HYDRO BRAKE STORMUATER DETENTION

     SYSTEM DEMONSTRATION  IN CLEVELAND,  OHIO

          Dual  Combined Sewer Overflow
         Pollution Control  and Basement
                 Flooding  Relief
                        by
               Timothy M.  Matthews
               Paul  D. Pitts, Jr.
               R. Charles  Larlham
         Snell  Environmental  Group, Inc.
                 Stow, Ohio  44224
                Grant No.  G005370
                 Project Officer

               Ralph G.  Christensen
         Great Lakes Demonstration Program
        Great Lakes National Program Office
             Chicago. Illinois  60604
               Technical Advisors

                Richard P. Traver
                Douglas C. Ammon
   Municipal Environmental Research Laboratory
             Cincinnati. Ohio  45268
This study was conducted in cooperation with the
   Municipal Environmental Research Laboratory
  United States Environmental Protection Agency

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

-------
                                 ABSTRACT


     This research project was  initiated  with  the overall objective of determining
the ability of the Hydro Brake  flow control  device,  in  conjunction with offline
underground storage, to limit the rate of stormwater flow into combined sewers.
The .intended result of this control was the  reductions  of overflows (CSO) and
street and basement flooding during storm events.

     Three underground storage  tanks were constructed and outfitted with Hydro
Brakes of different flow rates.  The storage tanks were filled,  and their rates
of discharge were measured to establish discharge curves for  the Hydro Brakes.
The Hydro Brakes were then monitored for  performance during actual storm events.
During the study, each Hydro Brake was downsized and retested for discharge
rates, permitting the evaluation of six sizes  of Hydro  Brakes.   Homeowner surveys
were also undertaken to evaluate the effects of the  Hydro Brake/storage instal-
lations on flooding.

     Results of the draindown tests were  evaluated in terms of their  comparison
with discharge curves of orifices of equivalent size.  Measured  storm flows were
similarly evaluated.  In addition, one year, two year,  five year and  ten year
return period storm flows were  identified from storm frequency tables, and
discharge hydrographs and storage needs were then calculated  from those storm
flows and the observed discharge curves.

     It was demonstrated that the Hydro Brakes did release storm flows to
combined sewers more slowly, and at a rate more nearly  independent of head,
than orifices of equivalent size.  The use of  off-line  storage tanks  appeared
to reduce the incidence of street and basement flooding.

     This report was submitted  in fulfillment  of  Grant  No.  G005370 by Snell
Environmental Group, Inc. under the sponsorship of the  U.S.  Environmental
Protection Agency.  This report covers the period from  September, 1980 to
August, 1981, and work was completed as of January  31,  1982.
                                       IV

-------
                                 CONTENTS

Foreword	-Hi
Abstract	-..,.	iv
Figures	".	vii
Tables	viii
Abbreviations and Symbols  	   ix
Acknowledgment 	   x
   1.  Introduction  	   1
   2.  Conclusions 	   7
   3.  Recommendations 	   9
   4.  Method of Approach  	   11
           General Procedures  	   n
           Description of Hydro Brake Control  Devices  	   H
           Description of Installations  	   12
           Simulation of Design Conditions 	   15
           Storm Flow Monitoring	18
           Precipitation Records 	   l9
           Water Quality Sampling  	   19
           Survey of Service Area Population  	   20
           Data Analysis	20
   5.  Results	22
           Discharge Curves  	   22
           Storm Hydrographs 	   29
           Storm Water Quality 	   33
           Homeowner Surveys 	   37
   6.  Design and Performance Evaluation 	   42
           Design Concept  	   ^2
           Design Criteria 	   42
           Storm Simulation and Design	43
           Evaluation of Original Design . ,   ,	44

-------
           Hydro Brake Redesign	   46
           Evaluation of Hydro Brake Redesign 	   47
   7.  Comparative Evaluation of Flow Regulator 	•	   52
       Installations  	   52
           Rochester, New York	   52
           Nepean Township, Ottawa>  Canada  	   56
           Borough of York, Ontario, Canada 	   62
           Comparison of Installations  .  . ^	   62
   8.  Alternative Evaluation 	   67
           General Screening of Alternatives   	   67
           Description of Viable Alternatives 	   68
           Cost Estimates of Viable  Alternatives  	   68
References	   70
Bibliography  	   71
Appendices	   72
   A.  Formulae, Head vs. Volume Tables	   72
   B.  Homeowner Survey 	   78
   C.  Water Quality Data	,   80
   D.  Photographs and Sediment Measurements   	   87
   E.  Design Storm Hydrographs 	  100
   F.  Cleveland Design Report  	  112
   G.  Hydro Brake Demonstration Project
       Santee Drainage Area - Rochester, New York 	  147
       (O'Brien and Gere Engineers,  Inc.,  1981)
                                   VI

-------
                                   FIGURES

Number                                                              Page
   1      Study Area Location Map	    3
   2      Drainage .Area and Control  Structure Locations 	    4
   3      Overflow Chamber M-15 	    5
   4      W. 170th Street Installation  	   13
   5      W. 177th Street Installation  	   14
   6      Puritas Avenue Installation 	   16
   7      W. 170th St. Head-Discharge Curves:  Original Design  .  .   23
   8      W. 170th St. Head-Discharge Curves:  Redesign 	   24
   9      W. 177th St. Head-Discharge Curves:  Original Design  .  .   25
  10      W. 177th St. Head-Discharge Curves:  Redesign 	   26
  11      Puritas Ave. Head-Discharge Curves:  Original Design  .  .. 27
  12      Puritas Ave. Head-Discharge Curves:  Redesign 	   28
  13      Hydrographs at W. 177th St. Control Structure -
          Storm of June 8-9, 1981	   32
  14      Hydrographs at W. 177th St. Control Structure -
          Storm of July 13, 1981	   34
  15      Hydrographs at W. 170th St. Control Structure -
          Storm of July 13, 1981	   35
  16      Hydrographs at W. 170th St. Control Structure -
          Storm of August 7, 1981	   36
  17      Basement Flooding Survey Responses  	   40
  18      Street Flooding Survey Responses   	   ^
  19      Head-Discharge Curves - Santee Drainage Area  	   53
  20      Head-Discharge Curve - Santee Hydro Brake 	   54
  21      Storm of May 11, 1981 - Santee Drainage Area	   55
  22      Inflow-Outflow Hydrographs - Nepean Twp.
          Area 1 - October 30 -31, 1976	   57
                                     vii

-------
  23      Inflow-Outflow Hydrographs - Nepean Twp.
          Area 2 - May 23, 1977	58
  24      Inflow-Outflow Hydrographs - Nepean Twp.
          Area 2 - June 25, 1977	59
  25      Inflow-Outflow Hydrographs - Nepean Twp.
          Area 2 - June 29, 1977	60
  26      Inflow-Outflow Hydrographs - Nepean Twp.
          Area 2 - July 1, 1977	>	61
  27      Comparison of Hydro Brake Discharge Curves  	  64
                                   TABLES
Number                                                               Page
   1      Hydro Brake Drainage Areas  	  31
   2      Summary of Sampling Results 	  38
   3      Combined Sewer Overflow Reduction at M-15 	  43
   4      Original Hydro Brake Design - 5 year Design
          Storage and Discharge Requirements  	  45
   5      Comparison of Hydro Brake Discharge Rates -
          Original Design vs. Redesign  	  47
   6      Design Hydrograph Parameters - 1/2 hour storm
          Rainfall and Intensity  	  49
   7      Comparison of Hydro Brake Discharge Rates and
          Storage Volumes for Original Design vs. Redesign  	  49
   8      Summary of Hydro Brake Control/Retention for 1/2 hour
          Duration Design Storms  	  51
   9      Analysis of Peak Flow Attenuation for 2 in. Unit	56
  10      Comparison of Rated and Actual Hydro Brake Discharges ...  65
  11      Cost Comparisons of Storm Flow Control Alternatives ....  69
                                   vm

-------
                    LIST OF ABBREVIATIONS AND SYMBOLS
ABBREVIATIONS
CFS/cfs      — cubic feet per second
cm           — centimeter
cm/hr        — centimeters per hour rainfall  intensity
CMP          — corrugated metal pipe
CSO          — combined sewer overflow
ft^          — square feet
ft3          — cubic feet
ha           — hectare
in           — inch
in/hr        — inches per hour rainfall intensity
km           — kilometers
L            — liter
L/s          — liters per second
mm           — millimeter
M            — meter
ml/1         — milliliters per liter
mg/1         — milligrams per liter
RCP          — reinforced concrete pipe
                                   ix

-------
                              ACKNOWLEDGMENTS
     During the course of this evaluation, special assistance was provided by
the City of Cleveland's Divisions of Utilities Engineering and Water Pollution
Control, the Northeast Ohio Regional Sewer District and the National Weather
Service.  Their cooperation is sincerely appreciated.

-------
                                SECTION 1

                              INTRODUCTION
BACKGROUND

     The City of Cleveland has long been plagued with most of the well-
documented problems which beset urban centers  served by  combined sewer sys-
tems.  Such systems are subject to overflow problems as  a matter of design.
Combined sewer overflows (CSC) are designed into these systems as the
primary method of relief when flows exceed the capacity  of the receiving
sewers.  However, while it provides hydraulic  relief for receiving sewers
and sewage treatment plants, CSO often carries heavy loads of pollutants
to streams and water bodies.  This is especially true of the "first flush"
waters from an intense storm.

     CSO occurrences can be eliminated, or their impacts attenuated, by a
variety of documented, generally effective and acceptable methods.  However,
these methods are not always effective in relation  to another set of prob-
lems often associated with combined sewer systems - basement and roadway
flooding.  Those CSO problem solutions which include flow retardation may
exacerbate flooding problems.

     In addition, CSO itself often provides relief  only  for areas downstream
from most of the sewer system.  Upstream areas have no external area to which
overflow may be routed.  Thus, the system is relieved by surcharging into
basements, or to the ground (usually the roadway) surface.  Surcharging and
flooding represent very evident health hazards. Structural damage as a
result of hydrostatic pressure and/or wash-out, as  well  as damage to per-
sonal possessions, are also frequent occurrences.

     For all of these reasons - water quality, health, safety and property
damage - control of CSO and combined sewer surcharging has become very
important in recent years.

     Over the past several years, numerous investigations have  been under-
taken to determine cost effective methods of abatfng  CSO without making
flooding worse, or to abate flooding and surcharging without  increasing
CSO.  Much of this effort has concentrated on upstream stormwater, and
methods of retaining it without worsening local flooding.

     One method has been to provide storage of upstream  stormwater, with
release to receiving sewers after downstream stormwater  has  drained.
This approach avoids upstream flooding by not allowing stormwater  into
sewers until capacity is available, and avoids downstream surcharging
and CSO by permitting downstream flow to be conveyed  away before upstream
flows can arrive.

-------
     Various approaches  to  upstream storage  have been devised:  roof storage,
roadway storage, in-line or off-line storage with  reduced discharge orifices
and offline storage with discharge pumps.

     The City of Cleveland  and the EPA Great Lakes  Demonstration Program under-
took the investigation of the effectiveness  of off-line storage with release con-
trolled by a proprietary device, the Hydro Brake,  which purportedly permits
discharge at a relatively constant rate.

PROBLEM DEFINITION

     The Puritas Avenue - Rocky River Drive  area of the City,  chosen for
this Investigation, encompasses approximately  115  (46.6 ha) acres.  Of this, 9.0
acres (3.6 ha) contribute flows to the Hydro Brake Structures  (Figures 1 and 2).
This area is served entirely by combined  sewers.   The area was primarily developed
during the 1920's as a medium density residential  neighborhood.

     The combined sewers in the neighborhood carry sanitary flows, as well  as
roof drain, weeping tile, driveway drain  and street flows.  All flows from  the
area are ultimately conveyed to a CSO chamber  (M-15 - Figure 3) via a 42 inch
(1.1M) sewer, from which a  36 inch (0.9M)  sewer conveys flows  which do not
exceed its capacity to the  Southerly Sewage  Treatment Plant.   Flows in excess
of the capacity of this  sewer are discharged into  a 60  inch (1.5M) CSO sewer,
which connects to a 78 inch (2.0M) sewer  discharging to the Rocky River.
River.

     Downstream of M-15, 507 acres (205 ha)  of land are tributary to the 78 inch
(2.0 M) sewer.  Land uses are industrial  and residential, and  during storms which
produce CSO from M-15, backwater from this  area affects performance of
the Puritas Avenue area sewers.

     Overland storm drainage within the demonstration area  is  discontinu-
ous, resulting in street flooding at low  road  points during sewer system
surcharge.  Basement flooding also occurs throughout the  area  during severe
rainfall events.

     During the design phase of this project,  an  actual storm  event with
cumulative rainfall approximately equal to a five-year  storm was  used  in  a
sewer system response simulation to further define the  extent  of the  CSO  and
combined surcharge problems.  The Simulation indicated  that the  Puritas Avenue
42 inch (1.1H) sewer was surcharged at an average loading ratio  of  2.0,  (i.e.,
it was subjected to a flow rate double its capacity).   Tributary sewer ratios
averaged 1.5, and ranged from less than 1.0 to 3.0.  An analysis  of this  storm
indicated that some basement floodings would occur every  six  to  nine months,
with CSO occurring every four to six months.H'

OBJECTIVES

     The primary objective of this investigation  was to evaluate the abi-
lity of the Hydro Brake to effectively regulate specific  design flows  from
stormwater storage structures to such an extent that receiving sewers  could
be protected from surcharging and creating CSO conditions.

-------
          CLEVELAND
          HOPKINS
          AIRPORT
                                            LEGEND
                                            (NO SCALE)


                                              STUDY AREA


                                          •   RAIN GAUGE LOCATIONS
FIGURE-I
STUDY AREA LOCATION MAP

-------
^ 	 r
-^~-"

*s
o
CD
»•
>
h-
N
Jf

NQIANA AVE.

      FAIRVILLE
                                                                                           OVERFLOW CHAMBER
                                                                                               M-15
                                                                         LEGEND

                                                                      (NOT TO SCALE)

                                                               LJ  COMBINED SEWER DRAINAGE


                                                               I.J  CONTROL STRUCTURE DRAINAGE AREA


                                                                    LOCATION OF STRUCTURES
FIGURE  2     DRAINAGE  AREA  AND  CONTROL STRUCTURE  LOCATIONS

-------
          INV. 768.66






        —TOP 772.00
FIGURES  OVERFLOW CHAMBER M-15

-------
     The secondary objective of this  investigation  (although undoubtedly the
primary concern of the residents of the study  area) was to reduce
flooding in the study area.

SCOPE OF STUDY

     Three off-line underground stormwater storage  units were  installed in
the study area, and a Hydro  Brake was installed  in  each.  The  following moni-
toring and evaluation tasks  were then performed:

     1.  Five storm events were monitored for  inflow,  hydraulic  level in the
         unit and discharge.  Discrete and composite water quality samples
         were taken from the storage tanks and analyzed, and observations
         of sedimentation were made.   Discharge  curves were developed and
         analyzed for the three devices.

     2.  In addition to the  above monitoring and  analysis, observations
         of the operating characteristics of the  devices were  made, home-
         owner interviews were undertaken, and other similar installations
         were comparatively  analyzed from reports.

     3.  From the above information, an analysis  of the efficacy and cost
         effectiveness of the off-line storage/Hydro Brake system as a CSO
         attenuation and flooding relief approach has  been prepared.

-------
                                SECTION 2

                               CONCLUSIONS
     This study has examined the performance  and  design  concept of the
Hydro Brake method of inlet control  for regulation  of peak runoff rates
and temporary storage of storm flows.   This application  was evaluated in
relation to its ability to reduce combined sewer  surcharge and resultant
combined overflows.

     This evaluation offers the following conclusions:

     1.  The Hydro Brake device does regulate flow  rates at relatively
         constant levels once an effective operating head has been de-
         veloped.  Conversely, the device behaves as an  orifice below
         the effective range of heads.   Hydro Brake flow rates above the
         effective operating head are substantially lower than those
         for an orifice or other clear opening of the same size.

     2.  The flow regulating capability of the Hydro Brake causes re-
         ductions in combined sewer overflow  peak rates  and total
         volumes by reducing the stormwater inflow  rate  to the sewers
         upstream of the control point and by delaying the drainage of
         storm runoff.

     3.  By removing the peak rate surge from the sewer  system, combined
         overflow pollutant loadings are reduced  because the  first flush
         effect is dampened.

     4.  Hydro Brake regulation of peak inflow rates is  effective in
         alleviating sewer surcharge and basement flooding problems.

     5.  Percentage reduction in peak flow rates  by the  Hydro Brake
         device is dependent upon the percentage  of total runoff which
         can be intercepted, as well as the level of control  and dis-
         charge rate desired.  Discharge rates may  be lower than avail-
         able excess capacity in receiving sewers,  depending  upon the
         availability of surface storage and/or the feasibility and
         expense of additional below grade storage.

     6.  Inlet control for purposes of storm  water  flow  rate  regulation
         may be accomplished with orifices,  but the orifice size must
         be smaller than the Hydro Brake that would be required for  the
         discharge rate desired.  The use of  an orifice  results  in a
         larger range of discharge rates and  a higher peak discharge
         rate over a given range of heads when compared  with  a  Hydro
         Brake of the same size as the orifice.

-------
     7.   For effective  application of the Hydro Brake control/storage
         technology,  it is  important that the design approach includes
         accurate characterization of drainage areas and sewer hydraulics
         to properly  identify  site specific release rate requirements.^
         The level  of control  desired determines the required storage
         volume,  and  the characteristics of the site determine whether
         to employ above grade or below grade storage, or a combination
        .thereof.

     8.   Where surface  ponding is an acceptable form of stormwater storage,
         the application of Hydro Brakes aloneris more cost-effective
         than Hydro Brakes  used in conjunction-with off-line, below
         grade storage  structures.  Both applications» however, are more
         cost-effective than other combined flow alternatives where both
         surcharging  and overflows are the prevailing problems.

     9.   The design of  inlet control/storage systems and the construction
         of below grade storage structures and related appurtenances are
         the major cost elements in the application of these systems.
         The cost of  the Hydro Brakes is a small portion of the total
         project  cost.

    10.   Because  the  Hydro  Brake is a specialty item, sufficient lead
         time must be allowed  for manufacturing and delivery delays.
       ' Installation of the device is relatively simple where the proper
         clearances have been  provided in control structures.

    11.   During the first 18 months of operation, the Hydro Brake control/
         retention structures  exhibited minimal maintenance requirements.
         Solids deposition  in  the storage tanks is almost negligible and
         should not increase through time.

    12.   Fouling  of small  diameter Hydro Brake control devices with stormwater
         debris is possible where inlets and catch basins are not trapped.
         This occurred  in both the Rochester and Cleveland projects.

    13.   The minimum  practical size of Hydro Brake devices appears to  be
         approximately  2 inches (5 cm), provided that inlet structures
         are trapped  or otherwise constructed  to capture debris this size
         and larger.   2 inch (5 cm) units were used  in catch basins  in
         Cleveland without any reported incidents of plugging.  Smaller
         sizes of Hydro Brakes are possible where stormwater flows are
         relatively clean,  as  has been suggested for downspout control/
         rooftop  storage systems.

     In summary,  the  Hydro Brake is an effective and feasible means  of con-
trolling storm water  flows. Applications are  practical  and  inexpensive
because the inlet control concept makes optimum use  of existing facilities.
Flexibility in stormwater storage options allows the concept to be adapted
to account for site differences and varying levels of control.  Finally,
potential cost savings  through use of  the Hydro Brake technology  offers
methods of stormwater control  where other alternatives are prohibitive
in cost.
                                   8

-------
                                SECTION 3

                             RECOMMENDATIONS
     As a result of this study,  certain design,  evaluation and opera-
tional recommendations  are proposed for existing and  future Hydro Brake
control/storage technology applications.

     1.  Design of Hydro Brake control/retention systems must include
         accurate simulation of storm inflow effects  on receiving sewers
         to properly identify maximum allowable  Hydro Brake discharge
         rates.  Receiving sewers should be monitored to check the
         validity of hydraulic modeling efforts  and to verify the proper
         selection of Hydro Brake discharge rates.

     2.  Design storm runoff characteristics must  be  properly identified
         in relation to selected Hydro Brake discharge rates.  An inflow/
         outflow continuing relationship is necessary to determine  the
         optimum size stormwater storage requirements.

     3.  Identification of site specific characteristics is necessary to
         properly identify runoff characteristics  as  well  as the best
         combination of surface and subsurface stormwater  storage.

     4.  Surcharge indicators and receiving sewer  level recorders should
         be utilized as necessary to identify potential applications
         and measure the effectiveness of existing installations.

     5.  Surface ponding effects of potential Hydro Brake  control/storage
         applications should be evaluated by selective installation of
         control devices in catch basins prior to  detail design.

     6.  Hydro Brake control/storage designs should be evaluated for  use
         in existing separate storm sewer systems  for sediment  control
         and receiving stream flooding relief.

     7.  Hydro Brake control/storage applications  should  be  considered
         in storm sewer design for cost savings  through  pipe size
         reduction, and to avoid transferring upstream flooding
         problems to downstream areas.

     8.  Hydro Brake regulated catch basins should be inspected frequently
         and cleaned as necessary to remove deposited debris.

     9.  Hydro Brake, regulated catch basins, as  well  as  unregulated inlets
         and catch basins, tributary to retention sturctures  should be
         designed to trap and prevent large debris and floatables  from
         interfering with Hydro Brake control devices.  Traps  should  also
         be employed to control the release of odors  from combined  sewers.

-------
10.  Retention tanks should be inspected  after  every major storm event
     to establish a history of debris  and silt  accumulation to deter-
     mine ongoing maintenance and inspection  requirements and verify
     unobstructed entry to Hydro Brake control  devices.

11.  Hydro Brake control/storage project  construction  inspection should
     include flooding or pressure testing of  plugs  inserted in abandoned
     catch basin leads where stormwater flows are to be  rerouted to
     control/storage structures.  Faulty  plugs  reduce  stormwater
     capture and effectively counteract the regulated  release function
     of the Hydro Brake.

12.  Periodic inspection is necessary  to  check  the  structural integrity,
     corrosion and performance of all  control/storage  structures and
     appurtenances.  Leaks in structures  and  abandoned catch basin
     leads should be identified and sealed to insure the control/
     storage system is operating as designed.
                                 10

-------
                                  SECTION 4

                            METHOD OF APPROACH

GENERAL PROCEDURES

     Various data collection and analysis tasks  were undertaken  to accom-
plish the evaluation objectives of this study.  The methods  described here
were directed towards measuring the performance  of the  Hydro Brake device
as well as evaluating its applicability in relation to  other methods of
storm water control.

     Field tasks were devised to record the ability of  the Hydro Brake
devices and storage structures to reduce peak storm flow rates and delay
storm drainage, thereby reducing receiving sewer surcharging and decreasing
the frequency and magnitude of combined sewer overflows. Additional tasks
included water quality monitoring to determine runoff pollutant  loadings
to receiving sewers, and sedimentation measurements in  the storage struc-
tures for evaluation of maintenance requirements.  Homeowner surveys were
performed as well to further document the effectiveness of storm water
control obtained.

     Data analysis efforts involved desk top modeling to evaluate design
conditions and measured storm events, and the derivation of  discharge curves
from field simulation of design storm conditions.  Further analysis was per-
formed to provide cost comparisons with alternate storm water control methods
and an evaluation of other applications of the Hydro Brake device.

DESCRIPTION OF HYDRO BRAKE CONTROL DEVICES

     The Hydro Brake is a proprietary flow regulator device  which purpor-
tedly acts as an energy dissipator by imparting." cvortex pattern to the flow
passing through the device.  It is a static device which is  said to develop
control energy from the head above the unit, and its physical geometry.   The
resistance to flow is described as increasing with increasing head, thereby
reducing the rate of increase of the discharge from the device.   This head-
discharge relationship results in a much "flatter" rating curve  when  compared
with the discharge from unrestricted openings or orifices of the same size.

     The Hydro Brake units examined in this study are described  as  the
"horizontal conical" type.  Each unit is constructed as a frustum of  a  cone,
having a sealed lower base and an open upper base, which is  the  discharge
side of the device, the diameter of which describes the size of the unit.
The cone is oriented horizontally such that its axis defines the effective
direction of flow.  Flow entry is accomplished through  a slot along the face
of the cone between the two bases.  The orientation of this  entry slot  and
the conical shape combine to produce the spiral flow pattern inherent with
the flow regulating capability of the Hydro Brake.  Photographs  of the  Hydro
Brakes presently in use in the study area may be found in Appendix D.
                                     11

-------
     During the course of this evaluation,  it was  determined that a reduc-
tion in the size of the Hydro Brake control  devices was required, in order
to reduce the discharge rates.  This subject is  discussed in detail in
Appendix F, Cleveland's Redesign Report.  This study, therefore, discusses
both the original design Hydro Brakes and the redesigned units.  All data
collection and evaluation tasks were performed for both the original design
and redesign conditions.

DESCRIPTION OF INSTALLATIONS

     Each Hydro Brake installation consists  gf- a storm water retention struc-
ture located at the low point of a drainage  area with a Hydro Brake regula-
tor device installed at the effluent end of  each structure.  Discharge is
to the existing combined sewers.  Minor storm sewer construction and plug-
ging of cath basin lead was accomplished in  the  immediate vicinity of the
Hydro Brake structures to direct runoff to  these units.  Catch basins
located in more remote locations of each drainage  area were modified through
installation of .05 cfs (1.4 L/s and .25 CFS (7.1  L/s) Hydro Brake devices.
When surface runoff rates exceed these values, storm  flows bypass the catch
basins, flow along the street gutter system, and drain to the retention
structures.

West 170th Street Installation

     The W. 170th Street Hydro Brake control structure consists of one 48
inch (1.2 M) diameter round corrugated metal pipe, (CMP), 163 feet  (50M)
long, sealed at both ends to form a tank.   The storage volume is approxi-
mately 2,000 ft3 (57M3).  Catch basins are  connected  through an 18  inch  (.45M)
pipe at both ends as shown in Figure 4.

     The tank is buried about seven feet  (2.1M)  to the invert.

     The Hydro Brake is located at the discharge end  of  the  tank and  is
inserted in a twelve inch (.30M) pipe which discharges to the 21 inch
(.53M) combined sewer.  There is no manhole at the junction  of  the  21  inch
(.53M) combined sewer and the twelve inch  (,30M) effluent line.

     The original Hydro Brake unit had a manufacturer's  discharge  rating
of 2.0 cfs (57 L/s).  On July 21, 1981, a  new  unit rated at  1.25 cfs  (35.4  L/s)
was installed.

West 177th Street Installation

     The W. 177th Street Hydro Brake control structure consists of two  156
foot (47.5M) long, 87" x 63"  (2.2 M x 1.6M) cross  section corrugated  metal
arch pipes, buried about eight feet  (2.4M)  to  the  invert, with  a  total  volume
of 10,000  ft3  (283M3).  The tanks have a series  of catch basins connected
to them as shown in Figure 5.  The two tanks are also connected together
by a 24 inch  C.61M) CMP.  The tanks  drain through  an 18  inch (.46M) CMP to
a manhole  where another 18 inch (.46M) RCP pipe  is connected,  tying in  a
series of  catch basins.  The  Hydro Brake unit is inserted in the 12 inch
(.30M) effluent line from this manhole and  is  drained to the 18 inch  (.46M)

                                     12

-------
%
            LEGEND
          (NOT TO SCALE)
          COMBINED SEWER
          STORM SEWER
          CATCH BASIN
          DIRECTION OF FLOW

                                YDROBRAKE  LOCATION
                                        63'-48"(49.7m- 122cm.)
                                            21" COMBINED SEWER
                 /
                                          (53cm.)
                               W 170 ST.
                          PLAN   VIEW
T!
 I
 I
   
-------
       LEGEND
     (NOT TO SCALE)
     COMBINED SEWER
     STORM SEWER
     CATCH BASIN
     DIRECTION OF FLOW
                        HYDROBRAKE  LOCATION
             W 177 ST.
           156'-  87 "X 63" (47.5m.-22!XI60cm.) O
            156'— 87 "X 63" (475m.-221X 160cm.) Q
                PLAN  VIEW
i   w
                      PROFILE   VIEW
	*•  J? J/_7_ ST.


iri
H
i|
^w
	 IT
_J!
I ' 1
                                                  HYDROBRAKE LOCATION
 FIGURE 5       W. 177 ST.     INSTALLATION
                                  14

-------
combined sewer on denshire Avenue.

     The original Hydro Brake unit that was  installed  had a manufacturer's
rating of 1.5 cfs (42 L/s).  A replacement unit  was  installed with a rating
of 0.25 cfs (7.1 L/s), on July 21, 1981.

Puritas Avenue Installation

     The Puritas Avenue Hydro Brake control  structure  is located under the
eastbound curb lane on Puritas Avenue between" W.  170th and W. 172nd Streets.
The storage tank is a corrugated arch pipe 170- feet  (52M) long with a cross
section of 95" x 67" (2.4M x 1.7M).   It is buried about ten feet (3M) to
the invert.  Total volume is 5,800 ft3.(164M3).   At  the downstream end
the tank there is an 18 inch (.46M)  spiral corrugated  pipe leading to a man-
hole containing the Hydro Brake unit.  This  manhole  has an invert approximately
3.3 feet (1M) below the invert of the tank.   An  18 inch (.46M) effluent line
from the Hydro Brake manhole discharges to the 3' 6" (1.1M) brick combined
sewer.  There is no manhole at that point, making access to the 18 inch (.46M)
effluent line very difficult.  An illustration of this control structure  can
be seen in Figure 6.

     The original installation consisted of  7.0  cfs  (197 L/s) rated Hydro
Brake.  On August 1, 1981, a new unit rated  at 1.0 cfs (28 L/s) was instal-
led.

