
160,000
420,000
to
-------
(I) Physical characterization of the microscreen.
(2) Description of the response surfaces of the subprocesses in the
microscreen.
(3) Evaluation of the general applicability of the response sur-
faces for a variety of influent solids sources and microscreen
fabrics.
On overview summary of the experimental program is presented in Table 4.
Eighteen run sets (numbered 0 to 17) were made using Units A and B, and
a total of 34 individual runs were made using II different fabrics.
The field program was conducted at the San Leandro and Concord sewage
treatment facilities, both located in the San Francisco Bay area
(Appendix A). The San Leandro facility has dual biological waste treat-
ment systems, a standard rate activated sludge system for treatment of
domestic wastes and a high-rate trickling filter system for treatment of
combined domestic and industrial wastes. The Concord facility is
equipped with a high-rate trickling filter system followed by an aerobic
pond system and is used to treat both domestic and light industrial wastes,
The first phase of activity in the field program was conducted prior to
the start of the run sets. Phase 2 consisted of Run Sets 0 to 4, 16, and
17; Phase 3 consisted of Run Sets 5 to 15 as outlined in Table 3.
Individual operating protocols were used for Run Sets 0 to 4 and stan-
dardized operating protocols were used for Run Sets 5 to 15 (Mode A) and
Run Sets 16 and 17 (Mode B). These and the analytical procedures used
in the field program are described in Appendix B.
24
-------
TABLE 4
OVERVIEW OF EXPERIMENTAL PROGRAM
TREATMENT
PLANT
San
Leandro
Concord
RUN
NO.
0
1
2
3
4
5
6
6
7
6
9
10
11
12
13
14
15a
15b
16
17
DATE
(1971)
16-17 Feb
19-20 Feb
23-24 Feb
25 Feb
2 Mar
3 Mar
4 Mar
4 Mar
5 Har
9 Har
10- Mar
17 Mar
18 Mar
19 liar
25 Har
26 Har
30 Mar
30 Mar
1 Apr
2 Apr
- i
WASTE
SOURCE
A3
AS
AS
TF
AS
AS
AS
AS
AS
TF
TF
TF
TF
. TF
PE
UTF
PE
TF
-6P
TF
UNIT
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
B
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
A
B
MICROSCREEN
FABRIC
act, ss
21pSS
30M SS
21 „ SS
30M SS
21 M SS
30M SS
40p SS; 21pSS
30W SS
21p SS
12-15U SS
20u Polyester
23u SS
25w Polyethyl.
15-18u SS
10p Nylon
lOii Nylon
23p SS
15-lBu SS
21p SS
18-22p SS
Zip SS
20p Polyester
lOii Nylon
12-lSii SS
10M Nylon
12-15p SS
18-22M SS
12-15P SS
18-22u SS
12-15ii SS
18-22P SS
12-15)i SS
18-22p SS
Varying
15-18)i SS
SUCRUN
BACKl.'ASII
PRLSSURES
(psiy)
7-1/2 to 20
4-1/2 to 32
8 to 35
5 to 32
25
10 to 40
10 to 15
10 to 20
30
30
15, 23, 30
15, 23-30. 35
15, 23, 30
15, 23-30. 35
15, 25. 35
15, 25, 35
25, 35
15, 20. 25
15
15
15
15
15, 20
15, 20
15
15, 20,25
15
15, 20, 25
20
20
20
20
15 to 20
20 to 30
20
20 to 30
NOTES
Fabric acclimatization run
Fabric acclimatization run
24-hour run
24-hour run
Backwash subprocess run
Backwash subprocess run
12-hour run
12-hour run
Final shakedown run
Final shakedown run
Start of Mode A operations
)F1n1sh of Mode A Operation
)Mode B Operation
Notes: (I) AS - clarified activated sludge of fluent
(2) TF - clarified trickling filter effluent (high rate at San Leandro and Concord)
(3) PE - primary effluent - trickling filter recycle mixture used as Influent to trickling filter
(4) UTF - trickling filter effluent (unclarlfled)
(5) OP - oxidation pond effluent
25
-------
SECTION VI
RESULTS OF FIELD INVESTIGATIONS
The field investigation was developed in consideration of the observations
presented in Section IV and with the overall objective of defining quan-
titively the behavior of the pilot plant systems. The basic data require-
ments of the model development necessitated that process behavior be
examined from a steady-state rather than a feedback basis as appeared
appropriate (Section IV). Because of the overriding requirement for a
steady-state profile of the process, the approach taken in data analysis
was to plot against a time-scale all of the parameters observed in the
individual runs, and to select from these profiles the points in time,
and associated parameter values, for which process behavior was judged
to be quasi-steady-state. With the steady-state criterion used (discussed
below), it was possible to obtain over 50 quasi-steady-state points
which, in the subsequent presentation, have been treated as representative
of microscreen process behavior in the quasi-steady-state mode. The
basic data obtained in the field investigations are presented in Tables 14
to 30 of Appendix C.
The steady-state criterion used in the data analysis required that process
performance be stable with respect to suspended solids removal, drum
speed, backwash pressure, and solids loading. Trapping efficiency in
terms of suspended solids removal was measured across the screen (drum
pool ->• effluent) rather than through the unit (influent -»• effluent) in
view of the "concentration effect" described in Section IV. Because of
the multi-dimensional response surfaces that are operative in the micro-
screen process, it is recognized that the above criterion may have biased
the selection of the data, i.e., that one or more of the variables con-
sidered or not considered in selecting a steady-state point may have been
located in a zone of instability in the response surface region sampled
(in a region where a slight shift in the magnitude of the variable would
result in a significant shift in process efficiency). Within the scope
of Information that could be developed in the present study, the re-
searchers had no other alternative but to accept the above criterion.
Trapping efficiency across the screen was computed as follows:
f = ioojE- d)
where: f = Suspended solids removal efficiency, percent
MC = Mass of suspended solids retained per unit area of screen
over the screening cycle (see Figure 2)
Ml = Mass of suspended solids loaded per unit area of screen over
the screening cycle
(Note: All symbols are defined in the Glossary of Section XI)
The parameters MC and Ml were calculated as follows:
[Q.X? - Q Xp] (mg/l ' l/min)
S (sq m/min)
27
-------
[Q,Xp] (mg/l ' l/min)
Ml = LJ_ (3;
S (sq m/min)
where: Q = Influent volumetric flow rate (l/min)
0 = Effluent volumetric flow rate (l/min)
E
X = Drum pool suspended solids concentration (mg/l)
X = Composite effluent suspended solids concentration (mg/l)
S = Rate of screen presentation, sq m/min
The hydraulic parameter was computed as:
XpHLA
(units of cm-sec)
pS
where: p = Density of suspended solids (assumed to be 1.05 gm/cc)
H| = Headless across screen, cm
A = Submerged area of screen, sq cm
It was found from an analysis of the PSD (particle size distribution)
data obtained for the influent, drum pool, and effluent suspensions that
the PSD's of the particles could be characterized as _l_pg-normal. The _
PSD's were characterized in terms of two parameters, d and CT.QQ where d
is the mean (50 percent!le) particle size on the log-normal distribution,
^84 3^-ile 's ~^e 84.3 percent!le particle size and:
lQg(d84.33-ile/7)
-LOG = —= (4)
log d
It is noted that PSD was the only parameter defined for the solids sus-
pensions examined in the field program. Gross empirical parameters such
as FiIterabi Iity Index, SVI, and sludge age were neither considered nor
measured in the field program for reasons cited in Section IV.
The parameter values for the steady-state points obtained from the data
analysis are tabulated in Table 30 of Appendix C. Most of the steady-
state points were obtained for runs in which clarified activated sludge
effluent (AS) and clarified trickling filter effluents (TF) were used as
pilot plant influents. Three steady-state points were obtained with
28
-------
primary effluents (PE), ten with unclarified trickling fll.ter effluents
(UTF), and two with oxidation pond effluents (OP).
The numerical values of MC, Ml, and the hydraulic parameter for the steady-
state points as reported in Table 30, represent process performance on
an overall run basis and not at any point during or at the end of the
screening cycle. That is, the parameters MC and Ml represent solids
retention and loading respectively given not only the positive (separa-
tion) effect of the separation subprocess but also the negative effects,
on trapping efficiency, of the transfer and backwash subprocesses as
described in Section IV. The hydraulic parameter is computed in terms
of an overall head loss across the screen and an averaged flow rate over
the screening cycle. Nineteen of the steady-state points (identified by
number in Table 30) were used for subprogram model calibration purposes
as described in Appendix C.
SEPARATION SUBPROCESS
Introduction of Solids to Drum Pool
The initial transfer taking place in the microscreen process is the trans-
fer of influent to the drum pool. As observed previously, solids are
transferred to the drum pool not only from the influent but also as a
result of solids recycling to the drum pool resulting from inefficiencies
and/or inadequacies inherent in the transfer and backwash subprocesses.
The drum pool was also postulated to act as a solids transformer, in that
particle shear caused by turbulence in the drum pool is expected to
result in a transformation of the characteristics of the PSD's. Because
of either circumstance (solids recycle and/or particle shear), the con-
centration and PSD characteristics of the solids suspension in the drum
pool are not expected to be uniform throughout the pool; additionally,
it is apparent that the concentration and PSD characteristics of the
influent and drum pool suspensions will be dissimilar in all but fortui-
tous circumstances.
Within the range of sensitivity of the data, it was possible to examine
the impact of two operational variables (PB, backwash pressure, and S,
rate of screen presentation, area/time) on the parameters defining the
influent and drum pool suspensions (Xs; d; and QLQQ as defined above).
The approach used in defining the response surfaces for each parameter
was first to examine the impact of varying S on parameter ratios (drum
pool/influent) at constant PQ and then to examine the effect of varying
DB at constant S.
The relationship between the ratio of drum poo! and influent suspended
solids concentration and speed (at constant Pg of 15 psig) is illustrated
by the data on Figure 4. (Because of transient influent quality, several
data points were observed at values of XpYX5- less than 1, as shown in
Figure 4). A curve of best-fit relating the ratio Xp/Xs and speed is
shown in Figure 4; the lower boundary of the curve was assumed to be
defined by X§/X? ->• I as S ->• 0, for which case the process cagnot be operated
without rotation of the drum. The relationship between Xp/X and S de-
fined by the best-fit curve has two distinct zones; the first (for
29
-------
RELATIONSHIP BETWEEN DRUM POOL AND INFLUENT SUSPENDED
SOLIDS CONCENTRATION AND DRUM SPEED
AT CONSTANT BACKWASH PRESSURE
c/)Q-,cO-<
X X
w
o
§
o
CO
Q
I—I
J
O
O
O
DH
o
o
CO
CO
2 3
DRUM SPEED, S, sq m/min
30
-------
c c
0
-------
RELATIONSHIP FOR DRUM POOL AND INFLUENT SUSPENDED SOLIDS CONCENTRATION
AND BACKWASH PRESSURE, DRUM SPEED = 1.8 to 2. 5 sqm/min
x x
o
o
o
CO
a
M
co
o
o
a,
O
o
V)
CO
20
eft
o
15
S : 1.8 to 2.5 sq m/min
-o
8
•n
o
20 25
BACKWASH PRESSURE (Pg)
30
35
-------
FIGURE 6
en O.IOTI-I
UJ
o
en
o
o
cc
O
tr
2
UJ
O
co
UJ
-o
O
Note:
For Pilot Microscreens
S (sq m/min) = (0.8sq m/rev)( W.rpm)
Data for all backwash pressures
O
O
O
Xs
—| = l.4 + (0.725)(S-2)
O
o
o
o
o
o
I
1
12345
DRUM SPEED,S , sq m/min
RELATIONSHIP BETWEEN DRUM POOL AND INFLUENT SUSPENDED
SOLIDS CONCENTRATION AT VARIOUS DRUM SPEED
33
-------
of drurr. speed and pressure by the optimization of the backwash subprocess
The relationship between d /d, and S is shown by the data presented in^
Figure 7; the backwash pressure associated with each datum is also indi-
cated. Two general trends can be observed from the data:
(1) d" /d" decreases with increasing S.
(2) The backwash pressures associated with each datum increased
with Increasing drum speed.
On the basis of these trends a best-fit curve has been developed as shown
in Figure !_,_ the lower bound of the curve being defined as dp/d. - I at
S = 0, and dp/d , = 0.75 at S = 8 sq m/mln. TheJIrnited available data
precluded, an examination of the impact of S on dp/d( at constant PB, and
PRon dp/d | at constant S, but given the foregoing context, it appears
that the date, set represents a situation in which the negative effects
of shear on dp/d, were countered over the range of 0
-------
RELATIONSHIP BETWEEN DRUM POOL AND INFLUENT MEAN
PARTICLE SIZES AND DRUM SPEED AND BACKWASH PRESSURE
2.5
UJ
Nl
cn
UJ
o
i-
01
CL
p
UJ
^
O
0
CL
^
or
Q
UJ
— 2 0
UJ
o
t-
cc
<5
CL
1.5
•z.
UJ
S
\—
"Z.
LU
^3
U.
—
Q. M
-o
T3
0.5
o
-
— —
in
O
— P psig —? —
B' JX °9
X
o l2 o15
3O 3O
00 03O 030
~~ _^^
o-« " . _______^^^
IO <*12 38 0
._ O o 22
IO Q WJ(? O
10 o o^o ^>o
1 1
0 2345678
DRUM SPEED, S , sq m/min
^
O
c
H
-0
-------
RELATIONSHIP BETWEEN (cr - cr ) AND DRUM SPEED AND BACKWASH PRESSURE
LOG-P LOG-I
40.4
+ 0.2
o
b^ -°-2
-0.4
-0.6
12 psig = PB
3O
10
IO
15
o 39
I
4 5
DRUM SPEED, S , sq m/ min
-------
Trapping Efficiency
The trapping efficiency of the microscreen process has been viewed
previously from two perspectives in the present study, the anticipated
response surfaces from the simulation model, and the updated physical
model (Section IV).
Given, in the present study, that trapping efficiency was measured on a
drum pool -> effluent basis, the measured efficiency data obtained in the
field program constitute observations of the net removal achieved by the
interactions of all three subprocesses (separation, transfer, and back-
wash), rather than by the separation subprocess alone.
In consideration of the above and the observations in Section IV, it
was possible to examjne the effect of the critical parameters of initial
trapping diameter, dp, and OIQQ only on an overall run basis. Thus, in
approaching the data analysis and presentation, each datum was classified
in terms of: the nominal pore size (NFS) of the fabric used during each
run, as a measure of the initial trapping diameter; dp; and tf|_C)G-P ^e
date were arrayed to permit the mapping and examination of specific planes
across a response surface_relati ng process efficiency (measured as MC/MI),
solids loading (Ml), NPS/dp (a normalized parameter assumed to relate in-
itial trapping diameter and mean drum pool particle size), and
In accomplishing the above, it was originally intended that the data be
viewed insofar as possible as being independent of wastewater source.
The premise of this original concept was that, within certain constraints,
a suspension could be described in terms of PSD character and suspended
solids concentration independent of solids source and history. It is
recognized in this approach that many factors may obviate this position,
the foremost being the manifestations of the different types of chemical,
electrical, and concentration forces operative in the various types of
suspensions examined, resulting in different levels of water-particle
interactions and behavior when cakes are formed from the suspensions
(e.g., porosity). Initial attempts at data analysis within the original
concept did not bear positive results, the range of sensitivity of the
data being a major factor precluding this. As a result, an additional
variable added to the response surface analysis as described subsequently
was wastewater source.
The steady-state results on trapping efficiency have been classified in
terms of:
(1) Four ranges of NPS/dp (<2; 2 to 4; 4 to 6; >6)
(2) Three ranges of I.2I)
(3) Five types of influent wastewater sources
(a) AS (clarified activated sludge effluents)
(b) TF (clarified trickling filter effluents)
(c) UTF (unclarified trickling filter effluents)
(d) PE (primary effluents)
(e) OP (oxidation pond effluents)
37
-------
An overview of the data set can be obtained from a review of the data
summary In Tabj_e 5. The greatest diversity of information in terms of
ranges of NPS/dp and a.Qop, was obtained for AS and TF sources. The
least information was obtained for PE and OP sources. The foremost trends
that can be observed from examination of the data are as follows:
(1) Independent of efforts made in the field program to maximize the
range of observation, most of the steady-state points fall into one of
two ranges of solids loadings on the basis of waste type; 06 gm/sqm for TF streams.
(2) For any given range of solids loading (Ml), and for each waste-
water source, a dec rea s i ng_ trend in efficiency occurs with increasing
values of cr,QQ_p and NPS/dp ratio.
To explore the latter trend in more detail, the individual overall run
data have been plotted in Figures 9 and 10. The data presented in Figure
9 are_ identified in terms of efficiency, Ml, and CTmG-P' independent of
NPS/dp and waste type, arid the data presented in Figure 10 are identified by
efficiency, Ml, and NPS/dp independent of waste type or cr,QQ_p designations.
These plots were developed in the above formats to identify tne trends in
efficiency over the range of Ml associated with each of the PSD character-
istics.
Three trend lines (one for each range of O|OG-P^ nave been drawn in
Figure 9 as best-fit curves relating efficiency and solids loading.
The best-fit curves of Figure 9 indicate the following:
(1) For any given range of CLQG-P' overall suspended solids removal
efficiency decreases at a decreasing rate with increasing Ml; a
doubling of the solids loading from 2 to 4 gm/sq m resulted in 12 to
15 percent reduction in efficiency, and a doubling in solids loading
from 4 to 8 gm/sq m resulted i n a 35 to 40 percent reduction in
ef f Iciency .
(2) The more uniform the PSD (ie., the lower the value of en OR-P^ •
the more efficient was the overall efficiency of suspended solids
removal; for example, at a solids loading of 2 gm/sq m, the efficiency
decreased from 7 1 percent for O|_QQ_P I.2I.
Four trend lines JTave been drawn in Figure IO as best-fit curves, one
per range of NPS/dp, relating efficiency and solids loading. The following
trends are indicated by the best-fit curves:
(1) For any given range of NPS/dp, efficiency decreases at a de-
creasing rate with increasing Ml.
(2) Efficiency at any given Ml is a function of the NPS/dp ratio,
i.e., a direct function of the fabric pore size and an inverse func-
tion of the mean particle size.
The trend lines of the data presented in Figures 9 and 10 are confirming
38
-------
TABLE 5
SUMMARY OF OVERALL RUN SUSPENDED SOLIDS
REMOVAL OBSERVATIONS
Wastewater Source
Clarified Activated Sludge
Effluent
Trickling Filter Effluent
(Clarified)
Trick! ing Fi Iter Effluent
(Unclari f ied)
Primary Effluent
Oxidation Pond Effluent
NPS
HP
<2
<2
<2
2-4
2-4
4-6
6-10
<2
2-4
4-6
4-6
6-10
6-10
6-10
2-4
4-6
6-10
6-10
6-10
2-4
4-6
4-6
>IOO
>IOO
aLOG-P
l .21
1 .21
1 .21
>l .21
1.21
-
Average Ml
(gm/sq m)
2.4
2.4
3.3
3.1
2.4
4.9
4.5
14.0
1 1 .5
16.7
12.7
9.5
1 1 .6
34.9
5.6
2.4
3.4
6.8
2.5
1 .0
3.6
1 .4
2.9
8.2
1
Average MC/MI
(Overal 1 Run)
(%}
I3.29
65.2
56.4
59.5
22.2
54.0
66.7
34. 4a
40.3
31 .8
24.4
47.7
12. 8a
9.9
33.3
61 .2
43. 3a
68. 7a
28. Oa
51. Ia
21. 8a
37. 8a
5.la
8.7a
Note: Single observation
39
-------
RELATIONSHIP BETWEEN OVERALL SUSPENDED SOLIDS
REMOVAL EFFICIENCY AND SOLIDS LOADING VS (T,
LOG -P
>-
100
LiJ
O
S
LU
cr
9
_j
o
to
o
UJ
o
z:
LU
CL
80
60
40
20
NOTES :
CODE
o
crLOG_p
-------
RELATIONSHIP BETWEEN OVERALL SUSPENDED SOLIDS
REMOVAL EFFICIENCY AND SOLIDS LOADING VS NFS / 3
100
o
o
LU
O
o 80
LL)
UJ
cr
CO
o
o
co
UJ
Q_
CO
=5
CO
(T
UJ
60
4O
20
A
O
NOTES: CODE NPS/dp
o
A
X
D
<: 2
2-4
4-6
> 6
DATA INCLUDED FOR ALL WASTEWATER
TYPES AND ALL CTT
LOG-P
IMPS/dp>6
TO 34.90
12 16 20 24
SOLIDS LOADING, MI, g suspended solids /sq m
28
32
-------
evidence of the validity of the trends predicted with the simulation model
(Section IV) and of the trapping mechanism incorporated therin. Addition-
ally, it is apparent that the specification of wastewater source was
tantamount to specifying a range of Ml values in which the process would
be operative. It is presently unknown if yet another waste type would
result in the microscreen process operating over a yet-undefined range
of Ml values. Two factors that may be associated with specification of
wastewater source, and which could not be examined within the scope of
the study, are the drainabiiity and porosity of the water-solids complex
itself.
When it is considered that the process efficiency as described in Figures
9 and 10 is defined on a drum pool -> effluent basis, and that the drum
pool Itself acts as an influent solids concentrator, it is apparent that
there exists a combination of operating conditions and solids loadings
at which the concentration effect in the drum pool will preclude the
microscreening process from achieving any solids removal whatsoever on an
influent ->- effluent basis. That is, there exists a combination of operating
conditions and solids loadings at which the product of the drum pool con-
centration ratio (X§/X?) and the ratio of effluent: drum pool solids
concentrations (XjVXp) is equal to, or greater than, unity. The condition
for which the probuct of the ratio (X^/X^) and (X§/X^) is equal to unity
can be designated as a zero performance boundary for the process. This
boundary condition for the microscreen process can be defined with the
information developed in the field program. For example, if it is assumed
that cr|_QG-p for a microscreen influent is
-------
in the literature that microscreen suspended solids removal efficiency
remained constant (effluent solids concentration proportional to influent
solids concentration) or increased with increased solids loadings (eg.,
Reference 5). It is impossible to compare the results of the present
study with those of previous investigations, the foremost of which are
as follows:
(1) Prior researchers without exception have not normalized their
solids loading and removal data with respect to drum speed, S. That
is, the drum speed parameter was not assumed to have an impact on
the suspended solids removal performance of the microscreen.
