EPA-600/2-77-071
August  1977
Environmental Protection Technology Series
                                                      Environmental Researeh Laboratory

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

-------
                               FOREWORD

      The Environmental Protection Agency was created because of increasing
public and government concern about the dangers of pollution to the health and
welfare of the American people.  Noxious air, foul water, and spoiled land are
tragic testimony to the deterioration of our natural environment. The complex-
ity of that environment and the interplay between its components require a
concentrated and integrated attack on the problem.

      Research and development is that necessary first step in problem solution
and it involves defining the problem,  measuring its impact,  and searching for
solutions.  The Municipal Environmental Research Laboratory develops new and
improved technology and systems for the prevention, treatment, and manage-
ment of wastewater and solid and hazardous waste pollutant discharges from
municipal and community sources , for the preservation and treatment of public
drinking water supplies, and to minimize the adverse economic, social, health,
and aesthetic effects of pollution. This publication is one of the  products of
that research; a most vital communications link between the researcher and the
user community.

      This  report  describes the evaluation, under normal  municipal operating
conditions , of a new technical approach for the regulation of combined sewer
flows .  The program  demonstrated that this approach can  provide municipali-
ties with more cost effective combined sewer regulation,  so that the pollution
of receiving waters from overflows can be minimized, at lower capital and
operating costs than required by currently used regulators.
                                      Francis T. Mayo
                                      Director
                                      Municipal Environmental Research
                                      Laboratory
                                   111

-------
                                ABSTRACT

      This report describes a USEPA Demonstration Program undertaken by the
Philadelphia Water Department, in which a novel combined sewer regulation
concept is evaluated under municipal service conditions.  This concept uses
fluidic technology to generate dynamic regulation characteristics without me-
chanical moving elements in the sewage flow. This concept had previously
been the subject of a USEPA Research Program.

      The report describes the chosen sites, the desired regulation functions,
design rationale and analyses, and construction details for two fluidic regula-
tor installations, representing small, and fairly large flow requirements . The
construction of the sites, regulators, and appurtenances,  together with labora-
tory tests and site calibrations are described, including difficulties encoun-
tered, and design modifications generated to achieve the desired system per-
formance .

      The two fluidic regulators are compared in detail with equivalent conven-
tional static regulators, from the standpoints of hydraulic performance; sur-
veillance, maintenance and repair; and initial and operating costs.  The report
concludes with recommendations for system and hardware design improvements.

      This report was submitted in fulfillment of USEPA Demonstration Grant
11022 FWR, by the Philadelphia Water Department under partial sponsorship of
the U.S. Environmental Protection Agency. The report covers the period from
February, 1971, to March, 1975, and the work was completed as of September,
1976.
                                   IV

-------
                            CONTENTS
Foreword	^
Abstract	iv
Figures	vi
Abbreviations and Symbols	viii
Acknowledgement	ix

   1.   Introduction	   1

   2.   Summary	9

   3.   Conclusions	10

   4.   Recommendations	12

   5.   Program History	  .  14

   6 .   Regulator Design Analysis	16

   7.   Design and Construction	26

   8 .   Regulator Calibration	46

   9.   Program Evaluation	57

  10.   Concluding Comments   	70

References and Bibliography	72

Appendices

   A.   Regulator Design Calculations	73
   B.   Calibration Data	79
   C .   Conventional Regulator Performance Analysis	89
Glossary
94

-------
                                FIGURES

Number                                                           Page.

   1   Schematic Arrangement, Fluidic Combined Sewer Regulator . .    3

   2   Conventional Mechanical Combined Sewer Regulator  ....    4

   3   System Schematic Arrangement using U-Tube Sensor	    6

   4   Site and Drainage Area, 67th/Callowhill Sts ., Fluidic
        Sewer Regulator		   17

   5   Predicted Hydraulic Performance, 67th/Callowhill Sts .,'
        Fluidic Sewer Regulator	   19

   6   Site and Drainage Area, Bingham St., Fluidic Sewer
        Regulator	   20

   7   Predicted Hydraulic Performance, Bingham St., Fluidic
        Sewer Regulator	   22

   8   System Schematic Arrangement using Diaphragm Valve ....   24

   9   Fluidic Sewer Regulator,  67th/Callowhill Sts . Site  .....   27

  10   67th/Callowhill Regulator Site, Before and After Modifi-
        cation 	  ...   29

  11   Site Design, 67th/Callowhill Sts.   ............   30

  12   Site Design, 67th/Callowhill Sts.	   3.1

  13   Site Design, Bingham St	33

  14    "     »   ,    "     "		34

  15    "     "   ,    "     "      	35

  16    "     "   ,    "     "	  . .  .	36

                                  vi

-------
                       FIGURES  (CONTINUED)



Number                                                             Page

  17   Bingham St. Fluidic Regulator under Construction	    37

  18   Bingham St. Fluidic Regulator Dam	    38

  19   Bingham St. Fluidic Regulator Interceptor Sensor	    39

  20   Bingham St. Fluidic Regulator U-Tube Sensor	    40

  21   Existing Slot Regulator Installation, 67th/Callowhill Sts. .  .    42

  22   Fluidic Regulator Installation,  67th/Callowhill Sts	    43

  23   Fluidic Regulator Installation,  67th/Callowhill Sts.   ......  44

  24   Fluidic Regulator Dip Tube Sensor, 67th/Callowhill Sts.  .  .    45

  25   Laboratory Calibrations, 67th/Callowhill Regulator .  ....    48

  26   Site Calibration, 67th/Callowhill Regulator	    49

  27   Water Level vs. Time, Bingham St. Fluidic Regulator
         Calibration	    51
  28   1/6 Scale Model, Bingham St.  Regulator  . .  .  .  .  .  . .  .    53

  29   1/6 Scale Model, Test Results	    54

  30   Modified Bingham St. Regulator Outfall Geometry	    55
  31   Interceptor Maintenance Log, 67th/Callowhill Fluidic
         Regulator, Fiscal '75	  .    59

  32   Interceptor Maintenance Log, 67th/Callowhill Fluidic
         Regulator, Fiscal '76	    60

  33   Hydraulic Performance Comparison, Fluidic vs. Con-,
         ventional Static Regulator,  67th/Callowhill Site	    62

  34   Hydraulic Performance Comparison, Fluidic vs. Con-
         ventional Static Regulator,  Bingham St. Site	    64
                                  VII

-------
                  LIST OF ABBREVIATIONS AND SYMBOLS

Q     Water flow, in general
Qtot  Total water flow pas sing through regulator
Qout  Water flow passing through fluidic regulator outfall discharge
Qin   Water flow passing through fluidic regulator interceptor discharge
cfs   Cubic feet/second, normal measurement of water flow
h     Hydraulic head of water upstream of regulator, in ft.
hn    Height of fluidic regulator inlet nozzle,  in ft.
hd    Height of fluidic regulator outfall discharge above regulator floor, or
         invert, in ft.
hD    Height of water level above regulator dam, in ft.
Wn    Width of fluidic regulator inlet nozzle, in ft.
a     Inlet nozzle aspect ratio, = hn/wn
A     Inlet nozzle, or orifice  area, in general, in ft. 2, or in.
An    Inlet nozzle area, in ft.2
DWF  Combined sewer dry weather flow, in cfs
D     Fluidic regulator flow diversion capability, in maximum percent of Qtot
         diverted to regulator outfall discharge
NPT   National Pipe  Thread, (F) female, (M) male
PVC   Polyvinylchloride, plastic material
PE    Polyethylene,  plastic material
V     Velocity, in general               ,
fps   Feet/second,  measurement of velocity
EL    Elevation above mean sea level, ft.
CD   Regulator, or orifice discharge coefficient
d     Diameter, in general; in ft. or  in. as indicated
p     Pressure, in general;  in in. of  water or as  indicated
                                  viii

-------
                         AC KNOWLEDGE MENTS

      The author wishes to acknowledge the technical support given by Mr.
Richard Field, Chief,  Storm and Combined Sewer Section (Edison,  NJ), USEPA,
Municipal Environmental Research Laboratory, Cincinnati, Ohio, and his staff
during the conduction of this program.                                   v

      Similarly, the author wishes to acknowledge the considerable, enthusi-
astic work performed by the Philadelphia Water Department under the direction
of Water Commissioner Carmen F. Guarino, including the Chief of the Water
Pollution Control Division, Mr. Michael D. Nelson; the Chief of the Intercep-
tor Service Section, Mr. William Barnes; and the Project Director, Mr. .M.
Stewart Cameron, and his  staff.  Without the efforts of this organization the
project could not have been accomplished.

      The author also wishes to acknowledge the Bowles Fluidics Corporation,
who  pioneered the basic technology behind  this  program, and who provided the
project with competent mechanical design,  model  shop construction, laboratory
test, and editorial services.

      The author finally wishes to express thanks and appreciation to Mrs . M •
Stewart Cameron for her gracious and helpful service in providing the author
with clean coveralls  during the field test and calibration phases of the program.

-------

-------
                               SECTION ,1
                             INTRODUCTION
PROJECT BACKGROUND

      The subject program has been established as part of the USEPA's overall
objective to improve the regulation of combined sewer flows to minimize over-
flows and thus relieve the resulting pollution of receiving waters.  This project
demonstrated and evaluated the f luidic combined sewer regulation concept.
This concept was previously investigated at the pilot-scale level by Bowles
Engineering Corporation under USEPA sponsorship (Ref .#1).  This earlier project
included the basic analytic and experimental exploration of the concept, and
the derivation of application design criteria. The project also included an in-
vestigation of several major cities in the eastern part of the U.S. to determine
both applicable sites and municipal interest in participating in a USEPA Demon-
stration Grant concerning the concept.  As  a result,  the City of Philadelphia
was selected, and has furnished two combined sewer sites applicable to the
evaluation of fluidic combined sewer regulators .  This  report describes the
results of the Demonstration Grant program.

CURRENT PRACTICE

      Combined sewer regulation is currently being performed by either static,
or dynamic regulators. The static systems are simple, have no  mechanical
moving parts,  are relatively reliable and easily maintained.  However, the
hydraulic performance of a static regulator is typically determined only by the
flow conditions in the combined sewer, independent of flow conditions in  the
interceptor sewer.  As a result,  these  devices tend to either overflow before
using all the available capacity of the interceptor during areawide light storm
flows, or surcharge the interceptor during areawide heavy storm  flows .  The
result is either unnecessary pollution of receiving waters,  or a flood/surcharge
condition to the interceptor sewer structure, and overloading of  the water  pol-
lution control plant.  Dynamic type regulators can provide much  better hydraulic
performance, in response to either or both combined and interceptor sewer con-
ditions .  These systems  are, however, much more complex, costly,  less re-
liable and more difficult to maintain in service.

FLUIDIC REGULATOR CONCEPT

-------
      The fluidic regulator concept, as illustrated in Figure  1, offers most of
the flexibility and performance of the dynamic regulator, yet retains the low
cost, simplicity, reliability of operation, and ease of maintenance of the  static
regulator.  The fluidic regulator consists of a structure with no mechanical'
moving parts in the sewage flow stream, and permanently open flow passages .
The construction is from standard,  non-corrosive materials such as concrete,
high grade plastics, stainless steel, etc. These characteristics render the
fluidic regulator minimally susceptible to common problems such as blockages,
corrosion, fouling, jamming, etc. affecting standard moving part regulators of
the type shown in Figure 2; and essentially equivalent to standard static regu-
lators .

      Another significant advantage of the fluidic regulator is that it is self •-
powered. Control energy is derived directly from the sewage flow, thus elimi-
nating the requirement, and initial  and maintenance costs  of external electric,
hydraulic,  mechanical, etc. power. The elimination of dependence on exter-
nal power assures greater reliability, particularly during storm events when
municipal power outages occur most frequently.

      The fluidic regulator obtains  operational flexibility comparable to that of
complex dynamic regulators through the sensing of water levels by simple  dip
tube sensors, which can be located in the interceptor, or upstream of the regu-
lator in the combined sewer, or both.  Remote command control of a fluidic
regulator can be implemented using an electrically actuated pneumatic valve
operated at low power levels easily transmitted over standard telephone lines .

REGULATOR OPERATION

      As shown in Figure 1, a large fluidic diverter is embedded in a dam across
a combined sewer.  One discharge is elevated and flows through the dam into
the outfall. The other discharges into the interceptor.  The regulation logic is
as follows:

         1)  Dry weather flow proceeds directly to the interceptor, since the
outfall discharge is elevated.  This action is similar to that of a conventional
static regulator.

         2)  With light to moderate  storm flow, the hydraulic head upstream of
the dam increases above the regulator inlet, causing it to flow full. The re-
sulting control action of the regulator occurs:  With full flow, the venturi
shape of the inlet generates a sub-ambient pressure at the control ports which
tends to aspirate ambient air into the regulator.  If the interceptor-side control
port were blocked, and the outfall-side port opened to atmosphere,, a pressure
differential would be exerted transversely across the incoming flow stream to-
ward the interceptor-side discharge, and the flow would exit entirely to that

-------

-------

-------
side, despite the fact that the upstream water level may be higher than the out-
fall discharge elevation. If the control port closure is reversed, the pressure
differential across the incoming flow stream would also reverse, and a signifi-
cant fraction of the flow would discharge over the outfall; the higher the up-
stream head,  the greater possible fraction of total flow over the outfall.

      In the simplest regulator configuration, the interceptor-side control port
would be connected to ambient atmosphere through a small orifice, while the
outfall-side control port is  connected to a dip tube level sensor located in the
interceptor. If the interceptor water level is below the dip tube, indicating a
condition of additional capacity in the interceptor, atmospheric  air is aspirated
freely into the outfall-side  control port, maintaining the regulator flow toward
the interceptor.  If the interceptor level rises above the end of the dip tube,
indicating a condition of interceptor near-capacity, the aspirated airflow into
the outfall-side control port is cut off, and the pressure differential will re-
verse toward the outfall side,  causing most  of the regulator flow to shift to the
outfall,  preventing interceptor surcharging.  The regulator thus prevents any
outfall flow as long as the interceptor has sufficient flow capacity.  Only when
its capacity is reached does outfall discharge occur, thus minimizing pollution
in the receiving waters .

         3) In heavy storm  flows the upstream head will rise above the dam
crest, and large outfall flows will occur over the dam.  With increasing hy-
draulic head,  however, the regulator will direct an increasing fraction of its
flow to the outfall, and the actual flow to the interceptor will be reduced,
thereby continuing the prevention of surcharging of the interceptor.

      The above-described  logic is basically digital in nature,  in which the
regulator acts as a flow switch, depending on the condition of submergence of
the dip tube in the interceptor. The fluidic regulator is also capable of propor-
tional regulation action, through the use of the sensor arrangement shown in
Figure 3  .  Here the simple dip tube is replaced by a multiple tube assembly,
whose open ends are graduated in elevation. Each tube is connected to the
connecting air line through  an  orifice,  so that the aspirated airflow reaching
the outfall-side control port is decreased proportionately as the interceptor
water level rises above the end of each tube.  As  the airflow into the control
port is attenuated, the internal pressure drops proportionately below atmos-
pheric ambient.  The line to this control port is also connected to the closed
end of a  "U-tube" sensor containing a  low volatility fluid. As the internal
pressure decreases, the fluid level in the U-tube  sensor closed end rises,
•while the level in the  open  end falls,  in the  manner of a manometer. A multi-
ple, graduated dip tube sensor assembly, similar to that in the  interceptor, is
suspended in the open end of the U-tube sensor, so that all of the tubes are
submerged when the U-tube fluid levels are the same.  Thus, as dip tubes are
successively  submerged by rising water level in the interceptor, the dip tubes
in the U-tube sensor are successively  uncovered, thus allowing an increasing

-------
                Multiple
                Dip Tube
                Level
                Sensor
U-Tube Sensor
 Combined
 Sewer
 Flow
            Outfall Discharge
                      Interceptor Discharge  51
Fluidic Regulator
                             Interceptor Sewer
                                                          Multiple Dip
                                                         i Tube Sensor
     Figure 3    System Schematic Arrangement using U-Tube Sensor

-------
airflow into the interceptor-side control port. This action results in a smooth,
proportional transfer of regulator flow away from the interceptor as its reserve
flow-carrying capacity decreases .  It will be noted that while the sensor ar-
rangement is more complex, no mechanical moving parts in the water flow are
required, thus preserving the basic intrinsic desirable characteristics of the
fluidic concept.

PROJECT OBJECTIVES AND APPROACH

      The specific project objectives and work description, as stated in Demon-
stration Grant (11022 FWR) agreement are as follows:

1.    Design, construction and operation of a fluidic regulator for a flow range
below 2 cf s with a minimum of reconstruction.  The unit will be capable of
demonstrating fluidic action on demand by use of city water to simulate storm
flow and a transparent top for observation.

2 .    Design, construction and operation of a fluidic regulator for a 4 cfs peak
dry weather flow.  This will demonstrate the use of a fluidic device at higher
flows on combined sewage. The automatic control of the overflow will be done
by sensing the interceptor level. No overflow will occur until the interceptor
reaches a pre-determined limit.

3 .    Evaluate the operation of above fluidic devices for one year, and relate
their performance, both advantages and disadvantages to conventional regu-
lators throughout the Philadelphia sewer system.

4.    Assembly of all data from the evaluation and testing program including
rainfall data, overflow vs . sensor height  so that the application of fluidics to
sewer regulator design can be demonstrated on a full-scale basis  .

