FR SJN135025 Clean1

Performance Comparison of Abutment and
Retaining Wall Drainage Systems
Prepared by:
Junliang Tao
Junhong Li
Sichuan Huang
Robert Liang
Ayse Ozdogan-Dolcek
William Likos
Prepared for:
The Ohio Department of Transportation,
Office of Statewide Planning & Research
State Job Number 135025
November 2017
Final Report

Technical Report Documentation Page
1. Report No. 2. Government Accession No. 3. Recipient's Catalog No.
FHWA/OH-2017-36
4. Title and Subtitle 5. Report Date (Month and Year)

November 2017
Performance Comparison of Abutment and Retaining Wall Drainage
Systems 6. Performing Organization Code
7. Author(s) 8. Performing Organization Report No.

Junliang Tao, ORCiD: 0000-0002-3772-3099
Junhong Li, Sichuan Huang, Robert Liang, Ayse Ozdogan-Dolcek,
William Likos
9. Performing Organization Name and Address 10. Work Unit No. (TRAIS)
The University of Akron

302 Buchtel Common
Akron, Ohio 44325-2102 11. Contract or Grant No.
State Job Number 135025
12. Sponsoring Agency Name and Address 13. Type of Report and Period Covered
Final Report
Ohio Department of Transportation
Research Section
1980 West Broad St., MS 3280 14. Sponsoring Agency Code
Columbus, OH 43223
15. Supplementary Notes

Project performed in cooperation with the Ohio Department of Transportation and the Federal Highway
Administration.
16. Abstract
Control of water infiltration and providing adequate drainage are critical to the performance of retaining walls and abutment
walls. Current Ohio Department of Transportation (ODOT) practice for drainage of structures specifies the use of a two-foot
porous backfill with filter fabric, which has a long performance history. ODOT is seeking alternative drainage systems that are
more cost- and time-effective, durable, and at the same time, have comparable or superior drainage capability compared to
current practice. A prefabricated composite drainage system (PCDS) is proposed in this research as an alternative structure
drainage system. The current state of practice of drainage systems for retaining wall and bridge abutment structures is evaluated
through a survey of county agencies in Ohio. Commercially available PCDS products as well as specifications for PCDS used
by other state DOTs are also reviewed and synthesized. Laboratory testing and evaluation of select PCDS products are conducted
to improve the understanding of their properties. It is found that tested values for some properties of some products do not match
those listed by the manufacturers. Recommendations on the selection of proper standard testing methods and suggestions on the
selection of factory of safety in design are discussed. Field performance of the PCDS system and the traditional drainage system
are evaluated with in situ instruments including piezometers, tiltmeters and flumes. The data analysis suggests that the PCDS
has comparable drainage capability to the traditional system. Field observation and feedback from the contractor reveal that
installation of PCDS systems are less labor-intensive and more time-effective. Cost analysis from ODOT historical bidding data
and the actual cost at the tested sites demonstrates that a PCDS system costs 40% less than the traditional process. Based on the
findings, draft specifications were developed to specify the material and construction requirements for a PCDS system.
17. Key Words 18. Distribution Statement
No restrictions. This document is available to the
Prefabricated Composite Drainage System, Retaining Wall, Abutment, Cost public through the National Technical Information
Service, Springfield, Virginia 22161
22.
19. Security Classif. (of this report) 20. Security Classif. (of this page) 21. No. of Pages Price
Unclassified Unclassified 180
Form DOT F 1700.7 (8-72) Reproduction of completed pages authorized
i

Performance Comparison of Abutment and
Retaining Wall Drainage Systems
Prepared by:
Junliang Tao, Ph.D.

Principle Investigator
Junhong Li
Sichuan Huang
Robert Liang, Ph.D., P.E.

co-invesigator
Department of Civil Engineering

The University of Akron
244 Sumner Street
Akron, Ohio 44325-3905
Ayse Ozdogan-Dolcek
and
William Likos, Ph.D.
Subcontractors at
University of Wisconsin, Madison
November 2017
Prepared in cooperation with the Ohio Department of Transportation

and the U.S. Department of Transportation, Federal Highway Administration
The contents of this report reflect the views of the author(s) who is (are) responsible for the facts and the accuracy
of the data presented herein. The contents do not necessarily reflect the official views or policies of the Ohio
Department of Transportation or the Federal Highway Administration. This report does not constitute a standard,
specification, or regulation.
ii

ACKNOWLEDGMENTS
This project was conducted in cooperation with the Ohio Department of Transportation (ODOT)
and the Federal Highway Administration.
The authors would like to acknowledge the support of ODOT’s Technical Liaison Committee:
• Mr. Jim Welter, Construction Structures Engineer, ODOT Office of Construction
Administration
• Mr. Sean Meddles, Assistant Administrator, ODOT Office of Structural Engineering
• Mr. Mohammad Tariq, Structure Engineer, ODOT Office of Structure Engineering
The authors appreciate the time and input generously provided by the members of the Technical
Liaison Committee.
Gratitude are also extended to the site engineers for their time and help during the installation of
the field instruments:
• Mr. Andrew Griesdorn, Construction Project Engineer, ODOT District 6
• Mr. Dave Polling, Construction Project Engineer, ODOT District 6
• Mr. Brian Gable, Construction Project Engineer, ODOT District 6
• Mr. Matthew Rotar, Transportation Engineer, ODOT District 6
• Mr. Jeremy Gilbert, Project Manager, Complete General Construction
The authors also would like thank Ms. Kelly Nye from ODOT’s Office of Statewide Planning
and Research for her time and assistance. Special thanks also go to the technical writer Sheila
Pearson in the Department of Civil Engineering at University of Akron for her help in
proofreading and improving the language of the final report. In addition, help and assistance
from students Ms. Hui Tao and Mr. Ganesh Pandey, research technician Mr. William Wenzel at
The University of Akron, as well as the instrumentation technologist Xiaodong Wang, are
appreciated.
iii

EXECUTIVE SUMMARY
Control of water infiltration and providing adequate drainage are critical to the performance of
retaining walls and abutment walls. Current Ohio Department of Transportation (ODOT) practice
for drainage of structures (ODOT Construction and Material Specifications Item 518) specifies the
use of a two-foot porous backfill with filter fabric, which has a long performance history. ODOT
is seeking alternative drainage systems that are more cost- and time-effective, durable, and at the
same time, have comparable or superior drainage capability compared to current practice. A
prefabricated composite drainage system (PCDS) is proposed as an alternative drainage method
for bridge abutments and retaining walls.
These drainage systems include ODOT’s current method of using two feet of porous backfill with
filter fabric, the prefabricated composite drainage system, and other promising drainage systems
such as ground tire chips. The specific objectives are to:
1) Develop specifications for the PCDS;
2) Select environmentally sound materials and procedures that are compliant with EPA
rules;
3) Quantify labor and materials savings of using an alternative abutment and retaining
wall drainage system, including the prefabricated composite drainage system; and
4) Select the best system for drainage on abutments and retaining walls.
To achieve the goals and objectives, the following tasks were taken:
• Task 1: State-wide survey among Ohio counties and contractors on the current state of the
practice for the structural drainage
• Task 2: Comprehensive literature review on other promising alternative drainage systems
used in other state DOTs, emphasizing commercially available PCDS products, DOT
specifications, and the potential for using tire chips as backfill materials
• Task 3: Laboratory testing and evaluation of select PCDS products to improve the
understanding of the various properties
• Task 4: Field evaluation and comparison of the performance of a drainage system using
two feet porous backfill and PCDS
• Task 5: Cost analysis and comparison of the two drainage systems
• Task 6: Development of material and construction specifications for the PCDS
The findings from the above tasks are summarized in this executive summary.
I. Current state of the practice for structure drainage in Ohio
• Based on the survey results, it can be concluded that porous backfill systems are the
predominant drainage systems used for retaining walls and bridge abutments in Ohio.
Alternative drainage systems used in Ohio include low-strength mortar fill and PCDS. The
owners (counties) are generally satisfied with the performance of the traditional drainage
system as well as the alternative systems. But the respondents also generally advocate the
iv

usage of PCDS if it can bring benefits such as lower costs, labor savings and a simpler
construction process.
II. Current state of the practice for PCDS
• An online search was conducted to collect information on commercially available

prefabricated drainage systems. A total of 17 manufacturers and 87 composite drainage
systems were identified, and their properties were summarized. The selection of PCDS
products should be based on design requirements, cost and availability.
• The design methodology for PCDS was developed by referring to existing research
findings, design guidelines by transportation agencies such as FHWA, AASHTO and state
DOTs: the required drainage capacity should be determined considering wall height and
in-situ soil properties. The ultimate transmissibility of the PCDS should be determined by
considering a factor of safety as well as various reduction factors. The design core strength
should be determined considering the abutment/wall type and height, in-situ soil properties
as well as allowable lateral movement at the top of the abutment.
• The hydraulic properties of the filter fabric should be determined considering the
permittivity, retention and clogging resistance criteria based on the properties of the
backfill and in-situ soils. The strength properties of the filter fabric should consider the
type and importance of the structures, survivability and the constructability.
• Through a survey on state DOT specifications, it is found that at least 21 state DOTs are
using PCDS products. Of these, 16 state DOTs have specifications on the material
properties of PCDS cores, five state DOTs provide design guidance of PCDS systems, ten
state DOTs specify the construction procedures, and seven state DOTs provide lists of
approved products.
III. Laboratory evaluation of select PCDS products
Five PCDS products from different vendors were acquired for index property testing in the
laboratory. Eight properties of the core or fabric were tested according to the corresponding ASTM
standards.
• The eight properties and the standards selected for testing include,
1) In-plane flow at various gradients and pressures of the core, according to ASTM
D-4716.
2) Compressive strength of the core, according to ASTM D1621.
3) Thickness of the core and the fabric, according to ASTM D1777 or ASTM D5199.
4) Permittivity of the fabric, according to ASTM D4491.
5) Grab tensile strength and elongation of the fabric, according to ASTM D4632.
6) Puncture strength of the fabric, according to ASTM D4833.
7) Trapezoidal tear strength of the fabric, according to ASTM D4533.
8) Apparent opening size and equivalent sieve of the fabric, according to ASTM
D4751.
Some of the methods were later found not to be ideal for PCDS, and recommendations of
replacement standards were made.
• Regarding the in-plane flow of the core, it was observed that:
v

o The presence of the fabric tends to reduce the in-plane flow rate significantly. In
this tested case, the reduction is about 42%.
o There were no significant differences between the flow rates for the seating periods
of 0.25 hours and 100 hours under the confining pressure levels in this study.
o Flow direction (machine direction or cross-machine direction) has a moderate
effect on the in-plane flow of the core.
o In-plane flow increases nearly linearly with the hydraulic gradient but does not
change much with the tested normal stress levels, which are not sufficient to
significantly deform the core.
o Compared to values listed by the manufacturers, the tested in-plane flows fall short
for all products when the fabric is bonded. When tested using the core only, the
resulting flow exceeded the listed value.
• Regarding the compressive strength, the tested values for the composite deviated from the
manufacturer values for the core alone. The differences varied from –25.5% to +12.6%.
For four of the five tested products, the tested values were lower than the manufacturer
values. It was later realized that this may be due to the testing method and conditions used.
It is recommended that ASTM D6364 “Standard Test Method for Determining Short-Term
Compression Behavior of Geosynthetics” be used to test the compressive strength of PCDS,
in lieu of ASTM D1621.
• Regarding the thickness of the core or composite, the tested average composite thicknesses
under two different levels of normal stress were very close to each other. For the separate
core, three samples were received, and the tested thicknesses were slightly less than the
manufacturer values. The thickness of a core is an indirect indicator of the drainage
capacity.
• Regarding the strength properties of the fabric, only one product showed significantly
lower values than the listed values among the three products tested. It was found that
different manufacturers test the puncture strength of the fabric using different ASTM
standards. ASTM D6241 is actually more suitable than ASTM D4833 for geotextile
materials.
• Regarding the hydraulic properties of the fabric, all nonwoven products showed smaller
tested permittivity than the listed values. Since permittivity and flow rate of a geotextile
product is correlated and the permittivity is a value normalized by the differential head
across the fabric, it is also recommended to use permittivity, instead of the flow rate, in the
specification. For every tested product, the tested apparent opening size (AOS) was
significantly larger than the listed value.
• The differences between the tested and listed values suggest that a factor of safety should
be used in the design when selecting a PCDS.
IV. Field implementation and evaluation
Based on the project schedules, locations, abutment/wall types, and other site conditions, two
bridges in Franklin County (FRA-270-1714 and FRA-270-1716) were selected for the field
monitoring and evaluation of the installation process and field performance of two drainage
systems: a traditional two-foot porous backfill with filter fabric and a prefabricated composite
drainage system. The PCDS at FRA-270-1714 was designed based on site conditions. Construction
plan notes were also developed to guide the installation of the PCDS. The forward abutments and
vi

wing walls at these two bridges were instrumented with piezometers in the backfill, tiltmeters on
the backwall, and flow flumes with water level sensors at the drainage outlets. A third site, ATB-
20-21.43 in Ashtabula County, was selected for additional observations of the placement of a two-
foot porous backfill. Major findings include:
• The number of workers required for installation of the two systems is similar. Both require
two or three workers.
• The installation of the traditional two-foot porous backfill system is cumbersome, and the
system can only be installed in stages during the backfilling process. This results in a longer
overall backfilling process. Further, the construction equipment (e.g., backhoe) is prone to
damage the filter fabric during construction, and it is difficult to control the thickness of
the drainage layer.
• The installation of the PCDS is a more straightforward process. It involves unrolling,
placing and securing the composite drainage board onto the back side of the abutment/wall
in horizontal or vertical courses. The joining of the different courses is of critical
importance, and caution should be exercised to properly seal the edges.
• Feedback from the contractors suggested that the installation of the PCDS was much easier
and faster.
• Piezometers were installed to measure the groundwater pressure in the backfill with time.
If a drainage system works properly, excess pore water pressure will be effectively
dissipated and the pressure will be be kept at very low levels at all times. At both FRA-270-
1714 and FRA-270-1716, the soil water pressure generally varied between –3 kPa and 3
kPa, which was relatively low, indicating effective drainage. Soil water pressure changes
with precipitation condition; it increases gradually after precipitation and decreases
relatively fast after precipitation stops. The actual drainage paths, though, are complex, as
indicated by the relative water pressure levels among sensors embedded at different
locations.
• Tiltmeters were installed to monitor the lateral movement at the top of the abutment. This
information can be used to predict the soil pressure status behind the wall (at rest, active or
passive). This information also helps evaluate the health condition of the wall itself. At
FRA-270-1714, the tilt angle changed between –0.025° and 0.075°, indicating that the top
of the abutment back wall moved between –0.16 inch to 0.47 inch, which was much smaller
than the specified 1.50 inches. The tilt angle of the abutment at FRA-270-1716 decreased
suddenly on June 7th, 2017 in response to construction activities; otherwise, the tilt angle
varied between –0.040° and 0.070°. The slight changes of tilting correlated very well to
the temperature changes, while no correlation between pore water pressure change and tilt
angle was observed.
• Two flumes were used to directly measure the flow rate of the water coming from the
collector pipe. No reliable quantitative flow rate data was obtained due to freeze-thaw
cycles at sites and the malfunction of the sensors due to fire accidents and manufacturer
flaws. Nevertheless, visual observations of the flumes during multiple site visits indicated
that water was drained out from the abutments, evidenced by the sediment trace left by the
draining water. Flowing water was observed at FRA-270-1716 on certain dates, but no
flowing was observed at FRA-270-1714, where the PCDS was installed. This might be due
to the fact that the outlet location at FRA-270-1714 was very far (>300 ft) from the
abutment and the drainage pipes were installed at a very flat grade (0.33%), while the outlet
vii

location was set very close (<50 ft) and the drainage pipes were installed at a steeper grade.
It was also realized that direct comparison of quantitative flow rate data at the two bridge
abutments should not be used to assess the performance of the two drainage systems, since
the two abutments had different geometry designs, backfill conditions (coverage area,
volume and final grades), and drainage pipe lengths and grades.
• It was noted that the quantity of sediments left on the flume at FRA-270-1714 was higher
than that at FRA-270-1716. A site reconnaissance revealed that some portions of the top
edges of the composite drainage board were not sealed properly, which may have caused
soil intrusion. This underscores the importance of quality control during the installation of
PCDS.
V. Cost analysis and comparison
• Based on a state-wide survey and ODOT historical bidding data, the typical unit total price
for porous backfill with filter fabric is $55 to $80 per cubic yard, or $4/SF to $5.9/SF of
wall. The material cost typically accounts for about 50% of the total cost.
• Based on limited ODOT historical bidding data, the unit total price for PCDS varied widely:
between $2/SF and $18/SF. The unit total price decreases with the increase of the plan
quantity. For a relatively large plan quantity (>1,000 SF), the cost for PCDS is more
competitive than the porous backfill system.
• The unit price of PCDS products ranges from $0.8/SF to $1.7/SF. The price varies from
manufacturer to manufacturer. For products from the same manufacturer, the price
typically increases based on the quality or the compressive strength of the product.
• Direct comparison between the cost of drainage systems at FRA-270-1714 and FRA-270-
1716 showed that:
o For the porous backfill system, the unit material cost was $2.4/SF of wall. The unit
labor and equipment cost was $2.2/SF of wall.
o For the PCDS, the unit material cost was $1.05/SF. The exact construction time
was not monitored, but feedback from the contractors suggested that the
construction of the PCDS was much faster and easier. A reasonable estimate of the
time savings is 25%, resulting in a unit labor and equipment cost of $1.65/SF.
o For FRA-270-1714, the contractor and ODOT decided to place a layer of granular
material, using ODOT CMS (Construction & Material Specifications) Item 703.17
for Item 304 between the drainage board and the native soil backfill. Considering
this replacement layer of granular material, the upper bound of the unit total price
for the PCDS was $4.44/SF, or a 3.5% cost savings compared to the traditional
system.
o Note that the additional layer of granular material behind the drainage board is only
optional. Under normal site conditions where the backfill material is well
compactable, this additional layer is not required, resulting in a unit total price for
the PCDS of $2.7/SF, or a 41% savings compared to the traditional system.
o The cost savings for using PCDS without a replacement granular layer at FRA-270-
1714 is consistent with the historical bidding record (47%) and findings from other
researchers (35%).
viii