(2) Neither the particle size distribution of the influent stream
nor the ratio of fabric NPS to a particle size parameter has been
identified and evaluated as a situation-specific variable affecting
microscreen performance.
(3) As noted in Section IV, very few microscreen investigations
have been documented with a sufficiency of information to permit
the calculation of liquid and solids balances and screen loading
rates.
(4) The microscreen process has not been viewed by previous researchers
as in fact a system comprised of three basic subprocesses as done here-
in.
Hydraulic Resistance
Hydraulic resistance characterized by the hydraulic parameter
(Xp H|_A/pS) as described previously, has been viewed from the perspec-
tives of the simulation as well as the physical model in the present study
(Section IV). It was observed in sensitivity tests with the simulation
model that changes in the defined porosity of the formed cake had a sig-
nificant impact on this parameter, the rate of change of hydraulic re-
sistance over the screening cycle decreasing with increasing solids
loading as the porosity increased. A variation of the magnitude of the
standard deviation of the PSD was found to have only a nominal effect on
the hydraulic parameter, the effect being one of decreasing the rate of
change of the hydraulic resistance over the screening cycle with increasing
solids loading as the standard deviation increased.
A major limitation of the pilot microscreen units utilized in the present
study was that headless could be measured only on an average basis for an
averaged flow through the submerged screen rather than at discrete points
across the screen, for flow at the discrete points. Thus, the calculation
of the hydraulic parameter from the field effort could be based only on
the total headless between drum pool and clear well, and the solids loading
parameter was based on the overall run Ml.
The hydraulic resistance data were analyzed in the same manner as were the
trapping efficiency data; all of the data were classified in terms of
NPS/dp, OLOG-P' and was+e source, and then examined at this level to
ascertain what relationships existed in each classification with respect
to solids loading. The entire data set is presented in Figure II, and a
43
-------
FIGURE
RELATIONSHIPS BETWEEN OVERALL RUN HYDRAULIC PARAMETER
AND SOLIDS LOADING- ALL WASTEWATER TYPES
1
.3.
115
ALL UTF & PE DATA
ALL AS & TF DATA-
CODE NPS/dp
o <2
+ <2
• <2
2-4
2-4
2-4
4-6
4-6
4-6
>6.0I
>6.0I
A
0
•
V
Q
WLOG-P
< 1.00
1.01-1.20
> 1.21
< 1.00
1.01-1.20
> 1.21
< 1.00
1.01-1.20
> 1.21
< 1.00
> 1.01
f - 1.05 q/tc (ASSUMED)
AS = CLARIFIED ACTIVATED SLUDGE EFFLUENT
TF = CLARIFIED TRICKLING FILTER EFFLUENT
UTF = UNCLARIFIED TRICKLING FILTER EFFLUENT
PE = PRIMARY EFFLUENT
_L
_L
4 6 12 16 2O 24
SOLIDS LOADING, MI, g suspended soiids/sq m fabric
_L
28
44
-------
consideration of the data set relative to the above classifications shows
that:
(1) A wide range of variation in both the hydraulic resistance para-
meter and solids loading was observed in the field;
(2) Two discrete best-fit curves can be defined, within the range
of sensitivity of the data, as follows:
(a) All AS and TF data independent of
(b) All UTF and PE data independent of MPS/d~p and O
(3) The general trend defined by the best-fit curves is one of the
hydraulic parameter increasing at an increasing rate with increasing
Ml.
Based on the relative locations of the two best-fit curves in Figure II,
it is apparent that the hydraulic resistance of the screen solids com-
plexes forming in the microscreening of UTF and PE streams is significantly
greater than that of the complexes derived from AS and TF streams. For
example, at an Ml level of 4 gm/sq m, a hydraulic resistance of 30 cm-sec
is indicated for the UTF/PE streams as compared with 3 to 5 cm-sec for
the AS/TF streams_. Because of the overlap of the two groups of streams
in terms of NPS/dp and O|_OG-P classifications it is likely that the
difference is associated with parameters other than the PSD characteris-
tics, particularly with factors associated with/manifested by the manner
in which the screen-solids complex is formed and passage of water through
the interstices of the complex occurs. Porosity is the only factor that
has been evaluated on a simulation basis, although it is recognized that
other factors may also be involved.
If it is assumed that porosity is a dominant factor in the relative
differences manifested by the data sets in Figure II, then it is possible
to estimate the relative difference in the porosity of the cakes formed
by the two groups of streams, using the relationship H|_~n as developed
from the Carmen-Kozeny relationship and described in Reference 4. On
the basis that the relative difference in hydraulic resistances are as-
sociated totally with a difference in the porosities of the formed cakes
at a given Ml, then the magnitude of the porosity of UTF/PE cakes is from
(3/TO)"' to (3i/6T -', or about one-half that of the AS/TF cakes.
In the calibration of the sub-program model to pilot scale data, it was
oeemed inadvisable to attempt to fit this variation in porosity, due to
the paucity of UTF/PE data for steady-state points. Accordingly, the
heuristicaIly-fit porosity in the subprogram model is probably somewhat
low for AS/TF effluents, and high for UTF/PE effluents.
TRANSFER AND BACKWASH SUBPROCESSES
Although the transfer and backwash subprocesses have been properly viewed
as discrete subprocesses in the update of the physical model presented
earlier in this chapter, each component subprocess could not be documented
separately as to its behavior in the field investigations due to the
physical configuration of the pilot plants. The major data analysis ef-
45
-------
fort was directed to examining the input/output and cleaning efficiency
characteristics of the backwash subprocess and to describing the response
surfaces relating the cleaning efficiency of the backwash subprocess with
the operating variables Pp and S.
Backwash System Characteristics
Initial tests were directed to ascertaining how applied washwater flow
(QB) was distributed as a function of the operating variables, the alter-
nate points of distribution being as throughput backwash flow captured
in the washwater collector (Qw), or as splash-back or splash-over not
recovered in the collector. The distribution of applied washwater was
examined independent of the separation subprocess by operating the drum
and backwash sprays with no influent entering the drum pool. This
approach was taken in recognition of the possibility that the absence of
particles in a screen-solids complex and/or particles jammed into the
pores of the fabric could result in not simulating the normal operating
behavior of the backwash system. However, the advantages in the approach
were that the behavior of the physical components in the backwash sub-
process could be examined independent of interferences from the separa-
tion and transfer subprocesses. A single fabric type (30y stainless steel)
was used in the evaluation.
The results of the experiment are presented in Figure 12, in which the
data are presented in the format of a relationship between recovery of
applied backwash flow as throughput washwater vs. backwash pressure at
varying levels of drum speed. The region of the response surface
relating these variables can readily be envisioned as having the follow-
ing characteristics:
(1) Recovery decreasing with increasing drum speed and increasing with
decreasing pressure.
(2) For a given level of drum speed, the recovery tends to saturate
at a maximal level with increasing pressure, the maximal level being
a function of drum speed. For example, maximum recovery was 52 per-
cent at S = 3.2 sq m/min, whereas maximum recovery was 48 percent at
S = 8.2 sq m/min.
(3) The backwash pressure required to achieve a saturation of re-
covery at a maximum level increases with increasing drum speed.
Based on the foregoing, the shape of a response surface can be speculated
upon over the entire range of variation of the operating variables S and
Pg, that shape being:
(1) Recovery decreasing with increasing S from a global maximum
recovery (undefined in the experiment) at S = 0 and maximum pressure.
(2) Recovery increasing at any speed with increasing pressure to a
saturation level that can be defined, relatively, as occurring at:
(a) PB >20 psig for S = 3.2 sq m/min
46
-------
COf RELATION CURVE-RECOVERY OF APPLIED BACKWASH
WATER AS THROUGHPUT BACKWASH WATER FOR NO INFLUENT FLOW
CL
I
100
cc
I
£ 80
i
o
a:
cc
UJ
>
o
o
UJ
cr
60
CD
Q
I] 40
CL
CL
LL
O
20
r
NOTE :
PB , psig
10
20
30
40
Applied wash water rate
QB, //min
7.5
10.7
13.0
14.9
Unit B - data for 30/z S3 screen
10
1
H
15 20 25
BACKWASH PRESSURE, P0 ,psig
30
35
40
ro
-------
(b) PD >35 psig for S = 6.6 sq m/min
ti
(c) PB >40 psig for S = 8.2 sq m/min
The above response surface shape demonstrates vividly several of the
hypotheses presented in Section IV, viz:
(1) That, as evidenced by the saturation of recovery: there exists
for any given combination of type and size of nozzle, backwash pres-
sure, and screen size a speed of rotation at which the synchronization
of the various components of the subprocess attains a maximal level.
(2) That the efficiency of the backwash subprocess is reduced with
increasing speed in part due to the increased tangential velocity
imparted to the impinging drops by the rotating drum, the result being
that the time available for the drops to traverse the screen and
fall into the collection trough is decreased.
It is readily evident from the foregoing that the variables S and PB are
Important factors to be considered in optimizing the utilization of the
applied backwash flow independent of considerations such as resultant
cleaning efficiency, type and size of nozzle, screen indentation, etc.
Fabric Acclimatization
A factor of concern in the experimental program was to ensure that the
microscreen fabrics had been appropriately acclimatized or pre-conditioned
with the suspension to be microscreened prior to starting a formal ex-
perimental run. In this context, fabric acclimatization was defined
conceptually as the accumulation of a stable level of residual solids
carryover on an initially virgin fabric as a result of preconditioning
of the fabric with the physical model. Inasmuch as acclimatization has
not been a documented concern in previous studies, a criterion was needed
to define the magnitude of pre-conditioning required to achieve the de-
sired stability, and to evaluate this factor as a variable in the exper-
imental program.
The approach used in developing an acclimatization criterion was to track
the rate of change of hydraulic resistance of a panel of backwash medium
removed from the microscreen drum at discrete time intervals over a day-
long test period. Only one of the two microscreens was operated, using
clarified activated sludge effluent as the feed stream, and holding the
operating variables S and PB constant insofar as possible throughout the
test period. The hydraulic resistance was measured as K, the slope of
the headless vs. superficial velocity curve (units of cm per cm/sec @
15°C), using the MTA methodology (Appendix B). The hydraulic resistance
of the backwashed panel of fabric/solids was then compared with the
hydraulic resistance (K°) of the virgin fabric to assess the rate and
relative impact of operation on the medium over time.
The results of the acclimatization test are presented in Table 6 and
illustrated in Figure 13. The rate of change of K with respect to time,
cumulative liquid loading, and cumulative solids loading is apparent
from a consideration of the best-fit curves presented in Figure 13. As
48
-------
TABLE 6
RUN 0 - FABRIC ACCLIMATIZATION EXPERIMENT
SUMMARY OF OBSERVATIONS
Run Time
(hr)
0
1.50
3.02
3.75
4.90
5.90
6.75
K
(AHL/AV)
(cm/cm/sec)
6 T=I5°C
1.35
6.45
7.40
11.50
9.05
9.50
6.25
Cumul ati ve
Liqu id
Flow-Through
(1 ,000 SL/sq m)
0
18.8
43.7
53.7
69.8
78.6
87.4
Cumu 1 at ive
Sol ids
Load i ng
(1 ,000 g/sq m)
0
1.47
2.75
3.23
4.02
5.08
6.05
Specific
Backwash
Rate
U/sq m)
0
I.I
1.2
1 .4
1.8
1.6
1.5
Average H.
during
Subrun
(cm)
0
12
17
16
18
16
18
Notes: (I) Screen: 21y Stainless Steel 200x630 mesh
(2) Run date: 16 and 17 February 1971
(3) Waste source: San Leandro clarified activated sludge
process effluent
(4) Screen area: "0.8 sq m
49
-------
O
0)
6
E
RELATIONSHIPS BETWEEN BACKWASH SCREEN HYDRAULIC CHARACTER AND
RUN TIME AND SCREEN LOADING PARAMETERS, FABRIC ACCLIMATIZATION-RUN 0
12
A
D
O
10
K vs CUMULATIVE SOLIDS LOADING
O
*£.
UJ
n 6
O
UJ
(n
en
3
Q
UJ
I
u.
O
UJ
0.
O
en
A
D
— O— —
K vs CUMULATIVE LIQUID LOADING
CODE BASIS
D Run time ,hr
O Cumulative liquid loading (10,000 liters/sq m)
A Cumulative solids loading (lOOOg/sq m)
O
VIRGIN FABRIC K
O
RUN TIME, t\r- CUMULATIVE LIQUID LOADING, (IO-£/sq m ); CUMULATIVE SOLIDS LOADINGS I03g/sq m)
-------
a general trend, K tended to saturate in a pattern of harmonic variation
about a mean value of K of 10 cm/cm/sec, a value which, for the selected
experimental/operationaI conditions, was over seven times greater than
the hydraulic resistance (K°) of the virgin fabric. The parameter values
at which the saturation level of K was approached was within: eight hours
operating time; 60,000 l/sq m of liquid loading; and 6,000 gm/sq m of
sol ids loading.
Based on the above results, all subsequent test runs in the experimental
program were done after a prior overnight acclimatization of the fabric
in the wastewater to be test-microscreened.
Response Surface for Backwash Subprocess
With the preliminary definition of the behavior of the backwash subprocess
obtained from the response surface for backwash flow recovery, a search
was made with the available data to map out the relationships between
cleaning efficiency, fabric pore size, drum speed, and backwash pressure.
In the case of each experimental or operating variable, the mapping
effort was approached in a single dimension using data selected from runs
where the parameters in the other dimensions were held constant. In
this manner (assuming superposition) the multidimensional response sur-
face could be sketched out across each dimension, and from this, con-
structed in all dimensions within the totality permitted by the data.
Cleaning efficiency was evaluated as the degree of recovery of the hy-
draulic capacity virgin fabric upon backwashing of the medium, and was
defined in terms of the ratio K°/K where K° and K are as defined above.
For a ratio value of 1.0, it was assumed that a cleaning efficiency of
100 percent was obtained.
Effect of Fabric Nominal Pore Size
The hypothesis presented in Section IV for the effect of fabric pore on
cleaning efficiency was that, for any given type and size of nozzle and
water pressure, there should exist a threshold pore size below which clean-
ing effectiveness declines rapidly and above which cleaning efficiency
is independent of pore size. In order to examine the validity of this
hypothesis, data were selected for all runs where PB was 15 psig (the
backwash pressure most commonly used) and speed was held within the range
of 1.3 to 2.4 sq m/min (the latter being the range of speeds in which
optimal cleaning efficiency was observed, as discussed subsequently).
Excluded from this data set were results obtained when nylon fabric
(NPS of 10y) was used because of the deformation characteristics of this
fabric.
The resultant data set is presented in Figure 14, and the relationship
between cleaning efficiency and fabric NPS is illustrated by the best-
fit curve. As a general trend, the least effectiveness was obtained with
fabrics having the smallest NPS, and the cleaning efficiency tended to
increase toward a maximal level of 0.90 to 0.95 at an NPS value of 16.5y,
and then to decrease slightly with increasing values of NPS. It is
tentatively concluded from these results that the cleaning efficiency
relationship exhibits a saturation with respect to NPS at a level of NPS
equal to or exceeding about I6y. On the basis of these observations,
51
-------
RESPONSE SURFACE RELATIONSHIP FOR BACKWASH EFFICIENCY AND FABRIC NOMINAL PORE SIZE
o
in
0.8
T>
o
o
^
^
o
»*
O
yj
o
u.
u_
UJ
X
I
o
CO
0.6
0.4
0.2
10
12
14
16 18 „ 20
FABRIC NOMINAL PORE SIZE,
NOTES:
o 20/j. POLYESTER -
a 23/z SS
* 25p POLYETHYL
v 15-18 // SS
x I8-22// SS
0 12-15 /i SS
PB=15 psig
S = 1.3 to 2.4 sq m/min
22
24
26
-------
only data obtained in runs with fabrics having an NPS greater the 16y
were used in the subsequent unldimensional response surface analysis.
Effect of Drum Speed
The hypothesis developed previously for the effect of drum speed on the
cleaning efficiency was that, for any given combination of type and size
of nozzfe, water pressure, and fabric NPS, there should be a speed of
rotation which optimizes the synchronization of the intersection of pores
and droplets, above or below which cleaning efficiency should decline
rapidly. Because it was observed during the field program that some of
the fabrics, particularly nylon, flexed much more than the stainless
steel fabrics, two discrete response surface relationships were mapped,
one for runs with stainless steel fabrics only, and the second only for
runs with nylon.
The cleaning efficiency vs. drum speed data for runs with stainless steel
fabrics and a constant operating pressure of 15 psig are shown in Figure
15. Data in this set were available for a range of speeds varying from
0.5 to 2.4 sq m/min. Based on the trend of the best-fit curve, the data
indicate that the cleaning efficiency increased at a decreasing rate
from a level of 0.4 at S of 0.5 sq m/min to a maximal value of about 0.95
at S of 2.4 sq m/min. The data generally indicate that the cleaning
efficiency was in excess of 0.85 over the range of speeds from 1.3 to
2.4 sq m/min. Given the range of sensitivity of the data, it has been
assumed from the results that 0.85, or 85 percent restoration of the
hydraulic capacity of the virgin fabric represents an optimal level of
cleaning efficiency for stainless steel fabrics as measured by the MTA
technique. Additionally, in terms of the previously discussed response
surface relating drum speed,backwash pressure, and recovery of applied
as throughput backwash flow, the drum speed range and operating pressure
at which optimal cleaning efficiency was observed with the stainless steel
fabrics represent conditions at which a saturation level of recovery is
expected to occur.
The available data on cleaning efficiency vs. drum speed for runs with
the nylon fabrics are presented in Figure 16. The range of operating
pressures for the data set is 20 to 25 psig and the range of drum speeds
is from 1 to 4 sq m/min. Because of discontinuities in the data set in
the range of speeds from 1 to 2 sq m/min, it has been necessary to specu-
late on the best-fit curve of the data defining the cleaning efficiency
vs. drum speed in this speed range. A pronounced peak can be observed
'n the relationship at a speed of 2.2 sq m/min, for which the cleaning
efficiency was 2.05, or 205 percent; i.e., at this point the hydraulic
resistance of the backwashed nylon fabric was about 49 percent of that
of the same fabric in a virgin state.
From observations of the behavior of nylon fabrics both in the physical
model and when assayed in the MTA apparatus, it was found that the nylon
fabric deformed into a catenary shape as flow passed through it, parti-
cularly as the solids-laden fabric passed through the zone of influence
of the backwash spray. The net effect of the flexure, as documented in
Reference 4, was to decrease the hydraulic resistance of the fabric.
What is unknown about the nylon fabric is whether or not its deforming
53
-------
RESPONSE SURFACE RELATIONSHIP FOR BACKWASH EFFICIENCY AND DRUM SPEED
STAINLESS STEEL FABRICS
1 0
o
o
in
o
O
0
0.8
0.6 —
O
UJ
y
U.
u.
UJ
I
d
o
CD
04 —
02
1
o
MOTES:
I5-I8// SS
18-22/i SS
21/z SS
23^ SS
PD = 15 psig
0
DRUM SPEED, S, sq m/min
-------
o
o
in
Of
o
0»
o
o
^
O
-------
characteristics changed with time of usage, exposure in wastewaters, etc.,
so that hydraulic resistance of the fabric decreased with use. Otherwise
stated, did or did not the nylon fabric stretch irreversibly with use,
the net result being that its resistance was reduced relative to that of
the fabric in its virgin state? If such were the case, then K° would
be greater than K, as was observed. Because of the improbability that
the hydraulic resistance of a backwashed fabric subjected to considerable
use would be less than that of a virgin fabric unless some form of ir-
reversible flexure/stretching/aging did occur, it has been assumed from
the results with the nylon fabric that this fabric did deteriorate with
usage. Additionally, on the'basis of this assumption, physical model
runs made with nylon fabrics were not used in subprogram model calibration,
Given the above qualifications, it is apparent that the response surface
relationship for cleaning efficiency vs. drum speed with the nylon
fabrics defines an optimal range of drum speeds (2 to 2.5 sq m/min)in
terms of backwash efficiency. The shape of the response surface is as
was predicted in the original hypothesis, and the range of speeds (2 to
2.5 sq m/min) and backwash pressures (20 to 25 psig) are within the
region at which the recovery of applied as throughput backwash flow is
expected to saturate (Figure 12).
Effect of Pressure
The hypothesis for the effect of pressure on the cleaning efficiency was
that, given any combination of nozzle type and size, speed of rotation,
and NPS, there should be a backwash pressure at which cleaning efficiency
is optimal (Section IV). This hypothesis can be modified in view of
the observations presented above that recovery of applied as throughput
backwash flow at any given speed saturates at a maximal level with respect
to pressure, the modified hypothesis being that for the given conditions
it is also expected that cleaning efficiency will saturate with respect
to pressure.
The data set used for examining the effect of backwash pressure was selec-
ted on the basis that only data from runs with stainless steel fabrics
and drum speeds in the range of 1.3 to 2.4 sq m/min should be used for
the evaluation. These constraints on developing an appropriate data set
were accepted as necessary in view of the relationships/observations on
the backwash subprocess presented above. The resulting, limited, data
set is presented in Figure 17, and on the basis of the best-fit curve
of the data, there is no apparent trend defined other than a slight in-
crease in cleaning efficiency with increasing backwash pressure over the
pressure range of 15 to 35 psig. That is, within the range of backwash
pressures in the available data set, cleaning efficiency was found to be
essentially independent of backwash pressure. Relative to the modified
hypothesis for the response surface, it can be reasoned that the cleaning
efficiency saturated with respect to backwash pressure at a Pg at or less
than 15 psig for the range of drum speeds from 1.3 to 2.4 sq m/min.
Generalized Response Surface
The following statements can be made about a generalized response surface
for the backwash subprocess on the basis of the above findings and quali-
ficat ions:
56
-------
RESPONSE SURFACE Rl .ATIONSHIP FOR BACKWASH EFFICIENCY AND BACKWASH PRESSURE
o
o
in
0)
o
O)
o
o
*:
^.
o
i
CD
1.0 —
0.8
0.6
O
z
UJ
I 0.4
u.
