Document Sample
421 Powered By Docstoc
					             North American Society for Trenchless Technology (NASTT)
                                   NO-DIG 2005

                                        Orlando, Florida
                                        April 24-27, 2005

Evaluating Sewer Rehabilitation Effectiveness: A Practical
Methodology and Case Study
Paul J. Batman1, Jim Shelton1, PE
     Malcolm Pirnie, Inc., Wilmington, DE

ABSTRACT: While there are many rehabilitation technologies available to reduce inflow and infiltration
(I/I), identifying the best and most cost-effective strategy or combination of strategies can be a difficult
task. A number of sewer rehabilitation projects using various rehabilitation methods are currently being
monitored to evaluate the effectiveness of those technologies. The objective of this work is to quantify the
effectiveness of the rehabilitation methods being used to better predict flow reductions and select
rehabilitation methods for future projects.


There are many different rehabilitation technologies available to reduce infiltration and inflow (I/I), and it is
sometimes difficult to identify the most cost-effective method. Studies are currently underway in Delaware
and Maryland to evaluate the actual I/I removal effectiveness of sewer rehabilitation projects utilizing
various rehabilitation methods.
Technologies included in the study include the following:
      •    Cured-in-place lining of the mainline (CIPPL)
      •    Cured-in-place lateral lining with full-circumference mainline seal (CIPLL)
      •    Air pressure testing and packer-injection grouting of mainline joints (mainline test and seal; T&S)
      •    Air pressure testing and packer-injection grouting of the lateral-mainline connection (logiball; LB)
      •    Manhole frame and cover (F&C) replacement
      •    Manhole rehabilitation (MHR) utilizing combinations of chimney seal, fiberglass reinforced
           cementitious liner, and injection grouting of base joints, wall joints, or pipe penetrations
      •    Sump pump disconnection

These technologies are being used individually and in various combinations for a number of projects to
address specific project objectives. The primary objective of these projects is to eliminate wet weather
overflows and basement back-ups and to restore hydraulic capacity by reducing groundwater and rainfall

                                                Paper B-4-03 - 1
derived I/I entering the collection system. Secondary objectives are to restore structural integrity and
balance long term O&M needs and costs with capital reinvestment required for a multi-decade
refurbishment of the collection system.
Table 1 summarizes the projects which are completed or currently in design, pre-construction, or
construction, and ultimately included in this ongoing evaluation.

Table 1. Projects
 Project Name                                                 Quantity             Rehabilitation Technologies
     Brandywood Frame and Cover Replacements and 71                                F&C Replacements
       Sump Pump Removals (Completed)            11                                Sump pump disconnections
     Talleybrook Mainline Test and Seal (Completed)          6,000 lf              T&S
     NA-2 Mainline Lining                                    12,000 lf             CIPPL, LB
     NA-2 Lateral Lining                                     55 laterals           CIPLL
     NA-2 Test and Seal                                      4,000 lf              T&S, LB
     NA-2 Manhole Rehab                                      40                    MHR
     Northcrest Afton Rehabilitation                         8,000 lf              CIPPL, CIPLL, T&S, LB, MHR
     Fox Point, Dorval Road, et. al.                         15,000                CIPPL, CIPLL, T&S, LB, MHR
     Ashbourne Hills Rehabilitation                          7,000 lf              CIPPL, CIPLL, CIPLL, MHR
     Claymont Heights Rehabilitation                         7,000 lf              CIPPL, CIPLL, MHR
     SP-24 Mainline Lining                                   32,000 lf             CIPPL
     SP-24 Lateral Lining                                    484 laterals          CIPLL
     SP-24 Test and Seal                                     7,000 lf              T&S, LB
     SP-24 Manhole Rehab                                     50 manholes           MHR
     Hillcrest Orchards Mainline Lining                      11,000 lf             CIPPL
     Hillcrest Orchards Lateral Lining                       206 laterals          CIPLL
     Hillcrest Orchards Test and Seal                        11,000 lf             T&S, LB
        and Manhole Rehab                                    90 manholes           F&C Replacement

The objectives of the rehabilitation effectiveness evaluations of these projects are to:
      •    Confirm that completion of a rehabilitation project actually results in reduced I/I.
      •    Quantify the effectiveness of the rehabilitation methods selected to better predict flow reductions
           and increase effectiveness of the selection of rehabilitation methods for future projects.
This paper will focus on the method of data collection and rehab effectiveness analysis and will present
two completed projects (Brandywood Frame and Cover Replacements and Sump Pump Disconnections,
and Talleybrook Mainline Test and Seal) as case studies.

