EPA-600/2-77-189
September  1977
Environmental Protection Technology Series
                       CONTROL OF  SEWER  OVERFLOWS
                                   BY  POLYMER INJECTION
                                    Municipal Environmental Research Laboratory
                                          Office of Research and Development
                                         U.S. Environmental Protection Agency
                                                 Cincinnati, Ohio 45268

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental Health Effects Research
      2.  Environmental Protection Technology
      3.  Ecological Research
      4.  Environmental Monitoring
      5.  Socioeconomic Environmental Studies
      6.  Scientific and Technical Assessment Reports (STAR)
      7.  Interagency Energy-Environment Research and Development
      8.  "Special" Reports
      9.  Miscellaneous Reports

This report has been assigned to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment,  and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
 This document is available to the public through the National Technical Informa-
 tion Service, Springfield, Virginia 22161.

-------
                                         EPA-600/2-77-189
                                         September 1977
               CONTROL OF SEWER OVERFLOWS

                  BY  POLYMER  INJECTION
                           by

             R.  W.  Chandler  and  W.  R.  Lewis
               Water  Utilities  Department
                     City  of Dallas
                  Dallas,  Texas   75201
                   Grant No.  11020  DZU
                    Project Officers

                      Richard Field
            Storm and Combined Sewer  Section
              Wastewater Research Division
Municipal  Environmental  Research Laboratory (Cincinnati)
                Edison,  New Jersey  08817

                           and

                    Robert L. Killer
          U.S. Environmental Protection Agency
                        Region VI
                  Dallas, Texas  75201
       MUNICIPAL ENVIRONMENTAL RESEARCH LABORATORY
           OFFICE OF RESEARCH AND DEVELOPMENT
          U.S. ENVIRONMENTAL PROTECTION AGENCY
                 CINCINNATI, OHIO  45268

-------
DISCLAIMER
This report has been reviewed by the Municipal Environmental
Research Laboratory, U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the U.S.
Environmental Protection Agency, nor does mention of trade names
or commercial products constitute endorsement or recommendation
for use.
11

-------
FOREWORD
The Environmental Protection Agency was created because of
increasing public and government concern about the dangers of
pollution to the health and welfare of the American people.
Noxious air, foul water, and spoiled land are tragic testimony to
the deterioration of our natural environment. The complexity of
that environment and the interplay between its components require
a concentrated and integrated attack on the problem.
Research and development is that necessary first step in
problem solution and it involves defining the problem, measuring
its impact, and searching for solutions. The Municipal
Environmental Research Laboratory develops new and improved
technology and systems for the prevention, treatment, and manage-
ment of wastewater and solid and hazardous waste pollutant dis-
charges from municipal and community sources, for the preserva-
tion and treatment for public drinking water supplies and to
minimize the adverse economic, social, health, and aesthetic
effects of pollution. This publication is one of the products
of that research, a most vital communications link between the
researcher and the user community.
One source of water pollutants is un-controlled overflows
from sanitary and combined sewers. This report deals with one
possible method for the reduction or elimination of such over-
flows.
Francis I. Mayo
Director
Municipal Environmental Research
Laborary
-Ill

-------
ABSTRACT
In the past, the operator of a sewage collection system has
had three alternatives for dealing with overloaded sanitary sewers;
ignoring them, diverting them to storm sewers and streams, or
pumping to other locations. An EPA-sponsored research program
entitled, “Polymers for Sewer Flow Control,” Contract No. 14-12-
34, suggested a possible alternative system wherein the capacity
of a sewer might be increased by the injection of certain water-
soluble chemicals to reduce turbulent friction. This concept was
further developed and demonstrated during this project, EPA Grant
entitled, “Elimination or Reduction of Sanitary Sewer Overflows
in the Bachman Creek Sewershed,” which was executed in Dallas,
Texas. This report was prepared to help operators of sanitary
sewage collection systems determine the feasibility of using
turbulent friction reduction, designing an injection facility,
choosing a friction reducing material, and evaluating the results.
This report was submitted in fulfillment of Grant No. 11020 DZU
by the Water Utilities Department of the City of Dallas under the
sponsorship of the U.S. Environmental Protection Agency. This report
covers a period from May 1969 to December 1973.
iv

-------
TABLE OF CONTENTS
Disclaimer . ii
Foreword ill
Abstract iv
Table of Contents v
List of Figures Vi
List of Tables Viii
Acknowledgments ix
Sections
1. Conclusions 1
2. Recommendations 3
3. Introduction 4
4. Polymers as Turbulent Friction Reducers 6
5. Selection of Materials for Friction Reduction 23
Applications
6. Equipment for Polymer Injection 37
7. Process Control and Instrumentation for Polymer 56
Injection
8. Precautions in Storing and Handling Polymer 60
9. Evaluation of a System for Potential Use of 62
Friction Reducing Agents
10. Relationship Between Rainfall and Sewer Overflows 68
11. Hydraulic Line Analysis and Computer Modeling 80
12. Preliminary Instrumentation and Flow Measurement 89
13. Results of Polymer Injection in Surcharged Gravity 98
Lines
14. Results of Polymer Injection in 6” Force Main 107
15. The Constructed Injection Station 113
16. On Line Operation of the Injection Facility 122
Appendix A - Friction Reducing Materials Tested 133
for Conformance to Performance Specification
Appendix B - City of Dallas Specification for High 134
Molecular-Weight Water Soluble Friction-Reducing
Additives NO PA-106-4061-70
Appendix C - Sewer Modeling Program 139
Appendix D - Bachman Creek Input Data 142
Appendix E - Computer Output from Modeling Program 148
Selected Bilbliography 163
Metric Conversion Table 169
V

-------
FIGURES
No. Page
1. Velocity Profile for Fluid Flowing in an “Ideal’ Pipe 8
2. Symbol Definitions for Gravity Flow in a Pipe 8
3. Velocity Profile for Fluid in the Laminar Regime 10
4. Velocity Profile for Fluid Flowing in the Fully 10
Developed Turbulent Regime
5. Typical Graph of the Output of a Rapid-Response 11
Flowmeter Measuring Turbulent Flow
6. Graph Showing Typical Relationship Between Flow And 13
Head Loss in a Gravity Sewer With and Without Poly-
mer Addition
7. Element of Fluid Moving in a Pipe Showing Derivation 15
of Shear Stress
8. Graph of the Relationship Between the Logarithmic 21
Profile Intercept Function and Dimensionless
Relaxation Time
9. Diagram of Apparatus For Screening Friction-Reducing 24
Materials
10. Graph of Relaxation Time versus wall Shear Stress for 33
Concentrations of Percol 155 and Polyox WSR-30l
11. Trailer-Mounted Equipment 40
12. Water-Dispersed Injection System 41
13. Modified Injection System 42
14. Light-Weight Variable-Flow Polymer Injection
Apparatus
15. Typical Eductor Construction 48
16. Operating Characteristics of a Typical 1 1/2 Inch 49
Disperser
17. Polymer Clod Separator . 52
18. Suggested Process Control Scheme for Polymer 58
Injection
19. (a) Polymer Injection to Change Head and Flow 65
19. (b) Polymer Injection to Change Flow Only 65
19. (c) Polymer Injection to Change Head Only 65
20A. Location Map 70
20B. Collection System Map 71
20C. Profile of Bachman Creek 18” Line 72
21. Rainfall Record-Bachman Watershed (January-
December 1969) Data From Four Stations
22. Rainfall Record - Bachman Watershed (January-December 78
1970) Data From Four Stations
vi

-------
No. Figures(contiflUed) Page
23. Ratio of Peak to Average Sewer Flows Versus 83
Population (Based, on National Average)
24. Example Flow Network Showing Indexing Convention 85
25. Purge-Tube Level Meter Installed in Conventional Manftle 95
26. A Multiple-Input Recording Station 96
27. Example Recording Showing a Building and Preceding
Storm Flow
28. Storm Stage at Station 164+33 (Bachman Trunk) 101
on March 20, 1970
29. Storm Stage at Station 35+40 (Bachman Branch) 102
on April 25, 1970
30. Storm Stage at Station 164+33 (Bachman Trunk) 103
on April 1970
31. Results of Injection Test 5 on Heads at Stations 104
164+33 and 166+21.58
32. Results of Injection Test 3, 4a, and 4b on 105
Surcharges 35+40
33. Effect of Polymer Injection Test 7 on 23 September 106
1970 on Overflow 1 Station 29+35 (Bachman Branch)
34. Flow Through 6 Inch Pressure line Without Polymer 111
Addition
35. Flow Through 6 Inch Pressure Line With 111
0.74 lbs/mm Polymer Addition
36. Comparison of Results for Laboratory and Field Tests 112
on 6 Inch Pressure Line w/O.74 lbs/mm Polymer Addition
37. Elevation of Polymer Injection Station Locating 114
Major Components
38. Sketch of Main Control Panel Showing the 115
Location of Controls, Meters and Indicators
39. Schematic of Process Control System 117
40. Plan of Injection and Metering Vault Showing l 1
Location of Major Components
41. Graph of Manual Injection Test at Various Feed Rates 124
42. Graph of Manual Injection Test at Constant Feed Rate 125
43. Graph of Automatic Injection Test at Constant Feed 126
Rate-Level controlled at 45 Inches
44. Graph of Automatic Injection Test at Constant Feed 128
Rate-Level Controlled at 30 Inches
45. Graph of Automatic Injection Test with Feed Rate 129
Proportioned to Flow
46. Graph of Automatic Injection of May 1, 1974 131
Polymer Feed Rate Proportional to Flow
47. Graph of Downstream and Local Levels During Automatic 132
Injection of June 9, 1974.Polymer Feed Proportional
to Level
vii

-------
TABLES
No. Page
1. Laboratory and Reduced Data for Polyox WSR-301 at a 25
Concentration of 10 wppm
2. Laboratory and Reduced Data for Polyox WSR-301 at a 26
Concentration of 50 wppm
3. Laboratory and Reduced Data for Polyox WSR-301 at a 27
Concentration of 100 wppm
4. Laboratory and Reduced Data for Percol 155 at a 28
Concentration of 10 wppm
5. Laboratory and Reduced Data for Percol 155 at a 29
Concentration of 50 wppm
6. Physical Constants Utilized for Calculations 30
7. Constants for Use in tne Equation OKr 31
w
8. Results of Comparative Solution Time Tests 44
of Dry and Slurried Polyox WSR-301
9. Results of Comparative Injection Tests of 46
Dry and Slurry Feeders
10. 1970 Rainfall Record 73
11. 1970 Rainfall Record 75
12. Observed Overflows in the Bachman Creek Watershed 79
During Calendar Year 1970
13. Land Use Parameters Used in Model 82
14. Results of Polymer Injection Tests 100
15. Lift Station Pump Data 107
16. Results of Lift Station Tests 108
viii

-------
AC KNOWLEDGMENTS
The accomplishment of the experimental project prerequisite
to the preparation of this type of manual would not have
been possible without a great deal of support and aid from
a large number of people in the Dallas Water Utilities
Department. Special thanks are due to Henry J. Graeser,
Director, for his firm support over a long period.
The work of N. C. Glaze, E. 0. Buch, and the Special Projects
field crews was essential to the successful operation of
the injection facility.
The design of working components of the station, and the
production of graphs, figures, and the text were ably support-
ed by the engineering, drafting and clerical personnel in the
Special Projects Section who withstood the pressure of too
many deadlines and many revisions.
ix

-------
SECTION 1
CONCLUSIONS
The work performed under this demonstration grant has
shown that it is possible to utilize friction-reducing
chemicals to prevent or limit overflows in a working sewer
line by establishing an automatic injection system for the
materials. It has also been shown possible to increase the
capacity of a “package” sewage lift station by injection of a
slurry of friction-reducing polymer into the pump intake.
Three methods of polymer feed control were utilized
in the demonstration work; constant rate, flow proportional,
and level proportional. The first of the three is the
simplest and least expensive in terms of equipment and has
proven to be adequate under most conditions.
Polymer dosages directly from the dry material have
been in the range of 15 to 50 parts per million, considerably
below that used previously in large sewers. This fact
engenders the possibility that polymer injection may be more
economical than previously believed.
Two polymer types from two different manufacturers
have been utilized; Union Carbide Polyox WSR-301 (polyethylene
oxide) and ICI America 4430 (polyacrylamide copolymer). There
are no noticeable differences in friction-reducing properties,
but a considerable difference in mixing properties exists.
Two problems which have limited the usefulness of the
polymer injection facility are instrumentation failures
and polymer lumping. The first problem can only be resolved
by the equipment manufacturers. The second problem,
polymer lumping,h been solved by a re-design of the polymer
feed-funnel to •incorporate a trap for the lumps.
Insufficient data has been obtained to permit the
production of a user’s manual to permit general application
of friction-reducing chemicals.
One type of polymer tested, ICI America 4430 poly-
acrylamide, has a wetted specific gravity greater than
1.0, thereby producing some polymer accumulations near the
1

-------
bottom of the mixing tank. Mechanical agitation with
a slow speed paddle would make the injection rate more
constant.
2

-------
SECTION 2
RECOMMENDATIONS
It is recommended that the station continue to be oper-
ated for the purpose of controlling overflows in the study
area and also provide the additional flow-head loss data
desired.
A slow-speed mechanical stirrer should be added to
the mixing tank to produce a more uniform slurry of the
difficu1t” materials.
3

-------
SECTION 3
INTRODUCTION
SCOPE AND PURPOSE
The City of Dallas has been studying a means whereby
the overflows from a sanitary sewer during periods of wet
weather may be eliminated without resorting to expensive
new construction or alteration. The means being studied
is the addition of friction-reducing chemicals to a sewage
stream in order to reduce the head loss caused by turbulent
friction in the sewer pipe.
In order to demonstrate this phenomenon, a complete
drainage area, the Bachman Creek watershed, was chosen
as the study area under EPA Demonstration Grant 11020
DZU, with the goal being the elimination of overflows by
the “as—required t automatic injection of the friction-reducing
chemicals.
This report presents data gathered during the conduct
of the program, between November 1, 1969 and April 15, 1974.
BACKGROUND
The Bachman Creek trunk sewer is a branched flow system
consisting of approximately 45,000 feet of trunk line,
700,000 feet of collector lines (not including house laterals)
which serves a drainage area of about 8,000 acres. The
main trunk lines follow natural drainage channels and in
most locations consist of unreinforced concrete pipe embedded
in concrete poured in channels cut in the limestone bedrock.
Access to the trunk lines is limited to personnel on foot
except at those locations where a street crosses the line
route.
During periods of heavy rainfall, water from either
illegal connections, inflows, or infiltration enters the
collection system to such an extent that the ultimate
capacity of the line is exceeded. These excess flows are
relieved through manholes and overflows constructed to
protect the property owner along the trunk.
4

-------
Because of the relative inaccessibility of the lines,
and the fact that the existing line is adequate to carry
normal dry weather flows, an alternative to reconstruction
or construction of relief lines was preferable. Elimination
of the entry of the wet weather flows into the system
is the ultimate solution; an on-going infiltration/inflow
abatement program is approaching this permanent solution
by inspecting, re-engineering, repairing, and replacing
the sewers in the area. However, the elimination of excess
flows is a time—consuming process, and elimination of
overflows in the interim period is necessary.
The background experimental work for the present program
was reported in EPA Report, “Polymers for Sewer Flow Control ,‘
fl020 DIG, August, 1969. A comprehensive bibliography
of other publications concerning the theory and practice of
friction reduction is included in this report.
PROGRAM DESIGN
The program was divided into four phases, each of which
generated data used as input in succeeding phases. The
four phases were:
A. Study Phase - Instrumentation, analysis of the
drainage system, computer modeling, and preliminary
injection tests.
B. Design — Design of the injection station and its
ancillary equipment.
C. Construction - Construction and check-out of the
equipment.
D. Demonstration and Operation - Demonstration of
polymer effectiveness, analysis of results for
application elsewhere, and preparation of maintenance
and operation documentation.
5

-------
SECTION 4
POLYMERS AS TURBULENT
FRICTION REDUCERS
HISTORY OF TURBULENT FRICTION REDUCTION
In 1948 a Dutch researcher named Toms noted that certain
chemicals dissolved in water altered the flow characteristics of
the fluid in a manner which could not be explained using classic-
al mathematical techniques. Upon further investigation he found
that the results obtained during the measurement of viscosity
varied with the rate at which the fluid was sheared. That is,
the viscosity was not a constant ratio of shear stress and shear
rate.
This phenomenon, which later became known as the Toms’
Phenomenon, remained a laboratory curiosity until it was “re-
discovered” by companies working with fluids used for the stimu-
lation of oil wells. Researchers using certain natural gums
with the generic name “guar gums” for viscosity control found
that a dilute solution of these gums in water exhibited lower
friction losses in pumped systems thereby increasing the effi-
cienty of such systems.
Because a guar gum was a natural polymeric material, a
dramatic research effort in polymer chemistry was begUn to dis-
cover even more efficient man-made materials.
Having successfully applied polymers to oil-field friction
reduction, personnel at the Western Company of North America began
looking for other areas in which friction losses were significant.
Three additional applications of the phenomenon were found; the
reduction of friction and noise for submerged projectiles
(torpedoes), the reduction of friction on the hull of fast war-
ships and the augmentation of flow capacities in pipes used for
product transport, including sanitary sewage and stormwater.
It is this last and most difficult application which is the
subject of this manual.
6

-------
THE NATURE OF TURBULENT FRICTION
Since most problems in sanitary collection systems arise
when sewer pipes cease behaving as open channels, that is, when
the cross section of the pipe is completely filled with the
flowing fluid, the discussion of turbulent friction will be lim-
ited to full-pipe flow. In addition, since laminar flow so
rarely occurs in practical sewer systems, flow in a laminar
regime will be largely ignored except as a basis for discussion.
If one imagines a “perfect fluid system, that is a pipe
and fluid which is completely devoid of friction or other dis-
turbing influence, each particle of fluid will move through the
pipe exactly parallel to the walls of the pipe. This type of
flow is illustrated by Fiqure l In this ideal system, the
fluid molecules do not coilioe or interact with each other or
with the walls of the pipe and there will be no energy transfer
within the confines of the pipe. The velocity of every parti-
cle will be exactly equal to that of every other particle and
the velocity profile will be shown in Figure 1. If we introduce
energy into the fluid in this situation oy sloping the pipe, as
in Fiaure 2, it is easy to see that all the potential energy
available will be converted to kinetic energy and we can write
the specific energy equation.
Potential Energy = = Kinetic Energy (1)
Where t Z difference in height of the enas of the line referred
to datum
V=velocity in fluid.
g= gravitational constant.
If the fluid particles are allowed to interact with each
other and the pipe wall, but the particles are still required to
travel parallel to the centerline of the pipe, we arrive at a
flow which is illustrated by Fiaure 3. At the pipe wall, which
is motionless, the fluid in cbrtta t with the wall is also motion-
less and the velocity of particles increases with increasing
distance from the pipe wall until a maximum is reached at the
pipe centerline. This condition approximates laminar flow.
Under these conditions, there is triction loss, heat is generated
throughout the fluid and at the pipe wall. This heat raises
the temperature of the pipe and the fluid, and is not available
for moving the fluid. Referring again to Figure 2, we must now
write the specific energy equation as:
7

-------
Figure 1. Velocity Profile for Fluid Flowing in an”Ideal” Pipe
(friction factor = 0)
Figure 2 . Symbol Definitions for Gravity Flow in a Pipe
Flow In
Flow Out
L
8

-------
Potential Energy= z Z= h +V 2
L7g (2)
Where hLis head loss in the same units
as E Z.
The symbol h represents the energy lost as heat. The
energy loss due tb friction under laminar conditions is the
minimum possible in a practical fluid system, and is generally
used as a reference value for determining the effectiveness of
friction reduction. Energy losses in laminar flow are largely
due to internal friction and increase in direct proportion to
the viscosity of the fluid. For the purposes of discussion,
these losses will be referred to as viscous friction.
The two cases of flow discussed above required that all of
the fluid particles travel in straight parallel lines. In prac-
tical pipe flows it is found that this is rarely the case. In-
stead, when the fluid is viewed on a microscopic level, the
particles appear to be in random motion, colliding with each
other and the pipe wall. This is the condition which is called
turbulence. When inspected on a microscopic scale the velocity
profile can be illustrated as in Figure 4 . As in the previous
case, the velocity of the individual fluid particles at the pipe
wall is zero. However, unlike the previous case, higher velo-
cities are found much nearer the pipe wall, and the central core
of the flow exhibits a relatively uniform velocity.
The collisions that occur in this flow state generate heat at
a higher rate than our laminar model because now we have more fre-
quent contact between particles and the pipe wall and there are
particles with velocities which are in directions which oppose
the flow. In fact, a flow meter which responds rapidly to flow
velocity will indicate as shown in Figure 5. Therefore, it may
be concluded that turbulent losses are largely due to inertial
effects.
If we write our specific energy equation for the turbulent
case, and assume that one may differentiate between heat losses
caused by viscous friction and turbulent, or inertial friction,
we have:
Potential Energy = L Z = (hL + hT) + V 2
2g (3)
Where: The symbol h represents the energy lost becauze of
turbulence. To simplify calculations the term in parenthesis
generally written as hf where:
hf = hL + hT = Total friction loss (4)
9

-------
Figure 3.
Velocity Profile for Fluid in the Laminar Regime
(friction factor > 0)
ti,
Figure 4.
Velocity Profile for Fluid Flowing in the Fully-Developed
Turbulent Regime
(friction factor > 0)
10

-------
F low
Indicated
F low
Time
Figure 5. Typical Graph of the Output
of a Rapid-Response Flowmeter
Measuring Turbulent Flow*
*Rouse, Hunter (ed.). Engineering Hydraulics. New York, John
Wiley and Sons. Inc., 1950. p. 86.
11

-------
The purpose of writing two terms is to emphasize the fact that
the term “friction reduction” applies only to turbulent friction;
that is, those losses attributable to inertial effects.
EFFECT OF POLYMERS ON ENERGY LOSSES
A great number of researchers have attacked the problem of
determining the reasons that polymers dissolved in a fluid reduce
turbulent friction losses. This research has resulted in many
theories which range from the attenuation of turbulent eddies to
the thickening of the laminar sublayer along the fluid-pipe inter-
face. There have also been a number of attempts to predict
the behavior of any material as a friction reducer, but the prob-
lem has proven so complex that practical applications of the phe-
nomenon have relied almost entirely on the results of experiment-
ation. It is for this reason that the writer will make no great
effort to explain the mechanism of friction reduction; only the
gross effect will be discussed.
The polymers used for friction reduction in water are obviously
water-soluble, but they also have a number of other characteris-
tics, which govern their behavior. First, the polymers have very
high molecular weights; the more efficient materials have mole-
cular weights in the range of 4,000,000 to 8,000,000. For com-
parison, the molecular weight of water is only 18. Some exper-
imenters have had some success with materials with molecular
weights as low as 500,000, but in general, these materials are
less effective as friction reducers. A second necessary chara&
teristic of the polymer materials is that their length to diamet-
er ratio be large. There are many materials with high molecular
weights which have low LID ratios. More simply, not all poly-
mers act as friction reducing agents. In fact, many high mole-
cular-weight polymers have the opposite effect because they
drastically increase the viscosity of the water and thereby in-
crease viscous friction.
The logarithmic graph of Figure 6 is a typical example used
for explanation.
The abicissa of this graph represents the total head loss for a
given length of pipe, and the ordinate represents the flow rate
through the pipe. The lower curve on the graph is typical of a
head-discharge relationship for a gravity sewer pipe of moderate
age, and can be represented by a power-law equation, that is:
Q=kt h (5)
Where: Q flow In convenient units
K = constant Including length, diameter, and friction
factor
12

-------
4 —
Ij i i 2
1 jii :
- I HHiiT :f 1
10 20
- - - -
30 40 50 60
Head Loss (tth in feet)
Graph Showing Typical Relationship
between Flow and Head Loss
in a Gravity Sewer
with and without Polymer Addition
13
10
9
8
7
6
1
1
Figure 6.