UJ
0.2
10
20
BACKWASH PRESSURE , PB , psig
NOTES:
SS
v 15-18 // SS
0 18-22/i SS
S = 1.3 to 2.4sqm/min —
O
30
40
-------
(1) With other than the nylon fabric of 10y NPS, cleaning efficiency
was found to increase with increasing NPS to a maximal value of 0.90
to 0.95 at an NPS of 16.5 y, and then to decrease slightly with i n-
creasi ng NPS.
(2) With stainless steel fabrics, cleaning effeciency was found to
increase with increasing speed to a maximal level at or in excess of
0.85 over a range of speeds from 1.3 to 2.4 sq m/min. The lack of
data precluded examining the response surface for speeds in excess
of 2.4 sq m/min.
(3) With other than the 10y nylon fabric, cleaning efficiency was
found to be independent of backwash pressures in the range of 15 to
35 psig and for a speed range of 1.3 to 2.4 sq m/min.
Relative to the third statement above, the implications of the response
surface for recovery of applied as throughput backwash flow are that a
saturation with respect to recovery is expected to occur for the pres-
sure range (15 to 35 psig) and speed range (1.3 to 2.4 sq m/min) in
which the response surface relating cleaning efficiency and backwash
pressure was mapped. If it is assumed that a direct relationship exists
between cleaning efficiency and recovery in terms of the variables S and
PB, then it is anticipated that the cleaning efficiency response surface
will saturate at a maximal efficiency at increasingly greater backwash
pressures as drum speed is increased.
It can be concluded from the foregoing analysis of the backwash subprocess
that the efficiency of the subprocess can be described in terms of a
multi-dimensional response surface which could be defined in terms of
two variables, Pg and S, in the present study. It is apparent from these
findings that the region of operation of the backwash subprocess must be
selected in consideration of at least these two variables if the perfor-
mance of the subprocess is to approcah an optimal level.
Yield
Based on the constraints of the physical system used in the investigation,
the yield of raw influent wastewater as effluent product water was com-
puted as:
Y = " (12)
QE
where: Qr- = Effluent flow rate, l/min
Qw = Throughput washwater flow rate, l/min
An empirical correlation curve was developed between Y and QE (specific
effluent flow rate, l/min - sq m) and is illustrated in Figure 18. The
trend of the empirical correlation indicates that the yield increases at
a decreasing rate with increasing Q^ from a minimum of 87 percent at Qp
of 90 l/min - sq m to over 98 percent at QF of 700 l/min - sq m, and to
a saturation value of about 99 percent. About 80 percent of the obser-
vations (data points) used in developing the correlation were at yields
58
-------
RELATIONSHIP BET\ EEN YIELD (EFFLUENT BASIS) AND SPECIFIC EFFLUENT FLOW RATE
100
Y = 100- I/ (0.00055 Ql )
100 200 300 400 5OO 60O 7OO
SPECIFIC EFFLUENT FLOW RATE, Q£/SUBMERGED SCREEN AREA, liters/minute/sq m
80O
-------
equal to or greater than 97 percent. The empirical correlation curve of
the yield data can be described by the equation:
Y= 1°°- '/(0.00055QE1'1) (13)
Equation 13 was incorporated into the subprogram model as described in
Appendix D.
SUMMARY-RESPONSE SURFACES FOR MICROSCREEN PROCESS AS DETERMINED IN THE
FIELD STUDY
The response surface relationships describing the field program results
were developed in consideration of the implications of the response sur-
faces developed with the simulation model and an ongoing evaluation of the
realities imposed on the study by the physical model (i.e., the pilot
plant system). The basic data requirements for development and evolution
of the mathematical model required that process behavior be viewed on
a steady-state basis. To this end criteria were established for defining
quasi -steady-state data. With these criteria and the col lected data, over
50 quasi -steady-state data points were defined which were used to analyze
process behavior and which, in the analysis, have been treated as repre-
sentative of microscreen process behavior in the quasi -steady-state mode.
Steady-state data points were obtained for 21 runs with clarified acti-
vated sludge effluents, 18 runs with clarified trickling filter effluents,
7 runs with unclarified trickling filter effluents, three runs with primary
effluents, and two runs with oxidation pond effluents. Each data point was
classified as to influent and drum pool solids concentration and PSD char-
acteristics, and operating variables. The data points were then analyzed
within the contexts established for evaluation of the implication of the
simulation model and physica l/conceptua I models of the process. Individual
data points were selected from the set of steady-state data points to
formulate (map out) the response surface relationships needed for complet-
ing model development and implementing model calibration.
The ramifications of these relationships were discussed in Section VI and
are summarized below.
Separation Subprocess
The role of the drum pool as a solids conceetrator (reservoir) in the
microscreen process was defined quantitatively by relating observations
on the influent stream solids concentration and PSD characteristics
(d and 0|_OG^ w'~l"h the corresponding parameters for the drum pool suspen-
sion. Within the range of sensitivity of the data it was found that:
(I) The ratio X§/X? (drum pool suspended solids concentration: influ-
ent suspended solids concentration) varied independently of backwash
pressure variation, but that it increased at an increasing rate in
the range 0
-------
(2) The ratio d /d| mean drum pool particle size: mean influent
particle size, in a log-normal PSD) was independent of speed in the
range 0• effluent basis rather than on an influent -»• ef-
fluent basis in order to view the behavior of mechanisms operative in the
separation subprocess across the screen. The field measurements of trap-
ping efficiency were developed on a drum pool •* effluent basis. Response
surface relationships were mapped from the data set relating trapping
efficiency (suspended solids removal efficiency between the drum pool and
the effluent) and the PSD characteristics of the suspensions at any given
level of solids loading (Ml). The shape of these surfaces can be described
as:
(1) Trapping efficiency decreasirvg from maximal levels for a, 0/>_p
increasing from I.00 and for NPS/dp (ratio of fabric nominal pore
size: mean drum pool particle diameter) increasing from 2 (this
portion of the response surface correlating with the predictions of
the simulation model).
(2) Trapping efficiency varying inconsistently in the regj_on of
the response surface where a.np D was less than I and NPS/dp less
i i r-* LUo—r '
than 2.
(3) Trapping efficiency decreased at a decreasing _r_ate with in-
creasing solids loading for any given level of NPS/dp and CT|_OG-P'
The design is that the relationships can be used, in conjunction
with the relationship for X^/X-j* defining the degree of concentration
of influent solids in the drum pool, to identify regions of the trap-
ping efficiency relationship in which a microscreen will approach a
zero performance boundary in any application. That is, it is possible
using the above relationships for trapping efficiency to define the
combinations of operating variables and solids loading in which the
microscreen process has zero performance capability in terms of sus-
pended solids removal on an influent -> effluent basis for a given
application, given a definition of the PSD characteristics of the
solids to be microscreen.
The hydraulic resistance of the solids-screen complex could be measured
only on an average ( overall-run) basis on the entire screening cycle in
the field program. It was not possible, with the information base develop-
ed in the study, to characterize the headloss and flow rate profiles at
any point across the drum. The data analysis indicated that four discrete
response surface relationships could be defined relating the overall-run
hydraulic resistance, solids loading, and PSD characteristics, the general
trend of the curves being that overall-run hydraulic resistance .Increased
at an increasing rate with increasing Ml, and as a function primarily of
the wastewater source of the suspension being screened and secondarily of
61
-------
The PSD characteristics of the suspension. At a solids loading (Ml) of
4 gm/sq m, it was found that the hydraulic resistance of screen-solids
complexes formed from microscreening of primary effluents and unclarified
trickling filter effluents was six to ten-fold greater than the hydraulic
resistance of clarified activated sludge and trickling filter effluents,
the difference being associated with and manifested by the manner in
which the screen-solids complex is formed and passage of water through
the interstices occurs.
Transfer and Backwash Subprocesses
The transfer and backwash subprocesses could not be examined discretely
in the present study because of constraints imposed by the physical model.
Consequently, the major effort was devoted to examining the input/output
and cleaning efficiency of the backwash subprocess.
As a result of the data analysis, a generalized response surface was
formulated for the backwash subprocess relating cleaning efficiency and
operating and experimental variables. Cleaning efficiency was defined as
K°/K, or the ratio of the hydraulic resistance of the virgin fabric (head-
loss/unit velocity) to that for a panel of backwashed medium removed from
the drum; it was assumed that a cleaning efficiency of 100 percent was
obtained for a K°/K ratio of 1.0. The generalized response surface can be
described as follows:
(1) Cleaning efficiency was found to increase with increasing fabric
NPS to a maximal level of 0.90 to 0.95 at an NPS of 16.5, and then
to decrease slightly with increasing NPS.
(2) Cleaning efficiency was found to Increase with increasing speed
to a maximal level at or in excess of 0.85 over a range of speeds
from 1.3 to 2.4 sq m/min. The lack of data precluded examining the
response surface for speeds in excess of 2.4 sq m/min.
(3) Cleaning efficiency was found to be independent of backwash
pressure in the range of 15 to 35 psig and for a speed range of
1.3 to 2.4 sq m/min.
(4) The implication of information acquired on the recovery of
applied as throughput backwash flow is that the cleaning efficiency
response surface will saturate at a maximal efficiency at increasingly
greater backwash pressures as drum speed is increased.
It was concluded from the above findings on the backwash subprocess that
the region of operation of the backwash system of a microscreen unit must
be selected in consideration of backwash pressure and speed of drum rotation
if the performance of the system is to approach an optimal level.
Yield
An empirical correlation was developed between the yield (defined as
ratio of effluent product water to influent raw water) and the specific
effluent flow rate (qE, l/min - sq cm). The trend of the correlation
was that yield was found to increase at a decreasing rate with increasing
62
-------
specific effluent flow rate, from a minimum observed yield of 87 percent
at q£ of 700 l/min - sq cm. About 80 percent of the observations used in
developing the correlation curve were at yields equal to or exceeding 97
percent.
Additionally, it was concluded from observation of the physical model
that the transfer and backwash subprocesses are the weakest links in the
microscreen process. Because microscreen units are designed traditionally
as package systems rather than on a situtation-specific basis, both of
the above conclusions relate to how microscreen units should be designed,
an area of concern presently being dealt with only by the limited number
of firms presently manufacturing microscreening equipment.
63
-------
SECTION VII
DEVELOPMENT OF SUBPROGRAM MODEL
TRADE OFFS AND ASSUMPTIONS OF THE SUBPROGRAM MODEL
Theoretical Structure
A number of decisions were made in extracting a final model version from
the complex of constraints and realities. The foremost among these deci-
sions was that of maintaining a theoretical structure as the basis for
the model, rather than going to a wholly empirical formulation. The
field investigative program results have shown the microscreen to be a
process which is variable in nature and very situation-specific; the
development of an empirical model formulation from results with these
indications would yield no more than an additional small territory on
the overall map of process performance. Rather, the maintenance of a
theoretical framework and the interpretation of field program results
in the light of the theoretical framework are consistent with the inves-
tigative context of developing broad delineations of the process maps
and providing methodology for transferabiIity of process-related infor-
mation from one specific context to another.
Steady-State
A further decision in developing the subprogram model was related to the
mode of achieving a steady-state model. The "best" model formulation as
presently conceived would be time-variant, expressing the feedback rela-
tionships in the process; steady-state behavior would be determined by
time-averaging. The achievement of such a model formulation, however,
was well beyond the scope and intent of the present study. Accordingly,
steady-state behavior was modeled by aggregating the transient character
of the process to quasi-steady-state behavior at the field program/data
base level, rather than by postponing this aggregation and modeling the
transient behavior.
Drum Pool Solids Concentration Effect
The phenomenon of drum pool concentration of solids is a composite result
of many causes. The resultant composition and concentration of solids
in the drum pool is expected to be a result of the mechanisms operative
in the turbulent shear of particles, the fall-back of cake, and the
splash-over of applied backwash water. A detailed model of such a pro-
cess, with the purpose of determining solids character and concentration
in the drum pool as a function of external gross parameters, would require
the time-variant, feedback-type model noted above. The approach taken to
quantitatively define the phenomenon of solids concentration in the
drum pool was to correlate, empirically, the drum pool solids concentra-
tion and PSD characteristics with the corresponging characteristics for
the influent suspension and with the drum rotational speed. The ratio-
nale associated with this effort was that drum rotational speed was
assumed to be a measure of the turbulence, and resultant particle shear
occurring in the drum pool, and a measure of the rate of cake erosion
and splash-over in the transfer and backwash subprocess. The model, as
formulated, thus requires that an influent mean particle size (d ) and
CTLOG-I be included in DMATX. '
65
-------
The use of an errpirical correlation between influent and drum pool solids
concentration and characteristics and the drum speed variable integrates,
in one step, a number of mechanisms and sub-processes which are operative
in the overall microscreen process; in particular, the effects of particle
shear and flocculation are thus assumed to have balanced out to a quasi-
steady state between the influent character, turbulent shear, splash-
over, and fall-back. Having eliminated all these effects in terms of the
model, it became possible to maintain the mechanism of the solids separation
subprocess as used in the operational model (viz., geometric trapping).
Thus, the basic concept of filtration across the drum pool from the
theoretic/simulation formulation was retained in the subprogram model
formulation, but the identity of influent and drum pool character that
was present in the operational model could no longer be held in the sub-
program mode I.
Microscreen performance was found to be sensitive to the parameters d
and °LOG, both as predicted in the model and as observed in the field.
Given that a good predictive model will be only as good as the input
values for these parameters, and that the range of variation of the PSD
characteristics for the types of effluent was found to be significant,
it is apparent that gross correlations of the PSD characteristics as a
function of type of influent source represents a tenuous, if not totally
unsatisfactory, approach. Accordingly, no such correlation is included
in the subprogram.
Backwash Subprocess
The backwash function is more stochastic than deterministic and appears
to be highly situation-specific. Backwash throughput and cleaning effi-
ciency are affected by a number of variables which are inherently un-
definable, e.g., the manner of installing the microscreen fabric, which
will affect the flexing and stretching behavior of the fabric. Further,
the magnitudes of splash-back and splash-over, which are prime deter-
minants in the recovery of applied as throughput backwash flow, will be
dependent on the specific design features of the components of the back-
wash system, etc. Accordingly, a simple, empirical formulation was
selected which, though inadequate to model the backwash process at a
high level of sophistication, provides the needed input to the overall
model at a level of sophistication commensurate with its intended utili-
zation. This choice was made with the recognition that the backwash
model used in the subprogram is not to be considered as a definitive
statement of the behavior of the backwash process.
Costs of Microscreen Installations and Operation
Ideally, a process performance-cost model should be capable of generating
response surfaces relating process performance levels and the least costs
necessary to achieve the levels of performance; this type of response
surface implies an optimal design procedure has been pursued so that the
least cost performance is in fact being obtained. Design flexibility for
most sanitary engineering processes, however, is limited by tradition and
equipment availability, so that, in general, cost-optimizing choices are
possible only within a narrow range, if at all. For the case of the
microscreen process, the problem is further compounded by the fact that
both design and equipment manufacture rest in the same hands; there is no
66
-------
generally accepted process of microscreen design and cost-estimating,
thus the design decision, as noted previously, is generally of a binary
(yes or no) type in regards to utilization of a given microscreen at a
given cost, all information on design and cost being supplied by the
manufacturer. Further, the number of manufacturers of microscreening
equipment is presently less than ten, additionally limiting the region
of choice and, hence, the potential for optimization of the design of a
microscreen installation.
Microscreen units, as supplied, are available only in a discrete number
of sizes from the manufacturers. The units built by the different manu-
facturers are not mutually overlapping as to size, screen-mounting
equipment, backwash systems, nominal submerged screen area, and in the
manner in which the hydraulic capacity of the units are designated. Thus,
the costing problem is further exacerbated by the availability of only
discrete cost information for defined types and sizes of microscreen
units where, properly, (for modeling purposes) it should be possible to
define a continuous cost function (itself optimized for any given micro-
screen size). Inasmuch as the microscreen manufacturers hold as proprie-
tary the information used for cost-estimating of microscreens, it is not
possible to develop such a cost curve independently, except by specu-
lating as to the probable nature of costs incurred. In light of the
above, the only course available in the present investigation was to
obtain empirical data from manufacturers and generate continuous cost
curves as a function of gross parameters. It is recognized that this
approach is far from optimizing.
One particular problem which is apparently unique to the microscreen
process is that there is a high capital cost associated with material
whose expected life is significantly less than that of most other pro-
cess facilities; i.e., the screening fabric. There is at present no field
experience in tertiary treatment applications of sufficiently long dura-
tion to give adequate estimates of actual screen life; available informa-
tion from the manufacturers indicated that a useful life of nine to ten
years could be obtained, but these estimates were given without substan-
tiation of actual field observations. The nature of the fabric is such
that it is quite susceptible to being rendered useless through mechanical
accident, while the chemical characteristics of secondary effluents in
some instances will cause continual degradation of the fabric. It was
necessary to include cost of fabric replacement as an amortized capital
cost; the capital cost of fabric is typically upwards of 50 percent of
microscreen capital cost, hence the amortized cost of fabric is a signi-
ficant part of the total cost of a microscreen, and is sensitive to the
assumed screen life.
SUBPROGRAM MODEL ELEMENTS
The above considerations represent the basic trade-offs and assumptions
used in developing the various model elements and the over-all structure
of the subprogram model. The program is comprised of five basic elements:
concentration, trapping, hydraulics, backwash, and cost, each of which
is described below.
67
-------
Concentration
The concentration element is designed to yield drum pool suspended solids
concentration and PSD characteristics as a function of influent solids
concentration, PSD characteristics, and drum rotational speed. The ratio
of drum pool solids concentration to influent solids concentration was
derived empirically as a function of drum rotational speed, as were the
ratios of mean particle size and standard deviation.
The distribution of particles within the drum pool is assumed to be log-
normal by number of particles, and unimodal, and is generated by the
computer subprogram as the number of particles in each size class, utili-
zing the drum pool PSD characteristics and solids concentration for each
influent. In terms of the stream vector, it is assumed that the suspended
solids parameter and all other parameters associated with suspended
solids are concentrated similarly, whereas the concentration of dissolved
components remain unchanged as the liquid stream is transferred both into
the drum pool and subsequently through the medium into the clear well.
Trapp i ng
The trapping mechanism of the subprogram model is geometric, and is
governed by a continuously diminishing "effective trapping diameter".
The particles retained upon the cake are those particles whose diameters
are greater than the effective trapping diameter. The effective trapping
diameter is initially some fraction of the fabric nominal pore size; as
the cake is built, the trapping diameter is taken as one-fifth of the 15th
percent!le diameter of particles in the retained cake layer. This number
is assumed to be representative of the pore size,of the cake. Mathemati-
cally, a variable number of iterations (layers) of cake are used to
describe the trapping process; the number of layers utilized depends upon
the quantity of particles impinging upon the cake, and is reduced as this
quantity increases, so that at each iteration, a layer not more than
50 d^tj in thickness is built, in order to insure an accurate integration.
Continuity between retained, approaching, and passing particles in each
size class is used to determine the number of particles in each size
class that pass through the cake and screen to the effluent. At each
step, the velocity of approach of particles to the cake is determined by
the hydraulic mechanism, described below.
HydrauI ics
For both trapping and hydraulic consideration, the submerged arc of the
drum is treated as being composed of eight radial segments; all calcula-
tions are on a unit width basis, and constant head over each segment of
the drum arc is assumed. Thus, the throughput will be largest for the
entering segment, and lowest at the emerging segment. Inasmuch as the
subprogram determines the requisite width of drum as a function of head,
rotational speed, and influent flow calculations of throughput are
based on an arc of unit width; the total throughput for this arc is then
divided into the total required flow capacity: to determine the required
number of such arcs of unit width. The throughput for each segment is
calculated, and is integrated around the drum to obtain the unit-arc
throughput.
68
-------
The methodology for calculating throughput is based on the calculation
of the hydraulic resistance coefficient for the combination of cake and
screen. Laboratory experiments early in the project indicated that
the assumption of laminar flow through the cake and screen was not wholly
justified, hence a quadratic expression of pressure as a function of
velocity was utilized. The coefficients are defined by the structure of
the cake, and are calculated utilizing a form, of the Carmen-Kozeny Rela-
tionship for calculating head-loss through mixed beds of particles. As
each incremental layer of cake is built in the trapping iteration, the
coefficients of the quadratic expression are calculated as functions of
the then-current cake particle composition, and the quadratic equation
is solved for velocity as a function of the externally defined head on
the drum. Thus, both the throughput and the rate at which particles
approach the cake are determined for each iteration and for each segment.
A running total of the incremental throughput is maintained, and in
this manner the total throughput is determined as the sum of the through-
put through each segment.
Backwash
A simple, empirical backwash model was selected for use in the subprogram.
The parameters which must be determined by the backwash model are the
flow rate in the backwash waste stream, and the condition of the fabric
at entrance to the drum pool (i.e., at submergence). The concentration
of solids in the backwash waste stream is obtained by assuming that all
residual solids on the fabric (retained solids less fall-off) are "cap-
tured" by the throughput backwash flow (that portion of the applied back-
wash flow that is transferred to the collection trough). The screen
condition at entrance is described by two parameters: The effective
trapping diameter, and the hydraulic resistance. The phenomenon of fab-
ric acclimatization has been described previously; it is assumed that
any reduction in initial effective pore diameter is due to this acclima-
tization process, and is independent of backwash. Thus, backwash is
assumed to affect only the hydraulic resistance of the screen. Based on
an analysis of the backwash process, a typical ratio between clean screen
and post-backwashed hydraulic resistance coefficient is 0.85, and this
coefficient is taken as constant for all conditions of backwash. An
empirical correlation of yield (Y) with specific throughput rate (Q-)
was developed and has been utilized in the backwash model to determine
the quantity of backwash flow by back-figuring from the yield, rather
than by calculating yield as a function of throughput and backwash.
Five components of cost were utilized in the subprogram. Capital cost
is calculated from manufacturer's data, organized on a cost/submerged
screen area basis inasmuch as the required submerged area to pass the
flow rate is calculated within the subprogram, and capital cost is based
upon this parameter assuming a continuous cost function. Operating,
replacement, and maintenance costs are assigned to four areas; fixed
cost of operation, variable cost of operation, energy costs, and cost
of amortizing fabric replacement. The fixed cost of operation is assumed
to be one hour/day, independent of size, on the basis of information
69
-------
provided by manufacturers; a wage cost (CWAGE) of $3.64/hr (Reference 6,
1971) was assumed. The variable cost of operation was assumed to be a
function of flow rate, using the relationship as follows:
COSTO(N,2) = (38.*(SMATX(2,ISI)**.I9))-35.*(CCOST(N,I)*(10~3)
Energy cost in kwh were obtained from information provided by manufac-
turers, and were correlated with submerged area, assuming an average rota-
tional speed of 5 rev/min. This cost was adjusted proportionally for
microscreens operating at other rotational speeds. The screen replace-
ment costs were derived assuming an average fabric cost of $60/sq ft,
with an optimistic fabric life of nine years at an interest rate of
4.5 percent.