VI. Development of the draft specification
Based on the proposed design methodology, recommendations are made on the required properties
of the PCDS. Draft specifications are developed for the PCDS by referring to the specifications of
other state DOTs, specifications provided by the PCDS manufacturers, and the findings from this
research.
• The recommended required PCDS properties depend on abutment/wall type, height and
backfill and in-situ soil properties. Tables listing requirements for separate core and fabric
were developed.
• Two draft specifications for PCDS were developed, referring to the ODOT CMS format.
One specification discusses material requirements, including the following components:
o 712.XX.01 Description
o 712.XX.02 Materials
o 712.XX.03 Prequalification/basis of approval
The other specification discusses mainly the construction processes, including the
following components:
o XXX.01 Description
o XXX.02 Fabrication
o XXX.03 Materials
o XXX.04 General
o XXX.05 Prefabricated Composite Drainage System
o XXX.06 Pipe
o XXX.07 Scuppers
o XXX.08 Excavation
o XXX.09 Method of Measurement
o XXX.10 Basis of Payment
Note that symbols “XX” or “XXX” are used to indicate the newly proposed item numbers. Such
temporary symbols are also used in the main text and appendices of this report. These symbols
should be replaced to numerical digits after approved by ODOT and before incorporated into the
existing ODOT Construction and Material Specifications.
VII. Recommendations
Based on the findings from this research, the following recommendations are made:
• Continue to monitor the performance of the drainage systems at FRA-270-1714 and FRA-
270-1716 for a longer period of time, so as to evaluate their long-term performance.
• Incorporate the developed draft specifications for PCDS into the ODOT CMS. There are
two options to specify PCDS in ODOT CMS: one is to develop a new designated Item
XXX, similar to the existing CMS Item 518, and the other is to incorporate PCDS into
CMS Item 518, “Drainage of Structures.”
ix

Table of Contents
EXECUTIVE SUMMARY ........................................................................................................... iv
List of Tables ............................................................................................................................... xiii
List of Figures .............................................................................................................................. xiv
CHAPTER 1. INTRODUCTION ................................................................................................... 1
1.1. Problem Statement ............................................................................................................... 1
1.2. Goals and Objectives of the Study ....................................................................................... 2
1.3. Research Context ................................................................................................................. 2
1.3.1. Importance of abutment and retaining walls and proper drainage systems .................. 2
1.3.2. Criteria for a proper drainage system............................................................................ 3
1.3.3. Overview of drainage systems ...................................................................................... 3
1.3.4. Specific requirements for an efficient prefabricated composite ................................... 6
1.4. Organization of the Report................................................................................................... 8
CHAPTER 2. CURRENT ODOT PRACTICE FOR STRUCTURAL DRAINAGE .................... 9
2.1 Introduction ........................................................................................................................... 9
2.2. Modifications to CMS 518 .................................................................................................. 9
2.3. Survey of County Engineers and Contractors .................................................................... 10
2.3.1 Types of drainage system............................................................................................. 11
2.3.2 Cost of drainage systems ............................................................................................. 12
2.3.3 Performance of drainage systems ................................................................................ 13
2.3.4 Implementation of CMS 518 ....................................................................................... 14
2.3.5 Prospect of adopting PCDS ......................................................................................... 15
2.3.6 Survey of contractors ................................................................................................... 15
2.4. Summary ............................................................................................................................ 15
CHAPTER 3. ALTERNATIVE DRAINAGE SYSTEMS........................................................... 16
3.1. State-of-the-Practice of PCDS as Structural Drains .......................................................... 16
3.1.1 Commercial products ................................................................................................... 16
3.1.2 Design methodology .................................................................................................... 17
3.1.3 DOT survey.................................................................................................................. 20
3.1.4. Synthesis of the specifications on PCDS .................................................................... 24
3.2. Tire Derived Aggregates as Backfill and Drainage Layer ................................................. 24
3.3. Summary ............................................................................................................................ 27
x

Chapter 4. Laboratory Testing of Drainage Composites ............................................................. 29
4.1. Laboratory testing program and methods .......................................................................... 29
4.2. Summary of testing results................................................................................................. 30
4.2.1. In-plane flow ............................................................................................................... 36
4.2.2. Compression strength of the composite ...................................................................... 37
4.2.3 Thickness of the composite, core and fabric ................................................................ 38
4.2.4 Grab tensile strength and elongation of the fabric ....................................................... 39
4.2.5 Trapezoidal Tear Strength............................................................................................ 39
4.2.5 Puncture Strength ......................................................................................................... 40
4.2.6 Permittivity and flow rate ............................................................................................ 41
4.3 Summary and conclusions .................................................................................................. 42
CHAPTER 5. SITE IMPLEMENTATION AND PERFORMANCE EVALUATION .............. 44
5.1. Site Selection ..................................................................................................................... 44
5.2. Development of the plan notes for the testing sites ........................................................... 45
5.2.1 PCDS design ................................................................................................................ 45
5.2.2 Recommendations of PCDS products.......................................................................... 49
5.2.3. Installation plan for the PCDS system ........................................................................ 50
5.3. Installation of the drainage systems ................................................................................... 52
5.3.1. Installation of the PCDS ............................................................................................. 52
5.3.2. Installation of the traditional drainage layer ............................................................... 52
5.4. Planning and installation of the instruments ...................................................................... 54
5.4.1. Overview ..................................................................................................................... 54
5.4.2 Instrumentation layout and installation........................................................................ 57
5.5. Data collection and analysis............................................................................................... 61
5.5.1 Additional Weather Data ............................................................................................. 64
5.5.2 Soil water pressure data ............................................................................................... 65
5.5.4 Flume flow level and flow rate data ............................................................................ 74
5.6. Summary and conclusions ................................................................................................. 77
CHAPTER 6. COST ANALYSIS ................................................................................................ 79
6.1 Unit material cost of commercially available PCDS .......................................................... 79
6.2 Cost analysis based on survey and bidding history ............................................................ 79
6.3. Cost analysis on the drainage systems at FRA-270-1714 and FRA-270-1716 ................. 82
6.3.1 Abutment geometries ................................................................................................... 83
6.3.2 Material cost................................................................................................................. 83
xi

6.3.3 Labor and equipment cost ............................................................................................ 83
6.3.4 Unit total cost ............................................................................................................... 84
6.4. Summary and conclusions ................................................................................................. 85
CHAPTER 7. SPECIFICATION FOR PREFABRICATED COMPOSITE DRAINAGE
SYSTEM ....................................................................................................................................... 87
7.1. Recommendations .............................................................................................................. 87
7.1.1 Recommended specification for the core structure of PCDS ...................................... 87
7.1.2. Recommended specification for the geotextile fabrics of the PCDS .......................... 88
7.1.3. Recommended specifications for the construction of PCDS systems ........................ 89
7.2. Draft specification.............................................................................................................. 90
7.2.1. Draft Item 712.XX Prefabricated Composite Drainage Boards ................................. 91
7.2.2. Draft ITEM XXX Prefabricated Composite Drainage Systems for Structures .......... 94
CHAPTER 8. SUMMARY, CONCLUSIONS AND RECOMMENDATIONS ........................ 98
8.1. Current state of the practice for structure drainage in Ohio............................................... 99
8.2 Current state of the practice for PCDS ............................................................................... 99
8.3 Laboratory evaluation of select PCDS products ................................................................. 99
8.4 Field implementation and evaluation ................................................................................ 100
8.5 Cost analysis and comparison ........................................................................................... 102
8.6 Development of the draft specification ............................................................................. 103
8.7 Recommendations ............................................................................................................. 103
REFERENCES ........................................................................................................................... 104
Appendix A. Survey of Drainage System for Bridge Abutment and Retaining Walls in Ohio . 108
Appendix B. Commercially Available PCDS Products.............................................................. 112
Appendix C. Synthesis on specifications for PCDS systems ..................................................... 121
Appendix D. Data Source for Cost Analysis .............................................................................. 127
Appendix E. Flow Net and Coefficients of Earth Pressure ........................................................ 133
Appendix F Installation plan as shown in the marked plan sheets ............................................. 137
Appendix G Detailed Laboratory Procedures and Results ......................................................... 144
xii

List of Tables
Table 1. Recommended drainage reduction factors (Koerner and Koerner, 2007) ...................... 18
Table 2. DOT Specifications for PCDS ........................................................................................ 21
Table 3. State DOT specifications on the core structure of PCDS .............................................. 22
Table 4. State DOT specifications on the geotextile fabric of PCDS ........................................... 23
Table 5. Tested products and corresponding material types for separate core and fabric ............ 29
Table 6. Laboratory test results for Product A .............................................................................. 31
Table 7. Laboratory test results for Product B .............................................................................. 32
Table 8. Laboratory test results for Product C .............................................................................. 33
Table 9. Laboratory test results for Product D .............................................................................. 34
Table 10. Laboratory test results for Product E ............................................................................ 35
Table 11. Effect of bonded fabric on the in-plane flow rate of Product A ................................... 36
Table 12. Effects of seating period on in-plane flow rate ............................................................. 37
Table 13. Comparison of tested average compressive strength to the manufacturer value .......... 38
Table 14. Thickness of the composite and separate fabric and core ............................................. 39
Table 15. Comparison of tested tensile strength and elongation to the manufacturer value ........ 39
Table 16. Comparison of tested tear strength to the manufacturer value ..................................... 40
Table 17. Comparison of tested/estimated puncture strength to the manufacturer value ............. 41
Table 18. Comparison of permittivity and flow rate to the manufacturer value........................... 41
Table 19. Comparison of AOS to the manufacturer value............................................................ 42
Table 20. Soil characterization of the native backfill soil ............................................................ 46
Table 21. Gradation specified for Type C granular material (Item 703.16.C). ............................ 46
Table 22. Required hydraulic properties of the prefabricated composite drainage ...................... 49
Table 23. AASHTO M288 Class 2 Specification ......................................................................... 49
Table 24. Comparison listed properties for Product A with design values .................................. 50
Table 25. Site visit log .................................................................................................................. 62
Table 26. Summary of data availability and reliability................................................................. 64
Table 27. Unit Price of Commercially Available PCDS products................................................ 80
Table 28. Cost comparison of different drainage systems ............................................................ 85
Table 29. Assumptions for Calculation of the Lateral Earth Pressure .......................................... 87
Table 30. Core Strength Requirements for Walls with Different Heights .................................... 88
Table 31. Required PCDS In-Plane Flow Rate for Walls with Different Heights ....................... 88
Table 32. Recommended requirements for geotextile of the PCDS ............................................. 89
Table B-1. Properties of the core of PCDS products .................................................................. 112
Table B-2 Properties of the fabric of PCDS products................................................................. 116
Table E-1. Approximate Values of Relative Movements Required to Reach Active or Passive
Earth Pressure Conditions (Clough and Duncan, 1991) ............................................................. 134
xiii

List of Figures
Figure 1. (a) Drainage method Type 1: porous backfill around drainage pipe (used in Iowa;
White et al., 2005). (b) Drainage method Type 2: porous backfill with filter fabrics (used in
Ohio). .............................................................................................................................................. 4
Figure 2. Drainage method Type 3: prefabricated geocomposite drainage system (used in
Missouri; Missouri DOT, 2013) ..................................................................................................... 4
Figure 3. Illustration of prefabricated composite drains: (a) Sheet drain. (2) Combination drain . 6
Figure 4. Reported use of various types of drainage system behind abutment and retaining walls
by 34 county engineers in Ohio. ................................................................................................... 11
Figure 5. Percentages for different drainage systems used in projects in 34 responding counties.
....................................................................................................................................................... 11
Figure 6. Cost of porous backfill with or without filter fabric...................................................... 12
Figure 7. Averaged weight of cost components for different drainage systems. .......................... 12
Figure 8. Performance of porous backfill with or without filter fabric. ....................................... 13
Figure 9. Reported issues for porous backfill with filter fabric. ................................................... 13
Figure 10. Maintenance work conducted for wall drainage systems. ........................................... 14
Figure 11. Averaged challenge rating in implementing CMS 518. .............................................. 14
Figure 12. Benefits typically leading to the adoption of PCDS.................................................... 15
Figure 13. Location of the selected sites for field testing. ............................................................ 45
Figure 14. Particle size distribution of the natural soil on site...................................................... 46
Figure 15. Construction of the drainage systems. (a) The installation of the PCDS at the forward
abutment of FRA-270-1714. (b) The installation of the collector pipe underneath the PCDS.
Note that the drainage pipe was wrapped with fabric and embedded in porous aggregates. ....... 52
Figure 16. (a) The layout for the trench and fabric. (b) The installation (dumping) process. (c)
An installation that was not well-controlled. ................................................................................ 54
Figure 17. Geokon Model 4500S standard piezometer. ............................................................... 55
Figure 18. Geokon Model 6350 Vibrating Wire Tiltmeter. .......................................................... 55
Figure 19. (a) The purchased H flume. (b) Modified flume with approach section and connector.
....................................................................................................................................................... 56
Figure 20. The layout of the piezometers and tiltmeters. (a) Elevation view. (b) Plan view at
FRA-270-1714. (c) Plan view at FRA-270-1716. ........................................................................ 57
Figure 21 Installation of the monitoring system. (a) Installation of the piezometers at Elevation 1
at abutment FRA-270-1714. Note that the piezometers and the cables (protected with conduits)
were embedded in sand prior to backfilling. (b) Installed tiltmeter. ............................................. 58
Figure 22. The position of flumes installed at the drainage outlets. (a) Plan sheet details for the
forward abutment at FRA-270-1716. (b) Landscape plan showing the final locations of the
flumes............................................................................................................................................ 60
Figure 23. Flumes installed at (a) FRA-270-1714 and (b) FRA-270-1716. ................................. 61
Figure 24. Piezometer data at FRA-270-1714. (a) Water pressure and precipitation. (b)
Temperature .................................................................................................................................. 66
Figure 25 Piezometer data at FRA-270-1716. (a) Water pressure and precipitation. (b)
Temperature. ................................................................................................................................. 67
xiv

Figure 26. Piezometer data for a short time period (a) Water pressure and precipitation at FRA-
270-1714. (b) Water pressure and precipitation at FRA-270-1716. 5.5.3 Tiltmeter data .......... 69
Figure 27. Tilt angle and temperature data from the tiltmeters (a) FRA-270-1714 and (b) FRA-
270-1716. ...................................................................................................................................... 71
Figure 28. Tilt angle and temperature data from the tiltmeters for a shorter period of time: (a)
FRA-270-1714 and (b) FRA-270-1716. ....................................................................................... 72
Figure 29. Tilt angle and temperature data from tiltmeters at FRA-270-1716 after June 7th. ..... 73
Figure 30. Flume water levels in the flume at abutments: (a) FRA-270-1714 and (b) FRA-270-
1716............................................................................................................................................... 74
Figure 31. Observation of water draining from the flumes on January 17, 2017: (a) FRA-270-
1716 and (b) FRA-270-1714......................................................................................................... 75
Figure 32. Observation of the flumes on June 18, 2017: (a) FRA-270-1716 and (b) FRA-270-
171. (c) improper termination of the drainage board edges. ......................................................... 76
Figure 33. Histogram of bidding average unit price of porous backfill with filter fabric from
ODOT bidding history in 2013. .................................................................................................... 81
Figure 34. Comparison of unit price between PCDS and porous backfill with filter fabric. The
data from this study is also included. ............................................................................................ 81
Figure E-1. Raw flow nets behind a typical bridge abutment..................................................... 133
Figure E-2. Relationship between wall movement and earth pressure. (After Clough and Duncan
1991) ........................................................................................................................................... 135
Figure E-4. Relationship between wall movement and earth pressure for compacted soil. (After
Clough and Duncan 1991).flat .................................................................................................... 136
xv

CHAPTER 1. INTRODUCTION
1.1. Problem Statement
retaining walls and abutment walls. Accumulation of water behind earth retaining structures will
result in extremely high hydrostatic pressure and frost pressure, which will exert a tremendous
amount of additional force on the walls. Inadequate or improper drainage of water could also lead
to soil collapse and erosion near the structures, which will result in instability of the structure or
unfavorable settlement. All of these factors could lead to structural damage to the retaining walls,
abutment walls, approach slabs, and expansion joints. The potential settlement under approach
slabs could also affect the ride quality of the roadway and even cause unsafe driving conditions.
Therefore, an efficient drainage system is vital for maintaining safety in abutments and retaining
walls.
Current Ohio Department of Transportation (ODOT) practice for drainage of structures (ODOT
Construction and Material Specifications Item 518) specifies the use of a two-foot porous backfill
with filter fabric, which has a long performance history. In general, the porous drainage system
must perform the following two functions: 1) to relieve the hydrostatic pressure by allowing water
to flow freely and to continue doing so throughout the life of the structure, and 2) to prevent
settlement by retaining the soil backfill and preventing the migration of soil particles. Overall, the
performance of the conventional porous backfill drainage system has been good except for the fact
that it typically requires a high-quality porous material, skilled workers, and considerable time for
installation. ODOT is thus seeking alternative drainage systems that are more cost- and time-
effective, durable, and at the same time, have comparable or superior drainage capability compared
to current practice.
A prefabricated composite drainage system (PCDS) is proposed as an alternative drainage method

for abutments and retaining walls. A PCDS, which consists of a drainage core wrapped with a
geotextile filter, must perform the same functions as a porous drainage system: the filter retains
the soil particles and allows water to flow into the drainage core, which collects the water and
directs it to the outside of the structure.
In recent years, several projects in Ohio have used PCDS behind some retaining walls, and
contractors have requested to substitute the PCDS for the conventional two feet of porous backfill
with filter fabric behind abutment walls in order to achieve cost savings. However, ODOT lacks
quantitative comparison data on the cost-effectiveness, constructability, and the drainage
performance of the two systems (i.e., PCDS and porous drainage system). A specification or design
guideline and a qualified products list for prefabricated composite drainage systems are also not
currently available. These deficiencies in pertinent information hinder the potential for wider
application of PCDS in abutment and retaining wall design and construction in the state of Ohio.
1