SIMULATION OF DESIGN CONDITIONS

Introduction

     In order to simulate design storm conditions (full storage utilization
and maximum head on the Hydro Brake) and to  field calibrate the Hydro Brake,
each storage tank was filled from street fire hydrants and a  record of the
drain down time was made.  Data collected was used to  derive  discharge
relationships as described below.

     The total storage volume at each location was calculated,  and then was
reduced to incremental volumes for each one-inch (2.5  cm)  reduction in water
depth.  Since two of the storage tanks were  corrugated arch pipe, and the
third was corrugated round pipe, geometric formulae were used to calculate
incremental volumes (Appendix A).

     All influent lines were plugged, as was the Hydro Brake, in each tank,
and the tank was filled from fire hydrants.   After filling, water  levels
in each structure were observed.  If the water level was observed  to  drop
prior to pulling the Hydro Brake plug, leaks in  the system or in  the  plug
were identified and corrected, if possible.   The Hydro Brake  plug was  then
pulled, and the time for each one-inch (2.5  cm)  drop in water level was
recorded, using a surveyor's level rod inserted  in each storage tank  at
the discharge point and a digital totalizing stop watch.  Automatic  level
recorders simultaneously recorded the fill and;drain procedures.   The effluent
elevation was taken as the invert elevation  of each Hydro  Brake device.   Each
installation was tested in this manner, and  retested when  revisions were  made
in Hydro Brake sizes.
                                    15

-------
           LEGEND
         (NOT TO SCALE)

         COMBINED SEWER

         STORM SEWER

         CATCH BASIN

         DIRECTION OF FLOW
  PU RITAS   AVE.
                            42"(
            07cm) COMBINED  SEWER
                       Q     170'- 95" X  67" (52m.-24IXI70cm.)  QJ
                        -HYDROBRAKE  LOCATION
                              PLAN  VIEW
    PURITAS   AVE.
          rr~
          ii
HYDROBRAKE LOCATION
  FIGURE 6
II
                           PROFILE   VIEW
    STORAGE  TANK

1701-95"X67"(52m.-24IX170cm.)
                             M
	                   I   I
 	I   L
                       42."{l07ctn)COMBINED  SEWER
                                                  ~rr
   PURITAS  AVENUE  INSTALLATION
                                    16

-------
     Before performing discharge tests  for each  Hydro  Brake structure,
head vs. volume relationships were computed for  each retention  tank.
Manufacturer's standard arch pipe cross sections were  used for  the  Puritas
Avenue and W. 177th Street locations.   The W.  177th Street tank is  a
48 inch diameter round pipe.  All three tanks  were constructed  with a
sTope of 0.5%.

    . For calculating incremental volumes, it was assumed  that the water
surface in the retention structures would be .horizontal during  storm
water storage conditions.  Profile drawings o-Tthe tanks  were oriented
at a slope of 0.5% and horizontal planes, representing one inch (2.54 cm)
increments in water depth, were constructed passing through the tanks.
Cumulative volumes were then computed  by increments through the full
depth of the tanks, primarily by using  the formula for an ungula of a
cylinder.  The circular cross section  defines  a  simple cylinder, and in
the case of the arch pipe tanks, the cross section defines a volume which
is a composite of partial cylinders having three different radii.   Sec-
tions through manholes and connecting  storm pipes were accounted for as
required.  Formulae and head vs. volume tables may be  found in  Appendix  A.

     Calculating incremental volumes of water  lost as  a function of time
for incremental changes in head produced a discharge rate curve for each
installation.  This curve was, of course, a variable head test  curve, based
upon a full head in the tank, and no additional  water  input during  the test.

     Each installation was tested in this manner, and  retested  when revi-
sions were made in Hydro Brake sizes.

Discussions of Calibration Procedures

Uest 170th Street Installation:

     For the calibration tests of the original unit as well as  the  new unit,
the Hydro Brake had to be removed from the tank to gain access  to  the pipe.
A sewer plug was allowed to fall into the combined sewer when  it was  released
to avoid obstructing the flow in the effluent pipe during the  drain down
test.  Calibration was then performed in accordance with  procedures described
previously.  In both cases the 48 inch (1.2M)  CMP was  filled  to the crown.
A catch basin on Martha Avenue leaked during the first test.   The  leakage
occurred at the plug intended to isolate it from the combined  interceptor.

West 177th Street Installation:

     In the manhole on denshire where the 12 inch (.30M) effluent line
meets the 18 inch (.46M) interceptor a 12 inch  (,30M)  sewer plug was
installed.  A plug was also installed in the 18 inch  (.46M)  RCP in the
Hydro Brake manhole to isolate the catch basins.   Calibration was  then
performed.  There were no leaks within the tanks.
                                    17

-------
     After the new Hydro Brake unit was  installed,  a recalibration was per-
formed.  The 18 inch (.46M)  RCP from the catch  basins was not plugged for
this test, and the additional  volume associated with these structures was
calculated from construction plans  when  the discharge curve was calculated.
Subsequently, a catch basin  at the  northeast corner of  Glenshire and W. 177th
was found to have a bad plug at the point where it  was  isolated from the
combined sewer.  This severe leak allowed the tanks to  be filled only to
about 3* 2" (.96M).  Therefore, only a partial  discharge curve was measured.
This causes calculated discharge values  to be too high  for the higher head
value.

Puritas Avenue Installation:

     The calibration of the  original unit was done  by wrapping foam rubber
around an 18 inch (.46M) sewer bag  and inserting it into the spiral corru-
gated pipe.  While this plug leaked slightly around the corrugations of
the pipe, the tank was still able to be  filled  to six feet (1.8M).  Cali-
bration was then performed as previously described.

     A recalibration was performed  on the new 1.0 cfs (28 L/s) Hydro Brake
unit, installed on August 1, 1981.   The  18 inch (.46M)  effluent line at the
combined sewer was plugged and the  tank  was filled.

     As the level in the tank approached 1.7 feet  (.52M) the plug  blew.
Following reinstallation, the maximum safe air  pressure was exceeded during
an attempt to better seal the discharge  line.  The  filling proceeded to
about 2.6 feet (.79M) when the plug blew out again. At this level there
was approximately 6 feet (1.8M) of  head  on the  Hydro Brake.  The filling
was stopped at that level and calibration was continued.

STORM FLOW MONITORING

Equipment

Bristol bubbler level recorders were installed  in  each  tank and at CSO
chamber M-15.  At West 170th and West 177th Streets, strip charts  set for
one-half inch (1.3cm) per hour were used.  Twenty-four  hour circular  charts
were used at the Puritas tank and at the Puritas overflow.  ISCO  1870
bubbler level recorders replaced the Bristol meters at  West 170th  and 177th
Streets for approximately the last  two months of the study, because of  re-
current problems with the air supply systems in those two meters.  All
equipment was installed in concrete meter vaults which  were placed on tree
lawns adjacent to the storage structures and connected  to  the  structures
with underground 3 inch (7.6cm) PVC conduit. Each  structure was  equipped
with hinged, locking steel access doors.  Monitoring  lines were secured in
the storage structures at the invert elevation  of  the Hyrdo Brakes at the
discharge end of each storage structure, and were  connected to the monitoring
equipment through the PVC conduit.   In  the Puritas  Avenue  structure, monitoring
lines were installed in the invert of the storage  tank  at  the  effluent  end,
approximately 3.5 feet (1.1M) above the  invert  elevation of the Hydro Brake
control device.

                                     18

-------
Data Collection

     Charts were pulled weekly or after storm events,  and  field observations
and equipment adjustments were noted.   The zero  level  point  for each level
recorder was adjusted so that the level record represented depth of storm
water above the invert elevation of the Hydro Brake  devices.   In the Puritas
Avenue structure, the zero level point was set at  the  effluent end invert
elevation of the storage tank.  At CSO M-15,  the difference  in elevation
of the bubbler tube and the overflow weir wall was measured  so that a chart
correction could be applied to the level  recordings  to accurately define any
potential overflow hydrographs.  Charts and field  notes for  each meter were
maintained in a common file.

Analysis

     Using the derived discharge curves for each location, outflow hydrographs
were prepared for significant storm events.  These hydrographs were com-
pared with inflow hydrographs derived  from recorded  rainfall  data to
evaluate peak storm flow rate attenuation and storage  utilization.

PRECIPITATION RECORDS

Equipment

     A Bel fort recording rain gauge with a twelve  inch (30 cm), double
traverse, eight-day chart was installed at Fire  Station No.  43 on Rocky
River Drive, approximately .25 mi (.4  km) from the center  of the study-
area.

Data Collection

     Charts were pulled weekly or after storm events,  and  concurrent storm
information was obtained from the National Weather Service and the Northeast
Ohio Regional Sewer District (NEOP.SD)  for storm  events. This supplemental
rainfall information was used during the Summer, 1981  monitoring period.

Analysis

     Hyetographs of rainfall data were converted into  run-off (inflow)
hydrographs for correlation with level and discharge data  from the  in-tank
level recorders.

UATER QUALITY SAMPLING

Equipment

     ISCO 1680 programmable sequential samplers  were installed in  the  same
manner as the flow level records in each end of the  West 170th Street  and
the Puritas Avenue storage tanks, and at the discharge end only  of the West
177th Street tank, because the influent line entered at that end also.
Level actuated Mercury float switches initiated  a variable time  contact
closure sequence to pulse the samplers.

                                    19

-------
Data Collection

     Upon initiation, the samplers collected discrete  samples  at intervals
of 2.5, 5, 10, 15, 30 and 60 minutes.   At the end  of the  first hour, a
relay eliminated intervening signals  and only 30 minute samples were col-
lected for the duration of the storm,  up to  a maximum  of  eleven hours unless
the 28 bottle collector was replaced.

     To accomplish the designated sample frequency,  a  timer was developed
which would activate sampling equipment on the..variable time schedule
described above.  The timing mechanism was activated by a mercury  level
switch which closed a circuit when storm flows were  detected in the storage
structures.

Analysis

     Discrete samples were analyzed for biochemical  oxygen demand  (BODs),
volatile suspended solids and total suspended solids.  Samples were com-
posited and analyzed for total organic carbon, chemical oxygen demand,
chlorides, sulfates, copper, cadmium,  chromium, lead and  zinc.

     Results were compiled and evaluated to  determine  whether  storm water
retention and peak attenuation could be shown.to have  an  effect on receiv-
ing water quality.

SURVEY OF SERVICE AREA POPULATION

Data Collection

     Two survey questionnaires (Appendix B)  were developed for the home-
owners of the study area.  The surveys were  conducted  as  house-to-house
surveys.  Residents were asked for information on flooding history prior
to construction.  The second survey was conducted at the  end ot the study,
and residents were asked about flooding during the study  period.

Analysis

     Homeowners' pre and post-construction responses were mapped  and  an
assessment of flooding conditions was developed.  Although post-construc-
tion responses indicate a reduction in flooding of basements  and  streets
during the study period, it is not possible  to prepare a  definitive
analysis of project effectiveness after only one year.

DATA ANALYSIS

     As noted previously, discharge hydrographs were computed using  water
level records from actual storm events and the stage discharge curves deve-
loped for each Hydro Drake.  Calculations of  inflow hydrographs,  represent-
ing surface runoff from the drainage area contributing to each retention
structure, were performed using the linearized subhydrograph modification
of the rational method proposed by Chien.(2)

                                      20  '

-------
     One-half hour duration storms were selected for evaluating the level
of control provided by each unit because the time of concentration  for
each drainage area approximates this duration.   Rainfall  amounts were
determined from the U.S. Weather Bureau Technical Paper #40,  "Rainfall
Frequency Atlas of the United States" for 1/2 hour storms having return
periods of one, two, five and ten years.  The standard time distribution
of storm rainfall (Huff, 1967)(3> was used to provide rain increment input
into the linearized sub-hydrograph method for computation of inflow hydro-
graphs for each retention structure.  Discharge hydrographs for these design
storms were then computed using the discharge" curves for each Hydro Brake
and the storage continuity equation:

     I - 0 = ds
             It

Where:  I is inflow rate

        0 is outflow (discharge) rate

        ds/dt is the change in storage per time increment

     A more thorough discussion of the methods introduced here and  the
formulae involved are presented with the discussion of results presented
in Section 6.
                                     21

-------
                                SECTION 5

                                 RESULTS
DISCHARGE CURVES
     Data collected in the variable head tests  described  in  Section  4 were
used to construct discharge curves for both original  and  replacement Hydro
Brake control units.  A plot of head v.s. flow-rate  was contructed from
the field record of falling head v.s. time and  calculations  of  partial volumes
v.s. incremental changes in head.  The head values used were averaged for each
increment of volume discharge, and flow rates were computed  from measured time
increments for each corresponding volume increment.   The  datum  point for each
Hydro Brake was taken as the invert of each control  device.   Best fit curves
were derived for each set of Hydro Brake test values  using a commercially
available curve fit program.  Head discharge data pairs were input and
regression analysis produced coefficients defining the equation of the best
fit curvilinear relationship between head and discharge valves.  For each
Hydro Brake, a discharge curve for an orifice having the  same size opening was
calculated, using the equation:  Q = .6a (2gh)l/2.

Where:  .6   is an orifice coefficient
         a   is the cross sectional area of the orifice
         g   is the force of gravity
         h   is the head on the orifice

     Figures 7 and 8 represent the discharge curves  developed for the
W. 170th Street structure, before and after redesign of the  Hydro Brake
control unit, respectively.  The curve for the  original device  indicates
a considerably higher discharge rate than the manufacturer's 2.0  cfs
(57 L/s) rating.  This can be explained in part by  the faulty catch
basin plug on Martha Avenue, which would increase the apparent  discharge
rate for the higher head values.  Repairs accomplished by the City of
Cleveland corrected this problem as indicated by the discharge  curve
for the redesigned control unit, which is rated at  1.25 cfs  (35.4 L/s).
Discharge rates at the higher heads indicate the new device  is  performing
close to its rating.

     The W. 177th Street discharge tests resulted in the  discharge  curves
presented in Figures 9 and 10.  The original Hydro  Brake  device exhibits
a lower discharge rate than an orifice of the same  size (7 in.  -  17.8  cm)
but considerably higher than the manufacturer's rating of 1.5 cfs (42  L/s).
This is also the case for the redesigned unit,  rated at .25  cfs (7.1 L/s),
but a leaking catch basin plug on Glenshire Avenue,  discovered  during
the second test, produces significant errors in the range of higher  heads,
heads, causing the discharge rates through the  Hydro Brake to appear to  be
too high.
                                      22

-------
       -(300)
    10-
    9-1
       -(250)
    7-
  ~- 5-
     ^
OC
UJ
  £4-
    3-
       _(200)
       -(150)
      	9" (23cm.) ORIFICE
      	^  9" (23cm.) HYDROBRAKE
                                CATCH BASIN LEAK
                                            MANUFACTURERS  RATING
                                            2.0 cfs  (57L/S)
                    \
I
 FIGURE 7
                               I
                  2'            3'
                 (.61)           (.91)
                HEAD IN FEET (METERS)
W.I70 ST. HEAD-DISCHARGE  CURVES : ORIGINAL DESIGN
                  (.30)
 I
 4'
(1.22)
 I
 5'
(1.52)
                                      23

-------
   2 —
CO
LU —-
X «   -
S U-   ^
co °
S
       -(75)
                                       	^r-    6"(l5cm.) ORIFICE
                                       	    6"(l5cm.) HYDROBRAKE
       -(50)
     --(25)
                                                     MANUFACTURER'S  RATING
                                                     125 cfs( 35.4 L/S)
                    I
 FIGURE 8
                  1           ^
                  2'            3'
                (.61)           (.91)
                HEAD  IN FEET (METERS)
W.I70 ST HEAD-DISCHARGE  CURVES: REDESIGN
                   (.30)
                                                          I
                                                          4'
                                                         (1.22)
  I
  5'
(1.52)
                                      24

-------
                                                  7 (18cm.)  ORIFICE
                                                 7" (18cm.)  HYDROBRAKE
     ' -75
      -50
                                                 MANUFACTURERS   RATING
tr co
a,!*
CD *-
       -25
                                                 1.5 Cts (42L/S)
                   I
                   I'
                  (.30)
                                I            I
                                2'           3'
                               (.61)          (.91)
                              HEAD IN FEET(METERS)
  I
  4'
(1.22)
 I
 5'
(1.52)
FIGURE 9
              W. 177 ST.   HEAD- DISCHARGE  CURVES :  ORIGINAL DESIGN
                                     25

-------
    1.5-1
    1.4-
        -(40)
    1.3-
        -(35)
    1.2-
        -(30)
    1.0-
    0.9-
        -(25)
    0.8-
co
o
 ~ o.7-r
    0.6 H
oen
cou.
    0.5-
    0.4-
    0.3-
        -(20)
        -(15)
        -(10)
                             	3"( 7.6cm.)  ORIFICE



                            	^   3" (7.6cm.)  HYDROBRAKE
                                       (CATCH BASIN LEAK)
                                                 MANUFACTURER'S  RATING
    0.2-
        -(5)
    0.1-
                                                 0.25 cfs (71 L/S)
                    (.30)
                   I             1


                  2*             3'

                 (.61)           (.91)

                   HEAD IN FEET (METERS)
 FIGURE 10
W. 177 ST.   HEAD-  DISCHARGE  CURVES



                        26
 4

(1.22)
REDESIGN
  5

(1.52)

-------
    14 —
    ra-
    10 —
    8 —
tO
LJ
 :_ 7
         -400
        — 350
        —300
          -250
       -200
                                                   	I6"(4lcm) ORIFICE

                                                   .	  16" (41cm) HYDROBRAKE
                          MANUFACTURER'S  RATING
CO
o
    4 —
    3 —
    2 —
        I
      -1-50
       I
                                                               70cts (I97L/S)
 \       I
 12
(.30)    (.61)
 1
  3
(.91)
  I      I
 45
(1.22)   (1.52)
                                              I       I
                                              67
                                             (1.83)  (2.13)
                                                              I       T^
                                                              89
                                                             (2.44)  (274}
10
FIGURE  II
                                   HEAD IN FEET (METERS)

               PURITAS AVE. DISCHARGE  CURVES : ORIGINAL DESIGN
                                          27

-------
                                      	5.5" (14cm) ORIFICE
                                      	-. 5.5"(l4cm)  HYDROBRAKE
    4-
     -f —
v>
iLl
UJ
< o
x    g"
o    c~
(f>
     I-
       -100
        - 50
                                               MANUFACTURER'S  RATING
                                               1.0 CFS  (28L/S)
 I      I
 I      2
(.30)   (.61)
                          I
                          3
                         (.91)
  I
 4
(1.22)
  I
  5
(152)
  I
  6
(1.83)
 7
(2.13)
 1       I
 8      9
(2.44)  (2.74)
                         HEAD IN  FEET  (METERS)
FIGURE 12      PURITAS AVE.   DISCHARGE

                                      28
              CURVES  —   REDESIGN

-------
     Figures 11 and 12 illustrate the discharge curves  for  the  Puritas
Avenue structure.  The downstream end of the retention  tank is  approxi-
3.5 feet (1.1 M) above the invert of the Hydro Brake control  unit  (see
Figure 6), so the effective operating range of heads on the control
unit is 3.5 - 9.0 feet (1.1 - 2.7 M).  Essentially no control occurs  between
0 f 3.5 feet (0-1.1 M)  of head because the volume involved is negligible
(44 ft3  - 1245 L).

     The discharge curve  in Figure 11 indicates that the original
control device performed  close to its rating of 7.0 cfs (198 L/s)  for
actual heads of 4 to 8 feet (1.2 - 2.4 M).   Data points from the second
discharge test (Figure 12) indicate that the discharge  rate began  to
stabilize around 2.0 cfs  (56.6 L/s)5 when the head on the control  unit
was 5-6 feet (1.5 - 1.8 M) or 1.5 - 2.5 feet (0.4 - 0.8 M)  depth in
the storage structure.  No measurements above this level were taken
because of the problems encountered during  the field test.

STORM HYDROGRAPHS

     In order to assess the performance of the Hydro Brake  structures
and control devices under actual operating conditions,  rainfall and
depth of flow measurements were performed for several storms.

Measured Storm Events

     Field records of actual storm events during the Fall,  1980 and  Summer,
1981 monitoring periods were disappointing in that the  rain events were  not  of
sufficient intensity to cause appreciable response in the storage/control
structures.  Before sampling equipment was  installed for the Summer,  1981
monitoring period, three moderate rainfall  events were  recorded on the  rain
gauge at Fire Station 43 on Rocky River Drive.  Total rainfall  of  1.87"
(4.75 cm) occurred June 8-9, 1981, 1.03" (2.62 cm) occurred June 22,  1981
and .78" (1.98cm) was recorded for June 25, 1981.  Peak intensities  were
approximately 2.3 inch/hour (5.8 cm/hr), 0.5 inch/hour  (1.3 cm/hr) and  0.4
inch/hour (1.0 cm/hr), respectively.  All three storms  caused slight short
duration overflows at the combined sewer overflow M-15  on Puritas  Avenue,
but it was not possible to estimate the involvement of  the  Hydro Brake  control
structures during these storms.  The-drainage areas which directly contribute
flows to the retention/control structures account for less  than 10 (ten)
percent of the total drainage area tributary to this overflow point.   No over-
flows at M-15 occurred while sampling equipment was in  place at the  control
structure.

     Three storm events which best represent the response of the Hydro
Brake structures to peak flow rates were selected for illustrative
purposes.  These events occurred June 8-9, July 13 and  August 7, 1981.
The Puritas Avenus structure is not discussed here because none of the
storms caused any appreciable response in this unit.  Storm intensities were
not of sufficient magnitude to produce stormwater inflow rates significantly
storms caused any appreciable response in this unit. Storm intensities  were
not of sufficient magnitude, to produce stormwater inflow rates significantly
greater than the Hydro Brake discharge rate.  Therefore, records of stormwater
levels in the storage tank  indicated negligible amounts of storage.

                                      29

-------
     Hydrographs were constructed for the W.  177th  Street  unit  for  the
first and second storms.  The 1.5 cfs (42 L/s)  control  device from  the
original design was in place for these events.   The W.  170th Street
unit was evaluated for the second and third storms.  The 2.0 cfs  -
(57 L/s) device was in place for the July 13 storm  and  the 1.25 cfs
(35.4 L/s) unit was in use for the August 7 storm.

Rainfall Data

     In attempting to construct hydrographs far actual  storm events to
illustrate inflow/outflow relationships for the'Hydro Brake installations,
it was found that the 8 day rain records from the Rocky River Drive
location could not be used to identify short duration rainfall  increments.
Subsequently, rainfall records were obtained from the National  Weather
service for Cleveland-Hopkins Airport and from Northeast Ohio Regional  Sewer
District for recording stations at John Marshall High School and  Brookpark
City Hall (See Figure 1).  Though these locations are 2-4 miles from the  study
area, they were expected to exhibit similar rainfall patterns and provided
records which could be reduced to five minute increments for use  in con-
structing inflow hydrographs.

Inflow Hydrographs

     A modification of the rational method was used to calculate  inflow
hydrographs for each drainage area, similar to the linearized sub-
hydrograph method proposed by Chien.^2)  Five minute rainfall  increments
were applied to the drainage area tributary to each structure  to  con-
struct subhydrographs, which were then superpositioned with a  five minute
delay between successive subhydrographs to construct the total  inflow
hydrograph.

     The formula used to construct each subhydrograph is:

          Q = CIA (2 tr)/(tr + tc)

Where Q = peak flow rate in cfs at time = tr
      C = runoff coefficient
      I = rainfall intensity in in/hr
      A = contributing drainage area in acres
     tr = time duration of rainfall increments
     tc = time of concentration of contributing drainage area

     The base of each subhydrograph is the sum of tr and tc.

     From Cleveland's redesign report  (Appendix F), the runoff coefficient and
time of area was estimated to be  .5 and 25 minutes, respectively.  Drainage
areas were calculated as the sum of street surface  and a 30 foot (9.1 M) setback
to  residential structures.  Table 1 summarizes  these values for each Hydro Brake
structure.
                                    30

-------
               TABLE 1.  HYDRO BRAKE DRAINAGE AREAS
                            Total  Area       Area to Hydro  Brake
Location	Acres (hectares)	Acres (hectares)
W. ^70 St.
W. 177 St.
Puritas Ave.
5.7
7.5
7.9
(2.3)
(3.0)
(3.2)
2.4
3.1
3.5
(1.0)
(1.2)
(1.4)
     The difference between total  area and area "directly contributing
to the Hydro Brake structures is the roof area and yard  area  which
drains to the combined sewers by means of area drains,  downspout con-
nections and footer drains.

Discharge Hydrographs

     Discharge hydrographs were constructed from values  of head vs.
time recorded during storm events and the head-discharge curves which
were derived for each Hydro Brake.  Inflow and outflow hydrographs
were then plotted for comparison purposes for the selected storms
mentioned previously.  Volumes were calculated by summing the areas
of trapezoidal sections under each curve.

     Figure 13 shows hydrographs at the W. 177th St. structure for  the
storm of June 8-9, 1981.  The original design Hydro Brake (1.5 cfs  -
42 L/s) was in place for this storm event.  The rainfall pattern represents
an average of the records from Cleveland Hopkins Airport and  John Marshall
High School, because this produces an invlow hydrograph which best  matches
the outflow pattern observed at the study area which is between these  two
rain gauge locations.  Because of problems with the level recorder,  the
outflow hydrograph constructed from the field record indicates too  small  a
volume for the rain observed for this event.  Therefore, a discharge hydrograph
was also calculated to illustrate the probable outflow pattern for  this storm.
This was accomplished using the calculated inflow hydrograph  and the storage
continuity equation:


     I - 0 = ds/dt

Where I = inflow
      0 = outflow
     ds = change in storage
     dt = time increment
                                      31

-------
      .-(5)
       -(75)
   2.5-
   2.0-
o

I
1.5-
    1.0-
   0.5-
                                                   RAINFALL INTENSITY

                                                   IN/HR.(cm./HR.)
                     •INFLJOW
                      (8382 FT.  -237,400 L)

                     -MEASURED OUTFLJOW
                      (4905 FT.3-138,900 L)
                   •CALCULATED OUTFLOW
                   (8181 FT.3-231,700 L)
       -(50)
       -(25)
                    2400
                                 1
                                0100

                               TIME OF DAY
  I
0200
0300
 FIGURE 13
           HYDROGRAPHS AT W.I77ST CONTROL STRUCTURE
           STORM OF JUNE  8-9,1981 -1.5 CFS (42 L/s) HYDROBRAKE

                                 32

-------
     These problems aside, this storm was  selected  to  illustrate the
ability of the Hydro Brake control  device  to  significantly  reduce  high
storm flow rates (the second peak in Figure 13), while lower flow  rates
pass through the unit with little control  evident.

  -,  Figures 14 and 15 present hydrographs at the W. 177th  Street  Ind
W. 170th Street locations, respectively, for  the storm of July  13,  1981.
The original design Hydro Brakes were still in use  at  this  time.
Rainfall records varied considerably among the four gauging locations,
such that inflow and outflow hydrographs did  no.t correlate  well.
It was noted that the depth of flow in all three retention  structures
reached a maximum of only 1-1.5 feet (.30 - .45 M),  well  below the
optimum operating range of the control devices (approximately 3-5  feet
(0.9 - 1.5 M)).  Though some flow attenuation occurred, it  appears
that only sustained high intensity storms  will produce any  substantial
peak rate reduction and storage volume utilization. This is because
of the relatively small drainage areas tributary to the control  points,
and the high rate operating ranges of the  Hydro Brake  devices.   Low
to moderate intensity storms (less than 1  in/hr or  2.5 cm/hr) will
exhibit unimpeded flow through the control structures, as was the  case
with most of the storms observed during the monitoring periods.

     Figure 16 shows hydrographs at the W. 170th Street location for
the storm of August 7, 1981, with very little control  of the storm flow
occurring.  This is as expected when the rainfall  record (Cleveland
Hopkins) is examined.  Intermittent high intensity  rainfall is  evident,
but is not sufficient to generate a high rate of surface runoff.

     The August 7 storm occurred after installation of the  redesigned
Hydro Brake control devices.  The flow record from the Puritas  Avenue
structure shows insignificant response to  this storm,  and no  records
were available from the W. 177th Street structure.   Shortly after  the
installation of the redesigned control unit at W.  177th Street  in  mid July,
1981, the device became plugged with a styrofoam  cup,  and the  retention
structure was 30% full of water for the remainder  of the monitoring period.
It was not possible to clear the unit before the  termination  of field investi-
gations in August, 1981.

STORM WATER QUALITY

Sampling Results

      It was originally intended to collect discrete samples of influent
and effluent flows for storm events at the three Hydro Brake structures.
as described in Section 4   Frequency of sampling  was 2.5,  5,  10,  15 and
30 minutes after the onset of a storm event, to observe any first flush
effects and loadings on the receiving combined sewer system.   Subsequent
samples after the  initial sequence were taken every half hour.
                                      33

-------

    2- '(5)
    3-
      -(75)
  2.5-
  2.0-
   15-
   1.0-
   .5 —
      -(50)
      -(25)
     0500
FIGURE  14
                             RAINFALL INTENSITY
                             IN/HR.(cmyHR.)
                             (JOHN MARSHALL)
                                            INFLOW
                               (4338 FT0-122,800 L)
                            —OUTFLOW,
                                            (2460FT. -69^700L)
      0600
0700
 I
0800
0900
                  TIME OF DAY
HYDROGRAPHS AT W.I77ST CONTROL STRUCTURE
STORM OF 7-13-81 - I.5CFS(42LA) HYDROBRAKE
                                    34

-------
                                    RAINFALL INTENSITY
                                    IN/HR.(cm/HR.)
                                   (BROOKPARK)
                                            INFLOW
                                            (6738 FT.  -190,800 L)
                                    	OUTFLOW
                                             (8304FT-235.200L)
     0600
     0800

TIME OF DAY
0900
1000
FIGURE 15     HYDROGRAPHS AT W. 170 ST. CONTROL STRUCTURE
             STORM OF 7-13-81 — 2CFS (57L/s) HYDROBRAKE

                                   35

-------
UJ
I
      2-
     1.0-
     0.9-
     08-
ci    0.7-
     0.6'
     0.5-
     0.4
     0.3--
     0.2-
     0.1'
             U
         (5)
         •(30)
         (20)
         (10)
                   u
                       RAINFALL INTENSITY
                       (CLEVELAND HOPKINS)
                       IN/HR.(cm./HR.)
                                        	INFLDW(3765 FT -106,600 L)

                                        	.OUTFLOW (3179 FT3- 90,000 L)
       1500
1600
1700
1800
1900
 FIGURE 16     HYDROGRAPHS AT W.I70ST CONTROL STRUCTURE
               STORM  OF AUGUST 7,198I-2CFS (57L/s) HYDROBRAKE

                                       36

-------
Sampling Results

     Samples were collected for  storm  events  as  summarized in Table 2.  Data
presented includes the range and average  of values measured at the locations
indicated.  Complete sampling results  are presented  in  Appendix C.. All values
represent samples taken at the effluent end of each  storage structure, immediately
upstream of the Hydro Brake control  devices.

     Due to the low levels and short duration of flows  in the, storage structures,
in addition to occasional  equipment  malfunctions,  it was not  possible to collect
samples from all locations for all storm  events.   In all but  one case, the
analyses of the samples collected did  not show any significant first flush
effects. -The loading on the combined  sewers  appears representative of normal
stormwater flows, as the comparison  in Table  2 indicates.

     The sampling also provides  no  indications of removals or deposition within
the storage structures, nor was  it possible to use the  sampling data to project
potential solids deposition problems.  This is partly because the  structures
have multiple influent points, and  in  the W.  170th Street and W. 177th Street
locations, part of the influent  flow enters the  storage structures immediately
adjacent to the effluent point.

Sediment Observations

     In addition to the sampling results, field  observations  and measurements of
the depth of solids, sediment and water along the length of each storage tank
indicate that solids deposition  and  debris accumulation should not be a problem
in the storage structures.  The  only exception is the redesigned Hydro Brake
control unit at W. 177th Street, which could  become  clogged with debris, as was
the case shortly after this unit was installed in July, 1981. This  is apparently
due to the small clear opening in this device.  The  Hydro Brake has  an opening
approximately 3 inches (7.16 cm) in  diameter, and became plugged with a styro-
foam cup on or about July 21, 1981.  This size opening  is smaller  than the
openings in the other Hydro Brake control devices and is, therefore, more sus-
ceptible to plugging.  Subsequent installation of catch basin traps  by the  City
of Cleveland should correct this problem.

     Photographs taken in each control structure during Fall, 1981,  along
with a record of solids depth measurements are presented  in  Appendix D.
These data were recorded after ths  structures has been  operational for
approximately 18 months, and indicate  minimal accumulation  of solids in the
retention structures over this time  period.

HOMEOWNER SURVEYS

     Two house-to-house surveys  were conducted during the  investi-
gation.  The first survey was conducted  prior to completion of con-
struction, and requested information on  past flooding experience.
The second survey was conducted during the first week of October,
1981, and requested information on post-construction flooding.
                                     37

-------
              TABLE 2 - SUMMARY OF SAMPLING RESULTS
                        STORM WATER PARAMETERS FROM
                                                            AND COMPARISON WITH
                                                            SELECTED LOCATIONS.
to
oo
LOCATION
Cleveland W  170
Cleveland W  170
Cleveland W  170
Cleveland W  177
Cleveland W  177
Ann Arbor, MI
Castro Valley, CA
Des .Moines,  Iowa
Durham, N.C.
Los Angeles, CA
Madison, WI
New Orleans, LA
Roanoke, VA
Sacramento, CA
Tulsa, Oklahoma
Washington, D.C.
DATE
7/21/81
8/8/81
8/17/81
7/13/81
7/28/81
1965
1971-72
1969
1968
1967-68
1970-71
1967-69
1969
1968-69
1968-69
1969
BOD5
AVG.
51
19
20
49
21
28
14
36
31
9.4
-
12
7
106
11
19
mg/1
RANGE
51
12-28
16-23
34-59
10-29
11-62
4-37
12-100
2-232
-
-
-
-
24-283
1-39
3-90
S.S.
AVG.
91
150
65
85
71
2,080
-
505
-
1,013
81
26
30
71
247
1,697
mg/1
RANGE
91
100-210
48-90
79-98
35-110
650-11,900
-
95-1,053
-
-
10-1,000 '
-
-
3-211
84-2,052
130-11,280
COD
AVG.
100
110
53
39
40
-
-
-
224
-
, -
-
-
58
85
335
mg/1
RANGE
-
-
-
-
-
-
-
-
40-660
-
-
-
-
21-176
12-405
29-1,514

-------
Basement Flooding

     Figure 17 shows pre and post-construction survey  responses  with
respect to basement flooding.  The distribution of responses  indicates
that, at least for Glenshire Avenue and W.  170th and 172nd  Streets,
the basement flooding problem may have been improved.   This conclusion
is further supported when the expository comments on the  questionnaire,
which indicated flooding historically occurred during  any heavy  (or
even moderate) rainfall, are considered in  light of a  two-inch rainfall
which occurred during a two-hour period July 9^ 1981.   In addition.
one Puritas Avenue resident indicated to field personnel  that he had
had no basement flooding, only siSce the installation  of  the  redesigned
Hydro Brake control device in July, 1981.  This information was  noted
August 8, 1981, one day after the August 7  storm.

Street Flooding

     As shown by survey responses illustrated on Figure 18, street
flooding problems on Glenshire and Martha Avenues have evidently been
alleviated to some degree.  However, W. 172nd Street apparently  still
experiences some flooding.  Also, storm water ponds around  a  catch
basin and yard area on W. 168th Street, where .05 cfs  (1.4  L/s)  Hydro
Brakes had been installed in the catch basins.  The City  of Cleveland
has since removed the control device from the problem  area.

     On balance, it can be concluded that both street  and basement
flooding have been reduced in the study area.  Basement flooding has
been reduced because combined sewer surcharging has been  alleviated  through
the use of Hydro Brake control devices, which reduce peak storm  inflow
rates to the combined sewers.  Street flooding has been reduced  through  surface
drainage improvements and the construction of below grade stormwater storage
structures.  However, it is expected that sustained high  intensity storm
would cause some street flooding (of a temporary nature)  because the entry
points to the combined sewers are restricted to eliminate surcharging of
the combined collection system.
                                      39

-------
                                                                                                      M
                                                                               LEGEND     .  '
                                                                              NO SCALE
                                                                      PRE-CONST     POST-CONST
                                                                       A  YES        •   YES
                                                                       •  NO         9  NO
                                                                      iiliii: STORAGE TANKS
FIGURE 17   BASEMENT FLOODING SURVEY RESPONSES

-------
                                                                              LEGEND  '
                                                                              NO SCALE

                                                                    PRE-CONST.    POST-CONST.
                                                                       A YES           • YES
                                                                       •  NO            O  NO
                                                                  i$8:|:$;:$:i::$ STORAGE  TANKS
FIGURE  18   STREET FLOODING SURVEY RESPONSES

-------
                                SECTION 6

                    DESIGN AND PERFORMANCE EVALUATION
INTRODUCTION

     This section presents a summary of the design elements  employed  by  Thiel
and Candaras d) in the original  design of the~..Cleveland Hydro  Brake  control/
storage system in addition to an  evaluation of the design and  performance  of
this system.  During the course of this study, it became apparent  that the
system was not performing as intended.   Therefore, the City  of  Cleveland under-
took a redesign effort to modify  the system to better respond  to actual  con-
ditions in the study area.  The elements of this redesign are  also presented
here.  All study tasks were duplicated  and an evaluation parallel  to  that  of
the original design was then performed.   ,

ORIGINAL DESIGN CONCEPT

     The Hydro Brake application  in Cleveland is based on the  idea of
regulating the inflow of surface  runoff into the existing combined sewers
so as not to exceed the capacity  of the collector system. This is
a radically different concept compared  to the traditional drainage
design approach of building collectors  capable of handling the  peak
flow rates generated by some five or ten year design storm.

     The Hydro Brake device is the means by which flow rates are regu-
lated, with any excess flows stored in  below grade structures  or by
surface ponding.  In a residential area such as the Puritas  study  area,
the practical limit to surface storage  is between curbs on the streets.

ORIGINAL DESIGN CRITERIA

     In order to design the flow  regulation and retention structures
constructed in the study area, the surface runoff parameters and
combined sewer flow patterns were modelled using the proprietary Dorsch
Hydrologic Volume Method.0)  This model was selected because  it purportedly
accounts for sewer surcharging and backwater effects from downstream
areas that would result from critical storm events over the  drainage
area.  This allowed simultaneous  simulation of surface runoff and sewer
system response.

Surface Runoff Parameters

     Drainage areas were divided into sub-basins, all of which were
assigned runoff parameters for land use, imperviousness, surface slope
and roughness, depression storage, infiltration rate and flow path.
Separate hydrographs were generated for different types of surfaces
as well as total surface runoff hydrographs.  This was necessary where
certain portions of the drainage area, such as roof areas, would be
routed directly to the combined sewers.
                                       42

-------
Design Rainfall

     The rainfall of July 19, 1972 was selected for design  purposes
because the greatest number of flooding complaints  occurred during this
storm event and it approximated a five year return  period rainfall.
Thi-s corresponds with the City of Cleveland's storm sewer design  standard
for a five year return frequency rainfall.   This storm event consisted
of three high intensity, short duration peaks, with each occurring ap-
proximately one hour apart.  Peak intensities (based on five minute
rainfall increments) ranged from 3.1 to 3.8 in/hr (7.9 - 9.8 cm/hr),
with a total accumulation of 1.96 in. (4.98 cm).

STORM SIMULATION AMD DESIGN

Model Ing Methodology

     Using records of the July 19, 1972 storm described above,  the
Dorsch HVM model was employed to simulate the response of the existing
sewer system, after which the simulation was repeated considering only
roof area and private area drain inflow sources to  the sewer system.
Surface hydrographs were developed for the remaining area to identify
storm flows which would be regulated into the collection system.  Finally,
inflow regulation rates were determined through a series of successive
iterations such that flow levels were maintained below basement ele-
vations.  These flow regulation rates are the discharge rates of the
Hydro Brake devices to be installed, and are based  on the available
excess capacity at those points in the sewer system where inflow regu-
lation is required.  An analysis of the surface hydrographs in  relation
to these flow regulation rates produced the required storage volume at
each inflow regulation point.

Control Device and Storage Sizing

     As a result of the design storm simulation, recommendations  were
made for storage tank sizes and Hydro Brake discharge rates, as sum-
marized in Table 3.  The design allowed for some surface ponding between
curbs on street surfaces.

                 TABLE 3.  ORIGINAL HYDRO BRAKE DESIGN
             5 YEAR DESIGN STORAGE & DISCHARGE REQUIREMENTS

                     Peak     Hydro Brake      Retention     Surface
                    Inflow     Discharge      Tank Volume     Storage
Location
W. 170
U. 177
St.
St.
Puritas Ave.
cfs
4.
5.
13.
3
8
0
(L/s)
022)
(164)
(368)
cfs
2.0
1.5
7.0
(L/s)
C57)
(42)
(198)
ft*
2074
8984
6024
(1000 L)
(58.7)
(254)
(170)
ft3 (1000 L)
1676
1246
246
(47
(35
(7.
.5)
.3)
0)
                                     43

-------
     The estimated surface storage is only practical  on lightly traveled
residential streets, where it is not likely to be a traffic  hazard  and
is very temporary in nature.  Draindown of this volume occurs  in ap-
proximately fourteen minutes for the W. 170th and W.  177th Street
locations, and less than a minute on Puritas Avenue.

EVALUATION OF ORIGINAL DESIGN

Combined Sewer Overflow Reduction

     The original design approach for the study" area  resulted  in recom-
mendations for a two phase plan to install seven Hydro Brake control/
retention structures, catch basin Hydro Brakes in the areas  tributary to
the control structures and storm sewers.  Modifications to overflow
chamber M-15 were also recommended to improve flow patterns.  Only  Phase
I was implemented, involving the control structures at W.  170th Street,
W. 177th Street and Puritas Avenue, in addition to catch basin Hydro Brakes
on W. 168th Street and the overflow chamber modifications.  As part of a
separate project, storm sewers and catch basin Hydro  Brakes  were installed
in the Mil burn Avenue area north of Puritas Avenue.

     Because all the recommendations were not implemented, the actual
reduction in overflows at M-15 is less than projected in the original
design report.  Overflow volume and peak rate reductions are a result
of the improvements made in the subject study area and the Mil burn  Avenue
area.  Estimates of the reduction in combined overflows at M-15 which
can be attributed to these improvements are presented in Table 4.  These
estimates take into account the removal of the Grayton Road  lift station
discharge which previously contributed flows to the Puritas  Avenue  sewer
upstream of the study area.  These flows were included in storm flow
projections in the original design study.  Peak flow rates were estimated
to be 8.4 cfs (238 L/s).  In the Spring of 1981, this flow was rerouted
to a sewer on Rocky River Drive which is not tributary to the  Puritas
Avenue trunk and overflow M-15.

     Because a large percentage of the overflows at M-15 is  related to
backwater effects from downstream areas, it is difficult to  estimate
overflow peak rate and volume reductions without using a surface runoff
and sewer system hydraulics model.  This effort is beyond the  scope of
this evaluation, so estimated reductions are based on the pre  and post-
design simulation from the original design study for those areas where
flow controls were installed.

     As presented in Table 4, simulation of sewer system response to
the five year design storm indicates that the peak rate of overflow
from chamber M-15 is approximately equivalent to the influent flow rate
to this control point.  This is true even with full implementation of
the design study recommendations, apparently because of backwater ef-
fects from areas downstream of M-15.  Therefore, estimates of overflow
rates resulting from Phase I improvements reflect the same relationship.
                                    44

-------
          TABLE 4.  COMBINED SEWER OVERFLOW  REDUCTION  AT M-15
           BASED ON DESIGN STORM SIMULATION  -  ORIGINAL DESIGN

Overflow        No Stormwater        Phase I S II        Phase  I
Parameter	Control	Implementation     Implementation	

Peak Inflow        122  (3455)            47   (1311)         89   (2520)
(cfs - L/s)

Peak Overflow      110  (3115)            38   flD76)         77   (2181)
(cfs - L/s)

Overflow Vol.  189,400  (5364)        53,790   (1523)    121,300   (3435)
(ft3 -1000 L)

     For all  practical purposes, the same volume of storm  water enters
overflow chamber M-15 under all  three levels of control, but the  overflow
rate and volume is reduced because of peak rate reduction  at various
inlet points  to the system as well as flow retardation until capacity  is
available after the subsidence of backwater  effects.

Reduction in  Flooding and Surcharging

     Based upon the results of the surveys of study area residents,
which were presented in Section 5, it is apparent that some reduction
in basement flooding has occurred as a result of the Hydro Brake  controls
now in place.  This indicates some lesser degree of sewer  surcharging
than was prevalent before the control/retention structures were con-
structed.

     To further examine the effectiveness of these controls in  relation
to sewer surcharging, the design simulation  results for individual sewer
segments were compared to pipe capacities.  Simulation flow values were
adjusted to reflect only the Phase I improvements described previously.
This analysis indicates that peak flow rates in the sewers in  the im-
mediate area of the Hydro Brake controls are generally less than  pipe
capacity, with a few segments at or slightly above capacity.

     It appears that some surcharging will still occur under design  storm
conditions, particularly in the Phase II areas where controls  have not
been installed.  This is the case in the U.  174th Street-Ponciana-Flamingo
area.  The Puritas Avenue sewer also exhibits surcharging  at all  points
between U. 174th Street and overflow chamber M-15.  It is  not  known  if
this condition causes any significant problems in the areas noted.
                                     45

-------
Evaluation of Control/Retention Sizing

     In the early stages of this evaluation,  particularly  during  the
Fall, 1980 field monitoring, it became  apparent that only  the  most severe
storm events would cause any appreciable response  in the control/retention
structures.  Subsequent review of the original  design criteria seemed
to indicate that a large percentage of  the drainage areas  tributary to
each retention structure was used in simulating response of the system to
a five year design storm.  However a significant portion of the runoff from
residential sites actually enters the combined  sewers directly by way of
area drains, footer drains and downspout connections.  It  appeared that
because of an oversight in the original  design  only roof areas were accounted
for in routing flows directly to the sewers.

     Once this apparent error had been  identified,  the City of Cleveland
prepared new runoff computations for the drainage  areas associated with the
W. 170th, W. 177th and Puritas structures (See  Appendix F). Given the
rainfall pattern employed in the original design,  it was determined that
a large portion of the subject drainage areas would have to be contributing
runoff to be able to generate the peak  rates  and volumes which were the basis
of design of discharge rates and storage volumes.

     There are several implications of  these  findings. It would  appear
that the storage tanks which had been constructed  would rarely be fully
utilized, especially up to the five year design level. For the original
discharge rates, the storage tanks were apparently significantly  over-
designed.  Furthermore, if too small a  proportion  of total runoff were
routed to the combined sewers, the available  capacity for  storm release
rate determinations would be too large.  Therefore, the Hydro  Brake  re-
lease rates would be too high and could negate  the intent  to reduce
surcharging and basement flooding.  Finally,  if a  smaller  proportion  of
total runoff were actually controlled through the  retention structures,
combined sewer overflow reduction would be overestimated.

     Time constraints and the scope of this evaluation did not allow
computer simulation of system hydraulics to check  the validity of these
concerns, nor was it possible to observe these  effects in  the  field
beyond the noted lack of response of the control/retention structures
to significant rainfall events.  No storms resembling a five year design
storm were recorded, and it was not possible to estimate  the response
of the combined sewers receiving storm flows.

HYDRO BRAKE REDESIGN

Design Concept

     After the Fall, 1980 monitoring period,  during which   insufficient
amounts of data were collected, the City of Cleveland initiated a pro-
ject to redesign the Hydro Brake discharge rates at the three retention
structures.  The intent of the redesign effort was to define runoff
parameters for representative design storms having various durations
and  return periods, and to reduce the Hydro Brake discharge rates such

                                     46

-------
that the storage facilities would have a  more frequent,  higher  percentage
utilization in relation to these storms.   A discussion of this  effort
has been prepared by the City of Cleveland and is  included in Appendix  F.

Design Criteria
  •v
     The areas which drain directly to the control/retention structures
were defined as the area between building set back lines, which includes
streets, sidewalks, driveways, tree!awns  and front yard  areas.   These
areas are presented in Table 1, Section 5.  A.runoff coefficient of  .5
was used with a time of concentration of  25 minutes (See Appendix F).
Rainfall distributions representative of  the Cleveland area were used in
defining runoff parameters.

Hydro Brake Sizing

     Runoff volumes were calculated and inflow hydrographs were con-
structed for each location.  Hydro Brake  discharge rates were selected
which maximized storage of five and ten year design storms while main-
taining a safe level of control over surface flooding.   Table 5 presents
a comparison of original design and redesign Hydro Brake discharge
rates.

          TABLE 5.  COMPARISON OF HYDRO BRAKE DISCHARGE  RATES

                                   Original Design    Redesign
                                   Discharge Rate      Discharge Rate
Location
W. 170th Street
W. 177th Street
Puritas Ave.
cfs
2.0
1.5
7.0
(L/s)
(57)
(42)
(197)
cfs
1.25
.25
1.00
(L/s)
(35)
(7.1)
(28)
     The new control devices were to be delivered and installed prior
to the start of the Summer, 1981 monitoring period.  Various delays in
delivery of the new Hydro Brakes and their installation postponed operations
under redesign conditions until July, 1981.  Therefore, data collection
during 1981 represents "before" and "after" redesign.

EVALUATION OF HYDRO BRAKE REDESIGN

Combined Sewer Overflow Reduction

     It was assumed that the combined reduction of 8.0 cfs (226 L/s)
in Hydro Brake discharge rates would result in a reduction of peak
inflow rates to 81 cfs (2294 L/s) at overflow chamber M-15.  From com-
parisons with values in Table 4, the estimated peak overflow rate is
69 cfs (1954 L/s) with an overflow volume of 106,400 ft.3 (3013 x 103 L).
This represents an estimated 37% reduction in the peak rate of combined overflow
over the previous conditions of no stormwater control.  Overflow volume reduction
is estimated at 44%.
                                     47

-------
Reduction in Flooding and Surcharging

     By reducing Hydro Brake discharge rates,  it  was  expected  that  flow
rates in the receiving combined sewers should  be  reduced  by  a  similar
amount.  It was not possible to determine from survey results  whether
th-is difference would be noticed by homeowners when  compared to  pre-
vailing conditions with the original  Hydro Brakes.

     As noted previously for the original design  Hydro Brakes, no sur-
charging in the vicinity of the control  devices is apparent  when the
sewer system simulation results are examined.  .As indicated  above,  this
condition should be further improved with lower rate devices in  place.
Those sewers on Puritas Avenue and H.  174th Street,  which would  ap-
parently be surcharged under design conditions with  the original Hydro
Brakes in place, would still be surcharged but to a  lesser degree.

     With the redesigned Hydro Brakes  in place, it is also expected that
the potential for surface ponding during severe storm events is  somewhat
increased.  An evaluation of design storm conditions presented below
indicates that any surface flooding which might occur will be  relatively
minor.

Evaluation, of Redesigned Hydro Brake Devices Using Storm  Simulation Techniques

     In an effort to evaluate the performance of the redesigned  Hydro
Brake control/retention structures in relation to design  storm conditions,
storm hydrographs and runoff volumes were calculated using the design
criteria proposed by the City of Cleveland and the hydrograph  construction
methods presented in Section 4.  As noted previously, one-half hour
duration storms were used because they would generate the highest  peak
flow rates over the drainage areas of interest.  This was done to  il-
lustrate the highest probable level of control (i.e. peak rate attenuation)
that could be provided by the Hydro Brake control devices.  However,
larger storage volume requirements are possible from longer  duration
storms having the same return frequency as the 1,2,5 and 10  year
storms examined.

     After construction of inflow hydrographs, discharge  hydrographs
were computed using the storage continuity relationship presented  in
Section 4.  These calculations were performed for the original Hydro
Brake discharge rates for comparison with design values,  and for the
redesigned discharge rates in order to project the expected  performance
of these units.  The head-discharge curves used were the  calculated
relationships from the field calibration tests of each Hydro Brake
device.  Hydrographs are presented in Appendix E.
                                      48

-------
     Table 6 summarizes rainfall  parameters  for the  one-half  hour
duration storms examined.

             TABLE 6.   DESIGN HYDROGRAPH PARAMETERS  -  1/2  HOUR
                      STORM RAINFALL AMD INTENSITY
Storm
Return
Frequency
1
2
5
10
Total
Rainfall
in. ( cm . )
.76
.91
1.19
1.37
(1.93)";
(2.31)
(3.02)
(3.48)
Peak
Intensity
in/hr. (cm/hr)
4.0
4.7
6.2
7.1
(10.2)
(11.9)
(15.7)
(18.0)
     Table 7 presents comparisons of Hydro Brake discharge rates  and
required storage volumes for a five year design storm as  computed by
simulation during the original design and by methods  employed in  this
evaluation.  Note that the drainage areas and design  storms used  in the
two methods are not equivalent.

           TABLE 7.  COMPARISON OF HYDRO BRAKE DISCHARGE  RATES
              AND STORAGE VOLUMES FOR 5 YEAR DESIGN STORMS

                             Original Design
Location
W. 170th Street
W. 177th Street
Puritas Avenue
Location
W. 170th Street
W. 177th Street
Puritas Avenue
Peak
Inflow
Rate
cfs (L/s)
4.3 (122)
5.8 (42)
13.0 (368)
Post- Construct! on
Peak
Inflow
Rate
cfs (L/s)
3.4 (96)
4.4 (125)
5.0 (142)
Peak
Discharge
Rate
cfs (L/s)
2.0 (57)
1.5 (42)
7.0 (198)
Evaluation
Peak
Discharge
Rate
cfs (L/s)
3.1 (88)
1.7 (48)
7.0 (198)
Storage
Required
ftj (1000 L)
3,750 (106)
10,230 (289)
6,270 (178)
Storage
Required
ft3 (1000 L)
380 (11)
3,470 (98)
None
                                     49

-------
     Table 8 presents a summary of the  expected  performance of the
control/retention structures for various  one-half  hour design storms
with the redesigned Hydro Brake devices  in  place.   It is projected that
storage capacity would be exceeded at the W.  170th Street structure for
the 5 and 10 year return storms.  For the 5 year storm, there would
be  340 ft3 (9630 L) and for the 10 year storm,  a  930 ft3 (26,340'L)
of stormwater in excess of storage capacity.   For  surface ponding at an
average depth of 4 inches (10 cm) on a  24 foot (7.3 M) wide road surface,
flooding would occur over a distance of 42  feet  and 116 feet (12.8 m and
35.4 M), respectively, and drain down at the  indicated peak discharge rates
would occur in 4.5 and 11.5 minutes. This  condition appears to be of
minor consequence considering that it would rarely occur and is of very
short duration.
                                     50

-------
TABLE 8.  SUMMARY OF HYDRO BRAKE CONTROL/RETENTION FOR
            1/2 HOUR DURATION DESIGN STORMS
Design
Control Storm
Structure Return
Location Period
W. 170th St. 1
2
5
10
W. 177th St. 1 '
2
5
10
Puritas Ave. 1
2
5
10
Peak Inflow
(Runoff) Rate
To Retention
cfs (L/s)
2.15
2.58
3.40
3.90
2.80
3.35
4.38
5.10
3.17
3.78
4.96
5.70
(61)
(73)
(96)
(no)
(79)
(95)
(124)
(144)
(90)
(107)
(140)
(161)
Peak
Discharge
Rate
cfs (L/s)
1.12
1.19
1.30
1.35
.43
.45
.48
.50
1.78
1.89
2.05
2.12
(32)
(34)
(37)
(38)
(12)
(13)
(14)
(14)
(50)
(54)
(58)
(60)
Retention Storage
Inflow Volume
.Volume Required
ff3 (1000 L) ft3 (1000 L)
3,340 (95)
3,940 (112)
5,220 (148)
6,000 (170)
4,280 (121)
5,130 (145)
6,670 (189)
7,730 (219)
4,810 (136)
5,800 (164)
7,700 (218)
8,710 (247)
1,280 (36)
1,690 (48)
2,640 (75)
3,230 (91)
3,350 (95)
4,120 (117)
5,550 (157)
6,540 (185)
1,570 (44)
2,220 (63)
3,570 (101)
4,430 (125)
Percent
Storage
Utilization
55
75
115
140
35
40
60
70
25
35
60
70

-------
                               SECTION 7

         COMPARATIVE EVALUATION OF FLOW REGULATOR INSTALLATIONS


     Flow regulator devices have been used and evaluated  in  several  communities
in the United States and Canada.  Evaluations  have included  emperical observa-
tions vs computer simulated performance of Hydro Brake devices,  and  evaluations
of such devices compared to straight orifice reduction, as well  as evaluation
against uncontrolled discharge.               - .

     Applications varied, as did the sizes of drainage areas and the types  of
unit configuration.

PROJECT SUMMARIES

Rochester, New York

     In Rochester, an off-line storage tank for stormwater  retention was  con-
structed for comparative evaluation purposes.   A Hydro Brake Standard 5-B-7
unit and a three-inch (75 mm) orifice were tested under controlled head con-
ditions.  A variable head test and a static head test were  run for each.
Unfortunately, size and rating data were not available for  the Hydro Brake,
making comparison of results something of a matter of faith.

     Head-discharge curves demonstrated that the Hydro Brake device  success-
fully retarded the discharge rate as heads increased.  The  observed  orifice
discharge rate was nearly in linear relationship to the head, as would  be
expected; for.a simple orif1ce> flow rate is proportional'to the square root
of the head.  The Hydro Brake discharge rate tended strongly to be asymp-
totic in relationship to the head.  Figure 19, redrawn from O'Brien  and Gere,
demonstrates th edischarge curves.  Figure 20 illustrates the relationship
between discharge rates derived during testing and discharge rates as
described by the manufacturer.  The difference is attributable to experi-
mental error, when compared to the difference between the Hydro Brake and the
three-inch orifice, according to the study report.

     Additionally, the performance of the Hydro Brake during a storm, which
occurred fortuitously during the testing period, was evaluated.  Inflow and
outflow were monitored and plotted, (Figure 21), demonstrating that  the Hydro
Brake effectively retarded peak flows and maintained relative discharge
constancy throughout the discharge period.
                                     52

-------
en
oo
             0.6 -
                 -(15)
             0.5-
             0.4-
                -(10)
             0.3-
         o

            0.2-
             0.1-
              0.
                -(5)
                          (.30)
                                            3" (7.6cm)  ORIFICE


                                            HYDROBRAKE

                                            (STD. 5-B-7)
               2          3'          4'         5'

              (.61)         (31)        (1.22)       (1.52)

                 HEAD IN FEET (METERS)
 6'

(1.83)
      FIGURE 19
HEAD - DISCHARGE CURVES    SAIMTEE  DRAINAGE  AREA  CALIBRATION

-------
       0.5-
       0.4-
       0.3-
      0.2-
       0.1-
           -(10)
           -(5)
	AS DETERMINED IN TEST ON
         MAY 13,1981'
                                                                            AS PER HYDROSTORM
DIFFERENCE  ATTRIBUTED TO EXPERIMENTAL
ERROR
                      I           I           I            !           I
                      12345
                     (.30)        (.61)        (.91)       (1.22)       (1.52)

                               HEAD IN FEET  (METERS)
FIGURE  20     HEAD-DISCHARGE  CURVE, S ANTE E HYDROBRAKE

-------
trt
en
             ,00 -J
             .80-*
             .60-
        Ig   -40H
             .20-
                 -(25)
                 -(20)
                 -(20)
                 -(10)
                 -(5)
                           20
      VOLUME = 2,000 CU. FT.
            (56,410 L)
                                                                         UNCONTROLLED  DISCHARGE  (INFLOW)
                                —HYDROBRAKE CONTROLLED DISCHARGE
                                   (OUTFLOW)
                                                                        VOLUME =1,700 CU.FT.
                                                                              (47,949 L)
 1
40
60         80
  TIME (WIN.)
100
120

140
       FIGURE 21      STORM OF MAY II, 1981 - SANTEE  DRAINAGE  AREA

-------
     Toward the end of this study,  it was  noted  that  the Standard 5-B-7
Hydro Brake was showing discharge curves that  were  not  appreciably different
from an orifice of similar diameter.   Debris in  the entry slot was discovered
to be the cause.  Although the piece  of lath had not  blocked the flow, as had
the styrofoam cup in the W. 177th Street (Cleveland)  installation", it had
affected the flow regulating characteristics of  the unit.

     The Rochester, New York, application  of the Hydro  Brake parallels the
Cleveland project.  The Rochester demonstration  project report is presented
in Appendix 6 to provide supplemental  information pertinent to the application
of this technology.

Nepean Township, Ottawa, Canada ^ '

     This evaluation consisted of the installation  of two oversize catch
basins, in which Hydro Brakes were installed.  One  catch basin drained 1.28
acres (0.519 ha), of which .39 acres  (0.158 ha)  was 86.67 ft3  (2,450 L)
and was outfitted with a Hydro Brake  of 0.6 in.  (15 mm) diameter orifice.

     The other catch basin drained 0.61 acres  (248  ha), of which 0.34 acres
(0.139 ha) was impervious surface.  This basin contained a storage capacity of
75.6 ft3 (2,135 L) and was outfitted  with  a Hydro Brake of 2.0 in  (50 mm)
diameter orifice.  Water levels in the catch basins were monitored by means of
continuous recording water level  sensing units.   Outflow was calculated from
the water level and rainfall data by  first calculating  inflow, and then using
discharge curves supplied by the  manufacturer, deriving discharge volumes.

     The Hydro Brakes demonstrably attenuated  the peak  flows  (Figures 22,
23, 24, 25, and 26, redrawn from  Gore and  Storrie), but the 0.6  in  (15 mm)
diameter device proved to be too  small for the storage  volumes provided.
It was replaced with a 4.0 in (100 mm) device, but  storm events  during the
study period did not produce great enough  peak flows  to exceed this  unit's
pass-through capacity.  Table 9 shows the  peak flow reductions obtained by  the
2.0 in (50 mm) device for several storms.   As  was the case  in  Cleveland,
accurate sizing of Hydro Brake devices during  design  proved to be  a  problem.

     No problems with debris in regards to Hydro Brake  operation were reported
in this study.  However, it was noted that solids buildup  in the catch basins
did occur, and that odor was a problem. Attempts to  trap  solids in  mesh  bags
were unsuccessful.

       TABLE 9  ANALYSIS OF PEAK FLOW ATTENTUATION  FOR  2  IN.  UNIT
Storm Event
May 23,
May 28,
June 25
June 29
July 1,
1977
1977
, 1977
, 1977
1977
Peak Rate
of Runoff
L/s cfs
5.60
2.94
2.95
.12
6.20
. .198
.104
.104
.145
219
•^'3 K<
Attenuated
Peak
L/s cfs
1.20
1.05
0.85
1.10
1.17
.042
.037
.030
.039
.041
% Reduction
of Peak
78.5
64.3
71.2
73.3
81.1

-------
tn
            .125-
            .100-
           .075-
        o
           .050-
           .025-
                                                             0.6" (15mm.)  HYDROBRAKE
                                                                    	INFLOW
                                                                    	OUTFLOW
                                                                                    0.009 CFS
                                                                                    (0.25 L/S)
                                           0.005 CFS
                                          '(0.14 L/S)
             I0:00pm     12:00     2'OOam    4:00
6:00       8=00
TIME OF DAY
10:00     12=00
 T    i     r
2:00pm    4:00
       FIGURE 22       INFLOW-OUTFLOW HYDROGRAPHS     NEPEAN  TWP   AREA I  -  OCTOBER  30/31,  1976

-------
01
00
          .250-
           .200-
           .150
      "•&  .100-
          .050-
               L(6)
               -(5)
               -(4)
-(3)
               -(2)
              3=00pm       3^30
                     4=00
                                                   2" (50mm)  HYDROBRAKE

                                                    	INFLOW

                                                    	OUTFLOW
                                         0.198 CFS'
                                         (5.6(L/S)
I                                                                             0.042 CFS
                                                                             [I.2L/S)
4=30
        I

       5'00

TIME OF DAY
                                                       5=30
6--00
 I

6:30
      FIGURE  23    INFLOW-OUTFLOW  HYDROGRAPHS   NEPEAN TWP  AREA 2  -  MAY 23, 1977

-------
en
            .125-
                 -(3)
            .100-
            .075-
        V)
            .050-
            .025-
                 -(2)
                -(I)
               IQiOOatti      11=00
         -0.104 CFS
         (2.95 L/S)
                                                -0.030 CFS
                                                (0.85 L/S
                                                               2" (50mm.)   HYDROBRAKE
                                                                 	INFLOW

                                                                 	OUTFLOW
12:00
 I

I'OO

 TIME OF DAY
 I

2=00
 I

3=00
4=00
5=00
      FIGURE 24        INFLOW-OUTFLOW  HYDROGRAPHS    NEPEAN  TWP.  AREA 2   JUNE  25,  1977

-------
CTI
O
               .150-
                   -(4)
               .125-
                   -(3)
              .100-
              .075-
          

         £
              .050-
              .025-
-(2)
                   -(I)
                                  2" (50 mm.)  HYDROBRAKE
                                        • INFLOW
                                  --- OUTFLOW
                                                                 -0.145 CFS
                                                                  (4.12 L/S)
                                                              r 0.039 CFS.
                                                             / (I.I L/S)


                                                              \
                 7=00   8=00    9=00    10=00   11=00    12=00   hOOpm   2=00   3=00    4=00    5=00    6=00  7=00

                                                 TIME OF DAY
       FIGURE 25     INFLOW-OUTFLOW   HYDROGAPHS    NEPEAN  TWP  AREA  2  JUNE  29, 1977

-------
    .250-
    .200-
    .150-
  CO
I;
     100 -J
    .050-
         -(7)
         -(6)
         -(4)
         -(3)
         -(2)
         -(I)
       3:00om
 I
4=00
                                        -0.219 CFS (62 L/S)
                                       2" (50 mm.)  HYDRO BRAKE
                                           	INFLOW
                                           	OUTFLOW
                                             -0.041 CFS (I.I7L/S)
 I
5:00
    I
   6:00

TIME OF DAY
 I
7=00
 I
8:00
9=00         IO--00
FIGURE 26      INFLOW—OUTFLOW  HYDROGRAPHS     NEPEAN TWR   AREA 2  —  JULY I, 1977

-------
Borough of York, Ontario, Canada (6)

     This study does not include evaluations  of Hydro  Brakes  In situ.
Rather, it uses as a given the ability of the device to  regulate flow to
a desired maximum discharge.   Based upon that ability, four areas of the
Borough of York are evaluated.

     The results of those evaluations  were translated  into required storage
capacities and regulated discharge rates for  each  area,  presented for 2,
5 and 10 year frequency storms.  Discharge rates are calculated from runoff
coefficients and existing sewer capacities to-p.rovide  zero surcharge flow
in existing sewers.

     Hydro Brake installation in the four areas is expected to permit the existing
sewers to carry the design flow without system modification other than  con-
struction of off-line storage, or implementation of roof storage in one
industrial area.

Euclid, Ohio

     Although no formal evaluation of the effectiveness  of the City's Hydro
Brake installations has been performed, telephone  interviews  with Mayor Anthony
Giunta and Mr. John Piscitello, Service Director,  elicited some informa-
tion.  Hydro Brakes were installed directly in catch  basins in areas which
experience basement flooding, in effect utilizing  street retention of
storm waters upstream of flooded areas.  Both the  Mayor  and the Service
Director feel that the devices have worked well, and  that flooding problems
have been greatly reduced, albeit not entirely eliminated.


COMPARISON OF INSTALLATIONS

Introduction

     Comparisons among the Hydro Brake installations  described above and
between those and the Cleveland installation are difficult.   Hard  data  is,
for the most part, extremely limited in the few reports  published  to date.
Although results are ascribed to various factors,  it is  not  possible to
compare their derivations, since such basic data as Hydro Brake orifice
diameter, calibration of storage structures,  discharge calibration,  etc.
are each missing from one or another of the reports.   Much of this  lack of
data is probably ascribable to the fact that the reports available  are
intended to apprise a client of results, rather than to  qualify  the
report for journal publication.

     In addition, each installation is different from the others,  and  only
the Santee Drainage Area in Rochester, New York, installation is  similar
to the Cleveland project.
                                    62

-------
     Finally, although Hydro Brakes  have  apparently  been  installed in at
least a dozen locations,  follow-up studies  are,  for  the most part, unavail-
able.  Thus, quantitative evaluations  and comparisons are limited to only
those described above.

DT-scussion of Calibration

     The Rochester, New York installation is  most  similar to the Cleveland
project.  An underground, off-line storage  structure with a Hydro Brake
regulator was installed to accept the  flow  from  approximately  50,000-55,000
ft2 (.465-.512 ha) as compared to the  Cleveland  installations.  Those
installations drained approximately  three acres  (1.215 ha) each.  The
Cleveland falling head calibrations  of Hydro  Brake discharge were, in fact,
modeled after the Rochester study.  Unfortunately, physical problems with
the sewer plugging efforts limited the usefulness  of some of the results
(see Section 4).  However, it was noted that  the discharge curves that
were developed did not indicate so dramatic a flow attenuation as did
the discharge curve developed in Rochester.

     Figure 27 displays the Cleveland  calibration  curves  against the
Rochester calibration curve for the  Hydro Brake  Standard  5-B-7.  Although
various sizes of Hydro Brakes are included, a consistent  pattern of attenua-
tion (discharge independent of head) should be evident for all.  As can be
seen, this seems to be the case.  However,  as shown  in Table 10, (which
includes data from the Nepean Twp. Ottawa Project) local  conditions may
affect Hydro Brake performance.

     As can be seen, observed flow rates  in the  Nepean Twp. and Rochester
studies more closely approximated the  expected rates than did  the observed
Cleveland rates.  The 1.0 cfs (28.3L/s) and 2.0  cfs  (26.6 L/s) units  in
the Cleveland study units were especially high in relationship to the
expected flows.  The difference shown  for the 2.0 cfs  (56.6 L/s) unit  is
not explainable by available data, although the  leakage  described in
Section 4 may have contributed significantly. In any  case, it is' pparent
that some attenuation of flow occurs for all  sizes of  Hydro Brake.

     The Nepean Twp. installations were somewhat different from the  Cleveland
and Rochester installations, in that the storage volume  provided was much
less per unit area of drainage.  However, if  street  storage is assumed
equivalent to off-line storage, the  ultimate  function  of the  units  is
quite similar.  Calibration data is  included  in  Table  10, but  the flows
were too low to be effectively displayed in Figure 27.

Discussion of Rainfall Event Monitoring

     Although differences in monitoring and calculation  methodologies  among
the studies make quantitative comparisons unuseable, it  is apparent  that
in Rochester, Nepean Twp. and Cleveland, the  Hydro Brakes did  reduce peak
flows.  Thiseeffect in turn increased  storage; thereby delaying entry of
stormwater into the sewer system.  Therefore, it can be  said  that  in all
three cases, additional sewer capacity was  made  available during the moni-
tored storm events.  Comparative flow rate  analyses  of ordinary orifices,

                                    63

-------
     7_ -(200)
     D ^™
        -(150)
     4 ——
s

oo
to
     2—
        -(100)
                                            I6in.(4lcin.) HYDROBRAKE
                                               (PURITAS AVE.)
                                                           9in.(23ctn.) HYDROBRAKE
                                                               W. 170 ST.
                                                 5.5in.(l4cm.) HYDROBRAKE
                                                           7in.( 18cm.) HYDROBRAKE
                                                                W. 177 ST
                                                              6in.(l5cm) HYDROBRAKE
                                                                    W. 170 ST
                                                                   3"(76cm.) HYDROBRAKE
                                                                     .  W.I77ST
                                                                     5-B-7 HYDROBRAKE
                      I (.30)
                                     I
                                    2 (.61)
                                     HEAD  IN FEET  (METERS)
 \
4(1.22)
50.52)
         FIGURE 27   COMPARISON OF HYDROBRAKE DISCHARGE  CURVES

                                        64

-------
en
                                              TABLE 10

                                   COMPARISON OF RATED AND ACTUAL
                                       HYDRO BRAKE DISCHARGES
Hydro Brake
Size
. in. (cm)
.6
2.0
3.0
3.0*
3.0
5.5
6.0
7.0
9.0
16.0
(1.52)
(5.08)
(7.62)
(7.62)
(7.62)
(13.97)
(15.24)
(17.78)
(22.86)
(40.64)
Manufacturer's
Rating (cfs)
cfs L/s
.005
.05
.24

.25
1.00
1.25
1.50
2.00
7.00
(.142)
(1.42)
(6.80)

(7.08)
(28.32)
(35.40)
(42.48)
(56.64)
(198.23)
Observed
Flow (cfs) at
3 ft. .Head
cfs L/s
.006
.04
.21
.44
.40
2..95
1.30
1.85
4.40
7.70
(.17)
(1.13)
(5.95)
(12.46)
(11.53)
(83.54)
(36.81)
(52.40)
(124.60)
(218.05)
Observed
Flow (cfs) at
5 ft. Head
cfs . L/s
--
—
.27
.58
—
3.95
1.45
2.50
5.50
9.10
—
—
(7.65)
(16.42)
—
(111.86)
(41.06)
(70,. 80)
(155.75)
(257.70)
Location
Nepean Twp.
Nepean Twp.
Rochester
Rochester
Cleve./W. 177
Cleve./Puritas
Cleve./W. 170
Cleve./W. 177
Cleve./W. 170
Cleve./Puritas
             *Straight Orifice

-------
similar in size to the Hydro Brake devices,  strongly suggest  that  the Hydro
Brakes were able to retard flow,  using a larger outlet  than would  have been
possible with a standard orifice.

     In addition  ,the Cleveland study indicates that by making  additional
sewer capacity available, basement flooding  and some street flooding was
re'duced.  The City of Euclid reports apparently similar results.   By the
use of Hydro Brakes in catch basins, Euclid  utilizes street storage,
thus reducing catch basin surcharging, and consequently basement flooding
caused by back-up flows from those catch basins.

     In summary, it would seem that the method of storage is  relatively
unimportant.  As suggested in the study prepared for the Borough of York,
street, roof and parking lot, or buried off-line tank storage should be
determined from the uses and needs of the project area; with  the Hydro
Brakes designed to best serve the chosen storage method.

     The Hydro Brakes themselves seem to have two outstanding advantages.
By retarding flow while permitting a larger  outlet, they seem less likely
to become fouled by refuse than would the smaller orifice. However, as
demonstrated by the redesigned 177th Street  Hydro Brake in Cleveland,
small orifice Hydro Brakes may themselves become fouled.  This  does not
negate the likelihood that the even smaller  orifice required  for the same
level of control would incur a proportionately greater  risk of  fouling.
In addition, their flexibility of design allows for a variety of appli-
cations, which increase their attractiveness as control devices for older,
overloaded sewer systems that may otherwise  require the alternatives of
replacement or relief sewer construction.

     Debris in two Hydro Brakes affected performance.  In the W.  177th
Street installation (Cleveland), a styrofoam cup wholly blocked the flow
of stormwater.  In the Standard 5-B-7 installation (Rochester)  a  piece of
lath apparently interrupted the vortex within the unit, and thereby
permitted stormwater flow to occur at a rate equal to a 3.5 in  (8.9 cm)
orifice.

     Generally speaking, maintenance was not described as a problem,
although the Nepean Township catch basin installations  were subject to
solids depositions.  Regular clean-out was necessary because  of odors.
                                     66

-------
                                SECTION 8

                         ALTERNATIVE EVALUATION
GENERAL SCREENING OF ALTERNATIVES

     Some alternatives for alleviating detrimental  combined  sewer over-
flow effects are not practical  for handling upstream sewer surcharge
problems.  These control  methods were identified and eliminated  from
further consideration.

Combined Sewer Relief Regulators

     This method of alleviating combined sewer overloads  is  not  viable
in areas similar to the Puritas Avenue study area.   There is no  where to
relieve the combined sewers without extensive interceptor construction,
and this option compounds the health and water quality degradation  problems
of conventional combined overflows.

In-Line Storage

     Where there is available trunk and interceptor capacity, this  is
a simple, cost-effective means  of controlling storm flows through in-line
regulators.  There is no excess capacity in the sewers of the subject
study area.

End of Pipe Methods

     Interception and transportation of combined flows at the point of
overflow does not relieve surcharging upstream of the control point,
and requires large interceptor construction to carry flows  to some
point for treatment.  Likewise, end of pipe treatment does  nothing  for
problems upstream and usually has lower pollutant removal efficiencies
compared with centralized treatment facilities.  Controls in upstream
problem areas reduce the pressure on downstream control points,  in  large
part eliminating the need for peak flow rate transportation  or treatment.

Off-Line Combined Storage with Gravity Discharge

     Grades in the study area are not sufficient to allow a  workable
gravity influent/gravity effluent combined flow storage approach.   To
take advantage of gravity return from off-line storage here  would  require
high capacity lift stations capable of handling peak runoff rates.
The reliability of such a system would also be jeopardized  by power outages,
so that back-up power would also be requird.
                                    67

-------
DESCRIPTION OF VIABLE ALTERNATIVES

Inlet Control

     The simplest control  requires only regulation  of peak  flows  at  com-
bined sewer inlet points.   Hydro Brake control  devices inserted  in catch
basins has been successfully implemented in several  areas,  as  noted
in Section 7.   This option is contingent upon the availability and
safety of surface ponding  to accommodate storm water storage require-
ments.

Inlet Control  and Storage

     This alternative is the central  issue of this  evaluation.  A
combination of inlet controls and below grade storage was  used for storm
flow regulation.  Relatively shallow storage facilities are required to
allow gravity discharge to the combined sewers.  This approach offers
considerable design and flow regulation flexibility, depending on the
level of control desired.

Off-Line Combined Storage  with Pumped Discharge

     Storage facilities must be deeper than the combined sewers,  and
require pumping to return  flows to the collection system.   Storage
volume and flow regulation rates are flexible, with the pump discharge
being the flow rate control mechanism.  Compared to surface runoff storage,
combined flow storage may have higher maintenance requirements because
of solids deposition and decomposition problems.

Sewer Separation

     This option eliminates combined sewers  but only with very
extensive construction and at great expense.  Sewer construction in
fully developed urban areas is often beyond the means of both  residents
and public agencies.

COST ESTIMATES OF VIABLE ALTERNATIVES

     Project cost summaries have been compiled from the City of Cleveland's
project cost summarie  for the Puritas Avenue and Mil burn Avenue projects,
the Santee Drainage Area/Hydro Brake Demonstration  Project report (O'Brien
& Gere) (?) and EPA construction cost summaries.  All costs were adjusted
to an ENR index of 3700 (January, 1982) and are reported as cost per acre
for rough comparability, although project scales are different.

     For estimation of off-line combined flow storage costs,  it was  assumed
that one retention dewatering lift station rated at 0.5 - 1.0 cfs (14 -
28 L/s) would be required for a 10 acre (4 ha) drainage area.  The only
difference between the Hydro Brake storage tanksand the combined flow
storage is depth; costs were based on Hydro Brake storage costs plus 25%.

     Table 11 presents a comparison of the costs for the various alter-
natives discussed.  The obvious benefit of the Hydro Brake concept  is
its cost-effectiveness when compared with other alternatives.

                                     68

-------
     TABLE 11.  COST COMPARISONS OF STORM FLOW CONTROL ALTERNATIVES
Alternative
Inlet Control
(Cleveland)
Inlet Control &
Project
Cost
$ 4,300
$345,000
Drainage
Area Cost Per
Acres (hectares) Acre ($/hectare)
31 (12.5)
25 (10. 1)
$ 140
$13,800
($350)
($34,200)
Storage (Cleveland)

Inlet Control   m      8,400
(O'Brien & Gerer '

Inlet Control         $22,100
Storage (O'Brien & Gere)(7)

Combined Storage
   (a)  Retention
   (b)  Pumping
   (c)  TOTAL

Sewer Separation
  35  (14.2)    $   240  ($590)
1.25  (.5)
      (.4)
      (.4)
      (.4)
  1    (.4)
$17,700  ($44,200)
$13,500  ($33,300)
  8,500  ($21,000)
$22,000  ($54,300)

$3(1.000- (  74.100-
 45,000   11KOOO)
                                      69

-------
                                 REFERENCES


1... Theil, Paul  E.,  and A.M.  Candaras,  Paul  Theil Associates Limited, Design
    of the Hydro Brake Stormwater Detention  Tank Assemblies for the Control
    of Combined  Sewer Overflow Pollution,  EPA  Demonstration Grant No. S005370,
    April, 1981.  (unpublished in-file  report).

2.  Chien, Jong-Song, and K.K. Saigal,  Urban Runoff  by  Linearized Subhydro-
    graphic Method.   Journal  of the  HYD.  Div., ASCE,  HY 8, Aug., 1974.

3.  Terstriep, Michael L., and John  B.  Stall,  The  Illinois Urban Drainage
    Area Simulator,  ILLUDAS.

4.  O'Brien & Gere Engineers, Inc.,  Hydro  Brake  Demonstration  Project -
    Santee Drainage  Area Progress Report,  June,  1981.

5.  Gore & Storrie Limited, Evaluation  of the  "Hydro  Brake" Flow Regulator,
    Report SCAT-7 for Canada  Mortgage and  Housing  Corporation.

6.  Paul Theil Associates Limited, Borough of  York Sewer Surcharging and
    Flooding Relief  Study by  Implementing  Hydro  Brake System,  June, 1977.

7.  O'Brien & Gere Engineers, Inc.,  Hydro  Brake  Demonstration  Project -
    Santee Drainage  Area - Rochester, New York,  February, 1981.
                                       70

-------
                           BIBLIOGRAPHY
Contractors Data Report - Nationwide Tabulated  Bid  Results  -  Ohio Listings.

Dames & Moore Water Pollution Control  Engineering Services„ Construction
     Costs for Municipal  Wastewater Conveyance  Systems:   1973-1977, EPA
     Technical Report 430/9-77-014, May,  1978.

Field, Richard, Anthony N. Tafuri  and Hugh  E\ Masters, Urban  Runoff Pol-
     lution Control Technology Overview,  Storm  and-Combined Sewer Section -
     Municipal Environmental  Research Laboratory, EPA -  600/2-77-047, March,
     1977.

Freeman, Peter A., Peter A. Freeman Assoc., Inc., Evaluation  of Fluidic
     Combined Sewer Regulators Under Municipal  Service Conditions, EPA Demon-
     stration Grant No. 11022 FWR, EPA -  600/2-77-071, August,  1977.

Mallory, C.W., The Beneficial Use  of Stormwater,  EOA - R2-73-139, January,
     1973.

Sarikelle, Somsek, Overview of Hydrograph and Routing Techniques and  Intro-
     duction to Modeling, in Proceedings  of Urban Runoff Seminar, University
     of Akron, December 10, 1976.

Sullivan, Richard H., et. al., American Public  Works Association, The
     Helical Bend Combined Sewer Overflow Regulator, EPA -  600/2-75-062,
     December, 1975.
                                   71

-------
1.
                          APPENDIX A

               FORMULAE, HEAD VS.  VOLUME TABLES

Ungula of a cylinder - calculation of volume  increments  for  head
vs. volume relationships.
Figure A-l.  Ungula of a cylinder

Volume = ^  [a(3R2-a2) + 3R2 (b-R) VI ]

Where b = depth of water measured at downstream end of storage tank
      r = radius of storage tank
     2a = width of water surface
      h = length of volume section (varies  with b  and  slope  of tank)
     2w = central  angle of chord defined by b  and  r
                                  72

-------
2.  Construction of Sub-Hydrographs (design storm evaluation)
       1
                           TIME
Figure A-2.  Design Storm Sub-Hydrograph


Q = CIA (2tr)/(tr + tc), tr«tc

Where Q = Peak flow rate in cfs
      C = Runoff coefficient
      I = Rainfall increment intensity in in/hr
      A = Drainage area in acres
     tr = Duration of rainfall increment
     tc = Time of concentration of drainage area
                                   73

-------
 3.  Hydrograph Routing - Design storm discharge simulation

      For convenience in hydrograph routing, it is assumed that the
"average of flows at the start and end of a small time increment is
 equivalent to the average flow during this increment.

      Therefore, the storage continuity equation:
      May be rewritten as:

                      II +  12 - DI + 02 = S2 - Sj

                         2         2        dt

      Where    0} = discharge at time tj
              02 = discharge at time t%
              \l = inflow at time t^
              I 2 = inflow at time t2
              S2 = storage  at time t2
              S1 = storage  at time t}
              dt = t2 - ti

      Rearranging yields:

                                     Q!) = 2S2/dt + 02
      All  terms on the left  side of the equation are  known, so the value
of the  right  side may be  calculated.  From  known relationships for
H  vs. S and H vs. Q, a graph of 2S/dt + 0 vs. 0 may  be constructed.
To solve  the  routing problem, successive inflow values are input to
determine corresponding discharge values (O^).  These discharge values
are then  used to repeat the calculation until all  influent flows have
been input through the relationship  to construct the corresponding
discharge hydrograph.
                                   74

-------
4.  Head vs. Volume Relationships

     Table A-l presents cumulative volumes in cubic feet for one
inch incremental changes in water depth in each Hydro Brake retention
structure.  Depth is measured at the downstream end of each structure
and includes volumes for catch basin leads; manholes and storm sewer
segments where applicable.
                             TABLE A-l
Depth*            Cumulative Volume in Cubic Feet
(Inches)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
22
24

W. 170 St.
0.5
2
7
13
23
37
54
75
100
129
163
200
242
287
337
391
451
516
585
610
675
743
811
880

W. 177 St.
1
6
15
30
51
80
116
161
214
275
346
426
516
615
723
842
970
1109
1257
1414
1578
1746
1914
2084
75
Puritas Ave.
1
5
15
30
52
83
121
169
226
293
368
451
540
633
730
830
932
1037
1143
1250
1359
1469
1579
1690


-------
Depth
(Inches)
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Cumulative
W. 170 St.
951
1021
1091
1161
1230
1298
1365
1429
1492
1553
1611
1667
1721
1773
1824
1874
1923
1971
2015
2067
2108
2146
2181
2214
2240
2261
2279
2293
2304
2311
2316
2318
Volume in Cubic
W. 177 St.
2260
2439
2622
2807
2993
3180
3369
3557
3746
3923
4123
4311 .
4498
4684
4868
5053
5236
5418
5598
5777
5955
6131
6305
6477
6647
6815
6980
7143
7303
7460
7614
7765
Feet
Pun'tas Ave.
1801
1914
2026
2139
2252
2365
2478
2590
2702
2814
2926
3037
3095
3257
3366
3474
3581
3688
3794
3899
4002
4105
4207
4307
4406
4503
4600
4694
4787
4878
4968
5055
76

-------
Depth
(Inches)
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
Cumulative Volume in Cubic
U. 170 St. W. 177 St.
2319 7912
8056
8195
8330
8461 _
8586 ~ -
8706
8819
8925
9023
9114
9196
9271
9338
9398
9452
9498
9539
9573
9601
9624
9642
9655
9665
9670
9673
9675
Feet
Pur Has Ave.
5141
5224
5305
5384
5460
5533
5603
5671
5734
5794
5850
5900
5945
5983
6015
6042
6062
6078
6089
6096
6100
6101





'b"  in  Figure A-l .
                             77

-------
                                APPENDIX B

                          HOMEOWNER SURVEY  FORMS

                        PURITAS  AREA  QUESTIONNAIRE

                            (PRE-CONSTRUCTION)
              ADDRESS
1.   Have you experienced basement  flooding?


    If so,  when and how often?
2.  What is the deepest that the water has  ever  appeared  in
    your basement:
3.   Does your street experience street flooding  during  a  storm
    or during a rainy season?
    If so, when and how often?
    If the street does flood,  at what location is  the water
    present?
                                   78

-------
                  PURITAS AREA QUESTIONNAIRE

                      (POST-CONSTRUCTION)
ADDRESS:
1.  Have you experienced basement flooding in recent months?
    If so, when, how deep? 	
2.  Does your street experience street flooding during a storm or
    during a rainy season? 	
    If so, where? 	
3.  Do you have any sewage odor problems?
    Explain when. 	
4.  Are these odors more noticeable during rain storms?
                             79

-------
SAMPLING LOC.:   W177
             APPENDIX C
STORM WATER QUALITY SAMPLING RESULTS
                     DATE OF SAMPLE:   July 13,  1981
               Time of Sample Taken (min.)
PARAMETERS 2.5 5.0
BOD mg/1 34 48
Suspended Solids mg/1 98 80
Volatile Suspended
Solids mg/1 15 13
Cadmium mg/1
Total Chromium mg/1
Chloride mg/1
Copper mg/1
Lead mg/1
Sulfate mg/1
Zinc mg/1
Total Organic
Carbon mg/1
COD mg/1
Settleability ml /I
10 --- 15 30 Composite
52 59 51
79 88 80
14 10 10
0.01
LT 0.01
	 6.3
0.02
0.10
	 8.5
	 0.16
12
39
LT 0.18
LT = Less Than
                                    80

-------
SAMPLING LOG.:  W170                    DATE OF SAMPLE:  July 21, 1981
                                  Time of Sample Taken  (min.)
PARAMETERS                               2.5
BOD mg/1                                  51
Suspended Solids mg/1                     91"""-
Volatile Suspended Solids mg/1            32
Cadmium mg/1                              0.01
Total Chromium mg/1                       0.01
Chloride mg/1                             13
Copper mg/1                               0.03
Lead mg/1                                 LT 0.01
Sulfate mg/1                              8.6
Zinc mg/1                                 0.15
Total Organic Carbon mg/1                 7
COD mg/1                                  100
Settleability ml/I                        LT 0.32

LT - Less Than
                                   81

-------
SAMPLING LOG.:  W177                    DATE OF SAMPLE:   July 28,  1981
                                  Time of Sample Taken  (min.)
PARAMETERS              90    JI20     J50     JSO    _2JO     Composite
BOD mg/1                10      29      17      19      21
Suspended Solids mg/1   61      56      40  /"  58      35
Volatile Suspended
  Solids mg/1           13      10       7      16      10
Cadmium mg/1            —      —      —      —      —     0.01
Chromium mg/1           —      --      —      --      —     0.01
Chloride mg/1           --      —      —      --      --     9
Copper mg/1             —      —      —      --      —     0.03
Lead mg/1               —      —      —      --      —     0.01
Sulfate mg/1            -	-      --     27
Zinc mg/1               —      --      —      --      —     0.18
                                                                 o
Total Organic
  Carbon mg/1           —      —      --      --      --     26
COD mg/1                —      —      —      --      —     40
Settleability ml/1      —      	LT 0.18

LT - Less Than
                                    82

-------
SAMPLING LOG.:   W177

 %
PARAMETERS             2.5

BOD mg/1

Suspended Solids mg/1

Volatile Suspended
  Solids mg/1            16
         DATE OF SAMPLE:   July  28, 1981

   Time of Sample Taken  (min.)
5.0      10      15      30      60
27
74
22
82
17
69 >/
18
56
15
110
13
72
 15
14
15
16
12
                                   83

-------
SAMPLING LOG.:  W170                    DATE OF SAMPLE:   August 8,  1981
                                        Time of Sample Taken (min.)
PARAMETERS                     id     Ii2      J£     li    Composite
BOD mg/1                        21      28      12      13    —
Suspended Solids mg/1          210     180  .\-100     110
Volatile Suspended Solids mg/1  65      60      30      30
Cadmium mg/1                    --      --      --      --    0.01
Total Chromium mg/1             —      --      --      --    0.01
Chloride mg/1                   —      —      —      —    11
Copper mg/1                   .  --      ~      -- '     --    0.05
Lead mg/1                       —      —      —      —    0.12
Sulfate mg/1                    —      --      —      —    12
Zinc mg/1                       —      —      --      --    0.26
Total Organic Carbon mg/1       —      --      --      —    36
COD mg/1                        -      -      —      -    110
Settleability ml/1              --      —      —      —    1.2
                                    84

-------
SAMPLING LOG.:   W170
PARAMETERS
BOD mg/1
Suspended Solids mg/1
Volatile Suspended Soilds mg/1
DATE OF SAMPLE:   August 17,  1981
   Time of Sample Taken -(min.)
30      60      90     120
19      19
58  "-  83
21      18
16
70
24
19
61
20
                                    85

-------
SAMPLING LOG.:  W170

PARAMETERS
BOD mg/1
Suspended Solids mg/1
Volatile Suspended Solids mg/1
Cadmium mg/1
Total Chromium mg/1
Chloride mg/1
Copper mg/1
Lead mg/1
Sulfate mg/1
Zinc mg/1
Total Organic Carbon mg/1
COD mg/1
Settleability ml/I
DATE OF SAMPLE:  August 17, 1981
Time of Sample
2.5 5.0 10
23 22 21
90 62 - - 50
31 24 20
—
—
—
—
—
—
—
—
—
~_ _«. __
Taken
15
20
48
20
—
—
—
—
—
—
—
—
—
„_
(min.)
Composite
—
—
—
0.02
LT 0.01
28
0.33
0.07
21
0.22
50
53
0.98
LT = Less Than
                                    86

-------
                                APPENDIX D

               STORAGE TANK PHOTOGRAPHS AND SEDIMENT RECORDS
     Photographs were taken in November, 1981, after the Hydro Brake
control/retention structures had been in operation approximately 18 months.
Each series of photographs starts at the downstream end of the structures
and proceeds up the tanks at approximately 25 foot {7.6 M) intervals.
Each photograph includes descriptive comments using the following format:

     Picture No.

     (a)  Location or approximate distance from downstream bulkhead, etc.
     {b)  Sediment measurements, comments - September, 1980
     (c)  Sediment measurements, comments - November, 1981

     Note that construction timbers were still in place in the Puritas
and W. 177th Street structures.

     Puritas Avenue~ ~

     #1. (a)  View of 16" (41 cm) Hydro Brake in place
         (b)  --
         (c)  --
                                    87

-------
#2.
#3.
(a)   West Bulkhead
(b)   1" (2.5 cm)  standing  water
(c)   1" (2.5 cm)  standing  water
 a)   25 feet (7.6 M)
 b)   1" (2.5 cm)  standing water
 c)   1" (2.5 cm)  standing water

-------
#4.
(a)
 b)
 c)
          50 feet (15  M)
          1" (2.5 cm)  standing water
          1" (2.5 cm)  standing water
#5.   (a)   75 feet (23 M)
     (b)   1" (2.5 cm) standing water
          1" (2.!
(c)
               ,5 cm) standing water
                               89

-------
#6.  (a)  100 feet (30.5 M)
     (b)  1" (2.5 cm)  standing water
     (c)  1" (2.5 cm)  Standing water
#7.  (a)  125 feet (38.1 M) - View of East Bulkhead
     (b)  1" (2.5 cm) standing water
     (c)  Dry
                                                              S
                                90

-------
W. 170th Street
#1.  (a)  Downstream Manhole with 9" (23 cm) Hydro Brake in place
     (b)  3" (7.6 cm) sediment
     (c)  1" (2.5 cm) sediment
#2.
 a)   25 feet (7.6 cm)
 b)   3" (7.6 cm) sediment
(c)   1" (2.5 cm) sediment
                                91

-------
#3.
(a)  50 feet (15 M)
(b)  3" (7.6 cm) sediment
(c)  1" (2.5 cm) sediment
 #4.   (a)   75 feet (23 M)
      (b)   Dry
      (c)   1" (2.5 cm) sediment
                                  92

-------
#5.
     100 feet (30.5 M)
 b)   1" (2.5 cm)  standing water
(c)   1" (2.5 cm)  sediment
#6.   (a)  125 feet  (38.1 M)
      (b)  2" (5.1 cm) standing water
      (c)  3" (7.6 cm) sediment & water
                                                 \
                                93

-------
H. 177th Street

#1.  (a)  7" (18 cm) Hydro Brake in place
     (b)  --
     (c)  --
     (a)  25 feet (7.6 M) - East tank
     (b)  1" (2.5 cm) standing water
     (c)  3" (7.6 cm) sediment
                                94

-------
#3.  (a)  50 feet (15 M)
     (b)  1" (2.5 cm) standing water
     (c)  2" (5.1 cm) standing water
           \
#4.  (a)  75 feet (23 M)
     (b)  2" (5.1 cm) standing water
     (c)  1" (2.5 cm) sediment
                                95

-------
#5.  (a)   100 feet (30.5 M)
      b)   1" (2.5 cm)  standing water
      c)   1" (2.5 cm)  sediment
 #6.  (a)  125 feet (38.1 M) view of South Bulkhead (East tank)
      (b)  2" (5.1 cm) standing water
      (c)  2" (5.1 cm) standing water
                                96

-------
#7.  (a)  North Bulkhead (West tank) - effluent pipe to Hydro Brake
          Manhole
     (b)  2"  (5.1 cm)  standing water
     (c)  2" (5.1 cm) standing water
#8.   (a)  25 feet (7.6 M)
      (b)  1" (2.5 cm) sediment
      (c)  2" (5.1 cm) sediment and water
                                97

-------
 #9.
#10.
 a)  50 feet  (15 M)
 b)  1" (2.5  cm) standing water
 c)  1" (2.5  cm) sediment
 a)
.b)
(c)
75 feet (23 M)
1" (2.5 cm) standing water
1" (2.5 cm) standing water
                                98

-------
#11.
(a)
(b)
(c)
100 feet (30.5 M)
1" (2.5 cm) standing water
1" (2.5 cm) standing water
#12.  (a)  125 feet (38.1 M) view of South Bulkhead (West tank)
      (b)  1" (2.5 cm) standing water
      (c)  1" (2.5 cm) standing water
                                 99

-------
                                APPENDIX E

               1/2  HOUR  DURATION DESIGN STORM HYDROGRAPHS
                                     RAINFALL INTENSITY
                                     IN/HR.(cm./HR.)
                                            INFLOW
                                     	OUTFLOW
                                            {6in./l5cm. HYDROBRAKE)
                                                             \
                                                               \
                                                                \
                                                                 \
                                                                  \
                                                                   \
         liliiiiiiiin^   I    i    i    i
         6   12   IB   24  30  36  42  48  54  60  66  72  78   84  90  96  102
                               TIME (MINUTES)
FIGURE E-l   W. 170 ST. DESIGN  STORM  HYDROGRAPHS
             10 YR. RETURN- 1/2 HR. DURATION (l.37ia-3.48cm.)
                                   100

-------
_.  3-
u.
o
I
LJ
                                    RAINFALL INTENSITY

                                    IN/HR(cm./HR.)
                                             INFLOW
                                       	OUTFLOW
                                             (6in./l5cm. HYDROBRAKE)
   2—
                                                       \
                                                         \
                                                          \
                                                           \
                                                            \
          i   r    i   i    i    i    i    i^    i    i    i   >   i    i    I
          6   12   18   24   30   36  42  48  54  60  66  72  78   84   90

                              TIME (MINUTES)


 FIGURE E-2   W. 170 ST. DESIGN STORM HYDROGRAPHS
             SYR. RETURN-1/2 HR. DURATION(U9ia-3X>2cm.)
                                    101

-------
                                    RAINFALL INTENSITY

                                    IN/HR.(cm./HR.)
                                            INFLOW

                                     	OUTFLOW
                                             (6in./!5cm. HYDROBRAKE)
iI    I    r
18  24  30  36
    42

TIME (MINUTES)
 1    I   I    T
48  54  60   66
T    r
72  78
                                                           t
                                                           84
FIGURE E-3   W. 170 ST. DESIGN STORM HYDROGRAPHS
             2YR. RETURN-1/2  HR. DURATION (.91 in.-2.3lcm.)

                                   102

-------
    5-
5
U-
o
UJ
   2—
       (5)
       (10)
                               RAINFALL INTENSITY
                               IN/HR.(cm./HR.)
       (15)
                                             • INFLOW
-(75)
                                 	OUTFLOW
                                       (6in./l5cm. HYDROBRAKE)
               2.15 CFS(60.9L/s)
                          »    I    I    T
                     24  3O  36  42  48
                      TIME (MINUTES)
                                    I    I
                                   54  60
                                           66
FIGURE E-4    W.I70ST DESIGN STORM HYDROGRAPHS
              IYR. RETURN-1/2 HR. DURATION (.76in.-l.93cm.}
                                     103

-------
                                   RAINFALL INTENSITY
                                   IN/HR.(cm./HR.)
                                             INFLOW
                                         	OUTFLOW
                                             (3ia/7.6cm. HYDROBRAKE)



                                                                    \
                                                             l    I    1
                                                            468  474  480
FIGURE E-5
                  TIME (MINUTES)

W. 177 ST. DESIGN STORM HYDROGRAPHS
10 YR, RETURN-l/2 HR. DURATION(l37ia-3.48cm.)
                                   104

-------
fe
                                     RAINFALL INTENSITY

                                     IN/HR.(cm./HR.)
                                              INFLOW
                                     	OUTFLOW
                                              (6ia/l5cnn HYDROBRAKE)
     I     I
24   30   36  42   48   54
                                                              I     I
                                                             420 426
                               TIME(MINUTES)


 FIGURE E-6   W. 177 ST. DESIGN STORM HYDROGRAPHS
              SYR. RETURN-1/2HR. DURATION (I.l9in.-3.02cra)
                                    105

-------
     5-
   4—
    3—
tn
LL.
o
I
1   *-
LL.
       -(5)

       (10)
       -(100)
           I
           6
                       RAINFALL INTENSITY
                       IN/HR.(cm,/HR.)
                  3.35 CFS (94.9 L/s)
                                               INFLOW
                         	OUTFLOW
                                {6ta/l5cm. HYDROBRAKE)
12
18
 I
24
I
30
  1    I
  36   42

TIME (MINUTES)
\
48
I
54
60
330  336  342
 FIGURE E-7    W. 177 ST. DESIGN  STORM HYDROGRAPHS
               2YR.  RETURN-1/2HR. DURATION (.91in.-2.31cm.)
                                      106

-------
                                     RAINFALL INTENSITY
                                     IN/HR.(cm;/HR.)
cc
                                             INFLOW
                                        	OUTFLOW
                                             (6in/l5cm. HYDROBRAKE)
               I
               12
18   24   30
I
36
1
42
1
48
1
54
60
 I    I     I
288 294  300
                               TIME (MINUTES)
 FIGURE E-8  W. 177 ST. DESIGN STORM HYDROGRAPHS
              IYR. RETURN- 1/2 HR. DURATION (76in.-1.93cm.)
                                      107

-------
                                      RAINFALL INTENSITY

                                      !N/HR.(cm./HR)
                                                 INFLOW
                                      	 OUTFLOW
                                                 (5.5in./l4cm. HYDROBRAKE)
_J
to
i
UJ
S3
ST.


I
FIGURE E-9
                         I    I    i   1   i    \    \    I   r
                     24  30  36  42  48   54  60  66  72  78  84
                            TIME (MINUTES)
PURITAS  AVE. DESIGN STORM HYDROGRAPHS
10YR. RETURN- 1/2 HR. DURATION (l.37in.-3.48cm.)
                                    108

-------
                                    RAINFALL INTENSITY
                                    IN/HR.(cm./HR)
                                             INFLOW
                                           - OUTFLOW
                                             (5.5in./14cm. HYDROBRAKE)
                            TIME (MINUTES)

FIGURE  E-IO  PURITAS AVE. DESIGN STORM HYDROGRAPHS
             SYR. RETURN-1/2 HR. DURATION (U9in.-3.02cm.)
                                   109

-------
                                     RAINFALL INTENSITY
                                     IN/HR.(cm./HR.)
                                               INFLOW
                                     	OUTFLOW
                                                (5.5in./|4cm. HYDROBRAKE)
          I    I   I    I    I    I   II    II
          6    12   18   24  30  36  42  48  54  60  66
                      TIME(MINUTES)

FIGURE E-ll    PURITAS AVE. DESIGN STORM HYDROGRAPHS
              2YR. RETURN- 1/2 HR. DURATION (.9lin.-2.31cm.)
                                    110

-------
to
fe
                                     RAINB\LL INTENSITY

                                     IN/HR.(cm./HR.)
ni  2-
INFLOW

• OUTFLOW
 (5.5in./|4cm. HYDROBRAKE)
                 \    I    f    I   I    I    I
                 18  24  30  36  42   48   54  60

                    TIME (MINUTES)
FIGURE E-12   PURITAS AVE. DESIGN STORM HYDROGRAPHS
              IYR. RETURN - 1/2 HR. DURATION (76in.-1.93 cm.)
                                    m

-------
          APPENDIX F


    HYDRO BRAKE/RETENTION


APPLICATIONS IN CLEVELAND, OHIO


       DESIGN CRITERIA


            AND


      PROJECT SUMMARIES
              by
       City of Cleveland
Department of Public Utilities
      Narrative Sections
              by
     J. Christopher Kocsan
      Technical Sections
              by
      Francis Toldy, P.E.
            112

-------
                                  INTRODUCTION


       The purpose of this appendix is three (3) fold.  First, the City of
Cleveland has recently constructed many projects which are similar in nature
to the evaluated grant project.  The City intends, at this point, to present
descriptions of each of these similar projects, including pertinent statistical
data, and to compare and contrast these similar projects to the evaluated grant
project.
                                                *" v
       Secondly, during the evaluation of the grant project, it became obvious
to all concerned parties that the underground retention tanks which had been
constructed were not being fully utilized as far as their storage capacities
were concerned.  Accordingly, new more restrictive Hydro-Brake flow regulators
were installed within these tanks during the summer of 1981.  The manner in
which these new sizes were determined by the City, is presented within this
section of the appendix.

       The final section of this appendix contains a Design Manual for Storm
Water Retention Facilities.  The City has taken one of its similar projects,
and by using that project as an example, it has demonstrated on a step-by-step
basis the manner in which an interested party would proceed with designing
this type of Storm Water Retention Facilities.  It is believed that the
information contained within this section will prove quite valuable to those
readers who are interested in constructing projects similar to the one which
has just been evaluated.


                         COMPARISON OF SIMILAR PROJECTS

       A comparison of the similar projects which have been constructed by  the
City of Cleveland since the time when the evaluated grant project became opera-
tional will now be presented.

       1.   Southwest Sewer Project - Phase I

            The Southwest Sewer Project - Phase I is a project in which the City
            of Cleveland utilizes an existing creek within the project area to
            receive excess storm water flow which has been created  by heavy rains.
            The City has achieved this by installing thirteen  (13)  4-inch Standard
            Hydro-Brake flow regulators within the existing catch basins which  are
            located at the uphill ends of Kirton Avenue.  The  flow  out of these
            catch basins has been greatly reduced because of  the  installation of
            this patented control device.  Accordingly, since  during  periods of
            heavy storms, the flow into these combined sewers  is  retarded by the
            installed Hydro-Brakes, and since these catch basins  are  also located
            at the uphill ends of Kirton Avenue, a majority of the  storm water
            run-off is bypassing the uphill catch basins and  flowing  into
            the catch basins which are located in the low point  area  of  Kirton
            Avenue.  These catch basins discharge their flow  directly  into  the
            existing creek.
                                      113

-------
     The Hydro-Brakes which were recently installed possess a discharge
     rating of .25 CFS at three (3) feet of head.  A total  of $8,450 was
     expended by the City for the purchase of Hydro-Brakes  concerning
     this project, as these Hydro-Brakes cost $650 each,including installa-
     tion.   The total drainage area along Kirton Avenue which is effected
     by the installation of these Hydro-Brakes is approximately four (4)
     acres.

     Since the project contained a number of different improvements in addi-
     tion to the Hydro-Brake installations, it is not appropriate to discuss
     a per cubic foot cost as far as storage space is concerned.  The con-
     struction cost of this project was $83,487.  In addition, because of
     the referenced additional improvements, it is also not appropriate to
     quote a figure concerning the cost per acre required in order to
     implement this project.

     The major element of the additional improvements which were made,
     above and beyond the installation of the referenced Hydro-Brakes, was
     the rebuilding of sixteen (16) existing catch basins located within
     the low point areas of three streets - Kirton Avenue,  Carrington
     Avenue and Erwin Avenue.  The storm flow from these catch basins was
     rerouted from the existing combined sewer system into a nearby creek.
     This obviously reduced the level of flow to the combined system within
     the area.  In addition, a small playground within the area was enlarged
     and landscaped, and it is now used as an above ground storm water re
     tention basin during periods of intense rainfall and sewer system sur
     charging.

2.   Southwest Sewer Project - Phase II

     The Southwest Sewer Project - Phase II is quite similar to the eval-
     uated grant project.  In this project, the City constructed four
     (4) underground storm water retention tanks within the project area,
     as well as utilitizing the available roadway surface for purposes
     of storm water storage during periods when intense rainfalls cause
     the combined sewer system in the area to become surcharged.  A summary
     of the characteristics of the four undergound retention tanks which
     were constructed is as follows:
        LOCATION

   1.  Crossburn Ave.
   2.  Crossburn Ave.
   3.  Bennington Ave,
   4.  Bennington Ave.
DIMENSIONS   CAPACITY
353' Long
 48" Dia.

400' Long
 48" Dia.

400' Long
  48" Dia.

371' Long
  48" Dia.

      114
4,436
Cu.  Ft.

5,026
Cu.  Ft.

5,024
Cu.  Ft.

4,662
Cu.  Ft.
 OUTLET
REGULATOR

6" Standard
Hydro-Brake

6" Standard
Hydro-Brake

6" Standard
Hydro-Brake

9" Standard
Hydro-Brake
 OUTLET
DISCHARGE
  RATE

 0.6 CFS @
5 Ft. of Head

 0.6 CFS G>
5 Ft. of Head

 0.6 CFS @
5 Ft. of Head

 1.5 CFS @
5 Ft. of Head

-------
All tanks were constructed of corrugated steel pipe, are circular in
diameter, and were positioned at low points on their respective streets.
A total of 13 catch basins along Crossburn Avenue now have their storm
flow diverted into the two newly constructed retention tanks instead
of directly into the combined sewer system.  A total of 14 catch basins
along Bennington Avenue have their storm flow diverted in a similar
manner.  All of the retention tanks eventually discharge their storm
flow through the Hydro-Brakes and into the existing combined sewer
system at the controlled rate of discharge which is regulated by the
Hydro-Brake.

A total of five (5) existing catch basins within the pro.iect area
have been equipped with four-inch Standard Hydro-Brakes in order to
increase the level of storm water runoff flowing towards the low
points of the referenced streets.  At this point, the surface runoff
from the area is dischargeed directly into the retention tanks
through the low point catch basins.  These Hydro-Brakes have discharges
which are rated by the manufacturer at  .25 CFS at 3 feet of head.

The construction cost of the entire project was $175,860.  The pro-
ject effects a total drainage area of 38.6 acres.  Thus, the project
cost $4,556 per acre in order to construct.  The total volume of
storage which is available within the constructed retention tanks is
19,175 cubic feet.  Thus, the project cost $9.17 for every cubic foot
of storage space created.

The costs pertaining to the purchase of the required Hydro-Brakes for
this project, including installation, were as follows:
    Description         Quantity  Price      Extension

9" Standard Hydro-Brake      1     $2,300       $2,300
6" Standard Hydro-Brake      3       1,675        5,025
4" Standard Hydro-Brake      5        915        4.575

                                    Total     $11,900
Thus, a total of $11,900 was  expended  by  the  City  within  the  applic-
able construction contract  for  the  purchase and  installation  of
Hydro-Brakes concerning this  project.
                          115

-------
3.   Southwest Sewer Project - Phase III

     This project, as far as scope of work is concerned, is a combination
     of the preceding two phases of this four phase series of projects.
     Within this phase, the City constructed one underground retention
     facility along West 127th Street between Astor and Crossburn Avenues.
     The referneced tank is 209 feet long, is 66 inches in diameter, and
     has a capacity of 4,965 cubic feet.  The tank's outlet is equipped
     with a 6-inch Standard Hydro-Brake-which has a discharge rating of
     .7 CFS at 5 feet of head.

     A total of seven (7) catch basins  in the area have their flow diverted
     into the newly constructed retention tank.  In addition, four other
     catch basins within the project area have also been equipped with 4-inch
     Standard Hydro-Brakes which possess discharge ratings of .25 CFS at
     3 feet of head.  This further relieves the level of storm flow which
     enters the combind sewer system during periods of sewer surcharging.
     Because of the implementation of this system, the number of incidents
     of surcharging within the project  area have been effectively reduced.
     The total drainage area which is tributary to the constructed reten-
     tion tank is 3.3 acres.

     The total expenses incurred by the City for the purchase of Hydro-Brakes,
     including installation, concerning this project is as follows:
        Description

     6" Standard Hydro-Brake
     4" Standard Hydro-Brake
Quantity  Price
   1
   4
$1,800
 1,050

 Total
Extension

 $1,800
  4,200

 $6.000
     Thus, a total of $6,000 was expended for the purchase and  installa-
     tion of Hydro-Brakes concerning this project.

     Along Longmead, Milligan and McGowan Avenues, which  are  adjacent
     streets located within the project area, thirteen  (13) existing catch
     basins were reconstructed with their storm  flow  being rerouted  from
     the area's existing combined sewer system into a nearby  creek.  This
     further reduces the storm flow entering the combined system within t
     he area.

     Because of these additional improvements, it is  improper to quote a
     project cost based upon per acre  of drainage area  effected.   It is
     also improper to quote a cost per cubic foot of  storage  space created.
     The construction cost of the project was $83,267,  while  the total
     drainage area effected by all elements of the project is 8.5  acres.
                               116

-------
4.   Southwest Sewer Project - Phase IV

     Within this project, the City constructed two (2) underground storm
     water retention tanks at low points along Carrington Avenue.  The
     first of these tanks is 300 feet long, 48 inches in diameter, and
     has a volume of 3,770 cubic feet.  The tank is not equipped with, a
     Hydro-Brake regulator, but with a discharge pipe which is six-inches
     in diameter and which discharges its storm flow into an existing
     18-inch combined sewer.

     The second tank is slightly smaller.  It is 250 feet long, 48 inches
     in diameter, and has a volume of 3,142 cubic feet.  This tank is
     also equipped with a discharge pipe which is six-inches in diameter,
     but which discharges its flow into a 20-inch combined sewer.  Both
     tanks are constructed of corrugated steel pipe.  A total of eight
     catch basins now discharge their storm flow into these two retention
     tanks.

     None of the catch basins which are located uphill from the two reten-
     tion tanks constructed along Carrington Avenue are equipped with
     Hydro-Brakes.

     The total storage volume created by the project is 6,912 cubic feet,
     with a cost per cubic foot of storage space created equaling $7.60.
     The total drainage area effected by the project is 16.1 acres.  The
     cost per acre effected was $3,621.  No funds were expended for the
     purchase of Hydro-Brakes concerning this project.

5.   Retention Catch Basin Demonstration Project
     Phase I-A

     This project is a continuation of the Storm Water Quality Control
     Program which was implemented by the City of Cleveland under the
     authority of this grant.

     This project included the  installation of 12 Hydro-Brakes  in existing
     Catch Basins just north of the area affected by the evaluated grant
     project.  These Hydro-Brakes were installed within catch  basins  located
     along Milburn Avenue, West 168th Street, West  171st Street,  West 173rd
     Street and West 176th Street.  The purpose of  these installations  was
     to allow the excess  storm  runoff to bypass the uphill  catch  basins  so
     that the runoff is collected at the street's low  point.   Eventually,
     this flow will be directly discharged into the new storm  sewer  refer-
     enced within the following paragraph.

     The project also included  the  installation of  a  new storm sewer in Mil-
     burn Avenue and West 168th Street.  This new sewer was connected into  an
     existing 78 inch sewer, thus eliminating the need  for  constructing
     underground retention facilities along either  street.   A  total  of 12
                               117

-------
            catch basins were rerouted from the combined system into the newly
            constructed storm sewer, thus providing relief to the existing-combined
            sewer system and reducing the number of incidents of sewer surcharg-
            ing in the area.  The construction cost of the entire project was
            $100,508, and the drainage area serviced by the project is 12.8 acres.

            As stated previously, a total of 12 Hydro-Brakes were installed with-
            in existing catch basins in order to utilize the roadway surfaces
            for storm water ponding, thus reducing combined sewer surchagring.
            The expenses incurred by the City within the applicable construction
            contract for the purchase and installation of these Hydro-Brakes
            were as follows:

               Description           Quantity   Price     Extension

            4" Standard Hydro-Brake      2      $600       $1,200
            2" Standard Hydro-Brake     10       290        2.900

                                               Total       $4,100
            Thus, a total  of $4,100 was expended by the City for the purchase
            and installation of Hydro-Brakes concerning this project.  Because
            of the additional improvements above and beyond the installation of
            the referenced Hydro-Brakes, the quoting of costs based upon per acre
            of area serviced or cubic foot of storage created concerning this
            project is irrelevant.

                         Summary of Projects 1 through 5


       The sewage flow from all five project areas is eventually discharged into
the 8 foot combined sewer which services West 130th Street and which flows in a
northerly direction.  This sewer terminates at Brook!awn Avenue where the excess
flow from the combined sewer is allowed to overflow into a dual 12 foot by 8 foot
storm sewer.  This occurs during periods when the 8 foot combined sewer is sur-
charged.  During dry weather, when no excess flow exists, the flow from the 8
foot combined sewer flows directly into a No. 5C sanitary line. (Please see
Table C)

       The Southwest Sewer Projects - Phases I through IV, are geographically
located near the Brooklawn Avenue Combined Sewer Overflow.  This means that these
four projects, some of which include the storage of storm water and thus reduce
the level of flow within the West 130th Street sewer, should have had a positive
impact upon the quality of water which  is routed into the streams and rivers within
the area.  It is believed that these projects have reduced the number of overflow
occurrences, as well as reducing the volume of combined sewage which overflows at
this location.  Although the City assumes this to be the situation,  because it
is the most logical result of the implementation of these projects,  it has not
been verified, as neither formal monitoring of the overflow nor a post construc-
tion evaluation study have been undertaken concerning these projects. The City
does not presently plan to undertake a  post construction evaluation  of Projects
                                     118

-------
1 through 5 as funds are rather limited and are not currently available for these
purposes.

       The sewage flow effected by the Retention Catch Basin Demonstration Project
Phase 1-A is also eventually routed into the West 130th Street sewer.  However,
because the location of this project is somewhat removed from the overflow in
question, its impact upon that overflow is not quite as profound or measurable as
the impact of the Southwest Sewer Projects, Phases I through IV.  Nevertheless,
the effect of this project on water quality within the area should not be
underestimated as the project directly effects tlie combined sewer overflow which
is located near the intersection of Puritas Avenue and Interstate 71 (M-15), as
well as indirectly effecting the performance of the Brook!awn Overflow.

       It does appear that the basement flooding within these areas has been
drastically reduced by the construction of these projects.  While the City used
to receive many complaints from the residents of these areas, the present rate
of complaints from these areas regarding basement flooding has been reduced
to a minimal level.  It also does not appear that street flooding has become a
problem in the areas where Hydro-Brakes were installed within the existing catch
basins in order to utilize surface storage technology.  Overall, the projects
seem to have had a positive impact upon the areas which they service and
the City is quite pleased with the results which have been generated by the projects,

       6.   Hamlet Avenue Sewer Project - Phase II

            The City of Cleveland constructed one retention tank in the sidewalk
            area of Hamlet Avenue within the scope of this project.  The tank
            which was constructed is 467 feet in length, 48 inches in diameter,
            and contains a volume of 5,868 cubic feet of storage space for
            purposes for storing storm water runoff.  The tank discharges through
            an outlet pipe which is 6-inches in diamter and into an existing
            No. 3 brick sewer. (Please see Table C)   The tank is constructed  of
            corrugated steel pipe and a total of five catch basins presently have
            their flow routed into the referenced tank.

            The project services a drainage area which encompasses two acres.
            The project's construction cost was $46,509, while the project
            cost $7.93 for every cubic foot of storage space created.  The  project
            can also be considered to cost $23,255 per acre of area serviced.
            No funds were expended for the purchase of Hydro-Brakes concerning
            the project, as none were installed at this location.

       7.   Adolpha Avenue Retention Facility

            Approximately one year after the completion of the Hamlet  Avenue Sewer
            Project - Phase II, it became apparent to the City that  the  completion
            of the referenced project had not solved all of the  basement  and street
            flooding problems which were present within the area.  At  that  time,
            the City's Department of Public Utilities began to  research  different
            alternatives which would lead to the elimination of  all  of the  prob-
                                      119

-------
            lems present within the area.  It was at this time that the City de-
            veloped and decided to construct a project which is now known  as the
            Adolpha Avenue Retention Facility.

            The Adolpha Avenue Retention Facility is an underground storm  water
            retention tank which was constructed in two sections.  The first
            section is 280 feet in length, 48 inches in diameter, and has  a
            storage volume of 3,518 cubic feet.  The second section is 232 feet
            in length, 60 inches in diameter, and has a storage volume of  4,555
            cubic feet.  Total storage volume of the entire tank is 8,073  cubic
            feet and the entire tank is constructed of Corrugated Steel Pipe.

            The tank discharges its flow through a 12-inch outlet which is  equipped
            with a 4-inch Standard Hydro-Brake.  This flow regulator has a dis-
            charge rating of .25 CFS @ 3 feet a head.  The total amount expended
            by the City within the construction contract for the purchase  of this
            Hydro-Brake was $300, including installation.  The tank discharges its
            flow into an existing No. 5 brick sewer.  (Please see Table C)

            A total of seven catch basins, five of which were existing and two of
            which were recently constructed, have their flow routed into the re-
            ferenced retention tank.  The total drainage area which is tributary
            to the retention facility is 2.9 acres.  The project's construction
            cost was $58,260.  Thus, to express expenses in comparative terms,
            the project cost $20,090 per acre of area affected or $7.22 per
            cubic foot of storage space created.


                           Summary of Projects 6 and 7

       The retention tanks constructed in Hamlet and Adolpha Avenues discharge
their flows into the existing Hamlet - Adolpha Sewer.  This sewer then carries
the flow into the No. 12 sewer running along East 65th Street.   (See Table C)
The East 65th Street Sewer is the main interceptor sewer for the area and
it flows in a northerly direction.

       There is a combined sewer overflow located  .7 of a mile north of the  point
where the Hamlet - Adolpha Sewer introduces its flow into the  East 65th Street
Sewer.  This overflow is located at the  intersection of East 65th Street  and Selma
Avenue.  However, because of the large capacity available within the East 65th
Street Sewer, and because of the large number of small sewers  which are tributary
to the East 65th Street Sewer, it is nearly impossible to measure the  positive
effects which these two projects have had upon the overflow under consideration,
and formal attempts to do so have not been undertaken.

       It does appear that the basement  and street flooding problems which  were
prevelant within the area have been reduced by the construction  of these  two
projects, as complaints to the City of this nature have been  reduced to minimal
levels within the past year.  Prior to this, complaints of  this  nature were quite
common as levels of street and basement  flooding sometimes  reached 18  inches.
Thus, the City is quite pleased with the  results of  the two projects,  as  a severe
problem has been eliminated with minimal  capital expense to the City.
                                     120

-------
8.   East 99th Street and Carr Avenue Area Retention
     Facilities

     This project is also quite similar in nature to the evaluated grant
     project.  The project included the construction of three storm water
     retention tanks at low points within a 12.3 acre drainage area.  Two
     tanks were constructed along Carr Avenue, while one tank was construc-
     ted along East 99th Street.  All three tanks were constructed of
     corrugated steel pipe and the totak.storage volume created by the
     construction of these tanks is 8,467/cubic feet.

     The two tanks constructed along Carr Avenue both discharge their flow
     into a common manhole with a 12-inch outlet pipe which is equipped
     with a 7-inch Standard Hydro-Brake.  This Hydro-Brake has a discharge
     rating of .9 CFS @ 5 feet of head and the City expended $1,220 for
     the purchase of the subject Hydro-Brake, including installation,
     within the project's construction contract.

     The larger of the two tanks constructed along Carr Avenue is 230 feet
     in length, 48 inches in diameter, and has a volume of 2,880 cubic feet.
     The smaller of the two tanks is 100 feet long, 48 inches in diameter
     and has a storage volume of 1,257 cubic feet.  The total storage volume
     which discharges through a 12-inch outlet pipe, and which is regulated
     by a 7-inch Standard Hydro-Brake, is 4,147 cubic feet.  A total of six
     (6) catch basins now have their flow routed into the two retention tanks
     which are under consideration.  The tanks in question discharge their
     storm flow into an existing No. 6 brick sewer.  (Please see Table C)

     The tank which was constructed along East 99th Street is 220 feet in
     length, 60 inches in diameter, and has a storage volume of 4,320 cubic
     feet.  A total of three (3) catch basins located along East 99th Street
     now have their storm flow diverted away from the combined sewer system
     and into the referenced retention facility.  The tank dishcarges its
     flow through a 12-inch outlet pipe which is equipped with 5-inch
     Standard Hydro-Brake.  This Hydro-Brake has a discharge rating of
     .5 CFS @ 5 feet of head and cost the City $950 to install, including
     installation.  The flow from the referenced facility is discharged
     into an existing No. 3 brick sewer.  (Please see Table C)

     The total cost incurred by the City for construction of this project
     was $83,202.  A total of 12.3 acres of drainage area was effected by
     the construction of the project.  This project cost the City $6,474
     per acre of drainage area effected.  The project also cost the City
     $9.83 per cubic foot of storage space created.

     A post-construction evaluation of the referenced project has not been
     undertaken.  However, preliminary indications are that the project
     has had a positive effect upon the combined sewer system within the
     area and that basement flooding has been drastically reduced.
                              121

-------
 9.    Parkridge Avenue Retention Facility

      Within this small  project, the City of Cleveland  constructed one (1)
      underground storm water retention tank along Parkridge Avenue.   The
      tank which was constructed is 300 feet in length, 48 inches  in  diameter,
      has a storage volume of 3,770 cubic feet, and is  constructed of Corru-
      gated Steel Pipe.   The flow from two catch basins within the area has
      been routed away from the combined sewer system and into the referenced
      retention tank.

      The tank discharges its flow through an outlet pipe which is 6-inches
      in diameter and into an existing 15-inch combined sewer.  The tank's
      outlet is not equipped with a Hydro-Brake.

      The project cost $29,018 to construct while the total  drainage  area
      which is tributary to the referenced tank is 4 acres.   Thus, the pro-
      ject cost $7.70 per cubic foot of storage space created and  $7,254
      per acre of drainage area effected.  The City has received many posi-
      tive responses from the area's residents concerning this project.

10.    Lakeview Road Area Retention Facilities

      The last of the projects which will be analyzed within this  appendix
      is the Lakeview Road Area Retention Facilities.  Once again, this pro-
      ject is very similar to the evaluated grant project.  The project in-
      cluded the construction of three underground storm water retention
      tanks along cross streets which intersect Lakeview Road.  The project
      also included the strategic placement of Hydro-Brakes within exsiting
      catch basins in order to take advantage of the roadway surface storage
      capacity which is available within the project area.

      The first of these tanks, which was constructed along Ohlman Avenue, is
      384 feet in length, 48 inches in diameter, has a storage capacity of
      4,826 cubic feet, and is constructed of Reinforced Concrete Pipe.
      The flow from six (6) catch basins has been diverted into this tank.

      The tank discharges its flow through a 12-inch outlet pipe which is
      equipped with a 5-inch Standard Hydro-Brake and into a 15-inch com-
      bined sewer.  The tank's Hydro-Brake has a discharge rated at  .5 CFS
      @ 5 feet of head and cost $1.075, including installation.

      Two catch basins located along Ohlman Avenue which are uphill  from
      the retention tank have been equipped with Standard 2-inch Hydro-Brakes
      in order to allow additional storm water runoff to be captured by the
      catch basins located within the low points of Ohlman Avenue.   The dis-
      charge of these Hydro-Brakes is rated at  .06 CFS @ 3 feet of head and
      the subject Hydro-Brakes cost $500 each,  including installation.

      The second tank, which was constructed along Saywell Avenue, is  428
      feet in length, 48 inches in diameter, has a storage capacity  of
      5,378 cubic feet and is constructed of Corrugated  Steel  Pipe.  The
      flow from six (6) catch basins within the area is  discharged into
                               122

-------
            the referenced tank instead of entering  the combined sewer system.
            The tank discharges its flow through a 12-inch outlet pipe which  has
            been equipped with a 5-inch Standard Hydro-Brake and into an existing
            18-inch combined sewer.  The tank's Hydro-Brake has a discharge rated
            at .5 CFS @5 feet of head by the manufacturer and cost $1,075',  in-
            cluding installation.

            In addition, two catch basins located along Saywell Avenue have also
            been equipped with Standard 2-inch^Hydro-Brakes in order to allow
            additional  storm water runoff to be captured at the catch basins
            located within the low point of Saywell  Avenue.  These Hyrdo-Brakes
            have a discharge rated at .06 CFS @ 3 feet of head and cost $500 each
            including installation.

            The third tank was constructed along Fairport Avenue.  This tank is 325
            feet in length, 48 inches in diameter, has a storage capacity of 4,084
            cubic feet, and is constructed of Reinforced Concrete Pipe,  A total
            of 6 catch basins now have their flow diverted into the retention tank.
            The flow from the retention tanks is then discharged through a 12-inch
            outlet which is equipped within a 5-inch Standard Hydro-Brake and
            into an existing 15 inch combined sewer.  The Hydro-Brake which has
            been utilized has a discharge rated at .5 CFS @ 5 feet of head and
            cost $1,075 including installation.

            As with the other two streets, two catch basins along Fairport Avenue
            were equipped with 2-inch Standard Hydro-Brakes in order to aid in
            the capturing of additional surface runoff in the low point area of
            Fairport Avenue.  The Hydro-Brakes which were installed have a dis-
            charge rating of .06 CFS @ 3 feet of head and cost $500 each includ-
            ing installation.

            The construction cost of this project was $177,143.  A total of 19.3
            acres of drainage area are effected by the project.  The total volume
            of storage space now available because of the construction of the
            three retention tanks is 14,288 cubic feet.  Thus, the project cost
            $12.40 for every cubic foot of storage space created, and $9,178 per
            acre of drainage area effected.

                          Summary of Projects 8 through 10

       These three projects are somewhat different than the previously considered
projects in that the effect which these projects have upon Combined Sewer Overflows
is minimal.  The construction of the East 99th Street and  Carr Avenue  Retention
Facilities did not effect the level of combined sewage flowing  into Lake Erie
in any manner, as the storage capacity which was created by the  project  is not
sufficient to have any effect upon the nearest Combined Sewer  Overflow, which
is located 1,050 feet to the west along East 88th Street.  The  project was under-
taken with the intention of reducing basement flooding in  an area where  this
problem was being experienced on a wide spread scale.  With thevconstruction of
the project, this problem has been almost completely  eliminated, as  virtually  no
complaints are being received from this area concerning this problem  which has
plagued this area in the recent past.
                                     123

-------
       The construction of the Parkridge Avenue Retention Facility also has a
minimal effect upon the level of combined sewage overflowing into Lake Erie.  The
nearest Combined Sewer Overflow is located 1,300 feet south of the project, where
a 48 inch combined sewer intersects with the six (6) foot Big Creek Interceptor.
The retention capacity created by the construction of a single retention tank on
Parkridge Avenue does not create enough storage volume to have a significant impact
upon the referenced overflow.  Once again, the intention of the City in construct-
ing this project was to eliminate basement flooding in an area which was badly
affected by this problem, and the elimination or reduction of combined sewer over-
flows was only a secondary consideration.

       The construction of the last project to be considered, the Lakeview Road
Retention Facilities, most probably had a profound effect upon the nearest Combined
Sewer Overflow, which is located at the intersection of Lakeview Road and Hopkins
Avenue, just three blocks north of the project area. Because of the large volume
of storage capacity created by the project, it is anticipated that the level of
and the number of occurances of combined sewer overflows at this location has been
drastically reduced.  However, without the benefit of a post - construction eval-
uation concerning the referenced project, it is impossible to quantify and verify
that a reduction in the level and the number of overflows occuring at this location
has taken place.

       As with the other two projects, the construction of the Lakeview Road Area
Retention Facilities has virtually eliminated the basement flooding problem which
was prevelant within the project area prior to construction.  Due to lack of funds,
an evaluation of the effect which this project has had upon the nearest Combined
Sewer Overflow can not be undertaken at this point in time.

                               General Comments

       It may appear from the information given within the preceeding text, as
well as from the information given within Table A and Table B, which immediately
follows these comments, that the installation of Hydro-Brakes significantly in-
creases the cost of a project as far as the acreage of drainage area effected, as
well as the cost per cubic foot of storage space created.  For example, from the
projects which are cited, those projects which do not utilize Hydro-Brakes cost
approximately $7.50 per cubic foot of storage space created, while the cost per
cubic foot of storage space created for those projects which do utilize Hydro-
Brakes ranges from $9 to $12.  In addition, for those projects cited, the cost
per acre of drainage area effected for projects which utilize Hydro-Brakes  is
generally higher than the cost per acre of drainage area effected for projects
which do not utilize Hydro-Brakes.

       This may lead the reader to believe that the utilization of Hydro-Brakes
within a project significantly increases the cost of that project.   In fact,
this is not the case.  The reader should notice from the text of  the appendix,
as well as from the information contained within Table A, that the actual  cost  of
Hydro-Brake regulators on a per unit basis is rather insignificant.  It should
also be noted that concerning those projects in which multiple Hydro-Brakes have
been utilized, the cost of these units is rather insignificant when  compared  to
the total construction cost of the project.  Therefore,  it  is purely coincidental
that the cost per acre of drainage area effected, as well as the  cost  per  cubic
                                      124

-------
foot of storage space created, appears to be significantly higher  for those pro-
jects which utilize Hydro-Brakes than for those projects which do  not.

       The reader may have also noted from the text of this appendix, as well as
from the information given within Table A, that it appears that  from project to
project, the price of Hydro-Brakes flucuates widely even though  the sizes of
Hydro-Brakes used within the projects may be equivalent.  The reason for this
phenomenon is the fact that the prices quoted for Hydro-Brakes within this  report
are the prices which were bid by the various contractors who constructed the
projects.  The prices for these units within eSch of  the applicable construction
contracts contain may different elements such as the  actual cost of the device,
overhead, profit, etc.  The inclusion of these considerations within the prices
quoted by the various contractors for the purchase and installation of these
control devices makes it impossible to quote the actual price which was paid for
the purchase and installation of Hydro-Brakes within  each of these projects.
                                      125

-------
                                                                                     TABU_A

                                                                               COMPARATIVE  SUMMARY
                                                                                      W"
                                                                               EVALUATED  PROJECTS
ro
                    PROJECT  TITLE

          1.    Southwest  Sewer  Project
                    Phase I

          Z.    Southwest  Sewer  Project
                    Phase II
         3.    Southwest Sewer Project
                    Phase  III
         4.   Southwest Sewer Project
                   Phase  IV
         5.   Retention Catch Basin
              Demonstration Project
                   Phase IA
         6.   Hamlet Avenue Sewer Project
                   Phase II

         7,   Adolpha Avenue Retention
                   Facilities


4
1
2

1
1
METHOD
OF
STORAGE
Surface
Ponding
Retention Tanks
Plus
Surface Ponding
Retention Tank
Plus
Surface Ponding
Retention Tanks
Plus
Surface Ponding
Surface Ponding
Retention Tank
Retention Tank
VOLUME
OF
STORAGE
H/A
19,148
Cu. Ft.
4,965
Cu. Ft.
6,912
Cu. Ft.
N/A
5,868
CU. Ft.
8.073
Cu. Ft. .
OUTLET
REGULATOR
N/A
One 9" Standard
Hydro -Brake
and
Three 6" Standard
Hydro-Brakes
One 6" Standard
Hydro-Brake
Two 6" Orifices
N/A
One 6" Orifice
One 4" Standard
Hydro-Brake
COST
OF
REGULATOR
N/A
$ 2,300
Each
1,675
Each
$ 1,800
Each
N/A
N/A
N/A
$ 300
Each
CATCH
BASINS
REGULATED
13
5
4
0
12
0
0
CATCH
BASIN
REGULATOR
4" Standard
Hydro-Brakes
4" Standard
Hydro-Brakes
4" Standard
Hydro-Brake
N/A
Two 4" Standard
Hydro-Brakes
and
Ten 2" Standard
Hydro-Brakes
N/A
N/A
/
COST
OF
REGULATOR
$ 650 Each
$ 915 Each
$ 1,050 Each
N/A
$ 600 Each
$ 290 Each
N/A
N/A

-------
                                                                                      TABLE  A
                    PROJECT TITLE

           8.  East 99th Street & Carr Ave.
                    Retention Facilities
           9.  Parkrldge Avenue
                    Retention Facilities
10.
               Lake view Road Area
                    Retention Facilities
                                                                               COMPARATIVE  SUMMARY

                                                                               EVALUATECTPROJECTS
(Continued)
METHOD
OF
STORAGE
3 Retention Tanks
1 Retention Tank
3 Retention Tanks
Plus
Surface Ponding
VOLUME
OF
STORAGE
8,467
Cu. Ft.
3,770
Cu. Ft.
14,288
Cu. Ft.
OUTLET
REGULATOR
One 7" Standard
Hydro-Brake
and
One 5" Standard
Hydro-Brake
One 6" Orifice
Three 5" Standard
Hydro-Brakes
COST CATCH CATCH COST
OF BASINS BASIN OF
REGULATOR REGULATED REGULATOR REGULATOR
$ 1,220 0 N/A $ N/A
Each
$ 950
Each
N/A 0 N/A N/A
$ 1,075 6 2" Standard $ 500 Each
Each Hydro-Brake
f\>

-------
                                                                                     TABLE B
               PROJECT TITLE

           1.  Southwest Sewer Prolect
                    Phase I

           2.  Southwest Sewer Project
                    Phase II

           3.  Southwest Sewer Project
                    Phase III

           4.  Southwest Sewer Project
                    Phase IV

           5.  Retention Catch Basin
               Demonstration Project
                    Phase IA

           6.  Hamlet Avenue Sewer Project
                    Phase II

           7.  Adolpha Avenue Retention
->                  Facilities    "
tvi
00         8.  East 99th Street & Carr Ave.
                    Retention Facilities

           9.  Parkridge Avenue Retention
                    Facilities

          10,  Lakevlew Road Area Retention
                    Facilities
COST SUMMARY
EVALUATED~PROJECTS
PROJECT
COST
$ 83,487
$ 175,860
$ 83,267
$ 53,503
$ 100,508
$ 46,509
$ 58,260
$ 83,202
$ 29,018
$ 177,143
DRAINAGE AREA
EFFECTED
4.0 Acres
38.6 Acres
8.5 Acres
16.1 Acres
12.8 Acres
2.0 Acres
2.9 Acres
12.3 Acres
4.0 Acres
19.3 Acres
PROJECT COST STORAGE SPACE
PER ACRE CREATED
N/A N/A
$ 4,556 19,148 Cu. Ft.
N/A 4,965 Cu. Ft.
$ 3,621 6,912 Cu. Ft.
N/A N/A
$ 23,255 5,868 Cu. Ft.
$ 20,090 8,073 Cu. Ft.
$ 6,474 8,457 Cu. Ft.
$ 7,254 3,770 Cu. Ft.
$ 9,178 14,288 Cu. Ft.
COST PER CUBIC FOOT OF
STORAGE SPACE CREATED
N/A
$ 9.18
N/A
$ 7.60
N/A
$ 7.93
J 7.22
$ 9.84
$ 7.70
.' $ , 12.40

-------
                                     TABLE C

        COMPARISON BETWEEN CITY OF CLEVELAND EGG SHAPED BRICK SEWER SIZES
                                       AND
                PRESENT DAY SEWERS WHICH ARE CIRCULAR IN DIAMETER
                         WATER WAY AREA
EGG SHAPED SEWER               IN              ...  EQUIVALENT CIRCULAR
REFERENCE NUMBER           SQUARE FEET        .. ^    SIZE IN INCHES

     1                        1.00                  '    15
     2                        2.52                      21
     3                        4.33                      27
     4                        6.35                      33
     5                        8.55                      36
     6                       10.90                      42
     7                       13.39                      48
     8                       16.00                      54
     9                       18.72                      60
    10                       21.54                      N/A
    11                       24.46                      66
    12                       27.47                      72
    13                       30.57                      N/A
    14                       33.74                      78
     Letter suffixes which are added to  sewer reference numbers  denote  the
     following type of construction:

          A - 1 Ring of brick around

          B - 1 Ring of brick around,  1  extra ring  on  arch

          C - 2 Rings of brick around

          D - 2 Rings of brick around, 1 extra  ring on arch

          E - 3 Rings of brick around
                                      129

-------
                          HYDRO BRAKE RESIZING DECISION


     As stated previously within the main body of this report, during the
course of the post-construction evaluation of the demonstration grant project,
it became evident to all concerned parties that the retention facilities which
had been constructed were not being utilized in an efficient manner.  Specifi-
cally, a large percentage of the storage volume created by the construction of
the three retention tanks was never being utilized.  This situation was verified
by the fact that during the course of 1980, the storm water which was captured
and stored within these retention tanks never reached a depth which exceeded a
few inches.

     It was apparent to the City that one of the following three factors was
adversely effecting the performance of the referenced retention tanks:

     A)   The retention tanks which were constructed were too large and
          were oversized.

     B)   The drainage areas which were tributary to each of the retention
          tanks were too small to capture enough storm flow to utilize the
          storage space available within these tanks.

     C)   The Hydro-Brakes which had been installed within each of the tanks'
          outlets at the recommendation of the designer did not restrict the
          discharge of the tanks sufficiently in order to take advantage of
          the storage capacity which had been created.

     It was fairly obvious that since the facilities had already been constructed,
Factors A and B could not be altered except at great expense to the City.  For  that
reason, Factors A and B were considered as given.

     Because of this situation, the City decided to attempt to modify Factor  C
so that the retention tanks which had been constructed would be utilized more
efficiently, thus taking advantage of the storage capacity which had been  created.
In order to do this, the City decided that the Hydro-Brakes within each of the
tanks' outlets should be removed and replaced with Hydro-Brakes which were more
restrictive in terms of their discharge.  This decision was made early during
1981, with the installation of the more restrictive Hydro-Brakes tentatively
scheduled for May 1, 1981.

     As indicated within the main body of this report, problems were experienced
with the delivery of the new Hydro-Brakes which were selected by the City  for
installation within the retention tanks.  After numerous delays, these devices
were finally received and installed in mid-summer of 1981.  This provided  the
City and its consultant a minimal period for evaluating and monitoring the per-
formance of the referenced Hydro-Brakes.  It should be pointed  out,  however,  that
during the period of time in which the Hydro-Brakes were being  evaluated,  much
meaningful data was gathered.  This data has subsequently  been  considered  within
preceeding sections of  this report.
                                      130

-------
     Prior to the City's decision to select alternative sizes of Hydro-Brakes
for installation, it was necessary for the City to make a number of assumptions
in order to be able to select those Hydro-Brakes which were of a proper size and
which possessed an appropriate discharge rating.  The assumptions which were made
are as follows:

     1.   The areas from building set-back line to building set-back line
          are tributary to the retention tanks.

     2.   The discharge of all Hydro-Brakes was .assumed to be relatively constant.

     3.   The Hydro-Brakes which were installed within the existing catch basins,
          and which discharge their flow directly into the combined sewer system,
          were not taken into consideration during the course of this evaluation.
          The reason why these Hydro-Brakes which had been installed within the
          existing catch basins were not taken into consideration is because the
          level of impact which these units have upon the volume of storage space
          utilized within the constucted retention tanks is minimal.

     4.   The volumes of the retention tanks as computed by Snell Environmental
          Group were assumed to be correct.

     Having made the preceeding assumptions, the City then used the following
methodology in order to arrive at its decision to request that the manufacturer
replace the originally selected Hydro-Brakes with Hydro-Brakes which were smaller,
and more restrictive, in terms of their discharge rating.

     1.   Areas which are tributary to each of the retention tanks were
          calculated.  (Table D)

     2.   The various drainage areas effected by the project were evaluated
          for various storm frequencies of one (1) hour duration.  (Table E).
          Following this analysis, the drainage areas were then evaluated for
          ten (10) year storms of various durations.  (Table F).  The one (1)
          hour, ten (10) year storm produced the largest volume of storage util-
          ized within the retention tanks.  It was determined that using this
          storm as a design criteria was far superior to using any other design
          criteria which had previously been recommended.

     3.   Hydrographs for a ten (10) year, one (1) hour storm were then developed
          for each drainage area.  (Figures 1, 2, 3).  A relationship between  hydro-
          graph volume and the rate of discharge was then developed to utilize the
          existing storage volume of the retention tanks located within the  drainage
          areas.

     4.   Hydrographs for a five (5) year, one  (1) hour storm duration were  then
          developed for comparative purposes.   (Figures 4, 5, and 6).

-------
     From the various hydrographs which were developed, the City was able to de-
termine which Hydro-Brake discharge rates would provide maximum utilization of
the storage space which is available in the constructed retention tanks.  In order
to make this determination, a 10 year storm of one hour duration was selected as a
design criteria for the previously mentioned reasons.

     Having selected the design criteria, the Engineering Section of the City's
Division of Water Pollution Control moved forward with its attempt to determine
the sizes and discharge rates of Hydro-Brakes inhich would utilize the storage
space which had been created in an efficient manner.  A trial and error methodo-
logy was used in that calculations of the volumes of storage space utilized for
each retention tank were computed under Hydro-Brakes of various sizes and possess-
ing various discharge ratings.  Results were compared, and those Hydro-Brakes
which provided the largest volume of storage space utilization, that is, those
Hydro-Brakes which stored the greatest volume of storm water runoff without creat-
ing a maintenance problem, were chosen for installation.

     The characteristics and capacities of each retention tank were considered
independently during the course of the analysis.  The Engineering Section of the
Division of Water Pollution Control arbitrarily determined that any Hydro-Brake
which possessed a discharge rating of less than .25 CFS would be a maintenance
problem because of clogging and possible blockage.  Taking these factors into
consideration, the existing retention tanks were modified by the installation
of new Hydro-Brakes.  The following is a summary of the characteristics of these
retention tanks after the new Hydro-Brakes had been installed, including a summary
of the tank's expected level of performance during a ten year storm of one hour
duration.

     1.   West 177th Street Tank
          Regulator:  3" Hydro-Brake
          Discharge rating:   .25 CFS

          Capacity of Retention Tank:
          Volume of water stored:
          Excess capacity:

          Puritas Avenue Tank

          Regulator: 5.5" Hydro-Brake
          Discharge rating:   1.0 CFS

          Capacity of Retention Tank:
          Volume of water stored:
          Excess capacity:

          West 170th Street Tank

          Regulator:  6" Hydro-Brake
          Discharge rating:   1.25 CFS

          Capacity of Retention Tank:
          Volume of water stored:
          Excess capacity:
8,984 cubic feet
7,900 cubic feet
1,084 cubic feet
 6,024 cubic feet
 5,000 cubic feet
 1,024 cubic feet
 2,074 cubic feet
 1,900 cubic feet
   174 cubic feet

-------
     This completes the analysis of the methodology used by the City of Cleveland
in computing the sizes and discharge ratings of the Hydro-Brakes which are currently
in place within the referenced retention tanks.  The figures and tables which
are referenced in this section will now be presented.
                                     TABLE D

                          Calculation of Drainage Areas

West 177th Street Tank
     (Right-of-way set backs) + (Right-of-way) = Drainage Area
     (50'+30'+30')1135' + (40'x290') = 136,450 Ft.2
          Drainage area =3.1 Acres
          Total area involved is 7.5 acres.

Puritas Avenue Tank
     (Right-of-way set backs - W.  172) + (Right-of-way set backs - W.  170) +
     (Right-of-way set backs - Puritas) + (Right-of-way - W. 172 & W.  170)
     (40'+30'+30')518' +  (40'+30'+30')222' +  (80'+30'+30')504' + 40'(78'+78')
         51,800 Ft.2   +   22,200  Ft.2     +   70,560 Ft.2     + 6,240 Ft.2

          Drainage area   = 150,800 Ft.
          Drainage area   =3.5 acres
          Total area involved is 7.9 acres.

West 170th Street Tank
     (Right-of-way set backs - Martha Rd.) +  (Right  of way - W.  170)
     (401 + 30')504'  + (40' + 30'  + 30')679'
          35,280 Ft.2     +   67,900 Ft.2 = 103,180 Ft.2
         . Drainage area   =2.4 acres
          Total area involved is 5.7 acres.
                                      133

-------
             TABLE E
Effect
West 177th Street Tank
Frequency
Depth
Total Volume
Run-off Volume
Puritas Avenue Tank
Frequency
Depth
Total Volume
Run-off Volume
West 170th Street Tank
Frequency
Depth
Total Volume
Run-off Volume
of Project on Storms of One

6 Mo.
.66"
7,500
3,750

6 Mo.
.66"
8,300
4,150

6 Mo.
.66"
5,600
2,800

1 Yr.
.90"
10,200
5.100

1 Yr.
.90"
11,300
5,650

1 Yr.
.90"
7,700
3,850

3 Yr.
1.30"
14,800
7,400

3 Yr.
1.30"
16,300
8,150

3 Yr.
1.30"
11,100
5,500
Hour Duration

5 Yr.
1.50"
17,000
8,500

5 Yr.
1.50"
18,800
9,400

5 Yr.
1.50"
12,900
6,450

10 Yr.
1.80"
20,400
10,000

10 Yr.
1.80"
22,600
11,300

10 Yr.
1.80"
15,500
7,750
Volumes are stated in cubic feet.
             134

-------
                                     TABLE F

            Effect of Project on 10 Year Storms of Various Durations

West 177th Street Tank

Duration                15 min.   30 min.   1 Hr;     2 Hr.     4 Hr.     6 Hr.
Depth                     1.13"     1.55"   1.80"     2.10"     2.20"     2.30"
Total Volume            12,850    17,600  20,400    23,800    25,000    26,000
Run-off Volume           6,425     8,800  10,200    11,900    12,500    13,000
Puritas Avenue Tank
Duration
Depth
Total Volume
Run-off Volume
West 170th Street Tank
Duration
Depth
Total Vol ume
Run-off Volume
15 min.
1.13"
14,200
7,100
15 min.
1.13"
9,700
4,850
30 min.
1.55"
19,500
9,750
30 min.
1.55"
13,300
6,650
1 Hr.
1.80"
22,600
11,300
1 Hr.
1.80"
15,500
7,750
2 Hr.
2.10"
26,400
13,200
2 Hr.
2.10"
18,000
9,000
4 Hr.
2.20"
27,600
13,800
4 Hr.
2.20"
18,900
9,450
6 Hr.
2.30"
28,900
14,450
6 Hr.
2.30"
19,800
9,900
                       Volumes are stated in cubic feet.
                                     135

-------
                     TIME - MINUTES
                                                    A  « 3.1 Ac.
                                                    C  « 0.5
                                                  T.C. « 25 MIN.

                                                  INFLOW

                                                  OUTFLOW 3" HYDROBRAKE


                                                  TANK CAPACITY 8984 C.F.
                                                              120
640
FIGURE 1    WEST 177 STREET TANK
            HYDROGRAPH 10 YEAR STOKM,   1 HR. DURATION
                          136

-------
                   6.13
6_
                                                        A  « 3.5 Ac.
                                                        C  - 0.5
                                                      T.C. « 25 MIN.

                                                      INFLOW

                                                      OUTFLOW 5.5" HYDROBRAKE

                                                      TANK CAPACITY 6024 C.F.
               I
              20
r
40          60
     TIME - MINUTES
80
100
120
            FIGURE 2    PURITAS AVENUE TANK
                        HYDROGRAPH 10 YEAR STORM,  1 HR. DURATION
                                    137

-------
5 -
                                                         A
                                                         C
                                                       T.C.
                                   2.4 Ac.
                                   0.5
                                   25 MIN.
                   4.19
4 -
                                                        INFLOW
                                                    —  OUTFLOW  6" HYDROBRAKE

                                                        TANK CAPACITY 2074 C.F,
2-
1 -
               i
              20
)           60

TIME - MINUTES
80
100
120
            FIGURE 3    WEST 170 STREET TANK
                        HYDROGRAPH 10 YEAR STORM, 1 HR. DURATION
                                 138

-------
  5-
                     4.64
to
u
  2 .
  1 -
                            A • 31. Ac.
                            C - 0.5
                          T.C. « 25 MIN.

                          INFLOW

                          OUTFLOW 3" HYDROBRAKE

                          TANK CAPACITY 8984 C.F.
                            40
     60

TIME - MINUTES
80
100
120
580
              FIGURE 4    WEST 177 STREET TANK
                          HYDROGRAPH 5 YEAR STORM, 1 HR. DURATION
                                        139

-------
1 -
                                                         A  -  3.5 Ac.
                                                         C  "  0.5
                                                       T.C.  -  25 MIN.

                                                       INFLOW
                                                   _  OUTFLOW

                                                       TANK CAPACITY 6024 C.F.
               20
40          60          80
      TIME - MINUTES
              FIGURE 5    PURITAS AVENUE TANK
                          HYDROGRAPH 5  YEAR STOBM,  1 HR. DURATION
                                    140

-------
5.
                  3.51
                               A  - 2.4 Ac.
                               C  » 0.5
                             T.C. - 25 HIM.
                             INFLOW
                             OUTFLOW 6" HYDROBRAKE
                             TANK CAPACITY 2074 C.F.
2-
1-
              i
             20
40          60
        TIME - MINUTES
100
120
140
            FIGURE 6    WEST 170 STREET TANK
                        HYDROGRAPH 5 YEAR STORM,  1 HR. DURATION
                                    141

-------
            Guidelines For Design of Storm Water Retention Facilities


     The third and final section of this appendix demonstrates the manner in
which an interested party should proceed when designing Storm Water Retention
Facilities which are similar in nature to those constructed under the auspices
of the evaluated grant project.  Our analysis will proceed by using one of
the comparable projects which has previously been described within this
appendix as an example.  The manner in which thjs project, known as the
Southwest Sewer Project - Phase II, was designedly the City of Cleveland
will be considered in detail.  It is hoped that from this description, other
agencies which desire to implement this type of storm water control technology
will be able to proceed with construction of their own facilities.  It should be
noted thaj: additional, more detailed information concerning the design of these
types of storm water retention facilities is available directly from the City of
Cleveland's Department of Public Utilities, Division of Water Pollution Control.

     The projects which were described within the preceding sections of this appendix
were designed by the City of Cleveland for construction in areas of the City
where both combined sewers are located and basement flooding problems are prevelant.
The purpose of these projects was to provide the residences within the respective
project areas protection from basement flooding up to the level of a five year
storm of one hour duration.

     Utilizing this design criteria, the following methodology was used by  the
City of Cleveland in designing the Southwest Sewer Project - Phase II.

     1.   The volume of storm flow which discharges directly into the existing
          combined sewer system from the house connections located within the
          project area was calculated using the Rational Method.  The following
          methodology was used to make this calculation.

          a.   The roof areas, and those yard areas (A) which drain into the house
               connections which are located within the project area, were  estimated.

          b.   Run-off Coefficients (C) for both the roof and yard areas within
               the project area were estimated.

          c.   The time of concentration at the proposed discharge point,  in this
               case a manhole, where the volume of flow can  be  checked  for  retention
               purposes, was estimated.

          d.   Intensities (i) were obtained from rainfall  intensity  curves.

          e.   The peak rate of flow was calculated using the  Rational  Method
               Equation. (Q=CiA)

     2.   The capacity of the existing  sewer was  calculated  using Manning's
          Equation for calculating velocity of  flow.

                                     (R2/3  S1/2)

          Q  (capacity) = velocity x cross-sectional area  of  pipe  = vA


                                      142

-------
     3.   A hydrograph at the subject manhole was constructed.  The volume of flow
          which was used in constructing the referenced hydrograph included all
          storm run-off tributary to the referenced manhole.

     4.   The capacity of the existing sewer was compared with the peak flow
          of the house connections as calculated in step number one.  Because the
          capacity of the existing sewer was greater than the peak flow, the peak
          flow was subtracted from the referenced capacity.  The difference be-
          tween these two figures was utilized /for Hydro-Brake design.

     The Construction of the Southwest Sewer Project - Phase II enabled the City
of Cleveland to utilize Hydro-Brake Control Technology for two purposes.

     A.   To control  the flow from the constructed retention tanks into the
          existing combined sewer system.

     B.   To control  the flow from regulated catch basins into the existing
          combined sewer system. This was accomplished by the installation of small
          Hydro-Brakes in existing catch basins within the project area.  These
          installations forced portions of the storm run-off within the area
          to bypass the regulated catch basins, which are connected to the existing
          combined sewer, and to flow directly into the catch basins which are
          .onnected to the constructed retention tank.  The implementation of this
          technology also provided for additional storage of storm water run-off,
          as run-off is stored on the surface of the street, as well as within the
          constructed retention tanks.

     In order to complete the storage tank design, a new hydrograph was constructed
based upon existing hydrographs.  This new hydrograph took into account the flow
calculation computed in accordance with Step No. 1.  The revised hydrograph also
considered the amount of run-off which was to be routed into the proposed retention
tanks and discharged through the Hydro-Brakes.  It was determined that all volume
not accounted for would be stored on the surface of the pavement.

     Accordingly, the available surface storage areas were surveyed and the volume
of storm water run-off which could be stored on the surface without causing serious
street flooding problems was calculated.  After this calculation had been made, the
volume of required storage, as determined by the revised hydrograph, was computed.
The volume of water to be stored on the surface was subtracted from the total
storage required, thus yielding the volume of storm water to be stored within  the
retention tanks. These retention tanks are usually constructed and  installed within
a sidewalk or treelawn area.
                                     143

-------
     As stated previously the Southwest Sewer Project - Phase II is to be utilized
as example of the demonstrated storm water control technology.  Accordingly the
reader's attention is drawn to the following calculations which were generated
using the data available concerning the Bennington Avenue Area, a sub-section of
what is now known as the Southwest Sewer Project - Phase II.

     1.   Calculation of storm flow which discharges directly into the Combined
          Sewer System.
          Description

          68 Homes
          34 Garages
          34 Yards
                      Area x Run-off .Coefficient

                           [68(25'x30')]x0.9
                           [34(20'xlO')]x0.9
                           [34(20'x25')]x0.5
                          Total  (CA)

                          45,900 Sq.  Ft.
                           6,120 Sq.  Ft.
                           8,500 Sq.  Ft.
                          60,520 Sq.  Ft.
Time of concentration
Intensity (5 yr. frequency)
                                          t= 15 minutes
                                          i=3.4 inches per  hour
                                Q=CiA=60,520 x 3.4

                                Q=CiA=205,768
          Conversion to CFS
                        _ 205,768 =
                        '  43,560
     2.   Calculation of the  capacity  of the  existing  combined  sewer system.

               Size of pipe                = 18"  Vitrified  Clay  Pipe
               Grade                       =  .5 %
               Roughness Coefficient,  N    = 0.015
               Hydraulic  Radius
                                 R =
Area
Permitneter
                                           K *~ w~™ -  ~ U • o / D
                                      144

-------
               Velocity                   v =  ( -^i )(R 2/3 S 1/2)

                                          v=(^f5)(-3752/3X.0051/2)
                                          v = 3.65 FPS
               Capacity                   vA = '(3.65) (1.77)
                                          vA = 6.5 CFS
     3.   The total volume of flow as determined by the revised hydrograph  is
          22,350 cubic feet.


          Utilizing this  information, it was determined that the  retention  tank
to be constructed would be controlled with a 6  inch Hydro-Brake which possessed a
discharge rating of 0.6 CFS.  The determination was also made to  regulate four
additional catch basins which are located upstream away from the  referenced
retention tanks.  Three regulated catch basins were modified by installing  4"
Hydro-Brakes, with discharge ratings of .25 CFS.

     With these modifications,  the total discharge into the existing combined
sewer system within the area was calculated as  follows:

               4.7 CFS +  0.6 CFS + (4 x .25)CFS = 6.3  CFS

     This level of discharge is less than the maximum  discharge rate of  the system
.which was calculated previously.  Thus, the design of  the  referenced retention
tank was continued with the following calculations being generated.

     1.   Volume discharged into the system through the Hydro-Brakes.

                      V = 1.5 x 70 x 60 = 6,300 Cubic  Feet.

     2.   Volume discharged into the system through house  connections.

                      V = ( 4'7xfx6° ) = 7,755 Cu. Ft.


     3.   Total volume to be stored.

                      V = 22,350 - 6,300 - 7,755 = 8,295 Cubic  Feet.
                                      145

-------
     4.   Estimate of volume required to flood an area, 220'  x 30', with an
          average depth of 6".

                      V = 200' x 30' x 0.5' = 3,300 Cubic Feet.

     5.   Total  volume to be stored in retention facility.

                      V = 8,295 - 3,300 = 4,995 Cubic Feet.


     In order to store the volume of water required, it was decided that 400 feet
of reinforced concrete pipe, 48" in diameter, would be used.   This yields the
following available capacity for storage purposes.

                      V = 400 Lineal Ft. x 12.56 Cubic Ft. = 5,024 Cu. Ft.

     The final calculation which needed to be made was the drain down time required
to discharge the entire volume of storm flow.

               Tank: 5,024 Cu. Ft. * 0.6 = 8,373 seconds = 140 minutes
             Street: 3,300 Cu. Ft. * 0.6 = 5,500 seconds = 92 minutes

          Total  drainage time = 140 minutes + 92 minutes = 232 minutes.

          This is equivalent to 3.9 hours.
                                  CONCLUSION


     This concludes Appendix F which has been authored by the City of
Cleveland for the purpose of relating the experiences which it has had concerning
the type of storm water control technology which has been evaluated during the
course of this demonstration grant project.  The City welcomes constructive
comments from other interested parties who have have similar experiences related
to this type of technology.  These comments should be addressed as follows:


                      Commissioner, Division of Water Pollution Control
                      City of Cleveland
                      1825 Lakeside Avenue
                      Cleveland, Ohio  44114

                      Attention: Engineer of Sewer Design
                                      146

-------
                APPENDIX 6

     HYDRO BRAKE DEMONSTRATION PROJECT

SANTEE DRAINAGE AREA - ROCHESTER, NEW YORK


                    by

      O'Brien & Gere Engineers, Inc.
         Syracuse,  New York 13221
              February,  1982
                    147

-------
             HYDRO-BRAKE DEMONSTRATION PROJECT
        SANTEE  DRAINAGE AREA - ROCHESTER, NEW YORK
GENERAL

     The  concept of inlet  control to reduce  downstream  sewer system
surcharging and resultant CSO discharges was  demonstrated in the Santee
Drainage Area in  the City of Rochester, New York.

     Past practices of removing stormwater runoff as quickly  as  possible
have  resulted  in moving  flooding  and  pollution  problems downstream,
potentially inducing  more  serious damage.   By  providing inlet control
such that the  rate of stormwater inflow  does  not exceed the  capacity of
the  existing collection  system,  problems  such  as  basement back-ups,
shock loadings  to downstream treatment facilities,  and CSO discharges can
be minimized or eliminated.

     Inlet control practices by their nature, increase the  frequency and
extent of surface flooding.   Surface flooding,  however, is  usually limited
to  short  term ponding.   In  most instances  this  is preferable to the
significant damage that may result from uncontrolled stormwater inflows.

     One type  of inlet control  method recently introduced into the United
States is a  device  known  as the Hydro-Brake.   This flow  regulator was
developed in Denmark and  is marketed in North America by Hydro Storm
Sewage Corporation.   The  Hydro-Brake  is  constructed of  non-corrosive
stainless steel, is self-regulating, contains no  moving parts, and  requires
no  power to operate.  The movement of water through the Hydro-Brake
involves a radial motion which dissipates  energy to control  the rate of
discharge  through  an  orifice.   As  the  hydraulic head on  the  unit
increases, the  radial   motion  or angular velocity increases,  thereby
decreasing the coefficient  of  discharge through the orifice.  At  lower
                              148

-------
heads, the angular velocity decreases, thereby increasing the coefficient
of discharge through the orifice.  In this way,  discharge  through the
orifice is kept relatively constant under  a varying range of heads.   A
schematic of a Hydro-Brake regulator is presented  in Figure 1.

     An  advantage of the Hydro-Brake unit, one that was investigated in
this study, is that for a  given sized orifice, an equally sized Hydro-Brake
(diameter  of Hydro-Brake  orifice  outlet  = diameter  of  orifice)  will
discharge less flow at identical  hydraulic heads.

PROJECT DESCRIPTION

     This demonstration  study consisted of two separate and independent
investigations.

     1.   The concept of off-line storage detention with controlled release
          using  a  Hydro-Brake  regulator.   (Off-Line  Storage  Tank
          Study)

     2.   The  concept  of utilizing increased surface/street ponding  of
          stormwater  to reduce  inflow to an  existing  combined sewer
          system.  (Catchbasin Study)

Each  concept offered a  viable  methodology to reduce the rate of inflow
into  the sewer system using flow restrictors.  Such  flow restrictors  or
regulators could  be  Hydro-Brakes,  or  specifically sized orifices  used  to
throttle flow.

     The demonstration  study  area is located on the west side of the City
of Rochester and  comprises 35  acres of primarily  single family residential
houses.   The entire  area is  served by a combined sewer system which is
tributary to  the West Side Trunk Sewer.  There are about 250 houses and
8 commercial establishments within the drainage area, many of which have
roof  leader  connections  to  the  combined  sewer  system.   Sixty-two
catchbasins  within the  area  are  presently connected  directly to  the
                                 149

-------
   OUTLET
   ORIFICE
(Typical 3.5 inch
  Diameter)
               DIRECTION
               OF FLOW
                                          INLET
                                                                  Typical  5 inch

               Figure  1.   Schematic of Hydro-Brake Regulator
                                         150

-------
combined  sewer  system.   A  schematic of  the  demonstration  site  is
presented in  Figure 2.

     A field survey was conducted in the Santee Drainage Area to acquire
necessary system  information.   The  information obtained  included sewer
sizes   and  slopes,   manhole  inverts,   the   number   and  location  of
catchbasins,  the number of house roof drain  connections to the sewer
system, and approximate street surface grades.

     The  Hydro-Brake  units required  for  this study were ordered  from
Hydro Storm Sewage Corporation in  December, 1980.   After  a  series of
unmet delivery dates, the Hydro-Brake units were finally received in late
April, 1981.  The Hydro-Brake designated for the off-line storage facility
was installed within  a week after delivery;  however, modifications to the
catchbasin Hydro-Brakes  were  required and installation of these units was
not completed until June, 1981.  These delays prevented the evaluation of
early spring  storms.

OFF-LINE STORAGE TANK STUDY

Background

     The  purpose of the off-line storage tank  study was  to evaluate the
effectiveness  of  off-line  storage  in  conjunction  with  Hydro-Brake
regulated  outflow,  on  the  reduction  of  downstream  sewer  surcharge
potential.

      Using  a reference storm  with  a return  frequency  of  2 years (60
minute duration, total rain  = 0.75 inches,  peak intensity = 1.5  in/hr), a
plot of cumulative runoff volume vs.  time  was  developed.  This  plot,
shown  in  Figure  3,  was  used  to determine  a proper Hydro-Brake
discharge  rate and  the capacity of the off-line storage tank.   A Hydro-
Brake discharge rate of 0.25  cfs at 50 inches of head,  and an off-line
storage tank capacity of 4700 gallons  were selected. The original design  of
the  off-line  storage tank  was  subsequently modified,   increasing the
capacity of the tank to  5700 gallons.

                                 151

-------
       en
             (Q
             c

             (D

             ro
 to
 o
 zr
 (D

 O)
 ct-
 »j.
 O

 O
 -h
             a.
             o
              i
             en
             3
o

o

ft-
3
ft-
o'

CO

n>
                 t
                              -N —







of
QC


K
! CURLEW S



i







i-
U)
z
i







&
z



•





j ,
1 
-------
       1.0  T
2 YEAR FREQUENCY  STORM
DURATION       =60 minutes
TOTAL RAIN      = 0.75 inches
PEAK INTENSITY  = 1.5  in./hr.
      0.8  •
  X
  I
  o
  0>
t «r
   E
   3
  O
      0.4  • •
      Q2  <•
                    10
            20
30
40
60
                                   TIME
          Figure  3.   2 Year Storm -  Cumulative Runoff  Volume vs.  Time
                          153

-------
     The off-line storage tank was  constructed using  sections of 54-inch
diameter reinforced  concrete  pipe.  The pipe  was  laid  end-to-end  to
create a tank 48 feet long.  A manhole was installed  at each end of the
tank to provide access to facilitate maintenance and monitoring.  The tank
was sloped at 0.008 to allow for drainage.

     Figure  4  presents  a  layout  of  the ;off-line  storage  facility area
including the number and location of catchbasins tributary to the storage
tank.   In addition,  the  piping route connecting these catchbasins to the
storage tank, and  the route of the outlet pipe from the storage tank  to
the combined sewer system at Villa and Santee Streets,  are  shown in the
figure.  Discharge  from  the  off-line  storage  tank  is controlled  by  a
Hydro-Brake (Standard  5-B-7)  regulator,  as shown  in Figure 5.  The
Hydro-Brake regulator has a 3.5-inch diameter outlet orifice.

     In order to restrict flow  in the outlet pipe from the existing storage
tank  until the  Hydro-Brake regulator was  installed, a 3-inch diameter
orifice, constructed  from  0.75-inch plywood, was  installed  at the tank
outlet.  Figure 6  presents an overall  plan  and profile  of  the off-line
storage facility.

Orifice Head-Discharge Tests

     Prior to the installation of the Hydro-Brake,  head-discharge tests
were conducted on  the  orifice in the off-line storage tank.  These tests
led to the development of a head-discharge curve for the 3-inch orifice on
the  outlet pipe from the storage tank.  The tests used  to develop  the
head-discharge curve, are discussed below.

Variable Head Test --

     The 6-inch diameter outlet  pipe  from the off-line storage tank was
temporarily  plugged  at the manhole at  Villa and  Santee  Streets.  Using a
nearby fire  hydrant, the storage tank  was filled.  The temporary plug in
the  6-inch  outlet pipe was then removed and the  level in the tank was
                                154

-------
             MICHIGAN
   Catch Basins
   New Piping Between
   Catch Basins and Off-
   line  Storage Tank
   Row
   Monitoring Location
  • Flow  Direction
      VILLA  STREET
 IB-—t CB*4
fCB'25

 CB'I
HYDRO-BRAKE
DEMONSTRATION  STUDY

 Off-Line
 Storage Control

 Scale:  |" . 50'
I-
UJ
LJ
cr
tn
   1*5
UJ
LJ
...J
          STREET
         BAUER  STREET
                             Fence
                                           •Off-Line  Storage  Tank
                                       RILEY  PARK
Figure 4.  Schematic of Off-line Storage Facilities Area

                        155

-------
Figure  5.  Hydro-Brake Regulator Installed in-Off-Line
           Storage Tank
                        -156

-------
Ol
            U3
            c
            fD
            CM
            22
            CU
            3
            Co
           O
           -t)
           fB
           O
           H,
           O
                                                                                                •N
J
24"Comb.
ti
3
SANTEE ST.
Exl«t CB U
T« 911.67 >»
Inv. '909.67
». B
il
36 1) Cornbiiwd Stwtr -
I
' . m cr==rrp\
Wtcharg'i Storag* Tank
— ;
IO
111
1
-£fc-
Exilt CB 2
/T = 31 1.66
/ Inv. = 909.64
„ Exilt CB
\y.B"CM.K^ T= 311
^TA.^^^ lnv=5
^f" 50" Comb.
.-1—4
Kn — 7 ^
^te C.M.R'
Exilt CB 9
T = 51 1.06
lnv.= 50637
MICHIGAN ST.
3
et
09.43


Exilt CB4
T 311.94
Inv. = S09.42
                                                                                        PLAN
           3
           ro
          rt-
          O
          O)
          IQ
          fD
          CD
          3
          n
                                         3*77
                                                     3*12
                                                                2t64
                                                                                                        Of-68 0)48  Ot20 CHOO
         ro
                                                                                   PROFILE

-------
monitored at one minute intervals.  The change in volume over each one
minute interval  (discharge) was plotted  against the average head on the
orifice during the  one minute interval.   The average head  here was
interpreted  as the average between head at the beginning of the interval
and  head  at the end of the  interval. A curvilinear  relation with  a
correlation coefficient equal to 0.94, was fitted to this data  and is shown
in Figure 7.

Static Head  Test —

     In this test, the depth in the off-line storage tank,  and therefore
the  head on  the  orifice,   was  held constant at  three different hydrant
discharge rates.   These three  points  were added to the plot of variable
head test data for  comparison.  The three observations fit well into the
expected  variance about the curve developed for  the variable head test,
and  therefore tended to confirm the results of that test.

Head-Discharge Curve —

     The head and  discharge data collected in the variable head test were
transformed by  taking their logarithms.  A linear regression function was
then fitted  to  the  plot  of  Log h  vs.  Log  Q;  where  h  = Head, Q =
Discharge.  The resulting  equation  was:

                    Log Q-= -0.6135+ 0.5394 Log h

By taking antilogs of this  equation, the following expression  is obtained:

                    Q = 0.2435 h °-5394

     The standard equation for discharge through an orifice is:

                    Q = CA  >/2gh~
                                 158

-------
    8
     E
     UJ
     
-------
     where         Q = Discharge
                   A = Area of orifice
                   g = Acceleration of  gravity
                   h = Head on orifice
                   C = Coefficient of discharge

     A comparison between the two equations, can be made.

     Test Data Equation (Q)  Standard Orifice  Equation  (Q)

        0.2435 h °-5394       =       CA/2gh

        0.2435 h °'5394       *       CA/2B  h0'5000

In the standard equation,  h is  raised to  the 0.5000 power.  The equation
obtained from the variable head test data shows  h raised  to the 0.5349
power,  consistent  (within experimental  error)  with  the  value  in the
standard  equation.  To solve for C, the coefficient of discharge, h  is
factored from  the equation.

                        0.2435 - CA>/2g~

For this test data, C * 0.62, which is consistent with most reported values
of  C - 0.60.   As a  result  of this observation, the  test  procedures
described above  were considered to be  appropriate in the evaluation  of
discharge from the off-line storage tank.

Hydro-Brake  Head-Discharge Tests

     Head-discharge tests were conducted on  a Hydro-Brake Standard  5-
B-7 unit installed in the off-line  storage  tank in the Santee Drainage
Area.   These  tests led to  the development of a head-discharge curve for
this Hydro-Brake. The tests used to develop this curve were the same  as
those  used  to develop the head-discharge curve for  the 3-inch orifice.
Data from these  tests  (variable head  and static  head)  are presented  in
                                 160

-------
Figure 8.  A curvilinear relation with a  correlation  coefficient equal to
0.80,  was fitted  to  the  variable  head test  data.  Again,  the three
observations  taken  in  the  static head test fit  well within  the expected
variance about this curve.

Head-Discharge  Curve —

     As for the orifice,  head and discharge data collected in the variable
head  test  were transformed  by taking logarithms.   A linear  regression
function  was then fitted to the plot of Log h vs. Log Q; where h = Head,
Q = Discharge.  The resulting equation was:

                    Log Q = -0.7897 + 0.3290 Log h

By taking antilogs, the following equation  was obtained:

                    Q = 0.1623 h °'3290

      In a general comparison with the orifice equation, in  which discharge
is  dependent on  head to the 0.5000 power,  it is seen that discharge from
the  Hydro-Brake depends  on  head  raised to  the  0.3290  power.  This
indicates  that compared to  an orifice,  the Hydro-Brake  will  tend  to
dampen discharges  associated with higher heads.  The curve, fit to the
variable head test data in Figure 8, shows this effect.

Hydro Storm  Head-Discharge Curve —

      Figure  9 shows the head-discharge relationship for a  Hydro-Brake
(Standard 5-B-7),with  a  3.5-inch  outlet  orifice,  as  defined   by  data
supplied by  Hydro Storm Sewage Corporation.   The curve developed from
the  Hydro-Brake testing at the off-line storage tank is  presented  for
comparison.

      Generally  the   relationship  provided  by  Hydro  Storm  slightly
underestimates discharge at given heads.  This is not considered to be a
                                  161

-------
ro
      in
      c
      (0
      CO
       n
       o>
      IQ
       o>
      -5
      (0
      Q.
      O
       I
      D3
      0)
HYDRO - BRAKE  TEST - SANTEE DRAINAGE AREA
                                               HEAD  (ft.)

-------
         <£}
          n>
cn
oo
          tu
          "S
         «£)
          (D

          O

          -S

          n>
          in
          o.

          o
          I
          TO

          0)
          2T
          ro

          a

          i
          o
          3
          t/J
          c+

          (U
         O
         3
         O
         n>
         o
HEAD - DISCHARGE CURV
- 	 	 	 	 	 - - 	 .-..._..-_ . .. - - -
---- 	 	 - - 	 • • — - 	

	 	 	 	 , . 	 , . 	 	 . . .
	 ._..,..,,... .. ._ 	 „._...,.
I":;:;:::::::::.: :. " ,..:_".::. ':..":;; :::.'.. .
	 - 	 	 - - 	 • - - - • •
	 _._ 	 	 	 J Y ' L , , 	 „ 	



... _ f) _ .. ._ .„,._.,
"_. << ^ x

•:.::.::::-. \$"~ r '.. ..:'. :. :::.':. \fl- '..:;.
1 C j /
. : : : . ~. . i .; ( ; . i . .; , . ^ " ' '."."":::.. t ^ 	 . ~ " .

- - - • ~ 	 K - - - 	 	 - - r - - • •• -
:::.:::::::: ^.;iji- :-:-i-i---i..:2-.L-::::-ii--:

.
'
f ' • - •• ^
::::*.:::::" *. * f ~:.~ ; -- ""..'„
~~~~ 	 7 " " 'n* " ""? 	 ^~?"7. "-" ..
	 . y .xx • ^x "
_..._.._,.„ ,. j -^ -• ^ .
.... ... . • ft . ' > - •
. .._,. . .. / / ^ ..._.
-------::--:---o— ^--^--- 	 ---- 	

^ES SANTEE HYDRO -BRAKE
•- • • • - 	 - 	 - - • • • — ......


	 	 ^ . -. _ 	 __._...._......_ 	 '..'
	 - - -- - 	 - 	 -
-- -• - - — .... - 	 	 -..._ .. .
.---•,< 	 - 	 	 -- - 	 	 -
-" - "• •" '• y<-'- '- • • '•'• •• • ""-- "..'.'. .: .: '.': . :.:'.':'.':.
~~.'~.fS.'\ '.' ": ~ ." : .,:.".":'." .. .: "~ .:
^ "*
f
,_..._... 	 	 ,,.....-, .... , . .. ,.,..-,__. _.-.„

- - 	 - 	 •- • - - - - : j - • -T 	



"*"-•• .... .. - - -. - .... „,..... - ..,,,-_.-,
."."-.'.'."' --". 	 " . - '. . . ". ." ,.."._'. .'.....
...... _..»».-:..... . 	 	 	
•..-*"-, ,-*' • •-• •• 	 • 	 - •
„-•="• - 	 --•• • • -• 	
'"" '"' " . ". ". . . "--•:----'2sIljf(rnh!i"S''ei1"".'r
	 - • o\ Miy 3tl >H -• -• • •


: .

                                                                                    HEAD  (ft.)

-------
serious  discrepancy  as differences  between  the  developed and  Hydro
Storm curves  can be attributed  to experimental and roundoff errors,  as
well as minor variations in the manufacturing process.

Storm Monitoring Results

     Fifteen  storms  were  monitored  during the period  from May 11  to
August  11,  1981  at the off-line storage tank in the Santee Drainage Area.
On  August 5th, additional discharge tests were conducted on the Hydro-
Brake regulator.  These tests resulted in the  development of a discharge
curve that  was  strikingly similar to a  discharge curve for a 3.5-inch
orifice.   Upon  further  investigation  at the  site,  a 15-inch long piece of
lath was  found  lodged inside  the  Hydro-Brake.   The  lath was  found
positioned in a manner  that would  have prevented  flow from developing a
radial motion  within  the  Hydro-Brake.  The  development of this  radial
motion is  necessary in order to dissipate energy  in the flow and  reduce
the coefficient of discharge through the Hydro-Brake's 3.5-inch  outlet
orifice.

     On  September 4th,  subsequent to the removal of the  piece of lath,
further  tests  were  conducted  on  the  Hydro-Brake  regulator.  The
discharge curve developed as a result of these tests  compared well with
the initial discharge curve, shown in Figure 8, developed in May.

     Four of  the fifteen  monitored  storms at the off-line  storage tank
occurred in August,  during the period of time that the Hydro-Brake was
determined  to  have been rendered ineffective due to the lodged piece of
lath.  Of the remaining eleven storms, five filled the storage  tank to at
least  20% of  its  maximum depth.   These  five storms  were considered
significant,  and the impacts of these  storms were analyzed in more  detail.

     On the evening of May 11, 1981 over  a one  hour-45 minute  period,
0.37  inches of  rainfall was  recorded at  the Rochester-Monroe  County
Airport.  Runoff from this storm was monitored  at the off-line  storage
tank.  Using the developed head-discharge curve  for  the Hydro-Brake in
                                  164

-------
the  storage  tank,  inflow   (uncontrolled)  and  outflow  (Hydro-Brake
controlled) hydrographs were developed.  These hydrographs are shown
in Figure 10.

     The inflow hydrograph  was developed using the equation:
     where         I  = inflow
                   E = outflow
                  As _ change in storage tank volume with respect
                  At "to change in time.

Values of outflow (E) were obtained from the Hydro-Brake head-discharge
curve  using  average values  of head on  the Hydro-Brake.  Values of -|-|
were developed from monitoring data at the storage tank.

     During  each  interval, average head was interpreted as the average
between  head at the beginning of the interval and head at the end of the
interval.
     The hydrographs presented in Figure 10 show that peak discharges
to the combined  sewer system  are  reduced by the Hydro-Brake installed
in  the  off-line storage  tank.   The  peak discharge rate of 1.038 cfs,
representing the peak uncontrolled discharge to the sewers, was reduced
by 75%.  The peak discharge rate through the Hydro-Brake was observed
to  be 0.260  cfs.  Total  volume discharged to the sewers in the first 50
minutes  of this storm (that portion of the storm where inflow to the tank
exceeded outflow) was reduced by 60%.

     Storms on June 4th, July 2nd, July 9th, and  July 20th were analyzed
in the same manner as discussed above.  The results of these analyses are
summarized in Table  1.  Figures 11-14 show the inflow (uncontrolled) and
outflow  (Hydro-Brake controlled) hydrographs  for each of these storms.
                                  165

-------
CTI
        (Q
         C
         to
         c+
         O
         O
         -b
         at
         CO
         I—*

         I

         co
         a
         3
         Ql
        in
         CD
         EU
                 C>~
                    IDO-
  1.00- •
                    .80"
UJ
g .60 +
O

a
                    .40 ••
                    .20 ••

                                                                KEY

                                                             Uncontrolled Discharge (Inflow)

                                                             Hydro - Brake
                                                             Controlled Discharge (Outflow)
                                20       40       60       80

                                                     TIME (min.)
                                                     1OO
120
140
160

-------
                                        TABLE 1.  STORM MONITORING RESULTS AT OFF-LINE STORAGE TANK
en
Date
5-11-81
6-04-81
7-02-81
7-09-81
7-20-81
Total Rain*
(Inches)
0.37
0.04
0.52
0.21
1.27
Peak Uncontrolled
Discharge (cfs)
1.038
0.331
0.996
0.536
0.327
Peak Hydro-Brake
Discharge (cfs) % Reduction
0.260 75.0
0.146
0.213
0.186
0.180
55.9
78.6
65.3
45.0
Total Uncontrolled
Discharge (gallons)
13,616
2,658
4,865
3,947
6,061
Total Hydro-Brake
Discharge (gallons)
12,716
2,366
5,296
3,937
5,990
Total Hydro-Brake Discharge ,«\
Total Uncontrolled Discharge *•*'
93.4
89.0
108.9
99.7
98.8
       *As recorded at the Rochester-Monroe County Airport.

-------
en
CO
          to
          c

          n>
           CO
           rt-
           o
          C-i
          c
          fD

          •P»
          VO
          CO
          C/i
          D>

          t^-
          n>
          n>
          tu
          tn
          n>
          n>
                     KEY


             Uncontrolled Discharge  (Inflow)


            Hydro - Brake Controlled Discharge

                                   '; (Outflow)
                                             10        20        30



                                                        TIME (min.)
40
50

-------
en
     to
      c
      -s
      ro
     »-•
     ro
      to
      r+
      O
      o
      -h
     vn
     CO
     CO
     O)
     3
     «-«•

     fD
     fD
     3
     _l.

     01
     tn
     ro
     ro
     03
til
O
cc
<

o
V)
Q
                                                                           KEY



                                                                    Uncontroled Discharge (Inflow) t



                                                                   Hydo - Brake ControHed Discharge (Outflow)
                                      10        20        3O        40

                                                            TIME    (min.)
                                                              50
60
70
80

-------
•vl
O
         (Q
         C
         CO
         c*
         o
         o
         -b
         vo
 vo
 CO
 CO
 Qi
 3
 c*
 n>
 n>

 o

 Qi

 3'
 Qi
to
 n>
        0)
        Qi
                       LOO
     .50- •





^   .4o4»
•
o
                 o   .30<
                 cc
                 o

                 S
                      .20
              .10
                                                                                        KEY
                                                                                   UncontroHed Discharge (Inflow),


                                                                                   Hydro - Brake Controlled Discharge (Outflow)
                          10
                            20        30        40        50

                                                TIME   (min.)
                                                           60
70
80

-------
Id
 c
 CO
 cH-
 O
 C-i
 c
 ro
 o
10
CO
CO
01
3
c^-
(D
n>

o

o>
_i.
3
O)
tjQ
0)
n>
cu
40-•
ai

£C

2
O

5
                                                                            KEY


                                                                       UncontroMed DJachage (Inflow)

                                                                                            '  /

                                                                       Hydro - Brake Controled Discharge (Outflow)
                              40        60        80

                                     TIME   (min.)
                                                                      100

-------
As in the storm on May  11, peak discharge  rates from these storms were
substantially  reduced  by  the  Hydro-Brake  regulator.   The  percent
reductions ranged from 78% on July 2, to 45% on July 20.

     Relatively little correlation can be seen between  total  rain and peak
discharges.  This is probably due, at least in part, to the 4-mile distance
between  the rain  gage at the Rochester-Monroe County Airport  and  the
Santee Drainage Area.

     Discrepancies between the total uncontrolled discharges and the total
Hydro-Brake discharges  are due primarily  to experimental and roundoff
errors.

Operation and Maintenance

     The maintenance required  at the  off-line  storage tank was minimal.
Over the four month demonstration period (May-August), only 2-3  inches
of fine debris built up  at the  lower end  of  the  tank.   Although these
depositions did not affect the normal operation of the Hydro-Brake unit, a
larger piece of stromwater debris did become lodged in the Hydro-Brake,
temporarily rendering the unit  ineffective.  Upon  removal  of  the debris,
operation of the Hydro-Brake returned to  normal.   Given the nature of
stormwater debris,  it may be  good practice  to check the Hydro-Brake
after every major  storm to ensure its normal  operation.

Installation Costs

     The following  represent the construction costs incurred installing
the off-line storage facility  described above:
                                  172

-------
                                       Costs Attributable   Costs Attributable
                         Total          to Storage Tank   to Collection  System
Labor                   $ 7,450             $4,800                $ 2,650
Storage Tank             2,320              2,320
Piping                    5,250                                    5,250
Equipment Rental          2,450              1,500                    950
 (Dump Truck,  Crane,
   Operator)
Materials                  2,065                500                  1,565
 (Gravel, Ready-Mix
   Concrete, Stone)
Restoration                 855                150                    705
 (Binder, Top Asphalt,
   Topsoil)
1 - Hydro-Brake Unit
  (5-B-7)                   525
                         $20,915             $9;795               $11,120 (ENR = 3510)
        The  costs of this alternative method to reduce the rate of inflow into
   an existing collection system, are comparable with costs  of  other off-line
   storage  alternatives  as  outlined  in  the 1978   Needs Survey:  Cost
   Methodology for Control  of  Combined  Sewer  Overflow  and  Stormwater
   Discharges  (NTIS PB-296 604).  The costs of this alternative, about $1.72
   per  gallon  of  storage,  compare favorably  with the $2.05  per  gallon of
   storage figure taken from the 1978 Needs Survey (ENR = 3510).
                                     173

-------
CATCHBASIN STUDY

Background

     The   purpose  of   the   catchbasin  study  was   to  evaluate  the
effectiveness of reduced inflow rates to the..combined sewer system on the
frequency  of  surcharge  conditions.   A  reduction  of  inflow  rates  was
accomplished by sealing a number  of  selected  catchbasins in the Santee
Drainage  Area.   Hydro-Brake  regulators  were  then  installed  in  the
remaining catchbasins in the drainage  area.  This combination allowed for
limited surface/street ponding of stormwater, thereby  reducing the rate
of runoff into the combined sewer system.

Homeowner Survey

     A complaint analysis was conducted to identify the existing flooding
problems  within  the  study  area.   Information was  obtained  through
distribution of  a homeowner survey, as shown in Figure 15, and review of
previous complaint records  filed with  Monroe County.  About 25 percent
of the total  number of surveys (241),  which  were hand  distributed to
homeowners in the  study area, were  returned.   Table  2 presents the
street location, number of responses,  and the number of responses which
indicated basement flooding problems.

         TABLE 2.  SANTEE  HOMEOWNER SURVEY  RESULTS*

                    No. of Responses             No. of Responses
     Street             Received               w/Basement Flooding
Michigan
Curtis
Emerson
Kestrel
Curlew
Santee
14
16
14
3
3
6
6
2
3
0
0
0

*Based on a 25% survey response.
                                  174

-------
                                                    FLOODING  SURVEY
     1)   Does your basement flood during rains?
                 D Yes         D No    --..

     2)   Have you previously filed a  complaint?
                 D Yes         D No    .

     3)   How many times per year does basement flooding occur?
     4)   How does your basement flooding frequency compare with some of your
          neighbors?
B                   about the same            (3 more than
                   less than

     5}   Is Street flooding a problem in your area?
                 n Yes         D No

     6}   Does basement flooding always occur along with Street floodina?
                 D Yes         D No

     7)   Does Street flooding ever encroach on your property?
                 D Yes         D No

     8)   Has your street ever been closed  due to flooding?
                 D Yes         D No

     9)   In your opinion, what location in your immediate area experiences the
          worst basement and/or street flooding.
     10)  Comments:
     Owner's Name:   	

     (address if different)

     Date:  	
     Building Address:
     Reach:  (Office Use Only)

     MH         to MH 	
Figure 15.   Homeowner  Survey  Questionnaire
                             175

-------
     A  review  of the  Monroe County complaint records indicated several
occurrences of basement flooding in the Santee Drainage  Area over the
last three years.  The complaint records, however, were so brief that  it
was  difficult to determine the impact that surcharging of the  combined
sewer system may have had on basement flooding.

Study  Description

     In order  to  determine  the extent of sewer surcharging  conditions
prior  to  the  sealing  and  installation  of  Hydro-Brakes  in  selected
catchbasins, sewer flows  in the study  area were  monitored.  Monitoring
was  accomplished  using  level sensing equipment at five locations within
the drainage area. Monitoring equipment used in this  study consisted  of
Manning Level Recorders (L-3000A), and  NB Electronic Manhole Meters
(GS).   The flow monitoring  locations are  listed below and are shown  in
Figure 16.

   - Emerson and Robin
   - Curtiss and Santee
   - Curttss -  2 MH west of Santee
   - Michigan - 2 MH west of Santee
   - Villa and Santee

     Storm  flows at these locations  were monitored in 1980, prior to the
sealing of any catchbasins or the installation of Hydro-Brakes.

     Preliminary stormwater modeling of the  sewer system in the Santee
Drainage Area  was  conducted  to help determine the  number and location of
catchbasins to be fitted  with  Hydro-Brake regulators.  A  field  survey of
street  surface grades was also conducted in order  to  locate  areas  of
natural depression that  would facilitate stormwater ponding.  The results
of this survey and the preliminary stormwater modeling indicated the need
for Hydro-Brakes at 11 catchbasin  locations.
                                   176

-------

 o
 O
 £1)
 0

 to
 CO
 01
 rt>



 CU
 —j,
 3
 0)
(O




 ro
 CU
Colchbosint   >»/   Hydro -Broke*    ©

     ii        w/  Seoled    X
     300'
                         200'
                                  400'

-------
     The flow monitoring conducted  in the Santee Drainage Area  during
1980 showed  frequent surcharging in the 12-inch combined  sewer along
Emerson  Street. To reduce the frequency and  magnitude of surcharging
in this sewer stormwater  inflow had to be restricted.  A reference storm
with a return frequency of 2 years (60 minute  duration, total rain = 0.75
inches,  peak intensity = 1.5 in/hr),  was  predicted to produce stormwater
inflow from roof leader connections on Emerson Street at an estimated rate
of  1.25  cfs.  When  combined with  the  normal dry  weather flow rate,
estimated  to  be 0.07 cfs, the remaining  capacity in  the Emerson Street
sewer was determined to  be 0.50 cfs. Inflow from each of the  two  Hydro-
Brakes on Emerson Street was therefore restricted to 0.25 cfs.  Hydraulic
computations for the remainder of the Santee Drainage  Area  showed that
an  inflow rate of 0.25 cfs through the remaining Hydro-Brake installations
was appropriate.   Since the  average  depth  from  street  surface  to
catchbasin invert was approximately  50 inches, and since it was desirable
to standardize the units selected, Hydro-Brake Standard 5-B-7 units with
a discharge of 0.25 cfs at 50 inches of head were  specified for each of the
11 catchbasin locations in the  Santee  Drainage Area.

     In the spring  of 1981,  the  necessary  catchbasins  were sealed  and
Hydro-Brakes were installed  in the 11  catchbasins.  Figure 16 shows the
location of the sealed catchbasins and the catchbasins with Hydro-Brakes
installed.   Early  into  the monitoring  program  in  1981, a  number of
residents in  the neighborhood,  inconvenienced by street ponding at the
ends of their driveways, opened  certain catchbasins in  order to prevent
ponding  in their areas.   These catchbasins are also shown in Figure 16.

     In 1981, subsequent to the sealing of catchbasins and the installation
of  Hydro-Brakes, flow  monitoring  was  resumed in the same locations used
during  1980.

Storm  Monitoring Results

     Information on existing system conditions was obtained by monitoring
depths  of storm flow,  at  the previously  indicated  locations,  from May
                                    178

-------
through  September 1980.  For each storm at each location, the maximum
depth of flow in the sewer was plotted against the  maximum 15-minute rain
intensity.

     In 1981 monitoring began in June, after the delivery and installation
of the Hydro-Brakes, and continued through mid-August.  As was done
in 1980, for each  storm the maximum  depth of flow in the  sewer was
plotted against the maximum 15-minute rain intensity.

     At  each   location,  1980  results  were  compared to those in  1981.
Figures  17-21  present  these  comparisons  for each  of the  monitoring
locations.  At Villa and Santee,  Curtiss, and Emerson and Robin (Figures
17-19), maximum flow depths obtained in 1981 for given rainfall intensities
were  generally lower than  those observed in  1980.  At Emerson and
Robin, a 12-inch sewer,  although sewer flows were generally reduced as a
result of the study,  surcharge conditions  during  most rainfall events still
persisted.

     At  Michigan  (Figure  20) and  Curtiss and  Santee (Figure  21)  no
improvement in 1981 data over 1980 data was seen.  These results may  be
attributed  primarily to the open  catchbasins, shown in Figure 16, near
these locations.

     It is  encouraging  to  note,  that  at  Villa and Santee (Figure 17),
downstream of each  of the other monitoring locations, flows in 1981 were
lower than those in  1980, Thus, although the study may not have  shown
consistent  improvements in  each  of  the  upstream sewer sections,  the
combined effect was a reduction in peak runoffs from the drainage area as
a whole.

Follow-Up  Homeowner Survey

     During  the last month of  the study, a follow-up  complaint  analysis
was conducted.  This  survey  concentrated  on those homeowners who
indicated basement flooding problems before the start of the study and on
                                   179

-------
   25-
   20
i
U-
O
en


a.
UJ
o
   15-
   10-
                      KEY

                       1980 Line of Best Fit

                       1980 Data

                       1981  Line of Best Fit
                       1981  Data
   A

 A
A A   A
                       A,-
-.-'   •
                 0.50
                       LOO

                     MAX  I|5( Inches)
             150
2.00
                                                                      2.50
 Figure  17.  Monitoring Results  -  Villa & Santee  Streets
                           180

-------
  13 T
  12 •
   II •
  10 •
   9-
I  84
F  6-
   5' •
   4-
   2.
    I-
                          AA
                        KEY
                	1980 Line of Best Fit
                  A  1980 Data
                —— 1981 Line of Best Rt
                  •   1981 Data
                 0.50
1.00           1.50
  MAX I|5 (Inches)
2.00
                                                                       Z5Q
 Figure   18.  Monitoring Results  -  Curtiss  Street
                            181

-------
   55 T
   50-
~ 45-


i
o
c

- 40-



1

"• 35J
o-  3CH
ui
Q
   25-
   20-
    IS
    10
    5,
                    KEY


               	I960 Line of Best Fit

                A   '980 Dota

               •      1981 Line of Best Fit

                •   1981 Data
                  050
1.00          L50

MAX I)5 (Inches)
2.00
2.50
Figure  19.  Monitoring Results -  Emerson &  Robin Streets
                            182

-------
    10
    9 •
S
1   8
w

1   e
LJ
o
X
<
    5 • •
    4 .
                x      A


''A
                                                    KEY
                                                  _ 1980 Line of Best Fit
                                               A   1980 Data
                                                    1981 Line of Best Fit
                                                •   1981 Data
t                                                    Indicates max depths of
                                                    flow greater than \Z"
                   0.50
LOO
MAX
                                         1.50
2.00
                                          (Inches)
2.50
 Figure  20.   Monitoring  Results  - Michigan  Street
                             183

-------
    25
~   20
o
c
O
u.
O


£
    15-
    10
                                               KEY


                                               Line of Best Rt
                                       A   1980 Data

                                         !•  1981 Line of Best Rt
                                       •   1981 Data
                  0.50          100          1.50

                            MAX  I(5  (Inches)
2OO
 Figure  21.  Monitoring Results  -  Curtiss &  Santee Streets
                        184

-------
those homeowners  closest to the  larger  street ponding  locations.  The
results of the survey were mixed,  with some howeowners noticing definite
improvements  and  others  indicating  no  change  in  the frequency  of
basement  flooding.  It  was  therefore  difficult  to  show  a  definite
correlation between the fevel of sewer  flow and the extent of basement
flooding.  The  inconsistent results  may  b,e  primarily  attributed to the
inability  to isolate stormwater  runoff  from  roof  leaders  connected  to
individual  house   laterals.  The  catchbasins  opened  by  neighborhood
residents  inconvenienced  by  street  ponding, may further explain the
inconsistent results obtained.

Operation and Maintenance

     In preparation for the monitoring  program in  1980,  each  catchbasin
in  the  Santee  Drainage  Area  was  cleaned.   Throughout  the  1980
monitoring  period,  only  minor  amounts  of debris  accumulated in  each
catchbasin.   In  the spring of 1981, prior to the sealing of catchbasins and
the  installation  of Hydro-Brakes, the  catchbasins were again  cleaned.
During  the 1981  monitoring  period,   only  minor  amounts  of debris
accumulated  in  the catchbasins.  This  accumulation  of debris did not
interfere  with  the normal operation of the Hydro-Brakes.  Larger pieces
of debris,  however, as at the off-line  storage tank,  may  render the
Hydro-Brakes ineffective.  In  order  to ensure the normal operation of the
Hydro-Brakes,  it  may  be good  practice to check each unit on a regular
basis, especially  following major  storms. The  catchbasins  themselves,
should be cleaned at least once every other year.

Installation  Costs

     The costs incurred  in developing  the inlet control system for the
catchbasin study described above are the following:

      Labor                                        $  765
      12 Hydro-Brake Units (5-B-7) @ $600/ea.      7,200
                                                  $7,965  (ENR = 3510)
                                  185

-------
     The  costs of  this  alternative,  about $230  per acre,  are very
favorable   when  compared  to the  conventional  alternative   -  sewer
separation,  at  approximately $42,000 per  acre,  as  estimated in the  1978
Needs Survey (ENR = 3510).

CSO REDUCTIONS

     By  attenuating  peak  stormwater  runoff,  sewer  surcharging  and
combined sewer overflow can be reduced.   Both methods demonstrated in
this study reduced the rate of inflow into the combined sewer system.
These  reduced  inflows  can  result  in  reducing  the frequency  and
magnitude  of surcharging in  the immediate localized tributary drainage
area,  and   in  the  reduced  frequency  and  volume  of  CSO  further
downstream in the combined sewer system.
                i
     At the off-line storage tank facility,  peak discharge rates and  total
volume  of  flow during peak  runoff  periods  were  reduced.   The  peak
discharge rate during the storm of May 11, 1981, was reduced by 75% and
the total volume of flow discharged  to  the  sewers during the peak runoff
period was  reduced  by 60%.  These  types of  reductions tend  to  reduce
overflow from the combined sewer system.  Since peak  stormwater flows,
especially  during  a  first-flush,  are  generally  associated  with  higher
concentrations of  pollutants,  any reduction in peak discharge rates and
resulting   overflows  may  show  significant Improvements  in  pollution
control.  Although  the  off-line storage demonstration was conducted over
a  small drainage area,  similar results would be expected using properly
sized facilities  with larger tributary drainage areas.

     The  mixed  results obtained as part of the  catchbasin  study and
confirmed in the follow-up homeowner  survey  can be attributable  in part
to  the indeterminate  and uncontrolled amount  of stormwater runoff from
roof-leaders connected  directly to the combined sewer system.  Due to the
resulting  surface ponding,  many homeowners did not perceive the Hydro-
Brakes to  be an effective solution  to their flooding problem.  In order to
change this perception  and to eventually gain acceptance of future Hydro-
                               186

-------
Brake  implementations,  it  may  be necessary  to  educate neighborhood
residents  on  the concept of inlet control through the  use  of Hydro-
Brakes.  Hydro-Brakes are able  to reduce sewer flow only at the expense
of increased  surface ponding.  The catchbasin study, in some cases, did
show  reduced levels of  sewer flow and  surcharging. It is felt  that this
will subsequently result in the reduced occurrence  of overflow  from the
combined sewer system.

SUMMARY AND CONCLUSIONS

     Two  different methods to reduce the rate of inflow  into the  combined
sewer  system were evaluated as a part of this  project.  The concept of
off-line storage  in conjunction  with  Hydro-Brake controlled  discharge,
serves to regulate downstream  sewer flows after  stormwater  runoff has
entered the  collection system.   In contrast,  the  surface/street ponding
concept with Hydro-Brake controlled inflows, serves to regulate the rate
of stormwater runoff before it enters the collection  system.

     Conclusions developed as a result of the off-line storage evaluations
are presented below:

 1.  The  Hydro-Brake regulator field performance  curve agreed  closely
     with the  performance curve  supplied by the  Hydro Storm  Sewage
     Corporation.  The field  unit indicated  flowrates  approximately 6
     percent higher than the design curve at 50 inches of hydraulic head.

 2.  Although  it performed reasonably well through a  number of storms,
     during  one  period  of time  the Hydro-Brake was rendered ineffective
     by debris present In normal stormwater runoff.

 3.  During  the period when the  Hydro-Brake was rendered ineffective
     by debris,  it failed to offer any improvement in flow restriction over
     a simple orifice of the same  diameter.
                                  187

-------
 4.   Compared  to  an equally sized orifice, the Hydro-Brake passed less
     flow at given hydraulic heads when operating as designed (i.e. with
     flow not impeded by debris).

 5.   The equation of discharge obtained for the Hydro-Brake Standard 5-
     B-7 regulator with a 3.5-inch diameter outlet orifice was as follows:

                   Q = 0.1623 h0'3290

     where         Q = Discharge in cfs
                   h = Hydraulic head in feet

 6.   Used in conjunction with off-line storage, the Hydro-Brake regulator
     effectively reduced peak stormwater runoff rates into the combined
     sewer system by as much as 78 percent.

 7.   Utilization  of  Hydro-Brakes to reduce  inflow  into combined sewer
     systems can result in a  reduction in the frequency and magnitude of
     combined  sewer  overflows.  In   localized  drainage   areas,  sewer
     surcharging  can  be  reduced at  the expense of additional  surface
     ponding.

     Conclusions  developed  as  a  result   of  the  catchbasin  control
evaluations are presented  below:

 1.   The   combination  of  surface/street  ponding   and  Hydro-Brake
     regulated  inflow  to  the combined sewer  system may not result in
     significant  impacts   in the  immediate  localized  tributary  area.
     However,  impacts may be felt further downstream as the combination
     did reduce peak runoffs from the drainage area as a whole.

 2.   In drainage areas where roof leaders are connected directly to house
     laterals,  the  effectiveness  of restricting inflow from street runoff
     through use  of catchbasin inlet control devices is reduced, since the
     fraction of runoff from roofs  is  not controllable by the catchbasin
     inlet control devices.

                                  188

-------
 3.   Even though the hydraulic grade  line of the collection system was
     reduced  by  means  of  inlet  control  devices,  the  perception of
     homowners in the demonstration  area failed to support  significant
     improvements.   The  existance of  surface ponding  and problems
     unrelated  to collection  system  conveyance  capacity  contributed to
     this perception.

     in general,  it should be noted that the implementation of the Hydro-
Brake  as a feasible static control device  is significantly  limited by the
present  capacity  of  the  supplier to  deliver  units  and  provide field
installation assistance.  (Supplier now claims that delivery problems have
been corrected.)

RECOMMENDATIONS

 1.   The design  of  the  Hydro-Brake   should  be  reviewed  to  develop
     modifications which  would minimize the impact of  stormwater debris
     on the optimal operation of the unit.

 2.   Use  of the  Hydro-Brake  should  be  evaluated  in conjunction with
     surface retention basins.  These results should be compared to those
     obtained  from   the  Hydro-Brake's   use  with  off-line  storage,
     particularly on a cost-benefit basis.

 3.   Further  studies  on  the effectiveness of  Hydro-Brakes in  reducing
     peak  stormwater  flows  and  the  frequency  of basement flooding,
     should be conducted by EPA at  other locations,  in  drainage areas
     where  roof leaders are not connected to house laterals.
                                189

-------