2.        METHOD

Data Collection
Rehabilitation effectiveness is evaluated using flow data collected with flow meters. Two types of flow
meter are used: radar-based (e.g. Marsh McBirney FloDars) and submerged sensor (e.g. Marsh
McBirney FloTotes). Submerged sensor flow meters require approximately one inch of water over the
sensor to accurately read flows. In larger basins where flows are high enough, this technology has been

                                                 Paper B-4-03 - 2
Placement of flow meters within a project area is critical. Flow meter locations are selected using the
following criteria:
    •      Size of the monitoring area – Between 5,000 and 15,000 linear feet of sewer main provides the
           most useful data resolution for flow metering.
    •      Accessibility – Meters must be easily accessible to download data and to perform routine
    •      Hydraulic conditions – Flow through the pipe should not be overly turbulent. Laminar conditions
           are best, and the pipe should ideally not experience surcharging.
Ideally, each project area is monitored for 3-6 months prior to the start of rehabilitation and for 3-6 months
after completion of rehabilitation in order to capture data from a significant number of storms. At least six
storms should be captured by the flow monitoring both before and after rehabilitation. The more data that
can be obtained, the more useful the effectiveness evaluation.

Data Validation and Truthing
Not all flow data are created equal. Flow data collected as part of this project are validated and truthed.
Confirming data are valid and true is critical before analyzing the data. If careful attention is not paid to
installation, truthing, routine maintenance, and data validation, the monitoring equipment can be
calibrated incorrectly, sensors can drift, and sensors can become fouled, any of which will result in
inaccurate flow measurements.
Validation is the process by which collected flow data are evaluated to determine whether velocity and
depth are correlating correctly with each other. Manning’s formula (Equation 1) is generally used to define
the relationship between depth and velocity "for unconstrained, uniform, steady-state, open channel flow”.
Q=        ( A × R 2 / 3 × S 1/ 2 )                                                                         [1]
In theory, measured velocity and depth should correlate according to Manning’s equation for open
channel flow. This relationship predicts that for each velocity there is a corresponding unique depth, and
vice versa. The validation process ensures that this relationship does not change over time, thus
confirming that the flow monitoring equipment is continuing to collect data reliably.
After validation, truthing is used to confirm that the flows calculated by the monitoring equipment are
accurate. Truthing can be accomplished by a number of methods, and the more truthing methods
employed, the greater the confidence that the flow readings are accurate. Truthing methods include:
    •      Volumetric weir readings – comparison of real-time flow readings with readings recorded by the
           flow meter.
    •      Flow balances – the sum of the results of individual flow meters throughout a monitoring area
           should equal the total flow exiting the monitoring area.
    •      Periodic verification of dry-weather flow – calculated as flow per inch-mile of pipe.
    •      Operational knowledge of the collection system (e.g., SSOs, sags in the downstream pipe,
           surcharging of the pipes and manholes, etc.).
Only valid and true data should be used.

Data analysis
The storm flow analysis in this evaluation uses flow data collected during and following a storm event,
referred to as the I/I period. The start of each I/I period is identified by the increase of the system flow
above baseflow, which should correspond to a measured rainfall event. Likewise, the end of the I/I period
is identified by the return to baseflow.

                                                Paper B-4-03 - 3
Two metrics for comparing and discussing I/I reduction are used in this evaluation:
    •                            Total system volume resulting from a rain event
    •                            Peak flow rate during a rain event
Total system volume resulting from a rain event is calculated by totaling the hourly flow volumes
measured during the I/I period. Peak flow rates are determined by reviewing the hourly data collected
during each rain event and identifying the highest measured flow rate.
There are two methods used for conducting the analysis:
    •                            Control Basin Method
    •                            Pre/Post Comparison Method
The Control Basin Method (CBM) of analysis is a correlation between the metrics of the basin undergoing
rehabilitation and the “simultaneous” metrics from a control basin. Scatter plots are generated with the
metric values from the control basin on the x-axis and the corresponding metric values from the
rehabilitation basin on the y-axis. Pre-rehabilitation data is plotted separately from post-rehabilitation data
and both sets are linearly regressed. The percentage difference between the slope from the pre-rehab
regression and the slope from the post-rehab regression yields the percentage reduction due to
rehabilitation. Figure 1 illustrates a typical scatter plot for the CBM using the total volume metric.