The relationship between cost/unit area and submerged area is illustrated
by the data in Figure 19 and Table 7. The curve-fit equation for the
model was taken as:
CCOST(N,1) = EFFA*300, dollars/yr (for EFFA greater than 150) (2)
CCOST(N,1) = (3860./SQRT(EFFA)*EFFA, dollars/yr (for EFFA less
than or equal to 150) (3)
where: EFFA is the submerged screen area in square feet.
The fixed cost of operation is expressed as:
COSTO (N,1) = 365.*CWAGE, dollars/yr (4)
A relationship for energy costs was obtained from an analysis of data
presented in Figure 20, and is expressed as follows:
COSTO(N,3) - 365.*CKWH*(EFFA*.4+12.5)*AFAC, dollars/yr (5)
where: CKWH = the cost in dollars of a kilowatt-hour.
AFAC = the ratio of design rotational speed to 5 rpm.
The screen replacement costs are expressed as:
COSTO(N,4) = 60.*EFFA*.1376/DMATX(3,N), dollars/yr (6)
where DMATX(3,N) is the design submergence factor.
OVER-ALL PROGRAM STRUCTURE
General Features
The subprogran has been developed +o be compatible with the EPA Executive
Program as a sub-routine, but certain additional options are present so
that the subprogram also acts as a simulation model, yielding additional
output data on particle size, mass loadings, and parameters for construc-
tion of design curves, as an aid to microscreen design and performance
description. This option is activated through common argument 1S2, the
70
-------
TABLE 7
SUMMARY OF COST DATA
MICROSCREEN MANUFACTURER
Manufac-
turer
A
B
C
D
E
Size
Dia.xWidth
5x1
5x3
5x5
10x10
4x1
5x6
5x8
5x10
5x12
10x10
10x10
4x4
6x8
6x4
4x2
Effective
Area (sq ft
13.3
40
55
210
9
66
88
105
133
210
190
24
92
46
12
Capital Cos1
(Feb.l970)
14,000
18,000
35,000
75,000
1 1,000
13,200
28,100
31,900
36,400
41,600
Capital Cost
(June 1971 )
(est.)
15,100
19,400
37,800
81,000
11,900
14,300
30,400
34,500
39,300
47,000
60,000
15,000
35,000
22,000
8,500
71
-------
FIGURE 19
RELATIONSHIP BETWEEN PURCHASE COST/UNIT EFFECTIVE AREA
OF MICROSCREENS AND EFFECTIVE AREA OF MICROSCREEN
10,000
UJ
UJ
CE 50OO
O
cn
O
CE
-------
DAILY PO\ ER REQUIREMENT PER EFFECTIVE AREA, MICROSCREENS
W 1"H HIGH-PRESSURE SPRAY BACKWASH SYSTEMS
300
250
c
o
S. 200
o
CM
O
U)
o
LL)
CT
X
5
150
100
50
Y =0.4X 4- 1.25
50
100
EFFECTIVE AREA,sg ft
150
200
r\)
o
-------
stream number of a second input stream, which is not used by the micro-
screen subprogram. Setting this number to zero produces the additional
output printing.
The behavior of the concentration element in the program is such that it
can appear that the microscreen is actually generating solids - such
behavior was, in fact, observed on an influent/effluent basis in the
field in transient behavior. This is a result of the explicit specifi-
cation of drum pool solids concentration rather than influent concentra-
tion in the subprogram model, in lieu of the calculation of drum pool
solids concentration through evaluation of the various solids transfers
in the process. Thus, a check was placed in the program which automati-
cal ly sets the effluent solids mass flow rate equal to the influent
solids mass flow rate, resulting in zero suspended solids removal effi-
ciency, whenever the combination of input and operating variables and
predicted process efficiency is such as to result in a predicted greater
effluent mass flow rate relative to influent mass flow rate.
Overall process performance is computed in the subprogram model on the
basis of the influent/effluent solids removal efficiency, which is calcu-
lated as the ratio of the respective mass flow rates in the influent and
effluent streams. After this parameter is calculated, the effluent stream
vector is calculated from the influent stream vector by multiplying all
particulate species concentrations by this ratio; as noted above, the
concentration of dissolved components in the influent stream vector is
assumed to remain unchanged by the process and is transferred directly
through the process to the effluent stream vector.
Variable Coefficients
The variable coefficients in the model formulation (eight in number) were
originally determined from laboratory experiments with Pelaspan particles
(References 3 and 4); for the final formulation, twenty-one steady-state
points were utilized and the variable coefficients were modified heuristi-
cally until a satisfactory fit of predicted and observed (field) trapping
and hydraulic data was obtained.
The results of this process are presented in Figures 21 and 22. In the
heuristic fitting process, no distinction was made as to the provenance
of the individual test runs, i.e.,the same values of the eight coeffi-
cients were applied to all sample points, rather than varying them in-
dividually or by groups to obtain "better" fit by"placing each point on
the line". With eight variable coefficients, it is possible, by suf-
ficient dissection expostulation, and finagling, with regard to the data
points, to put any point anywhere on a given line. In the present case
this was not done. The correlation coefficient for the data presented in
Figure 21 is 0.192; it is not known whether this is significant for a
steady-state predictor of a dynamic process.
A comparison of predicted vs. actual trapping efficiency across the drum
can be made from the data shown in Figure 21. The predicted values were
74
-------
FIGURE 21
PREDICTED VS ACTUAL MICROSCREEN SUSPENDED SOLIDS
REMOVAL EFFICIENCY ACROSS DRUM POOL TEST RUNS
10 20 30 40 50 60 70
ACTUAL SUSPENDED SOLIDS REMOVAL EFFICIENCY, %
CORRELATION COEFFICIENT = 0.192
75
-------
FIGURE 22
CALCULATE WIDTH OF MICROSCREEN FABRIC
REQUIRED IN TEST RUNS
(ACTUAL WIDTH = f )
RUN I.D.
—
-
1
— — ^^^^^^
NO.
19
IB
17
16
15
13
(RUN 12 OMITTED) 12
II
10
9
e
— 7
c
5
4
3
2
1
2 I
CALCULATED WIDTH ,ft
7t
-------
derived on the basis of the actual measured drum pool concentration and
PSD characteristics, rather than from the available empirical correla-
tions with the influent concentration and PSD characteristics and drum
speed. Test Run No. 12 is not included in the comparison on Figure 21
Inasmuch as the solids loading resulted in the computer program exceeding
its maximum permissible number of iterations; Test Runs No. 20 and 21,
representing oxidation pond effluents, could not be modeled since the
particle size as calculated from a log-normal distribution was too small
to provide any trapping whatsoever.
A comparison of hydraulic performance of the model relative to the proto-
type can be made from the data presented in Figure 22, where in the calcu-
lated required microscreen width is plotted for each run by identification
number. A perfect prediction of hydraulic performance would be a width
of one foot, the actual width of the pilot plant drum.
The results of a sensitivity analysis of model predicted performance
for variation in drum pool characteristics are shown in Figure 23. In
each rosette (constructed for the first seven test run points), each
arm represents a 100 percent variation; the vertical axis represents
standard deviation. The values shown at the end of each arm represent
the percent deviation of the predicted values for the modified condi-
tion^) relative to those predicted for the original condition(s). The
upper value at each arm corresponds to percent deviation in trapping
efficiency while the lower value indicates the percent deviation in the
hydraulics. The results of this sensitivity analysis indicate that the
process is largely sensitive to the average drum particle size, a 10
percent variation producing as much as a 20 percent variation in trapping
performance. Further,the variations are asymmetrical and depend upon
the local position on the response surface; i.e., the sensitivity is
dependent on specific local conditions of trapping. An inspection of the
original values of dp and GLQQ-P for the seven test points shows_that,
with the exception oT Test Run No. 1, sensitivity to changes in dp in-
creases with decreasing dp, as would be expected. The exception 'for
Test Run No. 1, with its nigh sensitivity; may be ascribed to the low
value tfiQQ_p to implying high uniformity of particle sizes and associated
higher efficiencies and sensitivities of trapping.
Program Listing
The contents of DMATX and OMATX for the microscreening subprogram are
presented in Table 8. The program listing for the sub-program model is
presented in Appendix D.
77
LIBRARY U.S.
-------
FIGURE 23
SENSITIVITY OF PERFORMANCE TO DRUM POOL CHARACTERISTICS
o
•0.7
10.4
0
0.488
1.39
-10.4
0
+ 4.1 -4.5
-4.3 + 2.5
+ 10.8
-3.4
0.377
1.18
-12.4
-1.8
+3.5 -9.7
'-3.4 -4.7
+ 14.4
+ 3.2
0.299
0.781
-20.0
- 17.8
+ 7.1
+ 0.1
-77
-20
+ 13.8
+ 2.0
0.325
0.873
- 21.5
- 15.6
+ 5.8 +0.15
-0.2 +2.2'
+ 5.4
-0.6
0.578
0.981
-5.9
-0.7
-O.I +0.57
'-4.4 +5.6
+ 5.2
+ 0.6
0.528
0.666
-6.1
+-3.16
-0.19
-5.1
+ 0.2
t 5.6
+ 4.7
+ 0.4
0.505
0.922
-69
0
-5.1
KEY
-O.I S
+ 0.1 D
MC/MI
WIDTH
+ O.IS
D : dp (microns)
-O.ID
78
-------
TABLE 8
PROGRAM PARAMETERS
DMATX(I,N) = I. activated sludge effluent
2. trickling filter effluent
3. primary clarifier effluent
4. oxidation pond effluent
DMATX (2,N) = drum diameter, ft
DMATX (3,N) = submergence %/\OQ
DMATX (4,N) = operating head, inches
DMATX (5,N) = drum speed, rpm
DMATX (6,N) = average nominal fabric pore diameter, microns
DMATX (7,N) = hydraulic resistance coefficient inches of head/gallon per min
DMATX (8.N) = influent mean particle size, microns
DMATX (9,N) = influent a, n~
LOb
OMATX (I,N) = DMATX (2,N)
OMATX (2,N) - DMATX (3,N)
OMATX (3,N) = DMATX (4,N)
OMATX (4,N) = DMATX (5,N)
OMATX (5,N) = DMATX (6,N)
OMATX (6,N) = DMATX (7,N)
OMATX (7,N) = drum width, ft
OMATX (8,N) = solids removed efficiency %/\QO
OMATX (9,N) = yield, percent
OMATX (IO,N) = submerged screen area, sq ft
owATX (II,N) = efficiency of screening between drum pool and effluent
79
-------
SECTION VIII
ACKNOWLEDGMENTS
A number of people contributed their efforts to the,conduct of the pro-
ject. Dr. T. G. Shea served as the project manager. Dr. R. M. Males
and Mr. M. A. Aaronson developed the mathematical/computer models and
assisted in the formulation,pf the conceptual models. Mr. J. D. Stockton
and Mr. J. McGiI I served-as project engineers for the field program and
design of the microscreen- units respectively. Mrs. Lillian Cors edited
and typed the final draft of the project report, which was authored by
Drs. Shea and Males. During the course of the study Mr. John Convery
and Mr. Joseph Roesler provided valuable input to the project as project
officers for the Environmental Protection Age-ney.
81
-------
SECTION IX
REFERENCES
1 "State of the Art of the Microscreen Process" report prepared by
Engineering-Science, Inc. for Federal Water Qua Iity Administrati on.
July 1970. (Contract 14-12-819).
2 "Theoretical Formulation of Operational Model for Simulation of
Microscreen Behavior, Report B-1", prepared by Engineering-Science,
Inc. for Federal Water Quality Administration, September 1970.
(Contract 14-12-819).
3 "Current State of Operational Model for Simulation of Microscreen
Behavior", prepared by Engineering-Science, Inc. for Federal Water
Quality Administration, November 1970 (Contract 14-12-819).
4 "Development of Field Data Acquisition Program for Pilot-Scale
Microscreens", prepared by Engineering-Science, Inc. for Federal
Water Quality Administration, January 1971 (Contract 14-12-819).
5 Lynam, B., Ettelt, G., McAloon, T., "Tertiary Treatment at Metro
Chicago by Means of Rapid Sand Filtration and Microstrainers", JWPCF,
41,2, pp 247-279 (February 1969).
6 "Employment and Earning Statistics for United States, 1901-1967.
Bulletin 1312-5," prepared by Bureau of Labor Statistics, United
States Department of Labor.
Note: References 1 to 4 have been submitted to: Clearinghouse for
Federal Scientific and Technologic Information, United States Depart-
ment of Commerce, Springfield, Virginia 22151.
83
-------
SECTION X
PUBLICATIONS AND PATENTS
No publications or patents associated with the project have been pro-
duced or are pending as of the date of this report.
85
-------
SECTION XI
GLOSSARY
Symbol Description
A Submerged area of screen
BEF Backwash energy flux
d Mean particle size of log normal PSD
f Suspended solids removal efficiency
K Hydraulic resistance of backwashed medium
K° Hydraulic resistance of virgin fabric
MC Mass of solids retained per unit area over the
screening cycle
mi Cumulative mass loading of solids per unit area
periphery
Ml Mass of solids loaded per unit area over the
screening cycle
n Porosity
NPS Norminal pore size of fabric
P Pressure
PSD Particle-size distribution
q Unit flow rate
Q Volumetric flow rate
S Rate of screen presentation
v Average superficial velocity
vc Cumulative volume of retained solids per
unit area
vi Cumulative volume of influent solids per
unit area
c
X Suspended solids concentration
Y Yield
Units
Area
Force/length
Length
Unit less
Length/1ength/t i me
Length/Iength/time
Mass/area
Mass/area
Mass/area
Dimension less
Length
Force/area
Volume/area-time
Volume/time
Area/time
Length/time
Volume/area
Volume/area
Mass/volume
87
-------
SymboI Descri ption Un its
to Drum rotational speed Rev./time
p Mass density of suspended solids in drum pool Mass/volume
a Standard deviation of log-normal PSD Dimension less
Subscript Description
B Backwash; applied backwash flow
E Process effluent
EI,E2..E7 Segregated-flow effluent channel
I Process influent
P Drum pool
W Throughput backwash flow
88
-------
APPENDIX A
DESCRIPTION OF SEWAGE TREATMENT FACILITIES
The pilot microscreen studies were conducted from January to April 1971,
at two wastewater treatment facilities located in the San Francisco Bay
Area. Jhe San Leandro Water Pollution Control Plant utilizes a dual
bPological waste treatment system. The facility is equipped with a
standard rate activated sludge plant and a high-rate trickling filter
facility both used for the treatment of about 8 myd domestic and indus-
trial wastes. The Concord Water Pollution Control Plant is equipped
with a high-rate trickling filter system followed by aerobic ponds,
and is used for the treatment of about 5 mgd of domestic and light in-
dustrial wastes. Descriptions of each of these facilities are presented
in the following sections.
SAN LEANDRO. CALIFORNIA WATER POLLUTION CONTROL PLANT
Facility Description
A layout of the treatment facilities at the San Leandro facility is
shown in Figure 24. Both domestic and industrial wastes are routed
through the sedimentation basins or the primary clarifier before distri-
bution to the activated sludge or trickling filter systems. The micro-
screen units were located midway between the secondary clarifiers of the
activated sludge and trickling filter systems (Figure 24). Clarified
effluent was withdrawn alternatively from the clear well of the clarifier
of either system, and transferred to the CPU (chemical pretreatment unit)
of the pilot microscreen system. The approximate locations of the points
of intake of microscreen effluent from either clarifi.er are also shown
in Figure 24.
Performance Characteristics
Because of the small size of the San Leandro facility (8 mgd), the
routine monitoring program conducted at the facility is very limited in
nature. Influent and clarified effluent suspended solids, 6005, settle-
able solids, and pH characteristics are monitored weekly for both the
activated sludge and trickling filter processes. Sludge volume index
(SVI) and sludge density index (SDI) are monitored on a daily basis in
~Jie activated sludge process. The results of the weekly and daily moni-
toring programs developed during the six week period of testing are
summarized in Tables 9 and 10 respectively. No additional information was
available to describe other characteristics of the biological processes
at the San Leandro facility; particularly lacking were data on solids
concentrations maintained in the aeration tanks and on sludge wastage and
recycle rates and concentrations.
During the testing period at San Leandro, the average flow rates to the
trickling filter and activated sludge processes were 4.3 mgd and 3.0 mgd,
respectively. The hydraulic loading to the trickling filter averaged
89
-------
SEDIMENTATION
BASINS
r{
ACTIVATED SLUDGE
AERATION TANKS
LEGEND
ffl
MICROSCREEN UNITS
CHEMICAL PRE-TREATMENT
UNIT
TRICKLING
FILTER
MAIN
PUMPS
nnnn
DDDD
nnnn
SECONDARY
CLARIFIER
ACTIVATED
SLUDGE
SYSTEM
CHLORINE
CONTACT
TANK
/I "CENTRIFUGES '
f rSLUDGE HEATING
1 /, ,
NO. I
PRIMARY
DIGESTER
--HJLJ
DOMESTIC
WASTE
INDUSTRIAL
WASTE
TO
SAN FRANCISCO
BAY
X EFFLUENT SAMPLING POINTS
FACILITY LAYOUT , SAN LEANDRO, CALIFORNIA, WATER POLLUTION CONTROL PLANT
-------
TABLE 9
SUMMARY OF WEEKLY MONITORING DATA DURING PILOT
MICROSCREEN PROGRAM. SAN LEANDRO, CALIFORNIA
LOCATION
A.S. Infl.
A.S. Cl. Effl.
T.F. Infl
T.F. Cl. Effl
A.S. Infl.
A.S. Cl. Effl.
T.F. Infl.
T.F. Cl. Effl
A.S. Infl.
A.S. Cl. Effl.
T.F. Infl.
T.F. Cl. Effl
A.S. Infl.
A.S. Cl. Effl.
T.F. Infl.
T.F. Cl. Effl.
WEEKLY
SAMPLING
DATE
18 February 71
23 February 71
3 March 71
11 March 71
CHARACTERISTICS
FLOW,
mgd
3.8
-
4.1
-
3.2
-
4.1
-
2.2
-
4.5
-
2.8
-
4.4
••
pH
7.3
7.4
7.5
7.4
7.1
7.3
6.9
7.1
6.7
6.9
6.8
6.9
6.8
6.9
6.8
6.8
SETTLEAGLE
SOLIDS, ml/1
8.0
0.0
3.5
0.0
10.0
0.0
2.6
0.0
5.0
0.0
0.0
0.0
4.0
0.0
8.0
0.0
SUSPENDED
SOLIDS, mg/1
240
40
120
48
252
32
672
64
156
16
188
48
244
16
130
24
BOD5
mg/T
363
15
255
77
420
58
518
165
230
22
230
97
368
22
350
106
VO
H
Notes: (1) A.S. - Activated sludge
(2) T.F. - Trickling filter
(3) Infl. - Influent
(4) Cl. Effl. - Clarified effluent
-------
TABLE 10
SUMMARY OF DAILY SVI AND SDI DATA
DURING PILOT MICROSCREEN PROGRAM,
ACTIVATED SLUDGE PROCESS
SAN LEANDRO. CALIFORNIA
Date
2-14-71
2-15-71
2-16-71
2-17-71
2-18-71
2-19-71
2-20-71
2-21-71
2-22-71
2-23-71
2-24-71
2-25-71
2-26-71
2-27-71
2-28-71
Characteristics
SVI
171
64
1 15
78
1 18
83
121
-
82
182
81
96
155
96
132
SOI
0.58
1 .20
0.87
1.30
0.60
1 .20
0.83
-
1 .20
0.5
1 .2
1.0
0.6
1 .0
0.7
Date
3-1-71
3-2-71
3-3-71
3-4-71
3-5-71
3-6-71
3-7-7!
3-8-71
3-9-71
3-10-71
3- II -71
3-12-71
3-13-71
3-14-71
3-15-71
3-16-71
3-17-71
3-18-71
3-19-71
Characteristics
SVI
154
161
145
158
-
-
-
168
134
188
177
172
-
156
182
92
89
181
80
SDI
0.7
0.6
0.7
0.6
-
-
-
0.6
0.7
0.5
0.6
0.6
-
0.6
0.5
1 .1
1 .1
0.6
1 .3
92
-------
21.6 mgad (million gallons per acre per day) and the BODj loading to the
filter averaged 239,000 Ib/acre-day, or about 30,000 Ib/acre-day/ft of
filter depth. The average solids loading to the trickling filter was
198,000 Ib/acre-day, or about 25,000 Ib/acre-dey/ftof filter depth. The
BOD5 removal efficiency averaged 67 percent and Itie suspended solids
removal efficiency averaged 83 percent in the trickling filter system
(relative to plant influent) during the test period.
The volume of the activated sludge aeration tank is 84,000 cu ft, which
provides a detention time of five hours at a flow rate of 3.0 mgd. The
BOD5 loading to the aeration tank averaged 35,600 tb/day or 420 Ib
BOD5/day/l,000 cu ft aeration tank, and the 6005 removal efficiency ave-
raged 92 percent relative to plant influent during the test period. The
suspended solids loading to the aerator averaged 22,900 Ib/day and the
suspended solids removal averaged 88 percent during the test period. The
SVI of the mixed liquor solids varied from 78 to 182.
The range of suspended solids concentrations in the clarified trickling
effluent varied from 34 to 55 mg/I and the suspended solids concentration
in the clarified activated sludge effluent varied from 10 to 33 mg/I in
the observations made during the test period.
Set-up of Pilot Microscreen System
A flow sheet illustrating the set-up of the pilot microscreen system at
San Leandro is shown in Figure 25. The chemical pretreatment unit was
deployed as a head tank to which either clarified activated sludge or
trickling filter effluent was transferred by means of two pumps (each
I HP Marlow Centrifugal pumps), and from which the wastewaters flowed by
gravity to the microscreen units. The flow rate from the head tank to
each microscreen unit was regulated independent of that to the other tank
by adjusting the height of free discharge from the transfer pipe into the
feed well of each microscreen unit. Tap water was used for the back-
washing operations in each unit, and the effluents from the microscreens
were disposed to the activated sludge secondary clarifier.