      Enlarging the foregoing, this project would evaluate fluidic  combined
sewer regulators using both the digital and proportional modes operation, as
described above.  In addition, the digital  mode would also be evaluated using
a simple diaphragm valve in combination with the dip tube sensor. The digital
operational  mode would be evaluated at a diversion site near 67th and Callow-
hill Sts. in  Philadelphia, and would be included in item #1 of the  Project
Objectives.  This site has been previously regulated using a plate-type static
regulator. Also located at this  site was th,e test set-up used in USEPA Demon-
stration Grant #11023  (Ref .#3), designed to evaluate the effectiveness of a
rotary microstrainer in improving the quality of combined sewer flows .  In
selecting this site for the subject program, it was felt that the availability of
the test  set-up would facilitate the testing and calibration of the < fluidic regu-
lator through the uses of existing facilities .

      The proportional operational mode of regulation was  planned as part of

                                     7

-------
item #2 of the Project Objectives. The selected site was a diversion point near
the intersection of Bingham St. and Tacony Creek.  This site had previously
been regulated using a diversion dam across the sewer, with an upstream,
manually positioned sluice gate to regulate  flow into a connecting sewer to the
interceptor.

      The fluidic regulators to be constructed at both sites would be evaluated
per items #3 and #4 in the Project Objectives .

-------
                               SECTION 2

                               SUMMARY
      Two Combined Sewer Fluidic Regulator Systems have been designed, con-
structed, and installed at two diversion sites in the City of Philadelphia, PA,
replacing conventional static regulators. The first, handling combined flows
up to about 2 cfs , has operated well for a period of over 3 6 months, providing
hydraulic performance similar to that of dynamic regulator with the same low
incidence of blockages and other service problems of comparable static regu-
lators .  The second, handling combined flows up to about 25 cf s, has operated
for a period of about 30 months, with virtually no blockage or service problems,
but has been prevented from reaching its desired hydraulic performance by un-
foreseen,  abnormal flow impedance in the connecting sewer to the interceptor.
This problem has unfortunately not been correctible within the scope of the
program.

      The project has accomplished 75% of its specific objectives;  100% for
the small unit, and about  50% for the large unit.

      A number of recommendations have been formulated for system modifi-
cations to improve regulator performance, mechanical design,  installation, re-
liability, and maintainability, and are  described within the report.

-------
                               SECTION 3

                              CONCLUSIONS
      The principal conclusions from this program is that the fluidic combined
sewer regulator can provide much-improved hydraulic regulation performance,
currently obtainable only from complex dynamic regulators, with equivalent or
improved reliability and maintenance performance, as compared to conventional
static sewer regulators . The improved hydraulic performance will result in
fewer and less serious overflows during light to moderate storm flows, with a
much reduced tendency to surcharge the interceptor during heavy storm flows .
The excellent reliability and maintenance performance will result in low opera-
ting costs,  and overall greater cost effectivity than either conventional static
or dynamic regulators.

      The basic design criteria and rationale  generated in the previous research
phase CRef. #1), which was used as the basis for design for this project, is
considered demonstrated for fluidic regulators handling flows up to 25cfs, and
probably practical for flows up to 50 cfs .

      The 5 in. x 5 in. inlet dimensions of the small regulator operated on this
program proved workable, but are judged to be the minimum practicable for
future fluidic regulator designs.

      The surveillance, maintenance, and repair procedures developed by the
Philadelphia Water Department for the fluidic regulators, proved simpler,
faster, and  somewhat less costly than those for the conventional static units
they replaced.  In all cases, these operations were performed with minimum
skill-level personnel.  These procedures were considerably simpler and less
costly than those required for conventional dynamic type regulators, which
generally require high skill-level personnel.

      The costs for retrofitting a fluidic regulator into the two chosen  sites
were estimated on the average at about 40% more than retrofitting a conven-
tional static regulator. The cost of installing a fluidic regulator into a new
location was estimated at only about 10% more than for the conventional static-
unit .  In general, the costs for retrofitting conventional dynamic regulators
into existing static regulator sites  are expected to be considerably greater
than for fluidic regulators.  Although specific cost ratios were not generated

                                     10

-------
due to the wide variation in dynamic regulator characteristics, it was found
that equipment costs for conventional dynamic regulators are much greater than
fluidic regulators per quantity flow handled; (for example, a 4 ofs mechanical
float-operated regulator recently cost the Philadelphia Water Department
$ 13,000; the corresponding fluidic regulator cost is estimated at $4,500).  In
addition,  it was estimated that site modification and installation costs would
run 25% to 50% greater for conventional dynamic regulators due to the combi-
nation of requirements for precision assembly, and alignment of  mechanical
assemblies, construction of additional flow chambers or passages,  and in-
stallation of operational energy supply facilities . A rough estimate of the
overall conventional dynamic/fluidic regulator retrofit cost ratio would be at
least 2:1.

      The costs for installing conventional dynamic regulators into new loca-
tions would be greater than for fluidic regulators for the same reasons given
above.  However, the total cost ratios would not be as great as for the retrofit
case, since the cost of basic excavation and construction common to either
type regulator,  generally exceeds the equipment and installation costs.

      In addition, a number of conclusions were reached regarding specific
points of regulator system and hardware design, construction, and maintenance.
These concerned the desirability of: locating interceptor level sensor air lines
outside the connecting sewer (to facilitate sewer maintenance); fluidic element
construction using prefabricated plastic or metal shells to achieve lower cost,
better dimensional control, elimination of nozzle erosion; the installation of
manual valves in sensor air lines to facilitate on-site calibration,  and routine
maintenance checkout; the exploration of alternate regulator-to-sensor commu-
nication techniques to reduce cost,  and the exploration of remotely controlled
regulator operation to facilitate computerized municipal area flow route control.
                                     11

-------
                               SECTION 4

                          RECOMMENDATIONS
      The recommendations resulting from this program fell into two categories:
recommendations to achieve the project objectives that were precluded by the
connecting sewer flow impedance problem encountered at the Bingham Street
site; and recommendations pertaining to regulator design improvements.

PROJECT OBJECTIVE RECOMMENDATIONS

      The recommendations to achieve project objectives are:

         1. Proceed with the necessary repair to the connecting sewer at the
Bingham St. site . The present condition prevents proper operation of any type
of combined sewer regulator.

         2. As an interim measure,  in order to fully demonstrate the propor-
tional mode Of operation in a larger size unit, the present installation should
be fitted with a removable invert that would reduce the inlet nozz;le area to
match the present flow-carrying capacity of the connecting  sewer.  Upon the
eventual repair of the connecting sewer,  the regulator would be returned to its
present condition, at which normal design operation could be expected.

         3. The present 4 in. air line to the  interceptor dip tube sensor should
be removed from the connecting sewer and a suitable alternate sensor and/or
communication means installed so that interceptor level regulation logic can
be demonstrated while permitting normal maintenance procedures in the connec-
ting sewer without the interference of the air line.

         4. An alternate, multiple dip tube sensor should be installed upstream
of the dam so that the U-tube sensor also can be fully demonstrated, at  least
during system calibration runs .

DESIGN IMPROVEMENT RECOMMENDATIONS

      Recommendations for system design improvement are:

         1.  In view of the practicality demonstrated by the diaphragm valve in

                                    12

-------
the digital operational mode, development should be started on a proportional
operational mode diaphragm valve,  which would be compared with the all-
fluidic U-tube sensor.

         2 . Development should be initiated on means for remote command
control of regulators, so that they can be integrated into large, computer-
controlled sewage flow networks .  The communication links could be either
hard wire (telephone line) or wireless in nature.

         3 . Suitable means should be designed into a regulator installation
to obtain direct access to the regulator control ports, while bypassing sensor
connecting lines, to facilitate the calibration and trouble-shooting of the
system.
                                   13

-------
                              SECTION 5
                            PROJECT HISTORY
      The Phase I effort was started during February, 1971.  Phase I work con-
sisted of regulator design analyses for the two selected sites. This work, per-
formed by the Bowles Fluidics Corporation during the spring and summer of 1971
included the inlet sizing; the prediction of regulator flow and diversion capa-
bility vs. upstream hydraulic head; and the design calculations and drawings
for the regulators, sensors, and connecting lines to the sensors.

      Phase II work consisted of the design of the  site  installations, and was
performed by the Philadelphia Water Department.  The 67th/Callowhill regu-
lation site construction was completed during the winter of 1973; the Bingham
St. site was completed in early summer 1973, completing Phase II.

      The calibration and checkout of the  67th/Callowhill site was completed
in October of 1973 . Calibration of the Bingham St. site was started during the
summer of 1973, and continued sporadically until the following spring, during
which it became evident that a modification in the  regulator outfall discharge
design was required.  A 1/6 scale model was constructed and tested, and a
design modification generated.  The installation was reworked during the
summer and fall of 1974, and additional calibration attempts were made during
October and November. During these calibration attempts, the potential for
normal performance was shown, but successful calibration was prevented by
excessive flow impedance in the 24 in. connecting sewer, which caused a
flow backup before full regulation action could be established. (This condition
had obviously been present in some degree through all the calibration attempts.)
Further calibration was suspended, pending a detailed investigation of the in-
terior of the connecting sewer,  which was conducted by the Philadelphia Water
Department during the spring and early summer of  1975. This investigation,
which employed a remotely controlled TV camera, showed some blockages,
which were removed, and considerable deterioration of  the sewer walls, which
permitted heavy infiltration from Tacony Creek and, in effect,  generated the
observed flow impedance.

      The results of the investigation were discussed at a meeting at USEPA,
Edison, NJ, on December  19, 1975, attended by representatives of Peter A.
Freeman Associates, Inc.   and the Philadelphia Water Department.  In view
                                    14

-------
of the large cost and expected time delay in repairing the connecting sewer,
and in consideration that the 67th/Callowhill regulator had operated success-
fully from both hydraulic performance and maintainability standpoints, it was
agreed that the basic objectives of the program had been sufficiently met,
that this Phase of the program should be concluded and reported.
                                   15

-------
                               SECTION 6

                       REGULATOR DESIGN ANALYSIS
GENERAL

      Several factors must be considered in the selection of a fluidic regulator
inlet size.  From the standpoint of pollution abatement, the regulator should
admit as much flow as possible before overflowing the dam, in order that the
maximum possible flow within the reserve flow-carrying capacity of the inter-
ceptor can be diverted to the treatment plant.  This is particularly important
in the  early stages of a storm event, where a maximum of pollution from both
sanitary flow and "first flush" flow is present. The desirability for maximum
flow thus calls for a maximum size inlet.   On the other hand,  a fluidic regula-
tor requires a minimum hydraulic head (approximately 2.5 inlet heights above
the inlet) in order to provide a practical range of regulation of about 50% of
the total regulator flow. This requirement, within the constraint, of the maxi-
mum dam height allowable within a particular  sewer, may limit the inlet size.
The size of  the connecting sewer to the interceptor can be another limitation
to inlet size, particularly in situations where a fluidic regulator is to be
retrofitted into an existing diversion site.  The fluidic regulator research pro-
gram (Ref. #1)  showed that for proper flow impedance matching of the regulator
to the  connecting sewer, the regulator inlet area should not exceed about 50%
of the  connecting sewer area .  As it happened, this requirement provided the
limiting factor for the regulator inlet areas for one of the selected sites.  The
effect  of improper impedance matching proved critical, as  described in Section
8, in the calibration of the Bingham St. regulator. The details of the  design
analyses discussed in this section can be found in Appendix A.

67th/CALLOWHILL SITE

      The location for this site is on the western boundary of  Philadelphia.
The drainage area is approximately 11.2 acres and land use is composed pri-
marily of two-story row type residential housing .  Figure 4 shows the drainage
area,  with dotted lines indicating the subdrainage areas and solid lines the
sewers. A  combined sewer system services the area and normal  dry weather
flow averages  1,000 gallons per hour.

       The regulator is located in a portion of Cobbs Creek Park.  The park runs
                                    16

-------
                                ; iV*+l7«7 *o  ' „>• ? WS 66  + l€37 9 r



                                f        **  '          ,   "
                                ,«.            •          I
                                        ••—-- •' '  ' ' TO ••"

                                        CALLOWHILL    >
      *»' -  j«;
 •
-------
parallel to Cobbs Creek, which is a tributary of the Delaware River.

      The original regulating device was an adjustable slot, or leaping weir,
feeding into a 24 in. outfall. The outfall has a theoretical capacity of 37 cfs.
The connecting sewer from the regulator to the interceptor sewer is 10 in. in
diameter, and runs 1500 ft.  with a change in elevation of 91 ft.

      The design maximum outfall flow of about 30 cfs permits a maximum dam
height of 12 in. above the sewer invert, assuming a maximum allowable sur~2
charge of 5 ft. upstream from the dam, with enlargement of sewer area to 2 ft.
above the dam. The connecting sewer area is 78.9 in.2; thus the inlet area
would be limited to about 40 in.2 from a hydraulic loading standpoint, corres-
ponding to an inlet size of 6.3  in. x 6.3 in.  However, with the established
dam height of 12 |n., and the maximum allowable excavation of the sewer in-
vert, the maximum available head was 14 in., which would not permit the re-
quired 2.5 inlet heights of head for 50% regulation performance. As a. result,
an inlet nozzle size of 5 in. x  5 in.  was selected.

      The predicted hydraulic performance for this regulator is shown in Figure
5. This prediction is based on the criteria developed (Ref. #1).,  Figure 5
shows an increasing available  flow regulation to the outfall, in terms of % of
total regulator flow, with increasing hydraulic head,  reaching a value of 71%
at a dam overflow head of 14 in. Also shown are the  total regulator flow, and
the predicted minimum interceptor flow. The minimum interceptor flow exceeds
the Philadelphia Water Department criterion of 2 x DWFmax.  es't'd , normally
 used in the design of static regulators .  This deviation was acceptable con-
 sidering the small values of the flow relative to the magnitude of the intercep-
tor flow, and the need to maximize the regulator inlet size to avoid blockages .

 BINGHAM STREET SITE

      This site is located in the near Northeast section of the City of Phila-
 delphia. The drainage area is one of the larger areas within the city, en-
 compassing 298 acres.  Land use is devoted to  light  industry, shopping areas
 and residential with an average imperviousness of 67%.  The sewer system
 handles a combined storm and  sanitary flow. This drainage area is shown on
 Figure 6.

       A large, unreinforced concrete arch  storm sewer, 7.50 ft. high, and
  10.25 ft. wide at the base, accepts the combined flow from the area. This
 main sewer terminates in a concrete outfall that diverts  storm water into the
 Tacony Creek.

       Within the arch section, a dam approximately 2 ft. high and a  manual
  3 ft. x 2 ft. sluice gate comprise the original storm water regulating device.
 Dry weather flow was contained behind the dam and entered a 24 in. diameter

                                     18

-------
                      Figure 5
   1.5-
   1.0--
CO
13
Predicted Hydraulic Performance
67th/Callowhill Sts.
Fluidic Sewer Regulator
                             5                    10
                    h = Supply Head, inches H2O
                                                                    X
                                  19

-------
Site and Drainage Area; Binghain St
Fluidic  Sewer Regulator
            20

-------
 connecting sewer through the sluice gate.

       The 24 in. sewer is constructed of vitrified pipe encased in concrete.
 It runs 450 ft. from the regulator to the Frankford High Level Collector on the
 west bank of the Tacony Creek.  Approximately 200 ft. of the concrete encase-
 ment serves as a low dam across the creek.

       The sluice gate and arched main sewer are located 25 ft.  below a park
 area on the low bluffs overlooking and paralleling the stream.

       As  described, the connecting sewer has an area of 3 .14 ft.2.  To provide
 the proper flow impedance match with the connecting sewer, and to allow for
 the approximately 60° elbow at the junction of the connecting sewer and inter-
 ceptor-side discharge, the regulator inlet nozzle area was set at 1.5 ft.2.  The
 inlet nozzle aspect ratio, a (height/width) was chosen at 1.5, giving a nozzle
 size of 12 in. x 18 in.  This configuration was chosen to obtain improved di-
 version performance with reduced overall horizontal dimensions, based on
 design criteria (Ref. #1),  in comparison to a square  (aspect ratio =  1) inlet.
 The upstream head at dam overflow was set at 3 .5 ft.  The ratio of upstream
 head to nozzle height was 2.33, which was felt to be adequate in view of the
 improved predicted performance available with the higher aspect ratio nozzle.

      The predicted hydraulic performance for the Bingham St. regulator is
 shown in Figure 7 . In this case, the criteria for maximum allowed flow to the
 interceptor = 2 x 'D^Fmax,est'd ~ 20 cfs is  not exceeded by the minimum inter-
 ceptor flow curve.  The maximum predicted flow regulation capability reaches
 about 65% of the total regulator flow at the dam overflow head of 3 .5 ft.

 SENSOR DESIGN

 67th St./Callowhill Site

      As described earlier, this site had been selected to demonstrate the
 digital mode of regulator operation.  Because of the length of the connecting
 sewer to the interceptor, about 1500 ft., it was  decided that the dip tube sen-
 sor should be located in a man-hole approximately 50 ft. upstream of the diver-
 sion site, thereby demonstrating regulator operation at minimum  construction
cost.  Two versions of digital operation logic were considered, as described
earlier in Section 2:  a) using a fixed air orifice on the interceptor-side control
port; and b) using a simple diaphragm air valve.  These are described as
follows,

Orifice Version —
      Using the criteria established (Ref .#1), orifice  and connecting line diam-
eters were established as follows:
                                    21

-------
Figure 7    Predicted Hydraulic Performance,
           Bingham St. Fluidic Sewer Regulator
          1                  2

      h = SUPPLY HEAD, FEET H2O
                      22

-------
         Interceptor-side Control port Orifice Dia.       7/32 in.
         Outfall-side Control port Orifice Dia.            3/8 in.
         Connecting Line I.D. (75 ft. run)               1-1/2 in.