                                                Linear (Post-Rehab)
                                                Linear (Pre-Rehab)
                                                                                                    y = 1.1949x
    Volume - Rehab Basin (MG)


                                                                                                                  y = 0.4746x



                                      0.0       1.0              2.0              3.0           4.0                5.0          6.0
                                                                      Volume - Control Basin (MG)

Figure 1. Example Scatter Plot by the CBM for Total Volume Metric

                                                                           Paper B-4-03 - 4
If the control basin was properly selected (i.e. it exhibits similar physical condition, I/I characteristics,
groundwater and rainfall conditions, and is geographically close to the rehabilitation basin) the
relationship between the two basins is linear because it is a direct comparison of metrics which occurred
during the same storm event.
The Pre/Post Comparison Method (PPCM) of analysis compares the post-rehabilitation flow data
collected from the rehabilitated basin with pre-rehabilitation flow data which was also collected from the
same basin. Scatter plots are generated with a given measure of rainfall (e.g. 1-hour rainfall, or total
rainfall volume) on the x-axis and the metric (total volume or peak flow rate) on the y-axis. As in the CBM,
the pre-rehab data are plotted as well as the post-rehab data. The data are regressed (either linearly or
logarithmically, depending on the data set) and the effect of rehabilitation determined by comparing the
slopes, as above. This method is more complex because of the comparison of storms with different
antecedent conditions (pre- and post). Figure 2 illustrates a typical scatter plot for the PPCM using the
peak flow metric, which has been treated as logarithmic.

                       3.0        Log. (Pre-Rehab)
                                                                                     y = 1.0385Ln(x) + 2.1473
                                  Log. (Post-Rehab)

     Peak Flow (MGD)



                                                                                            y = 0.3982Ln(x) + 1.0161


                          0.0''   0.25''       0.5''   0.75''     1.0''     1.25''       1.5''        1.75''       2.0''

                                                        1-Hour Rainfall (Inches)

Figure 2. Example Scatter Plot by PPCM for Peak Flow Metric


Below are two completed case study projects for which the above analysis methods were applied.

Case 1: Frame and Cover Replacement and Sump Pump Removal
In Case 1, a basin was targeted for manhole frame and cover replacements to reduce inflow resulting
from sheet flow entering the manholes and for sump pump disconnections to reduce clearwater
connections to the sanitary sewer which contribute large amounts of water to the peak flows.

                                                            Paper B-4-03 - 5
The basin contains approximately 34,000 lf of 8” pipe, 190 manholes, and 600 homes. Many of the
existing manhole covers had 4 large (1 ½” diam.) vent holes in the covers and large gaps between the
covers and the rimless frames. New frames and covers were installed which are built with gasketed rim
seals and single smaller (< 1” diam) vent holes. Approximately 71 frames and covers were replaced at a
total cost of approximately $90,000. About 3-5 months after the frame and cover replacements, 11 of 41
cross connected sump pumps in the basin were disconnected from the sanitary system by the utility as
part of a pilot sump pump disconnection program. The total cost for these disconnections was
approximately $20,000.
Flow data was collected from the rehabilitation basin for approximately 7 months before the start of the
frame and cover replacements, and flow data from the control basin was collected for 4 months before
the work. Data was collected for 8 months following the sump pump disconnections.
Using the Control Basin Method, there were six significant pre-rehabilitation rain events and twelve post-
rehabilitation rain events of significant size for the analysis. Small events which yielded no apparent
increase in system flows were excluded from the analysis. Storms used are generally greater than a 0.1”
1-hour rainfall, although some events with low 1-hour rainfalls which lasted several days were included in
the analysis. Two rain events with 1-hour rainfalls slightly over 0.1” were also removed from the analysis
due to extraordinarily high flow measurements given the amount of rainfall measured. Using the CBM,
Figure 3 illustrates the regression for total system volume. A 49% decrease in total volume following
rehabilitation was achieved. Likewise, Figure 4 illustrates the regression for peak flow. A 36% reduction in
peak flow was achieved.
Using the Pre/Post Comparison Method, Figure 5 illustrates the scatter plot and regression for total
system volume. Based on this analysis an 80% decrease at high rainfall levels following rehabilitation was
achieved. Figure 6 illustrates the regression for peak flow which yielded a 60% reduction.