1.2. Goals and Objectives of the Study
The goal of this study is to compare the performance of different drainage systems for abutment
and retaining walls and to select the best system. These drainage systems include ODOT’s current
method of using two feet of porous backfill with filter fabric, the prefabricated composite drainage
system, and other promising drainage systems such as ground tire chips. The specific objectives
are to:
1) Develop specifications for prefabricated composite drainage systems.
2) Select environmentally sound materials and procedures that are compliant with EPA rules.
3) Quantify labor and materials savings of using an alternative abutment and retaining wall
drainage system, including the prefabricated composite drainage system.
Upon completion of the study, the best drainage system for abutment and retaining walls will be
identified, and the corresponding specifications will be developed to guide the design and
construction of the selected drainage system.
1.3. Research Context
1.3.1. Importance of abutment and retaining walls and proper drainage systems
Abutment walls and retaining walls are simple structures, and their importance is often overlooked.
ODOT maintains one of the largest transportation systems in the nation, and bridges and retaining
walls are critical components in Ohio’s surface transportation system. Ohio has more than 43,410
bridges and more than 49,250 lane miles of interstate, US and state route roadways. Earth retaining
structures play a significant role in the safety and efficiency of the bridge and highway system,
and a large percentage of the new bridges in Ohio have semi-integral or full-integral abutments.
The application of retaining walls has also expanded greatly during recent years due to an increase
in urban development, transportation capacity demands, and construction in complex and difficult
terrain. Although failures of such earth retaining structures seldom result in loss of life, safety
remains a major concern, since these retaining structures support bridges and other transportation
features immediately adjacent to bridges or highways. Damage to abutment and retaining walls or
excessive settlement of backfill materials will affect the ride quality of the roads, disrupt traffic,
and incur large road repair costs.
A major reason for failures of abutment walls and retaining walls is related to drainage. Most
retaining wall failures occur after heavy rainfall. The lack of a proper drainage system behind
abutment and retaining walls will permit the accumulation of water behind the wall, thus increasing
the hydrostatic pressures, reducing soil strength, and causing soil erosion. These conditions can
lead to damage of the abutment and retaining walls and can cause damage or settlement at locations
adjacent to these walls, such as at the approach slabs and expansion and contraction joints. To
2

prevent such damage, it is necessary to drain away whatever water accumulates behind the
abutment and retaining walls using proper drainage systems.
1.3.2. Criteria for a proper drainage system
A proper drainage system should perform the following two functions (Berg, 1993. Holtz et al.,
1998):
1) Allow water to flow through, and to continue doing so throughout the life of the project.
2) Retain the soil particles and prevent their migration through the drainage system (if some
soil particles do migrate, they should not cause clogging of the downstream media during
the life of the project).
Accordingly, there are three criteria for the design of a proper drainage system to perform these
functions: the retention (or piping resistance) criterion, permeability criterion, and clogging
resistance criterion.
1.3.3. Overview of drainage systems
White et al. (2005) summarized the various drainage methods implemented by transportation
agencies in different states. Based on their review, drainage methods can be categorized into three
different types: (1) porous backfill around a perforated drainage pipe (Figure 1), (2) porous backfill
with geotextile fabric filters (Figure 2), and (3) a prefabricated geocomposite drainage system
along the wall face (Figure 3). Each of these drainage systems are briefly introduced in the
following paragraphs.
The method of porous backfill with a drainage pipe is shown in Figure 1a. This method has the
longest performance history and is the most widely used. It typically requires high-quality graded
porous materials in order to meet the retention, permeability and clogging resistance criteria.
According to ODOT specifications, porous backfill consists of gravel, stone, or air-cooled blast
furnace slag, with a No. 57 size gradation according to AASHTO M43. The disadvantages of using
this method are typically related to quality assurance and quality control during construction. For
example, White et al. (2005) inspected eight new bridges under construction in Iowa and found
that porous backfill was not used around the drainage pipe at most bridge sites, even though Iowa
DOT specified the use of porous backfill (as shown in Figure 1). They also found that insufficient
quality control measures had resulted in plugging of soil in the drainage pipes during and after the
construction.
3

(a) (b)
Figure 1. (a) Drainage method Type 1: porous backfill around drainage pipe (used in Iowa; White et al.,
2005). (b) Drainage method Type 2: porous backfill with filter fabrics (used in Ohio).
The method of porous backfill with filter fabric is illustrated in Figure 1b. The introduction of
proper filter fabric (geotextile) contributes to better retention and clogging resistance of the
drainage system and thus improves its long-term performance. This method has a relatively short
history of usage, and performance data are limited to approximately 45 years. This method is the
only drainage system currently specified by ODOT. A drainage system consisting of two feet of
porous backfill with filter fabric is shown in this drawing. ODOT’s Bridge Design Manual (BDM)
Item 303.2.3 (ODOT, 2007) specifies the design details of abutment drainage:
“The porous backfill immediately behind abutments and retaining walls should be
provided according to CMS 518. The porous backfill shall be effectively drained by
the use of a corrosion resistant pipe system into which water can percolate. …
Porous backfill shall be wrapped with filter fabric, CMS 712.09, Type A.”
Details on the filter coverage of the backfill, drainage pipes and outlets are also provided in BDM
303.2.3. ODOT’s Construction and Material Specifications (CMS) Item 518 (ODOT, 2013)
provides specifications for the construction of the drainage system, including the requirements for
the materials (i.e., the porous backfill, filter fabric, and pipes).
Figure 2. Drainage method Type 3: prefabricated geocomposite drainage system

(used in Missouri; Missouri DOT, 2013)
4

The method of prefabricated composite drainage system (PCDS) is illustrated in Figure 3. This
system typically consists of a drainage core wrapped with geotextiles. The application of PCDS as
a drainage system for abutment and retaining walls has a short history. However, it has become
increasingly accepted as an alternative for structure drainage due to its comparable performance,
lower cost, consistent properties, and ease of placement. A PCDS should be properly designed to
meet all three drainage criteria (retention, permeability, and clogging resistance). At the same time,
the geocomposite must have the strength and durability to survive construction and maintain
performance for the design life of the drainage system (JDAA, 1995). A number of state
transportation agencies have developed specifications for PCDS design and construction,
including those in California, Colorado, Indiana, Louisiana, Missouri, Nebraska, and New York.
In addition to these three drainage system types, drainage systems using a layer of tire chips
have also been proposed in a previous research program (White et al., 2005). The layer of tire chips
has a high permeability and an elastic nature, which helps to alleviate the potential for lateral
displacements of the bridge. A scaled bridge approach drainage model was developed by White
et al. (2005) to evaluate the performance of 13 abutments with different drainage systems and
backfill materials. The drainage system with the highest drainage capacity was the method of
tire chips (maximum flow rate: ~550 cm3/sec. time for water to drain: 1 sec), followed by
the PCDS method (220~380 cm3/sec. 1~4 sec), the method of porous backfill with filter fabric
(60~80 cm3/sec. 7~10 sec), and lastly by the method of porous backfill with drainage pipe
(30 cm3/sec. 10~12 sec).
Based on previous practices and the results of related research, especially the comprehensive
model tests conducted by White et al. (2005), it seems that much better drainage performance can
be gained if a prefabricated composite or a layer of tire chips is used as the drainage system behind
abutment and retaining walls in lieu of a porous backfill with filters, which is the current practice
specified by ODOT.
Reuse of ground tire chips in a drainage system is considered as a sustainable approach. However,
it should be noted that: (1) ground tire chips usually have similar size distribution characteristics
as porous backfill. While it can provide high drainage capacity, clogging might be a problem if no
filter layer is utilized. The combination of ground tire chips with fabrics might be a solution.
(2) While the elastic nature of the tire chips help alleviate the potential for lateral displacements,
the deformation and compaction of the chips might reduce the void ratio and, in turn, reduce the
drainage capacity in the long run. Therefore, the suitability of using tire chips as drainage systems
should be evaluated further.
5

1.3.4. Specific requirements for an efficient prefabricated composite
Core High Profile Core
Low Profile Sheet Core
Filter Fabric
Pipe Fittings
(a) (b)
Figure 3. Illustration of prefabricated composite drains: (a) Sheet drain. (2) Combination drain
A prefabricated composite drain usually consists of two components (Figure 3a): a polymeric core
and a filter fabric. The polymeric core provides strength and serves as a secure flow channel for
the collected water. It can be fabricated using different configurations (dimpled, mesh or netted)
with various thicknesses and strengths, and it can be formed into a one-sided or two-sided drainage
core. The filter fabric allows water to freely enter the drainage core while retaining the soil particles
to prevent clogging. The filter fabric used is typically a geotextile fabric composed of strong rot-
proof polymeric fibers formed into a woven or non-woven fabric. There are various commercially
available composite configurations, such as sheet drain, strip drain, and combined drain for
different applications. For continuous structural drainage behind abutments or retaining walls, a
sheet drain (Figure 3a) or a combination drain (Figure 3b) would be more suitable. A combination
drain typically includes a high-profile traditional core and a low-profile sheet core, which can be
connected to standard pipes to ensure a continuous and uninterrupted flow path for the collected
water. For walls having inclusions (such as soil nails) and for mechanically stabilized earth (MSE)
walls – which in both cases will prohibit full wall coverage – then strip drains or chimney drains
can be used.
To choose proper materials for maximum drainage performance, several physical and mechanical
parameters should be considered in the design process. Such parameters conform to three
categories of requirements (Cedergren, 1989; Holtz, et al., 1998; Christopher and Zhao, 2001): (1)
hydraulic requirements, (2) constructability requirements, and (3) longevity (durability)
requirements, which are summarized below.
Hydraulic Requirements: Hydraulic requirements ensure the drainage performance of the

composite drainage system. Flow capability can be characterized with index properties such as the
permeability or permittivity of the geotextile, the in-plane transmissivity of the drainage core, and
the soil retention ability or piping resistance. Clogging resistance can be characterized by the
apparent opening size (AOS) of the geotextile. It should be noted that the wall drains always work
under certain soil pressures. Therefore, the in-plane transmissivity tests should be conducted at
6

different gradients and normal pressures (ASTM D4716). A gradient of 1.0 and normal stresses of
38 kPa (5.5 psi) and 100 kPa (14.5 psi) are used in typical tests.
Constructability Requirements: Constructability requirements prevent material damage or failure

during the construction process. Placement and compaction of backfill will usually generate
considerable compression and shearing forces on the drainage system. Generally, therefore,
constructability requirements are related to the mechanical properties of the filter fabric and the
core. Specifically, the following strength parameters are considered in the design: compressive
strength of the core, puncture strength of the fabric, tensile strength of the fabric, and tear strength
of the fabric. Studies revealed that certain geocomposites exhibit yielding as the stress increases
(Koerner et al., 1985). The yield strength, or the compression/crush strength, represents the
maximum compression load that the core structure can support under a constant compressive strain
rate (ASTM D1621).
Longevity and Environmental Requirements: Although geosynthetic materials are typically inert
for most environments and applications, the filtration properties of the geotextiles may be affected
drastically in cases where the materials are exposed to certain chemical or biological activities.
Examples include chemical clogging by iron or carbonate precipitates and biological clogging by
algae and mosses (Holtz et al., 1998). If such clogging phenomena happen in the field, they can
reduce performance and generate increases in inspection and maintenance costs. Therefore,
consideration of biological or chemical potential is also recommended to be included in the design
process. Other soil environmental processes such as wet-dry cycles and freeze-thaw cycles may
also influence the field performance of the drainage system. Since the geocomposites will be
embedded in the soil after construction, they will not be exposed to UV light, which may have an
effect on the durability. However, special coverage of the geocomposites is required to prevent
UV exposure during transportation and storage. In addition, the geocomposites may also exhibit
“creep” behavior, and the continuing deformation may decrease the permeability and even cause
mechanical collapse. In design, the compression strength from a short-term test should be used
with a reduction factor to accommodate the realization that polymers can be creep-sensitive
materials (Hsuan and Koerner, 2002). A safety factor of 1.3 was recommended by Koerner (1999)
to account for creep.
It should be noted that for each of the index parameters mentioned in this section, there are
corresponding ASTM standards for testing. The specific tests and the corresponding standards are
listed below.
2) Elongation of the fabric, according to ASTM D4632
4) Grab tensile strength of the fabric, according to ASTM D4632.
5) Puncture strength of the fabric, according to ASTM D4833 or ASTM D6241.
7) In-plane flow at various gradients and pressures of the core, according to ASTM D4716.
8) Apparent opening size and equivalent sieve of the fabric, according to ASTM D4751.
7

To assist ODOT in making its decision to substitute the PCDS for the current two feet of porous
backfill with filter fabric, this research will identify the most efficient and environmentally sound
materials and installation procedures for PCDS, evaluate the potential cost benefits by quantifying
the costs, and develop corresponding design and construction specifications.
1.4. Organization of the Report
The remaining portion of this report is organized as follows:

• Chapter 2 summarizes the survey of Ohio counties on the current state of the practice for
structural drainage systems.
• Chapter 3 summarizes the literature review on other promising alternative drainage systems
used in other state DOTs, emphasizing the commercially available PCDS products, DOT
specifications, and the potential for using tire chips as backfill materials.
• Chapter 4 summarizes the laboratory testing results for five select PCDS.
• Chapter 5 describes the construction and performance of two (traditional and PCDS)
drainage systems, evaluated by in-situ instruments.
• Chapter 6 summarizes the cost analysis results based on ODOT’s historical bidding history
as well as the unit price of commercially available PCDS products.
• Chapter 7 presents the draft specifications for PCDS.
• Chapter 8 provides an overall summary and recommendations from this research.
8

CHAPTER 2. CURRENT ODOT PRACTICE FOR
STRUCTURAL DRAINAGE
2.1 Introduction
This chapter summarizes ODOT’s current practice. ODOT Construction and Material
Specifications Item 518 (CMS 518) provides specifications for the drainage of structures. The
drainage method used is referred to as the “two feet porous backfill with filter fabrics” method. In
CMS 518, various aspects in drainage system construction are included such as “fabrication,”
“materials,” “porous backfill,” “pipe,” “scuppers,” “excavation,” “method of measurement,” and
“basis of payment.” It should be noted that CMS 518 has been modified throughout the years. The
ODOT website archives the Construction and Material Specifications from 1995 to 2013. To
evaluate the current drainage method, efforts were made to track and analyze the modifications
made to ODOT CMS 518 over the last 20 years. A survey was also conducted among county
engineers and contractors to collect information regarding implementation of the current version
of CMS 518 (ODOT 2013).
2.2. Modifications to CMS 518
The research team has conducted a comparison among the various versions of CMS 518 and has
concluded that the specifications have been modified over the years. Tracking these changes helps
to evaluate the current specification and will also help to develop new specifications for PCDS.
The following summarizes the history of modifications regarding the porous material and the filter
fabrics.
For the porous material, modifications include:
(1) The minimum thickness of the porous material is 2 feet (0.6 m) in the current specification
(2013 version), whereas a thickness of at least 18 in (0.5 m) was specified in previous versions
(1995, 1997, 2002, 2005, and 2008).
(2) In the versions of CMS from 2005, 2008, 2010 and 2013, the porous backfill should have a No.
57 size gradation according to Table 703.01-1 in CMS 518, whereas in the earlier versions
(1995, 1997, 2002), CMS 518 specified the use of porous materials with a gradation of 100
percent passing the No. 2 (50 mm) sieve, 30 percent passing the ⅜ inch (9.5 mm) sieve, and
10 percent passing the No. 8 (2.36 mm) sieve.
(3) The current version does not specify the requirement for the compaction of the backfill, while
all previous versions did.
The increase of minimum thickness of the porous backfill probably intends to guarantee the
drainage performance by introducing some level of redundancy. Generally speaking, the change
of gradation specifications leads to better graded and relatively finer aggregates. Compaction of
backfill was required and should comply with CMS 603.11.D. Compaction is not required in the
new CMS 518 because Item 603.11.D was eliminated and incorporated into new Item 611, which
9

is a performance-based specification that requires compaction density to comply with the
constructor’s installation plan. In the new CMS 518, porous backfill is placed, but it is not
mechanically compacted. In fact, compacting the porous backfill is not practical when using
conventional compaction equipment. It is usually dumped into place in trenches in lifts. But it
should be noted that insufficient compaction may result in approach slab settlement problems.
For the filter fabric, on the other hand, specific laboratory performance parameters have been
specified including minimum tensile, puncture and tear strengths, maximum apparent opening
sizes (AOS), and minimum permittivity or permeability. Modifications include the following:
(1) Prior to 2008, the minimum permeability was specified as 1×10-2 cm/sec, whereas permittivity
(with a minimum permittivity of 0.5 sec-1) is specified in subsequent years. (Permittivity and
permeability are two terms used to describe the hydraulic flow capacity. Permittivity can be
obtained by dividing permeability by its thickness. If the fabric is 0.2 mm in thickness, a
permeability of 1×10-2 cm/sec is equivalent to a permittivity of 0.5 sec-1).
(2) Prior to 2002, burst strength was included as a specification parameter, but it was not included
afterwards. Burst strength should be considered to assess the survivability of the fabrics during
construction.
(3) Beginning in 2008, ODOT states that “the Department will determine acceptance of Type A,
B, C and D fabric according to test data obtained in the most current report from the National
Transportation Product Evaluation Program (NTPEP).” Since most of the tests ODOT
currently required are being performed by NTPEP and most of the current products on ODOT’s
qualified products list (QPL) are being tested by NTPEP, some revisions were made to old
DOT specifications to match the NTPEP testing.
2.3. Survey of County Engineers and Contractors
A survey was performed among Ohio county engineers in order to investigate the cost and
performance of various drainage systems behind bridge abutment and retaining walls constructed
in county projects. The questionnaire also inquires about the prospects of using prefabricated
geocomposite as an alternative option for drainage purposes. The original survey form can be
found in Appendix A. The survey was sent out by The University of Akron to Ohio’s 88 county
engineers in early November 2014. By the end of March 2015, 34 of the 88 county engineers had
responded to this survey. Responses to different aspects of the survey are summarized in greater
detail below.
10