CONCORD, CALIFORNIA WATER POLLUTION CONTROL PLANT
Facility Description
A layout of the treatment facilities at the Concord facility is shown in
Figure 26. Raw wastewater entering the plant is passed through the
nrimary sedimentation system (consisting of four clarifiers in parallel)
rrom which it is transferred through a two-section sump (Sump A) into the
trickling filter. Primary effluent was withdrawn from the first section
of Sump A for use in the microscreen testing.
Clarified trickling filter effluent is recycled to the second section of
Sump A, where it is mixed with the primary effluent prior to transfer to
the trickling filter. Trickling filter effluent (unclarified) was with-
drawn from Sump C for microscreen testing. Effluent from the secondary
clarifier is transferred to the oxidation pond system, and oxidation pond
93
-------
TRICKLING
FILTER
SECONDARY
CLARIFIER
BACKWASH
SOURCE
(CITY WATER TAP)
LEGEND
X CLARIFIED EFFLUENT SAMPLING POINTS
® LOCATION OF PUMPS
ACTIVATED
SLUDGE
SECONDARY
CLARIFIER
CHEMICAL
PRE-TREATMENT
UNIT
NOT TO SCALE
SET-UP OF PILOT MICROSCREEN SYSTEM
SAN LEANDRO, CALIFORNIA, WATER POLLUTION CONTROL PLANT
-------
SECONDARY
CLARIFIER
PILOT
MICROSCREEN
SYSTEM
LEGEND
STORM WATER
MICROSCREEN UNITS
CHEMICAL PRE-TREATMENT
UNIT
X EFFLUENT SAMPLING POINTS
WASTE WATER
/
JMP A
*,
1
PRIMARY
CLARIFIER
PRIMARY
CLARIFIER
PRIMARY
CLARIFIER
PRIMARY
CLARIFIER
STORAGE
POND TJ
FACILITY LAYOUT; CONCORD, CALIFORNIA
WATER POLLUTION CONTROL PLANT
GREASE
COLLECTOR
CHLORINATOR
LAB
CONTROL!
ROOM
I DIGESTER I
UTILITY BLDG
WATER
| 1 SUPPLY
POWER
| [SUPPLY
SLUDGE
THICKENER
INCINERATOR
EAST POND
ro
cr>
-------
effluent was withdrawn for microscreen testing from the outlet sump of
the oxidation pond system.
Performance characteristics
The average flow rate to the Concord facility was 5.2 mgd during the March-
April testing period, and the daily flow rate during this period varied
from 4.9 to 5.6 mgd. Because of the size of the Concord facility, the
monitoring programs conducted at the plant are limited in nature. Yearly
averaged data (from a monthly sampling program) are presented for BOD^,
suspended solids, settleable solids, and pH in Table II. On an annual
basis, the BOD removal in the trickling filter system averaged in excess
of 87 percent and suspended solids removal averaged 85 percent.
Each of the five streams used in the microscreen testing were grab-
sampled on 24 March 1971, and the results of analyses of these samples
are reported in Table 12. The TSS (suspended solids) concentration at
the time of sampling varied from 193 mg/l in the primary effluent to a
minimum of 30 mg/l in the clarified trickling filter effluent.
The surface area of the trickling filter is 0.374 acres and the average
loading to the filter during the study period was 5.2 mgd, equivalent to
13.9 mgad, or about 3.5 mgd/acre-ft of filter depth. The average BOD5
loading to the filter was 44,000 Ib/day, or 115,000 Ib/acre-day. The
average suspended solids loading to the trickling filter was 37,000 Ib/day,
or 100,000 Ib/acre-day. During this period, the monthly monitoring data
indicated that an average of 82 percent suspended solids removal and 78
percent BOD^ removal was obtained in the clarified trickling filter
effluent relative to that in the plant influent.
Set-up of Pilot Microscreen System
A flow sheet illustrating the set-up of the pilot microscreen system at
Concord is shown in Figure 27. The set-up used at Concord was similar to
that used at San Leandro in all aspects as described above. The effluents
from the microscreen units were disposed to the oxidation pond.
96
-------
TABLE 11
ANNUAL AVERAGE CHARACTERISTICS OF WASTEHATER STREAMS
CONCORD. CALIFORNIA WATER POLLUTION CONTROL PLANT
Source
Plant Influent
Primary Clarifier
Trickling Filter Effluent
Clarified Trickling Filter
Effluent
Final Effluent to Creek
(a) Fi 1 tered
(b) Unfil tered
Characteristics
BOD
mg/l
200-300
80-100
47
27
9
13
Suspended
Sol ids mg/l
235
85
53
30
-
31
Settleable
So 1 i d s m 1 / 1
15-20
2-10
1 .5-10
0
-
—
pH
7.6
7.6
7.9
7.9
-
10. 1
97
-------
TABLE 12
CHARACTERISTICS OF HASTEWATER STREAMS
USED AS MICROSCREEN INFLUENTS
CONCORD, CALIFORNIA HATER POLLUTION CONTROL PLANT
(Sampled 24 March 1971)
Source
Primary Effluent
Trickling Filter Effluent
Clarified Trickling Filter
Effluent
Oxidation Pond Effluent
Final Treatment Plant
Effluent
Characteristics
Suspended
Sol ids mg/ 1
193
65
41
57
65
Turb id i ty
JTU
44
20
14
1 1
13
Equ i va 1 ent
Turb id ity
JTU
47.5
27
15
13
15
pH
8.2
8.4
8.5
10. 1
10.3
Note: Equivalent turbidity: sample homogenized prior to
measurement of turbidity.
98
-------
CLARIFIED
TRICKLING FILTER EFFLUENT
CHEMICAL PRE-TREATMENT
UNIT (HEAD TANK) —
UNIT A
<*> &
UNIT
A
i- crtrcri i ICMT
B
t
1 c
EFFLUENT
TRICKLING FILTER EFFLUENT
SECONDARY
CLARIFIER IX
LEGEND
X EFFLUENT SAMPLING POINTS
® LOCATION OF PUMPS
PONDS
PRIMARY EFFLUENT
WELL WATER TAP
NOT TO SCALE
SET-UP OF PILOT MICROSCREEN SYSTEM
CONCORD, CALIFORNIA, WATER POLLUTION CONTROL PLANT
-------
APPENDIX B
OPERATING AND ANALYTICAL PROCEDURES
An overview of the experimental program was presented in Table 4 (Section V)
The operating objectives and procedures and analytical procedures used in
the experimental program are described below, usinq the run designations
of Table 4.
OPERATING OBJECTIVES
Fabric Acclimatization (Run 0)
The general objective of the fabric acclimatization run was to establish
a time criterion for accomplishing the "break-in" of a microscreen fabric,
i.e., to establish the time period required for development of a stable
residual solids carryover (after backwash) on an initially virgin fabric.
The basic premises of the experiment were that:
(I) Over a period of operating time, there will result an accumu-
lation of debris on the pores of the backwashed fabric.
(2) The level of accumulation will approach stability for a condition
of constant drum speed and backwash energy.
(3) The impact of accumulation of solids on the backwashed screen
can be assessed by comparing the hydraulic resistance (K) of the
backwashed fabric relative to that of the virgin fabric (K°).
In order to develop a fabric acclimatization criterion, the two microscreen
units were operated over an eight-hour time period in a fixed range of
headless values and process flow rates, varying PB (backwash pressure)
and S (drum speed) only as necessary to maintain the headloss in the fixed
range. The flow rates and solids transfers in the units were tracked
over time. The MTA analysis (described subsequently) was conducted on
backwashed screen panels removed from the drum at bi-hourly intervals
to determine the hydraulic resistance of the screen on a point-in-time
basis.
Long-Term Runs (Runs I and 5)
The objective of Runs I and 3 was to determine, on a continuous, 24-hour
basis, how the microscreen process behaved and how transient variation in
•nfluent quality and selected variation of the operating parameters on
Pp,S, and total headloss affected process performance (as measured by
efficiency of suspended solids removal across the screen). The funda-
mental understanding of how the variables interacted in the process, to
be derived from an analysis of data from these runs, was to be used in
the design of specific experimentation in subsequent runs.
The operating protocol for Run I was analogous to that of Run 0. During
the initial 24-hour run, it was found (because of the combination of
fabrics and wastewater sources selected) that it was not possible to
101
-------
achieve a six-inch headless during the run. For this reason, the opera-
ting protocol of Run 3 was revised as follows, in order that a headloss
equal to, or greater than, six inches was maintained at all times:
(I) As an initial step, the drum speed was adjusted to increase or
decrease H, , using a drum speed control setting range varying from
20 to 75 percent.
(2) If the above step did not result in maintaining HL at a level
equal to or exceeding six inches, then the backwash pressure (Pg)
was adjusted downward as necessary within an operating range of 15
to 30 psig.
(3) As a third measure, the influent flow rate (Q|) was adjusted
as necessary to maintain the desired headloss.
Run I was conducted over the 24-hour operating period as originally planned,
However, it was necessary to terminate Run 3 after 12 hours of running
time due to failure of a facility pump supplying the influent stream to
the head tank unit (HTU).
Backwash Subprocess Run (Run 2)
The overall objectiv e of Run 2 was to describe the effect of varying the
drum speed and the backwash pressure Pg alternatively on the hydraulic
resistance of a panel of the backwashea medium (fabric-residual solids
complex) as removed from the drum. A specific objective of the run was
to document the response surfaces relating each parameter with the clean-
ing efficiency as measured by the ratio of the virgin fabric hydraulic
resistance (K°) and the backwashed medium hydraulic resistance (K), i.e.,
defining cleaning efficiency as equal to K°/K.
The operating approach entailed the following steps:
(I) A headloss range of six to eight inches was maintained for both
units; Unit A was operated at a constant backwash pressure of 25 psig
and Unit B at a constant specific backwash rate of two liters/min per
sq m/min of drum speed.
(2) In Unit A, process performance and the hydraulic resistance of a
panel of backwashed medium was assessed after operation for one or
two-hour time periods at drum speeds of 3.9, 7.3, 10.2, and 12.7
rev/mi n.
(3) In Unit B, process performance and the hydraulic resistance of
the backwashed screen panel was assessed after operation for one or
two-hour periods of operation at backwash pressures and drum speeds
as follows:
(a) Pg, 10 psig (main header; drum speed, 4.0 rev/min
(b) Pg, 20 psig (main header); drum speed, 6.1 rev/min
(c) Pg, 20 psig (main and auxiliary headers; 10.3 rev/min
(d) Pg, 35 psig (main header); drum speed, 8.8 rev/min
102
-------
Runs 5 to 15
The overall objective of these runs was to extend the applicability of
the response surface relationships developed in Runs 0 to 4 by operations
with a diversity of process influents and the available selection of
microscreen fabrics. The operating approach used in these runs was de-
veloped as a result of experience acquired in Runs 0 to 4, and consisted
of the following:
(1) The basic operating objective of the approach was to maintain
a constant PB (backwash pressure), during the entirety of a 2-hour
subrun (three subruns per run).
(2) In order to maintain a constant PB level, the initial subrun
values of Q| and S (drum speed) were selected to provide a headless
equal to, or greater than, six inches throughout the subrun.
(3) A different value of P was used during each of the three subruns
in the run, and S was adjusted as necessary to maintain a headless
equal to, or greater then, six inches.
Runs 16 and 17
The overall objective of Runs 16 and 17 was to investigate process per-
formance in either of two ways:
(1) Run 16 (using Mode A) to assess the relationship between process
efficiency and Ml (solids loading, mass suspended solids loaded on
the fabric/unit area of fabric) at a constant dp (median particle
size of the suspended solids concentration in the drum pool).
(2) Run 17 (using Mode B) to assess the relationship between pro-
cess efficiency and dp at a constant value of Ml.
The operating approach for Run 16 was to conduct a sequence of microscreen
subruns, each using a successive dilution of HTU effluent as an influent
stream. The HTU effluent was diluted with tap water and the solids
concentration in the diluted stream was maintained constant during each
subrun by continuously adjusting (as necessary) the flow rate of tap
water into the HTU effluent. The solids concentration of the diluted
stream was measured on a real-time basis using the homogenized-samp Ie
optical density analysis discussed below. The control variables PB, S,
and Q| were maintained constant during the run.
The operating approach used in the conduct of Run 17 was based on using
Unit A to generate a throughput washwater stream in which the particle
size distribution of the suspended solids could be controlled, and to
use this stream to adjust the particle size characteristics of the in-
fluent stream to Unit B. The median particle size of the throughput
backwash stream from Unit A was adjusted from subrun to subrun by varying
the fabric used in the microscreen in each subrun. The solids concen-
tration of the combined stream (which comprised the influent to Unit B)
was held constant throughout all of the subruns by continuously adjusting
the flow rate of the HTU effluent. The solids concentration of the combined
103
-------
stream was measured on a real-time basis using the homogenized-samp Ie
optical density analysis. The control variables P , S, and Q( were main-
tained constant during the run.
Operating Procedures
The operating protocols for the experimental program were based on the
subrun, or unit of work in which all control variables were held con-
stant for a defined time period while process performance was being
monitored. As a general rule a run was comprised of three two-hour
subruns, during which the monitoring and analytical activities outlined
in Table 13 were conducted. The activities conducted in each subrun
included the following:
(1) Development of two-hour composite samples of influent, effluent,
and throughput washwater, to permit an assessment of the process
efflei ency.
(2) Measurement of the optical density (OD) of homogenized samples
of influent, drum pool, and effluent, and the throughput washwater
streams at hourly intervals.
(3) Measurement of the particle size distribution (PSD) of samples
from the drum pool at hourly intervals and from the effluent col-
lector at bi-hourly intervals. The optical density data were conver-
ted to values of total suspended solids using correlation curves re-
lating OD and TSS as developed on an ongoing basis during the field
program. Upon completion of the subrun, a panel of backwashed medium
was removed from the drum and subjected to the MTA analysis for
measurement of the hydraulic resistance (K) of the backwashed medium.
Analytical Techniques
The analytical techniques used in the experimental program were, with
exception of the MTA, PSD, and homogenized optical density analyses,
performed as described in "Standard Methods for Analysis of Water and
Wastewater", 13th Edition.
MTA Ana lys Is
The MTA (Medium Testing Apparatus) analysis was developed to permit
measurement of the hydraulic resistance of panels of microscreen fabric
under steady flow conditions. The MTA is shown in Figures 28 and 29,
and is designed to permit the continuous flow of filtered water from a
head tank into a test head and through the fabric panel. The MTA was
used in the experimental program both with virgin fabrics and backwashed
fabrics removed from the microscreen unit during experimentation. The
hydraulic resistance was measured as the slope of the headloss vs. super-
ficial velocity curve (units of cm per cm/sec at 15° C), and was desig-
nated by the symbols K° for virgin fabrics and K for backwashed fabrics.
In the MTA test procedure, the panel of fabric to be tested was loaded
into the pilot-scale MTA by clamping the panel to the support bar shown
in Figure 28. The test head could be moved laterally and the support
104
-------
TABLE 13
SUBRUN OPERATING SCHEDULE
SUBRUN
TIME
(mi nutes)
0
30
60
90
120
SAMPLING POINT AND ANALYSES
PSD
-
P, E
P, E
-
OD
-
1, P, E, W
1, P, E, W
-
TSS
(2-hr comp)
1, E, W
1, E, W
1, E, W
1, E, W
1, E, W
ACTIVITY
Start subrun
-
Monitor head losses, P , and S at half-hourly
i nterva 1 s
Stop subrun; remove screen panel and start MTA
ana lysis
H
o
Ui
Symbols: I = Influent
P = Drum pool
E = Effluent
W = Throughput backwash flow
PSD = Particle size distribution
OD = Optical density
TSS = Total suspended solids concentration
-------
MTA TEST HEAD
fO
Note: Not to scale
FLOW
PIEZOMETER TUBE AND BLEED
VALVE CONNECTION
•5"
FABRIC (EXPOSED AREA : 62.2 sq cm)
-------
P .OT-SCALE MEDIUM TESTING APPARATUS (MTA)
x- CONSTANT HEAD TANK
7
VALVE A
L- PRE-FILTER
HOSE VALVE
PIEZOMETER TUBES
BLEED VALVES
HXI—
n—
(SIDE VIEW)
-CONSTANT HEAD TANK
PIEZOMETER TUBES
|>k] VALVE A
VALVE B
-CLIPS-
O
VALVE B
-TEST HEAD (DETAIL FIGURE B-2)
FABRIC
-TEST HEAD
SUPPORT BAR
(END VIEW)
-------
bar vertically so that the test head could be located at any point on
the attached panel. After the test head had been located, the fabric
was pulled taut and the test head clamped over the fabric using C-clamps.
The test head is the key element in the MTA and consists of a 3-1/2 inch
ID (internal diameter) lucite column with a cross-sectional area of 62.2
sq cm. The 3-1/2 inch ID span across the test head was similar to the
span between the peripheral bands on the drums of the pilot microscreen
units; as was the case in the pilot units, no fabric supports were used
to minimize deflection or stretching of the fabric caused by the movement
of water through the test head.
The hydraulic components of the MTA include; a pre-fliter, constant head
tank, valves, and piezometer tubes as shown in Figures 28 and 29. Water
was introduced to the MTA by filling the head tank through the hose valve
with all other valves closed. The test head was then filled by opening
the bleed valves and Valve A slowly, releasing alI trapped air from the
system between the head tank and the test head. At this point, the bleed
valves were closed and air was purged from the piezometer tubes to com-
plete preparation of the unit for a test run.
A test run was conducted by adjusting the flow rate through the test head
with Valves A and B to attain a desired differential head-loss across
the fabric. The discharge rate through the test head was measured with
a bucket-stopwatch procedure, and the superficial velocity was determined
by dividing the average flow rate by the exposed fabric area. The MTA was
shut down by reversal of the above procedure.
Particle Size Distribution
The particle size distribution (PSD) analysis was conducted using the
techniques of photomicroscopy to record the presence of microscopic-
sized particles on slides, and statistical counting and plotting tech-
niques to develop and describe the particle size distribution character-
istics of the sample suspension. The photomicroscopic technique (using a
Polaroid Camera) was adapted after several alternatives were explored for
the following reasons:
(1) The technique permitted the direct observation of the particles
being recorded.
(2) Microphotography precluded the necessity of measuring individual
particle size and count at the time of observation, reducing the time
of direct observation.
(3) The quality of photography could be checked immediately after
the particles were photographed.
A statistical analysis was conducted with sets of photomicrographs of
several samples to assess the minimum number of particles that should be
counted to ensure that a representative particle size distribution was
obtained. It was observed that counts in excess of 300 particles per
individual sample did not increase the precision of the analysis as
measured by the coefficient of variation of the median particle size of
108
-------
the PSD (which was found to be log-normal for all suspensions observed
in the present study). On this basis the criterion was established that
a minimum of 300 particles should be counted per individual sample,
requiring that from three to ten individual photomicrographs be taken of
each sample.
The basic equipment and materials used in the PSD analysis consisted of:
(1) Tiyoda R Microscope
(2) Tiyoda Microscope Light Source
(3) Polaroid Camera and Olympus PN-P camera/microscope attachment
(4) No. 10F/3,000 speed Polaroid film (black and white)
It was found that the best photomicrographs were obtained at a shutter
speed of I/IOO second, a microscope light setting of 5 to 6 volts, and a
lens adjustment of 40x or lOOx.
The photomicroscopic procedure entailed the following basic steps:
(1) Mount sample of suspension on glass slide and cover with cover
si ip.
(2) Scan sample at 40x lens adjustment to define the distribution
of particles on the slide.
(3) Conduct photomicrography sequence to record a minimum of 300
particles; prepare a log identifying each photomicrograph, and iden-
tify on the Polaroid print all particles seen in the microscope.
The number and size of individual particles on each photomicrograph were
observed using a scaled circle template and magnifying glass. The par-
ticle size was taken as the maximum particle dimension, and the particle
counts were arranged into a frequency-particle size data set from which
probability plots could be made and the parameters of the PSD (median
particle size and standard deviation) ascertained.
Homogenized-Sample Optical Density
The homogenized-sample optical density analysis was done to provide a
fsal time method for tracking the concentration of particulate matter in
the influent and effluent streams of the microscreen during a subrun.
The sample was first homogenized over a standard two-minute time interval
to normalize the particle-size distribution of the suspension, after whicl-
the sample was allowed to stand for five minutes to release air entrained
during the homogenization. After completion of these steps the optical
density of the sample was determined and, by means of a correlation
curve relating the optical density and total suspended solids (TSS) con-
centration, the TSS concentration of the individual sample was estimated.
109
-------
APPENDIX C
FIELD PROGRAM BASIC DATA
110
-------
TABLE 14
INDIVIDUAL MICROSCREEN RUN DATA
(Run 1 Unit A)
Time
0930
1000
1100
1200
1300
1400
1510
1530
1600
1700
1800
1900
2000
2100
2130
2220
2300
2400
0100
0200
0300
0330 ,
0400 /
0500
0600
0700
0800
0900
0930
XP
(tng/S.)
52
45
169
44
59
106
97
-
153
59
93
79
-
45
38
33
34
32
33
32
37
55
47
43
46
50
50
33
32
«l
(i/m\ n)
232
236
221
216
244
102
90
-
110
195
167
167
-
216
223
233
227
222
220
239
239
186
221
216
226
215
202
218
214
XS
XE
(mg/i)
22
19
18
20
30
43
-
-
65
30
24
2°
"~
20
18
18
16
14
14
14
18
20
18
13
20
18
20
13 :
12
1
"E
(S,/mi n)
236
240
225
220
248
108
102
-
1 19
204
180
181
-
221
228
237
231
226
224
243
243
20!
225
221
233
220
208
224
220
S
(sq m/.
^m i n )
2.72
2.72
2.72
2.72
2.72
3.1
6.3
-
4.8
4.8
8.8
6.65
•"
2.32
1.36
1 .36
1.36
1.36
1 .36
1.36
1.36
3.84
2.08
1.76
2.96
2.08
2.08
2.08
2.08
1
Ml
(mg/ .
asq m)
4,430
3,910
13,750
3,500
5,250
3,480
-
3,510
2,380
1,750
1 ,990
_
4,180
6,230
5,670
5,680
5,150
5,340
5,540
6,500
3,940
5,000
5,270
3,510
5,100
4,800 '
3,460
3,240
l
MC
(mg/ .
sq m)
2,530
2,230
12,250
1,880
2,510
2.000
-
1,900
1 ,130
1,270
1,450
_
2,620
3,220
2,520
2,950
2,820
3,030
3,040
3,280
2,890
3,050
3,060
1,930
3,200
2,800
2,060
1,970
MC/M 1
<*)
56.9
57.0
89.2
53.7
47.8
57.4
-
-
54.0
47.5
72.2
73.0
_
62.6
51 .7
44.5
51.9
54.8
56.7
54.8
50.0 ;'
73.3
61.0
58.2
55.0
62.7
58.2
59.5
60.8
d~P
(y)
8.6
8.1
1
,
5.5
!
i
i
1
|
i
4.3 I
i
1
PB
(psig)
10
10
10
10
10
12
35
30
24
16
30
17
10
10
10
10
10
10
10
10
23
19
16
32
20
20
20
20
K
(sec)
2.21
3.36
TIME AVERAGES
Time
0930 -
1530 -
2130 -
0330
0930
1510
2100
0300
0930
0930
Ml
5,720
2,760
5,730
4,290
4,910
l
MC
3,900
1,670
2,980
2,620
2,830
1
! MC/M 1
! 1
j 67.9 ~
60.3
51.9
61 .2
57.6 ~
6-hourly
Notes:
I. Clarified Activated Sludge Effluent (San Leandro )
2. Date: 19 and 20 February 1971
3. Screen: 30y SS
4. K° = 0.59 sec.
5. Total Screen Area: 0.8 sq.m.
Ill
-------
TABLE 15
INDIVIDUAL MICROSCREEN RUN DATA
(Run 1 - Unit B)
Time
0930
1030
1130
1230
1330
1430
1530
1630
1730
1830
1930
2030
2130
2230
2330
0030
0130
0230
0330
0430
0530
0630
0730
0830
0930
xs
K
(mg/£)
41
46
45
124
522
858
167
76.5
45
-
43
-
38
34
32
34
32
54
47
50
59
50
59
38
°l
U/min)
167
162
162
107
54
34
87
175
227
-
250
-
252
239
250
309
217
227
212
230
233
234
247
xs
XE
(mg/1)
12
16
16
30
57
73
49
22
16
-
16
-
15
12
12
12
II
16
II
16
15
12
15
I '°
QE
U/min)
171
166
166
122
70
51
95
185
232
254
-
256
244
256
320
225
234
219
243
246
248
260
(sq m/
mi n)
3.13
3.13
3.13
7.84
7.84
8.44
6.58
4.90
3.87
-
3.87
-
2.17
2.17
2.49
3.69
3.69
3.69
3.69
5.73
5.73
5.73
5.73
(mg/
sq m)
2,180
1,465
2,340
1,695
3,590
3,470
2,200
2,740
2,650
-
2,775
-
-
3,910
3,530
3,360
2,700
3,200
2,920
2,860
2,340
2,000
2,380
1,650
(^
sq m)
1,520
625
1,495
1,230
3,085
3,028
1,490;
1,900
1,690
-
1,720
-
-
2,480
2,180
2^120
1,740
1,960
1,930
1,890
1,710
1,490
1,740
1,210
MC/MI
<*)
70.2
42. L
64.1
72.7
85.8
87.1
67.8
69.2
63.9
-
61 .9
-
-
63.4
61 .8
63.2
64.2
61.3
66.1
66.1
73.2
74.5
73.1
75.3
dP
(u)
6.0
5.1
5.7
PB
(psig)
9
9
9
19
19
21
33
20
18
14
10
10
10
13
24
24
24
24
20
20
20
20
K
(sec)
3.59
5.30
TIME AVERAGES
Time
0930 - 1530
1630 - 2130
2230 0330
0430 - 0930
Ml
2,460
2,590
3.340
MC 1 MC/MI
1,830
1,700
2,100
2,360 1,660
!
0930 - 0930 j 2,660 ' 1,620
74.6
65.6
62.9
70.4
.
68.3
6-hour Iy
24-hourly
Motes:
I. Clarified Activated Sludge Effluent (San Leandro)
2. Date: 19 and 20 February 1971
J. Screen: 2lu SS
4 . K° I . 35 sec
5. Total screen area: 0.8 sq m
112
-------
TABLE 16
INDIVIDUAL MICROSCREEN RUN DATA
(Run 3 - Unit A)
Time
1200
1230
1300
1330
1400
1430
1500
1530
1600
1630
1700
1730
1800
1830
1900
1930
2000
2030
2100
: 2135
' 2200
; 2230
XS
*P
(mg/ I)
108
1 13
1 10
1 10
113
122
119
126
129
144
187
260
450
314
290
270
167
153
151
144
158
173
"l
( j/min)
159
227
234
21 1
230
231
230
222
229
229
228
214
208
234
213
197
213
237
223
-
223
385
XS
XE
(mg/ a)
63
77
73
75
84
91
91
93
98
113
153
218
260
260
183
215
135
125
123
114
130
148
QE
( S/mi n)
163
231
238
214
233
236
235
226
233
232
232
216
212
239
218
201
217
241
228
-
227
391
S
(sq m/
min)
1.6
1.6
1.6
1.6
1 .6
1.6
1.4
1.4
1.4
1.6
1.6
1.4
1.4
1.4
1.4
1.4
1.4
1.4
1.4
-
1.2
1.4
Ml
(mg/
sq m)
10,720
15,900
16,000
14,500
16,200
17,600
19,500
20,000
21,000
20,600
26,600
39,800
67,000
52,500
44,000
38,000
26,100
26,000
24,000
-
29,400
47,600
MC
(mg/
sq m)
4,370
4,750
5,170
4,460
3,900
4,180
4,200
5,060
4,780
3,630
4,400
6,150
26,400
8,150
15,750
7,050
4,500
4,360
3,500
-
4,660
6,150
MC/M 1
(?)
41 .0
29.9
32.4
30.8
24.0
23.7
21.5
25.3
22.8
17.6
16.5
15.4
38.0
15.5
35.8
18.5
17.2
16.8
14.6
-
15.9
12.9
TIME AVERAGES
Time
1200 —1530
i'bOO - 2230
1200 - 2230
1
Ml
16,300
33,900
28,100
MC
4,510
6,900
6,410
MC/MI
27.6
20.3
22.8
Notes:
I. Trickling Filter Effluent (Clarified, San Leandro)
2. Date: 19 and 20 February 1971
3. Screen: 30p SS
4. «° 0.59 sec.
5. Total screen area: 0.8 sq.rrr.
113
-------
TABLE 17
INDIVIDUAL MICROSCREEN RUN DATA
(Run 3 - Unit B)
Time
1200
1230
1300
I33C
1400
1430
1500
1630
1700
1730
1800
1830
1900
1930
2000
2030
2100
2130
2200
2230
2300
Screen
40P SS
II
M
II
11
1?
11
21 M SS
II
"
ft
If
M
It
"
11
II
11
"
It
M
4
(mg/4)
101
1 15
108
1 10
1 17
124
108
138
169
280
406
340
280
213
167
164
162
138
147
202
194
°l
U/min)
121
159
227
212
236
224
292
237
242
245
220
244
247
204
239
245
245
253
250
xs
XE
(mg/n
71
77
83
81
89
92
94
_
150
210
254
254
187
197
134
123
1 18
114
123
140
146
-------
TABLE ]8
INDIVIDUAL HICROSCREEN RUN DATA
(Run 5 Units A and B)
PARAMETERS
PB (psig)
S (sq m/mln)
Q! U/mln)
QE (i/min)
QB U/min)
Qw U/min)
X^ (mg/i)
Xp (mg/i)
XE (mg/i)
Xp. (mg/i)
t jj
xj (mg/i)
Ml (mg/sq m)
MC (mg/sq m)
MC/MI (?)
dp
-------
TABLE 19
INDIVIDUAL MICROSCREEN RUN DATA
(Run 6 - Unit A and B)
PARAMETERS
PB (psig)
S (sq m/min)
P. U/min)
PE (t/min)
PB U/min)
Qw (i/mln)
X^ (mg/i)
Xp (mg/i)
XE (mg/i)
XE| (mg/i)
Xw (mg/i)
Ml (mg/sq m)
MC (mg/sq m)
MC/MI (?)
3p
-------
TABLE 2U
INDIVIDUAL MICROSCREEN RUN DATA
(Run 7 - Unit A and B)
PARAMETERS
PB (pslg)
S (sq m/mln)
Q U/min)
(JE (l/mln)
QD U/tnln)
D
Qw (i/mln)
X? (mg/4)
1
Xp (mg/i)
x| (mg/1)
Xp. (mg/t)
tl]
x£ (mg/JO
ff
Ml (mg/sq m)
MC (mg/sq m)
MC/MI (*)
_ — _
dp
ffLOG-P
dE (u)
3E| (y) _
J.I — —
K° (sec )
4.1
K (sec )
K°/K
Control
Variables
Process
Efficiency
P.S.D
MTA
BEFdO ) (dyne/cm) JBackwash
Yield (
-------
TABLE 21
INDIVIDUAL MICROSCREEN RUN DATA
(Run 8 - Unit B)
PARAMETERS
PB (psig)
S (sq m/min)
Q, U/mln)
Op (i/min)
0D U/min)
D
Ow U/min)
X^ (mg/JO
Xp (mg/JO
X^ (mg/JO
XJ^ (mg/JO
Xjj (mg/JO
Ml (mg/sq m)
MC (mg/sq m)
MC/MI (?)
d_, (u )
r
°LOG-P
dE (v)
dr-l (P)
b 1 '
K° (sec+l)
K (sec+l)
K°/K
BEFdO"3) (dyne/cm)
Yield (Q£-QW)/QE,
Control
Variables
Process
Ef f iclency
P.S.D
MTA
Backwash
%
UNIT A SUBRUNS
A
B
C
UNIT B SUBRUNS
A
25
3.2
255
263
1 1.9
4.0
54
92
38
36
1 ,060
7,380
4,220
57.3
2.0
0.91
3.5
2.6
3.45
3.88
0.890
1 1 .3
98.5
B
25
3.2
242
251
14
5.0
72
95
48
47
1,020
7,230
3,410
46.9
2.7
0.90
3.5
3.1
3.45
4.04
0.853
16.3
98.0
C
Notes: I. Type of waste: Clar I tied Trickling Filter Effluent4. BEF = Backwash energy flux
2. Date: 9 t'arch 1971 ' = Influent
E = Composite effluent
3. Screen size: Unit A - El = Channel one effluent
Unit B - I On Nylon p Qrum pool
B Backwash (applled)
W Backwash (throughput)
118
-------
TABLE 22^
INDIVIDUAL HICROSCREEN RUN DATA
(Run 9 - Unit B)
PARAMETERS
PB (psig)
S (sq m/mln)
Q, U/mln)
QE U/m!n)
9B U/min)
Q.. U/min)
W
X^ (mg/Z)
Xp (mg/£)
XE (mg/J.)
XE| (mg/4)
Xjj (mg/Z)
Ml (mg/sq m)
MC (mg/sq m)
MC/MI (?)
dp (u)
aLOG-P
d£ (y)
dV. (u)
K° (sec+l)
K (sec+l )
114 * "A JT-M
Control
Variables
Process
Efficiency
P.S.D
MTA
BEF(IO~3) (dyne/cmTj Backwash
Yield (QE-0W)/QE,?
UNIT A SUBRUNS
A
B
C
UNIT B SUBRUNS
A
15
3.2
257
262
9.3
4.3
34.7
68
30
27
547
5,530
3,060
55.2
4.4
0.9es
3.3
5.9
3.45
1.65
2.09
7.2
98.5
B
20
2.2
236
242
7.7
4.6
55.5
89
46
45
544
9,630
4,480
46.5
2.8
• 0.9es
2.6
4.9
3.45
1 .75
1.97
5.0
97.5
C
25
2.2
238
245
11.9
5.2
102
199
55.5
57
622
16,300
10,070
61.8
3.6
0.9es
3.6
5.7
3.45
1.68
2.05
11.3
98.0
Notes: I. Type of waste: Clarified Trickling Filter Effluent4. BEF = Backwash energy flux
2. Date: 10 March 1971 E = ColposSe ef f I uent
El = Channel one effluent
3. Screen size: Unit A -
Unit B - 23vi SS
P - Drum pool
B = Backwash (applied)
W = Backwash (throughput)
119
-------
TABLE 23
INDIVIDUAL MICROSCREEN RUN DATA
(Run 10 - unit A and E)
PARAMETERS
PB (pslg)
S (sq m/min)
0, (i/mln)
QE (i/mln)
QB (t/mln)
Qw (i/mln)
X^ (mg/i)
Xp (mg/i)
XE (mg/i)
X|L (mg/i)
X^ (mg/i)
Ml (mg/sq m)
MC (mg/sq m)
MC/MI (*)
dp /9E,*
UNIT A SUBRUNS
A
15
1.6
230
236
9.6
4.2
93.5
III
64.1
88
628
15,970
1 1 ,830
74.1
3.1
.733
5.5
4.3
1.53
"2.70
0.567
7.2
98.2
B
15
1.4
240
246
1 1.3
5.0
126
118
106
82
730
20,700
1,700
8.2
4.1
.82
2.8
3.4
1.53
1.58
0.978
10.5
98.0
C
15
1.3
237
243
9.6
4.7
120
no
118
99
700
22,450
--
--
5.2
.57
2.6
1.9
1.53
1.86
0.823
7.2
98.2
UNIT B SUBRUNS
A
15
I.I
213
218
9.3
4.1
102
115
96
94
440
21,450
1 ,700
12.7
3.4
.85
3.8
2.9
1.35
0.74
1.82
7.1
98.0
B
1 —
15
0.4
182
185
9.3
3.8
132
97
90
119
392
41,000
2,300
6.6
2.7
1.44
1.5
1.9
1.35
0.89
1.52
7.1
97.8
C
Notes: I.
2.
3.
Type of waste: Clarified Trickling Filter Effluent4. BEF = Backwash energy flux
I = Influent
Date: 18 March 1971 E = Composite effluent
Screen size: Unit A - !8-22y SS El = Channel one effluent
Unit B - 2ly SS P = Drum pool
B = Backwash (applied)
W = Backwash (throughput)
121
-------
TABLE 25
INDIVIDUAL MICROSCREEN RUN DATA
(Run 12 Unit A and B)
PARAMETERS
PB (pslg)
S (sq m/ml n)
Backwash (appl led)
W - Backwash (throughput)
122
-------
TABLE 26
INDIVIDUAL MICROSCREEN RUN DATA
(Run 13 - Unit and B)
PARAMETERS
PB (pslg)
S (sq m/mln)
Q, (A/mln)
QE (A/min)
0B (A/mln)
Qw (A/mln)
X^ (mg/A)
Xp (mg/A)
XE (mg/A)
XEL (mg/A)
XjJ (mg/A)
Ml (mg/sq m)
MC (mg/sq m)
MC/MI (?)
dp (w)
aLOG-P
dV (p)
dn (V)
El '
K° (sec"*"')
K (sec"1"1)
___
Contro I
Variables
Process
Efficiency
P.S.D
MTA
BEF(IO~3) (dyne/cm) j Backwash
Yield (QE-QW)/QE,!£
UNIT A SUBRUNS
A
15
1.3
105
II 1.4
9.6
3.5
26
43
29.3
20.8
306
3,510
985
27.3
~
2.8
3.0
2.12
3.13
0.678
7.2
94.6
B
15
1.3
78
83.4
9.6
3.5
38.7
62.6
35.3
34.3
760
3,760
1,480
39.3
3.3
0.85
1.7
2.3
2.12
3.69
0.574
7.2
95.8
C
15
1.3
78
84.2
9.6
3.5
68
67
42
36.3
330
4,070
1,330
32.7
3.3
0.77
2.0
2.8
2.12
2.34
0.906
7.2
95.8
'
UNIT B SUBRUNS
A
15
1.9
41
48.7
9.3
1.0
61.4
43.4
37.6
37.5
650
960
--
—
2.5
1 .57
3.3
3.0
3.45
6.06
0.570
7.1
97.9
B
20
3.5
57
66.5
10.7
1.0
55.3
100.8
33.3
38.6
600
1,640
1,010
61.5
1.9
2.01
2.9
2.6
3.45
1 1.5
0.302
9.5
98.5
C
25
3.9
53
64.1
1 1.8
1.0
52.7
143
28
35.7
1000
1,930
1,474
76.6
2.6
1.28
1.9
2.0
3.45
25.2
0.137
11.4
98.4
Notes: I. Type of waste: Primary Effluent.
2. Date: 25 March 1971
3. Screen size: Unit A - !2-l5y SS
Unit B - |0u Nylon
4. BEF = Backwash energy flux
I = Influent
E = Composite effluent
El = Channel one effluent
P = Drum pool
B = Backwash (applied)
W = Backwash (throughput)
123
-------
TABLE 27
INDIVIDUAL MICROSCREEN RUN DATA
(Run 14 Units A and B)
PARAMETERS
PB (pslg)
S (sq m/mln)
Qj (t/mln)
QE (t/min)
QB (t/mln)
Qw (t/mln)
X^ (mg/t)
Xp (mg/t)
XE (mg/t)
XE) (mg/t)
Xjj (mg/t)
Ml (mg/sq m)
MC (mg/sq m)
MC/MI (*>
dp
-------
TABLE 28
INDIVIDUAL MICROSCREEN RUN DATA
(Run 15 - Units A and B)
Time
UNIT A
0945
1100
1145
1245
1400
1430
1500
1700
1745
1815
1845
UNIT B
0945
1100
1145
1230
1400
1420
1500
1700
1730
1815
1840
Type of
Effluent
Primary
ti
ti
II
Clarified
Trickling
Filter
II
II
ft
II
tl
Primary
ii
ti
H
Clarified
Trickling
Filter
IT
II
It
II
tl
xs
xp
(mg/A)
114
,23
64
31
38
62
70
61
28
86
20
120
98
61
42
86
48
78
63
53
48
31
Q,
(i/min)
43
52
47
40
57
35
44
80
81
106
55
64
71
62
44
69
4^
87
82
76
52
35
XS
XE
(jl/mln)
80
85
48
18
21
27
41
43
32
18
10
74
90
40
20
47
l«
54
24
36
28
20
-------
TABLE 29
INDIVIDUAL HICROSCREEN RUN DATA
(Run 16 and 17 - Units A and 8)
Time
RUN 16 -
1420
1450
1935
2010
RUN 16 -
1330
1430
1450
1530
1935
2010
RUN 17 -
II 10
1 140
1240
1255
1600
1720
1815
1930
Screen
UNIT A
I2-I5W SS
"
It
It
UNIT B
I8-22V SS
t|
11
rt
"
11
UNIT B
I5-I8V SS
11
IT
tt
II
II
II
II
xs '
XP
(mg/*)
38
23
23
53
43
43
28
48
160
1 18
127
128
186
76
138
73
i 1
9,
U/min)
107
137
75
116
145
141
134
148
42
93
9.2
25
19
18
26
76
45
57
xs
XE
(mg/i)
61
38
-
23
58
43
38
-
23
14
II
26
8.9
24
20
33
47
-------
TABLE 30
STEADY-STATE MICROSCREEN OPERATIONAL DATA
PU;I a . UNIT .
i
i
i
i
i
i
i
i
i
3
3
3
3
5
5
5
5
5
6
6
6
7
7
7
7
7
9
9
9
3
S
A
A
A
A
B
B
B
B
A
A
B
B
A
A
A
B
B
A
A
B
A
A
B
B
B
B
B
B
3
B
WASTE
AS
AC
AS
AS
AS
AS
AS
AS
TF
TF
TF
TF
AS
AS
AS
AS
AS
K
AS
AS
AS
AS
AS
AS
AS
TF
TF
TF
TF
TF
MC
cm
2.2
2.8
3.0
3.2
1.07
1.72
2.15
1.91
4.5
6.0
1.5
3.0
.69
1 .95
2.06
1 .06
.32
.76
1.48
.39
.45
1 .87
1.14
1.84
1 .34
2.50
4.5
4.61
4.71
3.85
Ml
sq m
4.1
5.7
4.5
5.0
1.9
2.78
3.75
2.88
18.1
25.0
1 1 .6
15.0
1 .49
2.52
3.84
1 .49
2.43
3.20
2.05
1 .52
2.45
3.87
2.01
3.28
2.31
5.65
9.7
9.36
7.81
7.75
MC/MI
%
53.7
49.2
66.7
64.0
56.3
61.7
57.3
66.3
24.9
24.0
12.9
20.0
46.3
77.3
53.7
71 .2
13.2
23.7
72.2
25.6
3P
V
8.3
5.5
4.3
5.0
7.3
6.3
6.0
5.9
6.2
6.2
3.6
3.6
9.5
10.4
7.2
10.5
14.0
7.2
11.6
7.0
5.0
3.5
8.8
10.0
1 1 .2
3.1
2.7
3.8
3.9
2.2
"lOG-P
.38
.67
.81
.61
.52
.65
.65
.65
.63
.63
1 .05
1 .05
1 .05
1.05
1 .08
1.15
.43
1 .03
1.15
1 .08
1 .00
.95
1.17
1.20
1 .19
1.00
.81
.90
.75
.70
FABR 1 C
NFS
V
30
30
30
30
21
21
21
21
30
30
40
21
13.5
13.5
13.5
20
20
23
23
25
16.5
16.5
10
10
10
23
23
23
10
10
HL
cm
6.0
8.7
10.0
13.2
7.8
1 1.8
29.0
12.0
22.0
25.5
8.0
27.5
10.7
6.5
8.8
1 1 .5
16.2
9.6
6.6
9.5
7.5
8.0
13.2
15.0
5.6
10.5
14.0
19.5
20.5
19.2
QEQ
t/mln
240
232
212
217
165
250
225
225
235
235
245
255
240
205
245
245
232
245
170
260
255
255
208
255
120
270
242
235
270
255
SUBM. v
A E
sq cm cm/sec
3400
3630
3650
3730
3605
3680
4070
3690
3910
3930
3620
4010
3660
3570
3620
3680
3790
3670
3540
3660
3590
3600
3715
3755
3545
3660
3750
3860
3880
3860
.424
.384
.348
.349
.275
.408
.332
.366
.361
.359
.407
.382
.393
.345
.407
.399
.367
.401
.288
.427
.427
.426
.336
.408
.203
.443
.387
.366
.418
.396
PR ' S
7 ' S3,m/
pslq . mln
10
10
21
20
9
14
13
24
15
13
10
15
16
25
35
15
35
15
25
15
15
35
15
25
35
16
20
.25
25
35
2.8
1 .4
3.0
2.2
3.2
3.8
2.5
3.8
1.5
1 .4
2.6
1 .96
3.6
4.7
1 .1
4.0
2.3
1 .5
4.5
2.3
2.3
2.3
3.1
3.2
3.2
3.2
2.3
2.08
3.2
3.2
NPS
dp
U/U
3.62
5.47
6.98
6.00
2.88
3.33
3.50
3.56
4.84
4.84
II .1
5.84
1 .42
1.30
1 .88
1 .91
1 .43
3.19
1 .98
3.57
3.30
4.72
1 .14
1 .00
.89
7.42
8.52
6.06
2.56
4.55
xpHLA
cm-sec
1.91
4.57
3.55
6.38
2.22
2.81
8.92
3.27
29.5
51.2
7.77
45.7
1 .44
1.78
2.99
1 .52
3.65
2.68
1 .61
1 .21
1 .47
2.57
2.72
4.23
2.33
4.79
1 I .60
17.30
13.3
12.3
XS
XP
mg/i
46
33
51
50
44
43
33
49
90
125
108
143
23
63
18
25
24
20
54
14
22
36
30
42
66
70
89
84
94
93
TEST
RUN f
I
2
3
4
5
6
7
S
9
10
1 1
12
13
-------
TABLE 30
STEADY-STATE MICROSCREEN OPERATIONAL DATA
RIJII 1 UNIT
10
10
10
II
II
1 1
1 1
13
13
13
14
14
14
K
14
1
-------
PACE 1
// JOB
LOG DRIVE
0000
CART SPEC
0001
CART AVAIL
0001
PHV DRIVE
0000
V2 M08 ACTUAL 16K CONFIG 16K
// OUP
•DELETE
CART ID 0001
MICRO
OB ADOK 3FOT
OB CNT OOET
to
vO
// FOR
•LIST ALL
SUBROUTINE MICRO
INTEGER OSI.OS2
DIMENSION ZPPCmm,ZPD2(50),ZPNR(50>,ZPD<50lfPCT<50>,ZPNM50»
DIMENSION ZPDmO),ZPNC(50),ZSP(50),ZST(50),ZPNOC50)
DIMENSION PCTCI50I.PCTI I50),PCT0150»
COMMON SMATXI20,25> ,TMATX< 20,25 I.DMATXI 20,20), OMATXt 20,20), IPI25lt
. INP,IO,ISl,lS2,OSl,OS2,N,IAERF,CCOSTI20,5»,COSTO(20,5»t
. ACOST(20,5) ,TCOSTI20,5>
C INITIALIZE PARAMETERS FOR SUBPROGRAM
ZP1-.9
ZP2=l.
ZP3-1.
ZPPORI1I-.25
ZPPURI2)».H
RHO«I.
XMU».Ol
FH«2.
FT-.5
LPN'10
LSN=>8
ZPDMS'1.05
CONVERSION FACTORS (PROGRAM OPERATES IN METRIC SYSTEM >
ZCIN«2.5«
ZCFT'12.*ZC!N
ZCFT3=2CFT«»3
ZCLO-1000./2.20S
ZCGAL'ZCFT3/7.48l
ZCMIN=60.
ZCGPD=ZCGAl/ZCOAY
ZCKPM=ZC«tV/ZCMIN
ZCGRV=32.17.ZCFT
ZCHl;j=ZCIN3'ZCGRV/ZCIN2
CClfJVtRT INPUT TO METRIC SYSTEM
-------
PACE
ZDR-(OMATX(2tN))>ZCFT/2.
ZOP-(DMATX(4.,N) )»ZCHIN
ZUW-1..ZCFT
ZDAV-IOMATX(5,N) 1'ZCRPM
ZDA. (CMAIXI3,N> )«ZCREV
ZDAR*Z(M>2DP>ZDA
*F p.fiMATX I6.NI-.0001
2 F K = f M
If 1 -f T
HLC^ IDMATX17.N) I«IZCHIN«ZCFT2»ZCHIN»/ZCGAL
C THANSFfcR DISSOLVED SPECIES TO EFFLUENT STREAM
DO fctt !=• 11 t 17
666 SMS r « I I ,US1 I -SMAIXI I , ISl )
01 -nvAixie.N)
SI -0"ATX('),N)
C CALCULATE OKUM POOL PARAMETERS
C CALCULATE CONCENTRATION EFFECT IN DRUM POOL
IF (DMATX(b,NI-2.5) 21,21,22
1 .»( .06DMATXIS.N) I
C CALCULATE DRUM POOL STANDARD DEVIATION
1002 IF (DMAtX(b.N)-5.62) 1003,1003,100*
1003 XX1-.0088«OMATX(5 ,N1
CO TO 1007
100<. IF (OMATXI5.NI-7.5) 1005,1005,1006
1005 XX1-.20-.02666«DMATX(5,NI
CO TO 1007
1006 XXI -.90-. 12«OMATX(5,N1
1007 S-SI-XX1
SOL-ALOC(L)AV)«S
CO-CO'l.E-06
$ DAVDAV'.OOOl
SO-tXPI SOLI
F'l./SCRT I2.«ZCPI )/SOL
Xfl r, = DAV»SD»« (-<•.»
XMAX.OAV'SO***.
XPH-LPN
DO 9 l«l ,LPN
XI- 1
9 ZP02 I I )>XHIN>(XHAX/XNINI»*(Xt/XPN)
I P U 1 I 1 I • X M I N
DO 92 I»2,LPN
92 ZPDl I I >
-------
PAGE 3
93 ZPOU1-SORT»ZPDHII«ZPD2I II
ODL-ALOGUHAX/XHINI/XPN IN
00 94. 1«1,LPN
X«ALOG(ZPOt1)/DAV)/SOL
Y«CU»F»EXP(-1.»X«»2/2.)«ODL
94 ZPNI(I)*Y/(ZCPI»ZPDU)«3/6.»ZPDNSI
C BACKWASH FUNCTION
HLC*I1LC/.B5
NN*1
PV=HLC
K«0
IF (152) 200,210,200
210 WRITE (5,900)
900 FORMAT I1H ,12HSEGHENT DATA)
WRITE 15,9051
905 FORMAT!IH ,3HSEG,IOOH VI PV VIN VC/VI 0
EFN EFF1C PV2 PVI POR VC >
VI-0.
WRITE (5.906) KtVUPV
906 FORMATIIH ,13,IX,21E9.3,IX) I
200 CONTINUE
100 XSN*LSN«NN
OT«ZDT/XSN
DO 102 I-l.LPN
102 2PNCII)<0.
M-=LPN
CALL OISTK(ZPN|tXNIiAVOIiSOI»PCT!«SURFI,VOLIt019ltZPOtZPOUZPD2fMI
x«o.
v»o.
vc«o.
PSV-0.
PSV2-0.
K°0
DEF'ZFP«ZPS
PVI'O.
PV2=-0.
00 19 K-l.LSN
00 IB Kl-l.NN
IF (PV2) 4<>,44,55
44 V=ZDP/PV
GO TO 66
55 V'(-PV»SQRT(PV*»2«4.*PV2«ZOPll/(2.«PV2l
66 OY-VOT
DVI»OY«VOLt
PVI«PVI»ZDP/V«OVI
C CALCULATE TRUNCATED PSD
00 71 1=1,LPN
IF l2P02tI)-DEF) 71,71,72
71 ZPNRI1)=0.
WHITE (5.998) DEF,ZP02(LPN)
998 FORMAT I1H .4HDEF-,E9.3,IX,10HZP02(LPN)«,E9.)I
STOP
72 ZfURd ) = ZPNII n.ALOG(ZPD2(n/06F)/ALOGtZP02(I»/ZPOHII)
IF (ZPNR1 I)-ZPNI(I)) 74,74,75
75 ZPNRd I = ZPNI (I)
74 11*1+1
00 73 I'll,LPN
-------
PACE 4
7) ZPNIU I I-ZPNI < M
CALL OI$TR(ZPNR,XNR,AVDR,SO*,PCT,SUftFR,VOLR,015R,ZPD,ZPDl,IPD2.M»
EFF1C-VOLR/VOLI
OVC-DVI«EFFIC
IF ,<,<,6>
*«.6 FORMAT I HI .'.HEX IT I
CALL t» 11
12 Utf1-01^R/b.
DEFlsALOblDEFl+ZPl'IDVC/DlSRJ-IALOGIDEFn-AlOGIOEFM
CiEF-EXIMDEFLI
DO 13 I • 1 ,LPN
ZPNRIII«/PNKI I )«DV
13 icnct iI.JPNCM»»ZPNR(ll
SUKfR-SURfK«OY
VOLR-VOIR-OT
VC«VC < DVC
VC'.-VC/AVDI
V I • V I * D V I
VY»Dr
C CALCULATf POROSITV
IF (SIJR-ZPPORI3II 31,31,32
31 Ptm*{HPORI1I»ALOG(SORI/ALOC(ZPPOR(3))»(ZPPOR(2)-ZPPOR(1)I
CU 10 33
32 POR-/PPORI2I
3J CliUflliUC
X-X»DVC/tl.-PORI
PSV«PSV»SURFR»Il.-POR)/POR«»3
PSV2-PSV2«SURFR«»2/VOLR/POR»»3
IF (VCN-/FT) 60,60,61
60 OQU-VCN
CO 10 62
61 OOU-ZFT
62 PVHLC«2P2**.2*ZDMU»PSV2+ZP3>HLC*(ZFM-1.)>QOQ
PV2'Z>"i»ZDRHO«PSV
18 CONTINUE
ISPIKI=POR
ZSTIKI'X'l.EO^
VIN»VI/AVOt
VCVI-VC/VI
OEFN-DEF/AVOI
IF (IS?) Z25.230.225
230 WRITE 15,910) K.VI,PV,VIN,VCVt,DEFN,EFFIC,PV2.PV!.POR.VC
910 FORM&TI1H ,I3,1X,10(E9.3,1XM
225 CONTINUE
19 CONTINUE
2DO-ISKATX(2,IS1)»ZCGAL>1.E6)/2CDAV
N-ZDU/(V'ZOSPI
00 863 l-l.LPN
88} ZPrjOl I I«ZPNI II I«ZDO-ZPNC< I )«ZDSP«W
C TRANSMIT PARAMETERS TO OMATX
OfAT
,N)«DMATX(5,N)
I 3,N
-------
PACE 5
OMATX(5,N)«DMATXI6.N>
OMATX(6,N)«DMATX(7tN)
OMATX(7,N)«W
C CALCULATE SUSPENDED SOLIDS REMOVAL
CALL OISTRtZPNC,XNC,AVDC.SOC,PCTC,SURFC.VOLC«Ol5C,ZPD,ZP01.2P02tMI
CALL DlSTRIZPNO,XNO,AVDO,SDO,PCTO,SURFO.VOLOtDl50.ZPOtZPDltZPD2tM)
AX|«ZDO«ZPONS»VOLI»ZCDAY/ZCL8
AXB=S«ATX(10, ISl)»ZOQ»ZCOAY»l.e-6/ZCLfl
AXC«W«ZDSP»ZPDNS»VOLC»ZCDAY/ZCLB
AXG«ZPONS»VOLU«JCOAY/ZCLB
OMATXI ll.N)*AXC/AXI
If (AXn-AXBt 275,275,276
276 AXO=AXB
2T5 SSR-AXO/AXB
AY'AXU-AXO
SKAT XI 10,051 )«SSR«SMATX( IO.IS1)
OMATX(8,N»»SSR
EFFA-3. K15«DMATX(2,NI«H«OMATX( J.NI
OC*tSMATXI2,lSl)»2646.)/UFFA»ZCFT2»l.E-*»
OHAtX19,NI=100.-l./J5.5E-««QQ»«l.H
SMAIXI2,OS2>«SMATXI2,IS1)«I1.-OMATX<9,N)/100.)
C ASSUME BACKWASH WATER IS TAKEN FROM PROCESS EFFLUENT
C CALCULATE WASTE STREAM CHARACTERISTICS
DO 364 I'll ,17
M 369 SMATX1 I ,OS2 I-SHATXI 1, IS1)
10 SHATXI 10,OS2)'( AY/SHATXI2.0S2) )».12
^ DO 370 1*3,9
370 SMAIX(I,OS2l>(SMAIX(10,OS2l/SMArx(10,OSU»»SHATX«IiISl»
SMATXt2,OSll=SMAIX(2,ISl)-SHAfX(2,OS2)
DO 101 l«i,9
101 SHAIXI I,OSl)«OMATXl8,N).SMATX( 1,151)
C CALCULATE CUSTS
C COST ELEMENTS
C CAPITAL COST INSTALLED
C 0«M COSTS OF FIXED WAGES (YEARLY)
C OM COSTS OF VARIABLE MACES (YEARLY)
C ENtrtGY COSTS
C SCREEN REPLACEMENT COSTS (AMORTIZED OVER SCREEN LIFE)
OPArxtlO,UI=EFFA
IF (EFFA-150.) 641,841,842
841 CCOSHN, 1 )^l3ilfrO./SORT(EFFA) J.EFFA
GO TO 6<>3
842 CCOSTIH.I )>EFFA>300.
843 CWACfc=3.00
COSTO(N,1)-CWAGE«365.
COSTOIN,2)='I38..ISMATX(2,IS1I»».19»-35.»(CCOST»N,U»1.E-J)
CKWH»DMATX(1,20)
AFAC=GMArx(5,NI/5.
COSTQ(N.il= 3t5.«CKWH»(EFFA«.4+12.5l»AFAC
CSCRN<60.
CCSCR'CSCHN«EFFA/OMATXI3,N)
COS 10 ('.',<. I «CCSCR«AFSCR
C AFSCR= AMORTIZATION FACTOR FOR 9 YR LIFE AT 4.5 PERCENT
C OPTIONAL OUTPUT PRINTING
IF (IS2) 600,610,600
610 CONTINUE
-------
PACE
OJ
£>•
ORUH POOL RETAINED EFFLUENT*)
.EO*
.EO*
.EO*
I AVOI.AVDC.A/00
SOI.SOC.SDO
0151,015C,0150
ll< AVG. OIAM. ,2X,3F10.2,8H
IH STD. DEV. .2X.3F10.2)
IH 015 .2X.3F10.2)
, KX,'INFLUENT ORUH RESIDUAL
J3* FORHATI1H ,//)
WRITE 15,110)
110 FQRMATI1H ,11X,
AVOI-AVOI-l.EO*
AVOC'AVDC'l.EO*
AVliO- AVDO* 1 .EO*
D15I«015l«l
015C-DI5C*!
U I 'j 0 - U 1 5 U • 1
WHITE I'j.ll
Wkl IE 15, I 12
WRIIE 15, 1 13
111 f-URMAIIlH , IH AVG. OIAM. ,2X,3F10.2,8H MICRONS)
112 FORMAKIH
113 FOR^AI UH
119 FOP MA I ( 111
WH1TE I
WRITE(5,I 19)
WR1 IE I 5, I I* I AXB,AXI,AY,AXC,AXO
11* FChMATIlH ,11H TOTAL MASS,2X,5E10.3,7H LB/DAV)
WRITE(5,33*)
DO 3U? l-l.LPN
387 ZPD2II)>ZPD2(1I'l.EO*
INC'LPN/IO
LAST>10-INC
WRIKI5,120) IIPD2III. I-INC,LAST, INC)
DO 65', I«I ,LPN
PCT I 1 I l-PCTlI I 1*100.
PCICI I I-PCTCII 1*100.
65* PCTUIII-PCTOII)«100.
WRI IE(5,121 I IPCTI (I I,
WRI TEI5, 122) (PCTC(I),
WRITEI5,123) (PCTO(l),
DO 221 1*1,LPN
221 PCI I I I-100.*ZPNC I I)*ZOSP*W/IZPNIII)*ZOQ)
HRIIEI5,12*I IPCT(I), I-INC, LAST,INCI
INC'LSN/*
LAST-INC**
WRITEI5.33*)
HRI1E(5,10*) (I,I-INC,LAST,INC)
10* FORMAM1H , 7HSEGMENT ,8X,7H 1,*I7|
WRIIEI5,105) ZSTIU.IZSTll) , I • INC , L AST , INC I
105 FORMAKIH ,I6H THICK. (MICRONS) , IX, 5F7.2)
WRITE15, 1061 ZSPI1),(2SPII),I«INC,LAST,INC)
RETAINED EFFLUENT*I
I-INC,LAST,INC)
I-INC,LAST,INCI
I-INC,LAST,INC)
• 13H
,13H
,13H
, 13H
106 FORMAT IN
120 FOUH4! IH
121 FORMAT IH
122 FORMAT IH
123 FORMAT IH
12* FORMA1 IH
600 CONTINUE
RCTURN
END
VARIABLE ALLOCATIONS
SMATXIRCl«7fFE-7C18
101ICI-71UA
9H POROSI1Y,7X.5F7.3I
13H D1AMIHICRON),IX,F5.1,9F6.I>
DRUM PSD ,1X,F5.1,9F6.1)
RETAINED PSD,1X,F5.1,9F6.I I
EFFLUENT PSD,1X,F5.1,9F6.I)
PCT. REMOVAL,U.F5.1.9F6.H
TMATXIRC )-7C16-7630
IS1I IO-71B8
DMATXIRCI-782E-7510
IS2IIO-71B6
OMATXIRO-T50E-T1FO
OSHIO-718*
IP(tC)-71EE-71BE
OS2(ICI-7182
INP(IC).
NIICI-
71BC
71BO
-------
PAGE
lAERFMCt-TUE CCOST66
= 057A
= 0586
•=0b')2
• 0596
= 05AA
»05R6
'05C2
= 05CE
= 05DA
= 0566
= ObF2
«05F6
• 060A
= 0620
UNREFERENCED STATEMENTS
5
STATEMENT ALLOCATIONS
1009 *06CA 900 «060B 905 -0665 906
112 =0771 113 «077E 119 -078B 114
123 >0812 124 «0821 666 -094F 21
COSTCHRCI-70E4-T01E
ZPOIR
ZSTIR
ZP2IR
FMIR
ZCIN2IR
ZCMINIR
ZCCRVIR
ZDAIR
2FPIR
XPXSRIR
SUIR
ODLIR
01 IR
015KR
PV2IR
SORIR
DEF1 IR
VCVI (R
SDCIR
SUOIR
AXCIR
CWAGEIR
LPNI 1
KII I
•0130-OOCE
•0388-0326
•051C
-0528
-0534
= 0540
»054C
= 0558
= 0564
= 0570
= 057C
= 0588
= 0594
= 05AO
= 05AC
«0bfl«
-05C4
= 0")DO
= 05DC
*0568
= 05F4
= 0600
-0616
-0622
•071E 996 -0727
•07A8 104 »07BA
•098A 22 -0998
1004 -OAOO 1005 'OAOC 1006 -OA1B 1007 -OA28 5 -OA3B
200 -OB73 100 -OB73 102 -OB85 44
73 -OCAl 11 -OCD6 SI -OCE6 12
62 »CDC8 18 >ODEE 230 -OE1E 22$
101 >102E 841 '1066 842 «1078 843
CALLED SUBPROGRAMS
FALOC FEXP FSORT OISTR FAXB
FSIO FSTOX FSBR FOVR FAXI
REAL CONSTANTS
.900000E 00-0634 .1000006 01=0636
.100000E-Ol*0640 .200000E 01-0642
. 120000E 02>064C . 100000E 04^0646
.321700E 02*0658 . tOOOOOe-03»065A
. 145000E 01»0664 .625000E 01<0666
.8eOOOOE-02=0670 .750000E 01=0672
.400000E 01=067C .6000006 OU067E
.4?OUCOE 01=0688 .1000006 05=068A
,'j50000E-03 = 0694 .1100006 01-0696
.J65000E 03=06AO .3800006 02=06A2
.125000E 02-06AC .1376006 00=06AE
-OBDC 55 -OBE4
•OCEB 13 -0022
«OE38 19 »0638
-1083 610 -10FF
ACOSTIRO-701C-6F56
PCTIR
ZPNOIR
ZP3IR
FTIR
ZCFT2IR
ZCHRIR
ZCHINIR
ZDSPIR
ZFHIR
COIR
FIR
XIR
XNI (R
VCIR
VIR
SUKFRIR
OEHIR
ObFNIR
SURFCIR
SURFOIR
AXOIR
CKMHIR
LSNI 1
till
-0194-0132
-03EC-038A
• 051t
•052A
-0536
«0542
«054E
= 055A
*0566
= 0572
= 057E
= 05BA
= 0596
= 05A2
= 05A6
= 05UA
= 05C6
^0002
= 0'JD6
= 05EA
= 05F6
= 0602
= 0618
-0624
446 -0736 910
105 -07CA 106
23 -09AA 1000
9 «OA7A 92
66 -OC08 71
31 -0060 32
883 -OE5C 276
387 -1167 654
TCOSTIRCI-6F54-6E8E
ZPNKR
PCTCIR
ZP4IR
ZPONSIR
ZCIN3IR
ZCOAVIR
ZORIR
ZDKR
ZFTIR
OAVIR
XHINIR
YIR
AVOI IR
PSVIR
UYIR
VOLRIR
VCfJIR
ZOOIR
VOLCIR
VOLOIR
SSRIR
AFACIR
HI
INCH
•01F8-0196
-0450-03EE
-0520
= 052C
-0538
= 0544
= 0550
= 055C
'0568
= 05f4
= 0580
= 058C
= 0598
= 06A4
«05BO
= 05BC
= 05C8
= 0504
«05tO
'05EC
= 05F8
= 0604
>061A
= 0626
•073C 334 -0745
•0709 120 -07E5
•09C7 1001 -0908
•OAA7 93 -OAC4
•OC2E 72 -OC4F
-0083 33 -0089
-OF2C 275 -OF30
• UB9 221 -1219
FAOD FADOX FSUB FSUBX FHPV
FLOAT SWRT SCOMP StOFX SIOF
.300000E 00»0638
.500000E 00»0644
.220500E 01-0650
.2500006 01-065C
.9600006-02*0668
.2000006 00-0674
.8500006 00=0680
.1000006 07=068C
.15U0006 03=0698
.1900006 00=06A4
.250000E 00-063A
. 105000E 01*0646
.7401006 01=0652
.640000E-01-065E
.1190006 01=066A
.266600£-0l=0676
.0000006 00=0682
.3141506 01=068E
.3860006 04=069A
..3500006 02=>U6A6
FHPYX FOIV
SIOI SUBSC
110
121
1002
94
75
60
369
600
ZPPORtR
ZPDKR
PCTIIR
ZP5IR
ZCPI IR
ZCFT3IR
ZCGPOIR
ZOPIR
ZDARIft
HLCIR
XXlIrt
XMAXIK
PVIR
SOUR
PSV2IR
DVI IR
OlblUR
PbRIR
W(R
015CIK
015UIR
AYIR
CSCRNIR
NNI 1
LAST! I
•0004-0000
•025C-01FA
-04B4-0452
-0522
-052E
»053A
-0546
= 0552
= fJ5'jE
= 056A
= 0576
•0582
-05BE
= 059A
= 05A6
= 05B2
= 05I',E
-05CA
= Or)U6
= 05C2
«05EE
»05FA
= 0606
«=061C
= 0628
•074A ttl -075F
•07F4 122 "0803
-09E7 1003 -09F3
-OB2E 210 -OB50
-OC8E 74 -OC97
'ODBE 61 -ODC4
-OFAB 370 -OFEO
-1202
FOIVX FLO FLDX
STOP
.1500006 00-063C
.3141606 01-0648
.6000006 02=0654
.1400006 01=0660
.400000E-01=066C
.120000E 00=0678
.5000006 02=0684
.264600E 04=0690
.3000006 03=069C
.1000006-02=06A8
SNR
.1500006 01>063E
.254000E 01-064A
.2400006 02-0656
.5800006 00=0662
.562000E 01=0666
,1000006-05=067A
.5000006 01=0686
.1000006 03=0692
.3000006 01=069t
.4000006 00=0&AA
INTEGER CONSTANTS
10-06BO 8-06B1
2-06B2
3-0603
11-06B4
17»06B5
5-06B6
1-06B7
0-06B8 SOOO«06B9
-------
PAGE 6
1-06BA 4-0668 0-06BC
CORE REQUIREMENTS FOR MICRO
COMMON 4466 VARIABLES 1588 PROGRAM 5200
RELATIVE ENTRY POINT ADDRESS IS 0830 (HEX)
END OF COMPILATION
// DUP
•STORE MS UA MICRO
CART 10 0001 OB ADOR 3F24 OB CNT OOE6
•DELETE OISTR
CART ID 0001 OB ADDR 3F07 OB CNT 0010
// FOR
•LIST ALL
SUBROUTINE OISTRIPSOiSUMN,AVG.SO.PCTV,SURFiVOLtOlStZPDtZPDltZPD2«M
«l
DIMENSION PSOI501.PCTV150),ZPD(50),ZPDlt50>,ZPD2(50>
COMMON SMATX(20,25I«THATX(20,25),DMATX(20,20),OMATXI 20,201,IP<25»,
. INP,10,ISliIS2,OS1.0S2.N,IAERF,CCOSTt20,5),COSTO(20.5>t
. ACOST(20,St,TCOSTI20,5l
LPN-M
KPI-3.141519
SUMN'O.
SUMUL-0.
SUMU2»0.
SURF-0.
VOL-0.
00 11 I'l.LPN
XN-PSDI I )
0-ZPOII)
DL'ALOGIDI
SUMN=-SUMN»XN
VOLUM « XN*0>«3*ZCP!/6.
SUMN-SUMN»XN
SUMDL-SUMOL*VOLUM»OL
SUM02«SUMD2*VOLUM»OL»»2
SURF=SUKF»XN»0»»2«ZCPI
VUL'VOL»VOLUM
PCTVII I»VOL
11 CONTlNue
IF (SUMNI 20.20.30
20 AVC-0.
SD'O.
SURF«O.
VOL'O.
015*0.
RETURN
30 AVG*EXPISUMOL/VOLI
VA»L"iUMD2/VOL-(SUMOL/VOL»«»2
SUL'SCRTIVARL)
SO>EXP
-------
'AGE
(a
-4
si PCTvm-PCTvm/vot
00 32 l-l.LPN
IF IPCTV(I)-.15» 32.33.33
32 CONTINUE
33 DLl-ALOGIZPOlim
OL2=AlCGtZP02II)>
Pl-PCTVtl-U
P2*PCTVMI
OU5 = DL1 + I.15-P1I/IP2-PH«IOL2-DLH
D15-F.XPIOL15»
RETURN
END
VARIABLE ALLOCATIONS
SMATX(RC>'7FFE-7C1B THATXJRO-TC16-7830 DMATXIRO-762E-7510 OMATXIRO-T50E-T1FO IPI ICI-71EE-71BE INPIICI-71BC
IGUC) = 71BA IS1(IC)-71B8 IS2IIU-71B6 OS1(RC»»71B4 OS2IRCI-71B2 NIIU-71BO
IAERFI1U-71AE CCOST(RC)«71AC-70E6 COSTOIRCI-70E4-701E ACOSTIRCJ-701C-6F56 TCOSTIRCI-6FS4-6EBE ZCP1IK )«0000
SUHOLtR 1-0002 SUM02IR )-000« XNIK l«0006 OCR 1-0000 OL(R I-OOOA VOLUHIR I-OOOC
VARLIR I'OOOE SDL(R 1-0010 OLHR 1*0012 DL2IR I-OOU PUR 1-0016 P2IR 1-0018
OL1SIR );001A LP.NII 1 = 0020 III )>0022
STATEMENT ALLOCATIONS
11 *OOE1 20 -OOEF 30
CALLED SUBPROGRAMS
FALOG FEXP FSCRT FAOO
SUB IN
-0105 31 -012E 32 -0152 33 -015B
FSUB FMPV FOIV FLO FLOX FSTO
FSTOX FSBR
FOVR
FAXI
SUBSC
REAL CONSTANTS
.31A1S1E 01-0028
.OOOOOOE 00-002A
.600000E 01-002C
.150000E 00-002E
INTEGER CONSTANTS
1*0030 3-0031 2-0032
CORE REQUIREMENTS FOR DISTR
COMMON 4466 VARIABLES 40 PROGRAM 384
RELATIVE ENTRY POINT ADDRESS IS 0036 (HEX)
END OF COMPILATION
// OOP
•STORE MS UA OISTR
CART ID 0001 DB AODR 3FEO OB CNT 0010
// FOR
•lOCSICAROi 1403 PRINTER. DISK!
•LIST ALL
INTEGER OS1.0S2
C PROGRAM MAIN
C THIS PROGRAM TESTS THE SUB-PROGRAM AND ACTS AS A SIMULATION MODEL
DIMENSION NAMEm
COMMON SMATX(20,25),TMATXI20,25),OMATX(20,20».OMATXI20,20),IPI25),
. INP,IO,ISl,IS2,OSliOS2,N.IAERFtCCOST(20.5).COSTOI20,5t.
. ACnSTI20,5),TCOST«20,5»
C READ FORMATS FROM EXECUTIVE PROGRAM
C INITIALIZE TO ZERO
-------
PAGE 10
555 CONTINUE
00 5 1-1,20
DO 5 J-l,25
SMATX!I,J)-0.
3 TMATX!I,Jl-0.
DO 10 1-1 ,20
00 10 J-l ,20
OMATXI I,J)«0.
10 O^AIXII,JI-0.
DO 15 1-1,20
DO 15 J-l,5
CCOST(1,J)«0.
cusroii,ji»o.
ACOSTI 1,J)-0.
15 TCUSTII,Jl»0.
C RLAO Irj^UT DATA
READI2.60I (SMATXII,l),t-2,17>
60 FORMAT (Of 10.3/6F10.3)
READ(2,CO» (I)MATXI I ,20», 1-1,161
REAL! 1 2. 1201 K.N, I PROC, (NAME (l).l-1,31,1 SliI $2,OS 1.0S2, MI
120 FGKMAII I 1,IX,I2.2X, 12,1X.3A2.4X,I2,9X,I2,flX,12.8X,I2,28X,lU
SMATXI1 , ISl I«IS1
SMATX I 1 ,OSl MOS1
SMAFXI1,OS2)»OS2
RFAO(2,bO) I DM/MX(1,NI,1-1,16)
RCAOI2,70I IPHNf
70 FORMAril2l
C iHA-i'.Mir icRnr to MICRO THROUGH isz
IS2-IPHNI
WRI If(5,3001
300 FGMMATMH ,//,lM ,< HICROSCREEN MODEL •>
hft ITC IS.dOU NAME(l),NAME(2)(NAME I 3)
80* FORMAT I1H ,3A2,//I
IF (IPHNTI 20,25,20
20 wki re i'.,iioi
310 FCRHAIUH ,27HTRAPPING AND HYDRAULIC DATA)
24 CONTINUE
CALL MICRO
WRI IE 15,3201
J20 FORMAT!IH ,//!
WRirEIS.3301 (1,OHATX(I,N», 1-1,111
WRI IE(5,320)
WH I If 15,326)
326 FOAMAH1M , 10X,' IS1( ,6X,'OSl',6X, •OS2t I
WRI IE I 5, 335 I (l,SHATX(I.ISll.SMATX(1.0Sl),SNATX(ltOS2)i 1-ltlTI
330 FORVATIIM ,6HOMATXI,12,*H,N)«,E10.31
334 FORMAII1M ,3Ml» ,12,2X,Fr.3,2X,F7.3,2X,F7.3»
KRI It(5,3361
336 FORMAT!IH ,' CAP. COST FIX. OP. COST VAR. OP. COST ENERGV ••
«« SCRN. «FP. COST1I
WRITE I 5, 3*0) CCOSTIN,ll.COSTOIN.lI,COSTOIN,2I,COSTO(N,3I,COSTO(N,4
«)
3*0 FORMATI1H .SI2X.E1l.*lI
GO TO 555
1111 STOP
END
-------
UJ
PACE 11
VARIABLE ALLOCATIONS
SMATXUO-7FFE-7C18 TMATX(RCIuiC16-7830 DMATXIRCI-782E-7910 ONATXJRCI-750E-71FO IPHCJ-71EE-71BE
10IIO-718A IS1UU-71B8 IS2MCI-71B6 OS1UCI-71B4 OS2MO-71B2
1AERFIIO-7UE CCOST(RC)-71AC-70E6 COSTOJRCI-70E4-701E ACOSMRO-701C-6F56 TCOSMRCI-6F54-6E8E
III )«0006 JU 1-0008
UNREFERENCED STATEMENTS
till
KII I-OOOA
IPROCII I-OOOC
111(1 I-OOOE
INPMCJ-71BC
NIIO-71BO
NAMEU 1-0004-0000
IPRNTU )«0010
STATEMENT ALLOCATIONS
60 «0030 120 -0036 70
336 -OOA2 340 -COCA 555
FEATURES SUPPORTED
IOCS
• 00*9
-OOEA
300
5
• 00*8
• OOFO
804
10
-005C
•0128
310
15
CALLED SUBPROGRAMS
MICRO FLO FSTO
STOP SDFtO
REAL CONSTANTS
.OOOOOOE 00-001A
INTEGER CONSTANTS
1«001C 20-0010
CORE REQUIREMENTS FOR
COMMON 4*66 VARIABLES
END OF COMPILATION
// XEO
FSTOX FLOAT CAROZ SREO
SWRT
2S-001E
S-001F
2-0020
26 PROGRAM
650
• 0063
• 015F
320
20
•0075
•0220
326
25
•007A
•0224
330
1111
• 0069
•02A2
335 -0095
SCOHP SF10
SIOFX SIOIX SIOI
SUBSC PRNZ
17*0021
16*0022
3-0023
11-0024
0-0025
MICROSCREEN MODEL
U 1
DRUM POOL CONCENTRATION- 57.4199
SEGMENT DATA
SEC VI PV VIN VC/VI DEFN
0 0
.OOOE 00
1 0.927E-04
2 0
3 0
4 0
5 0
6 0
7 0
a o
AVG.
STD.
.1236-03
.147E-03
.168E-03
.186E-03
.202E-03
.216E-03
.230E-03
OIAM.
DEV.
0.690E 03
0.778E 04 0.177E 00 0.277E 00 0.163E 01
0.105E OS 0.237E 00 0.282E 00 0.160E 01
0.128E OS 0.283E 00 0.28SE 00 0.158E 01
0.148E 05 0.322E 00 0.288E 00 0.155E 01
0.166E 05 0.356E 00 0.290E 00 0.153E 01
0.183E 05 0.387E 00 0.293E 00 0.152E 01
0.199E 05 0.415E 00 0.295E 00 0.150E 01
0.214E 05 0.441E 00 0.297E 00 0.149E 01
DRUM POOL RETAINED EFFLUtNT
5.21 13.73 3.46 MICRONS
2.39 1.67 1.89
EFFIC
0.290E
0.299E
0.306E
0.312E
0.317E
0.322E
0.326E
0.331E
00
00
00
00
00
00
00
00
PV2
0.304E
0.417E
0.509E
0.589E
0.662E
0.728E
0.791E
0.649E
02
02
02
02
02
02
02
02
PVI
0.279E
0.558E
O.B38E
0.111E
0.139E
0.167E
0.195E
0.223E
00
00
00
01
01
01
01
01
POR
0.150E
0.1SOE
0.150E
0.150E
0.150E
0.150E
0.150E
0.150E
00
00
00
00
00
00
00
00
vc
0.2S7E-04
0.349E-04
0.422E-04
0.48SE-04
0.541E-04
0.593E-04
0.640E-04
0.685E-04
-------
OlS
2.02
7.32
INFLUENT DRUM RESIDUAL RETAINED EFFLUENT
TOTAL MASS 0.215E 02 0.426E 02 O.OOOE 00 0.126E 02 0.215E 02 LB/OAY
OlAHIHtCRON) 0.3
DRUM PSD 0.0
RETAINED PSO 0.0
EFFLUENT PSD 0.0
PCI. REMOVAL 0.0
SEGMENT
THICK. (MICRONS!
POROSITY 0
0.6
0.6
0.0
0.9
0.0
1
0.30
.ISO 0
1.2
5.0
0.0
7.1
0.0
2
0.41
.ISO
2.5
20. 5
0.0
29.2
0.0
4
0.57
0.150
5.2
49.9
0.0
71.1
0.0
6
10.
79.
30.
100.
31.
0.69
0.150
0.
5 21
* 94
9 63
0 100
2 99
8
0.80
150
.1
.9
.1
.0
.9
42.
99.
97.
100.
99.
5
3
7
0
9
as. 6
99.9
99.8
100.0
99.9
172.3
100.0
100.0
100.0
100.0
OMATXI 1,NI- 0.400E 01
r OMATXI 2, Hi- 0.420E 00
O OMATXI 3.N1- 0.228E 01
OMATXI 4,N)> 0.350E 01
OMATXI 5.NI« 0.300E 02
(JMATXI 6.N1- 0.160E-01
OMATXI 7,M- 0. 135E 01
UHAIXI B ,NI« 0. lOOt 01
CJMMXI 9.NI- 0.971E 02
OMATXI 10, N>- 0.717E 01
OMATXI ll.N)» o
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
CAP
0.
ISl
1.000
0.069
1.000
2.000
3.000
0.000
0.000
0.000
0.000
29.000
1.000
2.000
3.000
0.000
0.000
0.000
0.000
. COST
1033E 05
.297E 00
OS1
3.000
0.066
1.000
2.000
3.000
0.000
0.000
0.000
0.000
29.000
1.000
2.000
3.000
0.000
0.000
0.000
0.000
FIX. OP.
O.lOOiE
OS2
4.000
0.002
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
1.000
2.000
3.000
0.000
0.000
0.000
0.000
COST VAR. OP.
04 0.6661E
COST ENERGY
04 0.3926E 04
SCRN. REP. COST
0.1409E 03
HICROSCREEN MODEL
-------
U 2
DRUM POOL CONCENTRATION- 91.0999
SEGMENT DATA
SEG VI PV V1N VC/Vt DEFN 6PP1C
0 O.OOOE 00
1 0.
2 0.
3 0.
4 0.
5 0.
6 0.
7 0.
a o.
AVG.
STO.
015
105E-03
UbE-03
181E-03
20UE-03
2316-03
2526-03
272E-03
289E-03
OIAH.
OEV.
0.690E 03
0.9306 04 0.10TE 00 0.541E 00 0.858E 00 0.553E
0.1336 OS 0.151E 00 O.S46E 00 0.8346 00 0.563E
0.1656 OS 0.184E 00 O.SSOE 00 0.816E 00 0.5T1E
0.193E 05 0.211E 00 0.553E 00 0.600E 00 O.ST8E
0.2196 OS 0.23SE 00 O.S5&E 00 0.787E 00 0.5B4E
0.243E OS 0.2576 00 O.SS9E 00 0.77SE 00 0.5896
0.265E OS 0.276E 00 0.5616 00 0.764E 00 0.595E
0.286E OS 0.29<,E 00 O.S63E 00 0.7S3E 00 U.bOOE
DRUM POOL RETAINED EFFLUENT
9.83 19.97 3.93 MICRONS
2.82 1.99 1.84
3.18 10.42 2.02
00
00
00
00
00
00
00
00
PV2
0.499E
0.721E
0.89SE
0.104E
0.117E
0.1306
0.141E
0.152E
02
02
02
03
03
03
03
03
PV!
0.466E
0.972E
0.14SE
0.194E
0.243E
0.291E
0.340E
0.309E
00
00
01
01
01
01
01
01
0.
0.
0.
0.
0.
0.
0.
0.
POR
I50E
150E
1SOE
1SOE
1SOE
1SOE
1506
1SOE
00
00
00
00
00
00
00
00
ve
0.5T1E-04
0.812E-04
0.997E-04
0.1156-03
0.1296-03
0.141E-03
0.1526-03
0.163E-03
INFLUENT DRUM RESIDUAL RETAINED EFFLUENT
TOTAL MASS 0.193E 02 0.230E 02 0.923E 01 0.130E 02 0.100E 02 IB/DAY
OIAMIMICRONI
DRUM PSD
RETAINED PSD
EFFLUENT PSD
PCT. REMOVAL
0.3
0.0
0.0
0.1
0.0
O.S
0.6
0.0
1.5
0.0
1.8
5.0
0.0
11.4
0.0
4.2 9.8 22.5 51.8 119.0 273.3 627.9
20.5 49.9 79.4 94.9 99.) 99.9 100.0
0.0 11.3 63.5 91.1 98.7 99.9 100.0
47.0 100.0 100.0 100.0 100.0 100.0 100.0
0.0 21.6 100.0 100.0 100.0 100.0 100.0
SEGMENT
THICK.(MICRONS)
POROSITY
12466
0.67 0.95 1.35 1.66 1.92
0.150 0.150 0.150 0.150 0.150
OMATXI 1,N)-
OMATXI 2.NI-
OMATXI 3,N)«
OMATXI 4,N)>
OMATXI 5,N)>
OMATXI 6.NI-
OMATXI 7,N)«
OMATXI 6,N)«
OMATXI 9,NI-
OMA1XI1I ,N)»
0.400E 01
0.450E 00
0.342E 01
0.17SE 01
0.300E 02
0.160E-01
0.1176 01
O.S21E 00
0.973E 02
0.661E 01
0.563E 00
IS1
OS1
OS2
-------
1
2
1
4
i
6
7
8
9
10
11
12
13
14
15
16
17
CAP
0.
I.
0.
1.
2.
3.
0.
0.
0.
0.
26.
1.
2.
3.
0.
0.
0.
0.
. COST
9931E
000
089
000
000
000
000
000
000
000
000
000
000
000
000
000
000
000
FIX
04 0
3
0
0
1
1
0
0
0
0
13
1
2
J
0
0
0
0
.
. 1
.000
.086
.521
.043
.565
.000
.000
.000
.000
.S64
.000
.000
.000
.000
.000
.000
.000
OP.
09bE
4
0
35
70
105
0
0
0
0
475
1
2
3
0
0
0
0
COST
04
.000
.002
.050
.101
.152
.000
.000
.000
.000
.463
.000
.000
.000
.000
.ouo
.000
.000
VAR.
0.8
OP. COST ENERGY
I41E 04 0.1935E
SCRN. REP. COST
0.1214E 03
to
-------
1
Accession Number
w
5
Organization
ry I Subject Field & Group
$5D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
Engineering-Science, Inc., Cincinnati, Ohio 45226
Title
Investigation of Response Surfaces of the Microscreen Process
10
Authors)
Shea, Timothy G.
Males, Richard M.
16
Project Designation
EPA, WQO Contract No. 14-12-819
2] Noto
22
Citation
23
Descriptors (Starred First)
*Tertiary Treatment, *Fliters, Pilot Plants, Computer Models
25
Identifiers (Starred First)
*Microscreens, *Microstrainers, Treatment Process Design
27
Abstract
Field,
laboratory, theoretical, and state-of-the-art studies were conducted with
regard to utilization of microscreens for tertiary treatment applications. Field
studies were conducted with two pilot microscreen units, using a variety of screen
sizes and types, for activated sludge, trickling filter, and oxidation pond effluents,
Particle size distribution of the effluents (microscreen influents) were found to be
the key characterizing parameter in determination of treatment effectiveness. Overall
effectiveness of solids removal was low, and is ascribed to deficiencies in micro-
screen design practice for the transfer of screened solids from the screen to the
backwash system and out of the microscreen unit.
A computer model of the process was developed in a format compatible with the
EPA Executive Program for Optimization of Treatment Systems.
Abstractor
__JJiQQ±hy-G-
IrtNtitufion
fl r)C
:|OZ (REV. JULY 19»i9)
END, WITH CCTPY OF DOCUMENT
: WftTEH^ESCURCES SCIENTIFIC INFORMATION CENTER
U.S. DEPARTMENT OF THE INTERIOR
WASHINGTON. D. C. 20240
CPOJ 197O - 4O7 -891
------- |