 The recommended vertical position of the dip tube sensor was chosen to corres-
 pond to an upstream regulator hydraulic head = 4.5 in., or 7.5 in. below the
 dam crest.  (Note:  The regulator head is measured above the outfall weir ele-
 vation.)  This position provided regulator switching action when the  regulator
 total flow reached 0.83 cfs, which corresponded to the minimum interceptor
 flow maximum value, as shown on Figure 5.

 Diaphragm Valve Version—
      During the research program (Ref.  #1), it was found that improved flow
 switching to the outfall discharge was obtained if maximum airflow were ad-
 mitted into the interceptor-side control port, as against the partial flow ad-
 mitted through the fixed orifice.  A convenient method of accomplishing this
 was through the use of a diaphragm valve,  operated by the  suction generated
 at the outfall-side control port. It was recognized that the use of a moving-
 parts device represented a departure from the all-fluidic  system approach; how-
 ever, it was felt that in this case the system reliability would not suffer, since
 the device operated only on airflow, without contact with the sewage  flow.
 The operation of the diaphragm valve version is  shown schematically in Figure
 8.
      The closed side of the diaphragm valve is connected  to the dip tube
 connecting line, while the valve movement is connected to the interceptor-side
 control port. The diaphragm is spring-loaded so that the valve movement is
 normally closed when ambient atmospheric pressure is on the closed side of
 the diaphragm, a condition occurring when the dip tube is above water. When
 the dip tube is submerged, aspirated airflow is cut off, and the pressure  drops
 below atmospheric ambient, causing the  diaphragm to open the valve movement,
 allowing air to enter the interceptor-side control port.  This causes the regula-
 tor to transfer a  large fraction of its flow to the outfall discharge .

      The diaphragm valve was designed to start opening at -1 in. H2O
 pressure, and to be completely open at -2 in. H2O pressure.  The interceptor-
 side orifice and connecting line sizes were selected to be the same as for the
 orifice version,  permitting field interchangeability, if desired.

 Bingham St. Site

      This site was selected to demonstrate the  proportional, or analog mode
 of regulator operation, using the U-tube  sensor, as described in Section  1.
The multiple dip tube  sensor would be located in the interceptor, on the North-
east side of Tacony Creek, a distance of about  400 ft. from the regulator loca-
tion.  The U-tube sensor would be located in a  man-hole about 30 ft.  from the
regulator. The U-tube sensor fluid would be ethylene glycol (auto antifreeze)

                                    23

-------
                            Diaphragm Valve
Combined
Sewer Flow
      Fluidic Regulator
                                            Ambient Air
                                              Entry
                                                                 Dip
                                                                 Tube
                                                                 Level
                                                                 Sensor
Outfall Discharge
                                Interceptor
                                Discharge
                               Interceptor Sewer
              Figure 8    System Schematic Arrangement using
                          Diaphragm Valve
                                   24

-------
 on account of its low freezing temperature, low volatility, general chemical
 inertness, and general availability.

       Using Bernoulli's equation, the maximum suction at either control port,
 at maximum upstream hydraulic head, was computed to be about -12 in. H2O.
 This value would establish the minimum vertical dimension of the  dip tube
 sensors and U-tube, in order to prevent either sewage, or U-tube fluid from
 entering the connecting air lines.  It was found, however, that this value is
 very dependent on nozzle discharge coefficient, so that for operational margin,
" a value of 18 in. was  selected in the actual design.  In the event that the -12
 in. H2O value proved accurate, the dip tubes could be easily shortened to the
 correct lengths, whereas the reverse situation would require major rework of
 the whole U-tube assembly.

       Figure 70 from Ref. #1 was used to define the vertical locations of the
 dip tube ends, so that linear flow diversion with interceptor level change
 would be achieved.  This analytical procedure is described in Appendix A.
 Also included in the final design analysis was the effect of the higher density
 of ethylene glycol on dip tube length, and adjustment in dip tube length to
 compensate for the effect of the fluid volume contained in the dip tubes as
 they empty successively with falling U-tube fluid level.  This effect was mini-
 mized by selecting the section area of the U-tube large in comparison with that
 of.each dip tube.

       A connecting air line inside diameter of 4 in. was selected using Figure
 13 of Ref. #1.  Each dip tube inside diameter was set at 1 in., while an inside
 diameter of 12 in. was selected for the U-tube.
                                     25

-------
                               SECTION 7
                      DESIGN AND CONSTRUCTION
GENERAL

      This section describes the specific approaches used in the design and
construction of the fluidic regulator elements, their appurtenances,  and the
overall sites,  for both the 67th/Callowhill and Bingham St. installations .

DESIGN

67th/Callowhill Site

      The design objectives for this installation were to replace a small con-
ventional (1.05 cfs) static slot (or leaping weir) regulator with a f luidic type,
obtaining the much improved hydraulic performance of a dynamic type regulator,
with minimum reconstruction costs.

Regulator—
      The selected construction material was reinforced concrete.  The unit
was poured around a male core, contoured to the inside flow passage geometry.
Hoisting hooks and studs for attachment of the regulator cover and the securing
of the regulator to the upstream dam were molded into the concrete.  The regula-
tor cover was designed from transparent plastic, as an aid to demonstrating the
f luidic regulation action.  The regulator was designed so that interceptor flow
would be discharged through a drop opening  in the regulator base, while outfall
flow would be discharged over an elevated weir and through a shroud construc-
ted as part of regulator cover. Air and water seals were accomplished by a
foam rubber gasket between the cover and regulator body.  The outline of the
fluidic element shape was cut out of the gasket to permit viewing of the inter-
nal flow.  Control port connections were made by NPT (?) bosses in the plastic
cover. The regulator is shown before installation in Figure 9.

Sensors—
      The dip tube sensor was designed for construction from standard PVC
Schedule 40 pipe and fittings . Orifices were designed as  1/8 in. sheet PVC
discs, cemented into the control port fittings .  The diaphragm valve was de-
signed for construction from PVC shells, with a neoprene sheet diaphragm
                                    26

-------
 CD
 iJ
 CO
 M
 •)->
 CO
 10
co
 to
 0
 CD
CO
                                                                                                 \
0}
                                                     27

-------
clamped between. The valve was assembled using aluminum screws. The
valve spring was spring steel wire, protected from corrosion by rust-resistant
enamel. The valve flap was a PVC sheet disc, cemented to the diaphragm.

Site—•
      A comparison of the site before and after modification is shown in Figure
10.  As shown, the modification consisted of the removal of the curved  plate,
and minor excavation in the invert and crown of the sewer at the slot location,
for the installation of the fluidic regulator element and dam. The level sensor
was  located in an upstream manhole 75 ft.  from the regulator. The sensor was
protected from flow-borne debris by an aluminum deflector.  PE pipe,  1-1/2 in.
in diameter was  mounted .on the sewer crown to provide pneumatic communica-
tion between the sensor and regulator.  The diaphragm-valve was mounted on
the manhole wall, over the regulator.  Site design drawings for this installation
are shown  in Figures 11 and 12.

Binaham St. Site

      The regulator site at Bingham Street and Tacony Creek dictated a different
design  approach.  The larger size of the regulator element,  (DWF = 4 cf s)
necessitated the construction and hydraulic testing of the regulator on-site.

Regulator—
      The fluidic regulator element was designed for construction of reinforced
concrete, with a stainless steel splitter nosing. A 3/8 in.  thick stainless
steel plate would form the element cover, and would contain the connections
for the  sensor connecting air lines .  It would be placed as a unit below the in-
vert of  the outfall, and a concrete dam would be constructed above it.  A
number of possible locations were studied for the installation of the  regulator.
The  existing outfall structure is comprised of unreinforced concrete arch sec-
tions 7 ft. high  by 10 ft. 3 in. wide, approximately 15 ft. below the existing
ground  line.  Important considerations were the depth and volume of  excavation,
and the potential  structural problems that might result from cutting into the ex-
isting concrete  sections.  The three most promising approaches were:

         1.  The removal of the existing dam, and the reconstruction of a 20 ft.
 section of the outfall, with the regulator element located beneath a new, higher
 dam, and the sewer crown excavated to provide sufficient flow area over the
 dam for heavy storm flows .                                ;

         2.  The location of the regulator element below the existing dam, and
 below the  existing sewer invert. The regulator outfall discharge would flow
 into a closed parallel pipe below the existing sewer structure, which would re-
 enter the sewer near its  opening into Tacony Creek.

         3 .  The location of the regulator element as in 2., but with the regu-
                                     28

-------
24 in . Pipe — -v





Existing Slot Regulator
Before Modification

\

> MB^fe

.' •;.*.'•'.




1=
•I'.;,' '.


< '>• .'
» i
«•'',



-— V
9
t>
m


V/-/.,'r
. * A »
; • c
' *'/ ;
*. • frv
, '..'
, * •
« • »
• • * ' .
•(»>,, >
 i •



•''i i *?

S^

\ '
'-;.»



ES
\
-\N
•Jl

f t
\s-


\ »*. •
^
»t 1
V '
1 " <
•. 1
• *
• ,t.
'1 *
• >
r .

•.

4
*
- f
't>
* '
, ',
••
}>:
t
»
1
?W.T


••' v
'"* >


\s.

.I*, •'.'•! ''-'J
^
• «
&« ' t •
• .'» >,'"



• 10 ir
                                                              -Overflow
                 Dam
After Modification with
F_lui die Regulator

          Figure 10   67th/Callowhill Regulator Site,
                      Before and After Modification
                                                             Shroud
                                                               Overflow
                                                         10 in. Pipe
                                   29

-------
              X  fcg ^-£x.Inft.rctpttna Manhol
              . X.~ 5 TO BE MODIFIECT*><-_
                             PLAN

                             SCALE ' t"«30*-O"
ISO


ts'tr





'


pi . H4 5-j N?

[!


rt« S«u«^







-^





3 ff.





1
J











,

L-^J
v^!


	


SI'
fir.'






S"pu
.«


--


D,'«
,et





	
EL.jM.ja-
Vr-EI.Y 142,93


	 	


:pr^'*' •*


                            PROFILE
                              SMI.E
                              HOR. I'-M'-tf1
                              VERT. l"'S'-0"
        Figure 11    Site Design, 67th/Callowhill Sts .
                                   30

-------
OROUT IN HEGULATINO CLEMENT  	
      WITH MASTIC
li".OIA. P.E. PIPE
CONTROL LIKE—\
                                                                                                                    AM) GROUT WITH MASTIC
                                                                                                                    ....
                                                                                . ff.C. Pff*—^.    ~~

                                                                    DISH  INVERT TO MATCH REG. OPENING

                                                                    CLASS 33-1 CONC.
                                                                    0AM (SEE DETAIL)

                                                                    ASPIRATOR

                                                                    DIAPHRAGM VALVE            MODIFICATION
                                               l»c.
                                          WAV  EX,ST.
                                          SCALCi V4"« f-0"
                                                               • IWtf/MM'A*
                                                               *C Pr«
      r*/**^ •«' f if,'*,
      .n**

MANHOLE
                     MODIFICATION OF EXIST. INTERCEPTING MANHOLE
                                        PLAN
                                        SCALE i 3/4". I'-O"
    FLUIDIC
 REGULATOR!
    INSERT:
                                                                                 £LASS 35-1 CONC.->
                                                                                  CONTROL LINES]
                                                                               SHALL K INSTALLED
                                                                                SO THAT THE AREA
                                                                                  ABOVE THE O/
                                                                                  SHALL BE LEA!
                                                                                           -
                                                                                                                           O WIRE FABRIC
                                                                                                                             ALL AROUND
                                                                                                                       CENTER OF SLAB
                                                           •CONST. JOINT
                                                                                                SECTION D-.D
                                                                                                SCALE1 3/4"«l'-O"
                   I
                     SECTION   B-B
                     SCALE) 3/4"».l'-0!"
         ' DIA. RE. PIPE-



               DIP TUBE DETAIL
                     H.T.3.
                                                                                       fEMALE ELIOW
                                                                                         " OU. RE. PIPE
                                                                                                        END DETAIL
                                                                                                            H.T.8.
                                                                        CUT BRICK 8 CONC.
                                                                         AS SHOWN WITH
                                                                       FILLETS EACH SIDE
                                                                       OF TROUGH. FINISH.
                                                                         TO BE SMOOTH.
                              SECTION   C-C
                              SCALE- 3/4"ii'-0"
                                                                                                                    	DEFLECTOR-2l"LG.
                                                                                                                    4i4>* ALUM. ANGLE
                                                                                                                      ~  TO CUT OUT WALL
                         SECTION   A-A
                         SCALE' 3/4"' f-0"
                                                                              DIA.ili" SLOTTED HOLE FO1
                                                                              :G,T" OIA. BOLT ANCHOR

                                                     DETAIL OF DAM PLATE
                       Figure  12     Site  Design,  67th/Callowhill  Sts.
                                                                31

-------
later outfall discharge flowing into an open channel, 18 in. wide, cut into the
sewer invert with a starting depth of about 1 ft. near the element, tapering to
zero at a point about 100 ft. downstream.

      After evaluating the above with respect to hydraulic performance, struc-
tural feasibility, and cost, it was decided to pursue approach #1.  Site design
drawings are shown in Figures 13 through 16'.

Sens or-to -Regulator Communic ation—
      At this site, the distance between the interceptor sewer dip tube sensor
and the regulator element was about 400 ft.  Based on the control line size
criteria,  page 26, Ref. #1, a line inside  diameter of 4 in.  would be required.
Alternatives considered were pneumatic-mechanical, and electro-mechanical
systems.  In view of their greater complexity, the probable need for an auxiliary
energy source, and the desirability of demonstrating the "no-moving-parts"
fluidic approach, these alternatives were rejected.

      To save additional excavation and construction costs , it was decided to
locate the 4 in. control line inside the 24 in. connecting sewer.  The selected
control line material was PE. The line was to be secured in the manholes at
each end of the connecting sewer,  and allowed to float free in between, thus
eliminating the need to break into the connecting sewer to provide attachment:.

Sensors—
      The interceptor sewer dip tube  sensor was designed for construction from
a section of 4 in. PVC Schedule 40 pipe, with 1 in. PVC pipe dip tubes thread-
ed into the lower side.  The U-tube sensor was designed for construction of
sections of 12 in. diameter PVC pipe sections, welded to end pieces of 1/4
in. PVC sheet.  The dip tubes would be varying lengths  of 1 in. PVC pipe.

CONSTRUCTION   .,

Binqham St. Site

      Figure  17 shows this site under construction. The completed dam in-
stallation is shown in Figure 18. The interceptor and U-tube sensors are
shown in Figures 19 and 20.

      When construction was completed,  the existing manual gate upstream
from the dam was closed and dry weather flow was directed through the fluidic
regulator.  Controls were set for maximum flow diversion to the interceptor,
pending the field calibration phase .

67th/Callowhill Site

      Construction for the 67th/Callowhill site required only 7 working days.

                                    32

-------
                                                   CO

                                                   a
                                                   (0
                                                   -S
                                                   CQ
                                                   w


                                                  Q



                                                  5
                                                  CO




                                                  CO
33

-------
                                                                   Figure  14     Site Design,  Bingham  St.
                                                                                     S/ri 4'-0" dia, brick or cone,
                                                                                     manhole w/2'-oMdia, std.
                                                                                     C.I. cover B frame
                                                                                 -#6hoops

                                                                                 'dig. P.E.I.,
                                                                                 \terceptor j
                                                                                  If clamps
                                                                                 Tube-Sensor furnished
                                                                                  Wottr Dipt, ttt
                                                                                  ~.is! SptcMcattont
                                                                                Class 35-1 cone.
                                                                                                         9-$£
-------
                        . 4" DIA. R E. PIPE INSIfE
                      EX.24" DIA VIT. PIPE SEWER  '
 SELECTED CLEAN
. EARTH FILL—;
                     P.E. PIPES
                     |

                   C    24" DIA. VIT.
                  \%l X""PIPE SEWER
                                     —i-TO TAGONY INTER.

                                 JUNCTION MANHOLE
                                    SEE DETAIL
                                    SHEET NO. 3
                           SEWER

                            20-1 CONC.

                     VHIYMENT LINES
                         OR EXCAVATION
   24"  DIA. VIT. PIPE SEWER'
   WITH 4" DIA. P. E. PIPES
                "'"
ENLARGED PLAN
      SCALE: I"=IO'
                  FLUIDIC REGULATOR CHAMBER
                               PLAN
                                                                     .  TO BE REMOVE 3-
                                                                    'ERT GROUTED SMOOTH
                                                            FORMED INVERT TO NOZZLE -
                                                            CLASS 35-1 CONC.
                             SECTION A-A
                              SCALE:-J"ei'-o"
                                                      . RE. PIPE
                                                   FREE FLOATING
                                  SECTION  B-B
                                    SCALE: i"i i'-o"
                         Figure  15     Site Design, Bingham St,
                                         35

-------
f Sionittt iiiti pio>i
           ~~
licit to aiait fitn
J  h'Ja SS lack     f—>
#Jir>rr s Hjt	•-•—v   I
                                  i—Debris defector
                                    a,£*3/e stainless steel anale
                                    4 -O Ig, anchor -to MH wall wifh
                                    3-VK" did. S. 5. bolts
                                    H/screw oncho'S
                                  ,• -Dip tube level sensor
                                    PVCpipe
                                    Furnished by Water Oepl.
                                   Q see Special Specifications
                                                    .
                                                 . 24"d,'o. C.L
                                                ipe sewer
                                            'dia. P. f. pipe
                                          'from junction chamber
                           I   |   Ex. 4'-O"dia. brick sewer
                             **   facony Creek Interceptor
     tfiffi cost iron manhole
     frame and cover
     Clomp to wall with
     Gal, W, I wal/ type clomp

     Cip ?uh? level sensfv
         i   I
                PLAN
                                                                               SECTION  B-B
     IJ/4 " vitrified
     Jnvp' ' lining nor to
     be ottered in any way
NOTE- THE AS-BUILT PLAN,
DATED 4-1-'2l, INDICATES 1 HL
4'-O" D1A. BRICK SEWER IN
ROCK SECTION AS SHOWN
          NEW  MANHOLE ON  TACONY CREEK  INTERCEPTING  SEWER
                                                                                             Ex. 24" dia. C.I.
                                                                                             pipe sewer
                                                                                             LB.Kl.-f44.IO
                                                         SECTION  A-A
                                                            SCALE: £"= l'-0"
                              Figure  16     Site Design,  Bingham  St,
                                                       36

-------
                                                        I
                                                        (U
                                                        e:


                                                        18
                                                        CQ
                                                        I
37

-------
38

-------
Figure 19   Bingham St. Fluidic Regulator
            Interceptor Sensor
                                 39

-------
                                                      '!'	!|H|! ' I'l
                                                 iiiiliiiii	   : :
Figure 20   Bingham St. Fluidic Regulator
            U-Tube Sensor
                                   40

-------
This was attributed both to the smaller size of this regulator, and the fact
that it was prefabricated off-site, and required only insertion into place, The
existing brick manhole for the slot-type regulator,  as shown in Figure 21, was
replaced with a precast unit which also speeded construction. Completion of
the construction contract for both sites was accomplished by July 15,  1973 ,
well within the specified  70 working days.  The completed regulator install-
ation is shown in Figures  22 and 23.  The completed dip tube sensor installa-
tion is shown in Figure 24.
                                     41

-------
42

-------
-

'2>
 ^
  Figure 22   Fluidic Regulator Installation,

              67th/Callowhill Sts .
                        43

-------
Figure 23
Fluidic Regulator Installation,
67th/Callowhill Sts.
                   44

-------
Figure 24   Fluidic Regulator Dip Tube Sensor
            67th/Callowhill Sts.
                      45

-------
                                SECTION 8

                         REGULATOR CALIBRATION
67TH/CALLOWHILL SITE

      The calibration procedure for this regulator consisted of determining the
flow diversion ratios , relative to total regulator flow, for the outfall and inter-
ceptor-side discharges, as a function of upstream hydraulic head, for con-
ditions of either dip tube above water, or dip tube submerged.  Calibrations
were performed on this unit in the laboratory at Bowles Fluidics Corporation,
as well as at the field site.  Calibration details can be found in Appendix B.

Laboratory Calibration

      Two settings of outfall weir height were tested to establish the minimum
weir elevation for negligible trickle flow when maximum control was applied
toward the interceptor discharge.  Total regulator flow was established for the
calibration by the flow prediction criteria, page 20, Ref.  #1.  Outfall flows
were estimated using the flow continuity approach: Q=A x V, in which A =
flow cross-section area over discharge weir, and V= flow horizontal velocity
across the weir.  V was determined by measuring the horizontal  "jump" distance
of the discharge flow stream, while dropping a reference vertical distance,
assuming free-fall conditions, and neglecting aerodynamic drag on the flow
stream.  The interceptor discharge flow was computed as.  the difference be-
tween the total and outfall flows.

      Initial runs were made using the simple orifice arrangement of Figure 1.
Unacceptably large trickle flows occurred over the outfall discharge when
maximum control was applied toward the interceptor discharge.  The drop
opening area was enlarged, relieving the problem considerably.  It was found,
however, that trickle flows could only be reduced to a "negligible" level
(negligible was established as occurring less than 20% of the time, corres-
ponding to about 0.1% of the total flow) if aspirated airflow through the
orifice were cut off completely.  Since this condition could be obtained using
a diaphragm valve, the orifice configuration was discarded, and the balance
of the runs were made with one control port blocked, the other fully open,
depending on the direction of applied control. At the same time, development
was started on a diaphragm valve.

                                     46

-------
       Upon the availability of a diaphragm valve, the regulator calibration was
 repeated, using the valve. The results of both calibrations are shown in Figure
 25, together with the predicted values from the performance analysis .  The
 calibration values show a diversion margin over the predicted characteristic
 for all but the lowest values of upstream hydraulic head, thus indicating accep-
 table performance.

 Site Calibrations

       Site calibrations  were started in July, 1973, and completed in October,
 1973 . Storm flow was  simulated by opening fire hydrants upstream from the
 diversion site. Total regulator flow was established using the flow vs. head
 criteria of Ref. #1 and the sum of the measured flows from interceptor and out-
 fall discharges . These were determined using 90° V-notch weirs .  Early cali-
 bration attempts indicated low values  of outfall discharge flow rate, in com-
 parison to the laboratory calibrations .  It was found that insufficient time was
 being allowed in taking these readings, for the water level in a large tank up-
 stream of the  V-notch weir to stabilize, particularly at the high head readings .
 The flow measurement equipment for interceptor flow, used in the Microstrainer
 Evaluation Program (Ref. #3), also located at this site,  proved inoperative,
 necessitating the  construction of the second V-notch weir.

      The final calibration results are shown on Figure 26.  The total flow
 curves using the Ref. #1 criteria and the sum of the  discharge measurements
 agree well for all  but the highest head readings , again indicating that in-
 sufficient time was  being allowed before taking the outfall discharge
readings.  Diversion performance curves were  computed using:  a) the
 measured values of outfall discharge flows (curve D), and derived  values of
 outfall discharge flow obtained by subtracting measured interceptor flow from
the total flow prediction (curve D')- Also shown on Figure 26 is. the Ref. #1
 prediction of D.  The measured value curve agrees well with the predicted
values in the  middle range of hydraulic head, but drops  slightly lower at the
high values of head.  The  derived value curve exceeds the predicted value
for all heads above 7 in.  Both curves fall below the predicted range at low
values of head, as occurred during the laboratory calibrations.

      The general performance characteristics,  as shown by this calibration,
were considered satisfactory, and the  unit was  judged ready for active service .

BINGHAM ST. SITE

      The planned calibration procedure for this regulator was, as follows:  A
series of calibration runs Would be made with the ends of the interceptor dip
tubes successively plugged, simulating  a rising water level in the  interceptor.
Each run would be made by temporarily damming the regulator inlet, and allow-
ing the dry weather flow to build up behind the dam until the crest was reached.
                                     47

-------
                      Figure 25
                             Laboratory Calibrations

                             6 7th/C allowhill Re gulator
   1.4-
                   Qoutmax '  3-6-72 Calibration

                            ,  6-1-72

                      D     ,  3-6-72      "

                      D     ,  6-1-72
   1.2-
   1.0--100
en
•8
.84-80
•3  .6--  60
 G)
    .4
     40
     • 2--20
         (0
         B
         O
              (D

              1
          •8
          4-1
         O1

         •8
Q)
a
0
PU
S
1
         2
         ii
         Q
          T
                           T
                                           r
                          4          6


                   h = Upstream Head, inches
                                                     1

                                                     8
T-

 10
                                   48

-------
1.4
                 Figure 26   Site Calibration,
                            67th/Callowhill Regulator
                            Qtof fcef. #1)
                                                Qtot (Measured)
                   D,  (Ref-.#l Prediction)
                   h = Upstream Head, feet
                               49

-------
At this time, the inlet was quickly unblocked, providing a calibration run with
the hydraulic head decaying from 3 .5 ft. to zero over a period of about 3 min-
utes , after which normal dry weather flow conditions resumed.  Flow measure-
ments and hydraulic head readings would be taken at intervals during the run.
Total regulator flow would be established using the Ref. #1 flow-head criteria.
Outfall discharge flow would be measured using the continuity principle, in
which flow rate = outfall flow channel width x flow depth x flow velocity.
Flow velocity was obtained using a rotating cup instrument, and also pitot
head measurements. The initial runs were made with all the interceptor  .dip
tubes plugged, to determine the maximum  diversion capability.  Calibration
details can be found in Appendix B.

      The first attempt was unsuccessful in that hydraulic head could not be
built up behind the dam when the inlet was blocked. It was found that the
manual sluice gate structure in the by-pass channel had become damaged from
the effects  of acid in the sewage flow, and was not retaining the flow.  The
structure was repaired and another attempt was made.
                                                                        A
      In this attempt, hydraulic supply head was obtained in the desired
manner, but the diversion performance was well below the predicted value.
large piece of concrete was found in the bottom of the outfall channel that
could not be removed at that time, so maximum diversion runs were postponed
pending its removal. A run was made with full control toward the interceptor,
to evaluate the trickle flow over the outfall discharge.  On this run, the trickle
flow was negligible for a period of 15-20 seconds , after which significant flow
occurred.

      Of interest from this calibration attempt are the supply head time histories
of Runs 1 and 2, as shown on Figure 27. Each history shows an abrupt 3:1 de-
crease in slope in the region of 15-20 seconds after the start of the run, fol-
lowed by a gradual increase in slope.  The gradual increase in slope would be
expected as the horizontal dimensions of the impounded water decrease;
however, the abrupt decrease in slope indicates a 3:1 decrease in regulator
flow rate, presumably caused by a large increase in flow impedance.  The
piece of concrete found in the regulator did not offer an explanation, since it
could have affected only Run 1, even assuming that it entered the regulator
15 seconds  after the run start. The sudden appearance of outfall flow on Run
3  also occurred at about 15 seconds after the run start, and was apparently
associated with the change in regulator flow.

      On the next attempt, with the piece of-concrete removed, the diversion
performance was improved, but still unsatisfactory, with about 22% of the to-
tal flow being diverted to the outfall compared to an expected value of about
50%. The total flow is based on the Ref. #1 criteria and the measured supply
head. Readings were made at approximately the maximum supply head only,
because of the problem encountered during the previous calibration attempt.
                                     50

-------
                                                       •K
                                                       <0
                                                       a
                                                       c
                                                        CD
                                                       4!
                                                        (0
                                                        ra
                                                       o
                                                       o
                                                       0)
                                                       03
                                                       CD
                                                       S
                                               If)
UIBQ MOjeq saqouj '
                 51

-------
Measurements were also taken of control port pressures, and the outfall-side
showed an abrupt change from -12 in. H2O to -4 in. H2O after 15-20 seconds
of the run, indicating that the flow impedance problem was still present.  How-
ever, the diversion performance before the impedance change was sufficiently
below expectations that it was decided that model tests of the complete regu-
lator configuration should be made to determine the revisions necessary to
obtain the desired diversion performance.

Model Test Program

     A 1/6 scale model of the Bingham St. installation was  constructed and
tested,  and is shown in Figure 28.  The model was constructed with a remov-
able outfall discharge section,  since it was felt that this was the area requir-
ing modification.  The diversion performance of the model for the existing con-
figuration approximated the observed prototype performance, giving about 35%
diversion at the scaled maximum hydraulic head. The model test results for
the prototype, and three possible modifications are shown in Figure 29 .  Accord-
ingly, the model outfall discharge was widened from 3  to 7 in. at the weir crest,
maintaining the  discharge ramp invert profile.  Tests of the revised model con-
figurations showed close agreement between the model diversion performance
and the predicted values of the performance analysis .  The selected modified
configuration (model modification #2) is shown on Figure  30 . During the period
of revision of the actual installation, a detailed inspection was made of the
regulator, and a condition of erosion noted on the inlet nozzle walls . The con-
dition was not considered immediately serious, but indicated the desirability
of armoring this area on future installations, particularly where the flow veloc-
ity exceeded about 12 fps .

      A subsequent calibration attempt involved several runs in which diversion
performance was close to predicted values for a few seconds after the run start,
dropping  to much lower values after about 15 seconds . The outfall-side control
port pressures showed a corresponding change from -15 in. H2O to -5 in.  H2O.
On succeeding runs, an observer was placed in the special junction manhole.
In each case, the water level was seen to rise to a level about 2 ft. above the
connecting  sewer inlet after about 15-20 seconds after the run start.  This
condition acted as a hydraulic overload on the regulator, incapacitating its
regulation performance, and confirmed the flow impedance change suspected
from previous calibration attempts.  Further runs were  postponed pending a
careful inspection of the conditions in the connecting  sewer which might cause
the observed flow impedance change.  The inspection was performed, result-
ing in removal of some debris from the connecting sewer between the inlet and
 a manhole located on an island in Tacony Creek. This operation was signifi-
 cantly  impeded by the 4 in.  diameter air line.

      On the next calibration attempt, representative diversion performance
 was shown for a short period of time at the beginning of each run, until the
                                     52

-------
3* TWMLrtia^        '
               ,   *
                           53

-------
                 Figure 29   1/6 Scale Model Test Results
                -A—  Model Modification #1

                _v	    "        «        #2
                 yv
                         "        "        #3
   —  100
(0  r
1
O
o  I-   80
 o
 0)
6
I
1
 II
 P
    -   60
    -  40
       20
                    Ref.#l
                    Prediction
                         1.0
                                                Scaled Maximum
                                                Available Head
                                                           Prototype as
                                                           Originally
                                                           Constructed
                                               2.0
                                                                     3.0
                                   54

-------
              Existing Outfall Contour
  PLAN
                                                          T
Modified Outfall
    Contour
ELEVATION
                                                Steel Shroud
          Dam
      Figure 30   Modified Bingham St. Regulator
                  Outfall Geometry
                          55

-------
water backed up at the connecting sewer inlet.  It was concluded that a more
thorough investigation was necessary, that would include the section of the
sewer from the island manhole to the interceptor connection.  It was indicated
that a new inspection procedure, using a remotely controlled TV camera would
be employed.

TV Inspection

      In an attempt to determine the cause of the backup in the connecting
sewer,  several conventional techniques were tried initially.  The rodding
machine was tried first.  However, several attempts to run a rod from the
special junction manhole to the interceptor were unsuccessful.  An attempt
was then made to float a line through the sewer, since the normal  dry weather
flow was passing through unimpeded. This, too, was unsuccessful.

      Next a high pressure sewer jet machine was used, which apparently
cleared the blockage.  However, attempts to calibrate the regulator showed
that the flow impedance problem, while  reduced, had not been solved. At
this point, the decision was reached to  perform a complete detailed inspection
of the connecting sewer.

      A 24 in. air bag plug was inserted in the mouth of the fluidic regulator
and floated through to the leg to the interceptor. The bag was then inflated,
and the entire dry weather flow diverted over the outfall.  However when the
manhole at the junction of the connecting and interceptor sewers was inspected,
it was found that the connecting sewer was running half full.

      A TV camera inspection of the connecting  sewer was the next step.  With
great difficulty, a line was run through the sewer using the sewer jet machine.
The line was attached to the TV camera cable, which was then pulled back to
the special junction manhole. The TV camera inspection showed several dis-
tinct breaks in the concrete encased 24 in. vitrified pipe under the creek. It
also revealed  that the 4 in. air line was snaked inside the sewer, acting  as a
dam in  some places, and causing the difficulty  in rodding and running lines
through the sewer.  However, the polyethylene  pipe was in remarkably good
shape after the abuse it had received in the  various cleaning attempts.

      At this point it became obvious that any further calibration attempts
would be fruitless, with the observed level of infiltration. Also the repairs
to the sewer would be difficult and expensive, due to its location at the base
of a steep embankment and under a creek. A decision was made to suspend
further calibration attempts, and summarize  this phase of the program.
                                    56

-------
                                      SECTION 9

                                PROGRAM EVALUATION
      GENERAL

            This section describes the evaluation of the fluidic regulator concept,
      based on the results of the subject program.  The areas to be covered include
      the hydraulic performance in service, maintenance performance in service,  and
      a direct comparison between the fluidic regulators and their conventional static
      counterparts from the standpoints of hydraulic performance and costs.

      HYDRAULIC PERFORMANCE IN SERVICE

      67th/Callowhill Regulator

            This unit was placed in normal municipal service as of November, 1972.
      Because of its small size and close proximity to the Microstrainer Evaluation
      Project (Ref.  #3), specific flow recording equipment was not installed, and
      thus a quantitative evaluation of regulator hydraulic performance during storm
      events from November, 1972 to December,  1975, was not obtained.   However,
      the site was inspected frequently during this  period for demonstration purposes,
      and storm events were simulated using upstream street hydrants, as during
      calibrations . The demonstrations were given to representatives from the City
      of Philadelphia, the State of Pennsylvania, the  USEPA, other municipalities
      and jurisdictions , as well as consultant firms and others .  On all occasions
      the unit performed properly.  From this experience, it was concluded that the
      unit had performed normally during its service period, except for the  very few
      occasions of blockages .

      Bingham Street Regulator

            Because of the flow impedance problem described in the  previous section,
      this unit was not placed in municipal service, except as a conventional static
      regulator.

      MAINTENANCE PERFORMANCE IN SERVICE

      67th/Gallowhill Regulator

                                          57
_

-------
      After being placed in service, this regulator site was given the same
routine inspection and maintenance procedures as for other small static regula-
tor sites in the near vicinity.  Figures 31 and 32 show pages from the Phila-
delphia Water Department fiscal '75 and '76 Interceptor Maintenance Logs,
which describe the number of inspection visits, blockages or other malfunctions
detected, and corrective action taken for a number of nearby static regulator
sites. These sites are located along Cobbs Creek, and are also of the slot
type which the fluidic regulator replaced. As can be seen from the reported 3
and  1 blockages for fiscal '75 and '76,  respectively, the maintenance record
of the fluidic regulator is typical of nearby static units in terms of surveillance
required, number of blockages reported, and the time and effort required to re-
store the unit to normal service.  It was reported that the inspection procedure
developed for the fluidic unit was actually simpler and faster to perform, in
that blockages could  be detected by visual means, without the need  of probing,
or "rodding" as for the slot type units .  It was regarded that the fluidic regula-
tor,  from the standpoints  of surveillance and maintenance, was equivalent or
superior to the conventional slot regulator in the small size range typical of the
Cobbs Creek area, and that this comparison would likewise exist in  the larger
sizes .  It was felt that the fluidic regulator surveillance and maintenance
characteristics would be clearly superior to any conventional type of dynamic
regulator involving electro-hydraulic-mechanical elements .

      One significant mishap to the fluidic regulator occurred on October 19,
1974, when, through  vandalism,  the manhole cover was removed, the diaphragm
valve stolen, and the plexiglas cover of the  regulator cracked by dropping rocks
on it. A new valve was supplied, the cover repaired,  and the unit was restored
to service in about 1  week. It should be noted that the transparent plastic
cover was chosen to assist in the demonstration nature of the particular unit,
and that a normal construction approach using a heavy gage plastic or metal
shell sheathed in concrete would probably have survived the rocks without
significant damage.

Bingham Street Regulator

       Since the Bingham Street fluidic regulator was not placed in service, ex-
cept as a static unit, the official surveillance and maintenance records are less
complete than for the other unit. However, the fiscal175 Interceptor Maintenance
Log does show a total of  9 surveillance visits to the site with no problems re-
ported except for minor vandalism at  one manhole.  This regulator functioned as
a static  regulator from June, 1973 to the present, and  the only observed block-
age was the large piece of concrete that interfered with the second calibration
attempt. In the opinion of the Philadelphia Water Department, this particular
blockage could have  caused extensive  mechanical damage to an equivalent
hydraulic or float-operated dynamic regulator, necessitating considerable
 "down" time and expenditure for the replacement of parts.  In contrast, the
fluidic regulator suffered no damage, and was immediately restored to service

                                      58

-------
  -  i
•»? «
 ' * I
                                       •5*
                                    •  1
                                    .«  -p
*s
i 1
si
iJ
a 5i?
r«l
I ir
« -g s
g M
* fe «
                                                                 0)
oo
e
3
                                 59

-------
                                                                                                               a



                                                                                                           (0
                                                                                                           8.3
        O • «> »H
               S5°£   x
v    w

35, g  8£ 3
        "-«Q.d—• O U 1
S  S «  ** ~S
»*  Q. «  ..^j
                                     .     .
                                 «J  %> O  U
                                 * ** §
                                      '
                                                             O to  ** O  1-1 J*
                                              o *-. ~ * D-. -H
                                                               g  i
£  °  «

O\  .0 O
vo

*»  ^  W
                                                                                   O J3-»-* - o a 19 «

Sji'gTet;
*-,« u » to rta
  *a to u  iifc
  E   C ,C •-*•
    8—* a s>. «i!
    -1 2-» 4.Q
                                                                                           -
                                                                                          » o  ^ -* -«*ia
                                                                                                            0
                                                                                                                   PM
               "-.a e   -H
               i a o*»   oa
                                                                                         K oo o-e
                                                                                        ) a »•*» -»
                                                                                         e u-^
                                                                                        *
                                                                                                           CO
                                           Cn
-*  Q **  O rt **  O
        O rt **  O
        -»*c»»'-*
               —«*• o
               » e w a
               i.- o e •
        Cut*  C -«
                 «S?S *  R
                                      S  C C  C
                                                   g  g
                                 COCO
                                                   Q.  O O,O  O. f
   tV «• 4.

   Q. CO -»
                                                             "if  »;•»*•«*
                                                                                             «
                                                                                      u I- ^ r^-M t
  t « I ^ -HO
  o « — tl »-.
 - Aj •*• fVj ~  *M
  f   " 5 tf»
                                                        60

-------
 upon removal of the blockage.

 GREASE BUILDUP

      In the initial presentations of the fluidic regulator concept to the USEPA,
 the Philadelphia Water Department, and other agencies, concern was express-
 ed as to the effect of grease, normally occurring in combined sewers,  on
 fluidic regulator operation.  It appeared that the semi-cylindrical control
 "pockets" beneath the control ports would tend to accumulate grease,  eventu-
 ally blocking the control port.  No grease buildup was evident in the 67th/
 Callowhill unit, based on routine  surveillance inspections.  Such a buildup,
 if present, would have been easily visible through the transparent plastic cover.
 The Bingham Street unit received similar routine inspections, and was  also ex-
 amined in greater detail at the time of each calibration attempt and during the
 outfall discharge modification.  These examinations showed an inconsequential
 grease deposit on the regulator inlet walls at the DWF water level; no  accumu-
 lation in the control port pockets .  Based on laboratory test experience, this
 accumulation would have a negligible effect on regulator performance.   Also,
 laboratory experience shows  a high level of vortex flow in the control port
 pockets, which apparently supplied sufficient scouring action to keep  these
 areas clean. Consequently,  no clean-up of the regulator was felt necessary
 nor performed.

      It is recognized that the above represents only a small sample of the
 necessary municipal experience to determine the total effect of grease  on
 fluidic regulator operation; however, the initial inputs, based on the subject
 program, are highly encouraging.

 COMPARISON WITH CONVENTIONAL STATIC REGULATORS

 Hydraulic Performance

 6 7th/Callowhill Regulator—
      A performance comparison between the original leaping weir, or "slot"
type regulator used at the 67th/Callowhill site is shown in Figure 33.  Regula-
tor flow to the interceptor is shown plotted against total combined sewer flow
 reaching the regulator.  The slot regulator flow curve is shown as a dashed
 line; the maximum and minimum fluidic flow curves are shown solid.  Based on
Philadelphia Water Department data, the slot regulator was assumed set so
that overflow to the outfall started at a combined flow of 0.9 cfs .  Extrapolat-
 ing the- Philadelphia Water Department flow calculations , the interceptor flow
 is found to reach a  maximum of about  1.5 cfs  at  a combined flow of about 6 cfs,
then decrease slightly as the sewer flow velocity increases .  This reduces the
transit time  and vertical drop of  the flow stream  as it jumps across the slot
opening, and thus the admitted flow.  These calculations assumed free-fall
flow conditions across the  opening, clean separation of the flow along the

                                    61

-------
3  -
               Figure 33   Hydraulic Performance Comparison
                           Fluidic vs . Conventional Static
                           Regulator, 67th/Callowhill Site
 1  'r
      w
      m
      o
      I
      I
      Q)
      (1)
                                          Irite1
iatot
                         V Dam Overflow Starts, Fluidic Regulator
                                        Conve-
                  Calibrated Drop Flow using 6 in. Opening
                        Start of Overflow, Slot Regulator
                          •x.
                                                         Flow
                         JL
                                        Fluidic Regulator
                                       	,r i	
                          234
                      Total Combined Sewer Flow,  cfs
                     .1
                      6
                                    62

-------
 sewer invert, and a uniform vertical flow velocity profile.  Calculation details
 are found in Appendix C .

       The advantages of the fluidic regulator are directly apparent.  Dam over-
 flow with the fluidic regulator does not occur until the combined flow reaches
 1.9 cfs ,  or more than twice the corresponding value for the slot regulator.  At
 this condition,  the slot regulator diverts 1.3 cfs to the interceptor and 0.6 cfs
 to the outfall, while the fluidic regulator can divert as much as  1.9 cfs to the
 interceptor and none to the outfall if the  interceptor has the additional flow
 capacity, or as little as 0.5 cfs to the interceptor and 1.4 cfs to the outfall
 if the interceptor is near capacity.  The fluidic regulator minimum interceptor
 flow is only 38% of the slot regulator interceptor flow, thus the tendency to
 surcharge the interceptor is much reduced.  It should be noted that this sur-
 charge relief actually improves as combined flow increases, due to the increase
 in fluidic regulator diversion capability with upstream head.

       The performance of the fluidic regulator fits well with the practical occur-
 rence of pollution loading  during a storm event.  The maximum pollution loading
 results typically from a combination of sanitary and "first flush" flows occurring
 during the early part  of a storm before maximum combined sewer and interceptor
 flow capacity have been reached.  The fluidic regulator would thus be able to
 divert much more of this badly polluted flow to the treatment plant at the time it
 occurs, with much less surcharging tendency to the interceptor when the maxi-
 mum (and generally less polluted) storm flows occur.

 Bingham Street Regulator—
      A similar  performance comparison between the conventional static regula-
 tor and fluidic regulator, for the Bingham St. site is shown in Figure 34.  In
 this case, the conventional regulator was a 2 ft.  high x 3 ft. wide sluice gate,
 located on the North  side of the combined sewer in the sewer sidewall, up-
 stream of a 2 ft. high dam  then located across the sewer invert.  The sluice
 gate was  retained as  part of the fluidic regulator installation, but is kept
 closed during normal operation. The conventional regulator performance is
 shown as the dashed curve, the fluidic regulator performance shown as solid
 curves,  in computing the conventional regulator performance, the  gate dis-
 charge coefficient was estimated at 0.7,  considering that the sewer invert is
 at nearly the same elevation as the invert of the gate opening. A gate opening
 height of  0.96 ft.  was computed to correspond to an interceptor flow of 20 cfs,
 for a water level at the original dam crest elevation.  The corresponding inter-
 ceptor flow is plotted against total combined sewer flow, together with the
 maximum arid minimum interceptor flow curves for the fluidic regulator. Al-
though the type  of conventional static regulator is different for this site  (orifice
vs . slot or leaping weir), the performance improvement of the fluidic over the
conventional regulator is similar:  Larger  combined sewer flow before dam over-
flow, and the option of either much more flow to the interceptor if it has
additional flow capacity, or a much reduced tendency to surcharge  the

                                     63

-------
               Figure 34   Hydraulic Performance Comparison
                          Fluidic vs .  Conventional Static
                          Regulator, BinghamSt. Site
30
 20
 10
                                               Sluice Gate Opening

                                                   .96 ft. x 3 ft.
Dam Overflow for Fluidic Regulator
                         Dam Overflow for Sluice Gate Regulator
                                           eptor Flow
20          40         60         80

      Total Combined Sewer Flow, cfs
                                                         100
                                          120
                                   64

-------
interceptor if it does not have additional flow capacity.
Cost

      This section will present the costs of the fluidic regulators studied under
this program in terms of initial costs, and surveillance,  maintenance and re-
pair costs .  Each category of cost will be compared with those of comparable,
conventional static regulators, that might have been retrofitted into the two
subject regulator sites in the same manner,  and at the same time as the fluidic
units .  In addition, some general cost comparisons will be made with conven-
tional dynamic regulator systems, whose dynamic performance is closely  mat-
ched by the fluidic regulator.

67th/Callowhill Installation—

      Initial Cost—Based on the Consultant's and Contractor's bid sheets, the
major cost breakdown for this  installation is shown below. Figures are round-
ed to  the nearest  $ 10.

                           Item

         1  Excavation

         2  Build dam, chamber, regulator mounting,
             install regulator and appurtenances            5,700
         3  Install control  lines                              500

         4  Backfill and  reseed                                150
         5.  Consultant,  including engineering, liaison,
             and furnished regulator hardware               4,180

         6  Philadelphia Water Dept. engineering costs,
             (est'd. 200 m-h @ $10/hr.)  .	2,000
        Total
$13,480
      The initial cost of retrofitting a comparable conventional static regulator
is estimated as follows:
         1 Excavation

         2 Build dam, chamber, date, and connecting
             sewer

         3 Backfill and reseed
         4 Procured Hardware
         5 Philadelphia Water Dept.  engineering costs,
             .(est'd. 150 m-h @ $10/hr.)
        Total
$   900


  4,500

    150
    500

  1,500

$ 7,550
                                   65

-------
      The conventional regulator cost is thus estimated at 56% of that for the
present fluidic unit. It is expected that the fluidic regulator consultant cost
would be reduced to about $2 ,000 if the same installation were made again,
based on learning and experience from the first. The conventional regulator
would then be about 67% of the fluidic unit cost.  It can be seen that on a new
installation, which would include the additional costs  of excavating and con-
structing additional manholes needed for either installation, the percent diffe-
rence in cost would be quite  small, probably on the order of 10%,.

      Surveillance. Maintenance and Repair Costs—Initially,  in the program it
was felt that the fluidic regulator at 67th/Callowhill would require inspection
on the average,  twice per week.  This would be approximately the same level
of inspection given a slot, or other static regulator. However, experience
proved that the interval could be extended to about once per week.  Also,  the
simplicity of the fluidic regulator, with its lack of electrical,  hydraulic, or
other complex hardware, allowed inspection by existing interceptor service
crews, which consist of one  semi-skilled worker and two laborers . These job
specifications require no technical training.

      Each inspection consisted of visual observation  for unimpeded flow
through the transparent regulator cover, and checking the dam, inlet, and dip
tube sensor for accumulated  debris .  It was noted that the most troublesome
type of debris, in keeping with the small 5 in. x 5 in.  inlet dimensions , was
beverage  cans and plastic bottles. The originally estimated time for an in-
spection was one hour.  However, field experience showed that the actual
time required was about fifteen minutes , except when a blockage was encoun-
tered, and then  the time was about 1/2 hour. Using 1976 average wage rates
for the interceptor crews, the surveillance costs  are estimated as follows:
         Interceptor Serviceman (semi-skilled);
           $5.21/hr- * 1/4 hr./visit x 60 visits/yr.

         Two Laborers
           $4.92/hr. x 1/4 hr./visit x 60 visits/yr.

         Total Annual Cost of Surveillance
$ 78.15
  73.80
$151.95
 Note that the corresponding cost for a conventional static slot regulator is
 estimated at 1.5 x this amount, or $227.93 .

       Normal maintenance costs are essentially zero, since it has a minimum
 of moving mechanical parts, and these are constructed from corrosion-free
 material.  The only required repair in approximately 4 years of service was
 occasioned by vandalism, rather than normal service. This incident required
 the following costs.
                                     66

-------
          Material and Procured Parts:
               Diaphragm Valve
               PVC pipe fittings
               Miscellaneous
                Total
         Labor, 8 hrs . @ $5/hr.
         Total
 Bingham Street Installation—
$150
   10
	5_
$165
  40
$205
      Initial Cost—Based on the Consultant's and Contractor's bid sheets, the
 major cost breakdown for this installation is shown below:
                             Item                           Cost
         1  Excavation                                    $14,000
         2  Construction of regulator chamber, regulator,
             regulator installation                          23,570
         3  Rework junction manhole                         6,200
         4  Const, new manhole, install dip tube sensor      2,240
         5  Install Interceptor Discharge line                   820
         6  Install 4  in. PE cont. line                        3,600
         7  Backfill and reseed                                 150
         8  Consultant,  incl. engineering design,liaison,
             furnished hardware                             9,000
         9  Philadelphia Water Dept. engineering costs
             (est'd.  300 m-h @ $10/hr.)	    3.000
         Total
           $62,580
The initial cost of retrofitting a comparable conventional static regulator is
estimated as follows:
         1  Excavation
         2  Build chamber, dam, gate
         3  Rework junction manhole
         4  Install Interceptor conn. line
         5  Backfill and reseed
           $14,000
            18,000
             6,200
               820
               150
                                   67

-------
         6 Philadelphia Water Dept. engineering costs
             (est'd. 150 m-h @ $10/hr.)                   $  1,500

         7 Procured hardware   	1*000
         Total
$41,670
The initial cost of retrofitting a comparable conventional static regulator is
thus estimated at 67% of the cost of the fluidic unit.  It is expected that the
above consultant and regulator construction costs would be reduced at least
$5,000, based on the use of a prefabricated PVC element shell, and the ex-
perience gained from the present unit.  The retrofitted conventional static regu-
lator cost would then be about 72% of the fluidic unit cost, and again for a new
installation, the percent difference between the two approaches would be small.

      Surveillance. Maintenance, and Repair Costs—The routine surveillance
inspection procedure for this type of regulator is basically the same as for the
smaller unit, and is performed by the same type of crew.  The normal number
of annual inspections would be reduced to about 30/year, due to the larger
size, and the corresponding lower susceptibility to blockages . Thus the annual
surveillance cost would be 1/2 of the 67th/Callowhill unit, or $76 vs . $ 114 for
a comparable conventional static regulator. Normal maintenance and repair
costs for this fluidic regulator would be very nominal, and probably consist
principally of periodic replacement of ethylene glycol in the UHzube sensor.
It is expected that an amount of $50/year would cover materials and labor costs
for the fluidic unit.  The principal expected maintenance and repair cost of the
conventional regulator would be the replacement of parts of the manually posi-
tioned sluice gate due to corrosion and damage from debris .  Assuming a 10
year operating life, a replacement cost of $1,000, and an annual, labor cost of
$80  (16 m-h @ $5/hr.), the maintenance and repair cost for the conventional
unit is estimated at $180/year.

General Cost Comparison, Fluidic vs. Conventional Dynamic Regulator
System —
      Because of the wide difference in conventional dynamic combined  sewer
regulator systems (electrically or hydraulically operated gates, float, or flaw-
operated gates, siphons, inflatable dams, etc.),  a detailed comparison with
each type is not possible within the scope of  the  subject project.  The previous
discussions have concerned the costs of retrofitting fluidic and conventional
static regulators into the two chosen sites. The principal cost items affected
by the type of regulator are the regulator equipment cost, and installation and
checkout costs.  For the fluidic regulators , the 1973 estimated consultant
costs (which include engineering installation liaison, and furnished fluidic
hardware and appurtenance costs) adjusted by the experience of the present
program, were $2,000 for a 1.5 cfs regulator, and about $12,000 for a 25 cfs
regulator. A rough comparison with a conventional dynamic (mechanical float-
operated gate) regulator is generated as follows .  During 1976, the Philadelphia
                                     68

-------
 Water Department purchased a regulator of this type for $ 13,000, designed to
 handle flows up to 4 cfs .

      The consultant cost, adjusted to 1976, for a 4 cfs fluidic regulator is
 estimated at $4,500,  so that the cost of the mechanical dynamic regulator
 equipment is approximately 3 times the cost of the fluidic regulator equipment.

      In the case of electrically, or hydraulically operated gate-type dynamic
 regulators, the equipment cost ratio is probably higher, since the costs of op-
 erating energy supply equipment such as  pumps, compressors, or batteries,
 must also be included. For example, the  1964 capital cost of 4 cfs, hydrauli-
 cally-driven gate regulators used by the City of New York was $25,000 (Ref.
 #4, page 49).  Estimating the 1976 cost at $30,000, the  dynamic/fluidic  equip-
 ment cost ratio would be about 6.5.

      The costs of site modification, and regulator installation and checkout
 costs in retrofitting a conventional dynamic regulator are highly dependent on
 the specific regulator type, and site characteristics, so generalizations are
 very  approximate at best.  In the case of the mechanical-float, electric,  or
 hydraulically-operated types of regulators, one or more additional flow
 chambers, or passages, are required,  whose walls and openings require rela-
 tively close dimensional tolerances in construction to facilitate the assembly,
 alignment and checkout of precision mechanical assemblies such as gate frames
 guides, linkages, cylinders, gear drives,  motors, etc.  The mechanical assem-
 bly of a fluidic regulator is much less  critical in comparison.  In addition, the
 electric and hydraulically driven gate types require the installation of operating
 energy facilities such as electric power lines, potable water lines,  or com-
 pressed air lines, plus emergency stand-by facilities .  With such complex
 equipment, the labor costs required for assembly, checkout and operational
 and maintenance personnel training are necessarily much greater than for the
fluidic equipment.  Accordingly, site modification and installation and checkout
costs for conventional dynamic regulators are estimated at least 25% to 50%
greater than for fluidic regulators ,  Thus a rough estimate for the overall con-
ventional dynamic/fluidic retrofit cost ratio would be at least 2:1.

      For the case of  comparing the overall costs of  conventional dynamic/
fluidic regulator systems in new locations,  the above discussion applies as
well. However, the overall cost ratio will be less,  since the basic cost of
excavation and construction common to both system approaches (manholes,
connecting sewers, outfalls, main chambers, etc.) usually exceeds equipment
and installation costs, although it is highly dependent on particular site con-
siderations .
                                    69

-------
                                SECTION 10

                         CONCLUDING COMMENTS
      It was concluded from the results of the calibrations of the two fluidic
regulator installations that the fluidic regulator design criteria of Ref. #1 are
basically sound and accurate for the range of inlet sizes covered by the two
specific units , since predicted performance was demonstrated by both units .
It was also concluded that the criteria could be extrapolated to units with flows
up to 50 cfs, with a high degree of confidence  in the predicted performance.
The model test, conducted on the Bingham Street configuration, showed the
need for refinements to the basic Ref. #1 criteria, in the definition of outfall
discharge width as a function of inlet nozzle aspect ratio.  It was concluded
that the 5 in.  x 5 in. inlet nozzle dimensions of the 67th/Callowhill regulator,
while workable, represented the minimum desirable size for fluidic regulators .

      From the maintenance standpoint, the basic simplicity and capability of
the fluidic regulator to remain free from the effects of debris and other contami-
nants in sewage flow was strongly indicated, since very few blockages occurr-
ed, and the grease buildup on the inside flow passages was minimal.  It was
found that routine surveillance procedures for the fluidic regulators could
actually be simpler and faster than those for conventional static regulators ,
with a resulting significant cost saving.  Surveillance and maintenance could
be performed with minimum skill level personnel.  The indicated cost of repairs
and parts replacement was less than for the conventional regulators , since  no
movable or correctable mechanical elements operate in the sewage flow. The
maintenance characteristics of the fluidic regulator were regarded by the Phila-
delphia Water Department maintenance officials as clearly superior to any con-
ventional dynamic regulators in service in the city.

      From the design improvement standpoint, it was concluded that the  con-
necting air lines between the  dip tube sensor and regulator should preferably
not be run through the connecting sewer;  but if absolutely necessary, should  be
secured to the sewer crown.  The location of these lines at the sides,  or in-
vert presents  large problems in providing normal sewer inspection, cleaning,
etc., and significantly increases the probability of blockages .  The fluidic
element section of the regulator should be constructed as a prefabricated shell
of heavy plastic,  or corrosion-resistant metal. This would eliminate the diffi-
culties of holding close tolerances in concrete construction, and the possibil-
ity of inlet nozzle  erosion, while also reducing project costs. Also shown
                                     70

-------
during the program were:

       1.   The desirability of installing manual valves in control port air lines
to facilitate regulator calibration or checkout;

      2 .   The desirability of investigating other communication techniques
for extended dip tube sensor-to-regulator distances; and

      3 .   The desirability of investigating techniques for computerized remote
command-control of regulator operation.

      From the preliminary planning standpoint,  it was concluded that a thor-
ough investigation should be made of the flow capacity of connecting sewers,
in selecting sites for retrofitting fluidic regulators, since this appears to be
the principal factor in limiting regulator size.
                                   71

-------
                             REFERENCES
1.   Bowles Engineering Corp., Silver Spring, MD,  "Design of a Combined
     Sewer Fluidic Regulator", US EPA Report No. 11020 DGZ 10/69 (DAST-13),
     (NTIS-PB 188 914), 1969.  137 pp.

2 .   Fair, G. M., Geyer, J. C., Okun, D. A.,  "Water Supply arid Waste-
     water Removal, Volume I", John Wiley & Sons, New York, NY, 1966.
     approx. 250 pp.

3 .   Maher, M. B., Crane Co., King of Prussia, PA,  "Microstraining and
     Disinfection of Combined Sewer Overflows—Phase III", Report No.
     EPA-670/2-74-049,  (NTIS-PB 235 77I/AS), 1974.   82pp.

4.   APWA, "Combined Sewer Regulator Overflow Facilities",  USEPA Report
     No. 11022 DMU 07/70,  1970.   138pp.
                            BIBLIOGRAPHY
1.    APWA, "Problems of Combined Sewer Facilities and Overflows—1967",
      USEPA Report No. 11020	12/67, (NTIS-PB 214 469),  1967.   189pp.

2.    APWA, "Combined Sewer Regulation and Management, A Manual of
      Practice", USEPA Report No. 11022 DMU 08/70,  (NTIS-PB 195 676),
      1970.  133 pp.
                                   72

-------
                               APPENDIX A

                   REGULATOR DESIGN CALCULATIONS


CALLOWHILL & 67TH ST. LOCATION

Maximum Storm Flow Conditions:

       Design Storm Flow, Qst=30 cfs.
       Flow Area above dam enlarged to 2.0 ft.2
       Assume discharge coefficient, CD = 0.85, since only 1/3 of
         opening perimeter has sharp edge.

Then required upstream head=(Qst/CD A)2 x 1/64.4 = 4.84 ft.

       This value is within the allowable surcharge of 5 ft.

Regulator Nozzle Size Selection;

       Regulator Elevation Sketch:
           El = 152.88
                                   /  J
mien nc^~^--^
— J.oU . Uo T^>-
'
i .
' hmax
y y y /^ ^^/ i y
hn S ha
* h -^ *

/!__
f / f S / '
n .nas
                                                               £1 = 150.13
       Regulator Nozzle will be square, i.e.,  aspect ratio, a=l to provide
maximurii opening dimension for admittance of maximum size solids in sewage
flow. Then, from page 65, Ref. #1, for diversion capability of 50% of total
flow, supply head, h, must be 2.2 hn.  Allowing 1 in. = 0.083 ft. for regulator
structure depth,  then:

                                   73

-------
           max. available  h = 151.88 - 150.05 - 0.083  -  1.4 hn
                             = 1.747 - 1.4 hn

But max. available supply head = 2.2 hn for 50% diversion capability.
Then:      hn = 1.747/1.4 + 2.2 = 0.485 ft. = 5.82 in.   Select hn = 5 in. to
provide margin of diversion capability above' 50% at maximum supply head.
Then An = 25  in.2 = 0.174 ft.2  .

Regulator Performance Prediction;

       From  pages 20, 21 and 65 of Ref. #1, the following values of total flow,
Qtot;  % diversion of total flow, D; discharge coefficient,CD; maximum regulator
flow to the outfall, Qoutmax; minimum regulator flow to the interceptor,
Qintmin» an(* supply head have been determined for h = 5 in., and a = 1:
         h      h/tm
        (in.)

         3      0.6
         6      1.2
         9      -1.8
         12      2.4
         15      3.0
         18      3.6

Sensor Design;
Qtot
(cfs)
0.92
1.06
1.10
1.13
1.135
1.14
0.64
1.04
1.33
1.57
1.77
1.95
D
           7
          20
          43
          60
          70
          82
                      in
          0.04
          0.21
          0.57
          0.94
          1.24
          1.60
0.60
0.83
0.76
0.63
0.53
0.35
       From page 22, Ref. #1, for digital mode operation using bias orifice
configuration; Afcias orifice = 0.0015 An , and ASensor orifice = 0.004 An.
Then Abias orifice = 0.0375 in.2 and diameteibias orifice = O-22 in.;
Asensor orifice = 0.1 in .2  and diametersensor orifice = 0 .358 in. From page
26, Ref. #1, air line I.D.  was selected as 1.5 in.  The minimum dip tube
length above the  high water level to prevent water from entering the connecting
line was computed using Bernoulli equation.  Assume negligible velocity head
and negligible friction loss in flow entering nozzle .  Then:

       h=snstaticn * hvelocityn '  or nstaticn  = h " nvelocityn

where hstaticn = static pressure at nozzle,  and hvelocityn = velocity head  at
nozzle.
Then:  hvelocityn = 1/2PVn2 , and Vn = QtotAn = CD "\/6474~¥,

where   f» = mass density of H2O = 1.9379 slugs/ft .3 .
Then:  hstaticn = h - 1/2 CD2 (64.4/0.433 x 144)(1.9379)(h)
                = h ( 1 - CD  ) ,  where h is measured in in. H2O.

                                    74

-------
Then:  h at maximum storm flow = 12 in. + 14 in. =2.17 ft; h/hn = 5.2,
         and CD = 1.16.

Then:  hstaticn = -9 in. H2O .

At maximum sewer flow, the water level at the upstream manhole = 4.84 ft.
above dam, or at elevation  156.72 ft.  The sewer invert at this location is at
elevation = 152.20 ft. Assume lower end of dip tube is 0.5 ft. above invert.
Then dip tube length to prevent water from entering connecting line = 156.72
- 152.20 - 0.5 + 0.76 =4.78 ft.

BINGHAM ST.  LOCATION

Regulator Nozzle Size Selection;

       Regulator Elevation Sketch:
                                                      •El = 55.44
                                                   lmax
fa
f
1 / / / / J,
       Selected nozzle area,  An, is limited by the connecting sewer size, and
is selected = 1.5 ft.2 = 216 in.2. Compare performance characteristics of
nozzles with aspect ratio, a,  of 1 and 1.5.  Then nozzle dimensions are 14.7
in. x 14.7 in.,  and 12 in. x 18 in., respectively. For hn = 14.7 in. = 1.23
ft; hd = 1.4 x 1.23 = 1.72 ft.

Then:  Maximum available supply head = 55 .44 - 49.84 — 1.72
                                     = 3.88 ft. = 46.56 in.

Then:  hmaxAn = 3 .17; from  page 65, Ref. ttl, Dmax = 72%

For    hn = 18  in, = 1.5 ft.,  hd = 2 .1 ft, Then maximum available supply
head, hmax = 55 .44 - 49 .84 - 2 .1 = 3 .5 ft.  =42 in. Then hmax/hn = 2 -33,
and Dmax =  66%.

       The 12 in. x 18 in. nozzle was selected, since the predicted diversion
performance for both nozzles was close, while the horizontal dimension of the

                                   75

-------

 14.7 in. x 14.7 in. nozzle would be 23% larger, and would require a propor-
 tionately larger and costlier dam and regulator structure.

 Regulator Performance Prediction:

        From pages 20,  21 and 65, Ref. #1, the following values of D, CD,
 and regulator flows vs. supply head have been determined for hn = 1.5 ft.,
 and, a =  1.5:
                hAn
     Qtot
     (cfs)
D
                                                  oum
0.67
1.0
1.33
1.67
2.00
2.33
2.67
0..92
1.00
1.065
1.09
1.11
1.12
1.13
11.07
14.74
18.13
20.75
23.14
25.22
27.20
11
24
36
48
57
66
72
1.22
3.54
6.52
9.96
13.19
16.65
19.58
9.85
11.20
11.60
10.79
9.95
8.57
7.62
         h
        (ft.)

        1.0
        1.5
        2.0
        2.5
        3.0
        3.5
        4.0

Sensor Design;
        Digital Mode:  From page 22, Ref. #1, Abias orifice = 0.0015An,
Asensor = 0.004 An.  An = 216in.2.  Abias orifice = 0.323 in.2 and
diameter-bias orifice = 0.64  in.  Asensor = 0.865 in.2 and  diametersensor ==
1.05 in. From page 26, Ref. #1, an air line I.D. size of 4 in. was-selected,
assuming a line length to the Tacony Creek Interceptor Sewer of 500 ft.  The
minimum dip tube length to prevent water from entering the connecting line
was found using the same approach as for the other regulator.  At maximum
storm flow assume water level is 3 ft. above dam crest. Then h  = 3 .5 4- 3 =
6.5 ft.  Then h/hn = 4.35, and CD = 1.14 .  Then hstaticn = l •95 ft • Assume
end of dip tube is 27 in. =2 .25 ft.  above interceptor invert, or 1.75 ft. be-
low top. Then total tube length required = 1.75 + 1.95 = 3.7 ft.  = 44.4 in.
Thus a length of 45 in. was  selected.

       Analog Mode: The Philadelphia Water Department established the
following desired proportional regulation characteristics:
           Depth in
          Interceptor
          0 to 27 in.
          27 to 39 in.
          39 to 48 in.*
% of Qtot to
 Interceptor

     100
  100 to 25
   25 to 10
       % of Qtot to
         Outfall
            0
          0 to 75
         75 to 90
       *In this range of interceptor flow depth, the regulator is assumed to be
exerting maximum control toward the outfall.  The actual percentage of Qtot
                                   76

-------
would depend on the regulator upstream head.  The proportional range, of depth
is 27 in. to 39 in., a difference of  12 in. Assume flow regulation will step-
proportional over 5 equal steps, at depth intervals of 3 in.  Then system hy-
draulic geometry is sketched:
                  7	J	7	7	7	7  7
Use 45 in. for longest tube, as for digital sensor, to permit interchangeability.
From page 24, Ref. #1, the total dip tube area = 0.02 An = 4.32 in.2.  Since 5
dip tubes are used, each dip tube area = 0.864 in.2, with an I.D. = 1.05 in.
Use nom.  1 in. pipe.

U-Tube Sensor Design

       Figure 70, page 82, Ref. #1, shows the flow regulation fraction, R, as
a function of  sensor airflow, QA. This characteristic is approximated by the
equation:          QA =  3 QAmax (R - 0.75 R2)


Also assume that Asensor = 0.02 An R.

                                    77

-------
Then:

Substituting:
                                              t*
 QA - Asensor CD YAP/  or Ap = (QA/ASensor CD)

AP = K (1 - 0 .75 R)2 , where K = (3 QAmax/0 .02 An CD)
Ap vs . R is plotted, showing the ranges of Ap associated with successive dip
tube uncoverings, hence the required lengths of each dip tube, with proportion-
al increments of R.  For the purposes of conservatism in the experimental de-
sign APmax was assumed = 18 in. H2O.  (If calibration testing indicates that
this  value is excessive, the dip tubes could be easily shortened,,)

       di was  selected = d2  for simplicity  of design,  di was selected = 12 in.
The largest error in U-tube water level occurs at the uncovering of the first, or
shortest tube, when water flows from all tubes .
Tnen:              Total outflow volume = 4 x tube area x change in Ap +
                    tube area x  18 in. =17.42 in.3.

Then level error = outflow volume/U-tube area =  0.15 in.  This error is regard-
ed as acceptable.  Then dip tube lengths below quiescent liquid levels in U-
tube sensor, for water and ethylene glycol, are as follows:
          Tubes_in uncovering
                order

                  1
                  2
                  3
                  4
                  5
                Tube submergence, inches
                 water   ethvlene qlvcol
1.8
3.78
7.02
10.62
14.76
1.62
3.42
6.31
9.55
13.35
       l.OK


       AP


        .5K
                          Dip tubes #1 to #5
                           uncovered
               -Dip tubes #1 to #4
                 uncovered
                       Dip tubes #1 to #3
                         uncovered
                                  Dip tubes #1 and #2
                                   uncovered
                                          *ip tube # 1
                                          xmcovered
                                                        1.0
                                    R
                                     78

-------
                               APPENDIX B

                         SITE CALIBRATION DATA
CALLOWHILL & 67TH ST. LOCATION

Laboratory Calibration's

March 6,  1972—
       Notes: Tests run at Bowles Fluidics Corp., Silver Spring,  MD,  Total
Regulator flow, Qtot •  derived using Ref. #1 criteria. Outfall flow, QOut  >
determined by free-fall trajectory approach, see sketch below.   Maximum out-
fall diversion, Dmax /  measured with outfall-side control port plugged, inter-
ceptor-side control port open.
                     m = 23.25
                       = 1.938
                                                        Refence Level
Outfall discharge width = 7.5 in. Then discharge area = 7.5 c in.2 = 0.0521
c ft. 2 .  To determine flow velocity over discharge = V; let t = time to drop a
distance, m, to a reference level. Then:
          m = 16 .112  , or t = Vm/16.1  = Ym/4.012 .

                                   79

-------
Using above dimensions ,  t = 0.347 sec .  Then: b = V t, or V = b/0.347
(bin ft.)  or:  V = 0 .25 b' ,  (b' in in.).
Then:  Qout = 0.24  (0.0521) cb' = 0.0125 cb'  .
Run
No.
1
2
3
4
5
6
7
8
h
c
b'
Qout hAn
(in.) (in.) (in.)
3.25 0
4.5 1
7.0 2
10.0 4
10.75 4
5.5 2
3.8 5
5.1 2
.6
.5
.5
.0
.25
.25
.0
.25
To reduce trickle flows
raised
9
10
11
12
13
.14
June I

9
14
16
20
21
14
9
15
to
1 in. Then Qtot -
7.9 2
9.9 3
3.7 0
5.4 1
9.7 4
6.1 2
, 1972—
Notes:
previous runs #9
Figure
1
2
3
4
5
6
7
8
6 • Qout
4.1 0
6.5 3
10.2 4
5.0 1
5.2 1
6.2 2
7.2 2
8.0 2
.75
.75
.40
.25
.50
.50

Tests
to #14
17
20
7
9
20
15









an
CD
(cfs)
0
0
0
1
1
0
0
0
.07
.263
.5
.0
.12
.39
.06
.42
acceptable
0.0122







run at
0
0
0
0
1
0

cb' .
.57
.92
.03
.14
.10
.46

0.65
0.90
1.4
2.0
2.2
1.1
0.8
1.0
level

1.58
1.98
0.74
1.08
1.94
1.22

Bowles Fluidics
. Control
= 0.0122
.50
.50
.50
.25
.75
.0
.25
.5
6
14
18
10
13
15
15
16
cb'
.0
.5
.5
.0
.0
.0
.5
.0
•
0
0
1
0
0
0
0
0
0.93
0.99
1.07
1.11
1.12
1.04
0.96
1.01
Qtot
(cfs)
0
0
1
1
1
0
0
0
, the outfall

1.09
1.11
0.94
1.04
i.n
1.04


1
1
0
0
1
1

Corp. Same
ports connected to a

.04
.62
.02
.15
.28
.37
.43
.49

0.82
1.30
2.04
1.0
1.04
1.24
1.44
1.6

0.96
1.07
1.12
1.01
1.02
1.06
1.07
1.09
.68
.85
.14
.41
.48
.98
.75
.92
D
(%Qtot)
10
31
44
71
76
40
8
46
discharge weir was

.23
.41
.73
.97
.39
.04

basic
diaphragm

0
1
1
0
0
1
1
1

.78
.10
.44
.91
.96
.06
.15
.24

46
65
.4
14
79
44

setup as for
valve per

5
56
71
16
29
35
38
39
Site Calibrations

July 26, 1973 —
         Notes:  Regulator out-of-level condition was noted upon preliminary
inspection.  Basic regulator operation appeared similar to laboratory operation.
Need noted for increase in outfall weir height to prevent excess trickle flow
to outfall when regulator is controlled toward interceptor. Interceptor flow
measurement equipment was found inoperative.  No quantitative! data taken.
                                    80

-------
August 28, 1973 —
         Notes: Total flow derived from regulator supply head measurement
and Ref. #1 Q vs . h criteria.  Outfall flow measured using 90° V-notch weir
downstream of a large tank which receives the outfall flow.  Interceptor flow
determined as the  difference between total and outfall flows,  since interceptor
flow measurement  system of Microstrainer Project was still not functional.
Total flow to regulator controlled by adjusting opening of fire hydrants along
street upstream from site.  The outfall weir elevation was raised using an ad-
justable  weir.  (See sketch below.)  This weir was set so that its  elevation
was 2 .375 in. above the element roof at the discharge end to compensate for
an out-of-level condition of 2 in. in the regulator installation.
Galv. Sht. Steel
Attach with Screw
    Concrete
       Nails
Caulk edges after
  installation
Adjust height by
inserting wedges
Run
No.
1
2
3
4
5
6
h
(in.)
5.. 38
10.38
6.62
8.12
5.88
4.38
h/hn
1.05
2. .08
1.32
1.62
1.18
0.88
CD
1.04
1.12
1.06
1.09
1.05
0.97
Qtot
fcfs)
0.97
1.45
1.10
1.25
1.03
0.82
nweir
1.4
5.25
3.4
4.0
2.75
1.5
Qout
(cfs)
0.012
0.315
0.107
0.156
0.063
0.014
D
(% Qtot)
1.2
21.7
9.7
12.5
6.1
1.7
Since the indicated values of D were much lower than expected, although the
regulator appeared to be operating as during laboratory calibrations, a rough
check on outfall flow was computed as follows:  Water depth over discharge
weir was estimated at 4in. for h= 10 .38 in. and full control toward the outfall.
Then flow section area =4x 7.5  =30 in.2 =0.21 ft.2. Then center of flow
area is  2 in. above center of weir, and corresponding head drop is 10 .38 - 2 =
8.38 in. = 0.7 ft. Assuming a 50% loss in converting static head to velocity
head through the  regulator, then velocity head = 0.35 ft.  Then velocity
=  -y64.4x 0.35  = 4.75fps.   Then estimated %ut  = 4.75 x  0.21 =  1.0 cfs.
                                    81

-------
Then estimated D = 1.0/1.45 = 69%, which closely matches the laboratory
calibration value, thus indicating a problem in the V-notch weir flow measure-
ment.

September 25, 1973—
         Notes: Same basic setup as for previous  calibration.  Qout measured
using 90° V-notch weir and flow velocity x flow area over weir.  Flow velo-
city measured using pitot tube and propeller spinner.  Spinner was calibrated
at 1 revolution for 16  in. of water movement, independent of velocity.  Inter-
ceptor flow estimated using velocity x flow area through a 10 in. diameter
pipe.  Flow area estimated from flow depth at pipe centerline and also width
of flow surface. Velocity measured using pitot tube.

Run  1 -  Upstream hydrant opened 1-1/2 turns.
Qout    spinner turned 20 revs in 10 sec., depth over weir = 1.75 in.
         Aflow = 1.75x 7.5/ 144 = 0.0911 ft.2
         Vflow = 20 x 16/ 10 x  12 = 2 .67 fps
         Qout = 2.67 x 0.0911 = 0.24 cfs
         V-notch weir head = 3.5 in. =0 .292 ft.  Then QOut'  = 0 .11 cfs
Qint
Qtot
 D
         Compute flow area using water depth approach.
         pitot head = 9 in. H2O = 0 .75 ft. H2O .  Vfiow = 6.95 fps
         Flow area  = Area of segment = 1/2 r2 (A - sin A), A in radians
            A = 180° - 2 B, sin B = (r-d)/r

            r  = 5 in.,  d = 3 .75 in.

            B = 14.48°, A = 151.05° = 2.64 rad.

         Then flow  area = 0 .187 ft.2, and Qint =  1.3 cfs

         Compute flow area using water surface width approach.
         c = 9 in. Then cos B = 0.5 c/r, B = 25.84°, A = 2.24 rad.
         flow area = 0.126 ft .2, and Qint = 0.88 cfs
         Ref. #1 criteria, h = 4.63 in. = 0.385 ft., h/hn =
         CD = 0.99, then Qtot = 0.86 cfs
         Correlate with Qout + Qint •  Use Qint = 0 .88 cfs
0.93
           V-notch weir approach:  Qtot = 0.88+0.11 = 0.99 cfs
           Flow velocity x area  :  Qtot' = 0 .88 -f 0 .24 = 1.12 cf s
         Ref. #1 prediction for h/hn  = 0.93,  a = 1; D = 13% Qtot
         Laboratory Calibration; D = 17% Qtot
         V-notch Weir measurement;  D = 11%  Qtot
         Discharge V x A measurement; D = 2 1% Qtot
Run 2 -  Upstream hydrant opened  2 turns .
                                   82

-------
Qout     Depth over discharge weir = 4.5 in . ,  Aflow = 0.234 in . 2
         Vflow/' pitot head = 1.5 in. = 0 . 125 ft; VfiOw = 2 .84 fps
         Qout = 0.665 cfs
         V-notch weir head = 5 .5 in . =0 .458 ft .
         Qout' = 0.46 cfs

Qint     Vflow/' Pitot head = 8.75 in; Vflow = 6.85 fps
         Aflow; (water surface width measurement);  c = 8.75 in.
           cos B = 0.875,  B = 28.960,  A = 122 .08° = 2 . 13 1 rad.
           Aflow = 0.143 ft. 2;  Qint = 0.98 cfs

Qtot     Ref . #1 criteria;  h = 8 . 88 in . = 0 . 74 ft; h/hn =1.78
         CD =1.1;  Qtot = I-32 cfs
         Correlate with Qjnt + QOut/'
           V-notch weir measurement; Qtot =0.98 + 0.46 = 1.44 cfs
           Flow velocity x area     ; Qtot'=  0.98 + 0.67 = 1.65 cfs

 D       Ref . #1 prediction for h/hn = 1.78, a = 1;  D = 35% Qtot
         Laboratory calibration;    D = 60% Qtot
         V-notch Weir measurement; D  = 32% Qtot
         Flow velocity x area measurement; D  = 41%

         Using Ref #1 criteria for Qtot;
           .V-notch weir measurement;  D = 35% Qtot
           Flow velocity x area measurement; D = 5 1% Qtot

Run 3  -  Upstream hydrant open 2-1/4 turns.

Qout     Depth over discharge weir = 5 . 13 in .  =0.43 ft; Aflow = 0 . 2 67 ft . 2
         Vflow; Pitot head = 3 . 75 in . = 0 . 3 13 ft . ,  Vflow = 4.49 fps
         Qout = 1 • 198 cfs
         V-notch weir head = 7 .5 in; QOut' = 0 . 759 cfs

Qint     vflow' Pitot head = 7.5 in;  Vfiow = 6.344 fps
              ' (water surface width measurement);  c = 7.5 in.
           cos B = 0.75,  8 = 41.41°,  A = 97 . 18° = 1. 70 rad.
           Aflow = 0.0614 ft. 2  ,  Qint = 0.389 cfs

Qtot     Ref . #1 criteria;  h = 11.375 in. = 0 .948 ft .,  hAn = 2-23
         CD =1.12,  Qtot = 1-523 cfs
         Correlate with Qint + Qout/'
           V-notch weir measurement;  Qtot  = 0 .389 + 0 .756 = 1. 145  cfs
           Flow velocity x area measurement; Qtot = 0.389 + 0.926
                                                 = 1.587 cfs
                                  83

-------
 D       Ref. #1 prediction for h/hn = 2 .23 , a = 1;  D = 57% Qtot
         Laboratory calibration;     D =  76% Qtot
         V-notch weir measurement; D = 45% Qtot
         Flow velocity x area measurement; D = 79% Qtot

         From the above results,  it was concluded that measurements of inter-
ceptor flow using depth, or width of the flow in a circular pipe were inconsis-
tent as well as difficult to perform, and that a 90° V-notch weir should be con-
structed for this purpose.  The consistent difference between discharge flow x
area, and V-notch weir measurements of outfall flow was attributed to insuffi-
cient settling times before taking weir head readings.

October 4, 1973—
         Notes: Same setup as for the previous calibration, but interceptor
flow measured using a 90° V-notch weir.  Regulator out-of-level condition was
rechecked, and the following relationship to compute supply head was estab-
lished:  h = 12.125  - dwi ,  where  dwi was the water line distance  below the
dam crest, in in.   Also,  Qtot1 = Qout  +  Qint for comparison with Ref. #1
results.   And;  D' =. (Qtot - Qint)/Qtot  •
                                                 	Qout	
  Run     dwi     h      h/hn     CD    Qtot     hw      Q
  No.     (in.)   (ft.)     	    	    (cfs)   (in.)     (cfs.)
1
2
3
4
5
6
- (DWF , Regulator not flowing full ) -
5.5
1.375
0.25
4.50
7.135
0.55
0.90
0.99
0.64
0.42
1.32
2.16
2.38
1.54
1.0
1.06
1.12
1.13
1.09
1.02
1.1
1.48
1.57
1.22
0.92
0
4.0
7.4
7.6
5.4
2.5
0
0.156
0.757
0.788
0.340
0.051
                     Q-int
hw
3 .0
7.9
6.8
6.6
7.7
7.8
Q
(cfs)
0.078
0.885
0.613
0.561
0.819
0.852
Qtot'
(cfs)
0.078
1.041
1.37
1.35
1.16
0.90
D*
(%)
0
15
55
58
29
6
D'
(%)
0
20
59
64
33
7
   1
   2
   3
   4
   5
   6
*Outfall V-notch weir flow measurements appear somewhat low due to large
 time lag before reaching equilibrium conditions in tank upstream of weir.
 Time lag on interceptor V-notch weir should be short due to small upstream
 volume, hence accuracy of readings should be good.

BINGHAM ST. LOCATION
                                    84

-------
Site Calibrations

April 11,  1974 —
        Notes:  Total regulator flow derived using Ref .#1 criteria . Supply head
determined by measuring water level distance below dam crest, dw. Outfall
flow measured using flow velocity x area approach.  Flow velocity determined
using rotating cup instrument.  Flow area determined by depth of flow, dfiow,
over outfall weir x channel width, (1.5 ft.)-.  Run started when DWF backed up
to dam crest elevation, at which time gate over regulator inlet was  raised as
.rapidly as possible, (winching time about 10 sec.),  Readings of supply head
and outfall flow parameters taken at 15 sec, intervals .  Run 1 was a shake-
down run which did not include all measurements.
Run 1 —                                                  .
Time dw
(sec .) (in.)
q
15
30
45
60
75
9.0
105
120
135
150
*
Run 2
0
is
30
45
60
75
90
105
120
135
150

3
N
6
7
8
9
10
12
14
14
0
.5
.T.
.0
.5
.5
13
.3
.0
.0
.5
Vflow
(fps)
0
N.T.*
ti
4.0
4.0
4.5
4.5
4,0
3.5
3.5
N.T.
dflow
0
N.T.
ii
H
ti
M
it
ii
11
ii
it
h
(ft.)
3.5
3,21
-
3.00
2.88
2.79
2.73
2.64
2.50
2.33
2.29
hAn
2.33
2.14
—
2.00
1.92
1.86
1.82
1.76
1.67
1.55
1.53

CD
Qtot Qout
(cfs) (cfs)
1.125
1

1
1
1
1
1
1
1
1
.120
—
.11
.11
.10
.10
.09
.09
.08
.08
24

23
22
22
21
21
20
19
19
*•» »
.15
_ _
.14
.68
.12 -
.88
.32
.75
.84 -
.67 -
D
(% Qtot:

-
_
,.
—
—
'—
_
_
-
-
Not Taken
—

3
3
4
5
6
8
9
10
11
13

0
.0
.5
.0
.75
.75
.0
.0
.0
.5
.0

0
N.T.
5.0
4,2
4.2
4.4
4.4
4.4
5.0 ,
4.0
N.T.

0
N.T.
5.0
5.0
5.0
4.5
4.0
4.0
3.5
3.0
2.0

3.5
3.25
3.21
3 . 17
3.02
2.94
2.83
2.75
2.67
2.54
2.42

2.33
2.17
2.14
2.11
2.01
1.96
1.89
1.83
1.78
1.69
1.61

1
1
1
1
1
1
1
1
1
1
1

.125
.120
.120
.120
.110
.110
.105
.100
.100
.090
.080


24
24
24
23
22
22
21
21
20
20


.30
.15 3 . 13
.00 2.63
.22 2.63
.90 2.20
.39 2.20
.96 2.20
.62 2.19
.92 ,.,1.50
.21 -


_
13.0
11.0
11.3
10.8
9.8
10.0
10.1
7.2
-
                                   85

-------
       A large piece of concrete was found in the outfall-side of the regulator
after this run. When this could not be cleared, further diversion performance
runs were abandoned pending its removal.  Run 3 was made with the outfall-
side control port open, the interceptor-side control port closed, to check trickle
flow over outfall with full control toward the interceptor. Flow velocity and
depth readings were taken at approximately 30 and 60 seconds after the start
of the run.
Run 3 —
Time
(sec.)
0
30
60
dw
an.)
0
4.0
7.5
Vflow
(fps)
0
5.0
5.0
dflow
(in.)
0
4.0*
4.0
h
(ft.)
0
3.17
2.88
h/hn
2.11
1.92
                                             1.12   24.00
                                             1.105  22.55
                                                           Qout
                                                    2.5
                                                    2.5
                                                  (% Qtot)

                                                    10.4
                                                    11.1
       *Outfall flow depth was negligible until shortly before first reading,
         when it abruptly increased to about 4 inches .

May 14,  1974—
         Notes: Same setup as for previous calibration.  Vacuum readings
taken for outfall-side control port. Reading times are estimated, based on
Figure 27.
Est'd
Time
(sec.)
65
85
115
125
dw

(in.)
••JU— i ii f i
7.0
8.0
10.5
11.5
Vflow

(fps)
6.5
6.5
6.5
4.5
dflow

(in.)
6.0
6.0
6.0
4.5
h

(ft.)
2.92
2.83
2.68
2.54
h,


1
1
1
1
n


.94
.89
.75
.69
CD


1.11
1.10
1.09
1.09
Qtot

(cfs)
22.8
22.4
21.4
20.9
Qout

(cfs)
4.88
4.88
4.88
2.53
D

(% Qtot)
21.4
21.8
22.9
12.1
C.P.
Vac.
(in.H?.Q)
5.0
N.T.
4.0
2.0
  140
13.0
4.5   4.5    2.42   1.61  1.08   20.3  2.53   12.5
                                                                        0
Diversion readings were higher than on previous calibration, but much less
than predicted, based on derivation of Qtot using Ref .#1 criteria, Actual Qtot
may have been much less due to flow impedance in 24 in. connecting sewer.

1/6 Scale Model Test Program, May 20 to July 8, 1974—
         The test model is  shown in Figure 28 . The model was constructed of
laminated particle board waterproofed with clear lacquer. A clear plastic cover
allowed direct observation of internal flow patterns . Flow rates were determin-
ed by measuring the time taken to fill a 7-gallon container. The measurement
accuracy is estimated at 5-10% at max. flow rates; 2-5% at low flow rates . The
model was designed with a replaceable outfall discharge section.  Tests were
conducted on the prototype and three possible modifications .  Each modified

                                     86

-------
 configuration had the discharge weir widened from 3 in. to 7 in., with variations
 in the elevation at which the widening started.  The test results are shown in
 Figure 29 .  Note the large improvement in diversion performance obtained with
 all the modified configurations , as compared with the prototype .  These tests
 showed the need to supplement the Ref.  #1 design criteria to include design
 characteristics for outfall discharge width as a function of inlet aspect ratio, a.
 The proposed modifications, prototype, and selected modification configurations
 are shown on Figure 30.-  The  selected configuration represented a practical com-
 promise between improved diversion performance and excavation cost.

 September 27, 1974—
        Notes: Basic calibration setup as on previous calibration attempts.   Add-
 ed measurements were pitot head flow velocity, and interceptor-side control port
 vacuum.  Only maximum head readings were taken, at approximately 10 seconds
 after the run start, to avoid the flow impedance problem discussed earlier. The
 outfall width was  checked and found to be 42 in.
  Run
  No.

    1
    2
    3
    4
    5
    6
"w

(in.)
 6
 5
 6
 4
 5
 3
Vflow
R-C Ind.
ftps)
2.9
2.5
2.9
2.5
2.2
2.6
Vel. Hd.
pitot
(in.H2O)
N.T.*
"
3.0
2.6
1.2
N.T.
Vflow
pitot
ftps)
—
-
4.0
3.73
2.58
-
dflow
outfall
(in.)
8.0
8.0
9.0
9.0
7.5
9.0
h

ftt.)
3.0
3.08
3.0
3.17
3.08
3.25
h/hn


2.0
2.05
2.0
2.11
2.05
2.17
CD


•••^^•••••^H
1.115
1.115
1.115
1.12
1.115
1.125
              Qtot
   i
   2
   3
   4
   5
   6
Qout
R-C Ind.
(cfs)
6.77
5.83
7.60
6.60
4.80
6.80
Qout
pitot
(cfs )
_
-
10.53
9.80
5.60
—
D
R-C Ind.
(% Qtot )
29
25
33
28
20
28
D
pitot
(% Qtot)

-
45
41
27
—
                                                      C . P . Vac.
                                                     outfall   int'r.
                                                        (in.H20)
                                                      •8
                                                      -  8
                                                      •  8
                                                      •15
                                                      •  7
                                                      •  9
 - 4
N.T.
       * Not Taken
      Comments: On runs #5 and #6,  flow was observed to back up in the
special junction manhole, approximately 24 in. above the inlet to the connecting
sewer, at about 20 seconds after the start of the run.  Since the backup presum-
ably occurred on all runs, the values of Qtot/  as derived using Ref .#1 criteria,
are probably significantly larger than actual, since this  criteria assumes  no
                                    87

-------
hydraulic loading on the regulator.  Determination of flow velocity using pitot
head readings produced consistently higher values than indicated by the rotating
cup instrument. Interceptor-side control port readings were discontinued after
run, as the instrument malfunctioned.

November 14, 1974--
      Notes:  Basic calibration was as on previous calibration attempts . Pitot
head readings were taken with a floating sensor, which greatly reduced errors in
pitot tube submergence, occurring with hand-held readings .  This sensor was
directly calibrated in fps . The flow backup into the special junction manhole
after about 20 seconds of the run was seen to occur on two runs, and presumably
occurred on all runs . Thus the values of Qtot derived from the  Ref .#1 criteria
are probably higher than actual, and the values of D correspondingly lower.  Out-
fall-side control port vacuum readings were seen to drop from around -10 in.
H2O to about -4 in. H2O  at 15-20 seconds after the runs started,.  As on the
previous calibration, the  outfall flow velocity readings from the rotating cup
instrument were about 30% lower than from the pitot head sensor.
Run
No.
  1
  2
  3
  4
  5
  6
  7
aw

(in.)
 6
 6
 6
 5
 6
 5
 5
Vflow
R-C Ind.
(fps)
2.8
2.4
2.1
2.6
1.6
2.2
1.6
Vflow
pitot
(fps)
3.0
3.25
3.1
3.3
2.75
2.75
2.7
dflow
outfall
(in.)
8.0
7.8
7.5
8.5
8.0
8.0
7.5
h

(in.)
3.0
3.0
3.0,
3.08
3.0
3.08
3.08
h/hn


2.0
2.0
2.0
2.05
2.0
2.05
2.05
CD


1.115
1.115
1.115
1.120
1.115
1.120
1.120
Qtot

(cfs)
23.3
23.3
23.3
23.6
23.3
23.6
23.6
  1
  2
  3
  4
  5
  6
  7
Qout
R-C Ind.
(cfs)
6.53
5.46
4.59
6.44
3.73
5.13
3.50
Qout
pitot
(cfs)
7.58
7.39
6.78
8.18
6.42
6.41
5.91
D
R-C Ind.
(%Qtot)
28
24
20
27
16
22
15
D
pitot
(%Qtot)
33
32
31
35
28
27
25
                                                       C . P. Vac .
                                                     outfall    int'r
                                                        (in. H2O)
- 7
- 5
- 9
-10
-5
- 6
- 6
0
0
0
0+
+2
-1
0+
                                     88

-------
                                APPENDIX C

             CONVENTIONAL REGULATOR PERFORMANCE ANALYSIS


67TH/CALLOWHILL LOCATION

Compute Depth of Flow vs . Flow vs . Velocity

       Qmax  = 27.6 cfs, Vmax = 8.8 fps,  sewer I.D. = 2 ft.
       slope  = 0.02,  flow area = 3 .14 ft.2

At Qmax / flow area = 2 7.6/8 .8 = 3 .14 ft. 2, indicating full pipe flow.
Then d/D =1.0. Check slope by Manning's equation:
                 1/y   2/^5                  1 /     O /
       V= 1.49s    rH   /n,  &n= 1.49 s /2  rH/3 /V  rH =r= 1ft.
                                                       V = 8 .8 f ps
                               n = 0.024,  which appears reasonable,

Find Flow Entering Slot:

          / / / /  / L.JL.J- I /  /'.
                                            a' (shaded) = flow area
                                            not entering slot

                                            	'-•flow section after drop
                                  89

-------
Determine  y:
       fly time = t = x/v, y = 16 .112  , x set at 0 .5 ft. by PWD, then
                         y = 4.025/v2
       D = 2 ft., d' = d - y, V = 8.8 fps,  A -3.14 ft.2
       Q = 27 .6 cfs , assume n is .constant,  slot flow = qs = q - q'
d/D

0.6
0.5
0.4
0.3
0.25
0.2
0.15
0.125
0.1
0.075
0.05
d
(ft.)
9m^mu*m~mJfmif
1.2
1.0
0.8
0.6
0.5
0.4
0.3
0.25
0.2
0.15
0.1
q/Q

0.671
0.500
0.337
0.196
0.135
0.088
0.047
0.032
0.021
0.014
0.008
q
(cfs)
18.52
13.80
9.30
5.41
3.73
2.43
1.30
0.88
0.58
0.39
0.22
v/V

1.072
1.000
0.902
0.776
0.699
0.615
0.530
0.465
0.401
0.325
0.238
V
ffps)
9.43
8.80
7.94
6.83
6.15
5.41
4.49
4.09
3.53


y
(ft.)
0.045
0.052
0.064
0.086
0.106
0.137
0.200
0.240
0.323


d-y
(ft.)
1.155
0.948
0.736
0.514
0.394
0.263
0.100
0.010
-


d-y/D

0.577
0.474
0.368
0.257
0.197
0.131
0.050
0.005
-


q'/Q

0.630,
0.459
0.288
0 . 142
0.085
0.046
0.010
0.001
-


q1
(cfs)
17.39
12.67
7.95
3.92
2.35
1.27
0.26
0.01
—


qs
(cfs)
1.13
1.13
1.35
1.49
1.38
1.16
1.04
0.87
0.58
0.39
0.22
Compute Fluidic Regulator Flow for Overflow Conditions;

      Assume dam is sharp-edged weir.   Use Francis formula:
                      3/2
       QD = 3.331hD   ,  hD = hR - 1.167 ft., 1 = 2 ft.
hR
hD
QD
QReg
QD + QReg
D
Qoutmax
Qintmin
1.17
0
0
1.90
1.90
0.72
1.37
0.53
1.25
0.083
0.16
2.00
2.16
0.80
1.60
0.40
1.33
0.167
0.45
2.05
2.50
0.83
1.70
0.35
1.42
0.25
0.83
2.10
2.93
0.86
1.81
0.29
1
0
1
2
3
0
1
.0
.50
.333
.28
.15
.43
.88
.89
.26
1.58
0.417
1.79
2.18
3.97
0.89
1.94
0.24
1.66
0.500
2.35
2.25
4.60
0.90
2.03
0.23
1
0
2;
2
5
0
2
0
.75
.583
.96
.28
.24
.91
.07
.21
2 . 167
1.00
6.66
2.65
9.31
0.94
2.49
0.16
      Various Q's, q's, in cfs, various h's in ft.  Hydraulic element data ob-
tained from Table 14-3, Page 14-8, Ref. #2.  Above data is shown plotted on
Figure 33.

BINGHAM ST. LOCATION

Find Static Regulator Flow Relation:

      Regulator is manually positioned sluice gate,  maximum opening 3 ft.
                                    90

-------
wide x 2 ft. high.  Dam crest of original installation at elevation S3 .5 ft. .
Bottom of gate opening at elevation 51.5 ft. Dam width = 9 ft.  Determine gate
opening corresponding to dam overflow at 2 x maximum estimated DWF = 2 x 10
cfs = 20 cfs .  Assume gate inflow CD = 0.7, since bottom of gate opening is
very  close to  sewer invert.
           gate opening ~ ho
                                                               El 53.5 ft
                                                                  51.5 ft
Equations:

      h = 2 -ft.  - h0/2,  A0 = 3 h0

Then:       20  cf s = 0.7 (3ho)  (64.4 (2 - ho/2))

                  = 16.85 h0 (2 -h0/2) /2

Squaring both sides:

            400 = 284 h02  (2 - h0/2) =568 ho2 - 142 hQ3  , or

            h03 - 4 h02 -f- 2 .817 = 0 .   Solve graphically.  Find value of ho at
which left side of equation = 0.

            h0       1.1      1.0      0.9

Left side of  eq'n   -0.692   -0.183   +0.306

Plotting these values:
                                    91

-------
               -1.0--
Check:
Then:
           h     = 2 -ho/2 = 1.52 ft., A0 = 2.88ft.2
           QRef  = 0.7  (2.88)(64.4 x  1.52) 2  = 19.95  close enough to 20
Then relationship for QReg :          ,,    ^
           QReg= 0.7 (2.88)(64.4)/2  (h) /2  =  16. 178
                                                       ]/
Determine relationship for flow over dam, QD' vs .
for sharp crested weir:
                            3/2
            QD  = 3.33  1 hD      1 = 9 ft. ,
                        3/£
then        QD  = 30 hD   /  and  hD=h-1.52.
Determine Total Flow for Static Regulator:
                                                      Use Francis formula
h
1.52
2.0
2.5
3.0
3.5
4.0
4.5
QD
0
10.0
29.1
54.0
83.6
117.2
154.3
QRea
20.0
22.9
25.6
28.0
30.3
32.4
34.3
QD + QReg
20.0
32.9
54.7
82.0
113.9
149.6
188.6
                                   92

-------
Fluidic Regulator Total Flow:

      Determine flow characteristics of fluidic regulator under dam overflow con-
ditions,  considering maximum regulator control toward either interceptor or out-
fall .  Regulator diversion characteristics derived from 1/6 model test calibration
extrapolated.  QReg derived using Ref. #1 criteria. QD determined using
Francis formula:
then:
 QD =  3 .33 1
            QD  = 25.64 hD   .
                                  1 = 8 ft . ,  hD = h - 3 .5 ft . ,
h           2.25
h/hn        1.50
D(%QReg) 16.00
QReg .      19.00
Qout        3 .00
Qint
hD
QD
2.63
1.75
37.00
22.00
8.10
3
2
49
23
11
.00
.00
.00
.50
.50
3.38
2.25
59.00
25.50
15.00
3.75
2.50
69.00
27.00
18.60
4.13
2.75
74.00
29.50
21.80
4.50
3.00
80.00
32.00
25.60
5
3
83
34
28
.00
.33
.00
.00
.20
5
3
86
36
31
.50
.67
.0
.0
.0
6.0
4.0
88.0
38.0
33.4
16.00  13.90  12.00  10.50  8.40   7.70   6.40   5.80    5.0   4.6
         -                 0.25   0.63   1.00   1.50    2.0   2.5
                           3.33  13.32  25.60  48.90   75.3 105.3
QReg+QD   19.00   22.0023.50  25.5030.30  42.80  57.60  82.90  111.3  143.3
34.
      Various Q's, in cfs, various h's in ft.  This data is shown plotted on Figure
                                   93

-------
                                GLOSSARY
Fluidic Nomenclature
Nozzle:  Opening where fluid 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.

Splitter: A wall which divides the fluidic element exit area into two sections .

Attachment Walls:  Wall of element immediately downstream of control ports
      to which the nozzle flow 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 re-
      sult 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 element control ports due
      to suction effect of venturi nozzle.

Aspect Ratio:  Ratio of regulator inlet nozzle height to width.

Digital Operation:  Regulator operating having maximum flow diversion to
      either discharge.

Analog  Operation:  Regulator operation having a continuous range of diversion
      performance between discharges.

Diversion:  The capability of a fluidic regulator to  direct part of the entering
      flow toward the higher elevation discharge, away from the lower ele-
      vation discharge.
                                    94

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
 1. REPORT NO.
    EPA-600/2-77-071
                                                           3. RECIPIENT'S ACCESSION-NO.
          SUBTITL
            OFTF!UIDIC COMBINED SEWER REGULATORS
 UNDER MUNICIPAL  SERVICE CONDITIONS
              5. REPORT DATE
               August 1977(Issuing, Date)
                                                           6. PERFORMING ORGANIZATION CODE
 7. AUTHOR(S)
 Peter A. Freeman
              8. PERFORMING ORGANIZATION REPORT NO
 9,.E.EBJE.O.BMLN.G QRGAN.LZA.TION NAME AND, ADDRESS
   Peter A. Freeman Assoc.,  Inc.,  Box 2210, Rt. #4, Ocean
   Pines, Berlin, MD 21811
   under subcontract to:  Philadelphia Water Dept., 1180
  Municipal Sevices Bldg.,  Philadelphia, PA 19107
              10. PROGRAM ELEMENT NO.
              1BC611
              11. CONTRACT/GRANT NO.
              Grant  No.  11022 FWR
 12. SPONSORING AGENCY NAME AND ADDRESS
 Municipal Environmental  Research Laboratory--Cin.,OH
 Office of Research  and  Development
 US  Environmental  Protection .Agency
 Cincinnati, Ohio  45268
              13. TYPE OF REPORT AND PERIOD COVERED
              Final
              14. SPONSORING AGENCY CODE

                EPA/600/14
  Pi. SUPPLEMENTARY NOTES                                                              	
  .0.-Richard Field,  Chief,  Storm & Combined Sewer,Sec., Edison, NJ 08817,  201-321-6674
 Prototype.evaluation of .njpdel developed, .underprevious project as reported in  "Design
 of a Combined  Sewer  Fuiidic Requiator.USEPA Report No. 11020 GZU 10/69  (NTIS  PBl.88914
 16. ABSTRACT
 This report describes  the evaluation of two fluidic  combined sewer regulators  operated
 by the City of Philadelphia Water Department under typical  municipal service con-
 ditions.  The smaller  unit provided much better  hydraulic regulation performance than
 the conventional static  regulator it replaced, approaching  that of a complex,  dynamic
 regulator.  The larger unit demonstrated a similar performance potential which was  not
 practicably achieved because of unforeseen, heavy  infiltration in the connecting sewer

 The Philadelphia Water Department determined that  surveillance and maintenance costs
 for fluidic regulators were actually lower than  for  conventional  static regulators  and
 much lower than for conventional  dynamic regulators.   Retrofit costs for fluidic
 regulators were determined as about 30% greater  than  for conventional static units,
 DUt less than half of  those for conventional dynamic  regulators.   Considering  both
 lydraulic performance  and costs,  the fluidic regulator was  considered to offer greater
 cost effectiveness than  either conventional static or dynamic combined sewer regulators
 7.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS
                                                                        c.  COSATI Field/Group
  Combined sewers, Fluidic  devices, Regu-
  lators, Overflows-sewers,  Hydraulic gates
Diaphragm value, Low cost
Low maintenance, Inter-
ceptor variable diversion
Hydraulic sensors
                                                                              13 B
 8. DISTRIBUTION STATEMENT
     RELEASE TO PUBLIC
                                              19. SECURITY CLASS (ThisReport)'
                                                UNCLASSIFIED
                           21. NO. OF PAGES
                             J.UO
20. SECURITY CLASS (Thispage)
                                                                        22. PRICE
                                                UNCLASSIFIFH
EPA Form 2220-1 (9-73)
                                            95
                                                    U. S. GOVERNMENT PRINTING OFFICE: 1977-757-056/6't86 Region No. 5-11

-------

-------