-------
Ah = head loss in convenient units
n = empirical exponent
If n = 0.5 the equation is similar to the Darcy-Weisbach equation.
2
= f L V
(6)
and the friction factor, f, is a function of Reynolds’ Number.
If n = 0.54 the equation resembles the Hazen-WilliamS equation
for flow in fulipipes. In the graph shown, n 0.44. Since
this coefficient cannot be found in any classical pipe flow for-
mula, it Indicates that the example pipe exhibits loss character-
istics which cannot be reconciled with flow design equations.
Measurements of flows and heads showed that this apparently
anomalous behavior was the rule rather than the exception for
sewer pipes in Dallas. The importance of determining the head-
discharge relationship for a pipe before attempting friction
reduction cannot be over-emphasized; if this relationship is
not known the effect of friction reduction cannot be predicted
or evaluated.
The upper curve in Figure 6 represents the effect of poly-
mer addition to the flow in the pipe. This curve is typical of
a moderately good friction-reduction polymer. Note that for a
given head loss the flow is increased, or, conversely, for a
given flow the head loss is decreased. In practical pipe
systems, both effects usually occur simultaneouslY. It should
also be noted that at the lower head losses, in this case be-
low approximately 22 feet of water head, the apparent friction
reduction is decreased, and the effect of polymer addition
disappears at a head of about 17.5’ of water. This fact is
typical of all friction reducing agents.
Figure 6 is also typical of the behavior of friction re-
ducing polymers in that the graph of the sewage flow with
polymer addition is almost parallel to the graph of the sewage
without polymers. This is true in the range of velocities
practically obtainable in a gravity system. In pumped systems,
however, it is easily possible to produce turbulence of such
great magnitude that the physical structure of the polymer
molecule is destroyed. When this occurs the two lines will con-
verge at the upper end. The normal term applied to this phe-
nomenon is “shear degredation.”
By-passed in the discussion above was the reason for low
friction reduction at low head losses. This is explainable in
two parts; “onset shear stress” and “shear dependence of fric-
tion reduction.” Studies by various researchers have shown
that the polymers used as friction reducers do not become
14

-------
effective until some minimum shear stress at the fluid-pipe wall
is reached. This minimum shear stress is called the “onset
shear stress” and is a property of the particular polymer mole-
cule. The shear stress at the wall of a pipe can be calculated
using the geometric properties of the pipe and the head loss at
any flow rate.
Figure 7 illustrates the derivation of wall shear stress.
Consider a volume of any fluid of length L, bounded by a pipe
of diameter 0. On the upstream face there is a head of h
1-L h acting, and on the downstream face a head of h. The
following equation may be written:
(h+L h) ( D )- (h) (icD 2 )= (T )
4
Where = shear stress at the pipe wall,
Reducing the equation,
(irDL) (7)
and
It should be remembered
the flow in the pipe as
(t h) (1T) = T (rrDL)
4
4L
that Lh has
defined.
(8)
(9)
a specific relationship to
Figure ..
Element of Fluid Moving in a
Pipe Showing Derivation
of Shear Stress (T)
B
p+Ap
D
L
15

-------
A superficial inspection of the expression would indicate that
the wall shear stress increases in proportion to pipe diameter.
However, for a given flow the head loss (t h) decreases with
increasing pipe diameter. In fact, the head loss decreases
at a rate proportional to the fourth power of the inverse ratio
of the diameters of the two pipes, if the pipes have equal
roughness and a relationship such as the Darcy-Weisbach equat-
ion is assumed. That is:
Assuming two pipes of diameters 0 and Dof equal length L
Where D > D
2 1
and K K
i_ 2
D
1 2
2
then L h — KV
and hKV (Darcy-Weisbach Equation) (10)
for a given flow Q,
V = 4Q and V .L
irD 2 rD 2 (11)
1 2
thenV 2

V D (12)
1 2
Hence t, h V 2 D 2 2 D k
2 =( ) =( ( ) )( ) (13)
E h V — 0
1 1 2 2
The wall shears in the two pipes of different diameter can
then be related.
(it) 0 h 0 k
(T. )) L h, = (ç) =( •;-) (14)
The above expression is important to a user, since it becomes
obvious that a material which acts as a friction reducer in
one pipe may not work in another pipe because the onset shear
stress never occurs.
The shear dependence of friction reduction can be under-
stood by considering the region between point “A” and point “B”
on Figure 7. It is obvious that as the onset shear stress
16

-------
i Point “A”) is exceeded and friction reduction occurs, the
magnitude of the friction reduction Increases until some maxi-
mum is reached (Point “B”). Once this maximum occurs further
increases in shear stress have little or no effect on friction
reduction efficiency.
Therefore, for specific polymer solution in a specific
pipe, there is a range of flows in which there will be no fric-
tion reduction, a range of flows of which the efficiency of
friction reduction will be related to the flow rate, and there is
a third range over which the friction reduction efficiency is in-
dependent of the flow rate. Obviously, it would be desirable
to always operate in the most efficient range. However, in
gravity sewer systems this will not always be possible, and as
has been discussed above, there can be flow and head loss con-
ditions under which friction reducing techniques cannot be
appl ied.
MAGNITUDE OF FRICTION REDUCTION EFFECT
In the previous section it was pointed out that under cer-
tain conditions the addition of polymers to a gravity sewer
line can cause either increased flows, decreased head losses,
or both effects can occur together. No mention of the mag-
nitudes of these effects has been made thus far. This section
discusses the definition of friction reduction, the theoretical
limits of friction reduction, and the probable maximums which
can be expected under field conditions.
The definition of friction reduction utilizes the concept
of friction factor reduction, and bases a percentage change on
a comparison of friction factors in turbulent flow with the im-
aginary extension of the laminar friction factor graph. In the
form of an equation, percentage friction reduction is:
- f )x 100 (15)
Where p percentage friction reduction
friction factor of fluid in a pipe
f — friction factor of fluid with friction-reducing agent
— added
f projected laminar friction factor
1 =
R = Reynolds number (constant)
17

-------
The laminar friction factor is defined DY
f — 64
1 (16)
Where R Reynolds number
At the Reynolds Numbers of fully developed turbulent flows in
large pipes, the quantity represented by f is sufficiently small
as to be negligible. The friction reducti n equation can be re-
duced to:
= ( - —) x 100 (17)
For any specific pipe at a given flow velocity, head loss measure-
ments can be used to arrive at friction reduction efficiencies
directly:
l-t h) x 100 (18)
Where ip percentage friction reduction
h. head loss with polymer added
ht_ head loss without polymer *
This form of the expression has previously been used by Savins.
It is obvious then that the prediction of a friction factor
is the only step necessary in order to predict a friction reduct-
ion efficiency and the head loss-flow relationship. One would
then be able to engineer a solution to a friction-reduction prob-
lem. However, the prediction of the appropriate friction factor
is the stumbling point of the technology, since the friction
factor of a dilute polymer solution is a function of the follow-
ing parameters:
1. polymer characteristics
2. wall shear stress
3. velocity of flow
4. diameter of pipe
Seyer and Metzner**have suggested a form for the friction
factor equation as follows:
ayins, J, 6. kSt’r ss-Control1ed Drag ReductiOfl Phefl0mefbt
Rheologica Acta. (Darmstadt). 6:4, 1967
**Seyer, F.A. and A.B. Metzner, Drag Reduction in Large Tubes
and the Benavior of Annular Films of Drag-Reducing Fluids, Can-
adian Journal of Chemical Engineering. (Ottawa). 47:, Dec. 1969
18

-------
1 ( 1 _ )2 [ Am ‘ ) = B (8)-Am 2v’2 ]-G
212 (19)
Where f = Darcy friction factor
A= slope of the logarithmic velocity profile at the lamin-
ar sublayer boundary
R= Reynolds Number
B (e) = intercept function for logarithmic velocity profile
0= dimensionless relaxation time of polymer molecule
G= Empirical function, approximated by G 3.0 for
design purposes
E= dimensionless distance trom pipe wall y/r (r= radius of
pipe, y= distance from pipe wall)
The above equation reduces to the Nikuradse equation for smooth
pipes for a Newtonian fluid for which B (0) = 5.6., 0, A
2.46, and G 3.0. However, it is not possible to use the above
equation without a great deal of information including the def-
inition of , y, and B (0). These terms are interrelated by
additional equations. The parameter , is defined as the
ratio of the thickness of the laminar sublayer (near the pipe
wall) to the radius of the pipe. The value of ‘ ranges from
almost zero to about 0.2 and can be determined from the relation-
ship:
B (8) = (y x /r IPX 1) - Am (y x ITCA/p x ]_ 20
VI
and
- (21)
Where y = distance from pipe wall of the intersection of the
linear and logarithmic profile approximations
T fluid shear stress at the pipe wall
p = fluid density
v = apparent kinematic viscosity
r = radius of pipe
The function B (0) is a characteristic of the polymer solution
being considered and the w311 shear stress, but Seyer and Metzner
have postulated that the function is identical for all non-
19

-------
Newtonian polymer solutions. Their plot of B(O) as a function
of 0 is shown in Figure 8 . Assuming that their supposition is
correct, all that remains in the solution of a friction reduction
problem is to determine the relaxation time, 0, for the polymer
solution under the flow conditions of interest.
Previously mentioned was the fact that dilute polymer solutions
generally behave according to a “power law”, that is, the shear-
ing stress in the fluid is proportional to the shear rate raised
to some power which will range from 0.0 to 1.0 for “real world”
fluids. Therefore, it can be shown that the relaxation time,0,
is proportional to the shear stress, or:
K (22)
Where e = relaxation time
K constant of proportionality
= fluid shear stress at the pipe wall
n = a value to be evaluated by experiment
The suggested method for the evaluation of a potential
friction reducing material is to run small scale experiments
in which head loss and flow may be accurately measured for the
polymer concentration of interest. A friction factor may then
be calculated and inserted into Equation (19) along with the
appropriate Reynolds Number. As a first approximation the
dimensionless number may be assumed equal to zero. A value
of B (0) can then be determined. Using the graph of Figure 8, a
corresponding value of 0 car then be estimated. A logarithmic
plot of 0 versus t may then be constructed, and the value of
the constants K an n in Equation (22) computed. In order to
refine the procedure, a few iterations through Equations (19),
(20), and (21) will suffice to determine the value of
If the procedure outline above is repeated at a number of
different polymer concentrations, it will then be possible to
graph or tabulate the various parameters necessary to predict
the friction factor and hence the friction reduction efficien-
cy for the flow condition of interest.
As an estimate of the maximum friction reduction possible
with water-soluble polymers, the maximum and minimum values
which the function B (0) can take may be applied directly in
Equation (19) fo a typical flow condition. Using a Reynolds
Number of 1 x 10 which approximates the flow of water at 70°
Farenheit through a 24 inch diameter pipe at 5.3 feet per second,
the maximum and minimum values of the friction factor are:
Maximum f = 0.0112 Minimum f = 0.0030
20

-------
CD
0
•r•4
N)
U
Relaxation Time, 0
Figure 8.
Graph of the Relationship Between
The Logarithmic Profile Intercept Function
and Dimensionless Relaxation Time
-f-I -t ft I
______ H
ii
nhiin- t
4
40
30
20
10
C)
:t ±ffl Li ±
4111 fTITJTUt
ii
4-
H L
iTti
- f
- -
+ 1-
ii
Ij-
-4-
+
44
-t —
U
1-I
0 5 10 15
fl 1
20
30 35

-------
From Equation (17) the maximum friction reduction efficiency
for the case cited is then equal to 73 percent. Studies have
indicated that the maximum possible friction reduction is in the
vicinity of 80 percent, so that a material which would behave
as in the example above would be very satisfactory. Unfortunate-
ly, most successful polymer applications in gravity systems
have shown maximum efficiencies in the range of 40 to 50 percent.
CALCULATIONS OF REAL PROBLEMS
Equations 18, 19, 20, 21 and 22 as presented in the pre-
ceding discussion work well when applied to hhidealu pipes inthe
laboratory. However, some problems arise when these equations
are translated to practical applications; therefore a modific-
ation is indicated. Since the equations have been fitted to
empirical models of velocity profiles during their development,
these needed modifications are logically related to empir-
ical information gathered in full-scale field tests, and al-
though they have been checked against field-gathered data,
additional testing will be necessary before they will become
a standard of the industry.
The major problem encountered is one mentioned previously
concerning the failure of actual gravity sewers to perform
according to the textbook formulas. In checking Equation (18),
it was found that although it becomes the Nikuradse equation
when the proper constants are used, the Nikuradse equation did
not predict the friction factors actually measured. For this
reason it was necessary to first “fit° the equation to the en-
countered Newtonian flow conditions before it could be used
to predict Newtonian parameters.
Re-writing Equation (19) in a reduced form we arrive at the
following:
,. = (.3536) (l_ )2 [ 2.46 in ( RTf ) +B (e) ] -G 23
5.6568 2/2
The symbols are as previously defined. If it is assumed
that B (0) is a unique function of the polymer material and
concentration, and hence of relaxation time, then its value
in the equation is fixed, as is the Reynolds Number of the
flow (neglecting small changes in solution viscosity). How-
ever, if we inspect the constant symbolized by “G”, it is
found that the friction factor may be adjusted to fit data
by varying the value of G. Hence, if the characteristics of
a pipe transporting sewage can be measured so that tne fric-
tion factor and Reynolds’ Number for a few conditions are
determinate, a value of G may be picked which will satisfy the
data. The new value may then be utilized to predict the
effect of friction reduction on that particular pipe. One
caution in this respect Is that it Is conceivable that “G”
could become a function of Reynolds Number in some cases.
22

-------
SECTION 5
SELECTION OF MATERIALS FOR
FRICTION REDUCTION APPLICATIONS
Section 4 of this report dealt with the problem of
predicting the effect of polymers on a sewer line, and, as
an adjunct, discussed a method for extrapolating laboratory
data to field problem scale. This section presents typical
data that might be generated in a laboratory,acconipanied by the
extrapolated data described in the preceding discussion.
During the course of the friction-reduction investigation
in Dallas, thirty polymer products from eight different
manufacturers were thoroughly tested in the laboratory to
determine their friction-reducing properties. Of the
materials tested, eighteen were considered efficient enough
for serious consideration. The screening tests were performed
in an apparatus illustrated by Figure 9. The test sections
which consisted of tubing of various diameters were very
carefully fabricated to give conditions as near ideal as
possible. The tests were performed as outlined below.
A sample of the test material was dissolved in de-ionized
water in the manner prescribed by the manufacturer in the
proportions required to give the required concentration of
active friction reducer. This solution was gently agitated
for a sufficient length of time to insure a clear solution
with no lumps or “fish-eyes ’. Materials were tested immediate-
ly after solution agitation. Made-up solutions were not tested
more than once or retained longer than 30 hours.
Two gallons of the solution prepared as above were placed
in the pressure vessel with valve (F) in closed position.
With the pressure vessel open to atmospheric pressure, valve
(F) was then opened slightly until the tube (C) and fittings
were purged of air bubbles. Valve (F) was then closed.
Static pressure was built-up in the pressure vessel
by means of an auxiliary pressure regulator and air or
nitrogen source. The applied pressure was adjusted between
10-160 psi.
23

-------
Pressure Regulator
Air Supply
Precision Differential
Pressure Gauge
Pressure
Vessel
Valve F
Figure 9
Diagrairi of Apparatus for Screening Friction-Reducing Materials

-------
TABLE 1
*LABORATORY AND REDUCED DATA
FOR POLYOX WSR-301 AT A
CONCENTRATION
OF 10 wppm**
*Data taken in a 0.18 inch diameter test facility.
**Weight parts per million.
Velocity
(fps)
Reynolds
Number
Friction
Factor
T( )
0 VTW/p B(0)
(fps)
‘
(%]
13.88
21021
.0118
.551
6.5
.533
14.6
55
21.49
32549
.0096
1.08
7.6
.746
16.3
59
27.19
41184
.0090
1.63
7.8
.917
16.7
59
30.86
46749
.0093
2.16
1.06
56
34.45
52190
.0092
2.65
1.17
56
37.71
57113
.0092
3.19
1.28
55
40.58
61462
.0093
3.73
1.39
54
43.59
66020
.0094
4.32
1.49
52
45.68
69189
.0095
4.81
1.57
51
48.35
73234
.0095
5.40
1.66
50
49.96
75676
.0097
5.89
1.74
49
50.81
76958
.0098
6.10
1.77
49
25

-------
TABLE 2
*LABORATORY AND REDUCED DATA
FOR POLYOX WSR-301 AT A
CONCENTRATION OF 50 wppm**
*Data taken in a 0.18 inch diameter test facility.
**Weight parts per million.
Velocity
Reynolds
Friction
(fps)
Number
Factor
TU) B V’TW/p
( psf ) _____ ( fps )
18.25
22.33
30.98
35.37
39.97
45.16
51.46
54.50
58. 20
27644
33821
46g31.
53576
60541
68407
77949
82555
88166
B(O)
22.4
23.7
00684
.00624
.00556
.00536
.00528
.00524
00496
.00516
.00524
0.551
0.756
1.30
1 .62
2.05
2.59
3.19
3.73
4.32
11 .533
11.5 .624
.819
.914
1 . 03
1.16
13 1.28
1.39
1.49
72
73
74
74
74
73
26.3 74
72
71
26

-------
TABLE 3
*LABORATORY AND REDUCED DATA
FOR POLYOX WSR-301 AT A
CONCENTRATION OF 100 wppm**
Velocity
(fps)
5.31
9.08
12.02
15.45
17.72
22 . 08
25.98
29.88
33.31
36.01
41.87
47. 58
52.59
59.95
Reynolds
Number
Friction
Factor
.01584
iw
(psf)
0.108
0
5.6
/Tw/p
(fps)
.236
B(0)
12.8
p
(%)
53
8036
13759
.01080
0.216
.334
63
18208
.00924
0.324
9
.409
18.6
66
23397
.00784
0.454
.484
69
26841
.00704
.0534
.525
72
33448
.00640
.0756
.624
73
39335
.00596
.0972
.708
74
45264
.00548
1.19
12.75
.783
25.9
75
50450
.00524
1.40
.850
75
54541
.00512
1.74
.947
75
63423
.00508
2.16
1.06
75
72072
.00492
2.70
1.18
75
79659
90811
.00468
.00428
3.13
3.73
16.5
1.27
1.39
29.6
75
77
*Data taken in a
**weight parts per million.
0.18 inch diameter test facility.
27

-------
TABLE 4
*LABORATORY AND REDUCED DATA
FOR PERCOL 155 AT A
CONCENTRATION OF 10 wppm**
*Data taken in a 0.18 inch diameter test facility.
**weight parts per million.
Velocity
(fps)
18.41
Reynolds
Number
Friction
Factor
.00664
Tw
(psf)
0.546
0
11.3
fTW/P
(fps)
.530
B(o)
23.3
(%)
73
27880
26.50
40137
.0064
1.09
11
.750
24.0
71
31.06
47052
.00684
1.60
.908
68
34.84
52779
.00732
2.16
1.06
65
37.67
57059
.00784
2.70
1.18
62
40.68
61612
.00796
3.19
1.28
60
42.94
65043
.00844
3.78
1.40
57
44.57
67517
.00884
4.27
1.48
55
47.02
71224
.00896
4.81
1.57
54
49.38
74799
.00932
5.51
1.69
51
50.17
75992
.01044
6.37
1.81
45
28

-------
TABLE 5
*LABORATORY AND REDUCED DATA
FOR PERCOL 155 AT A
CONCENTRATION OF 50 wDDm
Friction 1 w 0
Factor ( psf ) ___
11W/P B(0)
( fps )
*Data taken in a 0.18 inch diameter test facility.
**weight parts per million.
Velocity
(fps)
18.50
29.79
38.43
43.49
50.41
56.50
59.83
65.98
Reynolds
Number
28028
451 23
58212
65877
76363
85590
90630
99949
.00636
00500
00452
00448
00423
• 0041 2
• 0041 2
• 00404
0.529
1 . 08
1.62
2.05
2.65
3.19
3. 56
4.27
11.8 .522
14.3 .746
16.3 .914
16 1.03
1.17
19.8 1.28
1.36
21 1.48
24.2 74
27.9 77
29.4 78
29.3 77
77
30.5 78
78
30.7 77
29

-------
When the above preparations were completed, a t1 run”
was made by opening valve (F) a preselected amount, measuring
the steady-state flow and frictional pressure loss (as
indicated by the differential pressure gauges).
Three tests were performed at a minimum of three
different flow rates, with fresh solution used for each
run, and the results at each flow rate were averaged for
reporting purposes.
Three generic groups of polymer materials tested in
the above manner are polyethylene oxides, polyacrylamides,
and polyacrylaniide co-polymers. The most efficient of each
generic type were tested in full-scale field tests. These
were:
Designation Type Manufacturer
Polyox WSR-301 Polyethylene Oxide Union Carbide
Percol 155 Polyacrylamide Allied Colloids
4430 Co-Polymer ICI America
The first two materials are presently available from
their manufacturers, but unfortunately production of the
third material has ceased. The laboratory and calculated
data for the materials which are still available appears
in Tables 1 through 5. All spaces in the tables are not
filled because inspection of the data shows that the
materials do not behave as “power law fluids” over the
full test range. Inspection of the tabulated wall shear
stresses produced under these test conditions indicate
that the polymer solutions suffered from shear degradation
at the high velocities.
Table 6 lists the physical properties which were used
in calculating the friction reduction parameters. Any
dimensionally homogeneous set of units may be utilized,
so long as they are used in all calculations.
TABLE 6
PHYSICAL CONSTANTS UTILIZED
FOR CALCULATIONS
IN TABLES 1 THROUGH 5
p=density of water=1.9388 slugs/cubic foot
30

-------
=Kinematic viscosity=1.059 square feet/second
g=gravitational constant=32.2 feet/second
w=specific weight of water=62.43 pounds/cubic foot
To make comparison of the materials simpler, Figure
10 is a graph of 8 (relaxation time) versus iu (wall shear
stress) for the various solutions. Using this graph, it
is possible to calculate the constants, K and n, to be
used in Equation (22). The constants determined in this
manner are displayed in Table 7.
TABLE 7
CONSTANTS FOR USE IN
THE EQUATION O=KT
FOR VARIOUS SOLUTIONS
Polymer Concentration K n
( wppm )
WSR-301 10 1.97 .1680
WSR-301 50 11.96 .1405
WSR-301 100 11.90 .248
Percol 155 10 11.03 -.0325
Percol 155 50 13.72 .293
The values of K and n for the Percol 155 at 10 wppm
are doubted, since the sign of n would indicate that that
solution gets more efficient as shear decreases.
An inspection of Figure io illustrates one weakness
of scaling from laboratory results. In order to produce
pressure drops which can be accurately determined, it is
necessary to operate the short tube used at much higher
shear stresses than those normally encountered in a
gravity sewer system. For comparison, an 18 inch diameter
sewer line with a hydraulic gradient of 0.77 percent
only develops a wall shear stress of 0.18 pounds per
square foot compared to the lowest average shear utilized
in the laboratory of 0.56 pounds. Laboratory tests
should be designed to cover the range of shear stresses
which are expected in large-scale applications.
31

-------
Figure 10 indicates that the Percol 155 material at
a concentration of 50 weight parts per million(wppm) is more
efficient than Polyox WSR-301 at a concentration of
100 wppm. This conclusion is correct under laboratory
conditions. However, the Percol was found to be more
difficult to disperse in the large scale equipment eventually
constructed. It was found that additional mixing energy
and water was required to form a good dispersion that was
pumpable. One other significant difference is the respective
onset shear stresses of the two materials. The onset
point for Polyox has been found to be concentration
dependent and is on the order of .012 pounds per square
foot at 10 wppm and .038 pounds per square foot at 50
wppm. The onset point for polyacrylamides is relatively
independent of concentration and occurs at a shear stress
of about .06 pounds per square foot. Therefore at very
low shears the Polyox will be a more efficient friction-re-
ducing agent. However, both materials are suitable
under the proper conditions.
If the materials for which data is given are chosen
as a friction-reduction material, the data may be used
directly. The fact that only two types are described
should not present a limitation to a potential user. In
many cases, the manufacturer of a potential material can
provide the data required for evaluation in the manner
described, or a moderate investment can equip a wastewater
treatment laboratory to perform the required tests.
Based on a potential user’s specific requirements,
the choice of a friction reducing agent should be made
after evaluation of the following properties:
1. The onset shear stress should be lower than that
anticipated in the “real world.”
2. The material should not produce gross solution
viscosity changes at low concentrations.
3. The material should work efficiently as a friction-
reducing agent in the laboratory tests.
4. The material should be as “dispersfble” in water as
possible.
32

-------
—
. 1— -
-
T i

Li1:1
1Th
H ILi L:+t
— — —
I I
II
T
i
1
II
F:f
1-

4 L
ftI I 1i
.2 .3 .4 .5 .6 .7.8 1
Wall Shear Stress (T , psf)
Figure . Graph of Relaxation Time Versus Wall Shear Stress for Various
Al
lJ,
a)
•1-4
F-
0
.rl
a)
25
20
15
10
9
8
7
6
5
4
3
2
1
(A)
I---
i -I
Th i
Th
ft
h
Legend:
Polyox WSR-301
0 10 wppm
‘ H 50wppm
: 4 100
• 155
• 10 wppm
A 50 wppm
• I L JJJ
-1-
.1
-I-i—t
2 3 456 7 8
10
Concentrations of Percol 155 and Polyox WSR-301

-------
PREPARATION OF POLYMER MATERIALS FOR INJECTION TESTS
During the course of the present and past projects,
many products suitable for use as friction reducing materials
have been tested and characterized. ppendix A is a list
of presently known acceptable polymer (1) materials, along
with the manufacturer’s name.
To obtain the polymer material used during the course
of the present project, all those manufacturers were invited
to submit a bid for any polymer material which complied
with a supplied performance specification. The specification
is included in this report as Appendix B.
The only material used during the preliminary tests
was Polyox WSR-3O1, supplied by Union Carbide. The material
was low-priced and exhibited friction reduction properties
which approached the theoretical maximum of 80% when tested
in the laboratory apparatus.
The polymer was delivered as a dry, granular material
which required protection from moisture, bacteria, and heat.
The portable equipment which was available for injection
required that the polymer be mixed in a non 2 lvent as a
slurry for dispensing. During earlier work the slurry
was made by making a gel of isopropyl aichol and then
suspending the friction reducing polymer in the resulting
viscous media. This approach was initially used for this
project; however, the slurry separates after two to three
days leaving tne liquid phase supernatent. In addition,
the anhydrous alcohol used constituted a fire and explosion
hazard.
These undesirable properties were not significant during
earlier field work, since the polymer slurry was generally
expended within 24 hours of preparation. However, the
execution of work described in this report was dependent
on rainfall events, requiring that the polymer materials
be prepared and stored until needed. It was therefore
necessary to devise a stable material for the application.
The first attempt to produce a mçre stable slurry material
consisted of the polymer mechanically mixed with a 40-per-
cent sodium hydroxide solution. This slurry was stable for
periods up to one week, but soon separated leaving the
liquid phase sub-natant. Another disadvantage of this material
is the obvious hazard of the caustic material to personnel
and equipment. __________________
e Products forminri nicellar structures are not included.
** The Weston Co. “Pclyicr fcr Sewer Flow ControL” USEP
Report No. ilO2O IGO8/69, August, 1969.
34

-------
The final slurry material prepared for injection made
use of the fact that the polymer being used, polyethylene
oxide, is insoluble in saturated brine. The brine was
prepared from rock salt and was relatively inexpensive.
It is possible to form chemical complexes in the brine to
suspend the polymer particles for extended time periods.
Slurry of this type was made up in steel drums for handling
convenience and stored for up to four months with only
slight degradation of friction reduction ability caused by
storage in the wet form. However, the slurry concentration
changed during storage as evidenced by a clear sub-natant
fluid. When the slurry was used for subsequent injections,
grab samples of the slurry were used to determine slurry
concentration.
The slurry used for the preliminary tests may be
prepared in the following manner:
A. Measure thirty gallons of super-saturated sodium
chloride brine into a standard 55—gallon drum.
B. Heat the brine to approximately 1500 F.
C. Stirring gently, sprinkle in one pound of Hercules
polymer FR-4 and one pound of Tamol (wetting
agent)
B. Stir until the FR-4 polymer is completely dissolved.
E. Allow the brine-polymer solution to cool to
approximately 85° F.
F. Increase the agitation of the polymer solution
until a large vortex is formed.
0. Slowly add the dry polymer to be suspended, continuing
agitation and rotating the drum to insure a uniform
slurry. The maximum amount of polymer which can be
suspended is dependent on the bulk density of the
material and ranges from about 80 pounds for the
low bulk density material to 100 pounds for material
with a high bulk density.
H. Without stopping the agitation add 860 milliliters
of saturated chromium chloride solution made by
adding an excess of the chemical to saturated brine.
I. Stir until the color is uniform, or for about 3
minutes, and withdraw the stirrers. Excessive
stirring will destroy the suspending properties
of the slurry.
35

-------
J. Transfer of the finished slurry to other containers
should be by decanting, since pumping can break the
slurry.
The proper preparation of this slurry material is largely
an art and those not skilled in its preparation may find it
difficult to produce a stable product on the first attempt.
The above described slurry is only suitable for the poly-
ethylene oxide materials or other materials insoluble in
saturated brine.
Three possible disadvantages to the slurry prepared as
above are inherent in its properties:
A. The slurry is extremely stable and requires
mechanical dispersion to insure adequate solution
in the treated line.
B. The brine used in the preparation may be objection-
able in the event that chlorides are a local water
quality problem.
C. The chromium ion used for complexing could become
a significant pollutant if large quantities of the
slurry were used.
36

-------
SECTION 6
EQUIPMENT FOR POLYMER INJECTION
GENERAL STATEMENT OF THE PROBLEI’1
The water-soluble polymers used for friction reduction
are furnished as dry granular solids which must be dissolved
in the sewage in the correct proporations to produce the
polymer-sewage concentration which has been previously
determined. Because of certain characteristics of the
polymer materials, it is not possible to simply “stir in”
the solid polymer as one might add sugar to coffee. It is
necessary to provide a means for effecting the dispersion
and subsequent solution of the material. Although this
particular requirement is the most critical stage of poly-
mer injection, there are other requirements which must be
met in the design of a polymer injection facility. These
requirements include:
1. Handling of packaged polymer materials.
2. Environmentally-controlled storage of polymer
material
3. Metering of dry polymer solids.
4. Dispersion of polymers into water.
5. Injection of dispersed polymer into sewer line.
6. Control of polymer injection process.
This section of the report will present design factors
which must be considered in assembling a polymer injection
facility, and specific recomendationS concerning equipment
which has been shown effective in field tests will be made.
The polymers used for friction reduction are varied
in chemical composition but share certain significant physical
characteristics. It is these physical characteristics which
determine the manner in which they are applied. The common
physical characteristics are as follows:
37

-------
1. Solubility - The polymers used for friction
reduction are very soluble in water, but the
viscosity of the resulting solutions restricts
the polymer-water ratio to low values if a
pumpable fluid is required.
2. Molecular Weight - The molecular weight of those
polymers found most effective range upward from
three million. The molecular weight of water is
only 18.
3. Physical Size - The extremely high molecular
weights of the polymer materials are paralleled
by the physical size of the molecules. For instance,
a molecule of a polyn r with a molecular weight of
4,000,000 is approximately .0012 inches long, and
should apparently be visible to the naked eye.
However, it is only 12 X iO inches in diameter.
4. Physical Form - Polymers for friction reduction are
granular, flake, or powder form material. The
individual particles may be spheres, platelets, or
discs which are made up of entwired individual
molecules. These dry molecules can be likened
to tightly-coiled springs with projections spaced
along the coils at regular intervals.
5. Specific Gravity - The specific gravity of most
water-soluble high molecular weight polymers
fall into the range 0.98 to 1.08, which makes
them neutrally buoyant.
6. Bulk Density - Air becomes trapped between the
particles producing bulk densities on the order of
11 to 60 pounds per cubic foot with a mean value
of about 25 pounds per cubic foot.
Two conditions necessary for complete solution of
water-soluble polymers are:
A. Each polymer particle must initially be surrounded
by enough water to satisfy the water uptake
requirements of all the polymer molecules which
constitute the particle, and
B. Once wet, the polymer particles must be kept
physically separated until solution is complete.
If these two conditions are not met the individual
polymer particles will agglomerate into sticky lumps which
are externally wet with essentially dry centers, normally
38

-------
called “fisheyes.” These fisheyes are very difficult to
dissolve and will not go into solution quickly enough to
act as friction reducers when injected into the sewer line.
PORTABLE FIELD TEST INJECTION EQUIPMENT
Earlier sewer line injection tests *were conducted
using a pump to lift a portion of the sewage stream and force
it through an eductor which provided a decreased pressure
to feed the polymer slurry. The system worked well in those
tests because the equipment could be placed immediately
adjacent to an entry into the line (Figure fl ) and suction and
discharge hoses could be short. However, those points selected
for injection for the present project were either totally
inaccessible for the trailer-mounted equipment or the hose
lengths involved (75-100 feet) were impractical.
The first modification to the equipment consisted of
replacing the gravity feed on the slurry trailer with a
positive displacement pump with a variable speed drive.
The slurry could then be pumped into an eductor in which
a stream of water from a fireplug provided the mixing energy
required (Figure 12). Through additional experimentation,
it was found that a good slurry could be adequately dispersed
by the turbulence of the sewer into which the injection
was made. The equipment was then simplified as shown in
Figure 13, eliminating the need for a water source and
the resulting long water hoses. This system was used for
the first and second injection tests.
During the second injection test attempt, a shear
pin in the pump broke while the slurry line (75 feet) was
full. Water was absorbed through the injection nozzle and
the hose plugged. When the pump was repaired and restarted
the resulting high pressure ruptured the hose, making it
necessary to abort the test. As a result of this mishap,
it was decided to construct a smaller, more portable
version of the injection rig to allow a closer approach
to the injection point. The result of this construction
is shown in Figure 14. It was essentially a specially
designed variable-speed, positive displacement barrel pump,
light enough to be handled by one man, yet capable of pumping
a viscoelastic slurry of the consistency of chassis lubricat-
ing grease. This devise was successfully used for the pre-
liminary injection tests, at rates up to about 4.5 pounds of
polymer per minute. Full-bore capacity was about 6.0 pounds
of polymer per minute.
*The Western Company, ‘Polymers for Sewer Flow Control”,
USEPA Report No. 11O2ODIGO8/69, August, 1969.
39

-------
FIGURE 1 1. TRAILER-MOUNTED EQUIPMENT
- .
:‘
40

-------
\ \ \
Figure 1 2. Water-Dispersed Inj ection Sv stem
FIRE
HOSE
— __ __
41

-------
POLYMER
STORAGE
GASOLINE _____ VARIAB ______
ENGINE DRIVE
t _- r
PUMP
Figure f3 Modified Injection System
42
+
9
- d
— 1
- 4 -

-------
L j T
4
F l g. 14 Light-Weight Variable-Flow Polymer
Injection Apparatus
43

-------
COMPARATIVE DISPERSION TESTS OF SLURRY AND DRY FEED
EQUIPMENT
Difficulties encountered in formulating and storing
slurry-type polymer suspensions made it necessary to
develop an alternative method for dispersing polymer
solids. A primary objective of this development was a
method which would allo the use of the materials in the
form in which they were received from the manufacturer.
Since no experimentation in sewer line friction reduction
has been performed using the dry materials without the aid of
a slurry or dispersing agent, it was decided to perform
comparative tests with the existing slurry feed equipment
and commercially available dry feed equipment.
The polymer slurry used In experimentation took
advantage of the fact that a liquid in which the polymer is
insoluble can serve as physical spacer to hold the
polymer particles separated until each particle could be
wetted requiring no mechanical agitation other than that
present in a turbulent flow stream. In fixed plants in which
polymer solutions are required, violent physical agitation
replaces the non-solvent uspacer and a solution can be made
directly. A series of laboratory experiments illustrates the
comparative behavior of the two systems. In these experiments,
a rotary viscometer was used to determine the time required
for the polymer forms to dissolve in solutions made from
slurries and dry powder. The dry powder solutions were pro-
duced by introducing the required amount of material into a
1000 milliliter beaker of water violently agitated by a labor-
atory stirrer, while the slurry solutions were agitated by the
action of the viscometer, alone. The polymer in both cases
was Union Carbide Polyox WSR-301. Table 8 gives the summarized
results of these tests which indicated that the dry polymer
took about 50% more time for solution.
TABLE 8 RESULTS OF COMPARATIVE SOLUTION TIME TESTS OF
DRY AND SLURRIED POLYOX WSR-30l
- SLURR lED_Pj DRYPOL ER_
Time to 50% of rime to 50%
CONCENTRATION Ultimate Visco Ultimate f Ultimate Ultimate
sity (sec.) Viscosity fiscosity Viscosity
( wppm ) —___ _________ centipoise ( sec.) — centipoise )
1000 43 44 65 45
2000 300 55 450 60
44

-------
To confirm the effectiveness of a dry feeder system,
parallel tests were run using the slurry used in the
experimentation and a Gaco*Dry Chemical Feeder. This
device is installed temporarily on a 24” line in the City
of Dallas which is subject to frequent surcharging.
Unfortunately, the dry feeder was capable of a maximum
throughput of only 0.8 pounds of dry polymer per minute,
a quantity insufficient to relieve the surcharge, but
adequate to produce velocity changes on the order of 10%.
The test proceeded as follows: polymer injection was
started using the slurry feed. Heads and velocities were
recorded continuously (head) or at 10 minute intervals
(velocities). After 30 minutes of injection, the injection
rate was changed. At the end of 1 hour, the slurry feeder
was shut down and the dry feeder immediately started. The
feed rate was changed after 30 minutes and the injection
stopped at the end of the second hour. Table 9 is a
summary of the results of these tests. All velocities were
determined with a Gurley direct reading rotary meter.
The line In which the test was run consists of
4,127 feet of 24 diameter reinforced concrete pipe with
five included manhole structures. At the velocities measured
during these tests, about 14 minutes should be required for
the sewage with polymer - sewage without polymer interface
to travel from the injection point at the first manhole to
the outfall at the fifth manhole. The data reflects this
‘travel time” in the amount of time required for a change
in injection rate to produce a change in flow rate.
The data indicates that the dry feeder is as effective
or perhaps more effective than the slurry feeder with each
producing a velocity increase of about 10%. This would
indicate that the times required for a polymer solution to
be formed from the two polymer forms (dry and slurry) are
either enuivalent or the differences are not significant
for fielu applications.
It should be noted that most polymer manufacturers
recommend “aging” of the polymer solutions before use to
permit adequate time for the most persistent polymer part-
icles to dissolve. At no time during the experimental
program was a solution prepared in which all the material
was dissolved without “aging.” The time required for aging
is dependent on polymer type, particle size, solution
temperature, level of agitation and concentration. From
qualitative observations of these solutions it can be
estimated that solutions made without aging time (as is
the case in a “quick-mix” system) would actually contain
gplv about 75 to 95 percent of the polymer bein9 mixed.
* A trademark of the Gaddis Manu acturing Company,
Bartlesville, Oklahoma.
45

-------
TABLE 9
RESULTS OF COMPARATIVE INJECTION TESTS OF DRY
AND SLURRY FEEDERS
INJECTION
RATE CONG. S VELOCITY % CHANGE
TIME (lbs/mm) (ppm) (ft/bOO ft) fps IN VELOCITY
Lminutes)
SLURRY
FEEDER :
1=0 0 0 4.08 4.55 0
15 0 0 4.08 4.50 -1.1
30 0.8 112 4.08 4.55 0
45 0.8 103 4.12 4.95 +8.8
60 0.8 103 4.12 4.95 +8.8
61 1.1 141 4.12 4.95 +8.8
75 1.1 143 4.08 4.90 +7.7
90 1.1 i35 4.03 5.20 +14.3
DRY FEEDER :
91 0.75 92 4.03 5.20 +14.3
105 0.75 96 4.08 4.95 +8.8
120 0.75 96 4.10 4.95 +8.8
121 0.33 42 4.10 4.95 +8.8
135 0.33 41 4.10 5.10 +12.1
150 0.33 41 4.10 5.10 +12.1
151 0 0 4.10 5.10 +12.1
165 0 0 4.12 METER FOULED
180 0 0 4.15 4.70 +3.3
46

-------
From the information gathered in the dispersion method
study, a number of conclusions were formulated:
A. Dry chemical feed into an eductor-type polymer
disperser is the perferred method of preparing
polymer solutions.
B. There were no packaged polymer dispersing units
of adequate capacity to meet the requirements of
the program.
C. Since a polymer solution cannot be pre-prepared in
anticipation of need, there will be some undissolved
polymers introduced into the treated sewer, resulting
in slight decreases in polymer efficiency.
JSPERSIOJi _ EU LP iENT FOR FIXED INSTALLATIOt4 .
The recommended method of dispersing dry polymer solids
in water is a standard piece of equipment called an “eductor-
type polymer disperser’ or more simply, an “eductor.” Figure
15 represents the basic construction of any eductor. It
consists of a water inlet connection, a polymer inlet,and a
discharge port for the mixture. Internally, it is simply
a Venturi tube which generates a low pressure area which
will draw air through the polymer inlet. Solid polymer
materials are dispersed first in the air, and are wetted in
the high energy mixing area in the lower part of the eductor.
Air vents are provided to avoid wetting the upper “dry’
area of the eductor in the event air flow through the
polymer inlet is cut off.
Eductors of this type are available in bronze, cast iron,
and stainless steel from manufacturers like Hercules Chemical
Company, Penberthy, Incorporated, and Shutte and Koering.
The physical configuration of eductors from various manufac-
turers varies but operational characteristics are similar
for all units. Figure 16 shows typical eductor performance.
It should be noted that the eductor requires some minimum
flow to operate, and that the permissible range of feed
rates is extremely broad. To utilize an eductor, it is
only necessary to establish a rate of flow which will
adequately disperse polymer at the maximum rate anticipated.
Once this flow rate is established, any solids feed rate up
to this maximum is possible without readjustment.
A type of eductor which has proved satisfactor is the
“Hootenanny” manufactered by C. E. Hooten Company in Miami,
Florida. This particular device is constructed of polyvinyl
chloride (PVC) and Teflon which suppresses build-up of caked
material on the interior surfaces. As provided from the
factory, the Hootenanny has no air vents, so when mounting,
a loose fit between the feed funnel and the eductor top
is recommended to suppress backsplash.
47

-------
Pressurized
Water
Inlet
Polymer
Inlet
Dispersed Discharge
High Energy Mixing Zone
Figure 15. Typical Eductor Construction
1
Air Vent Hole
48

-------
Water Flow Through Eductor, gpm
Figure 16. Operating Characteristics of
A Typical 1 1/2 Inch Disperser
.
E
a)
a)
a)
a)
24
22
20
18
16
14
12
10
8
6
4
2
0
Insufficient Dispersion
Operating Area For
Good Dispersion
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
49

-------
FEEDERS FOR POLYMER SOLIDS
There are basically two types of feeder systems for
dry polymer solids available; feed weighing systems and volu-
metric feeding systems. Both types of feed systems are
available with a variable feed rate adjustment. The former
system, which uses a weighing arrangement in a feedback loop
to control the rate at which solids are fed, has the advantage
that once calibrated any material may be fed at a controlled
rate; that is, for a given setting of the control, all
polymers will be fed at the same rate regardless of particle
size and shape. However, the weighing systems are generally
a great deal more complicated to maintain, and the initial
cost is greater than volumetric systems.
Volumetric systems require that calibration tests be
made for each material to be fed. These tests should be
run over a wide range of feed rates, since the speed-feed
rate relationship is not linear. Once calibration has been
determined for any batch of material, the feeders can
usually dispense material to within plus or minus two percent
of a set rate, an accuracy which is more than sufficient for
polymer injection work.
There are a number of suitable volumetric feeders
available, and the manufacturers can usually provide
information on the specific capabilities of their device
to perform with a specific material. The requirements
for specifying a suitable volumetric feeder include:
1. The feeder must be able to dose accurately and
repeatably.
2. The feeder should be independent of the depth of
material in the feed hopper.
3. The feed rate should be adjustable over the range
of feed required.
4. The feed rate should be controllable by standard
process control signal (e.g. 4-20 ma.).
A volumetric feeder which meets the above requirements
and has been field tested is the Model 105-2 manufactured by
Acrison, Incorporated, located in Carlstadt, New Jersey.
This particular feeder has two concentric helices, one
which preconditions the dry solids and the second which
extrudes the material. This feeder is widely used for
critical feed applications in the pharmaceutical industry.
50

-------
ANCILLARY FEED EOUIPMENT
Two items of equipment which have been found virtually
indispensable in a reliable solids feeding system are a
polymer ‘clod” separator and a solids level detector. The
first of these is a device to avoid feed problems and the
second acts as a safety switch in the event the first fails.
Invariably, a water soluble polymer will form dry
lumps or flakes when stored in stasis for long periods of
time. Ball-like lumps can form from the compression caused
by overlying polymer. Flakes generally form on an exposed
surface or on a surface which sweats. Many of the balls
will be broken by the feed mechanism, but the flakes will
generally pass through the feeder. Also, bits of foreigh
material may be introduced into the bulk storage tank during
loading operations. Any of these items can cause a blockage
in the intake of the polymer disperser and stop the
polymer feed. This condition must be avoided.
Figure 17 is a design for a “clod” separator which func-
tions adequately. The model shown was fabricated to order
by a local manufacturer from plans furnished by the engineer.
It consists of an inclined screen which is lightly vibrated
by an external inertial vibrator. The granular solids less
than one-half inch in diameter fall directly through the
screen into the polymer disperser inlet, and the “clods”
shaken from the surface of the screen are collected in
a plastic bag.
The second piece of ancillary equipment is a capacitance
solids level detector, such as that manufactured by Drexel-
brook, Incorporated. The probe of this device is mounted
in such a position above the vibrating screen of the polymer
separator that a stoppage which produces a solids build-up
in the feed funnel will react with the probe and shut down
the feed. This action will prevent a messy solids spill and
will make clearing of the stoppage easier. The Drexelbrook
unit chosen for this function ignores film and dust build—ups
on the probe, avoiding unnecessary shut-downs.
STORAGE HOPPERS FOR POLYMER MATERIALS
Polymer manufacturers recommend that the dry solids be
stored in the shipping containers until used. This recom-
mendation is based on the tendency of the polymer materials
to absorb large quantities of water from the storage
atmosphere, producing lumps, flakes, and in extreme cases
syrupy solutions. However, if one intends to operate an
un-manned, fully-automated polymer injection facility, this
51

-------
Viiu to MoUW’TZD ON
W T N SCRItM P.S
SECTION VA-A
Noi
Tlg r p .iL N r c*rc, YP.O
4A L S7AINL U 31( .. ( C !T
$CPZIN AN
POLYMER CLOD SEPARATOR
B CHMFkN POL’y’ME INJECTION STATION
Figure
DALLAS WATER UTILITIES
CITY O DAIta$. TIXAS
______
_-
- -.“ 3LtAJD1 QQK
a ..,
52

-------
implies that the polymer must be removed from its shipping
container and placed in bulk storage ready for use. To
insure that the materials will remain in a usable condition,
this storage hopper must provide for humidity control,
condensation prevention, ‘arch’ breaking, and free feeding.
The exact dimensions and shape of a storage hopper will
vary, depending on the project needs, but the basic functions
will remain the same. The bottom of the storage hopper
should be conical in shape, with the sides of the cone at an
angle of about 30 degrees with the vertical. The angle
exceeds the angle of repose of most of the polymer materials
and will provide satifactory feed conditions.
After polymer materials have been stored for a long
period of time, settling produces a packed structure which
will bridge the discharge port. This arch must be mechanical-
ly broken either by stirring, vibration, aeration, or any
combination. One satisfactory method of accomplishing this
function is by using a ‘vibrating hopper bottom” which forms
the bottom of the conical hopper and is mechanically excited.
The Acrison RP Hopper Bottom is an electrically-driven
product which uses a rocking motion to agitate the stored
material. In general, it is convenient to use a device
which is manufactured to mate directly with the feeder chosen,
thereby reducing installation problems and eliminating the
need for fabrication of special fittings.
If the polymer storage hopper is to be exposed to
drastic temperature changes, it requires that the walls
of the hopper be insulated to suppress condensation. A
satisfactory method for accomplishing the insulation is
by spray coating the interior surface with urethane foam,
and then painting the foam with a gloss finish latex paint.
The coating provides a surface to which the polymer does
not readily adhere.
Dehumidification of the polymer storage area is an
extremely important function, but one which is complicated
by the properties of the polymers. The polymers consist
of a wide grain-size distribution including some dust which
will be drawn through the circulation system of a dehumidifier.
This dust is then dissolved or melted in the apparatus causing
air flow stoppages and overflowing drains.
There are two basic types of dehumidifiers, the desiccant
type or the condensing type. The former type, such as the
“Honeycombe” dehumidifier manufactured by Cargocaire Incor-
porated, proved to be not sufficiently rugged to resist the
caustic action of the polymers, and dust drawn into the
system was melted by the heat dry air, causing air flow
53

-------
restriction. Therefore it is recommended that a condensing
type of dehumidifier be utilized.
Those portions of the dehumidifier that will come in
contact with the polymer dust should be of materials which
will resist caustic action. For example, alumimum coils,
enclosures, and ducting should be avoided unless the
surface is coated or plated with a more inert material
INJECTION OF DISPERSED POLYMERS
After the polymer has been fed through a disperser
and mixed with water, the resulting mixture gets to a
viscous state very quickly as more and more of the solids
go into solution. If the feeding and dispersing apparatus
were to be mounted directly over a sewer line so that the
dispersion could discharge freely into the sewage, the
problem of injection would be non-existant. However, in
many cases the dispersion will be performed at some distance
from the sewer line and the dispersion piped to the desired
injection point.
The design of the eductor-type disperser prohibits the
attachment of long discharge piping directly. The common
method of overcoming this is to allow the discharge of the
disperser to fall into a “buffer” tank, and then pump the
dispersion to the injection location. The size of the
pump and the volume of the tank should be such that the dis-
persion is not allowed to “age” more than about one minute,
the permissible aging time being determined by the solution
characteristics of the polymer used.
Referring to Figure 16, it can be noted that it is
possible to disperse approximately 8 pounds of polymer in
as little as 16 gallons of water. This corresponds to
a solution concentration of about 6 percent. At this
concentration, the polymer solution (fully dissolved, or
aged) has an estimated viscosity in excess of 100,000 centi-
poise. In other words, the solution could be formed into
balls and bounced. Therefore, the time which the dis-
persion resides in the buffer tank is very critical. Even
before the solution is fully aged, tha material can become so
viscous that the injected material will resemble a rope
and fail to disperse in the sewage as is required in order
to be effective.
The buffer tank should be made of a non-corrosive,
smooth-finish material such as stainless steel, polyethylene,
or polypropylene to withstand the caustic reaction of the
polymer solution and to facilitate wash down.
54

-------
A centrifugal pump will not suffice to pump the viscous
dispersion of polymer and water. A positive displacement
pump capable of passing the undissolved, dispersed solids
should be used. A progressive cavity pump such as a Moyno
screw pump with stainless impeller and neoprene stator is
satisfactory, but sometimes inconvenient in geometry. A
gear pump with cast iron case and bronze impellers such
as those manufactured by Worthington perform satisfactorily,
although some corrosion of the pump case will be encountered.
When sizing the pump and drive motor a viscosity of 2000
centipoise should be considered as the minimum.
If the dispersion is handled rapidly , no special pro-
vision for injection into the sewer line are necessary. A
satisfactory method consists of attaching the injection
piping flush with the inside wall of the sewer with the
discharge at right angles to the flow in the sewer. The
turbulence in the sewer line will complete the dispersion
process.
If, for some reason the polymer dispersion tends to reach
the extrudable state before reaching the sewer line, the amount
of water used in the initial dispersion must be increased,
or, additional water can be added to the buffer tank.
Supplemental mixing energy may be required if the latter
option is chosen.
55

-------
SECTION 7
PROCESS CONTROL AND INSTRUMENTATION
FQR POLYMER INJECTION
RELATIONSHIP BETWEEN INSTRUMENTATION_AND PROCESS CONTROL
Although It is physcially possible to inject a friction-
reducing material into a sewer line and qualitatively observe
the results (many experiments have been so performed), it
would not be possible to efficiently dose the material , con-
trol the process, or even evaluate the effectiveness of the
technology. The starting time of an injection will usually
be determined by a rising head in the sewer, but the rate
at which the polymer is to be injected is most efficiently
determined by proportioning to the flow rate. These two op-
erational aspects require the installation of static pressure
gauges at critical locations on the line and at least one flow-
meter at some point on the line. Hence, the equipment re-
quired for process control Is also the minimum instrument-
ation signals required to evaluate the effectiveness of
injection.
REQUIRED IROCESS CONTROL INSTRUMENTATION
Process control instrumentation is required in two separate
functional areas: (1) external to the injection facility to
provide “real world” data; and (2) inside the station to con-
trol the injection process. The external equipment is as dis-
cussed above. The internal equipment performs the following
functions:
1. Polymer Feeder Rate Control
2. Injection Pump Start-Stop Control
3. Polymer Agitator Start-Stop Control
4. Process Overflow Shut-Down
5. Process Failure Shut-Down
6. Injection Completion Clean-Up Control
7. Injection Start-Up Sequencing
The need for each of these functions will be discussed
individually, since some Installations may not require every
function. To start an injection process correctly, a number
56

-------
of things should happen in the following sequence:
a. Mixing water flow to the eductor should be established.
b. The polymer storage hopper should be agitated.
c. The polymer “clod” separator vibrator should be
activated.
d. The polymer feeder should be started.
e. The polymer dispersion injection pump should be started.
If the feeder starts much in advance of the agitator,
arches may form above the moving hopper bottom, requiring some
other action to restart the solids flow. If the solids feeder
is started before the water flow through the eductor is establish-
ed, the solids will clog the polymer Inlet. In a similar manner
simultaneous cessation of solids feed and water flow at the end
of the injection period will leave solids in the feed train
which can cause clogging during the next feed interval.
A control scheme which will execute the required func-
tions is illustrated in Figure 18. Beginning at the left of the
diagram a “run” signal which can be derived from a set point re-
lay which responds to head at critical points on the sewer
opens a solenoid valve, allowing water to enter the process.
The water pressure then activates a pressure switch which turns
on the bin agitator and clod separator, and begins a time “on”
delay for the polymer feeder. At the end of the pre-set delay
the feeder starts at a rate determined by the flow through the
sewage flowmeter, and a dispersion is discharged into the
buffer tank. A low level switch starts the injection pump.
During an injection a solids build-up in the clod separator
will stop the solids feed until the stoppage is cleared manual-
ly. In a like manner, a pump failure will activate a high
level sensor in the buffer tank, causing the solenoid valve
to close. The absence of water pressure will immediately
stop the solids feeder, killing the process until the problem
is corrected.
In normal operation, the disappearance of the “run” sig-
nal starts the timing cycle of a time delay “off”relay which
will hold the solenoid valve open, but will shut down the
solids feed. This action provides a clean water flush of
the dispersion and injection systems, leaving the process
ready to perform at the next “run” signal.
SELECTION OF PROCESS CONTROL COMPONENTS
The first step in selecting process control components,
once the control scheme is selected, is to decide on the pri-
mary signal amplitude which will be utilized in the system.
Standard signals include 1 to 5, 4 to 20, and 10 to SO
57

-------
Time
Delay
“Off,,
Start— T — —
Signal r
*
Water Supply — Bin
I Agitator
— I
Delay
“0
Polymer Supply
Feed Rate Control
U,
1
-j
I Tank Overflow
“S ’ t — — -
Polymer
Dispersion
Outlet
Buffer
Tank
Injection
Pump
To Sewer
Line
Figure 18. Suggested Process Control Scheme for Polymer Injection

-------
milliamperes direct current, and 1 to 5 volts direct current.
The 4 to 20 milliampere DC signal is one of the most common-
ly used, and there are many manufacturers who can provide
control elements and indicating devices for this signal range.
Which range is chosen is not important, but it is important
that the signals be compatible throughout.
A secondary signal for the direct operation of motor
starters, solenoid valves and operation equipment can be
either a low voltage DC or AC line current. The latter is
convenient, but greater care must be taken in the layout of
equipment to minimize shock hazards to maintenance personnel.
A few guidelines which should be considered during the
selection and assembly of process control equipment are as
follows:
1. The control system and the elements which comprise
the system should be as simple as possible.
2. Reliability of components should be stressed, with
particular attention to the state in which a failure
will leave a control.
3. All relays used in the system should be of the en-
closed type.
4. All connecting wiring should be color-coded or
marked at frequent intervals with easily readable
wire markers. Terminations on terminal strips should
be clearly marked with unique identifiers.
One piece of equipment which deserves special mention
is the sewage flowmeter. A meter should be chosen which does
not restrict flow in the line, and which will respond correct-
ly to non-Newtonian flows. This last restriction eliminates
the possibility of using Venturi tubes, orifice plates, and
flow nozzles. Two types of metering which are satisfactory
are electromagnetic velocity sensing and ultrasonic velocity
sensing. Satisfactory devices of the first type are manu-
factured by Fischer and Porter, Brooks Instruments, and others.
This type of meter should be ordered with a Neoprene or poly-
urethane liner. Ultrasonic electrode cleaners are not recommen-
ded. The second type of meter is manufactured by Nusonics,Inc.,
Sparling Meter Co., and others; and can be built into a sec-
tion of almost any type of pipe. Installation of the latter
device must include provisions for flushing across the trans-
ducer faces with clean water.
59

-------
SECTION 8
PRECAUTIONS IN STORING
AND HANDLING POLYMEJ
PROTECTION OF POLYMER MATERIALS
Shipping containers of polymer materials are either
polyethylene-lined fiber drums or polyethylene-lined multi-
layer bags. These containers should be stored in a covered
area and kept sealed until transferred into an environmental-
ly-controlled bulk storage container. The containers
should not be stored in close proximity to steam or hot
water pipes, heaters or other hot surfaces. Exposure to
direct sunlight should also be avoided because the resins
are thermoplastic with a low melting point.
SAFETY PRECAUTIONS
The polyethylene oxides are formed by the reaction of
a gaseous monomer and as a result the presence of unreacted
monomer is not a problem. As previously mentioned, the polymer
gives a basic reaction (solution pH = 10) so the dust can
cause minor irritation of mucous membranes and of the sensi-
tive eye parts. Allergic reaction is possible as with any
material. Polyethylene oxides are non-toxic and can be
ingested without causing difficulties. The large mole-
cule is relatively undigestable and will pass through the
system essentially unmodified. The FDA has recognized
and approved the use of Polyox in certain food uses including
packaging as well as a direct additive to malt drinks
(beer, ale) up to a proportion of 300 parts per million.
The Environmetal Protection Agency has approved Polyox
for unrestricted use as an inert ingredient in pesticide
formulations under Regulation Number 180.1001.
The polyacrylamides may contain small quantities of
unreacted monomers which are toxic until hydrolyzed by
mixing with water. They have a tendency to be more irritat-
ing than the polyethylene oxides and dust masks and goggles
are recommended.
60

-------
When the polymers are handled in a closed system, the
possibility exists that the concentration of dust could
build up to such a degree that a dust explosion could occur.
This fact should be considered in designing handling systems
such as bulk storage bins and pneumatic conveyers.
Containers in which polymers have been stored should
never be used for strong oxidizing agents such as potassium
permanganate ( t potash”), sodium hypochlorite (HTH, Chlorox),
or hydrogen peroxide. The polymer molecule is a long,
active hydrocarbon chain which can provide a concentrated
fuel source. A mixture of Polyox and HTH, when wetted with
water, will often burst into flame. Unauthorized use or
disposal of the polymer materials by persons unaware of the
potential danger should be avoided.
61

-------
SECTION 9
EVALUATION OF A SYSTEM FOR POTENTIAL USE OF
FRICTION REDUCING AGENTS
POSSIBLE APPLICATIPNS OF FRICTION REDUCERS
The operator of a wastewater collection system should con-
sider friction reduction as an alternative to other methods of
relief, such as pumping, based on a thorough engineering invest-
igation of the basic problem, including evaluation of economic
considerations. This section of the report presents a recommend-
ed sequence of steps In that investigation and evaluation.
It is probably safe to say that all operators of sanitary
sewage collection systems have faced, or will face, the problem
of gravity mains which are under-capacity because of unpredicted
population growth patterns, changes in land utilization after
the system has been built, or most often, because they are over-
loaded by water from infiltration and inflow sources. In most
cases, the permanent solution to overloaded sewers can be found
by re-engineering, re-building or rehabilitating the offending
sewers. However, occasions will arise which will require a
problem solution on an interim basis until a permanent solution
can be effected. It is in these cases that the application of
water-soluble polymers as friction reducers may be most useful.
In those cases when sewers overload and surcharge infrequ t-
ly, such as during major storms, it may not be economically feas-
ible to reconstruct a part of the wastewater collection system.
In these situations, it is possible that permanent polymer in-
jection points may be a suitable alternative. As in other pro-
jects, the decision to use this particular technology must be
based on the probability that the desired results can be ach-
ieved, the initial investment required, and the on-going opera-
tion and maintenance costs associated with the constructed
facility.
EVALUATION OF A COLLECTION SYSTEM SEGMENT
The first step in the engineering of a possible friction-
reduction application should be an analysis of that part of the
collection system in which surcharges are evident. This step
in the evaluation should be performed regardless of the action
contemplated.
62

-------
The evaluation should begin with a determination of the
frequency and approximate duration of the surcharging, and more
importantly, the frequency and duration of the resulting over-
flows. A rough estimate of the flow rate in the sewer will suf-
fice to make a first estimate of the cost of polymer injection.
A concentration of 50 parts per million may be assumed as a norm-
al value and the annual cost of polymer injection calculated as
fol lows:
Polymer CQSt = Flow Rate (gprn) x 8.33 pounds,’gal
x (50 x l0_6) x Duration of Overflow (mm)
x Annual Number of Overflows x $1.25/pound
If the polymer cost derived above is within reason, then the
evaluation can be continued with a thorough definition of the
flow-head loss characteristics of the system. Studies to deter-
mine flow rates should be performed in the most accurate manner
possible, and apparent anomalies should be carefully investigated
to insure that observed head losses are actually due to friction
losses. There have been a number of cases reported in which
obstructions, offset joints, sloppy manholes, or other loss—pro-
ducing elements were the causes of overflows. All such head-
loss producers should be corrected if possible, before the data
necessary to produce head_discharge curves is gathered.
Concurrent with the effort to gather the above data, con-
struction drawings should be reviewed to determine lcriticalu
elevations in the flow network. These “critical” elevations are
the points in the system at which the hydraulic grade line first
reaches an elevation sufficient to produce overflows. Care must
be taken to consider that overflows may occur through branch
lines and building service connections. Since the elevations of
these types of potential overflows may not be shown on system
construction prints, field surveys and the establishment of level
nets may be necessary.
When the critical elevations for the flow network have been
determined, these points may be plotted on a profile sheet. The
line connecting the points plotted in this way will establish
the upper limits of the hydraulic grade line which cannot be ex-
ceeded without causing overflows. A comparison should then be
made between the grade line so constructed and a hydraulic grade
line based on the values for flow and friction factor measured
during the field studies. This second line should be constructed
as if standpipes were to be placed at every overflow location,
thereby eliminating the overflows. Candidate injection locations
can be determined by inspection of the differences between the
actual hydraulic grade line and the required hydraulic grade line.
63

-------
The second check for the feasibility of friction reduction
can now be made by calculating the required friction-reduction
efficiency using Equation (17) and the existing and required
friction factors. As discussed in Section 4, the maxi-
mum possible efficiency is 80 percent, and the practical maximum
will be on the order of 50 percent. If the calculation perform-
ed indicates feasibility, the next step in the analysis is
justified.
Since the limiting value of the hydraulic grade line has al-
ready been established, the next operation is to calculate the
shear stress which the head loss characteristic of the grade line
represents. This is done by utilizing Equation (9). Once the
shear stress is determined, the value for various materials at
several concentrations may be determined using Equation (22)
and the appropriate constants previously determined by experi-
ment. These e values determine corresponding values of B(O)
(Figure 8), which can then be used to determine achievable fric-
tion factors. If it is possible to produce a friction factor
sufficiently small, it has then been determined that friction
reduction will result in head reductions or in flow increases.
RELATIONSHIP BETWEEN HEAD REDUCTIONS AND FLOW INCREASES
It has been pointed out in a preceding section that fric-
tion reduction can result in either head reductions or flow in-
creases, or a combination of both phenomena. If one is interest-
ed in stopping overflows, head reduction is usually required,
but is not always achievable even with high efficiency friction
reduction. There is no quick approach to determine if head re-
ductions will result from reduced friction losses.
A definitive analysis of the piping system upstream of the
point of polymer application is necessary. The need for this
analysis is illustrated in the following discussion.
Downstream of an injection location, the effective fric-
tion factor will be reduced. Using the Darcy-Weisbach equation
for frictional head loss, if the velocity (flow rate) through
the pipe with a reduced friction factor were to remain constant,
there would be a reduction in head loss proportional to the
reduction in friction factor. However, the flow in the pipe
immediately upstream of the injection point is also affected
since we have now reduced the head at the downstream end of
that pipe. The effect of this head reduction is to increase
the amount of sewage delivered to the injection point, thereby
increasing the head at the injection point. To further compli-
cate matters, it should be noted that the effect of the inject-
ed polymer on the friction factor can also be affected by the
change in flow conditions, requiring a recalculation of the
friction factor. The sketches of Figurel9 are Intended to aid
in the clarification of this concept.
64

-------
Figure g(a). Polymer Injection To Change Head and Flow
V
— — — —
Figure 19(b). Polymer Injection To Change Flow Only
Figure 19(c . Polymer Injection To Change Head Only
65
Hydraulic Grade Line
Without Polymer Q
With Polymer Q
Hydraulic Grade Line
Positive Displacement Pump

-------
The analysis of the flow system to determine the effect of
friction reduction will consist of the following steps:
1. For the material, concentration, and flow conditions
anticipated, determine the modified friction factor
of the sewer line downstream of the injection point.
2. Calculate the expected head at the injection point by
summing the head losses downstream of the injection
point.
3. Using the head calculated in Step 2, calculate the
increased flow in the upstream network. To be com-
plete, the calculation should be performed for every
pipe to the limits of the flow network.
4. Use the new flow found in Step 3 to recalculate friction
factor.
5. Repeat Steps 2 through 4 a sufficient number of times
that the change in the calculated flows become negligi-
ble.
Upon completion of the analysis just described it may be
found that the head reduction which will be produced will be
adequate to eliminate overflows, or it may be found that polymer
injection will result in flow increases only. In the event that
the calculations indicate the latter situation, the feasibility
of polymer injection for overflow relief is weakened, but there
is one additional possibility which may be considered.
The feasibility analysis discussed in the preceding section
was based on the premise that the collection system network was
filled, heads sufficient to produce overflow were existing, and
that head reductions were required “after—the-facto to stop the
overflows. The second oossibilitv would be to evacuate the ex-
cess water in the system at a rate sufficient to prevent the
system from becoming overly full by starting injection earlier
in the period of maximum input, thereby making space in the sys-
tem to accommodate the excess water at the peak input rate.
When an infiltration/inflow analysis is performed on a collect-
ion system segment, one of the requirements is to different-
iate between infiltration and inflow. In a graph of flow versus
time for a gravity sewer line, an inflow event is usually charac-
terized by a very rapid increase which begins soon after the be-
ginning of a storm, and whose duration Is related to the dura-
tion of precipitation. This type of flow increase produces rapid
surcharging and overflows, and the peak rate of excess flow will
generally exceed the capabilities of the friction reduction
technology. However, if these sources of rapid input can be
eliminated, the excess flows of infiltration which generally
66

-------
increase more slowly and endure longer, can be accommodated by an
earlier injection start. This technique would obviously also
apply to any input which causes a slow rise in system flows.
67

-------
SECTION 10
RELATIONSHIP BETWEEN RAINFALL AND
SEWER OVERFLOWS
During the performance of the preliminary tests, rainfall
data was collected from all of the Water Department gauging sta-
tions around the city. Four of the stations were selected as
representative of the drainage area, stations numbers 27, 41, 43,
and the official United States Weather Bureau Love Field Station.
The location of those stations in relation to the trunk line
is shown by the circled numbers in Figure 20A.
Tables 10 and 11 are a listing of the 1969 and 1970 rain-
fall records of the selected stations and an arithmetical average
of the four stations for each day that any precipitation was re-
corded at one or more stations. The 11111 found in the table indi-
cates rainfalls of less than 0.01 inch. Data reflects rainfall
for the 24 hours preceding the date on which the data was re-
corded. For instance the rainfall recorded for January 1, 1970
occurred between 0800, December 31, 1969 and 0800 January 1, 1970.
Figures 21 and 22 are bar graphs of the average rainfall re-
corded in Tables 10 and 11. The star symbols mark those periods
of time when overflows occurred in the drainage system. Un-
fortunately, the exact periods of overflow during 1969 were not
on record, but the rainfall data is presented for comparison.
Some anomalies can be noted on the bar chart. For instance,
although a rainfall which averaged 0.97 inches over the test
area produced overflows on March 20, 1970, a rainfall of 1.44
inches on March 17, 1970 did not cause overflows.
The problem of relating rainfall to flow in a sanitary sew-
er is not equivalent to relating rainfall to flow in combined or
storm sewer. In the case of a storm or combined sewer, a unit
hydrograph is constructed using the measured drainage area, rain-
fall intensity and a coefficient for runoff dependent on terrain
and ground cover. Infiltration, on the other hand, is dep-
endent not only on those variables listed above but on the type
of soil and the history of rainfall and temperature for some
period of time preceding the overflow. A qualitative example
of this is the comparison of the lags between rainfall and over-
flow start for two similar rains, May 31 and September 2. The
68

-------
overflow of May 31 started approximately six hours after the
rainfall began, but the lag on September 2 was only three hours.
A possible explanation lies in the difference of condition
of the ground at the two periods. The soil is largely a plastic
clay over weathered limestone. In the spring, the clay is swoll-
en and forms a relatively impermeable surface. On the other hand,
by September the ground is dry and cracked to considerable depth.
This same shrinkage can cause severe problems if a shallow line,
such as a house lateral, is not properly installed. The shrink-
age cracks provide a ready passage of water to the limestone,
which is relatively permeable.
A simple regression analysis was performed using the limit-
ed data available. The results of the analysis led to the
fol lowing concl us ions:
A. A total rainfall of at least 3.5 - 4.0 inches in any
20 day period will cause overflows.
B. A total rainfall of at least 4.5 - 5.0 inches in any
30 day period will cause overflows.
Further analyses will be possible as additional data is
gathered.
For future work, another rain gauging station has been in-
stalled in the northwest section of the drainage area. It is
anticipated that a more rational relationship between rainfall
and overflow can be developed with additional data over the life
of the project.
Table 12 relates the volume and duration of overflow with
the dates on which rainfall covered such overflows. It can be
seen that the period during which overflows persisted for the
various storms ranged from 9 to 39 hours. It should also be
noted that the overflow “patterns”, although similar, were not
the same for every flooding condition.
69

-------
I - PR& MI/iwifir /4 /If Cr/ON awr
RAm ’ CA ./ (
(IPSTREAM
EVEL SENSOR
0
YMER
I .
I
Oo vNsr.qsAM
LEVEL
SENSOR
2
\\ I
/ ( ‘8 qPNe q
\ J 8 / é’ (/ c 4 ’(
/
. .V WL L c &T7
//
M4’ c af w;
/
/
Ji #PE 20 if

-------
FIGURE 20—B

-------
DISTANCE (X io2 ft)
FIGURE 20-C PROFILE BACHMAN CREEK 18° LINE
.1.
4 )
4-
r ) c
I—
w
-J
w

-------
TABLE 10.
19 6 9RAJNFALLRECQRD(Values In Inches)
Weather
Month
Day
Station Number
Bureau
Average
27
41
43
January
February
March
April
May
June
16
29
30
1
14
15
20
21
22
3
6
8
15
16
18
23
24
25
31
5
13
16
17
27
5
7
8
9
15
17
18
26
27
29
1
4
24
0. 05
0. 00
2. 39
0. 07
0. 3
1.12
0. 3
0.46
0.45
0.45
0. 30
0.34
1.00
0. 85
0. 77
0. 69
0. 10
0. 02
0.33
0.10
1.15
0. 0
0. 60
1.19
2.38
4.50
0.35
0. 06
0.59
0.49
0.40
0. 0
0. 76
0. 60
0. 0
0. 25
0. 09
0.10
0. 14
1.70
0 10
0. 33
1.00
0.13
0.50
0.44
0.51
0. 29
0. 20
1.01
1.02
0.80
0.70
0.10
0.17
0.20
0.06
1.15
0. 0
0. 60
0.86
1.98
4.50
0.86
0.03
0.50
0. 65
0.50
0. 02
0. 78
0.15
0.04
0.27
0.10
0.13
0. 1
0.86
0. 14
0.36
0.94
0. 14
0.43
0. 52
0.44
0.34
0.19
0.80
0.84
0. 79
0. 75
0.12
0. 0
0.32
0. 15
1.05
0. 65
0. 0
1.38
2.26
5. 60
1.10
0.22
0; 52
0. 73
0. 63
0.32
0. 09
0.08
0. 06
0.30
0. 11
0.04
0. 1
1.90
0. 09
0. 32
0. 96
0. 09
0.49
0.34
0.40
0. 32
0.22
0. 55
0. 67
0. 68
0. 35
0. 39
0. 02
0. 14
0. 15
0.92
0. 0
0.40
1.14
2.05
4.96
0.46
0. 02
0. 25
0. 66
0.44
0.04
0.04
0. 09
0. 02
0.28
0. 10
0. 08
0. 85
1.71
0.10
0. 26
1. 01
0.10
0.47
0.44
0.45
0.31
0.24
0.84
0. 85
0. 76
0. 62
0. 18
0. 05
0.24
0. 11
1.07
0.16
0.40
1.14
2.17
4.89
0. 69
0.08
0.47
0. 63
0.49
0. 10
0.42
0. 23
0. 03
0. 28
0.10
73

-------
TABLE O
(continued) 1969 RAINFALL RECORD
Weather
Month
Day
Station Number
Bureau
Average
27
41
43
July
August
September
October
November
December
0. 0
1.05
0. 20
0. 27
0.45
0.78
0. 57
0. 20
0.45
0.12
0. 18
0. 0
1. 35
0.11
2.30
2.00
0. 58
0.86
1.71
0.05
0.52
0. 66
0.27
0.98
0.51
0.04
1.28
0.07
5
16
24
26
3
4
8
9
11
17
19
23
5
12
13
28
29
30
3
4
17
27
6
7
28
29
30
0. 0
0. 68
0.26
0.25
0.15
0.40
0. 60
0.33
0.20
0. 19
0. 0
0.33
1. 68
0.04
2.25
2.30
0. 63
0.22
2.17
0.15
0.41
0.47
0.20
1.02
0.50
0. 10
1.60
0.10
0. 0
1.75
0.44
0. 0
0. 74
0.92
0. 65
0. 0
0.25
0. 13
1.05
0. 0
2. 10
0. 04
0.96
2. 66
0. 75
0.26
1.83
0. 12
0. 56
0.95
0.22
1.25
0.59
0. 09
1.55
0.17
0. 0
1.69
0. 59
0.09
0. 0
0.98
0. 39
0. 08
0. 0
0. 0
0. 0
0. 0
1.71
0. 03
2.26
2. 13
0. 73
0.24
1.87
0. 11
0.43
0. 57
0.20
0.81
0. 50
0.11
1.63
0.14
0. 0
1. 29
0. 37
0. 15
0. 33
0. 77
0. 55
0. 15
0. 23
0.11
0.31
0. 08
1.71
0. 06
1.94
2. 27
0. 67
0.40
1.90
0. 11
0.48
0. 66
0. 22
1.02
0. 53
0. 09
1.52
0. 12
74

-------
TABLE 11 1970 RAINFALL RECORD (Values In Inches)
I 1
Month Day Station Number
41 43
Weather
Bureau
Average
January
February
March
April
May
2
5
6
1
2
6
7
8
15
16
23
24
25
28
2
3
1 1
12
17
19
21
10
16
17
19
25
26
29
30
31
I
23
27
28
30
31
0.22
0. 12
0.40
1.33
0.25
0.04
0.48
0.0
0.65
0.02
0.28
1.05
1.04
0.85
0.02
0. 60
0. 18
0.40
0. 20
1. 48
0.08
0.95
0. 10
0. 34
0.50
0.95
0.89
1.32
1.00
0.0
0.0
0.43
0.05
0.45
0.95
0.22
2. 0 5
0. 14
0. 12
0.42
1. 50
0. 13
0. 06
0. 52
0.08
0. 55
0.01
0.40
0.75
1. 10
0.75
0.03
0.85
0. 23
0.45
0.16
1. 10
0.06
0.92
0.14
0.45
0. 39
0.80
0.68
1. 50
0.0
0.0
0.0
0.45
0.0
0.44
0. 57
0.72
1. 50
0.05
0.33
0.34
1.31
0.0
0. 04
0.55
0.0
0.45
0. 20
0. 35
0. 68
1.35
0.76
0.0
1.25
0. 20
0.49
0. 25
1.85
0.05
1. 14
0.15
0.35
0.42
0.61
0.92
1. 80
0.0
0. 36
0.0
0.42
0.04
1.00
0.94
0.65
13.09
0. 15
0. 14
0.36
1.12
0. 28
0.03
0.48
0.02
0.39
0. 19
0.22
0.60
1.05
0.68
0.02
0.75
0. 14
0.38
0. 14
1. 32
0.03
0. 87
0. 10
0.31
0.13
0.81
0.81
1. 24
0.0
0.21
0.31
0.31
0. 12
0. 12
0.68
0.19
1. 96
0. 14
0. 18
0. 38
1,32
0. 16
0.04
0.51
0.02
0.51
0.11
0.31
0.77
1. 13
0.76
0.02
0.86
0.18
0.41
0.19
1 . 44
0.05
0.97
0.12
0.36
0.36
0.68
0.83
1.46
0.25
0. 28
0.08
0.40
0.05
0. 50
0.78
0.44
2.15
75

-------
TABLE 11 (continued) 1970 RAINFALL RECORD
Weather
Month
Day
Station Number
Bureau
Average
27
41
43
June
July
August
September
October
November
December
1
5
21
23
0.0
0.0
0.0
0.42
0.07
0.0
0.0
0.0
11
12
13
21
25
0.0
0.0
0.0
0.02
0.0
0.0
0.0
0.25
0.0
0.0
19
20
23
30
0.0
3.02
1.98
0.0
0.0
1.90
1.90
0.01
1
2
3
14
17
18
21
22
23
26
27
0.70
3.56
0.45
0.33
0.0
0.0
0.0
0.0
1.10
0.60
0.03
0.50
3.20
0.0
0.45
1.30
0.04
0.0
0.0
1.30
0.50
0.0
6
8
9
12
18
24
26
27
0.16
0.0
0.0
1.52
0.0
1.11
0.0
0.0
0.0
0.0
1.00
1.33
0.03
1.05
0.05
0.0
14
0.51
0.40
16
21
30
0.0
0.05
0.77
0.0
0,30
0.72
0. 20
0.02
0.23
0.45
0.0
0.11
0.27
0.0
0.06
0.0
2. 68
1.74
0.69
0.64
1.95
0.0
0.70
1.76
0.05
0.0
0.17
1.96
0.68
0.0
0.02
0.0
0.24
1. 27
0.02
1. 15
0.0
0.0
0.35
0.18
0.04
0.82
0.0
0.0
0.19
0.48
0.01
0.05
0.25
0.0
0. 20
0.95
1.09
1.61
0.73
0.55
2.15
0.04
0.72
0.72
0.0
0.03
0. 28
1.95
0.65
0.01
0. 07
0.03
0.34
1. 60
0.01
0.94
0.0
0.02
0. 32
0. 20
0. 0
0. 73
0.06
T
0. 10
0. 34
T
0. 04
0. 19
T
0.06
0.24
2. 17
1.81
0.36
0.60
2.71
0. 13
0.55
0.94
0.02
T
0.11
1. 58
0.61
T
0.06
T
0.39
1. 43
T
1.06
T
T
0.40
0. 10
0. 10
0. 76
76

-------
5—
4
3
U,
Q)
U
a -
I-
4-
• r
(t
1 - A M J J 0 N D
Figure 2L RairifaU Record - Bachman Watershed ( January - December 1969)
Data from Four Stations

-------
+ Overflow
5
4
0
-
cr -
+
4- 4
. 1
+
+
+
1 + - ____ ___
Jan Feb I Mar 1 Apr May Jun I lul Aug 1 Sept Oct Nov I Dec
Figure 22. Rainfall Record - Bachman Watershed (January - December 1970)
Data From Four Stations

-------
TABLE 1 2,
OBSERVED OVERFLOWS IN THE BACHMAN CREEK
DURING CALENDAR YEAR 1970
Showing Peak Rate Observed (gpm )
The symbol 4/ indicates the occurrence of
an event with no quantitative record.
The above table is a chronological list of
activity at observed overflow points. The
activities listed are for the following
days in calendar year 1970, respectively:
Feb. 25, March 21, April 25, May 30,
September 2 and SeDtember 23.
Duration (hrs)
WATERS HED
(See I’ T ote)
- .1
LO
Overflow Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
--
#
#
--
w
--
#
#
--
--
--
--
--
--
--
ZOO
6
20
18
R
‘
0

0
0
0
0
R
500
f
300
18
4’
W
‘
#
400

30
T
0
O
R
0

0

0
o
//
R
600

#
T
#
‘
#
#
250
39
‘
75
- -
/ 0
r
R
0

0

0

0

R
300
32
R
//
‘
‘
179
22
R
#
T
0 0

R
0

0

0

0

R
100
4
R
V
4’
‘
50

R
50
T
0
0
0
0
R
0
0
0

0

0

R
1000
6
R
200
W
200
W
200
1/
200
#
250
14
R
R indicates overflow removed or closed.

-------
SECTION 11
HYDRAULIC LINE ANALYSIS AND COMPUTER MODELING
GENERAL
Technical literature presents a number of ways by which the
relationship between flow, conduit properties and energy loss can
be expressed. In most cases, these expressions differ only by
the empirically-derived constants applied to make the mathemati-
cal formula and the physical model consistent.
Normal application of flow equations by engineers is through
the use of nomographs and tables, with occasional spot calcula-
tions to check the results obtained. Analysis of the flow system
of this project was accomplished through the use of nomographs,
tables, and direct calculation, but in addition, electronic pro-
cessing was used to permit more analysis, with an eye toward the
solution of a general “Branched netwOrk flow problem.
The basic equation chosen for use in analysis and modeling
is the Hazen-Williams Formula*expresSed in the form:
v=C 1 r° 63 s 0 ’ 54 o.ooi_0.04 (24)
v= average velocity in feet per second
I ,
C, a coefficient 0 f’roughflesS
r a/phydraulic radius , feet
s= energy loss per foot of pipe.
This equation can be transformed into the familiar “power
law” eSuation for round pipes by making the following sub -
stitutiOfls 2
Q= 1/4 lTD x vx 448.86
D/4 R (round pipes flowing full)
h/L=s
where:
Q= flow, in gallons per minute
D= pipe diameter in feet
h= head loss due to friction in feet of water
L length of pipe
“Handbook of HydraulIcs t ’, 4th Edition, 1954, McGraw-Hill.
80

-------
The resulting equation is: h= [ 9.76 x lO 5 L ] .85 25
C’ 85 D 4.865 (
The quantity enclosed in the bracket is a constant assianed
the name “K” for a given length of pipe; therefore, the fric-
tion loss in a length of pipe is proportional to the flow in
the pipe raised to a “power”, hence, a “power—law” equation.
It should be kept always in mind that dilute polymer sol-
utions do not obey this equation, since the equation was based
on experimental data for water, a Newtonian fluid. Therefore,
the computer model discussed in the following is only appli-
cable to a system before polymer addition is made. The com-
plex problem of calculating the effect of polymer addition in
flowing systems is discussed in Section 4 of this report.
The values of C 1 used in the analysis are shown below:
Pipe & Condition c 1
Extremely smooth, staight 140
Very smooth 130
Vitrified 110
Concrete 100
Tuberculated concrete 80
Small, rough concrete 60
Once a method of calculation was chosen, it was necessary to
make assumptions concerning quantities of sewage normally input
Into the line and the infiltration conditions which produce
overflows.
Table 13, “Land Use Parameters Used In Model”, was extract-
ed from WPCF Manual of Practice 9. “Design and Construction of
Sanitary and Storm Sewers” (1969). The average flows given in
this table are representative of the average design flows used by
the City of Dallas. For ease in programming, no variation with
tributary area was considered. The demographic data concerning
average-to-peak flow ratios shown in Figure 1 were also extracted
from the above reference and corresponds with the criteria es-
tablished for the Dallas systems.
COMPUTER MODELING
The purpose of modeling is to accurately simulate, in as
many ways as possible, the behavior of a physical system under
any perturbation desired, or required. The model developed
under this program is a mathematical representation of a “branch-
ed-network” flow problem. A branched network differs from a
closed network (as represented by a water distribution system)
81

-------
TABLE 13 LAND USE PARAMETERS USED IN MODEL
Land Use
Average
Density
(residential
uses in per-
sons per acre)
Average
Gallons/Day!
Acre
Flow (Q)
Mgd
Sq. Mile
Rural or Conservation
2. 5
250
0. 160
Institutional
-
500
0. 320
Low Density
Residential
5.5
550
0.352
Medium Density
Residential
9.5
950
0.608
Commercial
-
1500
0, 960
Medium-High Density
Residential
16
1600
1.024
High Density
Residential
25
2500
1.600
Industrial
-
4000
2. 560
82

-------
— - :I
- - -
- . 1
±IIL 1
+
H
. t ! ! :.
fI
i rLll
1:11
ii

I
tt
Th
I.
fli 1::
R’
+ plow
Ej
20
‘
r
:1
I I I
40 60 100
:1
.4 .—
II
1l j 1
• “
I Iii IlIr
200
I I I
400 600 1000
Population, in thousands
Figure 23
Ratio of Peak to Average Sewer Flows
Versus Population (Based on National
Averages)
U
-Il
.4- .
4,,-
ff Th t
0
a)
(D
C l )
(d
a)
I-I
a)
(
0
.4-,
a)
co
( )
- -4
--4
. 4 ,
0
0
-‘-I
4-,
c c,
-4
1:1
10
8
6
4
2
1
0.8
0.6
04
0.2
01
ft
-t -
-4-
-
4 4
44 .
n
U -
141
141 1-4
‘:41
Ht14T4
lilt
44.
4
8 10

-------
in the increased number of restraints and conditions as input
data. The model also differs from that used for design* in
that the operator does not have the freedom of generating pipe
sizes, slopes and geometry.
The computer program presented here is designed as an ‘on-
line” Fortran program, requiring the attendance of a person
knowledgeable of the purpose and method of the program to fur-
nish additional data or to change input parameters. The program
was prepared and run using the GE time-sharing computer service.
The following discussion Is presented as an example and user’s
guide.
INDEXING CONVENTION
The flow network is first broken into “lines” as shown in
Figure 24 with the index “1” being assigned to the most down-
stream line. The lines are then numbered, in order, by assign-
ing even numbers to the deadend branch lines and odd numbers to
those lines which are joined to other lines at their upstream end.
The next step in subdivision is to assign numbers to each
node or “entry” along the previously defined line, starting at
the most downstream end with “1” and proceeding upstream. An
“entry” is required for each input from a lateral line, each
change in pipe size or characteristic (e. g., roughness), and
for junctions with other “lines”. It should be noted that later-
als or lateral lines are distinct from “lines”.
The last index is that “serial”, which is either “1” or “2”.
The serial differentiates between two laterals entering at the
same entry. In the analysis of the Bachman Trunk Sewer, a “1”
indicates a lateral entering from the north or west and a “2”
indicates a lateral entering from east or south.
GEOMETRIC RESTRICTIONS
Constructed overflows are treated as special cases of later-
al lines.
The program is designed to accept a sewer network consist-
ing of one trunk line fed by “n” dead-end branch lines as illus-
trated in Figure 24 The number of branches (“n”) is limited only
by the capacity of the computer used to process the problem.
A branch line may not be split into sub branches.
* Zepp, Paul L., “A Computer Program for Sewer Design and Cost
Estimation”, Regional Planning Council, 701 St. Paul Street,
Baltimore, Maryland
21202 (April 1969).
84

-------
-Entry
3
(For Lines
& 3)
This lateral can be
identified by the
indices (5, 3, 1)
\
\
\
(Ti
\
3
-(For Line
\
‘ 1f2e 3
1)
7’
Figure 24. Example Flow Network Showing Indexing Convention

-------
To simplify processing, no more than two laterals may feed any
numbered entry. If more than two inputs are desired, two or
more adjacently numbered eptries may be defined to be connected
by pipes of length 1 x 10 3 feet.
- INPUT DATA REQUIREMENTS
Input Is accomplished by entering data into permanent”files”
(time-sharing systems) or “tapes” (batch systems). The required
data is defined in groups in the order required for processing.
The first data input describes the geometry of the system to be
analyzed. It consists of the number of lines, and then for each
line: (1) the line number; (2) the number of entries on the line
and (3) a “0” If the line dead-ends or a “1” if the line is join-
ed by other lines at its extreme.
The second data input describes the tributary areas along the lines
In the system. For each lateral or constructed overflow along
each of the lines the following data is required:
A. The line, entry, and serial identifying the lateral
(Or overflow).
B. The area of the tributary area in acres.
C. The land use factor In gallons/day/acre as shown in
Table 13.
D. The critical, limiting or overtlow elevation on the
lateral or overflow.
E. The estimated or measured total collector pipe length
In the tributary area.
F. The design roughness (C in Hazen-Williams Formula)
of the connected lateral.
G. The diameter and length of pipe connecting the point
which was determined to be critical to the trunk.
H. The probability (0 to 1) that any given tributary area
will contribute to infiltratiob in proportion to the
length of the collection system In the area. This input
requires engineering judgment, assigned by knowledge of
code restrictions, construction techniques, and field
measurements.
The third set of Input data consists of information to describe
the trunk line:
A. The line and entry numbers of the upstream end of the
pipe being described.
86

-------
B. The Hazen-Williams coefficient of roughness, the diam-
eter and the pipe length to the next downstream entry.
C. The elevation of the invertat the entry under consider-
ation.
D. An allowance for head lose due to bends, restrictions.
manholes, and grade changes.
THE PROBLEM SOLUTION
The flow problem is solved using a limited iteration technique,
limited in the sense that the operator of the program has the
option to continue interation, stop, or change the problem at
regular intervals during processing.
The processing proceeds as follow:
A. All data is entered; calculation of constants is per-
formed.
B. The flows in each lateral and in each section of the
trunk line under normal design peak daily flows are
cal culated.
C. The hydraulic heads at each entry and lateral node
are calculated from the flows.
D. The calculated elevations are checked against over-
flow elevations.
E. Identifying numbers, flows in all branches, elevations
and overflow rates are printed.
F. The operator enters an infiltration increment.
G. The infiltration increment is applied to all tributar-
ies and tributary flows calculated.
H. System flows and elevations are recalculated.
I. Elevations above critical points cause calculations
of overflows and reduced inputs.
J. System flows and elevations are recalculated and
checked for compatibility.
K. Steps I and J are performed three times, output is
printed, and the operator is queried for the option
to continue the problem or stop.
87

-------
Two successive outputs which are similar or within
operator determined accuracy criteria are used for a stop or
change problem decision.
A listing of the program is included in Appendix C.
The program shown, which is dimensioned to fit the analysis
of the actual system is the largest which can be processed by
a computer with a core capacity of 65,000 words. A change In
the program to accept a larger problem will require a larger
core.
The input data established for the study area is in-
cluded as Appendix D.
Appendix E is a run of the problem with the follow-
ing assumptions:
A. The ratio of peak-to-average flow is 1.44.
B. The probabilities of infiltration along the two
branch lines are approximately equal.
C. The allowed infiltration rates (average) are 0,
.008, .01, and .015 gallons/minute/foot of lateral.
The first output is a printout after five iteration
cycles of a peak daily flow. The second output Is a print-
out of the system flows at an infiltration rate of .008 after
five cycles. More cycles of iteration would eliminate the
overflows. The third output are the system flows after five
cycles with an infiltration rate of .01. The fourth output
is the result of five iterative cycles at an infiltration
rate of .015. The fifth and last printout shows the effect
of five additional iterative cycles with no change in the
infiltration rate. This final model is a good represent-
ation of the system under general overflow conditions.
The program as presented was designed for use at full-
pipe flow conditions and should not be used for computations
of varying or uniform flow in partially filled pipes.
88

-------
SECTION 12
PRELIMINARY INSTRUMENTATION AND FLOW MEASUREMENT
GENERAL
In order to provide a basis for the design of injection tests
and to demonstrate the effect of the injections, measurement
of flows, overflows, and hydraulic heads is required. The re-
quired and desirable characteristics of the measuring elements
were defined as follows:
Characteristics and desired features:
A. Ruggedness - As a field instrument, able to with-
stand rough treatment before and during installa-
tion. Preferably able to operate after submergence
in sewage.
B. Capability to measure flow in a surcharged line —
Efficient use of polymer materials require higher
than normal velocities; hence, surcharging is allowed.
C. Non-fouling - Experiments with standpipe and floats
in manholes indicated that projections foul rapidly.
D. Accurate - Precision to the limits of the method used.
E. Simple - For servicing requirements.
F. Recording - For unattended long-term measurements.
G. Easily interpretable data - Direct read-out in the
units required is preferable.
H. Semi-portable - To make movement from one measuring
point to another possible.
I. Adaptable - For mounting in conventional manholes, or
remote with taps sealed into the line.
The characteristics and features described above resulted from
the need to measure flow, pressure, or head under the following
conditions:
89

-------
4. Flow and/or head measurements through conventional
and type manholes.
B. Measurement of vertical flow out of specially-con-
structed overflow manholes with hinged lids.
C. Flow measurement in overflow pipes (horizontal)
ranging in size from 4-inch to 12-inch diameter from
beginning of flow to full pipe.
D. Flow measurement in vertical 4-inch overflows.
The instrument market was surveyed, information was solicited
from all the major equipment companies and many of the smaller
companies. The survey exposed the following general types of
flow and head measuring devices:
Float and Stilling Basin
Head measurement by mechanical means, limited to open
installations, extreme accuracy possible, subject to foul-
i ng.
Purge Tube Pressure
Head or pressure measurement by measuring the pressure
required to discharge bubbles against the head, with out-
put converted to mechanical or electrical output.
Ultrasonic Depth Measuremej
Measures distance from a known elevation to a liquid-air
interface, sophisticated method.
Hook Gauge
Manual or servo-operated to determine surface elevation
of a stilling basin, limited in range by mechanical con-
siderationS, subject to fouling.
Sounding Rod
Manual, for use in a stilling ba’sin.
Exposed Diaphragm Pressure transducer
Used for level in tanks, electrical output.
90

-------
Weir
Flow measurement, constructed or prefabricated and used
with depthrneasuring device, in open channels.
Venturi
Flow measurement, restricted to small diameters because
of required proportions, required full cross-sectional
flow.
Flume
Flow measurement, in open channels.
Propeller-type Steam Meter
Velocity only, requires a minimum stream depth and is sub-
ject to fouling.
Ultrasonic Doppler-effect Meter
In-stream velocity measurement, subject to fouling, sophis-
ticated method.
Psuedo—Sound Listenina Meter
Relative flow by turbulent noise generation, newly-develop-
ed.
Orifice plate
Flow in full pipe, high energy losses.
Dilution Meter (e.g., Fluorometer )
Requires constant rate injection, subject to fouling and
interfering substances.
Magnetic Flow Meter
Flow in full pipe, limited to 24-inch and smaller, sophis-
ticated.
91

-------
Turbine Meter
Flow in full pipe, subject to rapid fouling.
There are, in addition, combinations of and additions to the
above list if methods, rather than devices, are considered; but
all of the devices in common use, fall in one of the above
categories.
A question was raised early in the program concerning the use
of the fluorometer, which had been used in an earlier polymer
program. Experience with that device had shown that some of
the constituents in sewage, and even the pipe wall, can intro-
duce serious errors into flow measurement. For instance,
grease fouls the transmission cell wall, suspended colloids
are dyed by the injected chemical*and the pipe walls can absorb
the dye.
A survey of consulting engineers concerned with doing sewer
surveys and flow studies uncovered some interesting informa-
tion about the quality of flow measurements which they per-
formed for their customers. The majority of flow tests per-
formed use an empirical formula in combination with estimates
of line condition, construction data, and water depth measure-
ments to determine flows. The consultants opinion of the
accuracy of these methods vary from estimated error of 10- to 50-
percent.
The factor which seems to be susceptible to error in the cal-
culation of flow is the line condition or roughness, wnich co-
incidentally is the factor which also possesses a large sensi-
tivity in most flow equations.
Based on the above considerations, it was decided to use float-
type level recorders and dip-sticks in combination with tables
and nomographs to supplement a system of interconnected bubble-
purge level meters to be installed on the sewer trunk.
The bubble purge level meters are a variation on a system used
by some engineers for field survey work. The system consists
of a bottle of liqufied carbon dioxide discharging gaseous
carbon dioxide through a regulator into a tube equipped with a
sensitive pressure gauge. The devices built for this project
make use of the fact that a bottle of liquid carbon dioxide
holds essentially a constant pressure so that a constant rate
of discharge can be obtained by venting through an orifice . —
*Buchtela, K., et al, “Comparative Investigations Into Recent
Methods of Tracing Subterranean Water”, National Speleological
Society Bulletin Vol. 30, No. 3, July 1968 (70j.
92

-------
The pressure gauge was replaced with a semiconductor strain-gauge
pressure transducer and associated circuitry to provide an analog
signal proportional to the depth of submergence of the purge tube.
Figure 25shows one of the devices installed in a manhole.
The signal was brought out by drilling a hole through the man-
hole wall near the top, through which a cable was passed. Be-
cause of the remote locations of the measuring stations, it
was not possible to use existing telephone lines, so cables
were trenched into the ground, run overhead, or buried in pave-
ment as required. The overhead lines for which natural support
was used were the most troublesome in that breakage sometimes
occurred during the windstorm which accompany many of the
thunderstorms in the test area. The buried lines have given no
problems.
The signal cables from groups of measuring points were brought
to centrally-located recording stations. One of the stations
is shown in Figure 26. This station consists. of a power supply
to provide the operating voltage to all of the remote sensing
locations, a signal timer to sequentially connect each signal
source to a recorder, and a single-channel strip chart recorder.
The multiple signals are recorded side-by-side with a calibra-
tion signal and a zero check. An example of the record is
shown in Figure 27.
In those cases where a convenient entry, such as a manhole,
was not available, the level sensors were chained to convenient
structures or trees and the purge tube was run to the sensing
location. In small pipes, the tube end was sealed in place.
For type “S’ manholes (pressure manholes) in the stream bed,
iron pipe standards were welded into the manhole covers and
guyed to the bank to guard against the bombardment of flotsam
during flood stage. The purge tube was then inserted into
these standpipes.
The instruments which were exposed to gross temperature changes
required a n dification to Isolate the pressure transducer from
thermal stresses in the support. This was done by mounting the
transducer in a material having a low coefficient of thermal
expansion. This modification was not necessary for those
instruments mounted in the relatively constant environments
of a conventional manhole.
One additional method of flow measurement which was used during
the field survey is the “salt velocity” method. This technique
has been described by John Schmidt* for use in determining
discharge coeefficients and is very simple to apply to field
measurement problems. The technique avoids the clogging and
mechanical interference caused by solids when a rotating-cup
velocity meter is used.
93

-------
It should be kept in mind that the methods chosen for flow”
measurements during the preliminary injection tests did not
have the capability of performing during the tests, only be-
fore and after injections. This is because assumptions made
in designing most water flow-measuring devices assume the
properties of the fluid as Newtonian, a necessary condition
which is violated when the flow is non-Newtonian .
* Schmidt, 0. John, “Determination of Discharge Coefficients
by the Salt-Velocity Method”Journal Water Pollution Control
Federation, 1969.
94

-------
Fig. 25. Purge-Tube Level Meter Installed in Conventional
Manhole
95

-------
Fig. 26. A Multiple-Input Recording Station
96

-------
“ 4 •
• • ‘ t
.4 : :t•.
I - ‘
Fig. 27 Example Recording Showing A Building and Receding Storm Flow
I
I

: , ;:‘ : ‘ -:
- ; - : ,. .
¼0
-.1
S
t
-i
- J
1
i
4
a -
_‘t .-\

-------
SECTION 13
RESULT OF POLYMER INJECTION IN
SURCHARGED GRAVITY LINES
The purposes of performing polymer injection tests were as
follows:
1. To verify that the overflows from the Bachman trunk
sewer could be eliminated or reduced by the injection
of friction-reducing chemicals.
2. To determine design criteria concerning the injection
rate for the permanent injection station.
3. To establish a suitable location for the permanent
injection station.
Two injection locations were chosen for the 15” and 18” branches
of the trunk sewer, one for the most desirable location and one
to check the effect of injecting the polymer far upstream of the
required line section. One injection location on the 24” line
was chosen to check the effectiveness of the polymer after being
subjected to the destructive forces of a long run of pipe.
These five injection locations are indicated on the area map of
Figure 20 A. A secondary consideration in the selection of the
temporary injection points was the presence of an existing entry
into the line.
All data gathered during the early stages of the program in-
dicated that the problem line section was the 2,590 feet of 24”
line at the upper end of Bachnian Trunk and the 18” line of
Bachman Branch. This indicated the need for injection on the
Bachnian Branch, a location which would also satisfy the require-
ment for friction reduction in the 24” line.
Figure 28 is a surcharge-time plot for a complete storm flow as
recorded by one of the monitoring stations on March 20, 1970.
The average rainfall over the test area was less than one inch
and the rainfall was of low intensity so that the rise and fall
are gradual. The overflows caused by this storm were short-
lived and of low total volume.
98

-------
Conversely, the rainfall of April 25 was short ançl intense and
produced the flood waves shown in Figures 29 and 30. Notable
is the quick rise and the long decay recorded on both of these
graphs. The graph of Figure2g. was produced by a station about
8300 feet upstream of the station which produced the record of
Figure 30 . The lag in the front of the wave is indicative of
the distance between the two points on the line. The results
of an injection is shown on Figure 29 about hour 15. The same
injection appears on Figure 3 O at hour 17. The storm peaks
shown in Figures 29 and 30 are typical and represent flows at
the downstream measuring point of 10.7 MGD at peak. This com-
pares to a design of about 10.7 MGD with new pipe conditions.
Table 14 gives the dates, injection locations, injection rates
and results in terms of maximum head reduction. The locations
are keyed to the map of Figure 20 The type of slurry used is
also noted.
The philosophy of the injection tests was to inject polymer at
the prescribed rate until the head reduction ceased, then stop
injection and allow the system to come back to equilibrium. This
procedure was repeated at least once if possible to quarantee
that any reduction in head was truly related to the injection
of polymer and not caused by a coincidental phenomenon.
Figures 31, 32, and 33 show the results of some of the tests
as observed from the location noted on the figures.
Injections at locations four and five were not performed.
There were no significant surcharges or overflows on this branch
line during the conduct of the program.
Figure 33 indicates that a polymer rate of 4.5 pounds/minute
is in excess of that required for actual control of overflows.
This injection halted all overflows on the Bachman Trunk and the
Bachman Branch. The most efficient injection rate would be that
which holds the head at a safe” level below the overflow.
Injection of sufficient quantities of material to put all the
flow in the conduit would not be economical for any purpose ex-
cept experimentation.
99

-------
TABLE 14 RESULTS OF POLYMER INJECTION TESTS
TEST NO.
DATE
INJECTION LOCATION
INJECTION RATE
(lbs. 1mm.)
FRE-INJECTION
FLOW RATE
(gprn)
HEAD REDUCTION__
MAGNITUDE
(ft.)
LOCATION
NUMBER
STATION
1 25 April 70 1 35+40’ 1. 2 5000 2 35+40’
2 30 May 70 1 35+40’ ABORTE) BECAUSE OF EQIJIPIv ENT FAILURE
3 2 Sept.70 I 35-f it ’ 1.5 4500 1.0 35+40’
4a 2 Sept. 70 1 35- 4Q ’ 2.25 4500 1. 75 35f40 ’
4b 2 Sept.70 1 35+40’ 2.25 4500 1.5 164+332
5a, b, c 2 Sept.70 2 170+ 662 2.25 8000 2. 0 164+332
6 23 Sept. 70 3 128+00’ 2.25 3600 Q• 5 35+40’
7 23 Sept.70 1 35+40’ 4•5 5000 4 29 35’
* A large portion of the polymer was lost through an intervening overflow
-wwAll overflow stopped
Stationing on Bachmari Branch Line
Stationing on l3achmari Trunk Line

-------
Time in Hours
Figure 28. Storm Stage at Station 164+33 ( Bachman Trunk at Northwest Highway)
on 0 March 1970
8
6
4
z
a
(U
0
‘C
(U
(U
,— (U
‘C
u
6
18
24 30
48

-------
Time in Hours
Figure 29. Storm Stage at Station 35 + 40 (Bachman Branch at Walnut Hill Lane)
on Z5 April 1970
48
8
6
4
2
. 1
a 4
w
0
0
a)
a)
,—. a)
0
‘I
6 12 18 24 30 36 42

-------
Figure 30 Storm Stage at Station 164 + 33 ( Bachman Trunk at Northwest Highway)
on 25Apr11 1970
S
6
4
2
C)
- -1
C)
0
-Q
4- ,
C)
C)
(0
0
u
0
C l )
12
Time in Hours
42
48

-------
C t )
0
- Q
4

457
456
Figure 31
a
————— Sta. 166 + 21.58
Sta. 164 + 33
a
0
0
0
0:,
0
0
0
Sb
Sc
S
6
8
9
7
Time in Hours
Results of Injection Test S on Heads at
Stations 164 + 33 and 166 + 21. S8 (Below Northwest Highway)
10
11

-------
I I I I I I i
12 1 2
Noon
Time in Hours
Figure 32. Results of Injection Tests 3, 4a, and 4b
(Bachman Branch at Walnut Hill Lane)
3 4 5 6
on Surcharges at Station 35+40
a)
a)
0
-Q
a)
a)
a)
01 )
0
0
C l )
9
8
7
6
5
4
3
2
1.
0
10
—
ii

-------
Overflow Level
Figure 33
Time
Effect of Polymer Injection Test 7 on 23 Sept. 1970
on Overflow at Station 29 + 35 (Bachman Branch below Walnut Hill Lane)
4
3
2
1
0
‘p
0.
--4
Ow
‘p
0
.0
4 -J
‘p
‘p
a)
I -I
0
C)
‘-4
C/D
I
/
1 300
I
Limit of Observation
I
1330 1400

-------
SECTION 14
RESULTS OF POLYMER INJECTION IN 6” FORCE MAIN
While the main emphasis of this program is the overflow reduc-
tion in surcharged gravity sewers, a portion of the program re-
quired the determination of head loss reduction or pump flow
increase possible through the use of polymers in a lift station.
The lift station selected for tests was the Riverwood Station
in East Dallas. The station serves a small residential area
generating daily flows of about 30,000 gallons. This flow is
sufficient to require the pump to operate for about 3 minutes
twice an hour normally and more often during peak periods. This
site was chosen because it could be instrumented readily and
effectively and was representative of the many “package plants”
in the city. The prefabricated lift station was installed
adjacent to a sump which stored the sewage between pumping cycles.
The sump was emptied by one of the two pumps, pumping the sewage
through a 6-inch cast iron pipe for some 2,000 feet over a hill
to a suitable collector line. Table 15 lists the nameplate data
on the pumps.
TABLE 15 . LIFT STATION PUMP DATA
Make Smith & Loveless
Size 4 B
Capacity 250 GPM
Speed 1760 RPM
Head 65 feet
Power 15 Horsepower
Phase 3
Voltage 220-440.
Current 58.6 - 19.5
A review of the design criteria indicates the following:
1. Design static head - 35 feet of water
- 15.2 psi
107

-------
TABLE 16 RESULTS OF LIFT STATION TESTS
( Pe f rmed between March 10 and March 19, 1971 )
Polymer
.
(lb/mm)
-
Application
Based On
Initial
Polymer
Conc.
(ppm)
Rate
Fin 1
Polymer
Conc.
(ppm)
Initial
Flow
(GPM)
Velocity
(ft/sec)
Discharge
Pressure
(psi)
Flow
Increase
(%)
0
0
0
260
2.95
26
0
.26
120
90
350
3.97
25
34.6
.4
185
125
385
4.37
24
48.1
.74
350
210
425
4.82
24
63.5

-------
2. Total head at 400 gpm - 62 feet of water
- 26.9 psi
A 6-inch Foxboro magnetic flow meter was installed in the pump
discharge line to accurately measure the flow. A Westinghouse
type 44 recording a Pmeter was connected to the pump leads to
measure current variations. A corporation stop was installed
in the pump suction line for the polymer application. The
lightweight variable flow polymer injection device was used
to apply polymer to the pump suction during tests.
The low normal flow of sewage to the sump was augmented by a
fire hose discharging into the last manhole on the line leading
to the sump. This augumented flow resulted in shorter emptying
cycle, with the pump operating every ten minutes. This allowed
more tests to be performed over a shorter period.
Fifteen pumping cycles were observed of the augmented flow with
no polymer. The observed flow rate, and discharge pressure gave
base line data upon which to compare the polymer data. It
should be noted that at the rated total head the discharge was
35% low. Table 16 , Results of Lift Station Tests, lists the
flow data both from the base line tests and the subsequent poly-
mer tests.
The polymer tests were performed with applications of .26, .4,
and .74 pounds per minute of polymer. Five tests each were per-
formed using .74 and .4 pounds per minute of polymer with the
results as shown on Table 16. The polymer feed line plugged
during the second tests on the .26 series, therefore, the re-
suits shown are for only one test.
Following the polymer tests, the system was allowed to purge the
polymer from the lines, and the base line data observed. The
non-polymer flow rate and pressure returned to that originally
measured.
The change in flow rate from polymer application increased the
electrical current draw of the pumps. The ammeter showed
normal flow to require 19.5 amperes as stated on the nameplate.
The highest polymer application, .74 pounds per minute, result-
ed in a current draw of 22 ampere: or an increase of 12.5%.
Figures 34 and 35 are reproductions of recorder charts which
show the flow rates with no polymer added and at the maximum
injection rate. Chart values should be multiplied by 2 to ob-
tain actual flow rates. The charts were run at an accelerated
rate so that a complete rotation occurs every 24 minutes
rather than 24 hours. This permitted the average flow rates
to be calculated from the volume of the pump from pump turn-on
level to pump turn-off level. In this manner, the recorded
output of the flowmeter was verified. It should be noted that
109

-------
the “spiking” recorded Is an actual phenomenon, apparently
an amplification of normal short-term oscillations around the
normal flow* , or a result of varying solution efficiency of
the injected slurry.
One notable aspect of this experiment is the relationship be-
tween flow rate and discharge pressure. When working in a
gravity sewer, it is usually desirable to cause a decrease in
head while holding the flow rate constant; on the other hand,
for polymer applications to force mains it is desirable to
increase pumping capacity for a given discharge pressure. This
is exactly the result shown by the data; significant increases
in flow rate with negligible changes in discharge head.
Figure 36 is a graph relating polymer injection rate to per-
cent flow increase. The bottom line represents the results of
the field tests, and the top line represents a laboratory
test of the same material in a six-inch diameter line. It
should be noted that although the shape of the graphs are
similar, the flow increases for a given injection rate were
significantly lower for the field tests.
*Rouse, Hunter, “Engineering Hydraulics”, John Wiley and
Sons, Inc., New York, 1950, p. 86.
110

-------
I I
Figure 34 3 mm.
No Polymer Added
Figure 35
Polymer added at 0.74 at 0.74 lbs/mm
(Maximum injection rate)
a.)
•1-
a)
(n
0
I-
I-
‘V
400
200
I .
I- W
4-’
600
U,
C
0
I-
‘V
C.! 200
2 mm. 2 mm. 2 mm.

-------
50 100 150 200
Polymer Concentration (parts per million)
Figure 36. COMPARISON OF RESULTS FOR
6” TEST AND FORCE MAIN TESTS
6” Test
Facility
6” Force
Main
250
140 —
120
100 -
80 -
60 —
40 -
20
a)
C .)
S.-
a)
a)
(0
a)
S.-
C-)
0
U-
112

-------
SECTION 15
THE CONSTRUCTED INJECTION STATION
GENERAL DESCRIPTION OF THE FACILITY
The Bachman Polymer Injection Station is a fully automated
facility for the storage, mixing, and injection of selected
turbulent friction-reducing chemicals which generally are
amorphous, water-soluble, high molecular weight polymeric sub-
stances. The facility can be divided into six functional groups
as follow (Refer to Figure 37.)
A. Polymer de-druniming - Consists of a 600 cfm centrifugal
blower driven by a 3 horsepower, three phase electric
motor, 50 feet of three inch diameter flexible hose
and a centrifugal separation system which is an integral
part of the storage hopper.
B. Polymer storage - Consists of a cylindrical tank with
a conical bottom, lined with urethane foam insulation
and equipped with a dessicant-type dehumidifier and
humidi stat.
C. Polymer feed and metering - Consists of a vibrating
cone hopper bottom and a helical screw volumetric feeder
equipped with a variable-speed DC drive motor.
0. Polymer dispersing and injection - Consists of a water
jet eductor, buffer tank, and a positive displacement
gear pump with a three-phase AC drive motor.
E. Control circuitry - Consists of motor starter control
relays, time sequencing relays and two DC motor speed
control s.
F. Instrumentation - Consists of one 18” magnetic flow-
meter, three bubble-type level transducers, one thermo-
couple temperature transducer, one elapsed time meter
and one six-channel scanning strip chart recorder.
OPERATING PRINCIPLES OF THE POLYMER INJECTION STATION
A. Normal Operation
113

-------
VENT
BAcl LOw VALVE
w/ G rrorc
FIGUI
ELEVATION OF POL R INJECTION STATION
LOCATING V )JOR COFPONENIS
HI64 LEVE L EMiOc
Low L VE L SEp4 .c,R
- oPp j o-rTO?A
w/ AGIT&TOR
SPP CE %E 1-ER:
VOLUMETRLC. r Ec’E
t g co
EMe c. j FLOOD
Curo wrr e
114

-------
s4 ‘ 55 S6 S7
IN E.c’ R Pw ’ p Lc we’IN SOUPS
C c c 5 i
I 2..
ED R.. Mo-roR _____
t-4ou METER
IL 11
Loc
LEVEL RELP LEu . Re LEVEL R t i
S(6 L CoMv v R.
FIGURE 38
SE E CH OF ViMN COt’ffROL PANEL
SH I THE LOCATION OF
CONTI JLS, ETERS, AND INDICATORS
115

-------
Operation of the polymer injection mechanism is
initiated by a rise in the level of the free water sur-
face at any one of the three level-measuring sensors -
upstream at Royal Lane, downstream at Brookview Lane,
and in the vault which also houses the primary flow
measuring apparatus. These levels are converted to a
4 to 20 milliampere signal and displayed on the three
Beede Meters located immediately below the multipoint
recorder on the instrument panel. (Refer to Figure
38) These three meters are “percent of full-scale”
indicators, with 100 percent corresponding to a head
of 150 inches of water above the sensing point. These
meters are equipped with two manually-set relay point
ers; green for low set-noint and red for high set-point.
The position of these pointers determine the indicated
level at which two normally open sets of relay contacts
will close. Figure 39 is a schematic of the process
control system.
During stand-by operation of the station, with the
“Run-Auto-Test t ’ switch in the “Auto” position, signals
are received from the three level transducers and the
magnetic flowmeter, but these signals are not recorded.
If the indicator pointer (black) of any of the three
meters passes the green pointer, the recorder is acti-
vated. The signals recorded and their symbols are:
(1) Flow; (2) Sewage Temperature; (3) Local Level
(4) upstream Level; (5) Downstream Level; and (6)
Polymer Feeder Speed.
The red pointers on the three level meters are always
set at a higher scale position than the green, and
their positions determine at what water level the
injection station goes into an “Active” status. When
the indicator pointer (black) passes any one of the
three red pointers, the station will start operation.
Operation will continue until all three indicators
are at a lower scale position than the red set-point,
assuming that none of the emergency shutdown devices
are activated.
Once a signal activates the station, the following
sequence is executed automatically:
1. The solenoid valve opens, allowing process water
to flow through the polymer dispersing eductor,
thereby setting up an air flow through the intake
port of the eductor, and simultaneously closing
the pressure switch on the feed water line.
116

-------
POLYMER HOPPER
VIBRATOR
SCREEN
VI BRAT OP
SHAKER SCREEN—I
HI
LEVEL -
ALARM
LOW
LEVEL
ALARM
18” SANITARY
SEWER LINE\
—SCREW FEEDER
—POLYMER DISCHARGE
.-LUMP SEPARATOR
FEED FUNNEL
LUMP DISCHARGE CHUTE
EDUCTOR
-MIXING TANK
PUMP + MOTOR
METER V 0 MER SLURRY
DISCHARGE
ii
DOWN-
STREAM
LEVEL
SENSOR
--7---
MAGNETIC FLOW
METER
LOCAL LEVEL
SENSOR
I i ic i
BEEDE INDICATORS
WATER
FLOW
UPSTREAM
LEVEL
E N 50 R
S C HEM AT I C
FIGURE 39
OF PROCESS CONTROL SYSTEM
FEED
CONTROL
PANEL
BACHMAN POLYMER J JECTION STATION
117

-------
2. When the pressure switch closes, a time-delay
relay (adjustable 0 to 180 seconds) will start.
3. At the end of the pre-set period, a time-delay
relay closes, activating the bin activator and
polymer feeder.
4. The speed of the polymer feeder is determined
by any of five manually selected signals. The
signal is selected with the large rotary switch
located on the control panel at the upper right.
If this switch is placed in the “Manual’ position,
feeder speed is determined by the potentiometer
to the left of the “Run-Auto-Test” switch. If
this switch is placed in the “Flow” position, the
feeder speed is determined by the output signal
of the magnetic flowmeter, and a constant propor-
tion of polymer to sewage flow will be maintained.
If the selector switch is placed in the “Up
stream”, “Local”, or “Downstream” positions, the
feeder speed will be determined by the corres-
ponding level signal. This would normally be
the mode of operation, and the position chosen is
determined by establishing the most critical
level which must be maintained. Regardless of the
source of the feeder control signal , the speed of
the feeder is indicated on the tachometer located
on the control panel, and the speed is recorded
as percent of full scale on the strip-chart re-
corder.
Note : The hour-meter located on the panel is operational
when the polymer feeder is activated. The read-
ing of this meter can be utilized as an aid to
determining polymer usage.
5. The injection pump is activated by the lower
sensor mounted in the side of the stainless
steel mixing tank. A thermal-delay relay is
incorporated in the starting circuit, so the pump
will not start until the sensor has been covered
by the process fluid for the timing period of
this relay. The pump will continue to run
until the fluid level drops below this sensor.
The pump speed is controlled by the potentio-
meter to the right of the “Run-Auto-Test” switch.
The pump will normally run at such a speed that
fluid is removed from the tank at the same rate
that it iS being added.
118

-------
6. When the level in the sewer drops below the level
which initiates the operation as described above,
the bin activator and polymer feeder will stop
immediately. The process water continues to run
for a period of time determined by a second time-
delay relay adjustable from 1 to 30 minutes in
order to wash down solids which may be clinging
to the sides of the tank. At the end of the
delay period, the water flow will stop (solenoid
valve closes) and the injection pump will stop
when the lower sensor is exposed.
7. The station now returns to a “Stand-By’ status.
B. Tests, Indicators, and Emergency Shutdown
The operation described as normal sequence above can
be initiated for test purposes by placing the “Run-
Auto-Test” switch in the “Run” position to start the
sequence. Turning the switch to the ‘Auto” position
will start the wash-down sequence.
Seven indicator lights on the main control panel indi-
cate the status of the station at any time. When the
station is in a “Standby” mode, none of the indicators
should be illuminated. The meanings of each light,
reading from left to right on the panel, are as follows:
1. “ Water ” - This indicator is lit at any time that the
solenoid valve should be open. Certain interlocking
safeties may stop the water flow but leave the indic-
ator lit. These are: (a) flooded sump; (b) over-
flowing mixing tank; Cc) no water pressure; and (d)
a full polymer mixing tank (upper sensor covered).
2. “ Bin” — This indicator is lit when the bin activator
should be running. A lack of water pressure for one
of the reasons enumerated above, or a stoppage in the
polymer disperser eductor will stop the bin activator
and leave this indicator illuminated.
3. “ Feeder” - This indicator is lit when the feeder
should be running. Any of the abnormal conditions
described under “Bin” will also stop the feeder.
4. “ Pump ” - This indicator is lit if the fluid level
in themixing tank is above the lower sensor and
the injection pump should be operating.
5. “ Low Bin” - If this indicator is on, there is
less than ten cubic feet of polymer In the storage
bin.
119

-------
6. “ High Bin” - If this indicator is on, the storage
bin is filled to or past, the recommended maximum
fill level.
7. “ Solids Level” - This indicator is illuminated if
the polymer disperser eductor will not accept the
polymer feed. That is, the feed funnel is plugged.
C. The Injection and Metering Vault
Because the sewer line into which the polymer is in-
jected is located in the bottom of a creek channel,
it was necessary to construct a vault to house the
required metering equipment. It is in this vault that
the connection of the discharge from the metering and
mixing equipment was made.
In joining the fiberglass-reinforced pipe to the con-
crete sewer line, a polyethylene heat shrinkable
tubing was used as discussed in EPA Report No.
“Heat Shrinkable Tubing for Sewer Pipe Joints.”
Figure 40 is a plan view of the injection and meter-
ing vault showing the location of major components,
including:
1. Temperature transmitter for sewage temperature.
2. Magnetic flowmeter for sewage flow.
3. Pressure transmitter for the “local level”.
4. The polymer slurry feed line.
120

-------
LOWM TE
FI( JRE 40
PLAN OF INJECTION AND NEFERING VkJLT
SHOWING ThE LOCATION
OF 1 &JOR CcIIiJNENTS
(
T MPER TU E
T ArcTER..
ULTRA t I C
ELECT OtE
E
TR&NSM ITT R
POLYMER rEW L%WE
Cizcurr
121

-------
SECTION 16
ON-LINE OPERATION OF THE INJECTION FACILITY
GENERAL
Phase IV, the demonstration phase of the program, was designed
to show that overflows in the study area could be controlled
by the injection of friction-reducing chemicals into the sewer
line and that the injection could be unattended (fully auto-
mated). In addition, it was intended that the data gathered
during injection periods would extend the state of knowledge
of the effect of polymer addition to larger pipes than had
heretofore been utilized. Once the facility had been
“de-bugged”, it was extremely easy to demonstrate the control
of overflows in both manual and automatic control modes. How-
ever, the generation of data which could be generalized was
more difficult by orders of magnitude.
Earlier experiments in friction reduction were either per-
formed under laboratory conditions or the subtle nuances of
friction-reduction were ignored. Most of the field work per-
formed falls in the second category with one notable exception;
the work of Dr. R.H.J. Sellin in Wales. Dr. Sellin was fortun-
ate enough to have a pressure wastewater line which was fed from
a positive displacement pump, thereby making experiments at
constant flow and constant concentration possible.
Friction-reduction data from earlier field tests had been re-
ported in “percent flow increase” or “head reduction” with no
correlation between the two parameters. The difference bet-
ween these two parameters can be better understood by consider-
ing two experiments:
1. Constant Flow - Variable Head
2. Constant Head - Variable Flow
If one has a means whereby the flow through a pipeline can be
kept constant, such as Dr. Sellin’s pipe fed by a constant-dis-
placement pump, then friction reduction affects only the pr ssure
122

-------
at the pump discharge. On the other hand, if a pipeline is fed
from an overflowing head box, then friction reduction affects
only the flow through the line. Unfortunately, most practical
problems in dealing with sewage flow do not correspond to
either of these examples, especially when one considers flow in
gravity sewers.
Therefore, four types of experiments in friction reduction were
designed for the demonstration phase of the program:
1 . Manual control with fixed polymer feed rates to deter-
mine maximum head reduction as a function of polymer
feed rate;
2. Automatic control with fixed polymer feed rates to
determine the ability of holding a head in a band of
values around a pre-selected set point;
3. Automatic control with the polymer feed rate proportion-
al to the flow so that a constant polymer concentration
can be maintained;
4. Automatic control with the polymer feed rate pro-
portional to the critical sewer level to provide
smoothing of the flow.
Experiments of all four types have been performed. The experi-
ments are discussed in the following pages.
MANUAL CONTROL-FIXED FEED RATES
The first opportunity to check out the injection equipment
occurred on October 16, 1973 at a time before the instrument-
ation had been de-bugged, so the data obtained was limited to
level at the injection site. Figure 41 is a representation of
that data. The test was started while the level in the sewer
was rising. The injection was started at a rate of 1.46 pounds
per minute and maintained until the level dropped to the top
of the pipe. The injection was then stopped and the level
returned to its previous high. The injection was re-started at
a reduced rate of 1.1 pounds per minute. After the level
dropped about 0.7 feet, the feed rate was reduced to 0.85
pounds per minute. The object of this reduction was to
determine the minimum feed rate which would control the over-
flow. Note the reduced rate at which the level continued to
drop.
One difficulty with a test at constant injection rate is caused
by the reduced head; the effective concentration generally
increases as the level drops. Although there is a concurrent
phenomenon of increased flow initially, the flow system attempts
to establish equilibrium at a lower head; this leads to reduced
123

-------
5
0
1
I , ,
-a
>G)
—0
Or-
-Ic
— mm

- -I
— —
o
I-
-<-1
Orrt
)
-------
Local Level Flow
Temperature 63°F
Injection Rate 0.8 lbs/mm
Polyox WSR-301
N
\
\
FIGURE 42.
GRAPH OF FLOW AND LEVEL DURING DEMONSTRATION
OF MANUAL INJECTION OF MARCH 27, 1974
I ’
/
/
N
/
/
r
><
E
-J
Li..
f’ ) —
o1
a)
4-
w
w
-J
5
4
3
2
1
0
/
/
/
I
CD
4-
4-
CD
CD
4-
4-
CD
0 1 2 3

-------
11
10
9
8
7
f0
a)
=
I-
r’,) 0
4.)
U-
3
2
1
0
1 300
Figure 43.
Graph of Manual Injection of May 5, 1974
Feed Rate Constant Because of Low Polymer Supply
Local Level
Downstream Level
Intermittent
Polymer Feed
1400 1500 1600
Time of Da,y
1700

-------
flow which results in a higher concentration which results in
greater friction reduction, et cetera. The foregoing is a
simplified explanation of a very complex continuous function,
but serves to illustrate the enormous difficulty one faces In
analyzing large scale experimental data.
The second test was run as a demonstration on March 27, 1974.
Since a suitable rainfall could not be scheduled for this
demonstration, an inflow source was simulated by installing
a “valve” in the place of a manhole cover approximately two
miles upstream from the injection site and diverting the
stream flow through this valve. The valve utilizes an 18
inch diameter “pillow packer” as the valve element.
Figure 42 is a graph of the demonstration test. The injection
was manually controlled, with the injection rate set at 0.8
pounds per minute.
The third injection test performed under manual control began
on May 5, 1974 as an automatic test. However, when the tech-
nician arrived at the station, he found that the polymer being
dispensed was lumpy, causing the actual feed rate to vary. The
lumps resulted from a failure of the dehumidifier which caused
“crusting” of the surface layer. This layer was fed wnen the
polymer supply was low. It was necessary to abort the inject-
ion after three hours because of difficulties in clearing
the lumps. See figure 43.
AUTOMATIC CONTROL-FIXED POLYMER FEED RATE
With the control level set to start the injection at 45 inches,
the graph of Figure 44 as produced on April 21, 1974. Under
automatic control, the injection starts and stops at the preset
level. A series of oscillations of level results from this
type of control. Three on-off cycles appear on Figure 44. In-
spection of the graph indicates the time lag caused by the
build-up and purging of polymer over the 3,540 feet of sewer
line immediately downstream.
Figure 45 is a graph of an injection at the same rate (0.8
pounds per minute) with the control level set at 30 inches.
The level and flow graphs have similar characteristics as
before, only at a lower level. There are four short injection
cycles presented on the figure, with the last two of such short
duration that the flow rate was not stgnificantly affected.
AUTOMATIC CONTROL - POLYMER FEED RATE PROPORTIONAL TO FLOW
Because of repeated failures of the flowmeter only one success-
ful test of constant polymer concentration has been performed.
127

-------
Flow
Local Level
5 Downstream Level
Temperature 70°F
Injection Rate 0.8 lbs/mm
Polyox WSR-3O1
0
0
0
x
E
0
N.)
a,
a,
I I-
2
-J
LU
LU
-J
1
0 1 I I I
0 1 2 3
Time (hours)
GRAPH OF FLOW AND LEVELS DURING AUTOMATIC
INJECTION OF APRIL 21, 1974
f\ /
I
FIGURE 44.

-------
0
0
0
I-
x
E
:
0
-J

I . 4 )
N.)
-J
L&J
-J
5
4
3
2
1
0
0
I
A
I I
1 2
FIGURE 45.
F LOW
Local Level
Temperature 69°F
Injection Rate 0.8 lbs/mm
Polyox WSR-301
3
\
\/\I
—
— — — — . .—
Time (hours)
GRAPH OF FLOW AND LEVEL DURING AUTOMATIC
INJECTION OF APRIL 22, 1974

-------
This test is graphed in Figure 46, and occurred on May 1,
1974. The start-up level was set to 27 inches and the poly-
mer feeder controls set to yield 30 parts per million
concentration. Absent from Figure 46 are the rapidly changing
levels which have characterized tests under other types of
control. Rather, the flow graph showed variation as the in-
jection mechanism executed two start-stop cycles.
AUTOMATIC CONTROL - POLYMER FEED RATE PROPORTIONAL
TO LEVEL
Figure 47 illustrates the effect of allowing the magnitude
of the “Local Level” govern the rate of polymer feed. This
graph represents a test conducted on June 9, 1974. As the
level in the sewer drops, the feed rate is decreased, causing a
“rounding-out” of the level graph as it approaches some lower
value asymtotically. Theoretically, there will be a gradual
dampening of the curve variation when the feed rate is variable
and level controlled. The level will even out at a particular
level and maintain that level through control of the feed rate.
If the level control uses a fixed feed rate, there will be a
constant sawtooth pattern as the high head engages the feed
mechanism; the head is reduced as a result of the polymer feed
and a low head level disengages the feed mechanism. If the
fixed feed rate is not enough to bring the head down to the
cutoff point, the head will follow a pattern similar to that
which would exist if no polymer were being added. The only
difference would be that the head would be lower and the total
flow would be higher.
130

-------
5
4
CD
0
C D
I-
x
0
0
Ci .)
4 )
a)
a)
4-
I-
a)
>
ii
0
FIGURE 46.
Time of Day
GRAPH OF AUTOMATIC INJECTION OF MAY 1 , 1974
POLYMER FEED RATE PROPORTIONAL TO FLOW
Flow
Local Level
I
Downstream Level
0800 0830 0900 0930

-------
12
Figure 47
Local Level
Downstream Level
Fixed Feed
Variable Feed Rate 0.56 lbs/mm
4-
4-
1 300
Time of Day
Graph of Downstream and Local Levels Durinci Automatic
Injection of June 9, 1974. Polymer Feed Proportional to Level
11
10
9
8
7
6
5
4
3
2
1
0
1200
— -
c )
r.. a)
0
a)
a)
LL
I -
4-
4-

I-
4-
4-
ra
I-
ca 4-
9-

r
-w

-------
APPENDIX A
FRICTION REDUCING MATERIALS
TESTED FOR CONFORMANCE TO
PERFORMANCE SPECIFICATION
Product
Percol 139
Percol 155
Percol 351
Limiting Shear Stress*
(psi
.011
.029
.018
Manufacturer
Allied Colloids
RC-301
RC-322
Polymer 1100
FR-X
WCL 727
WCL 755
WI 3000
Separan AP3O
Separan AP273
NGL 3958
Polyox WSR 301
Polyox WSR 701
Polyox ERA
.011
.011
.018
.011
.011
.011
• 011
.015
.022
.011
.011
.018
.022
*Shear stress at which apparent degradation (or rupture) of the
polymer chains occur.
American Cyanamid
Betz Chemical Co.
Calgon Chemical Co.
Dow Chemical Co.
Stein, Hall and Co.
Union Carbide
133

-------
A7PE iDIX B
CITY OF DALLAS
SPECIFICATION FOR
HIGH MOLECULAR-WEIGHT WATER SOLUBLE
FRICTION- REDUCING ADDITIVES
No. PA-106-4061-70
I. PURPOSE
The purpose of this specification is to describe the performance
characteristics of organic polymer materials for use as a viscoelastic fluid
energy loss reducer in aqueous media.
II. GENERAL
A. These specifications are meant to include both natural and syn-
thetic high-molecular weight materials such as polyacrylamides, polyethyl-
ene oxides and guar gum formulations.
B. The material may be supplied as a dry material, solution or stable
suspension.
C. The material as supplied must not be highly hazardous in nature;
i.e., it must not be toxic, highly corrosive, explosive or highly volatile.
D. The container in which the material is supplied should be suf-
ficient to maintain the material in a usable state for a period of six months
under reasonable storage conditions. Sacks or bags are not generally
acceptable.
E. Each container should be clearly marked as to contents, pre-
cautions and storage instructions.
III. CHEMICAL AND PHYSICAL CHARACTERISTICS
The specifications shown in the following table are indicative of the
materials commonly accepted under these specifications. Materials which
134

-------
depart from these guidelines will be evaluated on the basis of performance
as a friction reducer.
PHYSICO-CHEMICAL PROPERTIES
Solubility in water Readily soluble
Usable in pHrange 4- 11,5
Storage stability temperatures
(for slurries or solutions) 35 F - 110 F
Molecular weight 500, 000
*
Particle size 95% passing 30 mesh
or smaller
100% retai ied on
50 mesh
*May be waived
IV. PERFORMANCE EVALUATION
A. Equipment
The equipment used in the evaluation test shall be a pressurized
straight-tube flow apparatus illustrated by the attached drawing. It consists
of a pressure vessel (A) fitted with a removable cover for filling and cleaning;
a thermometer (B) mounted through the wall of (A) such that the temperature
of the contents is determined; a tube (C) of stainless steel seamless tubing
having an inside diameter of 0. 18 inches and proportions as shown; pressure
taps (‘D) and (E) assembled in such a manner to produce a minimum stream
disturbance; laboratory-type differential pressure gauges (H) of ranges
selected to provide a resolution of not more than 0.5-percent of the measured
quantity; a throttling valve (F) to control the flow velocity; and apparatus
(G) manual or automatic to determine flow rate gravimetrically or volumetrically.
135

-------
B. Preparation of Test Material
A sample of the test material shall be dissolved in de-lonized
water in the manner prescribed by the manufacturer in the proportions re-
quired to give a concentration of active friction reducer of up to 50 parts-
per-million (ppm) by weight of water. This solution will be gently agitated
for a sufficient length of time to insure a clear solution with no lumps or
“fish-eyes. “ Materials shall be tested immediately after solution agitation.
Made-up solutions will not be tested more than once or retained longer
than 30 hours.
C. Method of Test
Two gallons of the solution prepared in (B) above shall be
placed in the pressure vessel with valve (F) in closed position. With the
pressure vessel open to atmospheric pressure, valve (F) will then be opened
slightly until the tube (C) and fittings have been purged of air bubbles.
Valve (F) will then be closed.
Static pressure is then built-up in the pressure vessel by means
of an auxiliary pressure regulator and air or nitrogen source. The applied
pressure should be adjusted between 10 - 160 psi.
When the above preparations have been completed, a “run” is
made by opening valve (F) a preselected amount, measuring the steady-
state flow and frictional pressure loss (as indicated by the differential
pressure gauges).
Tests should be performed at a minimum of three flow rates,
with fresh solution used for each run.
136

-------
D. Reporting of Results
Results should include the following information:
1. Material name or designation,
2. Material chemical genera (i. e., polyacrylamide).
3. Manufacturer.
4. Batch or lot number (if available).
5. Approximate molecular weight.
6. Remarks on solution appearance (i.e., clarity, etc.).
7. Temperature of material at the time of test.
8. A graph of pressure drop reduced to pressure loss
per 100 feet of length and velocity in feet per second.
To qualify under the performance requirements of this speci-
fication, the graph results from item eight, above, should fall below and
to the right of the dashed line in the figure below.
137

-------
2000
1000
500
400
300
200
100
. . t.
i
20 30 40 50
Fluid Velocity (Feet/Second)
138

.-. . . I 4
: :ii :iiii I:::::::::: .:::
A
r
4 3
4-4
0
0
U)
a
a
a)
S...
a)
U)
a)
S...
0..
‘1T ’1 ’ I!JIUIIJJI I TT TI
rm InmnTm “r’mTrm
100

-------
APPENDIX C
SEWER MODELING PROGRAM
100 DIMFNSIONJ 0(3,23,3), U(3 23, 3), PK( 3,23,3), FL( 323k 3), TEL( 323 3)
110 FF L Ll.)F, 1NFIt_,L0 C3,23),INF(3,23,2),DELo(3,p3,3)
120 INTEGEF ENT,SETP,ENTM, X(3),TT(3)
140 00 5,1=1,3
150 DO 5 ,J=1,23
160 00 S ,K=1 ,
171) TFL(T ,j ,K)= 1F4; DFLC(1,J , <)=fl;(1,j,’<)=0;FL(t,j,i.<,=Q;0(I,J,K)=r
1 R0 5:PK(I ,J ,K)=1
190 :FILE DEFINF ,LINES
200 F D(1)NLINE
21000 10 ,1=1 ,NLINES
220 F D( 1)LIN , ENTMAX(LIN), TT(LIN)
230 1Q:CONTINLJF
240 15:FE D( 1)LIN ,ENT, FF ,Ak,L1IF ,TEL(L1N, ENT, SEP) TPL C D ,PL, FPC
250 E D/12.
260 C(L1 N, ENT, EF ) =c k*LLIF/72U
270 ( 0(LIN,ENT,SEp)=c:(LIJ,wNT,SER)
275 IE(D)999,17,IA
2 0 16:PK(LIN ,ENT , Ek)=( )/((C?1.R5)*(L’? ))
2 1 00T0 1
2F 9 17: HK(LIN, FNT E )= I
290 1 :1NF(L1’4,ENT, ER)=TPL*F C
10IF(FNDFILF 1)15
3 fl FI D(2)0LTE( LL
330 20:FFAO(2)LIN,FNT,C, D,F’LsTEL(LIN ,ENT ,3) ,L05.S(LIN ,ENT)
340 D=D/12.
350PK(LIN ,ENI ,3) =( PL)/((C?1. S)*(L ?
360 IF(ENDFILE 2)20
370 P1
3 0 30:I=NLINES
390 40:J=ENTM X(I)
400 0(I ,J , 3) 0(1 ,J , 1)+AC( 1 ,J2)
Z 10 C’OT 60
4S20 50:0(I,J ,3)=Pfl(I ,J , 1)+’ 1)(1,J,2)÷0(I,,J÷1,3)
‘ 30 60:J=J-1
440 IF(J)70 ,70 ,50
450 70:1= [ -1
460 1 F( I) 100’, 100,75
70 75:I (TT(I)) 100 ’ 4fl,9fl
4 80 90:J=FNTMi X(1)
490 0(I,j,3)=A0CI,J, 1)+AC(I ,J ,2)+0(1+ 1 ’ I ,3)+O(I+2 , 1,3)
139

-------
500 J=J-1
510 IF(J) 100, 100s50
520 100 C NT1NMF
53( DO 170,1=1,NLINES
5’ 0 J=1
550 1F(t-1)9 9, 120’ 130
560 120:EL( 1. 1,3) 0VTF LL
570 EOTØ 160
580 130: =I;PP=9P/2
590 IF(PP-INTF(RP))999,140 ,150
600 l40:PP=EiNTMI X( I- I)
610 EL(1 1,3)=EL(I—I,RR ,3)
620 GOTO 160
630 150:PF=ENTMl kX(1-2)
6L 0 FL(1,1,3)=FL(I-2,PF,3)
660 160: FD=ENT 1 X( I)
670 DO 170,J2 ,00
680 CO= I ONF( DCL, j, 3))’ 1 • 8 5 ,0(I J, 3)
700 161:EL(I ,J 3)=EL( 1 ,J- 1. 3)+CD*F <( 1,J, 3)+ j )5 ( 1 ,J)
730 EL(1,J,3) MAX1(EL( 1,J,3),1EL(I,J ,3))
7L,fl 166:IF(P (I,j , 1)— 1)167, 168, 167
750 167:EL(1 ,J , 1)=EL(1, I,3)+s1c NF(c .ps(Ac(I,J , 1))’ 1 . 5) ’t C(t,J, 1))
760 +*p ((1,j,1)
765 EL(1J 1)=(’MI X1(EL(I,J , D,TEL(I,J,3) 1)
770 168:IF(PK(I,J,2)—1) 16’ ’ 170, 169
780 1.%5),4C(I , J ,2))
790 +*PK(I,J,2)
795 ELcI,J,2)=AMc xI(EL(1,J,2),TEL(I,J ,3)+ 1)
800 170:CONTINIE
8 10 I F( P1 10— I TF( F/ 10)) 999, 171’ 179
820 171:PF 1NT 600
830 P I T 610
840 DO 172,11,NLINES
850 0=ENTMPX(I)
860 DO 172,j=I,FF’
70 PkI ‘ T 621, 1, J, EL( I, J’ 3), CCI, J, 3)
880 00 172 , =1 ,2
TF(pK(I,J, <)_1)STRETCH,172,5T}(FTCk
900 ST ET( :P .INT
91C 172:CONTINLE
920 PFINT,”THF INFILTRATION SATE I PFF’ FNTLY’.,INFIL,”( LL0N /MIN/Fr
930 PRINT,”DO YOV t ISH TO C0NTINL F THE PFOPLFM (YES OF NO)?’
940 INPLJI,IA
950 lF(I -”N0”)I73,999 ,173
960 173:PIINT,”TYPF A VALVE FOP THE INFILTRATION RATE INCFE 1FNT
970 PRINT, “(DFCPEMENT)AFTEF- THE OVESTION F <.”
980 1 3P1’T,P
985 IF(P) 174, 179, 174
990 17A:INFIL=INFIL+E
1000 DO 177,1 l,NLtNF5
1010 RF=ENTMAX(I)
1020 DO 177 ,J1 ,BP
140

-------
1030 DO 177 K=1 ,2
10 U 0(T,J,K)=0(I,J,K)+INF(I,J,K)*P
1050 177: P 0( I, J, 1<) = DC I, j, K)
1060 00Th 210
1070 179:D0 200,I=1,NLINF
1080 BBENTM X(I)
1100 Dø 2O0 K=1,2
1110 IF(PK(I,j,K)-1)STAP,P00,STAP
1120 ST( F :CK=TEL(I,J,K)-ELCI,J,K)
1130 IF(CK)191 ,200 , 160
11 0 160:AC(I,J,K)=(I,J,K)+0.5* PSCAr(I,J,K))
1 1 5 IF(c C(T,J,K)-P(r,J,K))20o,2oe, 165.
1150 15:AQ(I,J,K)=0(I,J,K)
1160 GOTO 200
I 1 0 I91:AC(I,j,K)=t P(I,J,K)—0.5*ARS( U(I,j,K))
1165 A0(I J K)=c N1 X1CAC(I,J,K),0(I,J+1,3))
1190 IF( D(I,J,K))2o0, 192,200
1195 192:IF(D(I,j,K))200, 195,200
1210 195: PD( I, J, K) =- 10.
1230 200: DELU( I, J, K) =Q( I, J, K) — 0( I, J’ K)
12’ 0 210:P=P-’-l
1250 GOTO 30
1260
1270 +7X,6 OVERFLO ,,,7X,9H E FL0’ )
1290 610:F0 MAT(26X,3 4FT.,8X,3WC-F i,10X,3-{GPM,10X,3 -1GPM)
1290 620:F ii T(I6,I6,11X,F7.2,3X,F9.2,3d***)
1300 630:FøF MPT(12X,I6,5X,F7.2,3X,F9.2,3X,R9.2,3X,F9.2)
1310 999 STØP
141

-------
APPENDIX D
BACKMAN CREEK INPUT DATA
INPUT DATA FOR BACHMAN CREEK COMPUTER MODEL
Line
Entry
1-N;W
2-S;E
Serial
Area
in
Acres
Land
Use
Factor
Overflow
Elevation
(Feet)
Total Pipe
Length
(Feet)
c
Pipe
Dia.
(Inches)
Feed Pipe
Length
(Feet)
Factor
1 1
1 2 1 427.5 100
1 3 2 --- 448.5 100
1 4 --- 447.0 100
1 5 2 --- 446.5 --- 100
1 6 2 17.0 1600 455.0 2350.0 100 8 320.0 1.
1 7 2 31.0 1600 444.0 4070.0 100 8 150.0 1.
1 8 1 5.0 550 444.0 6520.0 100 15 267.0 1.
1 9 1 7.0 1600 444.4 7830.0 100 18 175.0 1.
1 10 2 41.0 950 490.0 7400.0 100 8 1800.0 1.
1 11 11.3 1600 446.0 1150.0 100 8 450.0 1.
1 11 2 122.0 950 500.0 5800.0 100 10 1700.0 1.
1 12 2 51.6 950 500.0 5400.0 100 8 1725.0 1.
1 13 2 89.0 950 473.93 7800.0 100 8 1200.0 1.
1 14 1 134.0 950 450.0 1392.5 100 10 2875.0 1.
1 14 2 59.0 550 474.0 5124.0 100 8 3370.0 1.
1 15 1 1.0 250 471.25 200.0 100 6 200.0 1.
1 16 2 2.0 500 471.5 1300.0 100 8 300.0 1.
1 17 2 1.0 250 482.0 2000.0 100 6 200.0 1.
1 18 1 79.0 550 493.0 7100.0 100 8 1050.0 1.
1 18 2 189.0 550 475.0 18325.0 100 15 450.0 1.
1 19 1 13.0 550 495.0 1100.0 100 8 1900.0 1.
1 20 2 579.0 550 485.0 60913.0 100 18 250.0 1.
1 21 1 47.0 550 497.0 4905.0 100 8 330.0 1.
1 21 2 307.0 550 487.4 28940.0 100 12 500.0 1.
1 22 1 60.0 550 487.5 5675.0 100 8 675.0 1.

-------
(Conti nued)
Line
Entry
1-N;W
2-S;E
Serial
Area
in
Acres
Land
Use
Factor
Overflow
Elevation
(Feet)
Total Pipe
Length
(Feet)
C
Pipe
Ijia ,
(Inches)
Feed Pipe
Length
(Feet)
Factor
2 2 2 41.5 550 515.0 4060.0 100 8 2600.0 1.
2 3 2 75.0 550 536.0 6175.0 100 10 525.0 1.
2 4 1 513.6 100 12 3.0
2 5 1 254.0 550 534.5 24275.0 100 15 1700.0 1.
2 5 2 46.0 550 545.0 3725.0 100 8 425.0 1.
2 6 1 47.0 550 553.0 3200.0 100 8 550.0 1.
2 7 2 15.0 550 585.0 1500.0 100 8 1200.0 1.
2 9 2 28.0 550 585.0 2800.0 100 8 1000.0 1.
2 10 2 32.0 550 587.0 3200.0 100 8 1000.0 1.
2 11 2 32.0 550 588.0 3200.0 100 8 1000.0 1.
2 12 2 29.0 550 590.0 3000.0 100 8 1000.0 1.
2 13 2 30.0 550 593.0 3000.0 100 8 1000.0 1.
2 14 2 30.0 550 596.0 3000.0 100 8 1000.0 1.
2 15 2 30.0 550 598.0 3000.0 100 8 1000.0 1.
2 16 2 39.0 550 599.0 3900.0 100 8 1000.0 1.
2 17 1 --- 588.0 100 12 3.0 1.
2 17 2 38.0 550 598.0 3800.0 100 10 1000.0 1.
2 18 2 40.0 550 599.0 4000.0 100 10 1000.0 1.
2 19 1 589.0 --- 100 12 50.0 0
2 19 2 41.0 550 597.0 4100.0 100 10 1000.0 1.
2 20 1 114.0 950 598.0 15200.0 100 10 1700.0 1.
2 20 2 85.0 950 602.0 8500.0 100 8 700.0 1.
2 21 1 1779.0 300 700.0 177900.0 100 10 100.0 1.
3 2 2 477.0 --- 60 4 30.1) 1.
3 3 1 140.0 550 500.0 7875.0 100 12 2200.0 1.
3 4 1 489.15 60 6 20.0
3 5 1 487.0 60 4 15.0

-------
(Continued)
Line
Entry
1-N W
Z-S E
Serial
Area
In
Acres
Land
Use
Factor
Overflow
Elevation
(Feet)
Total Pipe
Length
(Feet)
C
Pipe
Dia.
(Inches)
Feed Pipe
Length
(Feet)
Factor
3 6 1 488.0 60 4 15.0
3 7 1 --- 490.0 60 4 15.0
3 8 1 --- 493.1 60 4 20.0
3 9 1 212.0 550 515.0 21900.0 100 10 400.0 1.
3 10 2 14.5 550 520.0 1050.0 90 6 300.0 1.
3 11 2 142.0 550 520.0 11950.0 100 10 20.0 1.
3 12 1 128.0 550 532.0 11375.0 100 10 1125.0 1.
3 13 2 131.0 550 547.0 9125.0 100 10 800.0 1.
3 14 2 112.0 550 530.0 7425.0 100 10 50.0 1.
3 15 1 1156.0 550 600.0 115600.0 100 18 1000.0 1.

-------
(Continued)
Line No.
Entry No.
Station No.
I
Pipe Diameter Pipe Length
(Inches) (Feet)
Invert. Elevation
(Feet Above MSL)
Minor Head
Loss
(Feet)
U,
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
5+72.28
3
15+
17.75
4*
16+75.00
5
23+79.00
6
33
+ 25.00
7
47
+ 30.00
8
82
+ 45.00
9
87
+ 00.00
10
106
+ 50.00
11
116
+ 67.32
12
121
+ 23.00
13
128
+ 10.83
14*
150+
32.09
15
160
+ 70.00
160
164+
33.00
17
166
+ 21. 58
18
170
+ 66.0
19
181
+ 72. 14
20
186
+ 68.76
21 +
198
+ 78.00
22
206
+ 25.04
23 *
211
+ 90.00
30
30
30
36
36
36
36
36
36
36
36
36
36
24
24
24
24
24
24
24
24
24
572.28
945.47
157.25
704.00
946.00
405.00
3515.00
455.00
1950.00
1170.32
456.68
687. 83
2056.26
1037.91
363.00
188. 58
344.42
1106. 14
496.62
1209. 24
747. 04
564. 96
418.00
420.26
424. 04
425.00
426.60
427. 80
429.89
434.90
435. 50
438. 18
439.60
440.24
441.20
446.89
452.08
453.80
454.84
457. 06
466. 10
466. 14
470.37
472.99
475.24
.1
325
.2
.15
.1
.1
.3
.0
.1
.1
.1
.1
.93
.2
.1
.1
.2
27
.2
• 47
17
.32
+ Overflow
-w Change in Pipe Size
€Instrument Installation

-------
(Continued)
Line No.
Entry No.
Station No.
Pipe Diamete
(Inches)
Pipe Length
(Feet)
Invert. Elevation
(Feet P bove MSL)
Minor Head
Loss
(Feet)
0 i
2
Browning
Branch
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
3
Bachman
Branch
3
2*
3
44-
51-
6
7
8
9*
10
11
12
13
14
15
16
17 4- *
18
19 4-
20
21
3
4+
221 + 26.26
233+ 54.97
245 + 77. 38
259+ 36.67
262+ 19.4
293+ 34.65
297 + 24. 62
309 + 73.77
313 + 41.78
315 + 64. 55
317 + 06. 20
320 + 69.84
324 + 34. 35
328 + 08. 10
332 + 57.37
340 + 50.00
346 + 64. 10
349 + 00.00
355 + 40. 10
ORB
4+00.00
7 + 48.00
9+20.00
18
15
15
15
15
15
15
15
12
12
12
12
12
12
12
12
10
10
10
10
18
18
18
936.26
1228.71
1222.41
1359.29
282.73
3115.25
389.97
1249. 15
368.01
222. 77
141.65
363.64
364. 51
373.85
449. 27
792.63
614. 10
235.90
640. 10
1000.00
400. 00
348.00
172.00
481.78
496.58
506. 36
516.84
538.00
570.94
57 1. 52
573. 53
575. 59
576.77
577. 52
578. 14
578.76
579.39
580. 16
581.67
585.82
587. 40
591.58
601. 00
476.95
478. 52
479. 15
• 58
.3
.1
34
.2
.8
.2
.4
.1
.1
.1
.1
.1
.1
.1
.1
.1
.3
.1
.1
.01
• 15
.1
+ Overflow
* Change in Pipe Size
Ej 3 Instrument Installation

-------
(Continued)
Minor Head
Line No.
Entry No.
Station No.
Pipe Diameter
(Inches)
Pipe Length
(Feet)
Invert. Elevation
(Feet Above MSL
Loss
(Feet)
3
3
3
3
3
3
3
- NM
NM
3
3
6*
7 - I-
8+0
90
100
11
12
13
14
15
14 + 77.50
16 + 74.00
19 + 86.00
29 + 35.00
35 + 40.00
35+ 80.00
62 + 41. 50
75 + 00.00
79 + 66.30
97 + 50.00
ORB
18
18
18
18
18
18
18
18
18
18
18
557.50
196.50
3 12.00
920.00
605.00
40. 00
2661.50
1258. 50
966.30
1783.70
1000. 00
482.90
483.80
485.08
489. 15
492. 17
492. 20
505.60
513.86
517. 12
527. 7 5
550.00
17
.6
.01
.11
• 36
.1
.12
10
10
17
.1
+ Overflow
* Change in Pipe Size
Instrument Installation

-------
APPENDIX E
COMPUTER OUTPUT FROM MODELING PROGRAM
RUN #1
LINE
NØDE
ELEV.
FLOW
OVERFLOW
AF
-------
ELEV.
LINE NODE ‘F EA T.
2
1 21 70.37
471.37
2 471.37
1 22 472.99
473.99
1 23 475.24
2 a 475.24
2 2 481.78
2 482.78
2 3 496.54
2 .497.58
2 .4 506.36
507.36
2 5 516.84
1 517.84
2 517.84
2 6 538.00
539.00
2 7 570.94
2 571.94
2 8 571.65
2 9 573.66
2 574.53
2 10 575•59
• 576.59
2 11 576.77
2 577.77
2 12 577.52
2 578.52
2 13 578.76
2 579.14
2 14 579.96
2 579.97
2 15 581.15
2 581.16
2 16 582.51
2 582.53
2 17 584.72
1 584. 72
2 584.73
2 18 588.39
2 588.39
2 19 589.98
589.98
2 589.99
2 20 593.41
1 593.67
2 593.58
2 21 601.00
FLO
6PM
318.45
2536. 10***
25.85
1 68 • 4 S
2341 •
33.00
2308 •
I lF 6.%7***
1188.8 7***
22.82
1 166.05***
41 • 25
1 124.80***
• 00
I 124.80***
1 39 • 70
25. 30•
9 5 • 8 0***
25.85
933.9 5***
8 • 25
925. 70***
925. 70***
15.40
910. 30***
17.60
892. 70***
17.60
75. 10***
15.95
859. 15***
16.50
842 • 6 5* *
1 6 • 50
826. 15***
16.50
809 • 65***
21.45
788 • 20***
.00
20.90
767. 30***
22.00
745. 30***
• 00
22.55
722. 75***
108 • 30
80. 75
533. 70***
OVFPFLO
6PM
• 00
• 00
• oc
• 00
.00
• 00
• 00
• 00
.00
• 00
• 00
• 00
• 00
• 00
• 00
.00
.00
• 00
• 00
• 00
• 00
.00
• 00
CO
.00
.00
PF6 FLOk
(PM
313 • 45
25.85
1 66 . 3 5
33.00
22.82
41.25
• 00
1 39 • 70
25 • 3D
25.85
8 • 25
15.40
17.60
17.60
15.95
16 • 50
16.50
16.50
21 • .45
• 00
20.90
22.00
.00
22.55
1 08 • 30
80.75
149

-------
LINE NØDE AREA ELEV.
FT.
602.00
3 1 475.24
3 2 476.95
2 477.95
3 3 478.52
‘ 79.52
4 479.15
I Li 0.15
3 5
483.90
3 6 483.80
484.80
3 7 4 85.08
446.08
3 8 449.15
490.15
3 9 492.17
493.17
3 10 492.29
2 493.20
3 11 505.60
2 506.60
3 12 S13. 6
I 514.46
3 13 517.12
2 514.12
3 14 527.75
2 528.75
3 15 550.00
1 551.00
FLØ P
6PM
533. 70
1119 • 52***
1119 •
. 00
11i - r r.-
I 7•
7/.00
1042. 52***
• 00
1042. 52***
• 00
1042. 52***
• 00
1042. 52 **
• cc
1042. 52* *
• 00
1042. 52* *
116.60
925.92***
.7.97
91 7.95***
7• • 1 0
839 • 4
70.40
769 •
72.05
697. 40***
61 • 60
635 • 80* * *
635.80
OVEPFLO t,
G PM
• no
frFA FLO1
533.70
.00
.00
.00
77.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
.00
116.60
.00
7.97
.00
74.10
.00
70.40
.00
72.05
.00
61.60
.00
635.80
THE INFILTPATIØN RATE IS PRESENTLY .00 GALLØNS/MIN/F’T øE LATERAL
150

-------
RUN # 2
1
1
1
2
1
3
1
Lj
1
5
1
6
1
7
LINE
NØDE
AkEA
ELEV.
FLØW
ØVEPFLØ’
AEFA FLØ
FT.
GPM
OPM
GEM
419.00
7281.85* *
420.57
72%1.85**
424.04
72 81.85***
425.00
72%1.%5***
26.60
72Ffl.%5***
427.80
72%1.35*$-
2
428.80
46.00
.00
46.00
429.89
7235.85***
2
430.89
32.16
.00
82.16
I
3
1
434.90
435.90
7153.69***
54.91
.00
54.91
1
9
1
435.50
436.50
709%.78 **
73.84
.00
73.84
1
10
P
438.1%
439.18
7024.94***
98.15
.00
98.15
1
11
1
2
439.60
440.60
440.60
6926.79***
27.2%
162.30
.00
.00
27.2R
162.30
1
12
2
440.24
441.24
6737.21***
92.22
.00
92.22
1
13
2
441.20
442.20
6644.99***
146.95
.00
16.95
1
14
1
2
446.89
448.81
447.99
649%.04i***
234.70
73.4’i
.00
.00
238.70
73.44
1
15
1
452.51
453.08
6195.90***
1.35
.00
1.85
1
16
2
454.51
454.80
61F14.0S***
11.40
.00
11.40
1
17
2
455.59
455.84
6172.65***
16.25
.00
16.25
1
2
457.57
1i54.06
453.06
6156.40***
100.25
250.55
.00
.00
100.25
250.55
1
19
1
466.10
467.10
5%05.6 0***
15.95
.00
15.95
I
2C
2
463.60
468.71
5789.65***
805.75
.00
805.75
1
21
1
2
473.30
473.36
/i73.6%
49%3.39***
65.09
400.37
.00
.00
65.09
400.37
1
22
1
475.65
475.81
4518.43***
78.40
.00
78.40
1
23
477.57
4440.03***
2
1
477.57
1%30.11***
2
2
2
481.78
482.78
1%30.11***
55.30
.00
55.30
2
3
2
496.58
497.58
1774.80***
90.65
.00
90.65
151

-------
LINE
NØDF
AFF
ELFk’.
FL Y J
ØVFHLØW
F F6 FL( W
FT.
(:P 1
( F’
2
4
506.36
1684.15***
1
507.36
.00
.00
.00
2
5
I
2
516.S.’
517.8.a
517.84
16 8 4s15***
333.90
55.10
.00
.00
333.90
55.1
2
6
I
538.00
53q.OO
I95.15***
51. 5
.00
51.. 5
2
7
2
570.9.4
571.9 .4
1243.70***
20.25
.00
20.25.
2
8
S71. 9
1223. 1i5* ’
2
575.13
575.19
1223.45***
37. 8fl
.00
37. 20
2
10
2
577.32
577.40
1185.6S***
43.20
.00
. 42.20
2
11
2
578.60
578.68
1142.45***
43.20
.00
143.20
2
12
2
579.40
579.47
109 .25***
39.95
.00
39.95
2
13
2
581.18
581.25
1O59.30 **
40.50
.00
40.50
2
14
2
582.84
582.91
1018.80**
40.50
.00
40.50
2
15
2
584.43
584.50
978.3 0***
40.50
.00
40.50
2
16
2
586.19
586.30
937.80***
52.65
.00
52.65
2
17
1
2
588.91
584.91
568.95
8 5.15***
—33.75
51.30
33.75
.00
.00
51.30
2
18
2
593.48
593.49
867.60***
30.37
23.62
54.00
2
19
1
2
595.39
595.39
595.39
%37.23***
-33.75
10.38
33.75
44.97
.00
55.35
2
20
1
2
600.08
600.08
600.10
860.60***
14.37
27.89
215.53
120.86
229.90
1 i8.75
2
21
1
606.71
607.36
81 8.34***
818.34
.00
818.34
3
1
477.57
2609.92***
3
2
2
479.14
479.14
2609.92***
.00
.00
.00
3
3
1
480.65
480.89
2609.92**
140.00
.00
140.00
3
14
1
481.35
481.35
2469
.00
.00
.00
3
5
1
483.48
483.90
2469.92***
.00
.00
.00
3
6
1
48.4.77
484.80
2469.92***
.00
.00
.00
152

-------
ELEV.
FE.
FL OW
I
OVF FFE_@
1- p i
FF 1O’4
(- P
1
LINE NODE
3 7
3 3
3 9
3 10
3 ii
3 12
3 13
3 14
3 15
• on
P E t.
2
2
2
2
. 00
• 00
485.88
486 • 08
489 • 23
490.15
492. 17
493. 17
492.38
493.20
505 • 60
506 • 60
513.86
514. 4
517.12
518. 12
527.75
528 • 75
550.00
551 • 50
• 00
.00
2 I • R 0
2469 • 92***
• no
2469. 92***
.00
2469 .9.***
291.80
21 73 • I 2***
16.37
2161. 75***
173 • 70
1938 •
1 6 1 • 40
1826. 65***
145.05
1631 •
121 • 00
1560. 60***
1560.60
. 00
16.37
.00
173. 70
. 00
1
161 • 40
00
I 5. 05
• 00
THE INFLLTR TIcM4 PATE Is PRESENTLY .008 G, LLøNS/MINfFT ØF’ L TER4L
lOt • 00
.00
1560.60
153

-------
RUN # 3
LINE
NODE
(‘PF
ELFV.
FLOW
OVERFLOW
RE FLOW
FT.
GPM
6PM
6PM
1
1
419.00
92 17.89***
1
2
420.94
8217.89***
1
3
424.30
82l7.89* *
1
4
425.01
8217.89***
1
5
426.60
8217.89***
1
6
428.03
32 17. 39***
2
423.80
50.70
.00
50.70.
1
7
2
429.R9
430.89
8167.19***
90.30
.00
90.30
I
8
1
434.98
435.90
8076.89***
67.95
.00
67.95
1
9
I
435.69
36.50
8008.94***
39.50
.00
89.50
1
10
2
438.36
439.19
79 19.44***
112.95
.00
112.95
1
11
1
2
439.95
440.60
440.60
73 06.49***
29.58
173.90
.00
.00
P9. 8
173.90
1
12
2
440.61
44 1.28
7603.01 **
103.02
.00
103.02
1
13
2
441.53
442.61
7499.99***
162.55
.00
162.55
1
14
1
2
446.89
449.25
47.39
7337.44**
266.55
83.69
.00
.00
266.55
p3.69
1
15
1
453.33
453.88
6987.20***
2.25
.00
2.25
1
16
2
456.36
456.36
6984.95***
14.00
.00
14.00
1
17
2
457.69
457.70
697 0.95***
20.25
.00
20.25
1
1
1
2
460.12
460.61
460.19
6950.70***
114.45
287.20
.00
.00
114.45
287.20
1
19
1
466.81
467.10
6549.05***
18.15
.00
18.15
1
20
2
469.88
470.02
6530.90***
927.53
.00
927.58
1
21
1
2
475.61
475.63
476.10
5603.32***
74.90
458.25
.00
.00
74.90
458.25
1
22
1
478.43
473.68
5070.17***
89.75
.00
89.75
1
23
480.78
4980.42***
2
1
480.78
2 0 04. 15***
2
2
2
433.59
483.63
2004.15***
63.42
.00
63.42
2
3
2
‘496.58
497.5
1940.72***
103.00
.00
103.00
154

-------
LINE NØDF
2 4
2 5
2 6
2 7
2 S
2 9
2 10
2 11
2 12
2 13
2
2 15
2 16
2 17
2 18
2 19
2 20
2 21
3 1
3 2
3 3
3 4
3 5
3 6
I
1
2
1
2
2
2
2
2
2
2
2
2
I
2
2
p
2
1
1
FL F V.
4} EA FT.
506.36
507.36
516.94
517.84
517.94
538 • 00
539.00
570.94
571.94
572. 11
575.62
575.70
578.09
578. 19
579.52
579.62
580.41
580.49
582.38
592.47
584.21
59 . 30
585.95
586.03
587.95
588 • 00
590.79
590.79
590.94
595.65
595.65
597.74
597.74
597.74
602.99
603.00
602.99
610.71
6 11 • 48
480. 78
492.77
42. 77
484.65
484.95
485.53
485.53
488 • 2 1
488 • 2 1
489.69
489 • 69
FL 0
( ‘PM
1837. 72***
00
1937. 72***
382.45
62.55
1392. 72**
57.85
1334.9 7***
23.25
1311. 62***
1311 • 62***
43.40
1268 • 22***
49 • 60
1218. 62**
49.60
11 69 .
45.95
1 123.07***
46.50
1076. 57***
46.50
1030. 07***
46.50
983. 57***
60.45
923. 12***
-33.75
58 . 90
897.9 7***
11 • 62
8S6. 35***
-33.75
3.97
9 16. 1 3***
16.27
10.36
889. 50***
889. 50
2976.2 7 **
2976. 27***
—1.25
2977. 52***
155. 75
2821. 77***
00
2821. 77***
-1.25
2823. 02***
-1.25
OVER FL 0
( PM
• 00
.00
00
00
.00
.00
.00
• 00
00
• 00
.00
• 00
.00
33.75
.00
50.37
33. 75
59 • 58
244.03
155.39
• 00
1.25
.00
.00
I • 25
1 .25
REr L0
(- P 1
00
382.45
62.55
57.85
23.25
• 40
49 • 60
49.60
45.95
46 e 5Q
46.50
6.50
60.45
• 00
58.90
62.00
00
63.55
260.30
165.75
889.50
.00
155.75
.00
• 00
• 00
155

-------
•\‘ •- ‘ LvJ .r
LINE NØPF SF LEV.
FI Z
I
I
2
9
2
2
I
IhFC FL ’1
T.
1_Py
(:pM
F”•
3
7
491.II
9I.1I
282i .27***
-1.25
1.25
.00
3
495.37
495.37
2825.52 **
-1.25
1.25
.00
3
9
49 .46
498.97
2 2E.77***
335.60
.00
335.60
3
10
498.71
498.75
249 1.17***
18 .’ 7
.10
l . 47
3
Ii
508.21
508.22
513.86
514.86
2472.70***
197.60
2275.10***
18’ .15
.
.00
197.60
1RL .15
3
13
517.12
5)8.12
2090.95**
163.30
.00
163.30
3
14
527.75
528.75
1927.65***
135.85
.00
1 5.k’
3
15
550.00
551.94
179 1.80***
1791.80
.00
1791.80
THE INFILTRIATIØN RATE IS PRESENTLY •
GALLO” )S/MIN/FT OF LPTF L
156

-------
LINE
I
I
1
I
I
NODE
2
4
S
6
7
8
9
1 10
1 11
1 12
13
1 15
1 16
1 17
I 18
1 19
1 20
1 21
1 22
1 23
2 1
2 2
2 3
2 4
1 REi ELEV.
419 • 00
420.88
424. 1 4
1125.00
46.60
427 • 99
2 428.80
429 • 89
2 430. 39
4 314. 90
1 435.90
435.58
1 436.50
438. 18
2
439 • 67
440.60
2 440.60
440.29
2 441.32
441 • 20
2 442.81
4.6.89
1 447.89
2 448.35
453.52
1 453.53
455.87
2 455.88
‘457. 13
2 457.16
459 • 414
1 460.26
2 459.55
466.10
1 1467. 10
468 .89
2 468.96
474.55
1 474.67
2 474.78
477.51
1 477.61
479 • 90
479 • 90
483.21
2 4 33.28
496.58
2 497.58
506.36
1 507.36
RUN # 4
FL0
8062. 57***
8062. 57***
8062. 57***
9 062. 57***
8062. 57***
8062. 57 *
Lr j
r_ • ‘4
8000. 12***
110.65
7889. 47***
100.55
7788.92***
1 22 • 6 5
7660. 27***
I 4 • 9 5
7510. 32 4*
17.66
202.90
7289. 76***
130.02
7159 •
201 • 55
6958. 19***
63.03
1 09 • 3 1
6785 • 8
3.25
6782. 59***
20. 50
6762. 09***
30.25
6731 .34***
1 49 • 9 5
378 .92
6203. 07***
23.65
6179 •
616.07
5563. 3s**
99 • 42
301. 7
5162. 45***
59 • 06
5103. 3Ec***
2231 .27***
2231. 27***
83.72
2147 •
133.87
2013. 67***
—3.75
0’ ’FREL0 l
• 00
• 00
• 00
•
• 00
17.66
no
.00
.00
273. 14
• 00
• 00
.00
• 00
• (JO
• 00
.00
616.07
• 00
301 • 7
59 • 06
.00
• 00
3.75
P’EETA EL0I
2.45
110. 65
100.55
128 • 65
1 49 • 9 5
35 • 33
202.90
130.02
201 • 55
336. 17
I 0’ • 3 1
3.25
20.50
30.25
1 49 • 9 5
3 78 • $2
23.65
1232.14
99 • 42
602.95
118 • 12
$3. I c ?
133.87
.00
157

-------
•3 E FL I.
H . t t’.
LINE
2
N DF
5
A E T.
516. 8 i
i M
2017.42***
(PM
v.
1
S17.8
251.91
251.91
503.82
2
517. 8’
81.17
.00
<1. 17
2
6
1
538.00
539.0(
16 .33**
73. -’ 5
.00
73.85
2
7
7
570.94
571.94
11 .4 *-
30.75
.00
30.75
2
8
572.51
1579.73***
2
9
2
577.30
577 . 43
1579.73* *
57.40
.00
57.40
2
10
2
580.72
580.89
1S22.33***
65.60
.00
65.60
2
11
2
582.67
522.
1/,56.73* *

.
65.60
2
12
2
5..
S8’ .C()
l 3’ t.13* *
60.95
.00
2
13
2
S6.51
586.66
1330.1$*4z*
61.50
.00
61.50
2
l
2
525.9
589.11
1268. 68**
61.50
.00
61.50
2
15
2
591.25
591.40
1207.
61.50
.00
61.50
2
16
2
593.75
593.82
1145.68 **
39.97
3 977
79Q 5
2
17
1
2
597.81
597.81
597.81
1 105.71***
-33.75
14.61
33.75
63 .2
.00
77.90
2
18
2
605.13
605.14
1124.55***
15.37
66.62
32.00
2
19
608.1
605.14
608.14
1109.47***
-33.75
15.76
33.75
68.29
.00
.05
2
0
1
2
615.81
615.82
615.85
1127 . 47***
21.02
39.05
315.28
169.20
336.30
208.25
2
21
1
626.59
627.66
1067.40***
1067.40
.00
1067.40
3
1
479.90
2872.12***
3
2
2
481.77
491.76
2872.12’***
-11.25
11.25
.00
3
3
1
483.55
483.67
2883.37***
97.56
97.56
195.12
3
4
1
484.41
484.41
2785.R0***
-3.75
3.75
.00
3
.
1
487.03
487.03
27 9.55***
-11.25
11.25
.00
3
6
1
488.51
488.50
2800.80***
—11.25
11.25
.00
158

-------
LINF
NQ3L)E.
EL V.
FT.
,
ØVF FL .
FLOt
3
7
1
489.91
489.90
2812.05***
-11.25
11.25
.00
3
8
1
/i94.17
494.16
2823.30***
-11.25
11.25
.00
3
1
497.27
497.51
2834.55***
222.55
222.55
445.10
3
10
2
497.53
497.55
26 12.flO***
11.86
11.86
23.72
3
11
2
507.95
507.95
2600.14***
48.25
209.10
257.35
3
12
1
513.86
514.86
2551.89***
45.19
195.83
241.02
3
13
1
517. 5
518.12
2506.70***
104.46
104.46
208.92
3
14
2
527.75
528.75
2402.23***
32.43
140.54
172 . 7
3
T- E
15
1
INEILTRATIk)N
550.00 2369.80***
553.26 2369.80 .00
} TE IS PRESENTLY .015 GPLLØNS/MIN/FT
23 9 .%O
38 Lc\TE8 L
159

-------
LINE NODE AREA
1 1
1 2
1 3
4
—)
1 6
1 7
8
1 9
1 10
1 11
12
1 13
14
1 22
1 23
2 1
2 2
2 3
2
2
1
2
I
2
2
2
EL F V.
FT.
419 • 00
421 • 47
425.72
426.57
428 • 00
429 • 8?
429..7
430.65
430.89
437. 12
437. 12
438 • 00
438 • 00
441.37
442.77
443.37
443.40
444.21
444. 18
445. 21
445.31
446.92
449 • 20
450. 11
450.6
458 • 20
458 • 20
461 • 38
461 • 39
463.07
463.10
466. 16
466.98
466.27
474.60
474.65
478 • 45
479 • 69
485.27
485.39
486.08
488 • 58
88 • 9 1
491 • 18
491 • 18
494. 11
494. 18
500.96
501.08
9432. 35***
9432. 35***
9.32. 35***
9432. 35***
9432. 35***
p432. 35***
62.45
9369 • 9 0***
110.65
9259 • 25 **
100. 55
9 158 • 70***
128.65
9030. 05***
1 49 • 9 5
8 8fl. I0***
35.33
202.90
864 I.87***
130.02
851 1.%5***
201.55
8310. 30***
159 • 55
1 09 • 3 1
‘041. 44* *
3.25
8038. 19***
20.50
8017. 69***
30.25
7987. 44***
1 49 • 9 5
378.82
7458. 67***
23.65
7435. 02***
1232.14
6202.8 7***
99 • 47
602.95
5500. 50***
118.12
5382. 37’4 **
2061 • 25***
2061 .25***
83.72
1977. 52***
133.87
1 49 • 95
35.33
202.90
I
I S.) • U
20 1 • 55
336. 17
1 09 • 3 1
3.25
20.50
30.25
1 49 • 9 5
3_7% •82
23.65
1232. 14
99 • 42
602.95
118.12
RUN # 5
FL0 1 ;
OVFRFLO%
APE FLOW
6PM
6P ”1
( p
1 15
1 16
1 17
1 18
19
20
1 21
.00
62.45
.00
I10.6
.00
100.55
.00
128.65
• 00
• 00
.00
• 00
.00
1
76 • 62
• 00
.00
• On
.00
• 00
• 00
• 00
• 00
• 00
• 00
.00
.00
83.72
.00
133.P7
1
2
1
2
2
I
1
2
I
2
1
2
2
160

-------
LINE N DF AF E ELFV.
F T.
2 4 506.7%
1 507.36
2 5 516.84
1 517.84
2 517.84
2 6 538.00
1 539.00
2 7 570.9
2 571.94
2 8 571.91
2 9 574.77
2 574.90
2 10 576.68
2 576.84
2 11 577.75
2 577.92
2 12 578.40
2 578.55
2 13 579.76
2 579.91
2 14 580.96
2 581.11
2 15 582.04
2 582.19
2 16 583.16
2 583.40
2 17 584.70
1 584.70
2 584.78
2 18 588.82
2 588.32
2 19 590.64
1 590.63
2 590.64
2 20 597.60
1 597.60
2 597.60
2 21 608.39
1 609.45
3 1 491.18
3 2 493.62
2 493.61
3 3 495.90
1 496.34
3 4 496.94
1 496.94
3 5 500.16
1 500.16
FL0
C-PM
1843. 65**
- .35
1844. fl0***
503.82
81.17
1259 •
73.85
I 185. 15***
30.75
11 5’ . 40***
11 54. 40***
57.40
1097. 00***
65.60
1031 .
65.60
965.80***
60.95
904.8 r. ***
61 • 50
343. 35 **
6 1 • 50
781.8 5***
6 1 • 50
720. 35***
79.95
640. 40***
-256.29
77.90
8 18 • 79***
4.32
814. 47***
-256.29
I • 48
1069 • 28***
.66
I • 22
1067. 40***
1067.40
3321 • 1 2***
3321. 12***
-9.49
3330.6 1***
195. 12
3135. 49***
- .35
3135.8 4***
-9 • 49
ØVEP L@ W
0 PM
.35
• 00
• 00
.00
• 00
.00
.00
.00
.00
.00
.00
.00
.00
256.29
.00
77.68
256.29
82.57
335.64
207.03
.00
9 • 49
.00
.35
9 • 49
PPEI FLØt
0 PM
.00
503.82
81.17
73.85
_)(_ . I
57.40
65.60
65.60
6 (T) • () 5
6 1 • 50
6 1 • 50
6 1 • 50
79 • 9 5
• 00
77.90
82.00
.00
336.30
20% • 25
1067. ‘ 0
• 00
195. 12
• 00
• 00
161

-------
LINE NODE AF F ELEV.
F 1.
3 6 501.84
I 501.8 4
7 5fl3.5R
I 503.5
3 508.87
I 508.82
3 9 512.69
513.53
3 in 512.96
2 513.03
3 11 524.45
2 52L .Li5
3 12 529.79
1 529.97
3 13 533.58
2 534.00
3 14 539.61
2 539.61
3 15 550.00
1 553.26
FLOW
3145. 33***
-9 • 49
31 54.8 2**
-28 • 49
3183. 30***
-28. 1i9
3211 . 79 **
445. 10
2766. 69***
23.72
2742.9 5***
40.71
2702. 24**
114.39
258 7.8 5***
2C9 .92
2379 .92***
2369 .8
2369.80
0 V F f FL 0 W
C -fM
9 • 49
29 • 149
28 • 49
• 00
• 00
216.64
126.63
• 00
163.85
.00
AHA F L®W
f PM
• 00
• 00
• 00
4 5. 10
23.72
257.35
241 • 02
208 • ‘-1?
172.97
23 .90
OF LATERAL
THE INFILTR! TIøN R TE IS PPESFNTLY.015 GALLeN5/MIN/FT
162

-------
SELECTED BIBLIOGRAPHY
1. Atkinson, Bernard, ZdzislawKemblowski, and J.M. Smith,
“Measurements of Velocity Profile in Developing Liquid Flows”,
American Institute of Chemical Engineering Journal , Vol. 12,
No. 5, January 1967, pp. 17-20.
2. Baxter, Kerwin, “Data Analysis of In-House Friction Reducing
Polymers”, The Western Company, Research Division, Richardson,
Texas, November 1968.
3. Boggs, F.W. , and J. Thompsen, “Flow Properties of Dilute
Solutions of Polymers” ,U. S. Rubber Company, Research Center,
Wayne. New Jersey; Final Report, Part I, Contract No.
Nonr 3120(00), Office of Naval Research, Washington, DC,
February 1966, AD 666 581.
4. Boggs, F.W. , et al, “Effect of Solute on Turbulent Field”, U.S.
Rubber Company, Research Center, Wayne, New Jersey; Final
Report, Part III, Contract No. Nonr 3120(00) and N00014-66-
CO 322, Office of Naval Research, Washington, DC, December
1967, AD 666 581.
5. Carver, C.E.,Jr. , et al, “An Investigation of Velocity Profiles
in the Laminar Sublayer with Non-Newtonian Additives, Using
High-Speed Photomicroscopy”, University of Massachusetts,
Amherst, Massachusetts; ERI Report No. 69-3, Contract No.
Nonr 3357(07), Office of Naval Research, Washington, DC,
May 1969, AD 698 385.
6. Dodge, D.W., and A. B. Metzner, “Turbulent Flow of Non-
Newtonian Systems”, American Institute of Chemical Engineering
Journal , Vol. 5, No. 2, June 1959.
7. Elata, C. , and J. Tirosh, “Frictional Drag Reduction”, Israel
Institute of Technology, Haifa, Israel; Contract No. 62558-4093,
Office of Naval Research, Washington, DC, December 1964.
8. Fabula, A.G., J.L. Lumby and W.D Taylor, “Some Interpretations
of the Toms Effect”, Pennsylvania State University, University
Park, Pennsylvania, August 1965.
9. Hoyt, J.W. , “The Friction-Reducing Effects of High Polymers”,
Naval Ordnance Test Station, Pasadena, California.
163

-------
10. Leach, P.B. and K.C. Little, “Preliminary Experiments on Drag
Reducing Agents in Light Water Concentrate Solutions”, Surface
Chemistry Branch, Chemical Division, Naval Research Laboratory,
Washington, DC; NRL Memorandum Report 2030, August 1969,
AD 694 455.
11. Liaw, Gin-Chain, Jacques L. Zakin, and Garry K. Patterson,
“The Effects of Molecular Characteristics of Polymers on Drag
Reduction”, University of Missouri, Rolla, Missouri.
12. Lindgren, E. Rune, “Friction Reduction Effects on Turbulent
Flows of Water in Rough Pipes by Dilute Additive of High
Molecular Weight Polymer”, Bureau of Ships General Hydro-
mechanics Research; Program S-R0090l01, Research Contract
Nonr 2595(05), June 1965.
13. Little, Ralph C. , “A Review of 6. 1 Work Units in Drag Reduction
with Emphasis on Current Problems, Progress and Landmarks”,
Naval Research Laboratory, Washington, DC; NRL Memorandum
Report 1957, January 1969, AD 684 770.
14. Little, Ralph C. , “Drag Reduction by Dilute Polymer Solutions in
Turbulent Flow”, Surface Chemistry Branch, Chemical Division,
Naval Research Laboratory, Washington, DC; NRL Report 6542,
May 31, 1967, AD 654 160.
15. Lord, D.L. , B.W. Hulsey and L.L. Melton, “General Turbulent
Pipe Flow Scale-Up Correlation for Rheology Complex Fluids”,
Haliburton Company; Paper No. SPE 1680, Society of Petroleum
Engineers of AIME, Dallas, Texas, 1966.
16. Lumley, J. L. , “Drag Reduction by Additives”, Review of Fluid
Mechanics , Vol. 1, 1969, p. 367.
17. Lumley, J. L. , “The Toms Phenomenon: Anomalous Effects in
Turbulent Flow of Dilute Solutions of High Molecular Weight
Linear Polymers”, Applied Mechanics Review , Vol. 20, No. 12,
December 1967.
18. Merrill, E.W. , “Turbulent Flow of Polymer Solutions”, Department
of Chemical Engineering, Massachusetts Institute of Technology;
Contract No. Nonr-3963(lO), Office of Naval Research, Washington,
DC, April 5, 1965.
19. Metzner, A. B. , J. L. White and M. M. Denn, “Behavior of
Viscoelastic Materials in Short-Time Processes”, Chemical
Engineering Progress , Vol. 22, No. 12, December 1966, p. 81.
164

-------
20. Metzner, A.B., J.L. White and M. M. Denn, “Constitutive
Equations for Viscoelastic Fluids for Short Deformation Periods
and for Rapidly Changing Flows: Significance of the Deborah
Number”, American Institute Chemical Engineering Journal ,
VoL 12, No. 5, September 1966, pp. 863-866.
21. Metzner, A.B. , “Pipeline Design for Non-Newtonian Fluids”,
R&D Department, Colgate-Palmolive-Peet Company, Jersey City,
New Jersey.
22. Metzner, A.B. , and M. Graham Kerr, “Turbulent Flow
Characteristics of Viscoelastic Fluids”, University of Delaware
Newark, Delaware, February 1964.
23. Meyer, Warren A. , “A Correlation of the Frictional Characteristics
for Turbulent Flow of Dilute Viscoelastic Non-Newtonian Fluids
in Pipes”, American Institute Chemical Engineering Journal ,
May 1966, p. 522.
24. Oustenbout, R.5. , and C.D. Hall, Jr., “Reduction of Friction
in Pipes”, Society of Petroleum Engineers; AIME Paper No.
1596-G, October 1960.
25. Paterson, Robert W. ,“Turbulent Flow Drag Reduction and
Degradation with Dilute Polymer Solutions”, Harvard University,
Cambridge, Massachusetts; Contract No. N00014-67-A-0298-
0002, Office of Naval Research, Washington, DC, June 1969.
26. Paterson, G.K. , J.L. Zakin and J. M. Rodriquez, “Drag
Reduction-Polymer Solutions, Soap Solutions, and Solid Particle
Suspensions in Pipe Flow”, Industrial and Engineering Chemistry ,
Vol. 61, No. 1, January 1969.
27. Poreh, J. , et al, “Studies in Rheology and Hydrodynamics of
Dilute Polymer Solutions”, Israel Institute of Technology, Haifa,
Israel; Contract No. F610 57-68-C-00 51, Mathematical Science
Division, Office of Naval Research, Washington, DC, March
1969, AD 690 264.
28. Pruitt, G.T. and H.R. Crawford, “Drag Reduction, Rheology and
Capillary End Effects of Some Dilute Polymer Solutions”, The
Western Company, Research Division, Richardson, Texas; Final
Report, Contract No. 60 530-8250, Naval Ordnance Test 3tation,
Pasadena, California, July 1963.
29. Pruitt, G. T. and H. R. Crawford, “Effect of Molecular Weight
and Segmental Constitution on the Drag Reduction of Water
Soluble Polymers”, The Western Company, Research Division,
Richardson, Texas; Report No. DTMB-1, David Taylor Model
Basin, Contract Res earch Administration, Hydromechanical
Laboratory, Washington, DC.
165

-------
30. Pruitt,G.T. , Bernard Rosen, and H.R. Crawford, “Effect of
Polymer Coiling on Drag Reduction”, The Western Company,
Research Division, Richardson, Texas; Contract Nonr 4306(00),
David Taylor Model Basin, Contract Research Administration,
Washington, DC.
31. Ram, Arie, Ehud Finkeistein, and Chaim Elata, “Reduction of
Friction in Oil Pipelines by Polymer Additives’, I & EC Process
Design and Development , Vol. 6, No. 3, July 1967, p. 309.
32. Reusswig, G.H. , and F.F. Ling, “Reassessment of the Wall
Effect of Non-Newtonian Fluid Flow”, Air Force Materials
Laboratory, Air Force Systems Command, Wright-Patterson Air
Force Base, Ohio: Technical Report AFML-TR-68-207, September
1968, AD 678 575.
33. Savins., J.G. * “ Contrasts in the Solution Drag Reduction
Characteristics of Polymeric Solutions and Micellar Systems”,
Mobil R&D Corporation.
34. Savins, J.G. , “Drag Reduction Characteristics of Solutions
of Macromolecules in Turbulent Pipe Flow”, Society of Petroleum
Engineers Journal , September 1964, p. 203.
35. Savins, J.G., R.F. Burdyn, andG.C. Wallick, “Scaling Pumping
Requirements - Inelastic Fluids in Turbulent Flow and Inelastic!
Elastic Fluids in Laminar Flow”, Field Research Laboratory,
Socony Mobil Oil Company, Inc., Dallas, Texas.
36. Seyer, F.A., andA.B. Metzner, “Drag Reduction in Large Tubes
and the Behavior of Annular Films of Drag Reducing Fluids”,
Canadian Journal of Chemical Engineerinq Vol. 47, December
1969.
37. Seyer, F.A., et al, “Turbulent Flow Properties of Viscoelastic
Fluids”, University of Delaware, Newark, Delaware; Contract
No. Nonr 2285(03) Task No. NR062-2941, Office of Naval
Research, Washington, DC, 1967, AD 660 788.
38. Seyer, F.A., and A. B. Metzner, “Turbulence Phenomena in Drag
Reducing Systems”, University of Delaware, Newark, Delaware,
December 1969.
39. Shin, Hyunkrak, “Reduction of Drag in Turbulence by Dilute
Polymer Solutions” (Doctoral Thesis), Department of Chemical
Engineering, Massachusetts Institute of Technology, May 1965.
40. Slattery, John C., “Scale-Up for \Jiscoelastic Fluids”, American
Institute Chemical Engineering Journal , Vol. 11, No. 5, p. 831.
166

-------
41. Tiapa, Gerald A. , and Barry Bernstein, “Elastic Recovery and the
Toms Effect , Illinois Institute of Technology; Contract No.
N00014-67-A-0210-00l Task No. NRO41-438, Office of Naval
Research, Washington, DC, July 1968, AD 673 009.
42. Tulin, Marshal P., “Hydrodynamic Aspects of Macromolecular
Solutions”, Hydronautics, Inc. , Laurel, Maryland; Technical
Report No. 353-4, Contract No. Nonr-4181(00) NRO6Z-325,
Office of Naval Research, Washington, DC, May 1967,
AD 653 097.
43. van Driest, E.R., “The Damping of Turbulent Flow by Long-
Chain Molecules”, North American Rockwell, Inc., Anaheim,
California; Contract AF49(638)-1442, Air Force Office of Scientific
Research, Office of Aerospace Research, September 1967,
AD 660 883.
44. van Driest, E.R. , “Turbulent Drag Reduction Polymeric Solutions”,
North American Rockwell Corporation, Downey, California;
AFOSR 70-0593TR, American Institute of Aeronautics and
Astronautics, New York, New York, January 1970, AD 702 466.
45. Virk, Preetinder Singh, “The Toms Phenomenon-Turbulent
Pipe Flow of Dilute Polymer Solutions” (Doctoral Thesis), Indian
Institute of Technology, Kharagpun Massachusetts Institute of
Technology, November 1966.
46. Walsh, Myles, ‘Theory of Drag Reduction in Dilute High-Polymer
Flows”, California Institute of Technology; Naval Ordnance
Test Station, Physics of Fluids Conference, October 1966.
47. (The) Western Company, “Polymers for Sewer Flow Control”,
Water Pollution Control Research Series WP-20-22, August 1969.
48. White, Frank M. , “An Analyses of the Effect of a Polymer Additive
on Turbulent Wall Friction and Pressure Fluctuations”, U.S.
Navy Underwater Sound Laboratories, Fort Trumbull, New London,
Conn.; USL Report No. 881, December 1967, AD 666 818.
49. White, J.L., A.B. Metzner, “Constitutive Equations for
Viscoelastic Fluids with Application to Rapid External Flows”,
American Institute Chemical Engineering Journal , March 1965,
pp. 324-330.
50. White, J.L., A.B. Metzner, “Measurement of Normal Stresses”
University of Delaware, Newark, Delaware.
167

-------
51. White, W. D., and D. M. McEligat, “Transition of Mixtures of
Polymers in a Dilute Aqueous Solution”, ASME Publication ,
Paper No. 69 WA/FC-ZO.
52. Whits itt, N. F. , L. J. Harrington, H. R. Crawford, “Effect of
Wall Shear Stress on Drag Reduction of Viscoelastic Solutions”
Viscous Drag Reduction , Plenum Press, 1969.
53. Zimmerman, Barry, “How to Dissolve Polyox Water Soluble
Resins”, Union Carbide Corporation, Chemicals and Plastics,
R&D Department, Tarrytown Technical Center, Tarrytown, New
York, March 1970.
168

-------
METRIC CONVERSION TABLE
TO CONVERT...
acres
cubic feet
feet
feet of water
feet/sec
feet/sec
gal Ions
gal Ions
gal Ions/rim
I nches
I nches
Poise
pounds
pounds/cu ft
pounds/sq ft
pounds/sq in
Slug
temperature (F) —32
sq meters
cu meters
meters
kgs/sq meter
cms/sec
meters/mi n
cu meters
liters
I iters/sec
cent i meters
meters
Gram/cm. sec.
kilograms
kgs/cu meter
kgs/sq meter
a tmos ph e res
Kilogram
temperature (°C)
4,047.
0.02832
0.3048
304.8
30.48
8.29
3.785 x
3.785
0.06308
2.540
2.540 x l0_2
I .00
0.4536
I 6.02
4.882
0.06804
14.59
5/9
I NTO...
MULTIPLY BY...
169

-------
TECHN (CAL REPORT DATA
(Please read !ns.trucrions on the reverse before completing)
1. REPORT NO. 2.
EPA-600/2-77-18g
3. RECIPIENT’S 4CCESSIOr#NO
4. TITLE AND SUBTITLE
CONTROL OF SEWER OVERFLOWS BY POLYMER INJEC-
TI ON
5. REPORT DATE
September 1977(Issuirig Date)
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
R.W. Chandler and W.R. Lewis
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORG \NIZATION NAME AND ADDRESS
Water Utilities Department
Ci ty of Dal 1 as
Dallas, Texas 75201
10. PROGRAM ELEMENT NO.
lBC6l]
11. %OJTfl CT/GRANT NO.
11020 DZU
12. SPONSORING AGENCY NAME AND ADDRESS
Municipal Environmental Research LaboratQry_...
Office of Research and Development Cm., Oh
Environmental Protection Agency
Cincinnati, Ohio 45268
13. TYPE OF REPORT AND PERIOD COVERED
Final
14.SPONSORINGAGENCYCODE
EPA/6 0 0/14
15. SUPPLEMENTARY NOTES
P.O. Richard Field (20l)-321-6674 FTS 340-6674
16. ABSTRACT
In the past, the operator of a sewage collection system has had
three alternatives for dealing with overloaded sanitary sewers; ignoring
them, diverting them to storm sewers and streams, or pumping to other
locations. An EPA-sponsored research program entitled, “Polymers for
Sewer Flow Control,” Contract No. 14-12-34, suggested a possible alter-
native system wherein the capacity of a sewer might be increased by the
injection of certain water-soluble chemicals to reduce turbulent frictio
This concept was further developed and demonstrated during this project,
EPA Grant entitled, “Elimination or Reduction of Sanitary Sewer Over-
flows in the Bachman Creek Sewershed,” which was executed in Dallas,
Texas. This report was prepared to help operators of sanitary sewage
collection systems determine the feasibility of using turbulent friction
reduction, designing an injection facility, choosing a friction reduc-
ing material, and evaluating the results.
I?. KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS c. COSATI Field/Group
Combined sewers, Fluid friction,
Fluid flow, Overflows, Water pollu-
tion, Polymers, Addition resins
Water-so4jblechem c s
Injection facility,
Turbulent friction
reduction, Polymer
injection
l3B
lB. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report )
UNCLASSIFIED
21. NO. OF PAGES
180
20. SECURITY CLASS (Thispage
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9.73)
170

-------