                                            Post-Sump Pumps and F&Cs
                                            Linear (Pre-Rehab)
                                            Linear (Post-Sump Pumps and
    Volume - Rehab Basin (MG)

                                                                                                           y = 1.0705x


                                                                                                                y = 0.5486x


                                      0.0      0.5               1.0         1.5         2.0         2.5         3.0          3.5
                                                                       Volume - Control Basin (MG)

Figure 3. Case 1 Total System Volume Reduction by Control Basin Method

                                                                            Paper B-4-03 - 6


                                                     Post-Sump Pumps and F&Cs

                                                     Linear (Pre-Rehab)
   Peak Flow - Rehab Basin (MGD)

                                                     Linear (Post-Sump Pumps and F&Cs)
                                                                                                                             y = 0.5395x

                                                                                                                           y = 0.4132x


                                         0.0             1.0                 2.0               3.0           4.0                    5.0            6.0

                                                                             Peak Flow - Control Basin (MGD)

Figure 4. Case 1 Peak Flow Reduction by the Control Basin Method

                                                       Log. (Pre-Rehab)
                                                       Log. (Post-Rehab)
  Volume (MG)




                                             0.0''        1.0''            2.0''       3.0''         4.0''         5.0''                   6.0''   7.0''

                                                                                    24-Hour Rainfall (Inches)

Figure 5. Case 1 Total System Volume Reduction by Pre/Post Comparison Method

                                                                                      Paper B-4-03 - 7
                               Log. (Pre-Rehab)
                               Log. (Post-Rehab)

  Peak Flow (MGD)



                       0.0''           0.5''       1.0''             1.5''    2.0''         2.5''

                                                   1-Hour Rainfall (Inches)

Figure 6. Case 1 Peak Flow Reduction by Pre/Post Comparison Method

Case 2: Packer-Injection-Grouting
A basin containing 6,000 lf of 8-inch pipe was mainline test and sealed at a cost of $40,000 to reduce
infiltration and rainfall derived I/I through leaking pipe joints.
The Pre/Post Comparison Method was not used for this analysis because the duration of the
rehabilitation was prolonged over several months. The pre-rehabilitation data was collected during high-
groundwater conditions in the spring but the post-rehabilitation data was collected in the summer when
groundwater levels were at a low. For this analysis to be useful, pre and post- rehabilitation data should
be collected either within a short period of time or over an extended period (two or more years). Data from
the following spring might be used for post-rehabilitation, however groundwater levels from one year to
the next may vary widely, rendering the analysis qualitative.
Data from 5 suitable pre-rehabilitation storm events and 6 post-rehabilitation storms were used for the
analysis by the Control Basin Method. As mentioned above, rain events with greater than 0.1” 1-hour
rainfalls were used for the analysis. The analysis showed a 16% reduction in total system volume and a
13% reduction in peak flow. Figures 7 and 8 illustrate the respective scatter plots and linear regressions.

                                                    Paper B-4-03 - 8
                                                                   Linear (Post-Rehab)
  Total Volume - Rehab Basin (MG)

                                                                   Linear (Pre-Rehab)
                                                                                                                                      y = 0.2957x

                                                                                                                                                  y = 0.2495x


                                                         0.0        0.5           1.0            1.5          2.0         2.5             3.0            3.5        4.0

                                                                                          Total Volume - Control Basin (MG)

Figure 7. Case 2 Total System Volume by Control Basin Method

                                                                    Linear (Post-Rehab)
                                                                    Linear (Pre-Rehab)
                       Peak Flow - Rehab Basin (MGD)

                                                                                                                                                y = 0.2297x

                                                                                                                                                      y = 0.2002x


                                                             0.0     0.5        1.0        1.5          2.0         2.5         3.0             3.5       4.0       4.5

                                                                                           Peak Flow - Control Basin (MGD)

Figure 7. Case 2 Peak Flow by Control Basin Method

                                                                                                       Paper B-4-03 - 9

Manhole frame and cover replacements combined with sump pump disconnections were successful in
reducing total system volume by 50-80% and peak flow by 35-60%. Mainline test and seal was found to
reduce total system volume by only 16% and peak flow by only 13%. To achieve better reductions, future
work on this area will include packer-injection grouting of the lateral connections and pipe penetrations at
the manholes. Observation has indicated that both are issues in this project area and this work is planned
for 2005. Rehabilitation effectiveness evaluation following this additional work should show a marked
Over 15 contracts utilizing a variety of technologies and scheduled for release in 2005/2006 will ultimately
be included in this rehab effectiveness evaluation. If rehabilitation effectiveness evaluation continues
during these projects, there will be a substantial amount of data to continue quantifying technology-
specific rehabilitation effectiveness. Coupled with the severe levels of I/I in these project areas, there
should be some interesting conclusions brought to light over the next three years.

                                            Paper B-4-03 - 10

Shared By: