WATER POLLUTION CONTROL RESEARCH SERIES
DAST-13
Design of a Combined
Sewer Fluidic Regulator
J.S. DEPARTMENT OF THE INTERIOR • FEDERAL WATER POLLUTION CONTROL ADMINISTRATION
-------
Design of a Combined
Sewer Flu/die Regulator
The Development of Basic Configurations
and Design Criteria for Applications
of Fluidics in Sewer Regulators
FEDERAL WATER POLLUTION CONTROL ADMINISTRATION
DEPARTMENT OF THE INTERIOR
by
Bowles Engineering Corporation
9347 Fraser Street
Silver Spring, Maryland 20910
Program No. 11024 DGZ
Contract No. 14-12-486
October 1969
-------
FWPCA Review Notice
This report has been reviewed by the Federal Water
Pollution Control Administration and approved for
publication. Approval does not signify that the con-
tents necessarily reflect the views and policies of
the Federal Water Pollution Control Administration.
ii
-------
ABSTRACT
The objective of this program was to demonstrate feasibility, and
to develop a workable configuration for a combined sewer Fluidic regulator,
whose purpose is to minimize combined sewer discharge while protecting
interceptor sewers from overloading during storm flows. A second objective
•was to develop design procedures and criteria for the general application
of this concept to municipal sewer diversion requirements, including pre-
liminary investigations of construction methods, costs, and maintenance
requirements. A third objective was to establish a plan and location for
an operational demonstration of the concept with a cooperating municipality.
All objectives were successfully met. A generic Fluidic Regulator
configuration was evolved which diverts 0 to 75% of the combined sewer
flow away from the interceptor as a function of water level sensed in the
interceptor sewer, or combined sewer, in either an analog or digital
operational mode. Application design criteria were evolved for a range
of small to medium sized municipal sewers, in terms of a few basic
parameters. Projected installation costs are only slightly more than for
conventional diversion structures; while the anticipated construction and
maintenance requirements are simple and minimal.
The City of Philadelphia was established as the demonstration site,
and a demonstration unit should become operational in late 1970. Recom-
mendations were made for experimental activity to improve regulation
linearity; expand application size limit, and to better definitize con-
struction methods and costs.
This report is submitted in fulfillment of Contract 14-12-486,
between the Federal Water Pollution Control Administration and the
Bowles Engineering Corporation.
iii
-------
CONTENTS
ction
1
2
3
4
5
6
7
8
9
10
L Title
CONCLUSIONS AND RECOMMENDATIONS
INTRODUCTION
APPLICATION OF FLUIDIC REGULATORS
EXPERIMENT CONSTRUCTION AND
MEASUREMENTS
DISCUSSION
DEMONSTRATION PLANNING AND LIAISON
ACKNOWLEDGEMENTS
REFERENCES AND PUBLICATIONS
GLOSSARY OF TERMS AND ABBREVIATIONS
APPENDICES
Page
1
3
19
39
44
90
97
99
100
102
-------
FIGURES
Figure
Page
1 Final Sewer Regulator Geometry with
Blunt Splitter Nozzle 2" x 4"
Fluidic Sewer Regulator Switching Action 100%
Diversion to Interceptor Sewer No Aspiration ..
3 Fluidic Sewer Regulator Switching Action Diversion
Toward Interceptor Sewer Slight Aspiration 8
4 Fluidic Sewer Regulator Switching Action 50-50
Diversion, Aspiration at Both Controls 10
5 Fluidic Sewer Regulator Switching Action Maximum
Diversion to Combined Sewer No Aspiration 10
6 Typical Existing Diversion Structure 11
7 Schematic Arrangement - Fluidic Sewer Regulator .. 12
8 Fluidic Automated Irrigation System 15
9 Fluidic Irrigation Diverter Installed at Washington
State University Experimental Farm 15
10 Flow vs Supply Head vs Orifice Coefficient for
20
11
12
13
1 C
10
1 C
Discharge Coefficient vs Supply Head for Fluidic
Fluidic Sewer Regulator Geometry Dimensions
Control Line Diameter vs Line Length vs Nozzle
Pni-nm;»r« ctroAt Rtruo.turfi . Washinaton. D. C
21
23
26
27
27
29
vii
-------
FIGURES (Continued)
Figure Pa9e
1 7 Potomac Street Structure Details
Washington, D. C ............................. 30
18 Railroad Avenue Structure, Washington, D. C ..... 31
19 Railroad Avenue Structure Details
Washington, D. C ............................. 32
20 Intercepting Sewer in Cobbs Creek Park Slot
Regulator Installation, City of Philadelphia ....... 33
21 Potomac Street Structure No. 43 a Fluidic
Regulator Installation, Washington, D. C ........ 34
22 Anacostia Main Interceptor Structure No. 7
Fluidic Regulator Installation, Washington, D. C.. 35
23 Intercepting Sewer in Cobbs Creek Park Fluidic
Regulator Installation, City of Philadelphia ....... 36
24 Fluidic Sewer Regulator Test Layout .............. 40
25 Test Installation Showing Head Box .............. 41
26 Test Installation Showing Sewers and Regulator
Insert ........................................ 41
27 Sewer Regulator Test Insert ..................... 43
28 Fluidic Irrigation Diverter 1/2" x 1/2" Nozzle
Used in Predictive Analysis Tests ............... 45
29 Fluidic Irrigation Diverter 1/2" x 1/2" Nozzle
Operating at 1 00% Diversion .................... 45
30 Minimum Head vs Nozzle Height for 100% Diversion
of Fluidic Irrigation Diverters ................... 47
31 Bias Orifice Test Circuit ....................... 48
viii
-------
FIGURES (Continued)
Figure Page
32 Sensor Control Orifice Area vs Diversion
for Fluidic Sewer Regulators 49
33 Irrigation Diverter Exhibiting Analog Control 50
34 Irrigation Diverter in State of 100% Diversion 50
35 Changes in Setback Geometry 52
36 Changes in Splitter Geometry 52
37 Geometric Bias Test Configuration 53
38 Model Test Configuration Showing Combined and
Interceptor Sewers and Element Insert 55
39 Model Test Configuration Showing Combined and
Interceptor Sewers and Element Insert 55
40 Basic Irrigation Test Model with 2" x 1" Nozzle ... 56
41 Large Control Pockets on Irrigation Geometry
Nozzle 2" x 1" 56
42 Cutaway Sidewalls on Irrigation Geometry
Nozzle 2" x 1" 58
43 Short Sidewalls with Cutaway Nozzle 2" x 1" 58
44 Rounded Sidewalls with Cutaway Nozzle 2" x 1"... 59
45 Short Sidewalls with Splitter Upstream
Nozzle 2" x 1" 59
46 Short Sidewalls Rounded Splitter Nozzle 2" x 1" ... 60
47 Short Sidewalls Rounded Splitter Less Downstream
Setback Nozzle 2" x 1" . 60
ix
-------
FIGURES (Continued)
Figure Page
48 Analog Sewer Regulator Geometry Nozzle 2" x 4" .. 61
49 Analog Geometry with Larger Control Pockets and
Upstream Splitter Nozzle 2" x 4" 61
50 Analog Geometry with Long Nozzle, Pointed
Splitter Nozzle 4" x 4" 62
51 Final Analog Sewer Regulator Geometry
Nozzle 2" x 4" 62
52 Final Geometry with Pointed Splitter
Nozzle 2" x 4" 63
53 Maximum Diversion vs Supply Head for Fluidic
Sewer Regulators 65
54 Interceptor Slot.Configurations „. 66
55 Minimum-Maximum Diversion From Interceptor
vs Combined and Interceptor Weir Settings for
Fluidic Sewer Regulators 68
56 Maximum Diversion vs Combined Weir Height 69
57 Float Valve Mechanical Sensor 70
58 Float Valve Test Installation 70
59 Float Valve Sensor Test Setup .... „ 71
60 Float Valve Sensor Areas 72
61 Diversion vs Interceptor Water Level for Fluidic
Sewer Regulator 2 " x 4 " Nozzle 73
62 Diversion vs Sensor Area Change 2" x 4" Nozzle
Supply Head = 13.0" 75
-------
FIGURES (Continued)
Figure Pa9e
63 Diversion vs Sensor Area Change 4" x 4" Nozzle
Supply Head =13.0" ....................... o... 76
64 Diversion vs Interceptor Level for Fluidic Sewer
Regulators Aspect Ratio =0.5 Nozzle 2" x 4" ..... 77
65 Diversion vs Sensor Area for Fluidic Sewer
Regulators Aspect Ratio = 0.5 Nozzle 2" x 4" ..... 78
66 No-Moving-Part Sensor Test Setup ...... „ ........ 79
67 No-Moving-Part Sensor Push-Pull Bottles ......... 81
68 No-Moving-Part Bottles Installation ............. 81
69 Diversion vs Control for No-Moving-Parts
Push-Pull Sensor Nozzle 2" x 4" ................ 82
70 Air Water Flow Ratio vs Diversion ............... 82
71 Multiple Sensor Test Setup ..................... 84
72 Diversion vs Supply Head for Multiple Sensor
Control of Fluidic Sewer Regulator ............... 85
73 Diversion vs Supply Head for Multiple Sensor
Control of Fluidic Sewer Regulator ............... 86
74 Shrouded Discharge Low Velocity Flow
Interference .................................. & 7
75 Shrouded Discharge Low Velocity Longer Shroud
No Interference ............................... 87
76 Basic Irrigation Test Model 2" x 3" Nozzle ....... 88
77 Simulated Debris , Fouling Test .................. 88
78 Preliminary Demonstration Plan .................. 96
xi
-------
FIGURES (Continued)
Figure
Page
B-l Relationship of Discharge to the Differential Head
Across the Diverter. Fluidic Irrigation Diverter
8" x 8" Nozzle m
B-2 Discharge Coefficient vs Head Fluidic Irrigation
Diverter Test on 8" Diverter - Ft. Collins 112
B-3 Flow vs Normalized Inlet Head 1/2" x 1/2" Nozzle
12.5% Setback Blocked Port on Non-Control Side
of Regulator Orifice on Other Control 113
B-4 Flow vs Head for Fluidic Sewer Regulators 114
C-l Percent Diversion as a Function of Bias Orifices
1/2" x 1/2" Irrigation Element Supply Head = 6 hn .. 117
C-2 Percent Diversion as a Function of Bias Orifices
1/2" x 1/2" Irrigation Element Supply Head = 15hn. 118
C-3 Maximum Diversion vs Supply Head H9
C-4 Maximum Diversion vs Combined Weir Height
Element 710-1 Nozzle 2" x 4" Interceptor Exit
Area= 12.2 in2 120
C-5 Minimum-Maximum Diversion vs Combined and
Interceptor Weir Settings Element 829-1 Nozzle
2" x 4" Supply Head = 10.25" 121
C-6 Minimum-Maximum Diversion vs Combined and
Interceptor Weir Settings Element 818-1 Nozzle
4" x 4" Supply Head = 13.0" 122
C-7 Minimum-Maximum Diversion vs Combined and
Interceptor Weir Settings Element 818-1 Nozzle
4" x 4" Supply Head = 17.0" 123
C-8 Minimum-Maximum Diversion vs Combined and
Interceptor Weir Settings Element 818-1 Nozzle
4" x 4" Supply Head = 17.0" 124
xii
-------
FIGURES (Continued)
Figure Page
C-9 Diversion vs Height in Interceptor Element 710-1
Nozzle 2" 4" Sensor 2M Weirs Set for Maximum
Diversion .................................... 126
C-10 Diversion vs Sensor Area Element 710-1 Nozzle
2" x 4" Sensor 2M Weirs Set for Maximum
Diversion .................................... 127
C-ll Low Diversion, Low Interceptor Water Level
Float Valve Control ............................ 128
C-12 High Diversion, Loaded Interceptor Float Valve
Control ...................................... 128
C-13 Area of Sensor Orifices vs Height of H£O Interceptor
Sensor 2M .................................... 129
C-14 Flow Ratio vs Diversion Optimum Sensor and
Weir Settings ................................. 130
E-l Air Flow Through Orifice. „ ...................... 135
E-2 Air Pressure Drop Nomograph .................... 137
Xlll
-------
SECTION 1
CONCLUSIONS AND RECOMMENDATIONS
Conclusions*
1. The concept of a Combined Sewer Fluidic Regulator was found
feasible and practical, based on a series of scale model tests in which
the scale factor varied from about 1:6 to 1:20 of a "typical" municipal
diversion sewer. Of the three basic fluidic arrangements tested, (geometric
bias, fixed orifice bias, and variable orifice bias) only the variable orifice
bias provided complete, predictable analog flow diversion. The fixed
orifice bias provided good partial analog, or fully digital flow diversion,
which may be quite satisfactory for many system applications. The geo-
metric bias arrangement was abandoned early in the test program. While
mechanically simplest, it proved to be overly sensitive to dimensional
tolerances, and provided a significantly poorer range of flow diversion
performance.
2. A basic Fluidic element geometry was developed which reliably
diverts from 0 to greater than 75% of the combined flow from the interceptor
to the receiving waters outlet as either a digital or proportional function
of water level variation in either interceptor or combined sewer. This
characteristic has been demonstrated over a considerable range of inlet
heads. At low inlet heads (corresponding basically to dry weather flow),
all the flow enters the interceptor. The unit's performance remains rela-
tively unaffected by variations in inlet height-width ratio that would be
encountered in adapting to the normal range of municipal installations.
3. A design rationale has been evolved for establishing the
principal Fluidic Element geometric parameters to correspond to the
normal range of municipal combined sewer regulator requirements.
4. A preliminary analysis has shown that the installation of a
Combined Sewer Fluidic Regulator is similar in nature and overal com-
plexity to a conventional leaping weir, or side flow diversion structure.
It is estimated that the use of a Fluidic regulator would not increase the
cost of a large diversion structure, and would add only about 20% to the
cost of a small diversion structure. It is estimated that the modification
of an existing structure for a Fluidic Regulator would cost 20 to 50% of the
original installation cost, depending on size and degree of rework required.
*Terminology related to Fluidic technology is described in the Glossary.
-------
5. A simulated fouling test has shown a very low susceptibility
to fouling by solid or soft sheet-like debris in the water flow. It is
concluded that fluidic sewer regulators will require significantly less
maintenance and/or surveillance to assure proper operation in municipal
service.
Recommendations
1. It is recommended that a full scale pilot model of a Combined
Sewer Fluidic Regulator be constructed, tested and evaluated in a typical
municipal diversion point, or equivalent, based on the design criteria
developed in the subject program. The unit should be operated on typical
combined sewer flows, incorporating both normal dry weather and storm
flows. The installation should be adequately instrumented to provide
real time and recorded readings of all pertinent levels, flows, and control/
sensor pressures. The unit should be evaluated over a period of at least
one year to properly assess the combined effects of dry and wet weather
flows, including seasonal variations in each. This program should also
serve as a basis for establishing the nature and frequency of maintenance
or other services to keep this type of sewer regulator operable.
2. It is recommended that additional larger scale model testing
be performed to improve the accuracy of the performance prediction
criteria, when applying these criteria to the design of sewer regulator
structures with regulator inlet areas in excess of about 4 square feet.
3. It is recommended that additional design and testing be
performed to improve the linearity and predictability of the flow diversion
vs interceptor water level change characteristic for the no-moving-part
push-pull sensor. This will be highly desirable in those applications
where a network of sewer regulators is to be controlled from a central
municipal command center.
4. It is recommended that study and experimentation be performed
to evaluate the relative cost effectiveness of three suggested methods
of construction: plastic interaction region insert; cast concrete inter-
action region insert using a plastic reusable mold; cast concrete insert
using a disposable mold.
-------
SECTION 2
INTRODUCTION
The Combined Sewer Overflow Problem
The general problem which has prompted this program is the pollution
of natural water resources by overflows from combined sewers. According
to the American Public Works Association, roughly three-fourths of all
combined sewerage system overflows in the United States have their
sources in combined outfalls1. For this reason, existing combined sewers
are regarded as one of the most troublesome sources of pollution in this
country today.
The problem is materially aggravated by the distribution of these
combined sewerage systems. Serving more than one-fourth of the sewered
populace of the country, combined sewers are especially prevalent in
cities having populations in excess of 100,000. Such cities have gen-
erally been long established and their streets are underlain by such a
complex of sewage, water, transportation, electric, steam and telephone
lines that a separate sewer system is totally impractical. Beyond this,
of course, the time, the inconvenience and particularly the cost of con-
version to a separate system, even in many communities where conversion
could still be considered, also render this approach unacceptable so long
as any other solution is available.
For this reason, the Federal Water Pollution Control Administration
is currently investigating a number of alternate approaches to the problem
of reducing the substantial pollution problem of existing combined sewer
installations. Among the potential solutions being examined is that of
storing all overflows, either in the existing system or else in large
lagoons or underground reservoirs, and providing the capability of pumping
the stored waste water back into sewage treatment plants. By such means,
the treatment plants can be kept operating near full capacity at all times
with the combination of normal dry-weather (sanitary) flows and pump-
back flows. Study programs are either now in progress or have recently
been completed under FWPCA sponsorship to evaluate this type of in-
stallation for a number of cities.
^-''Report on Problems of Combined Sewer Facilities and Overflows, 1967,"
prepared by APWA under Contract No. 14-12-65, sponsored by the FWPCA.
-------
A necessary part of the storage concept is that of regulating the
flow of sewage into the treatment plant during heavy storms. Regulation
is required to prevent overloading of the interceptor sewers and/or the
treatment plant, itself, during heavy storm flows. Regulation also
assures that 100% of the dry weather flow is diverted to the interceptor
sewers, thus minimizing the required capacity and running costs of any
pumpback facility and reducing the tendency for sanitary solids to be
deposited in the storage reservoirs or low-use portions of the system.
Conventional regulation devices presently used in combined sewers
for regulating flow to adjacent interceptors range from several types of
manually-, float- or flow-operated gate valves to orifices, siphons, and
a variety of weir configurations. Such devices all suffer from a reliability
problem. This problem is caused by the environmental conditions which
prevail in sewers and by the characteristics of the sewage, itself. High
humidity coupled with acidic gases forms a corrosive atmosphere which
attacks most materials not actually immersed in the liquid. The flow,
itself, is both corrosive and charged with debris, including sticks,
leaves, newspapers, sand, stones and other solids. The result is that
the regulator devices tend to deteriorate rapidly, due to corrosion, to
physical damage, or to the fouling and jamming action of debris.
The moving-part devices are especially susceptible to these types
of damage. Although their performance can be excellent, such performance
can be secured only through frequent, periodic and expensive preventive
maintenance. Such no-moving part devices as weirs and orifices are,
of course, less susceptible to damage. They suffer, however, from the
fact that they are inherently very poor regulators even when operating
perfectly. There is, therefore, a definite need for a simple regulator
device which combines the superior reliability of no moving parts with
the improved performance available from properly functioning conventional
valves and gates.
The Regulator Problem
The problem of providing regulators that are accurate, reliable,
and easily maintained is unfortunately a problem that has largely eluded
solution thus far. There are, of course, passive regulator devices that
are relatively dependable. Such devices as weirs and orifices, for
example, have no moving parts and thus represent two of the more reliable
methods for flow regulation.
Such devices are also relatively poor in their performance as
regulators. Orifices, for example, regulate only to the extent that flow
is proportional to the square root of the combined sewer head increase.
Weir flow is proportional to the 3/2 power of the fluid head over the
-------
crest of the weir, for a rectangular weir installation. Other notch shapes
yield other relationships between weir head and flow (e.g. , a triangular
notch having the apex of the triangle down allows flow proportional to
the 5/2 power of weir head). Actual regulation, however, is usually in
terms of combined sewer head rather than interceptor sewer level for most
regulator configurations involving orifices or weirs. Hence, even if the
orifices or weirs offered appropriate accuracy, they could still not regu-
late in terms of the required control parameter which is the level of flow
in the interceptor sewer.
Other regulatory devices are available which offer substantially
improved accuracy and, in some cases, capability of control in accordance
with interceptor rather than combined sewer level. Fairly simple examples
of such devices include float-operated gate valves and tipping
gates used in small-size regulators. The former are gate valves actuated
by floats measuring the flow level in the interceptor. The latter consist
of gates having a pivot point below their center lines; excessive upstream
flow causes them to be closed, while low flow rates permit them to fall
completely open. Both of these devices could be made to regulate very
well if the regulated fluid were clean and non-corrosive.
Unfortunately, the atmosphere within the sewer is typically wet
and corrosive. The flow, itself, in addition to being corrosive is charged
with solids ranging from fecal matter to earth, sticks, stones, leaves,
rags, paper, and a whole variety of other objects. These solids produce
jamming of gates and fouling of mechanisms, while corrosion results in
rapid deterioration of mechanical parts. The result is that these other-
wise satisfactory simple regulators have proven short-lived and unreliable
in service.
Still other types of moving parts regulators have been developed
for large-scale regulation. For very large sewers, electrically- or
hydra ulically-opera ted sluice gates driven by interceptor level sensor
signals may be used. These large, complex and expensive installations
are capable of exceptional performance as regulators. Unfortunately,
they are also subject to the identical operating life and maintenance
problems that plague the float-operated and tipping-gate valves. Fre-
quent inspections and maintenance are required to keep the large devices
operational. Unfortunately, the jamming or stalling of a gate during a
severe storm is usually irreparable until the waters have receded to a
point where the installation may be safely entered by men. Hence,
although they do offer greater performance, this performance cannot be
obtained reliably without costly inspection and maintenance.
-------
The Use of Fluidics for Sewer Regulation
In seeking a regulation device or technique that combines the
reliability of weirs, orifices and other no-moving-parts devices with
the regulatory capability of moving-part valves, fluidics certainly
appears promising.
Concerning the reliability requirement, the majority of fluidic
devices have as one of their most obvious characteristics the elimin-
ation of moving mechanical parts. In effect, the fluid through its own
internal flow dynamics performs those functions which normally require
moving parts. As a result, most of the problems of moving parts in
fluid systems are eliminated. Such problems as wear, backlash and
slop, friction and binding, and the need for lubrication and seals, for
example, all disappear when moving parts are absent. Replacing the
moving parts are contoured flow channels constructed of materials re-
quiring only structural integrity and chemical compatibility with the
fluid involved. Hence, fluidic devices are obvious candidates for
sewerage system use from the standpoint of potential reliability.
In addition to their simplicity, however, fluidic devices also
offer a performance capability either duplicating or closely approximating
that of conventional moving parts valves for regulation of systems where
fluid flow is the principal controlled output. This occurs because most
of the sensors, transducers, interface devices and activators are re-
placed by simple flow channels through which the fluid itself supplies
control signals, feedbacks and power outputs in the form of such funda-
mental parameters as flow and pressure. In many applications, additional
simplification is possible because the energy which provides the control
signals is obtained directly from the main power source, thus eliminating
any need for auxiliary energy sources. Inasmuch as sewage regulation
is an example of such an application, the second criterion for considering
fluidic devices as potential regulators is also obviously met.
Basic operation of the Fluidic sewer regulator as shown in Figure 1
is as follows: When the interceptor control port of the Fluidic sewer
regulator is closed off, the jet stream is directed along the corresponding
sidewall shown in Figure 2. In this condition the interceptor control
registers a partial vacuum indicating a tendency to aspirate air as a
result of the venturi effect. In this state the combined control port is
open, allowing ample air to be aspirated which helps keep the jet stream
attached to the interceptor sidewall. As the interceptor control is opened
slightly, air is aspirated and the jet stream begins to pull away from the
sidewall as shown in Figure 3. In this state some flow begins to be
diverted into the combined discharge. As the interceptor control
-------
INTERCEPTOR
EXIT
COMBINED
OUTFALL
COMBINED
SEWER
CONTROL
INTERCEPTOR
SEWER
CONTROL
VENTURI
NOZZLE
Figure 1. Final Sewer Regulator Geometry with Blunt Splitter
Nozzle 2" x 4"
-------
Figure 2. Fluidic Sewer Regulator Switching Action
100% Diversion to Interceptor Sewer
No Aspiration
Figure 3. Fluidic Sewer Regulator Switching Action
Diversion Toward Interceptor Sewer
Slight Aspiration
-------
is opened farther, more air is aspirated and the jet stream gradually
swings away from the interceptor sidewall until it is in the center of
the diverter, when the control is fully open, giving 50-50 diversion,
see Figure 4. In a like manner, if the combined control is gradually
closed the aspiration will be restricted and the jet stream will be drawn
by the resulting pressure differential until, when the control is fully
closed, almost all the flow will be directed out of the combined dis-
charge, see Figure 5.
Fluidic devices can be considered to combine the best features
of both of the two types of conventional flow regulators. They have
the no-moving parts reliability of orifices and weirs. Yet they are also
capable of performance and installation flexibility entirely comparable
to that of servo-controlled valves or gates. Both first costs and mainte-
nance costs can be greatly reduced, in addition, while overall operational
reliability can be greatly improved.
Consider the problem of flow regulation at an existing diversion
structure. The diversion structure may presently utilize as simple a
regulator device as the leaping weir shown schematically in Figure 6.
Conversely, it may be equipped with a completely automatic sluice gate,
servo-opera ted, which limits flow diverted to the interceptor in accord-
ance with the remaining capacity of the interceptor line.
Figure 7 shows a possible arrangement for a fluidic flow regulator
which occupies the same diversion structure volume. The concept con-
sists basically of embedding a fluidic flow regulator element into the
weir of a conventional dammed-weir installation. One exit of the regu-
lator leads to the interceptor portion of the installation while the other
exit leads to the receiving waters. A small exit weir provided in the
interceptor exit of the diverter is significantly lower than a similar weir
in the combined sewer exit. Hence, normal low-velocity dry-weather
flow passes through the near side of the regulator and into the interceptor
line. Flow passages would be less than full under these conditions and
no flow would go over the outfall weir.
The fluidic regulator is designed for control by a balance between
pressures applied at control ports on either side of its intake nozzle.
One of these ports, shown on the near side of the regulator in Figure 7,
consists of a fixed area orifice through which air is aspirated into the
flowing stream. The opposite port is shown connected to a level sensor
located in the interceptor sewer. Under conditions of low interceptor
flow the area through which air is aspirated from the interceptor is larger
than the fixed area orifice. Hence, pressure on the far side of the flow
stream is greater than that on the near side and this AP maintains flow
-------
Figure 4. Fluidic Sewer Regulator Switching Action
50-50 Diversion, Aspiration at Both Controls
Figure 5. Fluidic Sewer Regulator Switching Action
Maximum Diversion to Combined Sewer
No Aspiration
10
-------
COMBINED
SEWER
INTERCEPTOR
SEWER
Figure 6. Typical Existing Diversion Structure
11
-------
Control Port
Elevated Exit Weir
Combined
Sewer
Outfall
Combined Flow
Weir
Communication Lines
Fixed Area Orifice
Simple Level Sensor
Air Slot
Interceptor Flow
Figure 7. Schematic Arrangement - Fluidic Sewer Regulator
-------
into the interceptor. As interceptor flow increases, however, the aspiration
area at the level sensor is reduced. This change in area produces a cor-
responding change in the AP, resulting in flow modulation to the combined
sewer outfall of the regulator as the level of liquid in the interceptor sewer
increases, see Figures 2 to 5.
The foregoing description of a fluidic regulator is intended primarily
to illustrate the operating principles of a system utilizing such devices.
Clearly, variations in geometry and size from that shown are possible.
Likewise, different forms and arrangements of sensors are also possible
including, for example, lead or anticipation sensors in the upstream com-
bined sewer, and in the upstream interceptor line. One element of design
which has not been discussed directly heretofore is the question of analog
versus digital operation. Both types of operation are possible, of course,
and the question of type is therefore best answered by simultaneously
considering both the needs of the installation and the design parameters
yielding a given level of performance for each type of regulator.
By nature, the properties of a fluidic regulator are a function of
the geometry of its internal flow passages. The basic requirement for
construction material, therefore, is simply that its geometry should be
unaffected by either the operating environment or the fluid passing through
it. Since this requirement is basically identical to that for the sewer,
itself, it follows that the materials and techniques normally used for
sewer construction should be equally applicable to fluidic sewer elements.
Hence, concrete, brick and stone should be entirely adequate in most
installations. The exception is in the narrow venturi section of the
inlet nozzle where relatively high fluid velocities occur. In a few in-
stallations where sustained high velocity flow occurs, a local "armoring"
may be necessary using a tough, smooth, non-corrosive material, such
as corrosion resistant steel, high quality plastic, or fiberglass.
In the construction of these elements dimensional tolerances can
be relaxed to as much as ± 5% except in the immediate vicinity of the
inlet nozzle and sidewalls where a ± 2% tolerance is necessary. For
reference, the latter is roughly equivalent to a tolerance of ± 1/4 inch
for a one-foot nozzle width. Construction of small flow control elements
can probably be done in concrete on a prefabricated basis. Large elements
can either be poured directly on site or else built up of prefabricated
subassemblies.
Program Approach
The technical approach for the subject sewer regulator program is
based on the work conducted by the Bowles Engineering Corporation toward
the development of a proprietary Fluidic automatic agricultural irrigation
13
-------
system. The irrigation system requirement is similar in many ways to
the sewer regulator system requirement described in the previous section.
Both handle relatively large water flows, operating at quite low gravity
heads. Both may handle water heavily charged with solid particles, and
on occasion, debris. Both are faced with significant environmental
corrosion problems; both require high reiliability with a minimum of
maintenance and monitoring. It is highly desirable that both systems
operate without external or auxiliary energy sources, other than the
main water flow, and both systems should be economical both in terms
of initial and operational costs.
A schematic arrangement for a Fluidic automatic irrigation system
is shown in Figure 8. This system is currently the subject of a U. S.
patent application. It is under consideration by the Hawaiian Sugar
Planter's Association for the automated irrigation of sugar cane, and a
similar system is currently being demonstrated at the Washington State
University Experimental Farm at Othello, Washington, for general
automated farm irrigation as shown in Figure 9. This system will
automatically irrigate a number of growing areas to the desired water
depth whenever water is applied.
The system consists of a number of large fluidic diverters connected
in series downslope. Water reaching the first diverter is directly initially
to the adjacent growing area, then switched downslope to the next di-
verter. Associated with each diverter is a water level sensor, which
senses a predetermined water depth in a typical furrow, and provides a
signal to the diverter to switch.
In operation, each diverter acts as a digital logic flip-flop. In
order to eliminate the need for positive pressure control flows, the
pressure in the diverter adjacent is reduced below atmospheric ambient
by narrowing the inlet nozzle in the manner of a venturi meter. Ambient
air is thus aspirated into the control ports to provide control flow. In
its simplest form the diverter flow is thus controlled by capping off as-
pirated air flow on the side of the diverter when water flow is desired,
and opening the opposite side. In the case of the irrigation diverters,
this arrangement is modified such that switching occurs when one control
port aspirates air continuously through a small orifice, while the opposite
port is either capped, or opened wide. Control flow air is allowed into
the diverter through the level sensor, and interconnecting line. The
level sensor consists simply of an inverted cup, whose open end is set
at the desired maximum water level in the furrow. When the water level
rises to the cup, air cannot enter, hence the first diverter switches its
flow to the next downslope diverter. A saucer is generally placed under
the cup so that the diverter will remain switched downslope if the water
in the furrow recedes before all the fields have been irrigated.
14
-------
Lateral
Distribution
Ditch
Furrows
Diverters
Crop Areas
Figure 8. Fluidic Automated Irrigation System
Figure 9. Fluidic Irrigation Diverter Installed at
Washington State University Experimental Farm
15
-------
Program Procedure
The procedure followed on the subject program has been to utilize
the above described technology and practice as a starting point. Of
specific use was the Fluidic element, sensor, and communication line
configuration data.
The program was planned to include a number of phases: a predictive
performance analysis; the construction of a test setup; the construction,
testing, and evaluation of a number of experimental Fluidic Regulator
models; the generation of an application design rationale for applying the
test results to practical municipal design requirements; a preliminary
configuration and cost analysis; and a planning and liaison effort for
evaluating a full scale pilot model under operational conditions.
A major technical objective in the course of the program was the
modification of a number of characteristics of the irrigation element con-
figuration. These are described below.
The principal modification was the need for proportional flow
diversion as well as the on-off, or digital type, diversion characteristic
of the irrigation systems.
A second modification concerned adding the capability to operate
with a considerable difference in elevation between the outlet to the
interceptor, and the outlet to the receiving waters. The interceptor
outlet must be lower in order that normal dry-weather, or sanitary flows,
be allowed to flow with minimum impedance to the interceptor, and then
onto the treatment plant. On the other hand, the outlet to the receiving
waters must be elevated to prevent the dry weather flow from entering
the receiving waters, except under storm flow conditions when the inter-
ceptor is running near or at its capacity.
A third modification involved the redesign of the overflow structure
so that very heavy storm flows can flow over, or around the regulator
without causing significant changes in its regulation performance.
Another possible modification, which is desirable for future sewer
systems, is the capability for remote control from a centrally located
command center. Remote regulator control is currently the subject of
experiments in several large municipalities, and in time will probably
be used in most large, and many smaller cities. However, due to its
futuristic nature, this requirement was not investigated during the initial
phases of the subject program covered by this report; nevertheless, it
appears promising based on Fluidic irrigation experience.
16
-------
Program Implementation and History
On February 4, 1969, the U. S. Department of Interior, through the
Federal Water Pollution Control Administration, entered into a contract
with the Bowles Engineering Corporation to conduct an initial research
and development activity. The purpose of this activity has been to establish
design and performance criteria for general application of fluidic devices
to a representative range of sewer sizes and locations as a potential means
for reducing or controlling combined sewer overflows.
The program was conducted under the personal supervision of Mr.
Peter A. Freeman, Principal Engineer and Manager of the Water Manage-
ment Group at Bowles Engineering Corporation. A total of four tasks were
involved in addition to the preparation of reports. These tasks and their
completion dates are summarized below:
Task I - Predictive Analysis. The purpose of this task was the
analytical establishment of general design criteria for an interceptor
sewer junction flow control based on fluidic technology. This task was
started on February 4, 1969, and completed May 15, 1969.
Task II - Design and Fabrication of Scale Model Junction. Task II
had the goal of preparing to substantiate the general design criteria de-
veloped in Task I by designing, fabricating, and assembling a scale
model junction suitable for test. This task was started about February 15,
1969, and was completed May 24, 1969.
Task III - Testing. The purpose of Task III was to perform a complete
series of tests of the model constructed under the preceding task. Specific
objectives of the tests were as follows:
o Substantiate basic design criteria.
o Perform flow and blockage tests using both small (sand, silt
and pebbles) and large (newspaper pulp, rags, confetti, lint
and tree limbs) contaminants.
o Generate preliminary cost estimates for installations of
various sizes.
o Test to determine the maximum practical lengths of air
aspirator lines.
o Study the effects of reduction of air aspirator line cross-
sectional area due to clogging or foreign material buildup.
o Determine the preferred material(s) for air aspirator lines.
17
-------
This task was started on May 5, 1969, and completed September 9,
1969.
Task IV - Liaison and Planning. The purpose of Task IV was to
establish liaison with one or more municipalities desiring to work toward
the design and installation of an interceptor junction with a fluidic flow
control. Once liaison was established, a Phase II plan covering the
design of a demonstration combined sewer flow regulator installation
was to be completed jointly by the cooperating municipality and EEC and
included in the Phase I final engineering report. This task was started
on July 8, 1969, and was completed September 3, 1969, with the agree-
ment by the City of Philadelphia to request a FWPCA Demonstration Grant
to evaluate a Fluidic Sewer Regulator.
This contract has been the first phase of a planned four-phase
program. The overall program purpose is to develop, build, install and
demonstrate a fluidic sewer regulator. As cited above, the purpose of
the Phase I contract reported on herein has been to establish design and
performance criteria on the application of fluidic devices to the problem
of reducing or controlling combined sewer overflows.
Succeeding phases of the program cover respectively the design,
construction and evaluation of a test system at an actual sewer site of
the cooperating municipality. Phase II provides for detailed design of
the test system including device design by EEC and A&E services by the
municipality. Phase III covers the construction and installation of the
test system. Phase IV provides for demonstration and evaluation of the
test system over a twelve-month time period.
As described later in Section 6, it is expected that Phase II through
IV will start early in 1970, and will continue through 1971.
18
-------
SECTION 3
APPLICATION OF FLUIDIC REGULATORS
The intent of this section is to discuss the application of Fluidic
Regulators to practical municipal combined sewer requirements. This
section will include specific design criteria and procedures, suggested
installations in typical municipal combined sewers, rough cost esti-
mates, and suggested approaches to the maintenance and service of
Fluidic Regulators.
It should be noted that the information here presented must be
considered experimental and preliminary at this time. It is expected
that refinements and improvements in the design and cost criteria and
procedures will occur in the course of forthcoming testing to be con-
ducted under a FWPCA Demonstration Grant.
Water Channel Sizing
The principal parameter affecting the water channel size is the
maximum desired flow into the diversion channel from the combined
sewer, before flow begins to bypass (over or around) the diversion
structure. Also to be determined is the desired inlet head at which
this flow occurs. The inlet head is established by the weir or dam
height at the regulator inlet nozzle. The inlet nozzle area is computed,
using the nomograph shown in Figure 10. (Note that the discharge co-
efficient, CD, is a function of the relative magnitude of the inlet head
as compared to the nozzle height. This relationship is shown in Figure
11. For first approximation purposes, a value of 1.0 is suggested.)
When the nozzle area, AN, is determined, the values of nozzle
height and width must be chosen, so that the best overall diversion
performance can be realized. A number of considerations affect this
choice.
1. Based on the program test results, a greater range of flow
diversion is obtained with higher aspect ratio nozzles; i.e. , hn*
large compared with wn*. Note that flow diversion represents
that part of the flow which does not enter the interceptor.
*hn is the height of the regulator supply nozzle.
wn is the width of the regulator supply nozzle.
19
-------
t:
50
40
30 -
20
|
>- 10
(0
Q
7.0 -
5.0
3.0
:
- 1.0
I .7
.5 -
.3 -
.1
.07-
.05-
s Orifice Coefficient
CD =.90 |=_
CD = i.oo
cD-i.10
CD=1.20
ssaEgs^;:3F3H:n^a_ 3 p
7.0
f-j •-••;—!—*-— -;
-
I
-
_
g
--
*"
B
U
-
BC
-
-
,
sc
'
c
o
-
.
5C
B
_
-
_ 1.0
Figure 10. Flow vs Supply Head vs Orifice Coefficient
for Fluidic Sewer Regulators
20
-------
1.20
K
_ J
nlet '
jpaH hf i
1 h
\
ft
K/
Fluidic Regulator
Discharge Head
0.85 '
Nozzle Height
0.80
+
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0
SUPPLY HEAD RATIO - INLET HEAD MINUS DISCHARGE HEAD \~ h(i
hn
Figure 11. Discharge Coefficient vs Supply Head for Fluidic Regulators
-------
2. Flow diversion is also strongly affected by the ratio of
inlet water head to hn. For ratios less than about 2, the range
of flow diversion drops off sharply.
3. A limiting value of undiverted flow must be determined
from the interceptor sewer flow handling capabilities, in that
the interceptor must be able to accept this amount of undiverted
flow when the combined sewer is running full. If the inter-
ceptor cannot accept this amount of flow, then the regulator
inlet nozzle area must be reduced, or the nozzle weir lowered.
It can be seen that the achievement of the best diversion charac-
teristics within the above indicated restraints may require several "trial"
designs. When wn is established, then the horizontal plane geometry
of the regulator can be established, using wn-based parameters shown
in Figure 12. Note that considerable flexibility is possible in the con-
figuring of both flow passages from a point downstream of the splitter
to the interceptor inlet orifice, or the outfall weir. A good working
value of the outfall weir height has been determined as (1.4) hn. A
good working value of the interceptor orifice width has been determined
as 1.4 wn, assuming a vertical dimension of hn.
Air Channel Sizing
Two procedures for air channel sizing can be used, depending
on the mode of regulation desired. Digital, or "on-off" action similar
to that used in irrigation systems, can be obtained with a small fixed
bias air orifice on the interceptor side; a larger fixed orifice will pro-
duce a hybrid type of action in which a limited degree of proportional
diversion occurs in the vicinity of the maximum diversion range (toward
the outfall). Below this range maximum flow into the interceptor occurs.
Complete analog action requires a variable bias orifice, such as the
push-pull sensor arrangement described later. Considerations for sizing
air channels for both modes of operation are described below.
1. Digital Action. The bias orifice (interceptor side) should
have an opening equal to .0015 AJJ, where A^j is the nozzle area.
The sensor orifice area should be . 004 AJJ.
The communication lines should be sized so that they produce
less than 10% of the pressure drop across either the bias, or control
orifices, when each is passing maximum airflow. As a guide to sizing
these lines, the static pressure at the bias port when the regulator is
switched to the interceptor side is -.3h, where h is the water head at
the inlet. (This is the maximum airflow condition.) Also the static
22
-------
asw
SIDEWALL
SETBACK
CONTROL
POCKET
VENTURI NOZZLE
Figure 12. Fluidic Sewer Regulator Geometry Dimensions
23
-------
pressure at the control port under the same conditions is about -. Ih
the water head at the inlet. This also corresponds to a maximum steady
airflow condition.
In a practical installation, the control orifice should be located
well above the interceptor sewer water line, so that sewage cannot be
drawn or "percolated" into it when the regulator is operating at high
head conditions.
2. Analog Action. For analog operation, the push-pull control
arrangement is recommended. The dip tube sensor located in the
interceptor sewer should have a maximum orifice area = . 02 AJJ.
The orifice should be shaped in the form of a vertical slot, or
alternately, a series of holes. The vertical distribution of the
orifice area is a function of range of interceptor water level over
which flow regulation is desired. Experiment has shown that a
linear variation of regulator diversion with interceptor water depth
will require a tapered area distribution, with the largest slot
width (or hole diameters) at the bottom of the dip tube.
In the experiments to date, a linear taper of the orifice slot has
been used. This has resulted in a rather non-linear diversion vs water
depth calibration, however. The reason for this is attributed to the fact
that the fluidic element's capability to aspirate airflow into a control
port drops off sharply as the main flow stream is diverted away from
that side of the element; thus, a relatively large decrease in area of
the control side sensor was necessary to produce a significant lowering
of pressure in that part, which would allow flow to start entering the
bias port. It is expected that a non-linear tapered slot, which widens
sharply near the bottom of water depth range, can be formulated to pro-
duce a much improved degree of linearity. Such a contour could be
formed by the area between a vertical line and a parabola. This can
be described by the formula:
ws = a (1 -1
T -^
where ws = sensor slot width
along contour
a = slot width at the
lower water level
limit
hs = slot height along
contour
D = maximum depth of
slot
24
-------
by simple integration, the area of the slot can be shown to be = 1/3 Da.
Additional testing will be required to fully evaluate the above or other
improved linearity contours.
A second dip tube sensor is required in the open side of the push-
pull controller. Its orifice should be constructed with the same distri-
bution of aiea as a function of variation in liquid depth, except that the
depth will be determined by the maximum control pressure difference
expected, the density of the controller liquid, and the relative cross
section areas of the open and closed sides. The most desirable choice
of the controller liquid is still open to experimentation. It appears that
the most significant requirements for the liquid are good chemical
stability, relatively low vapor pressure, low toxicity, and low surface
tension and/or viscosity. These requirements reflect the fact the liquid
must evaporate at a minimum rate; and when the regulator is called on
to operate, the ambient air can bubble through it with minimum restraint,
foaming, or chemical reaction.
Sizing of the communication lines can be accomplished by the
relationship shown in Figure 13. This graph shows the inside diameter
of the line as a function of line length and inlet nozzle area. In general,
pressure drops along the communication line of a Fluidic regulator oper-
ating in the analog mode will cause errors in the desired degree of
diversion. The maximum error will occur at the maximum airflow con-
dition which has been shown to occur near the 50% diversion point. In
formulating Figure 13, a maximum allowable error in diversion of 10%
has been assumed as a function of increased flow impedance from the
sensor orifice. The curves show the trade off of line I.D. (I.D. =
inside diameter) vs length, in order to remain within the tolerable im-
pedance increase limit. These relationships hold for variations in inlet
head (assuming a maximum value of discharge coefficient, CD) .
It is recommended that all dip tube sensors be relatively large in
diameter, as compared to the communication lines, to minimize the
chance of "percolating" sewage, or controller liquid into the communication
lines. It is recommended that communication lines and sensors be con-
structed from plastic pipe, in view of the very low operating pressures,
need for corrosion resistance, and low cost. It appears that several
types of plastic material would be equally satisfactory, including PVC,
ABS, PE, or possibly polypropylene. The sensor installations in the
interceptor (or combined) sewer should provide for deflectors to prevent
occlusion of the orifice slot by soft debris such as rags or newspapers.
A preferred location would be against a sidewall as shown in Figure 14.
If a sidewall installation is otherwise unsuitable, a midflow installation
could be made per Figure 15. In this arrangement, soft debris could be
caught by the sensor pipe, but the side deflectors and the downstream
location of the orifice slot would prevent interference with the basic level
sensing function.
25
-------
to
en
4-
w
tc
U
a
w
O
u
O
p 1
25
50
75 100 125 150
LENGTH OF CONTROL LINE IN FEET
175
200
Figure 13. Control Line Diameter vs Line Lengtn vs Nozzle Area
for Fluidic Sewer Regulators
-------
DEFLECTOR
COMMUNICATION
LINE
SENSOR
SLOT
ORIFICE
INTERCEPTOR
<^ FLOW
WALL LOCATION
Figure 14. Sensor Installation
MIDSTREAM LOCATION
Figure 15. Sensor Installation
27
-------
Installation
Description. In order to gain insight into the task of adapting
a Fluidic Regulator into a typical municipal diversion location, con-
struction layout drawings were obtained for several existing diversion
structures in the City of Washington, D. C., and Philadelphia,
Pennsylvania. These are shown in Figures 16 through 20. These
structures have been modified as shown in Figures 21, 22 and 23 to
outline possible arrangements and locations for Fluidic regulators,
including sensors, and control equipment. In each case the sizing
is only approximate, since the drawings are too limited in detail for
precise scaling. The upstream weir shown is higher than that for the
existing structure, since it is presumed that the Fluidic unit would be
sized to handle a considerably greater flow before bypass flow would
go over the regulator. This would permit a much greater flow to enter
the interceptor if the latter can accept it, thereby greatly reducing the
sewage flow into the receiving waters even during significant storm
flows.
Construction. The construction of the Fluidic regulators is not
shown in detail; however, the layouts are based on the assumption
that the Fluidic element interaction region would be constructed of
precast concrete shells, sealed against a concrete slab. This type
of construction minimizes the likelihood of air leaking into the element
interaction region. An alternate method of element construction would
be the insertion of a heavy molded plastic shell of the element inter-
action region, secured around the edges with concrete or other masonry
structure. Hard Polyethylene, or PVC, would be promising plastic
materials. These units could be fabricated by rotational casting. A
second alternate method would be the in-place casting of the element
in concrete around a disposable core contoured to the element's internal
geometry. For example, the core could be made of a plastic foam that
could be easily burned away, steamed away, or dissolved away with
solvents. A third alternative would be the on-site casting of the
element in concrete with reusable heavy plastic molding forms. It is
anticipated that the construction approach will be studied in detail
during the design task to be performed as part of the Demonstration
Grant activity.
Cost Estimates. At the time of writing of this report a detailed
cost estimate for a complete Fluidic Regulator Structure has not been
made, since such an estimate would require a detailed design for a
specific site, which will not be available until the next phase of the
program. The subject has been discussed with the responsible planning
and construction engineers in Washington, D. C. and Philadelphia to
obtain rough cost estimates of Diversion Structures with Fluidic Regu-
lators. The following responses were obtained.
28
-------
.c_4=_/i_/y
SCALE : /'=2O'
Figure 16. Potomac Street Structure
Washington, D. C.
29
-------
_jeii^
''
4'-0S -
4.17
^15*5, Inv. ~6.82(0vtrf/
/53s
6.29-
-6.29
!2aS/uice Gate
4.19
- Dam
I2'S-
(Overflow)
-6.89
DETAIL
i'= 10'-o'
Too of Dam
El. 8.30
. =6.8
Figure 17. Potomac Street Structure Details
Washington, D. C.
30
-------
PLAN
SCALE:l"-IOo'
Figure 18. Railroad Avenue Structure
Washington, D. C.
31
-------
70S
•Sluice Gate
f S/u/ce Gate
'3.O3 lO'S
-3.32 (Dam)
DETAIL
Figure 19. Railroad Avenue Structure Details
Washington, D. C.
32
-------
cm
co
2'-0" PIPE
.
\^\'.V..!;A'.;V :r.;-;.vT
OVERFLOW
10" PIPE
TO INTERCEPTOR
SEWER
Figure 20. Intercepting Sewer in Cobbs Creek Park
Slot Regulator Installation
City of Philadelphia
-------
SHROUD
CONTROL LINES
(OVERFLOW)
FLUIDIC REGULATOR
NO-MOVING-PARTS
SENSOR
(OVERFLOW)
rmrm.
•4'-0"S
FOULING COVER
Figure 21. Potomac Street Structure No. 43a
Fluidic Regulator Installation
Washington, D. C.
-------
(OVERFLOW)
FLUIDIC
REGULATOR
NO-MOVING-PARTS
SENSOR
INTERCEPTOR
WEIR
CONTROL LINES
REGULATOR
OVERFLOW
CHANNEL
DIP TUBE
FOULING
COVER
Figure 22. Anacostia Main Interceptor Structure No. 7
Fluidic Regulator Installation, Washington, D.C.
35
-------
00
OJ
2'-0" Dia. PIPE
DAM
.'." '. '•'. •'•.'•
X-
• XJ
'. '. ','*"' '',
ll"
t
^
; ,';'','•",
>
,' "A
'* .'• •
1
k
•
^jjjj
~
, 4,' • •
. . ' <
•'.*: -'
;•>.;;
"•'•.*'
V
^T
^
• ' i '
***
-^^
^
^S
-|
>' ;
y
'j|&
|
• A'
\
•'• 4
I
*N
.'.'A
!
/
/
|;
'*
»'.
'A
V
1
„• J^-jT
i
•.-•«'
i
X
/•s
..''.'I'- ',".'.".*'
/
1
•'/•>:'••*.
HROUD
/^ OVERFLOW
'/
>
10" Dia. PIPE
TO INTERCEPTOR SEWER
FLUID 1C REGULATOR
Figure 23. Intercepting Sewer in Cobbs Creek Park
Fluidic Regulator Installation
City of Philadelphia
-------
1. It was estimated that the cost of a diversion structure using
a Fluidic Regulator would range from equivalent, to approximately
25% greater than for a conventional (weir and side pipe) diversion
structure, depending on the size, and overall complexity of the
structure. For a large structure, with such features as a wet well,
leaping v;oir, elaborate manhole or other servicing access arrange-
ments, the incorporation of a Fluidic Regulator would add a
negligible fraction to the cost. For smaller structures, such as
the "slot" arrangement in use in Philadelphia, a modest increase
in cost would be required to incorporate a Fluidic Regulator. The
initial costs of Conventional Diversion Structure costs would
range from about $15,000 for the small, easily excavated instal-
lations, to over $50,000 for the larger, deeper, or more complex
installations. Thus the minimum cost for a Fluidic Regulator
installation might approach $20,000 while the maximum cost
would not be appreciably different from current practice.
2. The City of Washington, D. C. provided a rough cost
estimate for the modification of the current conventional structure
shown in Figure 17 to that indicated in Figure 21. The original
structure cost, at current prices, is estimated between $20,000
and $25,000. The cost of modification for a Fluidic Regulator
was estimated at $15,000. In this case the modification was
rather extensive, since the original sewer bottom would have to
be lowered to receive the Fluidic Regulator. In contrast, for the
projected modification of Figure 19, the Fluidic element could
be mounted completely within the existing combined sewer walls,
necessitating only a small hole for the diverted flow line. This
modification could probably be accomplished for significantly
less than $3,000.
3. The cost of Fluidic Regulator and auxiliaries is relatively
modest. Based on cost projections for large plastic irrigation
elements, a plastic insert of the element interaction region would
range in cost between $50 and $150, depending on size. The
cost of in-place cast concrete regulator structures is dependent
on many factors such as element size, concrete costs, labor
costs, etc. As an example, an element with a nozzle opening
of 1' x 1.5' will contain about 12 cubic feet, assuming a 3" shell
thickness. Assuming a concrete cost of $25/cubic yard, the shell
material cost would be around $13. The cost of construction labor
@ $4/hour would be around $20, and molding form rental would
probably be on the order of $50. Thus the total cost of the in-
stalled regulator, using either type of construction, less auxiliary
equipment, would be less than $100.
37
-------
4. The sensor, auxiliary controls equipment, and communication
lines would range in cost from $25 to $50 depending on size, This
equipment would be constructed of a high grade plastic material,
such as ABS, or PVC.
Service and Maintenance Procedures. Service and maintenance
requirements for Fluidic Sewer Regulators should be extremely straight-
forward. Since there are no moving parts, there are no requirements for
typical mechanism maintenance procedures, such as cleaning, lubrication,
replacement of moving seals, adjustment for wear, corrosion removal,
etc. The foreseeable situations requiring maintenance, or service, are
described as follows:
1. Cleanout. Removable cleanout hatches would be provided
in each outlet at a point abreast of the element splitter. These
would facilitate the removal of very large pieces of debris, or
occasional periodic flushing of accumulated solids and sediments
after long periods of only dry weather flow, if necessary. (Note,
experience on irrigation elements has shown that the increased
flow velocity through the nozzle, even under moderate flows should
minimize the sediment buildup in the element interaction region.
For example, irrigation elements that have been completely
"silted in" when the water is turned off, completely clear within
a few seconds when the water is again turned on.)
2. Auxiliary Equipment. For those installations using the
analog mode of operation, the fluid in the U-tube should be
checked for level at several month intervals. For analog or
digital mode operation all communication lines should be checked
periodically to prevent air leakage through seals, or joints.
It may be desirable to flush, or blow out communication lines
occasionally. This can be facilitated by installing fittings in
easily accessible locations, and flushing with potable water,
or compressed air.
38
-------
SECTION 4
EXPERIMENT CONSTRUCTION AND MEASUREMENTS
Test Setup
The performance of the subject program required a test setup suitable
for both the development of a satisfactory Fluidic Regulator element geometry
and its evaluation under simulated system conditions. Accordingly, a test
setup was designed to operate on an existing steel water test tank in the
EEC laboratory. This unit is shown in Figure 24. It consists of a large
wooden head tank which flows water into two closed conduits; one conduit
represents a combined sewer, the other an interceptor sewer. The latter
makes a 90° bend and runs across and under the combined sewer. The
Fluidic element fits at the junction so that one of its outlets discharges
back into the combined sewer, while the other outlet discharges into the
interceptor sewer. Both sewers discharge flow into the tank, from where
it is returned to the head tank through a large irrigation pump. The head
tank is equipped with bypass disc valves to control water head, while the
discharge from each sewer is controlled with a Bowles Engineering Cor-
poration proprietary flexible roller curtain gate. Photographs of the
completed unit are shown in Figures 25 and 26.
The test setup pump furnishes a total flow of about 1700 gpm. The
pump support structure has been designed to support two additional pumps
of the same type, thus giving the test setup the potential flow capacity
of about 5000 gpm. With one pump, each conduit can flow about 1/2 full
at a flow velocity of about 1. 7 feet/sec. If a scale factor of 10:1 were
assumed, a flow velocity of 17 feet/sec would be simulated. If all the
available flow passes through one conduit, the simulated velocity can
be doubled to 34 feet/sec, which can adequately simulate a flood flow
condition.
Experimental Fluidic Element Construction
The test setup is arranged so that the experimental Fluidic element
under test slides into place through the side of the combined sewer in
the manner of a desk drawer for easy removal and adjustment between
tests. Both combined and interceptor sewers are fitted with Plexiglas
sides in the vicinity of the element for visual observation of the element
performance.
The test elements were constructed of a stack of 1/2" marine ply-
wood laminations, a plywood floor, and a Plexiglas top, through which
39
-------
HEAD ADJUSTMENT 0 (.i (T\
CONTROL
WATER TANK (EXISTING)
WATER BOX
12 IN x 12 IN INTERCEPTOR DUCT
12 IN x 12 IN REGULATED DUCT
TEST ELEMENT BLOCK
ADJUSTABLE GATE
CIRCULATING FUMP
FLOW STRAIGHTENER
(AS REQUIRED)
Figure 24. Fluidic Sewer Regulator Test Layout
-------
Figure 25. Test Installation Showing Head Box
Figure 26. Test Installation Showing Sewers and Regulator Insert
41
-------
the aspiration control airflow was passed. The plywood laminations
were cut out to accommodate the maximum width dimensions anticipated
during the test program. A "basic" maximum nozzle width of 5" was
thus obtained. Element depths from 1/2" to 4" were obtained by adding
or subtracting plywood laminations. The basic element test insert is
shown in Figure 27. In the course of testing, internal geometry was
changed by bending strips of aluminum flashing to the desired contours
and anchoring these strips to the element floor and cover with strips of
caulking putty.
The combined sewer discharge weir was constructed of an aluminum
sheet, which was pivoted at the sewer floor to adjust the weir height.
The weir was sealed along the sewer sides with putty. The interceptor
discharge gate was designed initially to provide a narrow, full depth
passage to pass simulated dry weather flow without any head backup.
The gate height increased to full element depth along a diagonal. The
gate was vertically adjusted to vary the exit area.
Measurements
Water flow measurements were made by timing the interval required
to fill a 18. 7 gallon bucket. This provided a measurement accuracy
within 2%, except for timing intervals less than 10 seconds, where the
accuracies were on the order of 5%.
Water heads and weir settings were measured directly using scales
against the transparent Plexiglas sides of the test sections.
Control port pressures were measured using conventional water
manometers.
Control port airflows were established by measuring the pressure
differential across various sized, precision sharp edged orifice plates.
An inclined manometer was used for very small differential pressure
measurements.
42
-------
CONTROL LINE
COMBINED SEWER
SUPPLY
LOW
FLO
HEAD SCALE
\
OUTFALL WEIR
WEIR SCALE
ff INTERCEPTOR SEWER
EXIT ADJUSTMENT
PLEXIGLAS TOP \
Z
TEST REGULATOR
INSERT
ADJUSTABLE CURTAIN
GATE
INTERCEPTOR SEWER
TEST REGULATOR
INSERT
INTERCEPTOR EXIT
INTERCEPTOR EXIT
SCALE
PLEXIGLAS VIEW PLATE
Figure 27. Fluidic Regulator Test Insert
43
-------
SECTION 5
DISCUSSION
Predictive Analysis
The purpose of the predictive analysis was to obtain design criteria
from the established irrigation element configuration which would provide
a theoretical basis for the design of larger size model sewer regulators.
It was found after an initial effort, however, that a purely analytical
approach would have been prohibitive in time and cost, due to the general
mathematic complexity in handling two-phase flow field phenomena.
Consequently, it was decided that existing performance data available
from the testing of several large irrigation diverters, plus additional data
to be taken on a small diverter, could adequately provide the basis for
performance predictions of fluidic sewer regulator elements.
The test setup shown in Figures 28 and 29 utilizes a 1/2" x 1/2"
nozzle size irrigation diverter. Test data was taken to obtain mathe-
matical relationships for: Flow as a function of head, diversion as a
function of control orifice size, control pressure and flow as a function
of diversion and head, and geometrical modifications vs analog control
range. The results of the tests were graphically analyzed and compared
to theoretical predictions as well as actual data taken on large scale
models of the irrigation configuration. See Appendix A for log of tests.
Flow vs Head Analysis. In an effort to obtain design data to
predict the flow through water regulators for a range of supply heads
and nozzle sizes, flow data over a wide range of supply heads were
taken for a 1/2" x 1/2" nozzle diverter as part of the task 1 testing.
Additional flow vs head data were obtained from an 8" x 8" nozzle di-
verter from tests conducted by the Agricultural Research Service at the
Engineering Research Center, Fort Collins, Colorado. The data from
these two size diverters were used with the orifice equation to determine
the relationship of orifice discharge coefficient to the supply head as
shown in Figure 11. For the purpose of this study it was assumed that
the orifice coefficient did not change with aspect ratio*; however, for
aspect ratios much less than or greater than 1 this variable must be
considered. See Appendix B for a discussion of the orifice equation,
graphical data, and calculations of orifice coefficient. The orifice
. height of supply nozzle _ nn
*Aspect Ratio = width Qf supply nozzle - ^
44
-------
Figure 28. Fluidic Irrigation Diverter 1/2" x 1/2" Nozzle
Used in Predictive Analysis Tests
Figure 29. Fluidic Irrigation Diverter 1/2" x 1/2" Nozzle
Operating at 100% Diversion
45
-------
equation was then used along with the values of orifice coefficient
obtained from the report of test results from the Fort Collins irrigation
diverter to produce a nomograph of supply head vs flow through diverters
as a function of the nozzle areas and orifice coefficients. See Figure 10.
The head vs flow characteristic of the 1/2" x 1/2" irrigation diverter
tested was compared to the nomograph data and was found to closely
agree, even though the scale factor between the two diverters compared
was over 250:1. Also, head vs flow data were obtained in testing model
sewer regulators with 2" x 1", 2" x 4", and 4" x 4" nozzle sizes. A
comparison with the nomograph predicted flow values showed agreement
to within five (5) percent.
Testing of the small size diverter used for the predictive analysis
also yielded a graph of minimum supply head vs nozzle height for 100
percent diversion performance of the diverter. Comparison of this data
with the 8" x 8" Fort Collins diverter and the various size model sewer
regulators tested showed good agreement. This graph is shown in Figure
30 for diverters having square nozzles, or aspect ratio 1. Although this
curve applies specifically to the 100% diversion condition, it indicates
the relative ability of small elements vs large elements to provide flow
diversion for various head/nozzle height ratios. Accordingly, this
curve is useful in interpreting experimental test results taken on small
elements, on the effect of nozzle aspect ratio on diversion performance.
This will be described later.
Biasing Orifices. In order to obtain scaling information for control
orifice size the small 1/2" x 1/2" nozzle irrigation diverter was tested
with a fixed orifice of specific diameter on one control port and a range
of orifices on the opposite control as shown in Figure 31. The results
of the biasing orifice testing are shown graphically in Figure 32. See
Appendix C for actual data curves. Results showed that the bias orifice
for maximum diversion performance had to be 0.12 (10~2) AN/ where
AJJ = nozzle area; and the control orifice had to vary from nearly closed
to ten (10) times the area or approximately three times the diameter of
the bias orifice to provide maximum diversion control. The validity of
this data for reliable scaling was confirmed by its agreement with the
required control port sizes used on the 8" x 8" nozzle diverter tested
by the Agricultural Research Service at Fort Collins. Even though bias
orifice tests were performed on an irrigation configuration designed
primarily for digital operation, an analog control region was observed
which provided diversion control between 30 and 70 percent. See
Figure 32. Bias orifice control of irrigation diverters is shown producing
proportional diversion in Figure 33 and total diversion in Figure 34.
The bias control data were useful in designing a dip tube sensor con-
figuration to be used in the scale model test program and investigating
analog control.
46
-------
o
PS
o
o
a,
P
CO
7 8
h, NOZZLE HEIGHT, IN INCHES (ASPECT RATIO = 1)
10
Figure 30. Minimum Head vs Nozzle Height for 100% Diversion
of Fluidic Irrigation Diverters
-------
TO
INTERCEPTOR
SEWER
TO
COMBINED
DISCHARGE
ASPIRATED
AIR FLOW
FIXED BIAS
ORIFICE
ASPIRATED
AIR FLOW
VARIABLE
CONTROL
ORIFICE
DIP TUBE SENSOR
PROVIDES VARIABLE CONTROL
FROM RISING INTERCEPTOR
WATER LEVEL
Figure 31. Bias Orifice Test Circuit
48
-------
(O
O
1.2
1.0 ••
IN*
c
•l-l
O
•pH
M-l
§
1
a
O
u
lu
0
10
-------
Figure 33. Irrigation Diverter Exhibiting Analog Control
Figure 34. Irrigation Diverter in State of 100% Diversion
50
-------
As pointed out above, bias orifice tests showed that a region of
analog control existed in the performance of the digital irrigation diverter
tested. In order to gain some insight into the results of changes on
geometry to analog performance two basic modifications were tried on
the 1/2" x 1/2" nozzle diverter. The normal 12.5% setback of the side-
walls was changed to zero and 21%, and diversion performance data
taken, see Figure 35. Decreasing setback to zero reduced the diversion
considerably, whereas increasing setback to 21% had little effect. The
distance of the splitter downstream was changed by one nozzle width
with little effect in performance, see Figure 36. Because of poor access
to the internal geometry of the small size regulator, modifications were
limited. Since no improvement in analog performance was observed as
a result of these modifications, additional changes were planned for the
larger size scale model tests.
As a result of test data taken from the standard 1/2" x 1/2" nozzle
irrigation diverter and comparisons made with the 8" x 8" irrigation di-
verter, several significant and reliable scaling parameters were determined
as follows:
1. Flow through regulators vs supply head for any size regulator
nozzles, Figure 10.
2. Orifice coefficient vs supply head for any size regulator
nozzles, Figure 11.
3. Supply head vs nozzle height required for maximum diversion
performance, Figure 30.
4. Diversion vs control orifice area for fixed bias orifice areas,
Figure 32.
The testing carried out in this portion of the program also provided insight
into a test procedure and element design for the scale model testing.
The predictive analysis made from the tests provided a good foundation
for the starting of larger size regulator tests, particularly in choosing
supply heads and control sizes.
Sewer Regulator Model Development
Regulator configurations similar to that shown in Figure 37 having
a geometrical bias were tested for single control analog operation. In
such a configuration the water stream would normally attach to the left
sidewall and flow into the interceptor as long as the control sensor in
the interceptor was open to permit maximum aspiration. When aspiration
was reduced by interceptor flow level on the dip tube sensor, the water
stream would be pulled away from the left sidewall in proportion to the
51
-------
SETBACK
12.5%W
W
FILLER
SETBACK = 0
SETBACK 21%W
W
FILLER
IRRIGATION
GEOMETRY
MODIFIED
GEOMETRY
MODIFIED
GEOMETRY
Figure 35. Changes in Setback Geometry
IRRIGATION
GEOMETRY
W
MODIFIED
GEOMETRY
Figure 36. Changes in Splitter Geometry
52
-------
INTERCEPTOR
OUTLET
COMBINED
DISCHARGE
LARGE SETBACK
SMALL SETBACK
CONTROL TO
INTERCEPTOR
DIP TUBE SENSOR
Figure 37. Geometric Bias Test Configuration
53
-------
aspirated flow change until the jet was attached to the right sidewall
and all flow was directed to the combined discharge. Tests of this
geometry showed that geometrically biased elements could be made to
perform as described above; however, the geometry was extremely
critical. The maximum diversion performance was generally less than
60% into the combined discharge. Also, in the course of testing biased
configurations, it was not possible to duplicate any of the acceptable
designs. AE a result of this critical geometry problem this approach
was dropped as impractical and no further investigations were made.
Irrigation Configuration Modifications
In an effort to develop an analog diverter controlled by air aspiration,
a 2" x 3" nozzle size irrigation configuration was made to adopt to the
sewer simulation test setup shown in Figures 38 and 39. The basic
irrigation geometry shown in Figure 40 was tested with bias orifice control
as described above. In an attempt to obtain better analog performance,
tests were made with several splitter variations, including pointed, round,
flat, and concave shapes; and several weir settings. Results showed no
improvement in analog performance over the digital irrigation diverter.
The nozzle size was then changed to 2" x 1" (aspect ratio = 2) to provide
a greater range of flexible geometry variations. Variations were then
made in control pocket size, setback, cutaway of sidewall, sidewall
length, sidewall curvature, downstream sidewall setback, and splitter
bluntness and downstream location. See Log of Sewer Regulator Tests,
Appendix A, for a detailed account of these tests. -As a result of the
testing, a 2" x 1" element configuration and sensor were developed
having linear analog aspiration control over a 0 to 90 percent diversion
range. The results of the control parameters tested were as follows.
The size of the control pockets were varied from 0.875W to 3W,
see Figures 40 and 41 (W = width of nozzles, see Figure 12). As the
pocket size is increased above W the diversion control is reduced until
at 3W there is no aspiration control.
The setback of the regulator sidewalls was varied from 1/2W to
zero, see Figure 35. As the setback approached zero, control rapidly
deteriorated until at zero setback there was no control. As setback
increased above 1/4W total diversion decreased slightly and control
became more digital.
The difference in linearity and control range of rounded vs pointed
splitters was not significant; however, splitter location downstream of
the nozzle did cause considerable effect on analog performance and
linearity. Moving the splitter closer to the nozzle as shown in Figure 36
54
-------
Figure 38. Model Test Configuration Showing Combined and
Interceptor Sewers and Element Insert
Figure 39. Model Test Configuration Showing Combined and
Interceptor Sewers and Element Insert
55
-------
Figure 40. Basic Irrigation Test Model with 2" x 1" Nozzle
Figure 41. Large Control Pockets on Irrigation Geometry
Nozzle 2" x 1"
56
-------
Increases the total diversion range and improves linearity. If the splitter
is rounded as shown in Figure 46, performance is not degraded. A very
blunt splitter as shown in Figure 37 does degrade performance.
If the sidewall is cut away at the control cavity as shown in Figure
42, the water stream cannot be diverted significantly in either a digital
or analog fashion, regardless of other variations in geometry. If a short
sidewall is left downstream of the controls and then cutback as shown
in Figures 43 and 45 analog performance can be obtained with good total
diversion characteristics. As the length of the sidewall is made longer
than W, operation becomes more digital. As the length of the sidewall
is made shorter than W the diversion range is shortened. Rounded side-
walls as shown in Figure 44 did not improve performance. The setback
downstream of the sidewalls was also found to effect performance. A
setback of 1/8W produced better total diversion than did larger setbacks.
See Figures 46 and 47.
During the tests of geometrical variations on analog diversion
performance several sensors were tested to obtain best analog control.
A mechanical float valve was found to be best and was used for testing
the other parameters discussed above. A discussion of sensor tests
and designs follows.
A long neck nozzle was tried, see Figure 50; however, this was
found to reduce the venturi effect of the nozzle and consequently de-
graded diversion performance. Nozzle variations also showed that
performance was not affected by minor changes in the nozzle centerline
relative to the element splitter.
The best 2" x 1" nozzle configuration as shown in Figure 47 was
scaled to form 2" x 4" and 4" x 4" nozzle size regulator elements to
test geometry at different aspect ratios. Basic variations were made
in control pockets, see Figures 48 and 49, and splitters, see Figures
51 and 52. The regulator performance varied as a result of these changes
in the same manner as the 2" x 1" nozzle configuration. It was there-
fore assumed that an acceptable final geometry had been obtained. See
Figure 12.
Performance Tests - Final Configuration
After the final geometry was obtained as a result of the extensive
testing of numerous configurations as described above, the final con-
figuration was then tested in 2" x 1", 2" x 4", and 4" x 4" nozzle
regulator elements. The purpose of this testing was to obtain performance
design curves which could be used to design a sewer regulator for any
particular installation when given the requirements of the installation.
57
-------
Figure 42. Cutaway Sidewalls on Irrigation Geometry
Nozzle 2" x 1"
Figure 43. Short Sidewalls with Cutaway
Nozzle 2" x 1"
58
-------
Figure 44. Rounded Sidewalls with Cutaway
Nozzle 2" x 1"
Figure 45. Short Sidewalls with Splitter Upstream
Nozzle 2" x 1"
59
-------
Figure 46. Short Sidewalls Rounded Splitter
Nozzle 2" x 1"
Figure 47. Short Sidewalls Rounded Splitter Less
Downstream Setback Nozzle 2" x 1"
60
-------
Figure 48. Analog Sewer Regulator Geometry
Nozzle 2" x 4"
Figure 49. Analog Geometry with Larger Control Pockets
and Upstream Splitter Nozzle 2" x 4"
61
-------
Figure 50. Analog Geometry with Long Nozzle, Pointed Splitter
Nozzle 4" x 4"
Figure 51. Final Analog Sewer Regulator Geometry
Nozzle 2" x 4"
62
-------
Figure 52. Final Geometry with Pointed Splitter
Nozzle 2" x 4"
63
-------
Diversion Performance vs Supply Head. The difference aspect
ratio nozzles were tested at numerous supply heads from 6" to 18" and
the maximum and minimum flows were measured to obtain the diversion
range. All regulators tested were adjusted such that with the bias port
completely blocked as much flow as possible was directed to the inter-
ceptor. This was accomplished by adjusting the interceptor slot opening
and combined exit weir height. Design curves derived from this data
are shown in Figure 53. See Appendix C for actual data. These curves
show that regulators with nozzles of higher aspect ratios require less
head to obtain the same diversion. It is believed that the use of these
curves will result in conservative estimates of maximum diversion per-
formance, particularly for large sewers with very low aspect ratio nozzles.
This belief stems from the fact that data for the 0.25 and 0.5 aspect
ratio curves were taken from tests on elements with 1 and 2 inch nozzle
heights respectively, whereas the 1.0 aspect ratio curve was taken
with a 4 inch nozzle element. Figure 30 shows a significant increase
in head in terms of h, to achieve 100% diversion for small values of h.
Note that a relatively large change in required head occurs between
h = 1" and h = 4". This would indicate that the a = 0.25, and a = 0.5
curves of Figure 53 would have lain closer to the a = 1.0 curve had the
tests been made on elements with h = 4" for all aspect ratios. Referring
again to Figure 30, the decrease in required head from 4" up is quite
small, hence the diversion performance predicted from Figure 53 for
nozzle aspect ratios = 1 should be fairly accurate. With the testing
of large, low aspect ratio nozzle regulators, it is expected that the
accuracy of the lower aspect ratio curves of Figure 53 can be appreciably
improved.
Total flow through the regulator was measured for all nozzle sizes
at all heads to obtain head flow data. This was compared to the nomo-
graph obtained from the predictive analysis, see Figure 10 and Appendix B.
Agreement was within ± 5 percent for all nozzle sizes.
Weir Settings; Interceptor Exit Design. Several configurations of
the interceptor weir were investigated to obtain the most simple design
with the best performance, see Figure 54. Design #1 was a conventional
weir with a cut off corner that allowed low flow to pass without restriction.
The discharge area in this design was distributed across the channel and
provided an opening subject to fouling. Design #2 provided a larger
opening by shaping the exit area into a full depth rectangular slot along
the splitter wall where the high velocity stream would have a low im-
pedance path to the interceptor. This design exhibited basic functional
response problems in switching the flow to the interceptor by forming an
air pocket, see Figure 54. Proper performance was obtained with this
weir design by moving the opening next to the sidewall side of the element
as shown by Design #3 in Figure 54.
64
-------
en
O
CM
ffi
CO
w
ffi
u
CO
U
I— I
w
ffi
N
N
O
6h "
s
I—I
9 2h
w
ffi
o.
p
co
a = aspect ratio =
w
n
20
30 40 50 60 70
PERCENT DIVERSION FROM INTERCEPTOR
80
90
100
Figure 53. Maximum Diversion vs Supply Head for Fluidic Sewer Regulators
-------
AIR
POCKET
ADJUSTMENT
ADJUSTMENT
WEIR
SPLITTER
WEIR
ADJUSTMENT
^
"\>J" IV*^
SPLITTER
#2
WEIR
#3
Figure 54. Interceptor Slot Configurations
66
-------
In order to obtain design data concerning the proper interceptor
exit area settings, the combined weir was set at various fixed levels,
and maximum-minimum diversion data were taken for a complete range
of interceptor exit area settings and inlet heads. These data are shown
in Figures 55 and 56 in dimensionless parameters based on nozzle size
and area. These characteristics proved independent of supply heads.
Using Figure 56 we see that the optimum height of the combined weir
is 1.4 hn (hn = nozzle height). Using Figure 55 we see that the optimum
area of the interceptor discharge is 1.2 AN (AN = nozzle area). See
Appendix C for actual data curves used to obtain these design curves.)
Sensor Characteristics. The predictive analysis yielded bias
orifice criteria as discussed above. These data were found to hold for
using single dip tube sensors and a bias control orifice to effect digital
type control; however, it was found that the analog operation of the
regulator could best be obtained by a sensor which would operate inde-
pendently on each control and thus obtain maximum linearity. This type
of control also required a sensor with very different area relationships
than the bias orifice type. The development of the analog area ratios
and sensor schemes are presented in the following detailed discussion.
In order to determine the sensor area relationship required for
linear analog control as described above, a float valve was placed in
the interceptor line as shown in Figures 57 and 58 to provide control.
The schematic of the float valve test setup, Figure 59, shows how
regulator control was effected by interceptor water level. When the
water level is at the desired minimum point, float valve area #2 is
closed by the rubber flapper and all the flow through the regulator is
directed to the interceptor. When the interceptor is filled to the desired
maximum level, area #1 is closed and all flow is diverted away from
the interceptor to the combined discharge.
Three sensor area configurations tested in this setup are shown
in Figure 60 and performance data from these sensors are shown in
Figure 61. Sensor 1 consisted of small areas on the order of the bias
orifice areas recommended by the predictive analysis. Performance
with this size area was purely digital; that is, the interceptor water
level had to raise to its maximum height to close off the sensor area
#1 before any change in diversion occurred. At this point complete
switching of the stream occurred with no analog control. Sensor #2
incorporated areas five (5) times as large as the areas in sensor #1.
Performance with sensor #2 was still digital; however, a range of analog
control was observed, as shown in Figure 61. As a result of numerous
tests and observations with sensor #2, it was determined that maximum
sensor area is needed when the areas are equal and the regulator is
67
-------
CD
OO
X
w
Q
0)
O
0)
.I-J
c
(0
0)
0
'N
N
O
(0
0
1.6
1.2 -'
0.8 "
Combined Weir Height
Nozzle Height
0.4 --
-4-
10
20 30 40 50 60 70
PERCENT DIVERSION FROM INTERCEPTOR
80
90
100
Figure 55. Minimum-Maximum Diversion From Interceptor vs Combined and
Interceptor Weir Settings for Fluidic Sewer Regulators
-------
01
ID
2.4 -
2.0 -
N
O
1.2 -
3. Oh
Area of Interceptor Exit
Area of Nozzle
= 1.5
4.Oh
n
5.0hn 6.Oh
n
Supply Head
10
20
30 40 50 60 70
PERCENT DIVERSION FROM INTERCEPTOR
80
90
—I—•
100
Figure 56. Maximum Diversion vs Combined Weir Height
-------
Figure 57. Float Valve Mechanical Sensor
Figure 58. Float Valve Test Installation
70
-------
GATE
fO TO
COMBINED INTERCEPTOR
INTERCEPTOR
Figure 59. Float Valve Sensor Test Setup
71
-------
L
AREA = 0.012 in?
~Z-~ RUBBER FLAP
,
/ TOTAL AREA = 0. 033 in2
SENSOR 1
0.05 in2
\
|^» .55 «J
RUBBER FLAP
/ TOTAL AREA =0.16
SENSOR 2
AREA = 0.16 in2
\
RUBBER FLAP
V
SENSOR 2M
Figure 60. Float Valve Sensor Areas
72
-------
00
100
w
j
K
O
H
Qu
U
U
PH
O
w
O
O
H
W
U
w
1
g 80--
SENSOR 1, A = 0.033 in2
30 40 50 60 70
PERCENT DIVERSION FROM INTERCEPTOR
80
90 100
Figure 61. Diversion vs Interceptor Water Level for Fluidic Sewer Regulator
2" x 4" Nozzle
-------
dividing the flow evenly between the interceptor and combined sewers.
By gradually opening up the areas at the 50-50 operating diversion point
the optimum area designated as sewer #2M was established and per-
formance data recorded, see Figures 60 and 61.
Percent diversion vs change in sensor area was then taken for the
2" x 4" and 4" x 4" nozzle regulators, see Figures 62 and 63. Area
change was recorded as a ratio of sensor area change to regulator nozzle
area, as referenced from the condition where the interceptor control area
is fully closed and all flow is diverted to the interceptor. The ratio
corresponding to maximum diversion is then twice the area of the sensor
opening used for each control. Test results from both 2" x 4" and 4" x 4"
nozzle regulators yielded the same ratio of 40 (10~3) as the proper ratio
for linear analog control.
It follows, then, that the optimum control orifice area for analog
operation is 20 (10~3) AN, (AN = nozzle area). Digital operation is
obtained by using orifice areas equal to one quarter of the analog area.
Tests at numerous supply heads showed diversion performance vs area
change or interceptor level change was linear, over a complete range
of heads, as shown in Figures 64 and 65. (See Appendix C for actual
data curves.)
No-Moving-Part Sensors
Principles of Operation. In the course of this program a new design
for a no-moving-part sensor was investigated which could provide linear
analog control without the need for a mechanical float valve. A schematic
of the arrangement is shown in Figure 66. The system uses a control dip
tube in the interceptor sewer which is connected to the combined control
of the regulator and a sealed bottle. A second dip tube is connected to
the interceptor control and placed in a second bottle which is vented and
has a common fluid connection at its base with the sealed bottle. The
two bottles and connection operate in the manner of a U-tube manometer.
As the interceptor dip tube sensor is covered by rising water level, as-
piration through the combined control is reduced causing an increase in
vacuum. This increase in vacuum acts on the sealed bottle to raise its
fluid level. Since the two bottles have a common connection and the
dip tube bottle is vented, a change in fluid level between the two bottles
results. The change in fluid level of the bottles produces a change in
the exposed area of the vented bottle dip tube, which in turn controls
aspiration to the interceptor control. Therefore, an interceptor level
increase will increase the aspiration to the interceptor control by in-
creasing the opening of the dip tube slot through the bottle and decrease
the aspiration to the combined control directly. The jet stream will be
switched in a push-pull manner.
74
-------
Cn
ro
I
O
40 --
-------
20
30 40 50 60 70
PERCENT DIVERSION FROM INTERCEPTOR
80
90
100
Figure 63. Diversion vs Sensor Area Change 4" x 4" Nozzle Supply Head = 13.0"
-------
-vl
co
w
K
U
K
o
e
w
U
ex
Limit
10.0
8.0"
6.0--
W
I
PLH
O
E-t
sc
o
g Limit
4.0 •-
2.0
2h
n
10
H , , , 1 1
20 30 40 50 60 70
PERCENT DIVERSION FROM INTERCEPTOR
80
90
100
Figure 64. Diversion vs Interceptor Level for Fluidic Sewer Regulators
Aspect Ratio =0.5 Nozzle 2" x 4"
-------
oo
I
o
n
40
00
to
0)
o
w
G
O
CO
c
(0
.c
u
CD
13
g30
20 ••
0)
75
N
S, 10
(D
20
30 40 50 60 70
PERCENT DIVERSION FROM INTERCEPTOR
80
90
100
Figure 65. Diversion vs Sensor Area for Fluidic Sewer Regulators
Aspect Ratio = 0.5 Nozzle 2" x 4"
-------
INTERCEPTOR
CONTROL
COMBINED
CONTROL
TO
INTERCEPTOR
SEWER
TO
COMBINED
SEWER
DISCHARGE
VENTED BOTTLE
DIP TUBE
1
INTERCEPTOR
DIP TUBE
SENSOR
COMMON CONNECTION
INTERCEPTOR
Figure 66. No-Moving-Part Sensor Test Setup
79
-------
Push-Pull Sensor Tests. Several tests were made to investigate
the no-moving-part bottle sensor with different dip tubes and bottles,
see Figures 67 and 68. It was found that equal size bottles with equally
distributed dip tube areas did not provide enough gain and control was
very non-linear. A sealed bottle of twice the area of the vented bottle
was tried to increase sensor gain; however, this was only partially
successful. Control with the push-pull bottles as a function of inter-
ceptor water level did not affect diversion until the dip tube sensor was
over 50% closed; however, after diversion control started a reasonably
linear control range was observed, see Figure 69. In order to expand
the control range the interceptor dip tube was shaped similar to Sensor
#1 shown in Figure 60; however, no noticeable improvement in per-
formance was observed.
After testing of model sewer regulators was completed and aspirated
air flow data analyzed it was found, as shown in Figure 70, that the
aspirated air flow is high at the midpoint in diversion (see Appendix C
for actual data curve) and drops sharply as diversion approaches its
maximum value. Thus, only a low aspiration capability is available
to produce the initial change in the fluid levels of the push-pull sensor
bottle to obtain linear analog control. From a controls system stand-
point, this characteristic represents a region of very low gain. There-
fore, to provide linearity, this low gain region must be compensated by
a high gain in some other part of the system. The two-bottle, or U-tube
concept offers several approaches to provide such a localized, high
gain characteristic.
1. The dip tube orifices can be shaped so that a small change
in liquid level effects a large change in total orifice area. This
requires a non-linear tapered slot, which widens abruptly near
the bottom of its range.
2. The relative bottle diameters can be shaped. The increase
in vacuum on the combined port as the interceptor sensor is covered
causes a corresponding differential change in the liquid level
between the U-tube branches. Since only a fixed amount of liquid
is involved, the relative fluid level rise in the closed branch, as
compared to the level drop in the vented branch, is proportional
to the inverse ratio of the branch cross section areas. Thus if
the vented branch is necked down opposite the bottom end of the
dip tube slot, a relatively large change is liquid level will result
from a small increase in vacuum.
It was not possible to fully explore the above possibilities within
the scope of this initial program phase; however, it appears certain that
good diversion linearity can be achieved through one or both, so that proper
sewer regulation without moving mechanical parts is achieved.
80
-------
Figure 67. No-Moving-Part Sensor Push-Pull Bottles
Figure 68. No-Moving-Part Bottles Installation
81
-------
o
O>
(0
K
O
U
en
CO
C
•rH
O
20 40 60 80
PERCENT DIVERSION FROM INTERCEPTOR
100
Figure 70. Air Water Flow Ratio vs Diversion
82
-------
Multiple Sensors
In an effort to determine whether two fluid levels could be monitored
with dip tube sensors and added to provide control from a single control
port, the sensor combination shown in Figure 71 was tested. In this setup
one dip tube sensor was put upstream of the regulator nozzle in the com-
bined sewer and a second sensor was placed in the interceptor sewer.
Both sensors were then connected to the combined control to provide
diversion control. With the knowledge of the optimum area for analog
control as discussed above the total area of both sensors was selected
having an area ratio of about 3:1. The diversion performance of the
multiple sensors was measured as a function of supply head. Curve 1
of Figure 72 shows the change in diversion vs head over the full range
of the combined sensor with the interceptor level constant. Curve 2 of
Figure 72 shows the change in diversion resulting from change in head
without the combined sewer sensor connected to the control. The sig-
nificant difference between these curves shows that both sensors are
contributing to the diversion. Figure 73 consists of a similar pair of
curves taken at a different interceptor water level. Results of these
tests showed that the knowledge of control areas obtained from regulator
tests could be applied to multiple dip tubes as well as single ones.
Flow-Over Discharge
Diversion performance of the sewer regulator was tested for the
simulated condition of large upstream heads, caused by storm flows,
causing flow over the regulator and thus over the discharge. When the
head of water flowing over the discharge exceeds about fifteen (15)
percent of the upstream head the fluidic regulator flow is biased to the
interceptor and diversion performance is seriously reduced. In an effort
to avoid this problem, shrouded discharge tests were made to isolate
the two flows and provide an air-water mixing region, see Figures 74
and 75. For large flows over the regulator a short shroud will act like
a leaping weir; however, at low flows a short shroud produced interference,
see Figure 74, and effected diversion noticeably. A longer shroud as
shown in Figure 75 solved this problem.
From these tests it is concluded that for installations where the
outfall communicates to tidal waters or back water conditions exist to
produce a head at the outfall the fluidic regulator performance will show
significant degradation as the discharge head builds to about fifteen (15)
percent of the supply head regardless of shrouding.
83
-------
COMBINED SEWER
X DIP TUBE
1
COMBINED
SEWER
% Closed
"OPEN"
INTERCEPTOR
CONTROL
COMBINED
CONTROL
TO
INTERCEPTOR
SEWER
TO
COMBINED
SEWER
DISCHARGE
INTERCEPTOR
DIP TUBE
INTERCEPTOR SEWER
% Closed
Figure 71. Multiple Sensor Test Setup
84
-------
100
CD
cn
CURVE 1
COMBINED & INTERCEPTOR
SENSORS w/ INTERCEPTOR
SENSOR SET @ 83% CLOSED
CURVE 2
INTERCEPTOR SENSOR
ONLY,SET @ 85% CLOSED
70
9.0
PERCENT DIVERSION FROM INTERCEPTOR
Figure 72. Diversion vs Supply Head for Multiple Sensor Control
of Fluidic Sewer Regulator
-------
oo
01
CURVE 1
COMBINED & INTERCEPTOR SENSORS
w/ INTERCEPTOR SENSOR 100% CLOSED
CURVE 2
INTERCEPTOR SENSOR
ONLY, 100% CLOSED
••12.0
-• 10.0
9.0
60 65 70 75
PERCENT DIVERSION FROM INTERCEPTOR
O
CM
CO
B
-•ii.o 3
Figure 73. Diversion vs Supply Head for Multiple Sensor Control
of Fluidic Sewer Regulator
-------
Figure 74. Shrouded Discharge Low Velocity Flow Interference
Figure 75. Shrouded Discharge Low Velocity Longer Shroud
No Interference
87
-------
Fouling Analysis
In order to determine a Fluidic regulator's susceptibility to fouling,
tests were conducted with small branches of various sizes and shapes.
Maximum dimensions of branches ranged from 1.1 to 4 times the nozzle
height (hn), as shown in Figure 77. The channel upstream of the nozzle
was 6 hn square so branches could approach in any orientation. The
regulator used for testing is shown in Figure 76. Ten runs were made
with each of eight branches giving a total of 80 runs. Six cases of
lodged branches occurred, three on the converging section of the nozzle
and three on the splitter. The splitter in this case was relatively sharp
making it more susceptible to fouling by the forked sticks than other
regulator splitter geometries having the same functional characteristics;
see Appendix D for fouling data. In addition to tests with branches,
tests were run with pieces of paper to give some insight into the fouling
caused by large soft objects such as newspapers. No problems were
experienced except with a very large, 5 hn x 8 hn piece which caught
2 of 6 times on the pointed splitter. Later tests showed that regulators
with blunted splitters could be designed to perform as well as pointed
splitter types, therefore actual hardware utilizing blunted splitter designs
will be less subject to fouling.
Fouling tests showed that the Fluidic sewer regulator without an
upstream trap of any sort was relatively free of fouling. On the occasions
when fouling occurred, the fouling objects had a linear dimension of four
(4) or more times the nozzle height. A sharp splitter is more susceptible
to fouling than a blunt one, and that sharp corners near the regulator
nozzle should be well beveled, or rounded.
It is anticipated that field tests to be conducted as part of the
demonstration grant will resolve the fouling problems associated with
fluidic sewer regulators under actual conditions.
88
-------
Figure 76. Basic Irrigation Test Model
2" x 3" Nozzle
Figure 77. Simulated Debris, Fouling Test
89
-------
SECTION 6
DEMONSTRATION PLANNING AND LIAISON
General
One of the specific objectives of the subject program was to assist
the FWPCA in planning and establishing the succeeding phases of the
program; namely, the design, construction, operation, and evaluation
of the Fluidic Sewer Regulator concept in a typical metropolitan diversion
location. It is the FWPCA1 s intent that the work should be done by a
large municipality, under a FWPCA Demonstration Grant. Because of
the preponderance of combined sewers in the eastern USA and the relative
convenience in performing engineering liaison from both EEC and the
FWPCA, it was decided that candidate cities should be New York, Chicago,
Washington, D. C. , Philadelphia, and Baltimore.
Accordingly, a series of visits were arranged with the various
municipal sewer engineering chiefs as follows:
New York City, N.Y.
Chicago, Illinois
Baltimore, Maryland
Philadelphia, Pa.
Washington, D. C,
Official & Title
Mr. Joseph Cunetta
Deputy Director, Plants
Bureau of Water Pollution
Control
City of New York
Mr. Clint Keifer
Chief, Water & Sewer
Design Engr'g, Dept. of
Public Works
City of Chicago
Mr. John Prussing
Principal Engineer
Sewer Construction
Mr. Carmen Guarino
Deputy Water Commissioner
City of Philadelphia
Mr. James Robertson
Associate Director
Engr'g & Construction
Dept. Sanitary Engr'g
District of Columbia
July 14, 1969
July 25, 1969
July 28, 1969
August 5, 1969
90
-------
Questionnaire
A brief questionnaire was presented at each meeting to assist in
defining each city's capability and interest in seeking and performing
the subject Fluidic regulator demonstration uder a FWPCA Demonstration
Grant. The questionnaire contained the following:
1. Are relatively small diversion points available within the
municipal area for installing a small pilot model Fluidic sewer
regulator?
2. Is municipal funding available in keeping with the FWPCA1 s
75% - 25% funding policy?
3. Are there difficulties, or unusual conditions required to
secure the commitment of municipal funds to support 25% of the
program cost?
4. Would serious scheduling delays exist in the installation
of the demonstration program due to prior commitment, or un-
availability of engineering or construction manpower, or facilities?
5. Would the municipality consider the use of A&E contract
assistance in conducting the demonstration program?
6. Would the municipality encounter procurement difficulties
in contracting for sole source goods or services in view of BEC's
proprietary position in the Fluidic field in general, and in water
management devices in particular?
7. Has the municipality had prior experience in conducting
programs under FWPCA Research, or Demonstration Grants?
8. Would officials of the municipality be interested in visiting
BEC's laboratory and watching a demonstration of the experimental
model?
Visit Discussion
The results of each visit are summarized:
New York City. Mr. Cunetta and his staff showed strong technical
interest in the program concept, and test results. Questions were raised
on the effect of regulator diversion performance away from interceptor
when tide gates are backloaded with high tides; Answer - regulator would
work fairly well until tide level exceeded 25% of inlet head level. In
response to questionnaire items:
91
-------
1. Location for small regulators probably available, but some
study needed.
2. Funds could be made available, with approval of Board of
Estimates.
3. No specific approval difficulties - this type of program has
been approved in the past.
4. A starting delay from 18-24 months would be necessary due
to engineering work backlog.
5. NYC would consider A&E assistance; they suggest that BEG
should contract this effort directly, rather than NYC.
6. No specific problem procuring sole source, proprietary
equipment or services if good reason is present.
7. NYC is currently evaluating the "Ponsar" regulator concept
under FWPCA Demonstration Grant.
Summary — New York City has the technical interest, and financial
capability, but engineering capability would not be available for 1-2 years
- Good possibility for a Demonstration Grant program later, poor possi-
bility in near future.
Chicago, Illinois. Mr. Keifer showed moderate technical interest
in concept, but indicated that Chicago's system of sewers did not include
many small diversion points that needed regulation (most regulation is
done with very large electric motor driven gate structures), and foresaw
difficulty in getting approval to install a regulator in a currently non-
regulated diversion point. In response to questionnaire items:
1. Small diversion points available, but currently do not need
regulation (see above).
2. Present bond improvement funds very low - more bonds require
voter approval before a new program could be committed.
3. No particular difficulty in funding programs of this magnitude
if funds are available.
4. No particular difficulty in getting approval, if funds are
available.
92
-------
5. Chicago would probably not use A&E assistance for basic
hydraulic design - only for detailed structures once specifications
are established.
6. No difficulty in sole source, proprietary procurement of
equipment, or services, if good reason is present.
7. Chicago currently has several FWPCA Demonstration Grant
programs in progress.
Summary -- Chicago has an excellent technical capability for
the subject program, but currently no firm requirements for this device;
is very limited in available funds - poor immediate possibility for the
subject Demonstration Grant program.
Baltimore. Maryland. Mr. Prussing showed technical interest
but indicated that Baltimore had virtually no combined sewers. He
indicated that requirements for storm water diversion existed, if the
device could be adapted to this requirement (Answer - it could). In
response to the questionnaire:
1. Except for a few storm water diversion points, very few
diversion points are available.
2. Municipal funding is available.
3. No particular difficulties in obtaining approval of Board
of Estimates.
4. Baltimore has a significant engineering work backlog.
Consequently, considerable delay in starting would be anticipated.
5. A&E contract assistance would be a strong probability.
6. No difficulty in procurement of proprietary, sole source
equipment, or services if justification shown.
7. Baltimore currently has a FWPCA Demonstration Grant program
in progress at the Back River Sewage Treatment Facility.
Summary -- Baltimore has minimum requirements for subject device,
except possibly for storm water handling. Despite general technical
interest, this city has a considerable engineering work backlog, and thus
could not consider the subject demonstration program for a considerable
period of time - poor possibility for the near future.
93
-------
Philadelphia, Pennsylvania. Mr. Guarino showed strong interest
both in the technical progress made to date, and, the possibility of
evaluating the concept. Philadelphia has at present approximately 70
small mechanical float-operated regulators, in addition to a number of
large hydraulically operated units. The small units require at least
twice-a-week surveillance and close monitoring during storm conditions
to keep them operating. With this background, Mr. Guarino indicated
he was quite interested in evaluating approaches which might require
less monitoring. With respect to the questionnaire items:
1. Philadelphia has a considerable number of mechanical
regulators, also many "slot" diverters, which are basically
equivalent to the standard diversion dam and interceptor side
flow line arrangement.
2. Municipal funding is readily available for experimental
purposes upon the approval of the Water Commissioner.
3. Funding approval can be made directly by the Water
Commissioner.
4. Philadelphia has an experienced engineering staff who
would be available on short notice to undertake the subject
program.
5. A&E support would be considered, but is probably not
specifically required.
6. There would be no procurement difficulty for sole source,
or proprietary equipment or services, if sufficient justification
is present.
7. Philadelphia is currently active on a FWPCA Demonstration
Grant concerning microstrainers for combined sewer overflows.
Summary — Philadelphia appears to have strong interest, and
excellent capability for performing the subject Demonstration Program
- A top possibility.
Washington, D. C. Mr. Moorehead, acting for Mr. Robertson,
indicated a strong interest in the Fluidic concept, having accepted an
invitation to view the experimental model at EEC's laboratory in the
D.C. area where a pilot Fluidic regulator could be demonstrated. How-
ever, he cited local government policy which forbade any new construction,
or improvements to be made on existing combined sewers, since the city
94
-------
was embarked on a long term program to replace these with separate
sewers. Consequently, although the city would be willing to offer a
location for a demonstration model, it would not agree to conduct the
actual demonstration program under a FWPCA grant.
Summary — Despite a strong technical interest in the subject
concept, and a very cooperative attitude in offering to provide facilities
and other assistance to such a demonstration program, Washington,
D. C. could not be considered for the subject program.
Second Round Meeting
Following the initial series of meetings, Mr. Guarino and his
staff visited EEC on September 3 and viewed the experimental laboratory
model. Mr. Guarino confirmed the City of Philadelphia's desire to
participate in the subject Demonstration Grant program. At the time
of writing of this report, the Demonstration Grant Request is in prepa-
ration by his staff.
Demonstration Program Schedule
A preliminary Demonstration Program Schedule spanning 24 months
is shown in Figure 78. This plan would establish a 4 month design
phase, a 4 month construction phase, a 2 month checkout phase, and
a 12 month operational evaluation phase, and a 2 month reporting phase.
This plan will be finalized when the Demonstration Grant Request is
submitted to the FWPCA.
95
-------
1970
1971
(£>
cn
DESIGN
CONSTRUCTION
CHECKOUT
EVALUATION
REPORT
DRAFT
FINAL
I FMAMT T ASOND
~3
[
1
d
Hi
1
J FMAMT I ASOND
I
=•
Figure 78. Preliminary Demonstration Plan
-------
SECTION 7
ACKNOWLEDGEMENTS
Acknowledgement is made to the following chiefs of Sewer Engineering,
and their respective staffs, for their interest and the time taken in the con-
ducting of the interviews necessary for the Demonstration and Planning phase
of the subject program, as described in Section 6.
Mr. Joseph Cunetta
Deputy Director, Plants
Bureau of Water Pollution Control
City of New York
Mr. Clint Keifer
Chief, Water & Sewer
Design Engr'g Dept. of Public Works
City of Chicago
Mr. John Prussing
Principal Engineer
Sewer Construction
Baltimore, Maryland
Mr. Carmen Guarino
Deputy Water Commissioner
City of Philadelphia
Mr. James Robertson
Associate Director
Engr'g & Construction
Dept. of Sanitary Engr'g
District of Columbia
Acknowledgement is made to Mr. George A. Moorehead, Chief of
Systems and Planning, Department of Sanitary Engineering, District of
Columbia, Washington, D. C. , and to Mr. Carmen Guarino, Deputy
Water Commissioner, City of Philadelphia, for furnishing typical diversion
structure drawings, and the generation of preliminary diversion structure
cost estimates, as described previously in Section 3.
Acknowledgement is made to Mr. William Rosenkranz, Chief of the
Storm and Combined Sewer Pollution Control Branch, FWPCA; and to Mr.
Darwin Wright, Program Project Officer, of the same organization. These
97
-------
individuals have provided valuable guidance and background information
in the course of conducting the subject program.
Acknowledgement is made to members of the Bowles Engineering
Corporation staff as follows:
P. A. Freeman Program Director
R. Bean Research Engineer
J. Zaloudek Research Engineer
P. Cain Test Technician
P. Senes Test Technician
A. Freiling Model Shop
98
-------
SECTION 8
REFERENCES AND PUBLICATIONS
1. Report: "Problems of Combined Sewer Facilities and Over-
flows - 1967," prepared by the APWA under contract 14-12-65,
sponsored by the FWPCA.
2. Paper: "Performance and Operating Characteristics of a
Fluidic Irrigation Diverter," prepared byDrs. Howard R. Haise
and E. Gordon Kruse, of the ARS, SWC, USDA, Ft. Collins,
Colorado.
3. Article: "Air Pressure Drop Nomograph," F. Kaplan, Kaiser
Engineers, appearing in the April 1967 issue of Controls Engineering^
Magazine.
99
-------
SECTION 9
GLOSSARY OF TERMS AND ABBREVIATIONS
Fluidic Element Nomenclature
Nozzle - opening where fluid (water in this case) enters the fluidic
element having a geometry similar to a venturi throat.
Control Ports - openings immediately downstream from the nozzle,
where fluid (gas or liquid) is admitted to influence the direction of nozzle
flow.
Interaction Area - the volume of the element immediately down-
stream of the control ports and upstream of the splitter, containing the
attachment walls.
Splitter - a wall which divides the fluidic element exit area into
two (or sometimes more) sections.
Attachment Walls - wall of element immediately downstream of
the control pockets to which the water jet attaches by the coanda effect.
Coanda Effect - wall attachment phenomenon of a jet stream close
to a wall which provides a pressure differential to act on the jet stream
as a result of entrainment at the wall, producing a low pressure area.
Venturi - a constriction in a flow channel which produces increased
velocity and decreased pressure or suction at the constriction.
Aspiration - drawing of fluid (air in this case) into diverter inter-
action region by suction caused by the venturi action of nozzle.
Setback - offset of one portion of geometry to another measured
away from centerline of nozzle.
Aspect Ratio - ratio of height of diverter nozzle to its width.
Digital Switching - having only two functional states, either maximum
or minimum diversion.
Analog Switching - having a continuous range of diversion values
between maximum and minimum levels.
100
-------
Abbreviations
PVC - Polyvinylchloride
ABS - Acrylonitrile - Butadiene - Styrene
PE - Polyethylene
hn - height of regulator nozzle
hi - height of inlet head
hd - height of discharge
wn - width of regulator nozzle
a - aspect ratio
A - area of nozzle
101
-------
SECTION 10
APPENDICES
Page
A. Log of Sewer Regulator Element Tests 104
B. Orifice Equation 110
C. Performance Design Data 116
D. Fouling Analysis 132
E. Control Line Sizes 134
102
-------
APPENDIX A
Log of Sewer Regulator Element Tests
The test program was begun with tests of the 1/2" x 1/2" nozzle
irrigation diverter as part of the task 1 predictive analysis listed on the
first page. Testing was then continued on model sewer regulator vari-
ations starting from the basic irrigation geometry, and culminating in the
final analog regulator configuration. These tests are listed from June 20
to July 25. Several final configurations were then tested to investigate
minor changes in geometry and obtain performance curves and design
data (see tests from August 15 to September 4). Tests listed in the log
were for regulator elements for which test data were recorded and a
definite variable was tested. In many cases numerous testing was done
by visual observation with many variables and formal data were not
recorded.
103
-------
Log of Sewer Regulator Element Test&
Test
Date
4-4
4-7
4-9
4-17
4-17
4-17
4-18
4-18
4-18
4-24
4-24
4-28
4-28
Test
No.
1
1
1
1
2
3
1
2
3
1
2
1
2
Config.
No.
Irrigation
Element
Irrigation
Element
Irrigation
Element
Irrigation
Element
Irrigation
Element
Irrigation
Element
Irrigation
Element
Irrigation
Element
Irrigation
Element
Irrigation
Element
Irrigation
Element
Irrigation
Element
Irrigation
Element
Nozzle
Size
1/2x1/2
1/2x1/2
1/2x1/2
1/2x1/2
1/2x1/2
1/2x1/2
1/2x1/2
1/2x1/2
1/2x1/2
1/2x1/2
1/2x1/2
1/2x1/2
1/2x1/2
Sensor
Bias
Orifices
Bias
Orifices
Bias
Orifices
Bias
Orifices
Bias
Orifices
Bias
Orifices
Bias
Orifices
Bias
Orifices
Bias
Orifices
Bias
Orifices
Bias
Orifices
Bias
Orifices
Bias
Orifices
Parameter
Tested
Supply
Flow vs Head
Control Flow
vs Supply Flow
Control Bias.es
Head = 3.0"
Control Biases
Head =5.0"
Control Biases
Head = 7.5"
Control Biases
Head = 3.0"
Zero set back
Head = 4.0"
Head = 7.5"
Control
Airflows
Control
Pressures
Setback
Control
Airflow
104
Results of Tests
Useful productive data
Useful productive data
Some analog control
possible
Digital operation
More analog range
Same as 5.0" head
No control
No control
No control
Airflow vs % div.
obtained
Pressure diff. vs.
Diversion obtained
Curve obtained low
scatter of data
Airflow vs % div. showed
wide scatter
Diversion
Range
0-100
0-100
50-100
0-70
0-95
0-90
95-98
95-100
95-100
0-95
0-95
0-100
0-95
-------
Log of Sewer Regulator Element Tests (Continued)
Test
Date
5-14
5-14
5-14
5-15
5-16
5-16
5-23
5-26
5-29
6-16
6-16
6-17
6-17
6-18
6-18
6-18
6-18
6-19
6-19
6-19
Test
No.
1
2
3
1
1
2
1
1
1
1
2
1
2
1
2
3
4
1
2
3
Config.
No.
513-1
513-1
514-1
515-1
516-1
516-2
516-2
516-2
527-1
616-1
616-2
617-1
617-2
618-1
618-2
618-3
618-4
618-4
618-4
618-4
Mozzle
Size
2x3
2x3
2x3
2x3
2x2-1/2
2x3-1/4
2x3-1/4
2x3-1/4
2x3-1/4
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
Sensor
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Bias
Orifice
Float
Valve
Bottles
Bias
Orifice
Parameter
Tested
Single control
geometrical
bias
Weir setting
Cusp on split-
ter
Cusp - smaller
irrigation
Geometry
Wider nozzle
pointed splitter
Weir setting
Lower head
Fouling tests
Higher aspect
ratio
Smaller control
pockets
Vortex pockets
1-1/2"
Vortex pockets
2"
Increased set-
back decreased
wall angle
Splitter posi-
tion 1 " down-
stream
Vortex pockets
3"
New profile
cutaway side-
walls
Sensor
Sensor
Sensor
105
Results of Tests
Poor control range
Poor control range
Improved diversion
range
Reduced operating
range
Strickly bistable
operation
Improved diversion
range
Good range - still
bistable
Reduced range
Low percent fouling
Comparable to previous
tests
Strickly bistable
Some intermediate
values
Poor control
Reduced operating
range
Same as 618-1
No control
Nearly bistable
Small analog range
Sames as Test 1 above
Reduced operating
range
Diversion
Range
25-45
16-36
19-73
27-39
8-100
20-79
15-94
23-88
18-87
18-80
18-79
0-100
30-78
26-64
27-68
50-50
15-74
30-91
30-100
30-68
-------
Log of Sewer Regulator Element Tests (Continued)
Test
Date
6-20
6-20
6-20
6-20
6-20
6-23
6-23
6-23
6-24
6-24
6-24
6-24
6-25
6-25
6-26
6-26
6-26
6-26
6-27
7-1
7-1
Test
No.
1
2
3
4
5
2
3
4
1
2
3
4
1
2
1
2
4
5
1
2
3
Config .
No.
620-1
620-2
620-3
620-4
620-5
623-1
623-3
623-4
624-1
624-1
624-1
624-1
625-1
625-2
626-1
626-2
626-4
626-5
627-1
71-1
71-1
Nozzle
Size
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x1
2x4
2x4
Sensor
Bias
Orifice
Bias
Orifice
Float
Valve
Float
Valve
Bottles
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Float
Valve
Parameter
Tested
Setback = 0
Splitted moved
upstream
Setback
increased
Short sidewalk
added
Short sidewalls
Wall length
1,2"
Wall length
3.2"
Splitter moved
2" downstream
Curved side-
walls 1.2"
Control orifice
shade
Larger area
triangle
One half size
triangle
Pointed splitter
upstream
Blunt splitter
Splitter moved
downstream
Sharp splitter
Sidewall cut-
back shortened
1/2"
Rounded
splitter
Sidewall cut-
back shortened
1/8"
Nozzle
Combined
weir raised
106
Results of Tests
Not enough setback to
deflect jet
Same as Test 1 above
Same as Test 1 above
Improved analog control
range
Pour range - surging
Poor range
Improved range
Reduced operating
range
No significant improve-
ment
Slightly improved
linearity
Most linear performance
Much more digital
performance
Improved range
Reduced range
Improved range
Suring - splitter very
close
Not surging
Reduced range
Improved range
Did not match 2x1
performance
Improved range
Diversion
Range
50-50
50-50
50-50
0-77
0-81
11-52
19-100
8-71
11-83
15-86
12-83
13-75
11-79
10-45
6-73
7-100
11-91
19-81
0-86
4-68
7-81
-------
Log of Sewer Regulator Element Tests (Continued)
Test
Date
7-8
7-8
7-8
7-9
7-9
7-9
7-9
7-10
7-10
7-10
7-10
7-11
7-11
7-11
7-11
7-11
7-14
7-15
7-15
7-15
7-15
Test
No.
1
2
3
1
2
4
5
1
2
3
4
1
2
3
4
5
1
1
2
3
4
Config.
No.
74-1
74-1
74-1
78-1
78-1
78-1
78-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
Nozzle
Size
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
Sensor
Float
Valve 1
Float
Valve 1
Float
Valve 1
Float
Valve 1
Float
Valve 1
Float
Valve 1
Float
Valve 1
Float
Valve 1
Float
Valve 1
Float
Valve 1
Float
Valve 1
Float
Valve 2
Float
Valve 2
Float
Valve 2
Float
Valve 2
Float
Valve 3
Float
Valve 2M
Float
Valve 2M
Float
Valve 2M
Float
Valve 2M
Float
Valve 2M
'arameter
Tested
ensor low
lead
nterceptor
weir lower
ncreased
head 10"
nterceptor
weir higher
ncreased
head 13"
Lower head
8.5"
Low head
60"
Head 8.5"
Head 10.0"
Head 12.0
Head 6.0
Sensor head
10.0"
Lower com-
bined weir
ncreased
head 13.0
Head 7.5"
Sensor head
10.0"
Sensor head
8.5"
Head 11.0"
Head 14.0"
Raised com-
bined weir
Reduced head
11.0"
107
Results of Tests
Poor control
Slight increase in range
- digital
Improved range -
digital
Improved range -
digital
Slightly improved -
digital
Reduced range - digital
Further Reduced range
Still digital
Same as 8.5" head
Same as 8.5" head
Greatly reduced range
Increased analog range
Reduced range
Sames as 8.5" head
Reduced range
Analog performance
Linear analog
performance
Increased range
Reduced range
Range restored
Reduced range
Diversion
Range
35-54
3-40
4-61
5-70
1-69
3-55
2-47
1-78
1-75
1-78
1-38
2-76
5-58
2-79
0-60
0-74
3-69
2-81
3-47
3-81
1-69
-------
Log of Sewer Regulator Element Tests (Continued)
Test
Date
•••j^- i —
7-16
7-16
7-16
7-16
7-16
7-18
7-18
7-18
7-18
7-23
7-23
7-23
7-24
7-24
7-24
7-24
7-25
7-25
8-15
8-15
Test.
No.
1
2
3
5
6
1
i
2
3
4
1
2
3
1
2
3
4
1
2
1
2
3onfig. ]
No.
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
710-1
813-1
813-1
Nozzle
Size
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
2x4
1x4
1x4
1x4
1x4
1x4
1x4
4x4
4x4
1
Sensor
Float ]
Valve 2M
Float
Valve 2M
Float
Valve 2M
Float
Valve 2M
Float
Valve 2M
Multiple
Dip tubes
Single
Dip tube
Multiple
Dip Tubes
Multiple
Dip Tubes
Plugged
controls
Plugged
controls
Plugged
controls
Float
Valve 2M
Float
Valve 2M
Float
Valve 2M
Float
Valve 2M
Float
Valve 21V
Half
area 2M
Float
Valve 4
Float
Valve 4
Parameter
Tested 1
Bead 10.0
Head 13.0
Head 6.5
Head 6.0
Head 7.0
Sensors
Sensor
Sensor areas
Head
Combined
weir
Higher heads
Higher heads
Head
Head 12.0"
Head 8.0"
Head 10.0"
Diversion
Sensor
Weir settings
High com-
bined weir
108
1
Results of Tests
__
Linear
Increased range
Reduced range
Greatly reduced range
Increased range
Poor control range
Wide linear control
Good range - linear
Good control - 2 sensor;
w/head
Fine adj for optimum
setting
Higher max. diversion
range
Maximum range
Very short range -
still linear
Won't attach on inter-
ceptor side
Lower range - attach-
ment problem
Same switching problem
Best control range
No improvement
Not as good range as
211 A II
x4
No control switching
Diversion
Range
3-77
3-81
8-53
4-8
5-53
5-26
1-76
0-78
41-82
29-68
49-83
47-83
4-9
39-84
21-49
30-75
36-84
36-81
8-64
50-50
-------
Loq of Sewer Regulator Element Tests (Continued)
Test
Date
8-15
8-15
8-15
8-15
8-19
8-20
8-20
8-20
8-21
8-29
9-3
9-4
9-4
9-4
9-4
Test
No.
3
4
5
6
2
1
2
3
1
1
1
1
2
3
4
Config.
No.
813-1
813-1
813-1
813-1
818-1
818-1
818-1
818-1
818-1
829-1
829-1
829-1
829-1
829-1
829-1
Nozzle
Size
4x4
4x4
4x4
4x4
4x4
4x4
4x4
4x4
4x4
2x4
2x4
2x4
2x4
2x4
2x4
Sensor
Float
Valve 4
Float
Valve 4
Float
Valve 4
Float
Valve 4
Float
Valve 4-5
Float
Valve 6
Float
Valve 6
Float
Valve 6
Float
Valve 6
Float
Valve 2M
Bottles
Float
Valve 2M
Float
Valve 2M
Float
Valve 2M
Float
Valve 2M
'arameter
Tested
..ower head
.0"
Low combined
weir
Head 17.0
High combined
weir
Sensors
Sensors
Head 9.0"
Head 17.0
Flow over
discharge
Diversion
Sensor
Weir settings
Higher head
16.0
Lower head
8.0
Shrouded
discharge
109
Results of Tests
Narrow control range
Increased control range
Increase range
Increase range
Improved linearity
with area
Greater area - improved
linearity
Slightly improved
range
Slightly improved range
Greatly effects
performance
Performance matches
early 2x4
Narrow control range
- not linear
Affect performance
significantly
Slightly improved
control ranges
Lower control ranges
Large shroud need to
separate flows
Diversion
Range
40-50
10-50
18-71
8-69
11-62
12-60
12-64
12-71
4-30
0-80
0-63
2-75
2-82
0-52
43-66
-------
APPENDIX B
Orifice Equation
Q =
2gAh
Q = Flow through orifice in cubic feet per second (CFS)
CD = Orifice discharge coefficient (non-dimensional)
A = Area of orifice in square feet
g = Gravity constant 32.2 ft/sec/sec
Ah = Differential head in feet across orifice, i.e.,
upstream head minus downstream
In order to accurately use the orifice equation to calculate the
flow through an orifice the head and orifice coefficient must be known.
Test data taken from a fluidic irrigation water diverter by the Engineering
Research Service, at Fort Collins, Colorado, provide flow vs supply
head data for a known nozzle size diverter, see Figure B-l.
These data were used with the orifice equation to obtain the values
of discharge coefficient vs differential head, see Figure B-2.
The experimental curve of discharge coefficient vs supply head
was then used with the orifice equation to form the nomograph of Figure
10 covering a nozzle area range from 0.1 to 10 square feet, a head
range from 1.0 to 100 feet and orifice coefficients from 0.8 to 1.2.
The nomograph obtained from the orifice equation and irrigation
diverter (8" x 8") test data were then compared to the data from the
1/2" x 1/2" irrigation diverter used in the predictive analysis and the
2" x 4" fluidic sewer diverter, see Figures B-3 and B-4 with the fol-
lowing results.
Comparison of Test Data
1/2" x 1/2" Nozzle Irrigation Diverter, Figure B-3
A = 1.73 (10-3) sq ft Ah = .455 ft CD = 1.2 hi/hn = 12
110
-------
90-95% 96-1007]
Original romp length
l" ramp extension
3" romp retraction
6" ramp retraction
M
<«-
-o
<
O
4.0
3.0
Figure B-l. Relationship of Discharge to the Differential Head
Across the Diverter. Fluidic Irrigation Diverter
8" x 8" Nozzle
111
-------
-------
•
-------
114
-------
Flow from Figure B-3 measured
Q = 5 gpm = 1.11 (10~2) cfs
Flow from nomograph Figure 10
Q = 1.1 (10-2) Cfs
Agreement of flows was as good as the reading accuracy of the nomograph
or approximately 1%.
2' x 4" Nozzle Sewer Regulator, Figure B-4
A = 5.5 (10-2) sq ft Ah = .915 ft CD = 1.2
Flow from Figure B-4 measured
Q = 228 gpm = . 507 cfs
Flow from nomograph Figure 10
Q - .51 cfs
Agreement better than 1% indicating the nomograph of Figure 10 is a
reliable source of design data for fluidic sewer regulators.
115
-------
APPENDIX C
Performance Design Data
Diversion vs control orifice areas for various bias orifice areas
were taken from the 1/2" x 1/2" nozzle irrigation diverter at various
supply heads, see Figures C-l and C-2.
Diversion vs supply head data were taken for sewer regulator
geometries having nozzle and aspect ratios as follows:
Aspect Ratio Nozzle Size
2 2" x 1"
1 4"x4"
.5 2"x4"
.25 1" x 4"
Actual data from test of these model regulators are shown in
Figure C-3. Inconsistencies in the data curves resulted from minor
variations in the geometries of the regulators tested. The misplace-
ment of the data points from the 4" x 4" nozzle regulator were caused
by a long nozzle which reduced the venturi effect, see Figure 50.
Diversion vs height of the combined discharge weir were measured
over a complete range of weir settings and supply heads as shown in
Figure C-4. The test data show that raising the combined weir increases
its resistance and diverts more flow to the interceptor. As the combined
weir is lowered diversion increases until the weir gets so low that an
air pocket forms along the top of the element and destroys the wall attach-
ment effect by equalizing control pressures and permitting direct com-
munication from ambient air to the interceptor control port. The water
flow is supercritical in this case. The combined weir tests were taken
for an interceptor weir exit area of 12.2 in^ which was within the operating
zone of the regulator. Figures C-5, C-6, C-7, and C-8 show data taken
for maximum and minimum diversion as a function of the interceptor weir
area which produces a loading effect somewhat like that of the combined
weir. If the interceptor area is too great the coanda effect won't form as
a result of supercritical flow conditions. Too small an area affects
maximum dry weather flow diversion to the interceptor, so that some
flow goes over the combined weir. Data were taken for two nozzle sizes
and three supply heads, see graphs.
116
-------
a .O*K>' D'd
PERCENT DIVERSION AS A
OF BIAS
H+milMIIMIIIIIIIIIM
-------
CO
-------
-------
ro
o
-------
PERCENT DIVERSION F*OM
-------
I I
I ,
-------
!
HEIGHT of NoT.-Z.Le
PERCENT DIVERSION) FROM
-------
I. I
-------
Percent diversion vs the height of water in the interceptor data
were taken for the best analog geometry with the most linear sensor to
determine diversion range and linearity over a wide range of heads,
see Figure C-9. Results showed that linearity was good over the oper-
ational head range; however, diversion performance dropped sharply
for supply heads below 3.5hn (hn = nozzle height). Diversion performance
showed improvement with head until a head of 5 hn was reached at which
point best performance is reached. Figures C-ll and C-12 show the
combined interceptor sewer test setup with the float valve, used to obtain
the diversion data discussed above.
In order to represent the change in interceptor level as a more meaningful
parameter, the sensor float valve calibration was expressed as area change
of sensor over area of regulator nozzle as follows: (see Figures C-10 and
C-13)
Sensor Area Change Ratio = - - - — (10 3)
AN
AA^ = Change in combined control area
AA2 = Change in interceptor control area
AN = Area of nozzle
Reference Point
Flow diverted 100% to interceptor
AA, = 0 AA9 = 0
1 £.*
As the interceptor water level increases area AA2 (interceptor control
area) is opened until the 50-50 diversion point is reached. As the inter-
ceptor level continues to change AA]_ (combined control area) is closed until
at 100% diversion from the interceptor, AAi is fully closed and AA2 is fully
opened giving a sensor area change ratio of:
AC + AC = -- , Ac = Area of control orifice
AN AN
Air flow through the control ports of a fluidic sewer regulator as a
function of the percent diversion was plotted as an air/water ratio for a
2" x 4" and 4" x 4" sewer regulator being controlled by a mechanical float
valve, see Figure C-14. Test observations and data confirmed the air flow
125
-------
r .
• i,
wer/es SET pae MAX/MUM
-------
I
I
-------
Figure C-ll. Low Diversion, Low Interceptor Water Level
Float Valve Control
Figure C-12. High Diversion, Loaded Interceptor
Float Valve Control
128
-------
f I
-------
w
o
-------
characteristic as shown. As the control port is opened from a maximum
diversion setting aspiration begins and increases until the 50-50 diversion
point is reached when the port is fully open. Sensor tests showed that
further increase in sensor area has no effect; that is, the regulator is
getting all the air it needs. As the diversion increases above the 50 per-
cent level, the aspiration in the open control line actually drops as a result
of the decreased venturi effect when the jet stream is directed away from
the control pocket.
131
-------
APPENDIX D
Fouling Analysis
The table shown on the following page is the test data taken as
part of the fouling analysis. The test configuration used, see Figure 76,
had a very pointed splitter which was much more susceptible to fouling
than more rounded types which were found to have the same diversion
performance. Half of the fouling hang up caused by the twigs were caused
by the forks snagging the pointed splitter. The remainder of the hang ups
caused by twigs resulted in the long (10") twig lodging crosswise at the
nozzle. It is quite possible that under actual conditions where a reasonably
long approach line supplies the regulator, long objects such as twigs and
rolls of paper will align themselves in the water stream with their longest
dimension in the direction of the stream, allowing them to pass through the
regulator rather than hang up crosswise.
132
-------
Fouling Test Data
Regulator Element Nozzle Size 2" x 3-1/4"
Supply Head 9-1/4"
Interceptor Weir Open
Code
Letter
Fig.
4-16
A
B
C
D
E
F
G
H
-
-
-
-
Subject
Length
Twig- 6. 5"
Twig- 3.0"
Twig- 6.0"
Twig- 6. 5"
Twig- 4.5 "
Twig- 3. 5"
Twig- 4.0"
Twig- 10.0"
Paper- 2 "x3"
Paper- 4 "x5"
Paper- 2 "x6"
Paper-10"xl6"
Number
of
Forks
2
0
2
0
2
1
1
1
-
•
-
-
Total
Passes
10
10
10
10
10
10
10
10
10
10
10
10
Passed
Through
5
10
8
9
9
10
10
5
10
10
10
8
Hung Up
Then
Passed
Through
3
0
1
1
1
0
0
2
0
0
0
0
Hung
Up
2
0
1
0
0
0
0
3
0
0
0
2
Remarks
Fork caught splitter
Large fork caught
pointed splitter
Caught in neck of
nozzle
Hung up on splitter
TOTAL HANG UPS
133
-------
APPENDIX E
Control Line Sizes
The control line I.D. vs length vs regulator nozzle area data as shown
in Figure 13 was derived with the following considerations.
The maximum airflow through the control port occurs at the 50-50 diversion
point, see Figure 70.
The maximum airflow for fluidic sewer regulators is 5 x 10~2 times the
water flow through the regulator, see Figure 70.
The optimum sensor area for linear analog control of a sewer regulator
is 20 (10~3) AJJ (A}j = area of nozzle -in2), see Figures 62 and 63.
Knowing the maximum area and maximum airflow we can calculate the
pressure drop across the sensor orifice with zero control line length,
using Figure E-l.
Using the sensor area vs diversion characteristic for linear analog control
shown in Figures 62 and 63, it was assumed that a degradation in di-
version performance of 10% would be tolerated as a result of the control
line length to be added to the sensor control orifice.
The area change corresponding to a 10% degradation in diversion per-
formance, from Figures 62 and 63, is 5 (10~3) AJJ (A^ = area of nozzle).
It was then assumed that the same maximum control airflow would pass
through an orifice with new area equal to the optimum sensor area minus
the area change corresponding to the 10% degradation in diversion per-
formance:
New area = [20 (1Q-3) - 5 (10~3) j
= 15 (10~3) Ajsj (AN = area of nozzle)
Using the maximum airflow and this new area the pressure drop across the
new area is obtained from Figure E-l.
The difference in pressure drops (AP) between the pressure drop across
the optimum sensor orifice area and the sensor orifice area which would
provide a diversion change of 10 percent is then the tolerable pressure
drop in the control lines when the control orifice is fully open and airflow
is at its maximum value.
134
-------
10.0
0.
0.01 0.1 1.0
FLOW (Q) IN CUBIC FEET PER SECOND (CFS)
10.0
Figure E-l. Air Flow Through Orifice
-------
The pressure drop (AP) was then used with the nomograph showing Figure
E-2* to determine the minimum I.D. (internal diameter) of a 100 foot
control line. By using various scale factors of AP with the nomograph
of Figure E-2, a family of curves of control line I.D. vs length of control
line vs regulator nozzle area was obtained. (See Figure 13.) The air
pressure drop nomograph (see Figure E-2) used to obtain the control pipe
size is demonstrated through the following example:
The temperature of the fluid on the "t" scale is connected
with the exhaust pressure of the pipe on the P scale intersecting
index A at point a.
P = 0 psig, t = 700p
Point a is then connected to the flow rate through the pipe
as shown on the "V" scale and intersecting index B at point b.
V = 1.0 scfs = 60.0 scfm
Point b is then used with the allowable pressure drop for
100 foot of control pipe from scale &PIQO am* ^e intersection of
a line through these two parameters gives the minimum required
control line size.
AP = .04 psig
Minimum I.D. = 2.9 inches
*This figure is based on an article "Air Pressure Drop Nomograph," by
F. Kaplan, Kaiser Engineers, appearing in the April, 1967, issue of
Controls Engineering Magazine.
136
-------
Figure E-2. Air Pressure Drop Nomograph
Pressure - PSIG
Pressure Line Drop/100 Feet = PSIG
Flow Through Line - SCFM
Temperature of Fluid - deg F
I.D. of Tubing - inches
3o—
/ :
Inda* Index /
r« '
\
\
\
-5 \
\
— 10 \
— IS
— 20
—30
—40
— 50
— 60
— 70
— 80
— 90
— \oo
j— ISO
\
-2oo
— 250
— 3oo
P In
,
*?;n
J50—
40 —
30 —
2P —
^_^
\IO=:
\~
x-
4—
3 —
o
o
!f-/.o —
&.^1-
-1 W —
*) f\^
^
0.2—
al —
.05 —
,04 —
.03 —
D2 —
jo\ —
.ooe —
^)06 —
ex-aW—
\
i
-^— ^
^ ^~ ^.
^^7
\ /
\»
1
\ ,
\
\ 1
\ i
V
/\
/ \
/ \
/ \
/ \
/ \
/
/
»-P>?essuRE DROP
PEPMlTTCO IN
CONTROL UH£
PC* 100'
In
' V
1000 —
7*o —
500 —
400 -j
3oo f-
1
20O —
1
• 100 —
b-76^
50 —
AO —
30-
"320 —
Ul
to"
u. —
7 -
5 -
4 —
3 —
\ ^ _
\
' "V
05 -
04 ~
0.3-
0.2 -
0.1 -
01 —
Jex job-
r, 2M-
-500 —
/ 1.6-
—
15 —
^ 0/3 —
El
— • —
—
— o
— -10
20
0./9 ^
30 p.j8 —
.e
Q"
M
Ol
C
-a
/"
-------
BIBLIOGRAPHIC: The Bowles Engineering Corporation. Design of a Combined.
Sewer Fluldic Regulator FWPCA Publication No. DAST-13 October 1969.
ABSTI&CT: The objective ot this program was to demonstrate feasibility, and
to develop a workable configuration for a Combined Sewer Fluidtc Regulator,
whose purpose Is to minimize combined sewer discharge while protecting
interceptor sewers from overloading during storm flows. A second objective
was to develop design procedures and criteria for the general application of
this concept to municipal sewer diversion requirements, including preliminary
Investigations of construction methods,, costs, and maintenance require-
ments. A third objective was to establish a plan and location for an oper-
ational demonstration of the concept with a cooperating municipality. All
objectives were successfully met. A generic Fluldic Regulator configuration
was evolved which diverts 0 to 75% of the combined sewer flow away from
the Interceptor as a function of water level sensed in the Interceptor sewer,
or combined sewer, In either an analog or digital operational mode. Appli-
cation design criteria were evolved for a range of small to medium sized
municipal sewers, in terms of a few basic parameters. Projected installation
costs are only slightly more than for conventional diversion structures; while
the anticipated construction and maintenance requirements are simple and
minimal. The City of Philadelphia was established as a potential demon-
stration site, and a demonstration unit should become operational In late
1970. Recommendations were made for experimental activity to improve
regulation linearity; expand application size limit, and to better definitive
construction methods and costs. This report is submitted In fulfillment of
Contract 14-12-486, between the Federal Water Pollution Control Adminis-
tration and the Bowles Engineering Corporation.
ACCESSION NO:
KEY WORDS
Combined Sewers
Low Cost
Low Maintenance
Fluldic
Regulator
Variable Diversion
No-Moving-Parts
u
BIBLIOGRAPHIC: The Bowles Engineering Corporation. Design of a Combined
Sewer Fluldic Regulator rWPCA Publication No. DAST-13 October 1969.
ABSTRACT: The objective of this program was to demonstrate feasibility, and
to develop a workable configuration for a Combined Sewer Fluldic Regulator.
whose purpose Is to minimize combined sewer discharge while protecting
Interceptor sewers from overloading during storm flows. A second objective
was to develop design procedures and criteria for the general application of
this concept to municipal sewer diversion requirements. Including preliminary
Investigations of construction methods, costs, and maintenance require-
ments. A third objective was to establish a plan and location for an oper-
ational demonstration of the concept with a cooperating municipality. All
objectives were successfully met. A generic Fluldic Regulator configuration
was evolved which diverts 0 to 7596 of the combined sewer flow away from
the Interceptor as a function of water level sensed In the Interceptor sewer,
or combined sewer. In either an analog or digital operational mode. Appli-
cation design criteria were evolved for a range of small to medium sized
municipal seweis, In terms of a few basic parameters. Projected Installation
costs are only slightly more than for conventional diversion structures; while
the anticipated construction and maintenance requirements are simple and
minimal. The City of Philadelphia was established as a potential demon-
stration site, and a demonstration unit should become operational In late
1970. Recommendations were made for experimental activity to Improve
regulation linearity: expand application size limit, and to better definitive
construction methods and costs. This report Is submitted In fulfillment of
Contract U-12-486, between the Federal Water Pollution Control Adminis-
tration and the Bowles Engineering Corporation.
BIBLIOGRAPHIC: The Bowles Engineering Corporation. Design of a Combined
Sewer Fluldlc Regulator FWPCA Publication No. DAST-13 October 1969.
ABSTRACT: The objective of this program was to demonstrate feasibility, and
to develop a workable configuration for a Combined Sewer Fluldic Regulator,
whose purpose Is to minimize combined sewer discharge while protecting
interceptor sewers from overloading during storm flows. A second objective
was to develop design procedures and criteria for the general application of
this concept to municipal sewer diversion requirements, including preliminary
investigations of construction methods, costs, and maintenance require-
ments. A third objective was to establish a plan and location for an oper-
ational demonstration of the concept with a cooperating municipality. All
objectives were successfully met. A generic Fluldic Regulator configuration
was evolved which diverts 0 to 75% of the combined sewer flow away from
the Interceptor as a function of water level sensed in the Interceptor sewer,
or combined sewer. In either an analog or digital operational mode. Appli-
cation design criteria were evolved for a range of small to medium sized
municipal sewers. In terms of a few basic parameters. Projected Installation
costs are only slightly more than for conventional diversion structures; while
the anticipated construction and maintenance requirements are simple and
minimal. The City of Philadelphia was established as a potential demon-
stration site, and a demonstration unit should become operational in late
1970. Recommendations were made for experimental activity to Improve
regulation linearity; expand application size limit, ami to better deflnltlze
construction methods and costs. This report Is submitted In fulfillment of
Contract 14-12-486, between the Federal Water Pollution Control Adminis-
tration and the Bowles Engineering Corporation.
ACCESSION NO:
KEYWORDS
Combined Sawers
Low Cost
Low Maintenance
Fluldic
Regulator
Variable Diversion
No-Moving-Parts
ACCESSION NO:
KEYWORDS
Combined Sawers
Low Cost
Low Maintenance
Fluldic
Regulator
Variable Diversion
No-Moving-Farts
------- |