2.3.1 Types of drainage system
Figure 4. Reported use of various types of drainage system behind abutment and retaining walls by 34
county engineers in Ohio.
Figure 5. Percentages for different drainage systems used in projects in 34 responding counties.
Firstly, this survey investigates the types of drainage system that have been constructed for
abutment and retaining walls in Ohio counties during the past 10 years. Available options for this
multiple choice question include porous backfill with filter fabric (PBF), porous backfill without
filter fabric (PB), tire chips with filter fabric (TCF), prefabricated composite drainage system
(PCDS), and others. The reported use of different drainage systems by 34 county engineers is
shown in Figure 4. It was found that the most common drainage systems that are adopted in Ohio
county projects are PBF (46.6%) and PB (46.2%) (Figure 5). Only one out of the 34 county
engineers has experience with constructing PCDS behind walls. In addition, none of the 34 county
engineers has ever adopted TCF behind retaining walls. It should be noted that only three county
engineers mentioned the use of low strength mortar fill for drainage purpose.
11

2.3.2 Cost of drainage systems
The cost of each drainage system is investigated, and the results are presented in Figure 6 and
Figure 7, which show the reported prices for porous backfill with and without filter fabric,
respectively. It clearly indicates that the cost for PB is typically below $55 per cubic yard, and it
is generally less expensive than the PBF. The unit price for PBF has a wider range and typically
varies from $55 to $85 per cubic yard. A unit price of $67 to $75 per cubic yard is slightly more
common.
Figure 6. Cost of porous backfill with or without filter fabric.
Figure 7 shows the averaged weight of different cost components in various drainage systems. The
cost components include material, labor, equipment and other factors. The weight of each
component in each drainage system is averaged among all survey results. Only one county reported
the weight of cost components of PCDS and only two counties reported the weights of low strength
mortar fill. Therefore, the data for averaged cost components for these two drainage systems
should be viewed with caution. Generally, the cost for material accounts for the largest fraction of
the total cost of drainage systems with backfill (PB, PBF, low strength mortar fill). Based on the
only response, the material cost for PCDS is much less than the labor cost. Since the results are
based on a relatively small response sample size, it is necessary to conduct a more detailed cost
analysis. This analysis will be presented in Chapter 4.
Figure 7. Averaged weight of cost components for different drainage systems.
12

2.3.3 Performance of drainage systems
General performance:
The performance of each drainage system is rated as “excellent,” “very good,” “good,” “average,”
and “poor.” Figure 8 shows the survey results for performance of PB and PBF. Generally, the
performance of porous backfill, either with or without filter fabric, is viewed positively by county
engineers. Porous backfill with filter fabric is considered to perform better than porous backfill
without filter fabric. Five county engineers graded the performance of low strength mortar fill: two
rated this system as “excellent,” two rated it as “good,” and one rated it as “average.”
Figure 8. Performance of porous backfill with or without filter fabric.
Potential issues:
The respondents were asked to report potential issues encountered during service life of the CMS-
specified PBF method. Ten potential issues were listed in the survey responses. Only 12 out of the
34 responded county engineers reported 6 of the 10 issues. It should be noted that this question
offered multiple options, and two or more issues may be indicated on the survey by the same
county engineer. Figure 9 indicates that clogging of the weep hole or drainage pipe is the most
common issue, followed by the appearance of a sink hole near the top wall, erosion behind the
walls, and soil loss from weep holes and the drainage pipe. Wall bulging and weed infestation in
weep holes were also reported.
Figure 9. Reported issues for porous backfill with filter fabric.
13

Maintenance:
Maintenance work conducted for wall drainage systems varies from county to county. Out of the
34 county engineers, 16 reported that maintenance work is conducted annually. In the remaining
counties, maintenance work is either conducted non-routinely or is not considered. Among the
measures that are adopted to maintain the wall drainage system, inspection of weep holes and
outlet is the most common, as shown in Figure 10. Only 16% of the respondents reported cleaning
of the weep holes. This is in accordance with the reported good performance of porous backfill.
Figure 10. Maintenance work conducted for wall drainage systems.
2.3.4 Implementation of CMS 518

The respondents were asked to rate the challenges encountered in the construction of porous
backfill with filter fabric in accordance with CMS 518. Available options include “placing and
compacting of backfill materials,” “placing of filter fabrics,” “placing of drainage pipes/weep
holes,” “QC/QA during construction,” and “other challenges.” Ratings vary from 1 to 5, with 5 as
the most challenging. The rating for each choice is averaged based on all the responses, as shown
in Figure 11. It was found that “placing filter fabric” and “QC/QA during construction” are
considered more challenging than placing backfill materials and weep holes.
Figure 11. Averaged challenge rating in implementing CMS 518.
14

2.3.5 Prospect of adopting PCDS
Since few county engineers have experience with using PCDS, only a limited number of counties
responded to the survey questions on PCDS. Ten out of 34 county engineers indicated that they
would consider the use of PCDS upon the suggestion of the consulting designer or the owner. The
expected benefits of adopting PCDS are summarized in Figure 12. If PCDS are adopted, the
respondents expect it will lead to better drainage performance, simpler construction process, time
savings, labor savings and standardized product or QA/QC measures. Better drainage performance
is viewed as the primary expected benefit. Among the 34 counties, only two counties have reported
using PCDS products (the products were J-Drain and TerraDrain). Two counties suggest
constructing in accordance with specifications from the manufacturers. One county suggests
constructing based on the DOT engineer’s guidance or the contractor’s own judgement.
Figure 12. Benefits typically leading to the adoption of PCDS.
2.3.6 Survey of contractors

Questionnaires were also sent to more than 10 contractors in Ohio. Unfortunately, only a few
contractors agreed to participate, and only two of them (Shelly & Sands and Kokosing) completed
the survey. In contrast to the responses from the counties, it was reported that, for the PCDS, the
material cost accounts for the major portion of the total cost (80%), while labor only accounts for
18%. Again, further cost analysis should be conducted due to the limited number of responses.
According to the contractors, PCDS is mainly adopted to reduce labor costs and to simplify the
construction process. Commonly adopted commercial PCDS include J-Drain, TerraDrain, Mell-
Drain and Amerdrain. PCDS is typically installed following specifications from the manufacturer
and using the contractor’s own judgement.
2.4. Summary
Based on the survey results, it can be concluded that porous backfill systems are the predominant
drainage systems used for retaining walls and bridge abutments in the state of Ohio. Alternative
drainage systems used in Ohio include low strength mortar fill and PCDS. The owners (counties)
are generally satisfied with the performance of the traditional drainage system as well as the
alternative systems. The respondents generally advocate the usage of PCDS if it can bring benefits
such as less cost, labor saving and simpler construction process.
15

CHAPTER 3. ALTERNATIVE DRAINAGE SYSTEMS
As indicated in the survey results, there are several other drainage systems in addition to the
traditional porous backfill methods, including PCDS, low strength mortar fill and tire chips. This
chapter synthesizes the collected information on these alternative drainage systems. Special
attention was paid to the current state-of-the-practice of PCDS: commercially available products,
design methodology, construction specifications, and its applications by state DOTs. Also included
is a summary of a literature review on the application of tire chips as structural backfill and
drainage layers.
3.1. State-of-the-Practice of PCDS as Structural Drains
3.1.1 Commercial products

An online search was conducted to collect information on commercially available prefabricated
drainage systems. It was found that products from numerous manufacturers are available for
composite drainage systems. It should be recognized that a synthesis study on commercially
available PCDS has been conducted previously (Rixner et al., 1986; Koerner and Hwu, 1989;
Hsuan and Koerner, 2000). In this study, additional manufacturers and new products are located.
The index properties of the products are also updated. A total of 17 manufacturers and 87
composite drainage systems were identified and are listed in Appendix B.
For each system/product, we have obtained the datasheet provided by the manufacturers. The
thickness of the core typically ranges from 0.25 in to 1 in. The compression strength varies with
different brands and different materials, ranging from 4,500 psf to 80,000 psf. Note that
polyethylene is a high-density material and is used to achieve high compressive strength. The
hydraulic behavior of the PCDS is typically characterized using the in-plane flow rate (ASTM
D4716). The tests are typically conducted with a gradient of 1.0 and compressive pressure of
2,088~3,600 psf, which intends to model gravity flow conditions such as those behind a vertical
retaining wall. As shown in Appendix B, the in-plane flow varies from 5 gal/min/ft. to 23 gal/min/ft.
The compression strength and the in-plane flow rate of the core structure should satisfy the
construction and drainage capacity requirements as discussed in Section 3.4.
The properties of the geotextile component of the PCDS are also summarized in Appendix B.
The elongation (ASTM D4632) varies between 15% and 70%, and values near 50% are more
common. The puncture strength ranges from 35 to 850 lbs. The grab tensile strength varies
from 80 lbs. to 350 lbs. and the tear strength varies from 40 lbs. to 115 lbs. Regarding the
hydraulic behavior, the apparent opening size for the fabrics ranges from No. 40 sieve to No. 100
sieve. The filter flow rate ranges from 26 gal/min/ft. to 200 gal/min/ft. The permittivity ranges
from 1.00 sec-1 to 2.80 sec-1.
It is apparent that there are a variety of PCDS products with wide ranges of mechanical and
hydraulic properties on the market. The selection of PCDS products should be based on design
16

requirements, cost and availability. The design methodology of PCDS products is presented in
Section 3.1.2. The cost will be discussed in Chapter 6.
For most systems, guidelines on the installation process are also provided by the manufacturer.
The installation guidelines suggested by the manufacturers and the specifications developed by
state DOTs will be synthesized and presented in Section 3.1.4 and Appendix C.
3.1.2 Design methodology

The design of the drainage system includes determination of the strength and drainage capacities
required for selection of PCDS core and fabrics.
In-plane flow
The maximum flow rate requirement can be obtained using flow net analysis:
⎛ F⎞
q = kh ⎜ ⎟
⎝ N⎠
where k is the permeability of the backfill; h is the height of the abutment/wall; and F and N are
the number of flow channels in the flow net and the number of potential drops, respectively.
For vertical walls (Appendix E),

F
=1
N
With the calculated maximum flow rate, the required and allowable transmissivity can be obtained:
q
θ req =
i
θ allow = θ req ⋅ FS
FS is the factor of safety, which accounts for site, construction, and material and design
uncertainties. The values are larger than one and depend on the importance of the structure. A
factor of safety of 3 is considered adequate for retaining wall applications.
To choose a drainage product, the ultimate transmissivity should be calculated by considering a

various reduction factors:
θ ult = θ allow ( RFIN ⋅ RFCF ⋅ RFCC ⋅ RFBC )
The ultimate transmissivity θult is the tested value in the lab or the value listed by the manufacturer.
For field application, reduction factors should be applied to calculate the allowable transmissivity
since the drainage capacity decreases due to a variety reasons such as infiltration of the fabric into
the core space (RFIN), creep deformation of the materials (RFCF), chemical clogging (RFCC) and
biological clogging (RFBC), and other factors. While the most reliable way to determine these
factors is through rigorous laboratory tests, normally they can be obtained via a recommendation
17

table developed by Geosynthetic Institute (Koerner and Koerner, 2007). The reduction factors can
then be determined considering the application area and the site conditions.
Table 1. Recommended drainage reduction factors (Koerner and Koerner, 2007)
For retaining wall and abutment applications, an overall reduction factor of 2.2 is suggested:
RF = RFIN ⋅ RFCF ⋅ RFCC ⋅ RFBC = 1.4 ⋅1.2 ⋅1.1⋅1.2 = 2.2
Core Strength
The compression stress acting perpendicular to the geocomposite comes from the self-weight of
the backfill and any surcharge loads to the upper ground surface. The hydraulic pressure behind
the wall can be assumed to be zero. The earthquake effect is also ignored. It is recommended that
the method to calculate the lateral earth pressure complies with that for design of abutment and
retaining walls, according to the ODOT Bridge Design Manual and AASHTO LRFD Bridge
Design Specifications. Generally, the compression stress can be calculated using conventional
Coulomb’s theory on lateral earth pressure. The maximum pressure occurs at the deepest part of
the wall, which can be expressed as
σ R = γ HK + σ q K
18

where σR and σq are the resultant stress and the surcharge stress at the ground surface, respectively;
γ is the unit weight of the backfill soil; H is the wall height; and K is the coefficient of earth pressure.
The selection of K is critical and depends on the types of the backfill soil and the structure.
In general, for conventional abutment and walls, the at rest lateral earth pressure coefficient K0 can
be used. For semi-integral or integral abutments, passive lateral earth pressure coefficient KP can
be considered (ODOT BDM 303.2.1). Note that the movement required to mobilize full passive
pressure is about 1% to 5% of the height on which the passive pressure acts, depending on the type
of backfill (AASHTO LRFD C3.11.1). In bridge design, the allowable lateral movement of the top
of the abutment is generally limited to 1.5 in. Therefore, for high walls, the allowable lateral
movement may not fully mobilize the full passive pressure. Some value between the K0 and KP
can be used. An equation developed based on experimental data can be used to estimate the lateral
earth pressure coefficient considering lateral movement at the top of the abutment (England and
Tsang, 2001):
⎛ Δ ⎞
0.6
K move = K 0 + ⎜
⎝ 0.03H ⎟⎠
KP
As suggested in AASHTO LRFD 3.11.2 (AASHTO, 2007), the effect of additional earth pressure
induced by compaction shall be taken into account where activity by mechanical compaction
equipment is anticipated within a distance of one-half the height of the wall. The method proposed
by Clough and Duncan (1991) is recommended to estimate the compaction induced residual stress.
However, the effects of compaction only have appreciable effects on the soil at shallow depths.
The deeper the backfill, the smaller the compaction effect on the lateral earth pressure coefficient.
Under normal compaction pressures, the compaction effect usually diminishes at depths greater
than 30 ft. Therefore, the compaction effect on lateral earth pressure can be neglected when
calculating the design pressure for the composite drain board for walls higher than 30 ft.
More discussions on selection of K is detailed in Appendix E.
The surcharge stress can be caused by pavement structures, approach slabs and traffic loading. In
the calculation, the maximum surcharge stress value or a combination of the values should be
considered. It is clear that the resultant stress increases with wall height and depends on the
properties of the backfill soil and the surcharge loading.
It should be recognized that this calculated resultant stress should be the minimum strength of the
PCDS core. In practice, a factor of safety is often applied to ensure the long-term performance of
the drain core structure. The selection of the factor of safety depends on the criticality and severity
of the structure. For the design of bridge abutment and retaining walls drainage, a safety factor of
5 is recommended (Holtz et al. 1998).
Fabric:
As discussed in Section 1.3.4, the geotextile should be selected based on the permeability criteria,
the retention criteria, the clogging criteria and the survivability and durability criteria. The detailed
19

criteria can be found in Holtz et al. (1998). In general, the selection of geotextile depends on the
property of the backfill and/or in-situ soils.
The permeability criteria can be characterized with the permittivity or permeability of the fabric.
Retention capability and clogging resistance can be characterized using the apparent opening size
(AOS) of the geotextile. AOS is a property that indicates the approximate largest particle that
would effectively pass through the geotextile. The strength requirements or the survivability
criteria can be determined by the severity of installation conditions. The FHWA design criteria
developed by Christopher and Holtz (1985) was proved to be an excellent guide for predicting
filter performance. In addition, AASHTO M288 also specifies the geotextile property requirements
for subsurface drainage.
The survivability and durability parameters include the grab tensile strength and elongation, the
puncture strength, the tear strength and the ultraviolet resistance. The AASHTO M288
specification for geotextiles (Class 2) is recommended to be utilized in this regard. In fact, ODOT
CMS 712.09 specifies the geotextile fabrics, and Type A fabrics are specified for underdrains and
slope drains applications (Table 712.09-1 in ODOT CMS). When comparing the values specified
by ODOT and those specified by other state DOTs and AASHTO M288 Class 2, it was found that
ODOT CMS 712.09 Type A geotextile has much lower required values than AASHTO M288
Class 2. The tensile strength of ODOT Type A is also lower than that specified by most other
DOTs. In abutment/wall drainage application, the compaction equipment is usually heavy and the
compaction energy is relatively high. AASHTO M288 specification Class 2 requirements are
recommended.
An example design of the abutment drainage system is provided in Chapter 4, and the
recommended specification for ODOT future application is discussed in Chapter 5.
3.1.3 DOT survey
This section summarizes the state of the practice of the application of PCDS as drainage systems
behind retaining walls and abutments by state DOTs. The Geosynthetic Research Institute (GRI)
published a detailed assessment of the use of PCDS behind vertical retaining structures in 1995
(Wilson-Fahmy and Koerner, 1995). In that report, specifications from nine state DOTs were
collected. In 2000, PennDOT published the result of a new survey and updated the usage of PCDS
by various state DOTs. It was reported that twelve states were using PCDS behind wall or abutment
structures. Seven of the 12 states have detailed material property specifications for the core
structure.
In this research, the research team updated the current usage of PCDS in state DOTs. Information
was collected mainly through online searches of the websites of state DOTs. Special attention was
paid to synthesize the available specifications on the material properties, design method and
construction procedures for PCDS systems. It is found that at least 21 state DOTs are using PCDS
products. Of these, 16 state DOTs have specifications on the material properties of PCDS cores,
five state DOTs provide design guidance on PCDS systems, ten state DOTs specified the
construction procedures, and seven state DOTs provide approved product lists (Table 2).
20

Table 2. DOT Specifications for PCDS
Specifications on PCDS
State DOT Product List
Material Design Construction
Alaska √ √
Arizona √ √ √ √
California √
Colorado √ Same as New York
Georgia √ √ √
Hawaii √ √
Illinois √ √
Kansas √ √ √
Louisiana √ √
Michigan √
Missouri √
Nebraska √
New York √ √
North Carolina √
South Carolina √ √ √
South Dakota √
Tennessee √
Utah √ √
Virginia √
Washington √* √
West Virginia** √ √
*: Only specifications on geotextile fabrics are available.
**: Used only as a pavement edge drain, not as a wall drain.
Regarding the core structure of the PCDS, the material specifications from various state DOTs are
summarized in Table 3. In general, three properties are specified: the thickness, the compression
(crush) strength, and the in-plane flow rate. The thickness of core structure typically ranges from
0.2 in to 1 in. The compressive strength of the core ranges from 2,880 to 15,120 psf, with 5,760
psf as the typical value. The measurement of the compressive strength complies with ASTM
D1621. It should be noted that different states have different considerations when implementing
ASTM D1621. For example, DOTs in Louisiana and Virginia determine the compressive strength
at a deflection of 20% of the core, while the DOT in Missouri determines it at a deflection of 18%.
The DOT in Michigan determines the compressive strength using its own standard MTM 411. The
determination of in-plane flow rate of the core complies with ASTM D4716. The test conditions,
including the normal confining pressure, hydraulic gradient, and seating time, vary among the
DOTs. The normal confining pressures varies from 1,440 psf to 6,000 psf. A hydraulic gradient of
1.0 and a seating time of 15 minutes are typically used. However, some states require a much
longer seating time to consider the effect of creep: DOTs in Virginia and South Dakota require a
21

seating time of 100 hours and South Carolina requires 300 hours. Under various testing conditions,
the specified in-plane flow rate of PCDS ranges from 4 gal/min/ft to 15 gal/min/ft. Virginia’s DOT
also specifies the thermal stability of the core, requiring that the core shall retain at least 75% of
its ultimate strength when subjected to temperatures of 0°F and 125°F for a period of 24 hours.
Regarding the geotextile fabrics of PCDS, properties specified by state DOTs are summarized in
Table 3. It should be recognized that ODOT specifies the use of geotextiles for the purpose of
drainage. According to ODOT CMS 712.09, a geotextile of Type A should be used for drainage.
The specifications for Type A geotextiles are also included in Table 4. The permittivity of the
fabrics ranges from 0.1 sec-1 to 1 sec-1.
Table 3. State DOT specifications on the core structure of PCDS
Core
ASTM D ASTM D
ASTM D 1621 ASTM D 4716
State 1777/5199 1621
DOT Compressive Flow Rate@ Compressive
Thickness Deflections Duration
Strength i=1.0 pressure level
(inch) (%) (hrs)
(psf) (gal/min/ft) (psf)
Arizona 0.23 6,000 — 4 3,000 —
California — — — 4 5,000 —
— — —
Colorado 0.25 2,880 10
Georgia — — — — — —
Hawaii — 5,760 — 15 2,090 —
Illinois <1.125 —
<20 @
10 6,000 —
6000 psf
Kansas 0.23 10,000 20 — 5,000 —
Louisiana 0.31 7,920@ε=20% — — — —
— 4,000@ε=18% — — — —
Michigan
(MTM 411)
Missouri — 6,000@ε=10% — — — —
New York — / — 4 1,500 —
North
0.25 ~ 0.50 5,760 — 5 — —
Carolina
South
0.3 ~ 0.5 14,000
— — 3,600 300
Carolina
South
— — — — 1,440 100
Dakota
Utah 0.4 15,000 — — — —
Virginia — 5,760@ε=20% — 15 1,440 100
22

Table 4. State DOT specifications on the geotextile fabric of PCDS
ASTM D ASTM D ASTM D ASTM D ARIZ 730/ASTM D ASTM D

ASTM D 4632 ASTM D 4632 ASTM D 4632
3786/3787 4533 4833/6421 4651/4751 4491 4355
Geotextile Class in
State DOT Grab/Apparent Apparent
State DOTs Hydraulic/Mullen Ultraviolet
Grab Tensile Seam Strength Elongation (% Trapezoidal Puncture Opening Size Permittivity (per
Burst Strength Resistence/St
Strength (lbs) (lbs) in each Tear (lbs) Strength (lbs) (U.S. Standard second, min)
(psi) ability (%)
direction, max) Sieve Size)
Arizona DOT -- 90 -- 35 min., 115 max 140 30 30 30 ~ 140 0.5 70
Class A 40 0.5
California DOT Class B 157 -- 50 210 -- -- 60 0.2 70
Class C 70 0.1
Class A 180 160 50 80 50% strength
Colorado DOT -- -- >50 --
Class B 80 70 25 25 retained
Flow Rate (50~350
Georgia DOT -- 65 -- 40 -- -- 30 -- --
gal/min/ft, GDT87)
Hawaii DOT -- 110 -- 70 -- -- 70 70 ~ 120 / --
Kansas DOT -- -- -- -- -- -- / / / --
Class B 90 Sewn Seam -- 140 30 30 300 µm 1
Class C 130 Strength (2600 50 210 40 40 300 µm 1
Louisiana DOT --
KN/m width, min
Class D 180 50 290 50 75 212 µm 1
ASTM D 4437)
Michigan DOT -- 90 -- -- 200 45 65 210 µm 0.5 --
Missourri DOT -- -- -- -- -- -- -- -- / --
New York DOT -- 247 -- 50 -- 90 -- 40 0.5 --
North Carolina DOT -- -- -- -- -- -- -- -- -- --
<600 μm (when
<50% in situ soil
passing 75 um);
Ohio DOT -- 80 -- -- -- 25 25/140 <300 µm (when 0.5 --
>50% & <85%
in situ soil
passing 75 µm)
South Carolina DOT -- 80 -- 50 150 40 50 -- 1 70

South Dakota DOT -- -- -- -- -- -- -- -- -- --
Woven 248 223 <50 90 495
430 µm (when
<15% in situ soil 0.5 (when <15% in
passing 75 um); situ soil passing 75
250 µm (when um); 0.2 (when
Class 2 in
>15% & <50% >15% & <50% in
Utah DOT ASSHTO M Non- -- 50
158 142 >50 57 310 in situ soil situ soil passing 75
288 Table 1 Woven
passing 75 µm); um); 0.1 (when
220 um (when >50% in situ soil
>50% in situ soil passing 75 um)
passing 75 µm)
Virginia DOT -- 180 -- -- -- -- -- 50 0.5 --
23

3.1.4. Synthesis of the specifications on PCDS
The information collected from both industry and state transportation agencies is synthesized to
provide a reference to develop the ODOT’s own specifications on PCDS. Considering the space
constraints of the report, the synthesis can be found in Appendix C.
3.2. Tire Derived Aggregates as Backfill and Drainage Layer
In addition to PCDS, ODOT also would like to assess other alternative drainage systems for wall
and abutment structures. In the proposal, the research team cited a previous work on the use of tire
chips as drainage system (White et al., 2005). In this section, a brief review is presented on
applications of tire chips in civil engineering, especially in wall and abutment construction. The
potential advantages and disadvantages are summarized.
General applications and advantages
Tire derived aggregates (TDA) are increasingly adopted in civil engineering construction projects,
and the use of TDA is considered as a green and sustainable approach. The low compacted dry
density, high hydraulic conductivity, and low thermal conductivity make tire shreds suitable for
use as retaining wall backfill. The use of TDA as retaining wall backfill is a relatively new
application that has been studied more in the past decade. Many studies have been conducted to
investigate the feasibility of using TDA as backfill behind retaining walls (Ahn and Cheng, 2014;
Cecich et al., 1996; Drescher and Newcomb, 1994; Humphrey, 2006; Humphrey and James, 2010;
Humphrey et al., 1993; Humphrey and Sanford, 1993; Humphrey et al. 1998; Lee et al., 1999;
Tweedie et al., 1998). These studies showed that TDA backfill performed similarly or even better
than the conventional granular backfill.
The lateral earth pressure for tire shred backfill can be only about half of the value for conventional
granular backfill (Tweedie et al., 1998), allowing for a more economical wall design. Cecich et al.
(1996) concluded that a substantial cost savings could be realized through the use of shredded tires.
The retaining wall backfill applications often include a TDA/sand mixture (Hoppe et al., 2013).
Texas DOT suggests a value for the coefficient of lateral earth pressure at rest to be 0.4. When the
active condition of the retaining wall (that is, the retaining wall moves slightly outward) is reached,
Tweedie et al. (1998) recommend a value of 0.25 for Ka (active lateral pressure coefficient).
Experimental study also revealed that a mechanically stabilized earth (MSE) wall with TDA
backfill performed better than a wall with traditional granular backfill when the MSE wall was
subjected to seismic loading (Xiao et al., 2012). TDA backfill is also shown to help reduce void
development within the abutment backfill.
The high hydraulic conductivity of tire shreds makes them suitable for drainage applications. In
an experimental study by White et al. (2005), TDA was demonstrated to yield the greatest drainage
capacity at 552 cm3/sec, as compared to the drainage capacity of a geocomposite drainage system
at 383 cm3/sec.
TDA is potentially suitable to help reduce the adverse effect of the freeze-thaw process. Firstly,
TDA has large pore sizes such that pore water cannot rise up toward the freezing boundary.
24

Secondly, the thermal resistivity of tire shreds is about eight times greater than the value for typical
granular soil (Humphrey, 2006a). Therefore, tire shreds can be used as an insulating layer to limit
frost penetration beneath roadways (ASTM D6270-98). As a result, the subgrade frost heave
during the winter in cold climate regions can be reduced. In addition, the subgrade support during
the spring thaw can also be improved, since TDA provides better drainage. Potential problems
relating to this application include differential icing and reduced subgrade strength if the TDA
layer is placed close to the surface layer. The effectiveness of insulation properties of TDA
decreases as TDA size and density increase and as the thickness of soil cover increases (Hoppe et
al., 2013).
Regarding bridge abutments, the use of TDA as backfill may also bring additional benefit due to
its elastic nature. It is well known that the expansion joints used in the conventional bridge
abutments may result in substantial maintenance costs. By comparison, integral abutment bridges
have a demonstrated history of reducing both the initial construction cost due to economic use of
materials and the life-cycle maintenance cost (VTrans, 2008). The cost benefit of the integral
abutment bridges has led to its widespread use in North America. However, uncertainties for the
integral abutment also exist due to thermal expansion and contraction, creep, and shrinkage of the
bridge deck as well as the changes in the properties of traditional backfill caused by thermal cycling.
Specifically, cyclic buildup of lateral earth pressures on the stem wall cause seasonal cyclic
movement of the abutment wall. As discussed previously, the elastic nature of TDA may help to
accommodate the lateral movement of the bridge deck and abutment. In addition, the high thermal
resistivity of TAD may prevent thermal cycling–induced buildup of lateral pressure. Through
numerical modeling, a recent study (Mitoulis et al., 2014) also illustrated the structural benefits of
introducing a compressible inclusion of TDA between the abutment and the backfill soil: (i) the
structure–soil interaction effects between the abutment and the backfill are controlled, (ii) the
approaching slab settlement can be prevented, (iii) the ratcheting effect and the increasing of
passive pressures towards the abutment is controlled, and (iv) not only can the residual stresses
induced at the deck can be reduced, but the additional bending moments that affect the prestressing
needs of the deck can also be drastically reduced. Such analysis indicates that TDA is a promising
backfill alternative; however, more research is necessary to confirm its benefits.
Cost and availability
In general, tire shreds are economical when used in lieu of backfilling aggregates. According to a
report by California Integrated Waste Management Board (2001), the approximate cost of
shredded tire is $20–$30/m3, or $15.3–$23/CY. In comparison, backfilling aggregates are typically
~$50/CY. There were also some case studies which reported the cost involved in using tire shreds
as backfilling materials. In a project on US Route 42 near Roseburg, Oregon, it was reported that
the cost of the tire shreds delivered to the site was $30/ton, with placement costing an additional
$8.33/ton; the total cost based on CY was $27/CY; on this job, a state subsidy reduced the cost to
Oregon DOT to $13/CY (Upton and Machan, 1993). In another project located in Milpitas,
California, it is reported that the total in-place cost for tire shreds was $35.84/CY (including the
costs of purchase, delivery, and placement), before considering state subsidy. Cecich et al. (1996)
did an economic analysis comparing the estimated costs for both shredded tires and sand used in
retaining wall construction. It was concluded that by using shredded tires instead of sand as the
backfill material, the material costs can be reduced by 81% to 85%, and the total cost savings can
25

range from 52 to 67%. In a highway embankment project in Portland, Maine, the use of tire shreds
as a lightweight fill saved the Maine Turnpike Authority a total of $300,000 (California Integrated
Waste Management Board, 1999).
Shredded tires are widely available in Ohio. According to Ohio EPA, more than 12 million scrap
tires are generated in Ohio each year. There are more than 20 tire recycling facilities distributed
all around the state of Ohio and more than 90% of the scrap tires in Ohio are recycled. The
beneficial use of shredded tire includes tire derived fuels, ground rubber, landscaping materials,
and playground surfaces. While there is no statistical data showing the exact amount of shredded
tire available for use in civil engineering constructions in Ohio, a rough estimation is made based
on a nationwide study on the consumption of shredded tire in various markets (Rubber
Manufacturer Association, 2015). In that study, it was found that about 7.7% of the tire shreds
were used for civil engineering applications. Based on this, the estimated tire shreds available for
backfilling purpose in Ohio is around 13,600 tons or 27,000–40,000 CY.
Special considerations and potential disadvantages
Use of TDA in civil engineering also has some potential disadvantages. Decisions should be made
after assessing the suitability of TDA in specific applications. The following aspects should be
taken into consideration:
1) Since the elastic modulus for TDA is 1 to 2 orders lower than that of granular soil, TDA backfill
needs to be overbuilt to accommodate the expected settlement. Furthermore, when TDA is
used as a drainage layer, the compression of TDA reduces the void ratio and thus the hydraulic
conductivity.
2) The hydraulic conductivity may also be reduced due to biochemical or physical clogging
(Geosyntec Consultants, 2008). In environments that are rich in microbes, biological reactions
may cause inorganic matters to deposit. As a result, the pore space and thus the hydraulic
conductivity are reduced. For example, Rowe and McIsaac (2005) conducted column tests
permeated with landfill leachate for up to two years. It was reported that the hydraulic
conductivity of TDA dropped from 0.7 to 2 cm/s to between 10-5 and 10-6 cm/s after one year,
under a compressive pressure of 3,132 psf. Therefore, the authors suggest that gravel be used
in leachate-critical zones. Apart from biochemical clogging, physical clogging may occur as
sediments build up over time with the TDA voids. ASTM D6270-98 suggests that the tire shred
layer should be wrapped with woven or nonwoven geotextile to minimize the infiltration of
soil particles into the tire shred voids (ASTM, 2004).
3) Internal heating is a potential issue in tire shred backfill. The possible causes of internal fires
include the oxidation of steel belts and rubber, free access of air and water, retention of heat
in thick TDA fills, large amounts of exposed steel belts, and the presence of inorganic and
organic nutrients that enhance microbial activity (ASTM D6270-98, ASTM, 2004; Scrap
Tire Management Council, 1997). To address these causes, the Ad Hoc Civil Engineering
Committee (1997) developed design guidelines to minimize internal heating of tire-derived
fills. Thermal gravimetric analysis by Moo-Young et al. (2003) showed that tire shreds are
generally stable at temperatures up to 200˚C.
26

4) The potential release of contaminants from TDA materials may raise some environmental
issues if not properly addressed. As an example, some problems in using TDA as a drainage
layer have been encountered in a project in Maine. Iron bacteria or rust from the steel belts in
the shredded tires have created aesthetic issues with the effluent emanating from a TDA
subgrade French drain (Maine DOT, 2003). Other studies (Twin City Testing, 1990; Downs,
et al., 1996) have shown that metal can be leached in low pH environments and that organics
can be leached in high pH value environments. Therefore, it is recommended that tire shreds
should be used in environments where the soil and groundwater are around a neutral pH value.
A study by O'Shaughnessy and Garga (2000) suggested that the possibility of release of
toxicants is higher when using newly discarded tires. Therefore, it is better to use scrap tires
that have been stockpiled or have been used for some time for TDA-reinforced earth structures.
5) In addition to meeting the practice guidance by ASTM D6270-98 (ASTM, 2004) on addressing
the internal heating issue, Humphrey (2006b) provided some gradation requirements for the
selection of a TDA material to use as backfill:
• A minimum of 90% (by weight) must have a maximum dimension, measured in any
direction, of 12 in., and 100% must have a maximum dimension, measured in any
direction, of 18 in.;
• A minimum of 75% (by weight) must pass the 8-in. sieve;
• A maximum of 50% (by weight) must pass the 3-in. sieve;
• A maximum of 25% (by weight) must pass the 1.5-in. sieve;
• A maximum of 1% (by weight) must pass the No. 4 sieve; and
• All TDA particles must have at least one sidewall severed from the tread.
3.3. Summary
systems were identified, and their properties were summarized. The selection of PCDS
• The design methodology for PCDS were developed by referring to existing research
type and importance of the structures, survivability and the constructability.
27

state DOTs specify the construction procedures, and seven state DOTs provide approved
product lists.
• A brief review is presented on applications of tire derived aggregates (TDA) in civil
engineering, especially in wall and abutment construction. Its potential as a drainage
backfill is assessed. While its low cost, status as a recycled material, as well as its low
compacted dry density, high hydraulic conductivity, and low thermal conductivity make
TDA a suitable backfill material for abutment and retaining walls, some major
disadvantages also should be considered during the decision making process:
1) compression-induced decrease of hydraulic conductivity of TDA backfill has not been
well studied, 2) biochemical or physical clogging of TDA is not well understood, and
3) internal heating and the potential release of contaminates from TDA materials may cause
internal fires or raise some environmental issues if not properly addressed.
28

Chapter 4. Laboratory Testing of Drainage Composites
ODOT does not currently specify the use of composite drainage boards for applications behind
abutment or retaining walls. ODOT has only a limited understanding of the various properties of
composite drainage boards and would like to test the index properties of several type of
commercially available products in order to gain a better understanding. While this effort can also
help ODOT develop future programs for verification testing of candidate products to be included
in the qualified product list (QPL), the objective of these tests was not to develop such a QPL.
4.1. Laboratory testing program and methods
Five products were acquired from different vendors for index property testing in the laboratory.
As discussed in Section 1.3.4 and Section 3.1.2, the criteria for an efficient drainage system include
permeability criteria for both the core and fabric as well as retention and clogging resistance
criteria for the fabric. In addition, the system should possess enough strength to survive the
construction process and to perform in a satisfactory manner during its design life. Therefore, the
hydraulic and mechanical index properties should be specified. A series of tests was selected to
ensure the composite drain complies with the aforementioned criterion. The tests were performed
according to corresponding ASTM standards:
6) Puncture strength of the fabric, according to ASTM D4833.
8) Apparent opening size and equivalent sieve of the fabric, according to ASTM D4751.
The basic features of the five products tested are listed in Table 5
Table 5. Tested products and corresponding material types for separate core and fabric
PRODUCT ID Core Type Fabric Type

A Dimpled drainage core Needle punched nonwoven
B Dimpled drainage core Woven monofilament
C Dimpled polystyrene core Needle punched nonwoven
D* Dimpled drainage core Nonwoven filter fabric
E Formed polystyrene core Nonwoven filter fabric
* Product D also has a waterproof membrane on the back side.
All samples were shipped as composites with fabric bonded to the core. Note that the material
properties listed in the product data sheets for most commercially available products are for
29

separate core and fabric. In order to compare the tested values with the ones listed by the
manufacturers, separate components were requested. However, only separate fabric materials for
Products A, C and E were obtained. Since the bonded materials will actually be used in the field,
direct testing with the bonded composites is more appropriate. This was particularly the case for
the in-plane flow and compressive strength, where only bonded materials were used in this study.
The laboratory tests were conducted at the University of Wisconsin-Madison (UW) through a
subcontract. The test results can be used for design and selection of proper prefabricated drainage
systems, and they can also provide a measure for acceptance/rejection of shipped products.
4.2. Summary of testing results
Summary results for each product are provided in Tables 6-10. Detailed procedures and results are
provided in Appendix G.
30

Table 6. Laboratory test results for Product A
31

Table 7. Laboratory test results for Product B
32

Table 8. Laboratory test results for Product C
33

Table 9. Laboratory test results for Product D
34

Table 10. Laboratory test results for Product E
35

4.2.1. In-plane flow
It was reported in previous research that fabric intrusion can cause reduction of the in-plane flow
capacity of geocomposite. When determining the in-plane flow rate, the hardness of the load platen
and backfill material can affect the degree of fabric intrusion. It was found that geotextile intrusion
shows the largest reduction in geocomposite drainage in-plane flow performance (2 ~ 5 times the
in-plane flow reduction due to compression of the core). Geotextile intrusion is product-specific
and is most significant with soil backfill. According to ASTM D4716, for index testing or for
acceptance testing, the contact surfaces should be prescribed by the material specification. In the
absence of a specification, rigid sub and superstrates can be used to minimize the variables
impacting the test results. For performance testing, it is recommended that the nature of the
material in contact with the geocomposite in the field should be modeled. A rigid platen simulates
similarly rigid surfaces (such as concrete walls) while a layer of rubber membrane or representative
soil placed between the platen and the geocomposite specimen can be used to simulate the in-situ
soil. In this study, only a rigid platen was used.
Since separate components were not available for several products, the effect of the bonded fabric
on the in-plane flow was evaluated using Product A. In addition, the in-plane flow capabilities for
each composite were tested under different seating periods, flow directions, hydraulic gradients
and confining pressures. The results are summarized as follows.
Effect of bonded fabric
Table 11. Effect of bonded fabric on the in-plane flow rate of Product A
Product In-Plane Flow Rate (gal/min/ft)

ID Tested Tested (Core without Difference Manufac. Val.2
(Core+fabric) fabric) (core)
A 14.5 24.8 42% 22.7
It is clear that the existence of the fabric tends to reduce the in-plane flow rate significantly. In this
tested case, the reduction is about 42%. This is caused by the fact that the bonded fabric would
block part of the clearance space of the core. For the core alone, the tested value was comparable
(slightly larger in this case) to the manufacturer value.
Effect of seating period (creep)
For each product, both short-term and long-term in-plane flow rates were tested. The short-term
test was conducted with a seating period of 0.25 hours (15 minutes), while the long-term test was
conducted with a seating period of 100 hours. This test was conducted to determine if the product
shows any performance degradation (creep) with time. The results show that there were no
significant differences between the flow rates for the seating periods of 0.25 hours and 100 hours
for Products A, B and E, which showed no deformation or creep under the test loading. Most of
the products actually showed a higher flow rate for the longer seating period, a result that was not
expected. In particular, the flow rates for Products C and D were not well matched for the 0.25-
hour and 100-hour seating periods. The flow rates increased by 42% and 9% for Products C and
36

D, respectively, under a normal stress of 3,600 psf and a hydraulic gradient of 1.0. These results
can be attributed to the measurement uncertainty, as it is certain that all testing procedures were
conducted under the same conditions, and no problems were encountered during the tests.
Table 12. Effects of seating period on in-plane flow rate
Product ID In-Plane Flow Rate (gal/min/ft)

Tested (core + fabric) Manufacturer’s
Value2 (core)
Short Term Long Term Difference
A 14.5 14.9 3% 22.7
B 17.5 18.1 3% 21
C 11.8 16.7 42% 17
D 14.5 15.8 9% 18
E 17.5 17.3 -1% 21
Effects of flow direction
For each product, the tests were conducted in two directions, the machine direction and the cross-
machine direction. During manufacturing, the dimples may have slightly different dimensions
along and across the machine direction. The test results show that, in general, the machine direction
results in a slightly larger in-plane flow rate for most products (Products B, C, D, and E). For
Product A, the effect is mixed. Generally speaking, the effect of flow direction is moderate (15%-25%).
Effects of normal stress and hydraulic gradient
For each product, the in-plane flow tests are tested under three levels of normal stress (i.e., 720
psf, 2,000 psf, and 3,600 psf) and three levels of hydraulic gradient (i.e., 0.1, 0.5, and 1.0). It is
clear that in-plane flow increases nearly linearly with the hydraulic gradient but does not change
much with the tested normal stress levels, which are not sufficient to significantly deform the core.
4.2.2. Compression strength of the composite
As shown in Table 13, the tested compressive strength values for the composite deviate from the
manufacturer values for the core alone. The differences vary from -25.5% to +12.6%. For four of the
five tested products, the tested values were lower than the manufacturer values. There are several
possible reasons: 1) the manufacturers typically report the strength of the core only; 2) the tests were
performed using a single layer of the composite, which has a thickness less than 1 inch minimum
thickness requirement by ASTM D1621; and 3) a loading rate of 0.1 in/min was used, which was a
relatively high loading rate for a sample that is thinner than 1 in.
37

Table 13. Comparison of tested average compressive strength to the manufacturer value
Product ID Tested Manufac. value Difference (tested vs. manufacturer)

Avg. compr. Max. Compr. Strength
strength psf (kPa)
psf (kPa)
A 13,865 (664) 18,619 (890) -25.5
B 17,505 (840) 18,000 (861) -2.75%
C 14,067 (674) 15,000 (719) -6.22%
D 16,896 (810) 15,000 (719) 12.64%
E 17,530 (840) 18,000 (862) -2.61
Note that although most of the manufacturers list the compressive strength values tested according
to ASTM D1621 “Standard Test Method for Compressive Properties of Rigid Cellular Plastics,”
an alternative testing method, ASTM D6364 “Standard Test Method for Determining Short-Term
Compression Behavior of Geosynthetics” is more suitable for drainage composites. In fact, ASTM
D1621 was designed with regard to the behavior of cellular material under compressive loads. It
requires a specimen height of at least 25 mm (1 in.) but no greater than the width or diameter of
the specimen. The specimen is then loaded at a rate of crosshead displacement of 0.1 in./min for
each 1 in. of specimen thickness. This causes confusion as to whether a constant rate of 0.1 in/min
or a rate of 10%/min should be used. Furthermore, the geosynthetic products are usually less than
1-in. thick. For example, the tested products in this study were about 0.4 in. in thickness. ASTM
D6364 is designed specifically for geosynthetics, and it requires the specimen to be tested at a
strain rate of 10% of the nominal thickness per minute or 1 mm/min (0.04 in/min), whichever is
greater. Theoretically, ASTM D6364 is more suitable for geocomposite drains. But to be consistent
with the manufacturers’ tests, ASTM D1621 is followed in this study. It is recommended, however,
that ASTM D6364 should be adopted in future testing programs.
4.2.3 Thickness of the composite, core and fabric
The nominal thickness of the geosynthetics was determined following ASTM D5199-12 by
measuring the distance between two parallel surfaces of test specimens confined under a specified
normal stress (20 kPa or 2 kPa). The tested average composite thicknesses under two different
levels of normal stress were very close to each other. For the separate core, three samples were
received, and the tested thicknesses were slightly less than the values reported by the manufacturer.
38

Table 14. Thickness of the composite and separate fabric and core
Product ID Fabric 1 Core Composite Difference

(as tested
vs.
manufac.)
Tested Tested Manufac. Tested (Avg.) Core
(Avg.) (Avg.) value (cm)
(cm) (cm) (cm)
at 2 kPa at 20 kPa
A 0.05 (1.21) 0.41 (10.5) 0.43 (11) 0.43 0.44 -4.40%
B 0.03 (0.68) 0.37 (9.3) 0.40 (10.2) 0.38 0.39 -7.50%
C2 0.05 (1.21) ND 0.40 (10.2) 0.39 0.41 ND
D2 ND ND 0.40 (10.2) 0.42 0.43 ND
E 0.04 (1.06) 0.40 (10) 0.44 (11) 0.43 0.43 -9.00%
1 No fabric thickness was defined in the manufacturer data.
2No separate core samples were obtained.
4.2.4 Grab tensile strength and elongation of the fabric
Three products (Products A, C and E) were tested. Two of them (Products A and C) showed higher
tested values in the weaker principle direction as compared to the manufacturer values. They also
exhibited larger apparent elongation at the breaking load.
Table 15. Comparison of tested tensile strength and elongation to the manufacturer value
Product Weak direction Manufac. value Difference (as tested vs.

ID manufac.)
Grab tensile Apparent Grab tensile Apparent Grab tensile Apparent

strength (lb) elongation strength (lb) elongation strength (lb) elongation
(%) (%) (%)
A 159 55.1 160 50 5.60% 10%
B ND ND 365 ND ND ND
C 159 55.1 160 50 5.60% 10%
D ND ND 100 60 ND ND
E 105 46.9 160 70 -28.1% -33%
4.2.5 Trapezoidal Tear Strength
Three products (Products A, C and E) were tested. Two of them (Products A and C) showed higher
tested values as compared to the manufacturer values. The trapezoidal tear strength for Product E was
not listed in the manufacturer data sheet.
39

Table 16. Comparison of tested tear strength to the manufacturer value
Product ID Weak direction Manufac. value Difference (tested vs.

manufac.)
Tear strength (lb) Tear strength (lb) Tear strength (lb)
A 65.9 60 9.83%
B ND NA ND
C 65.9 60 9.83%
D ND 60 ND
E 55.5 NA ND
4.2.5 Puncture Strength
Different standards are specified to determine the puncture strength: ASTM D4833 “Standard Test
Method for Index Puncture Resistance of Geomembranes and Related Products” or ASTM D6241
“Standard Test Method for Measuring the Damage Resistance of a Fiber-Reinforced Polymer-Matrix
Composite to a Concentrated Quasi-Static Indentation Force.” ASTM D4833 was originally designed
for geomembrane materials and specifies the use of a solid steel rod with a diameter of 8 ± 0.1 mm
[0.315 ± 0.004 in.] having a flat end with a 45° = 0.8 mm [0.315 in.] chamfered edge contacting the
surface of the test specimen. The tested strength using ASTM D4833 is also called pin puncture
strength. It is recommended that geotextile and geotextile related products be tested using ASTM
D6241, which specifies the use of an indenter having a smooth hemispherical tip with a diameter of
13.0 ± 0.3 mm [0.50 ± 0.01 in.] and a hardness of 60 to 62 HRC. The tested strength using ASTM
D6241 is also called CBR puncture strength. Research has been conducted to correlate the pin and
CBR puncture strengths, considering different geotextile types and test conditions (Van Dyke, 2014).
For nonwoven geotextile under common test conditions (using constant rates of compression of 300
mm/min and 50 mm/min for pin and CBR tests, respectively), the following equation can be used to
estimate the CBR puncture strength for nonwoven geotextiles:
StrengthCBR,estimated = 5.27 × Strength pin,measured
While ASTM D4833 was used in this study, some of the manufacturers used ASTM D6241. The CBR
puncture strengths were estimated using the pin puncture strengths. Three products (Products A, C and
E) were tested. Two of them (Products A and C) show higher estimated values as compared to the
manufacturer values. Product E showed a lower value in the test as compared to the value listed by the
manufacturer.
40

Table 17. Comparison of tested/estimated puncture strength to the manufacturer value
Product Tested Test Estimated Manufac. Manufac. Difference (tested

ID avg. method CBR value method vs. manufac.)
(lb) strength (lb) (lb)
A 89.6 ASTM 472 410 ASTM 15%
D4833 D6241
B ND ASTM ND 675 ASTM ND
D4833 D6241
C 89.6 ASTM 472 410 ASTM 15%
D4833 D6241
D ND ASTM ND 250 ASTM ND
D4833 D6241
E 69.5 ASTM 366 90 ASTM -22.8%
D4833 D4833
4.2.6 Permittivity and flow rate
Most products have both the flow rate and permittivity values listed on the manufacturer data sheet.
Permittivity is the volumetric flow rate of water per unit cross-sectional area per unit head under
laminar flow conditions, in the normal direction through a geotextile. Therefore, the permittivity
depends on the head. The constant head method was adopted in this test and a head of 50 mm was used.
Four products (Products A, B, C and E) were tested for permittivity and flow rate. Separate fabrics
were received from the vendors directly for Products A, C and D. The fabric for product B is woven
textile, which can be debonded from the core. All nonwoven textiles (Products A, C and D) showed
lower permittivity in the tests than the manufacturer values. Note that since a constant head of 50 mm
is specified for the constant head method in ASTM D4491, the ratio between the flow rate and the
permittivity should be constant. However, for some products (Products B and E), this constant ratio
was not used. It may be that the manufacturers used a different constant head for the tests. Therefore,
it is recommended that the permittivity, instead of the flow rate, should be used in the specification.
Table 18. Comparison of permittivity and flow rate to the manufacturer value
Product Tested Manufacturer values Difference (tested vs.

ID manufac.)
Permittivity Flow rate Permittivity Flow rate Permittivity Flow
sec-1 gal/min/ft2 sec-1 gal/min/ft2 Rate
(l/min/m2) (l/min/m2)
A 1.08 79.5 1.5 110 -28.0% -27.9
(3233) (4481)
B 2.04 150.5 2.1 145 2.86% 3.79%
(6124) (5904)
C 1.08 79.4 1.5 110 -28.0% -27.9
(3233) (4481)
D ND ND NA 140 ND ND
(5700)
E 1.4 100 1.8 110 -22.2% -9.12%
(4074) (4483)
41

4.2.7. Apparent Opening Sizes (AOS)
Apparent opening size (AOS) is the apparent largest hole size in the geotextile, and AOS
determines the ability of geotextile to retain soil particles. Four products (Products A, C, D and E)
were tested for AOS. The tested values are, in general, larger than those provided by the
manufacturers. All four manufacturers suggest an AOS of 0.210 mm or US No.70 sieve, but the
tested results are between 0.32 mm~0.42 mm, or US No. 50~No. 40 sieves.
Table 19. Comparison of AOS to the manufacturer value
Product ID Tested Avg., in mm Manufac. Value, in mm Difference (tested vs. manufac.)

(US Sieve) (US Sieve)
A 0.32 (50) 0.212 (70) 51%
B ND 0.43 (40) ND
C 0.32 (50) 0.212 (70) 51%
D 0.42 (40) 0.212 (70) 98%
E 0.36 (50) 0.210 (70) 71%
4.3 Summary and conclusions
Five PCDS products from different vendors were acquired for index property testing in the
laboratory. Eight properties of the core or fabric were tested according to the corresponding ASTM
standards.
• The eight properties and the standards selected for testing include:
2) Compressive strength of the core, according to ASTM D-1621.
3) Thickness of the core and the fabric, according to ASTM D-1777 or ASTM D-5199.
4) Permittivity of the fabric, according to ASTM D-4491.
5) Grab tensile strength and elongation of the fabric, according to ASTM D-4632.
6) Puncture Strength of the fabric, according to ASTM D-4833.
7) Trapezoidal tear strength of the fabric, according to ASTM D-4533.
8) Apparent Opening Size and Equivalent Sieve of the fabric, according to ASTM D-4751.
Some of the methods were later found not ideal for PCDS and recommendations of
• Regarding the in-plane flow of the core, it was observed that
o The presence of fabric tends to reduce the in-plane flow rate significantly. In this
tested case, the reduction is about 42%.
of 0.25 hours and 100 hours under the confining pressure levels used in this study.
o Flow direction (machine direction or cross-machine direction) has a moderate
effect on the in-plane flow of the core.
42

o In-plane flow increases nearly linearly with hydraulic gradient but does not change
much with the tested normal stress levels, which are not sufficient to significantly
deform the core.
o Compared to the manufacturer-listed values, the tested in-plane flows fall short for
all products when the fabric is bonded. When tested using the core only, the tested
value exceeded the listed value.
manufacturer values for the core alone. The differences varied from –25.5% to +12.6%.
For four of the five tested products, the tested values were lower than the manufacturer
values. It was later realized that this result may due to the testing method and conditions
used. It is recommended that ASTM D6364 “Standard Test Method for Determining Short-
Term Compression Behavior of Geosynthetics” be used to test the compressive strength of
PCDS, in lieu of ASTM D1621.
under two different levels of normal stress were very close to each other. For the separate
core, three samples were received, and the tested thicknesses were slightly lower than the
manufacture values. The thickness of a core is an indirect indicator for the drainage
capacity.
• Regarding the strength properties of the fabric, only one product showed significantly
lower values than the listed values among the three products tested. It was found that
different manufacturers test the puncture strength of the fabric using different ASTM
standards. ASTM D6241 is actually more suitable than ASTM D4833 for geotextile
materials.
• Regarding the hydraulic properties of the fabric, all nonwoven products showed lower
tested permittivity than the listed values. Since the permittivity and flow rate of a geotextile
product are correlated and the permittivity is a value normalized by the differential head
specification. For every tested product, the tested AOS was significantly larger than the
listed value.
• The differences shown between the tested and listed values suggest that a factor of safety
should be used in the design when selecting the PCDS. Based on the discrepancies
observed, a proposed factor of safety (FS) of 3 is reasonable.
43

CHAPTER 5. SITE IMPLEMENTATION AND PERFORMANCE
EVALUATION
To evaluate the effectiveness of the PCDS systems, two bridges were selected to implement both
the traditional two-foot porous backfill drainage system and the PCDS system. Instrumentation
including piezometers inside the backfill, tiltmeters attached on the bridge abutment and
monitoring flumes at the drainage outlet were installed on site to monitor the change of pore water
pressure behind the abutments and the wing walls, the tilt angle of the abutments, and the drainage
flow rate at the two bridge abutments. This chapter begins with a discussion of the site selection
and the development of plan notes for the testing sites, followed by the observations of the
installation processes and the data analysis on the performance of the two drainage systems. The
findings on the field performance the two systems are summarized at the end of the chapter.
5.1. Site Selection
It is desirable to have two structures (abutment or retaining wall) instrumented for the field testing:
one structure with PCDS and the other with the system currently specified by ODOT (i.e., two feet
of porous backfill with filter fabric). It is also desirable that two site locations that are close to each
other be used to eliminate bias induced by different environmental conditions. In addition to the
two sites selected for the field tests, the research team requested to observe the installation process
of the traditional drainage system at a third site.
A total of 13 projects were considered for site selection. The plan sheets for all the projects were
reviewed in detail. Based on the project schedules, locations, abutment/wall types, and other site
conditions, two bridges (FRA-270-1714 and FRA-270-1716) were selected for the field evaluation.
The abutments and wing walls at FRA-270-1714 had a PCDS system, while at FRA-270-1716, a
system having the traditional two feet of porous backfill with filter fabric was installed.
Bridges FRA-270-1714 and FRA-270-1716 are two bridges at the intersection of Interstate 270
(I-270) and US Route 33 in Dublin, Ohio. Both bridges cross over I-270. FRA-270-1714 is part
of the RAMP ES and FRA-270-1716 is part of the RAMP WN (Figure 13). Both bridges were
built with continuous curved steel plate girders and have reinforced concrete composite decks. The
abutments and wing walls are supported with continuous spread footings. The typical elevation of
the top of the rock formation is about 860 ft, or about 20 ft below the footings.
In addition, one bridge, ATB-20-21.43, was selected for additional observation of the placement
of two feet of porous backfill.
44

Figure 13. Location of the selected sites for field testing.
5.2. Development of the plan notes for the testing sites
5.2.1 PCDS design
Site conditions
The abutments and wing walls for bridges FRA-270-1714 and FRA-270-1716 are all cantilever
walls. The heights of the abutments for both bridges are 30 ft; the total coverage area of the
abutment and two wing walls for FRA-270-1714 and FRA-270-1716 are 5,859 SF and 6,249 SF,
respectively. According to the cost analysis based on ODOT historical bidding data (Chapter 6),
the wall coverage areas for these two bridges is large enough that it becomes more economical to
use a PCDS as the drainage system.
On the plan sheet, it was specified to use Type B granular material, Item 703.16.C, to fill the void
created by excavating for the abutment footings. Item 203 was specified for the approach
embankment. Natural soil was mainly used to fill the void between the constructed abutment and
the pre-existing slope to construct the embankment. Note that before the construction of the
abutments, part of the approach embankment had been placed, and the grade of the existing slope
is shown in Figure 15. Vegetation was well developed on site for erosion control. No further
actions were taken to remove the organic materials. The natural soil was a granular material with
a wide range of particles sizes. Cobble and gravel sizes were common on site. Soil samples without
cobble and gravels were collected for laboratory characterization. It was recommended that for
45

drainage design purposes, only soil passing the 4.75 mm (US No. 4) sieve should be used when
the soil contains particles 25 mm and larger. The particle size distribution (<4.75 mm) and
parameters from the sieve analysis are shown in Figure 14 and Table 20. The liquid limit is 29,
and the plasticity index is 13. Based on the AASHTO Classification System, the native soil is
classified as A-2-6. According to the Unified Soil Classification System, the soil is classified as
clayey sand (SC).
Compaction tests were also conducted in the lab. The optimal moisture content is 15%, and the
maximum bulk density and dry density were 135 lb/ft3 and 117 lb/ft3, respectively. The
permeability of the native soil, when compacted to 98% of its maximum dry density, was
1.21×10-7 m/sec, which is very low. Note that the organic content was not measured. During
construction, the contractor and ODOT decided to place two feet of Type C granular material
(ODOT Item 703.16.C) between the PCDS board and the natural soil backfill at FRA-270-1714.
The Type C material has a gradation as shown in Table 21. For drainage design, the worst-case
soil is the one with highest coefficient of permeability, since the more permeable soil controls the
maximum drainage flow. In this case, the two feet of granular material is more permeable, but the
majority of the backfill soil is clayey sand, which has a much lower permeability. In the drainage
design, an intermediate permeability of 5.0×10-5 m/sec was assumed.
Table 20. Soil characterization of the native backfill soil
Parameter D50 D60 D85 PI LI

(mm) (mm) (mm)
Value 0.22 0.3 0.75mm 13 29
100
80
Percent passing
60
40
20
0
10 1 0.1 0.01
Sieve size (mm)
Figure 14. Particle size distribution of the natural soil on site.
Table 21. Gradation specified for Type C granular material (Item 703.16.C).
46

Drainage Capacity
The required drainage capacity of a bridge abutment site depends on the on-site soil permeability
and site topography, vegetation coverage, abutment height, precipitation characteristics and other
factors. For preliminary design, the drainage capacity can also be obtained through flow net
analysis. A schematic showing the flow net behind a typical wall or abutment can be found in
Appendix E. The steps to obtain the drainage capacity are as follows:
1) Calculate the maximum flow rate to the drainage board per unit width (1 foot):
⎛ F⎞
q = kh ⎜ ⎟
⎝ N⎠
⎛ 5⎞
= ( 5.0 × 10 −5 ) ( 3.28 ) ( 60 ) ( 30 ) ⎜ ⎟
⎝ 5⎠
= 0.2952 ft 3 min⋅ ft
= 2.2 gal min⋅ ft
F and N are the number of flow channels in the flow net and the number of potential drops,
respectively.
2) Determine the flow gradient within the drainage board:
i = 1.0 ,
for gravitational flow
3) Calculate the required transmissivity,

q
θ req =
i
= 2.2 gal min⋅ ft
4) Determine the allowable transmissivity of the board,

θ allow = θ req ⋅ FS
= ( 2.2 ) ( 3)
= 6.6 gal min⋅ ft
5) Determine the ultimate transmissivity of the board (as tested or listed by the manufacturer),
θ ult = θ allow ( RFIN ⋅ RFCF ⋅ RFCC ⋅ RFBC )

= ( 6.6 ) (1.4 ⋅1.2 ⋅1.1⋅1.2 )
= 14.7 gal min⋅ ft
47

Compressive Strength
Assuming a friction angle of 31° for clayey sand, K0 is 0.426 based on AASHTO (2007).
Considering a total surface and live load of 650 psf, a bulk density of 135 psf and a wall height of
30 ft, the lateral earth pressure can be estimated using the following equation:
σ R = γ HK0 + σ q K0
The calculated allowable strength of the drainage board is 2,002 psf. Further considering a factor
of safety of 5 to resist creep effects and to account for field conditions, as suggested by FHWA
(Holtz et al. 1998), the design strength of the drainage board is 10,000 psf.
Fabric design
Fabric is placed in direct contact with the retained soil to 1) allow water to flow through and 2) to
keep the soil particles in place. If some particles do migrate, they should not cause clogging of the
downstream media. Therefore, there are three criteria for the design of a fabric, namely, the
permeability criterion, the retention criterion and the clogging resistance criterion. In addition to
these criteria are constructability and strength requirements to prevent material damage or failure
during the construction process.
Retention capability and clogging resistance can be characterized with apparent opening size (AOS)
of the geotextile. AOS is a property that indicates the approximate largest particle that would
effectively pass through the geotextile. Permeability criteria can be characterized with permittivity
or permeability of the fabric. The strength requirements or the survivability criteria can be
determined by the severity of installation conditions. The FHWA design criteria developed by
Christopher and Holtz (1985) was proved to be excellent for predicting filter performance. In
addition, AASHTO M288 also specifies the geotextile property requirements for subsurface
drainage. According to FHWA (Holtz et l., 1998):
1. For nonwoven fabric and steady flow conditions, the retention criterion is
AOS≤ 1.8D85=1.35 mm.
For nonwoven fabric, dynamic flow conditions, the retention criterion is
AOS≤ 0.5D85=0.375 mm.
2. The clogging criterion is
AOS≥ 3D15, when Cu>3.
When Cu≤3, select the geotextile with the maximum AOS value based on the retention
criterion.
3. The permittivity/permeability criteria are
a. For a critical application, the permeability criterion is
kgeotextile ≥ 10ksoil = 5 × 10-4 m/s
b. For a soil with 15% to 50% passing the No. 200 sieve, the permittivity criterion is
Ψ ≥ 0.2 sec-1
48

According to AASHTO M288 Class 2:
1. The retention criteria are

a. For 15 to 50% passing 0.075 mm sieve, AOS≤0.25 mm.
b. For cohesive soils with a plasticity index greater than 7, AOS≤0.3 mm
2. The permittivity criterion is
a. For 15 to 50% passing 0.075 mm sieve, Ψ ≥ 0.2 sec-1
Based on the comprehensive considerations, the minimum requirement for the hydraulic properties
of the prefabricated composite drainage board are summarized in Table 22.
Table 22. Required hydraulic properties of the prefabricated composite drainage
Property Test Method Requirement

Permittivity ASTM D4491 >0.2 sec-1
AOS ASTM D4751 ≤0.25mm
For survivability or constructability, it is not recommended to use the requirements from ODOT
CMS 712.09 Type A for abutment/wall drainage application, since these were originally designed
for underdrain and slope drain applications. In an abutment/wall drainage application, the
compaction equipment is usually heavy, and the compaction energy is relatively high. AASHTO
M288 specification Class 2 requirements are recommended (Table 23).
Table 23. AASHTO M288 Class 2 Specification
Property Test Method Minimum Value: N (lb)

Grab strength ASTM D4632/D4632M 700 (157)
Trapezoidal tear strength ASTM D4533/D4533M 250 (56)
Puncture strength ASTM D6241 1,375 (309)
5.2.2 Recommendations of PCDS products
Based on the design specification, a 30-ft conventional abutment retaining native soil backfill
requires a drainage capacity of 14.7 gal/min-ft and a core strength of 10,000 psf with a factor of
safety of 5. The in-situ soil requires a fabric with permittivity greater than 0.2 sec-1 and an AOS
smaller than 0.25 mm. Furthermore, the fabric should meet AASHTO M288 Class 2 strength
requirements. The Product A is recommended (Table 24).
49

Table 24. Comparison listed properties for Product A with design values
Property Manufacturer value Design value

Thickness-core 0.40 in ≥0.4 in
Compressive 16,500 psf 10,000 psf
strength-core
Flow-core/composite 21 gal/min/ft 14.7 gal/min/ft
Fabric permittivity 1.5 sec-1 0.2 sec-1
AOS 0.212 mm ≤0.25 mm
Grab Strength 160 lb 157 lb
Puncture Strength 410 lb 309 lb
5.2.3. Installation plan for the PCDS system
A preliminary plan for the installation of the PCDS system, as included below, is developed for
contractor use.
INSTALLATION OF PCDS SYSTEM
1. General Information
This document provides basic guidelines for the installation of PCDS system at bridge
FRA-2701714. These mainly include the surface preparation, placement, joining,
connections to the pipe outlet, repair, and compaction of backfill.
2. Surface preparation
(1) Before placing the geocomposite material, clear the ground at the site of sharp objects,
cobbles, large rocks and vegetation that will prevent intimate contact between the
surface and the drainage system. Excavate rocks protruding above the surface to level
the ground surface. Backfill any voids and over-excavations with structural backfill
material, as required to construct neat lines. The finished surface should be smooth and
conform to the design grade.
(2) The concrete surface of the abutment and wing walls against which the geocomposite
drain is to be placed shall be free of soil, debris and excessive irregularities that will
prevent continuous contact between the concrete surface and the drain material.
3. Placement
(1) Unroll geocomposite directly onto the prepared surface. Dragging geocomposite
material on ground will not be permitted. Stretch the geocomposite to remove any
creases or wrinkles. Do not expose geocomposite to the elements for longer than 5 days
between laydown and cover. The single fabric surface shall be in contact with the
backfill material.
(2) Install the geocomposite in horizontal courses, with the first course resting on the top
of the footing.
(3) The geocomposite shall be in direct contact with the wall and secured with concrete
nails longer than 2 inches, together with appropriate washers or wood battens of no less
than 4 square inches in area. The spacing of the nails shall be no more than 4 feet in
50

both directions. There shall be at least one horizontal row of nails in each course of
geocomposite.
(4) At the bend corners along the back face of the abutment, use separate courses whenever
possible. Avoid bending the geocomposite in the inner corners. Make sure the channels
inside the geocomposite are not blocked by bending and joining.
(5) Perforations shall be made in the core where the wall drain will lay against a weep hole
but note that the fabric shall not be damaged.
4. Joining
(1) Horizontal seams shall be formed by flapping the extra geotextile fabric extending from
the upper geocomposite course and lapping it over the top of the fabric on the next
lower course. The fabric flap shall be securely fastened to the lower fabric by means of
a continuous strip of 3-inch-wide, waterproof plastic tape.
(2) As an alternative method of splicing, either horizontally or vertically, rolls of
geocomposite drain material may be joined together by turning back the fabric at the
roll edges and interlocking the cuspated sheets approximately two inches. For
overlapping in this manner, the fabric is then lapped over and tightly taped beyond the
seam with the three-inch-wide, waterproof tape. Interlocking of the core shall always
be in the direction of the water flow.
(3) To prevent soil intrusion, all exposed edges of the geocomposite drainage core shall be
covered by tucking the 4-inch fabric lap over and behind the core edge. The bottom,
side, and top edges of the geocomposite shall be covered with a suitable cap formed by
folding a 6 in. flap or a 12-in.-wide strip of geotextile over the edge and securing with
a continuous application of 3-in.-wide plastic tape.
5. Connection to the pipe outlet
(1) The same drainage collector pipes as shown in the plan sheet can be used.
(2) Place the first course of geocomposite on top of the drainage pipe. Use the extra filter
fabric from the drainage board to wrap the pipe and secure the fabric on the back side
of the board continuously using tape, as shown below.
(3) It is suggested to place a thin layer of granular material around the pipe.
51

5.3. Installation of the drainage systems
5.3.1. Installation of the PCDS
The installation of the PCDS at the two abutments of bridge FRA-270-1714 basically followed the
installation instructions as suggested in Section 5.2.3. Minor modifications were made by the
contractors: the contractors installed the drainage boards in vertical columns instead of horizontal
courses. Wood battens and nails were used to secure the adjacent columns on to the wall (Figure
15). Three workers were involved in the installation: one for placement and alignment, one for
nailing, and the third for lift operations.
(a) (b)
(c) (d)
Figure 15. Construction of the drainage systems. (a) The installation of the PCDS at the
forward abutment of FRA-270-1714. (b) The installation of the collector pipe underneath the PCDS.
Note that the drainage pipe was wrapped with fabric and embedded in porous aggregates.
5.3.2. Installation of the traditional drainage layer
The traditional drainage layer at the forward abutment of FRA-270-1716 was installed in stages
with the backfilling process. First, the geotextile fabric was laid down on the ground and the
collector pipes were aligned at the bottom corner of the abutment and wing walls (Figure 15d).
52

The first layer of porous backfill aggregates were then dumped on top of the fabric. The aggregates
were manually trimmed to a layer with an approximate width of two feet measured from the wall
and an approximate lift of one foot. The fabric would be then lapped onto the top of the aggregates
and the first lifts of backfill were then applied. After one lift of backfill, the fabric was then flapped
back onto the backfill side and another lift of drainage aggregate was placed. This process was
repeated until the drainage layer reached the bridge seat elevation. The placement of the drainage
aggregates repeatedly interrupted the soil backfilling, leading to a longer overall time required for
the backfilling process. Furthermore, as shown in Figure 15d, the backhoe ran directly onto the
fabric, and this might cause damage to the fabric. The staged process also made it difficult to
control the thickness of the drainage layer.
The research team also visited bridge ATB-20-21.43 in Ashtabula County to observe the
installation of the traditional drainage system. Slightly different from the process used at
FRA-270-1716, the geotextile fabric was held at a distance of two feet away from the back face of
the abutment, with hanger rebar fixed on the top of the abutment (Figure 16a). After one lift of
backfill was compacted, the drainage aggregates were then dumped into the “trench” between the
wall and the fabric, which was achieved by shaking a backhoe bucket full of aggregates (Figure
16b). Two to three workers were needed for the installation of backfill: one backhoe operator and
one or two coordinators. Since the backhoe operator cannot precisely control the backhoe bucket,
the coordinator should give the operator instructions regarding when and where to “shake” the
bucket to unload the aggregates so as to fill the “trench.” The uncertain nature of the “shaking
process” might also result in uneven filling and material waste, since not all the materials were
dumped precisely into the trench (Figure 16c). Although the thickness of the drainage layer can be
better controlled using this method, the dumped drainage aggregates could not be compacted. The
process was also tedious and time-consuming. It took about five minutes to finish “dumping” a
single backhoe bucket. It took about two days to finish backfilling an area along the wall that was
five feet deep and about 100 feet long.
53

~2feet
Fabric
(a) (b)
(c)
Figure 16. (a) The layout for the trench and fabric. (b) The installation (dumping) process.
(c) An installation that was not well-controlled.
5.4. Planning and installation of the instruments
5.4.1. Overview
The Forward (East) Abutment of FRA-270-1714 and Forward (East) Abutment of FRA-270-1716
were instrumented to monitor and evaluate the drainage performance of the two systems. Three
different types of instruments were installed: a piezometer, a tiltmeter and an H flume.
Piezometers: A Geokon Model 4500S standard piezometer was selected to measure the ground
water pressure in the backfill. If a drainage system works properly, excess pore water pressure will
54

be effectively dissipated, and the pressure should be kept at very low levels at all times. The
transducer uses a pressure-sensitive diaphragm with a vibrating wire element attached to it. The
diaphragm is welded to a capsule that is evacuated and hermetically sealed. Fluid pressures acting
upon the outer face of the diaphragm cause deflections of the diaphragm and changes in the tension
and frequency of the vibrating wire. The changing frequency is sensed and transmitted to the
readout device by an electrical coil acting through the walls of the capsule. Piezometers incorporate
a porous filter stone ahead of the diaphragm, which allows the fluid to pass through but prevents
soil particles from impinging directly on the diaphragm.
Figure 17. Geokon Model 4500S standard piezometer.
Tiltmeters: A Geokon Model 6350 vibrating wire tiltmeter was selected to monitor the lateral
movement of the top of the abutment. The information collected from the tiltmeter can be used to
predict the status of the soil pressure behind the wall (at-rest, active or passive). This information
also helps in the evaluation of the health condition of the wall itself.
Figure 18. Geokon Model 6350 Vibrating Wire Tiltmeter.
Because it has adjustable brackets, the tiltmeter can be easily mounted on either a vertical or a
horizontal surface. When at rest in a vertical configuration, a pendulous mass inside the sensor,
under the force of gravity, attempts to swing beneath the elastic hinge on which it is supported,
but it is restrained by a vibrating wire. As the tilt increases or decreases, the mass attempts to
rotate beneath the hinge point and the tension in the vibrating wire changes, altering its
vibrational frequency. This frequency can be measured by readout devices.
H flume with water level logger: Two 0.5 H flumes, purchased from Virtual Polymer Compounds,
LLC, were used to directly measure the flow rate of the water coming out from the collector pipe.
The size was determined by the expected maximum flow rate in the field; 0.5 H flumes are capable
of measuring the flow rate in the range of 0.18–149 gal/min. The purchased flumes only had the
measurement section as shown in Figure 19a. An approach section was later designed and bolted
55

to the original flume section. A pipe connector was also added to connect the flume and the
drainage outlet in the field (Figure 19b). These additional sections were made of aluminum, which
is light but robust. The top of the approach section was covered with an acrylic sheet to prevent
ingress of external debris and to ensure that the measured flow was solely due to the drainage water
rather than direct precipitation.
Original flume section

Connector
Approach section
(a) (b)
Figure 19. (a) The purchased H flume. (b) Modified flume with approach section and connector.
A WL 16 water level logger (Global Water, Texas), which is a submersible pressure sensor with a
self-contained data logger, was used to determine the water level inside the flume. The water level
can then be converted to flow rate based on the calibration curve developed in the lab. To facilitate
water level measurement, a stilling well was attached to one side of each flume. A stilling well is
a chamber connected to the main flow in the flume through an inlet, typically a 2” (5.08 cm) round
coupling. As the stilling well is isolated from the main flow in the flume, the liquid surface in the
stilling well is still. By dampening out the non-flow related oscillations of the main flow, the
stilling wells provide an excellent area in which to mount secondary flow meters. The floor of a
stilling well extends below the zero elevation of the flume (typically extending 3” or 7.62 cm
below the floor of the flume). This depression results in standing water in the stilling well at all
times – even when the flume is experiencing no flow.
56

5.4.2 Instrumentation layout and installation
At each abutment, one tiltmeter and six piezometers were installed.
Tiltmeter
Elevation 2 (around 898’) 6ft
Elevation 1 (around 884’) Piezometers
9ft 6ft 3ft
(a)
(b)
(c)
Figure 20. The layout of the piezometers and tiltmeters. (a) Elevation view.
(b) Plan view at FRA-270-1714. (c) Plan view at FRA-270-1716.
57

Piezometers: The piezometers were installed at two elevations during the backfill process:
Elevation 1 was located at the bottom of the abutments and wing walls, at an elevation of 884 ft.
Four piezometers were installed at Elevation 1 at each bridge. Three piezometers were aligned
perpendicularly to each abutment (Figure 20): P1-1, P1-3 and P1-2 at FRA-270-1714 and P2-1,
P2-2 and P2-3 at FRA-270-1716. The perpendicular distance from the back face of the abutment
for piezometers P1-1, P1-3 and P1-2 were 3 ft, 9 ft and 18 ft, respectively; the perpendicular
distance for piezometers P2-1, P2-2 and P2-3 were 4.5 ft, 9 ft and 18 ft, respectively. One
additional piezometer was installed near the wing wall at each bridge: P1-4 at FRA-270-1714 and
P2-4 at FRA-270-1716; the distance of the piezometers from the wing wall was 6 ft at both bridges.
Elevation 2 was located at the halfway point of the abutment, at an elevation of 898 ft. Two
piezometers were installed at Elevation 2 at each bridge: P1-5 and P1-6 at FRA-270-1714 and
P2-5 and P2-6 at FRA-270-1716. They were located at a distance of 6 ft from either the abutment
or the wall; detailed locations are shown in Figure 20. Note that the backfill behind the wing walls
was sloped. Based on the final grades shown in the plan sheets, the embedment depths of P1-6 and
P2-6 were 6 ft and 7 ft, respectively.
To install a piezometer, the tip was first placed in a filter fabric bag filled with clean, uniformly
graded sand. The sand in the bag was then saturated with water, in order to saturate the piezometer
tip. The cables for all the piezometers were protected with 2-in PVC conduits and routed to a
common location where the data logger was installed. Before soil backfilling, a thin layer of sand
was placed on top of the sensors and cables to protect them from damage during compaction.
Tiltmeters: The tiltmeters were installed at the front face of the back wall after the bridge
superstructure was completed. The tiltmeter at FRA-270-1714 was designated as P1-7 and the one
at FRA-270-1716 was designated as P2-7. The cables were routed to the data loggers.
Piezometers
Figure 21 Installation of the monitoring system. (a) Installation of the piezometers at Elevation 1 at
abutment FRA-270-1714. Note that the piezometers and the cables (protected with conduits)
were embedded in sand prior to backfilling. (b) Installed tiltmeter.
At each abutment, the cables from the six piezometers and one tiltmeter were routed to a single
data logger (LC-2×16, Geokon) (Figure 21). More details regarding the locations of the sensors
and data loggers, as well as the cable routes, can be found in the marked version of the plan sheets
in Appendix F.
58

Flumes and water level sensors: The flume has a connector which can be connected to the non-
perforated CPP pipe at the drainage outlet as shown in the plan sheet (Figure 22a). Shown in Figure
6 is the forward abutment of bridge FRA-270-1716. For bridge FRA-270-1714, the outlet position
was changed from the initial design due to the fact that the original outlet position was a little
higher at elevation than at the bottom of the footing. A revised plan (Figure 22b) shows where the
outlet pipe turns at the end of the wing wall and terminates at ES station 6879+50. This change
required an additional 300ft of 6-in drainage pipe, which had to be installed at somewhat flat grade
(0.33%). This change of outlet location at FRA-270-1714 may have affected the quantity of the
water flow measured by the flume. More discussion of this is included in Section 5.5.4. The final
positions of the monitoring flumes are highlighted in Figure 22 on the following page.
59

Connect to the flume
(a)
Flume
(b)
Figure 22. The position of flumes installed at the drainage outlets. (a) Plan sheet details for the forward
abutment at FRA-270-1716. (b) Landscape plan showing the final locations of the flumes.
60

Water level sensor in side
the stilling well
Water level sensor in side
Data logger
the stilling well Data logger
Figure 23. Flumes installed at (a) FRA-270-1714 and (b) FRA-270-1716.
For accurate measurement, a flume should be leveled on a base of concrete or suitable building
materials. To install the flumes at the sites, a concrete slab was first poured at the outlet location,
and the flume was then leveled on top of the slab, connected to the drainage pipe, and sealed with
sealant. The flume was then fixed in place using zero-shrink grout. The stilling well was then filled
with water, and a water-level sensor was inserted into the stilling well. A pipe assembly was
designed to protect the data logger. Figure 23 shows the installed flumes. Drawings that illustrate
the field installation details can be found in Appendix F.
5.5. Data collection and analysis
All Geokon vibrating wire sensors (piezometers and tiltmeters) were connected to Geokon data
loggers. The data sampling interval was set to 15 minutes. The flume water-level sensors had self-
contained data loggers, and the sampling interval was set to 15 minutes. Data was collected
(downloaded) during monthly site visits. Since the end of November 2016, when the data loggers
were installed, a total of 12 site visits have been made. In addition to regular data collection trips,
site visits were also made when the data showed abnormal trends or the site experienced accidents.
Two major accidents occurred on the site that caused the damage of the monitoring system at
FRA-270-1716 and a loss of data. Two technical operational accidents resulted in the loss of data
at FRA-270-1714. A site visit log summarizing the site visit activities and observations is provided
in
61

Table 25. The data availability for each sensor type and duration are summarized in Table 26.
The data from the vibrating wire sensors has been consistent and reliable since installation.
However, the flume water level sensors were found to be not ideal for use in the field. In this
section, the data from piezometers and tiltmeters were analyzed. The flume water-level data is only
briefly discussed.
62

Table 25. Site visit log
Date Activity Note

Location of the bridges were located. Natural soil
4/14/16 First site visit.
samples were taken.
Observation of PCDS
6/11/16 Forward abutment at FRA-270-1714.
installation.
Installation of Piezometers
6/13/16 Eight piezometers were installed.
at Elevation 1.
Installation of Tiltmeter at
8/11/16
FRA-270-1716.
Installation of data loggers The cables at FRA-270-1716 were found to be
11/29/16
and flumes Day 1. damaged from previous construction activities.
Installation of data loggers The cable for the tiltmeter at FRA-270-1714 was
12/1/16
and flumes Day 2. found damaged.
First data collection.
12/20/16 Installation of tiltmeter at The tiltmeter at FRA-270-1714 was fixed
FRA-270-1714.
The flume at FRA-270-1716 had water flowing in it,

continuously draining at around 11:30AM. Water
decreased and was dripping at around 3:00PM.
Data collection. Data logger
1/17/17 There was no flowing water in the flume at
maintenance.
FRA-270-1714, but sediments on the flume bottom
were obvious, indicating historical flow events.
The Geokon data logger at FRA-270-1716 was
damaged from flooding, and data was lost.
The pipe assembly at FRA-270-1716 was burned

Site reconnaissance after and a sensor was damaged.
fire accident on site on Jan. New PVC pipe was installed.
2/7/17 23, 2017. At the time, the water depth in the flume at
Downloaded data for FRA-270-1716 was about 1 cm, while there was no
FRA-270-1714. flowing water in the flume at FRA-270-1714.
A new data logger at FRA-270-1716 was installed.
63

Date Activity Note
The distance of the new sensor tip at FRA-270-1716

Data collection. Installation
to the flume bottom was 4.2 cm.
of replacement water-level
2/23/17 Water-level sensor at FRA-270-1714 was brought
sensor at FRA-270-1716.
back for recalibration. Multiple freeze-thaw cycles
Data logger maintenance.
might induce zero drifting of the sensor.
Reinstalled the flume water

Distance from sensor tip to flume bottom was about
3/1/17 level sensor at
3.1 cm.
FRA-270-1714.
Changed battery for Geokon logger at
FRA-270-1714. Data from March 1 to April 1 was
Data collection. Data logger lost due to data collection settings.
4/16/17
maintenance. The data collection settings of both data loggers
were changed from ‘Append data to existing file’ to
‘Create new file then append data’.
Changed battery for flume water level sensor logger

at FRA-270-1716.
The flume cover at FRA-270-1716 was damaged
Data collection. Data logger from falling stones.
5/7/17
maintenance.
No flowing water during visit, but sediments left on
flume bottom indicated historical drainage events.
Flumes was cleaned.
No flowing water during visit, but sediments left on

flume bottom indicated historical drainage events.
Data collection. Data logger
6/18/17 Data wrapping incident occurred again and the
maintenance.
Geokon logger data (05072017-06132017) at FRA-
270-1714 was lost
From previous data analysis, it was found that the

Data collection. Took the thermal couples of the sensors were not activated
6/27/17 flume sensors back for therefore the data was not corrected for temperature
service. change.
Uploaded proper Geokon logger settings since
Data collection. Changed battery for Geokon loggers.
7/25/17 Reinstallation of the two The initial flume sensor readings were 5.6cm at 75F
flume sensors. at FRA-270-1714.
64

Table 26. Summary of data availability and reliability
Duration Data availability and reliability

FRA-270-1714 FRA-270-1716
From To Piezometers Tiltmeter Flume Piezometers Tiltmeter Flume
11/29/16 12/20/16 x ⍻ x ⍻
12/20/16 1/17/17 ⍻ x x ⍻
1/17/17 2/7/17 ⍻ x x x
2/7/17 2/23/17 x x
2/23/17 3/1/17 x x ⍻ x x ⍻
3/1/17 3/11/17 x x ⍻ x x ⍻
3/11/17 4/1/17 x x ⍻ ⍻
4/1/17 4/16/17 ⍻ ⍻
4/16/17 5/7/17 ⍻ ⍻
5/7/17 6/18/17 x x ⍻ ⍻
6/18/17 6/27/17 ⍻ ⍻
6/27/17 7/25/17 x x
Note: indicates that the data is both available and reliable.

x indicates that the data is not available.
⍻ indicates that the data is available but not reliable.
5.5.1 Additional Weather Data
Efforts were made to correlate the soil water pressure and flume flow rate to precipitation data.
Precipitation data was extracted from a reliable weather data source, the Automated Surface
Observing System (ASOS), which has an online portal for data extraction (available at
https://mesonet.agron.iastate.edu/request/download.phtml?network=OH_ASOS). ASOS provides
hourly precipitation data. The data from a weather station closest to the site should be used. The
weather station selected in this study is located at Ohio State University airport (Columbus, Ohio),
which is about 3 miles from the site.
After processing the data, it was found that the precipitation data does not reflect small rain events,
as indicated by the weather condition records. For example, On April 14, 2017, there was a light
rain in the region, but the rain gauge at the OSU Airport station gives zero precipitation. Therefore,
in addition to the ASOS data, the rain events from the weather records available at the Weather
Underground website (https://www.wunderground.com/history/airport/KOSU/) is also included in
the plots. Weather Underground not only provides weather reports for most major cities across the
world on its website but also provides a platform for observations from members with automated
personal weather stations (PWS). In this study, weather events at a weather station located in the
Ohio State University airport were extracted.
65

5.5.2 Soil water pressure data
The variation of water pressure with time for FRA-270-1714 and FRA-270-1716 are shown in
Figure 24 and Figure 25, respectively.
The water pressure at FRA-270-1714 generally varied from –3kPa to 3kPa, which was relatively
low. Negative pressure indicates an unsaturated state. The soil water pressure can be changed for
many reasons. Water pressure increases mainly due to infiltration and the subsequent rise of the
groundwater table. It can decrease due to evapotranspiration and drainage. Infiltration is the
movement of water from the soil surface into soil. The subsequent movement of infiltrated water
in the unsaturated zone of a soil, or the redistribution of the water, can involve exfiltration and
evaporation, capillary rise, recharge and drainage. Capillary rise is the upward movement of the
water from a saturated zone to an unsaturated zone above. Recharge is the downward water
movement from an unsaturated zone to the saturated zone below. Infiltration and redistribution
depend critically on the material properties and hydraulic properties of soils.
The groundwater table typically changes seasonally. Generally, the groundwater table is highest
in springtime when precipitation is greatest and evaporation and plant uptake are low. During
summer and fall seasons, the evaporation and plant uptake are high and may affect infiltration, and
recharge becomes low, causing the groundwater table below the land surface to become lower. At
the subject site region, the water table is generally lower than the bottom elevation of the abutments.
Although no dedicated water level observation wells were installed, observations from the
abandoned pull box at FRA-270-1716, which has no bottom plate and connects directly with the
backfill soil, indicates that the groundwater table can reach the abutment bottom level (Elev. 884
ft) in early spring. It is highly possible that the groundwater table never goes appreciably higher
than the bottom of the abutment. This is also evidenced by the low pressure readings from sensors
embedded at the bottom levels (P1-2, P1-3 and P1-4).
It can also be observed from Figure 24 that the soil water pressures tend to increase after
precipitation events, indicating infiltration. The infiltration process may take as long as 5 days. In
winter and spring, the pore water pressure decreases mainly due to drainage, since evapo-
transpiration is low. If no precipitation events occur for a number of days, drainage continues and
the water pressure goes below zero, indicating an unsaturated state for the soil. An exception was
in winter, when snow pack melted. A close look at the data between January 29 and February 2,
2017, reveals that no precipitation events occurred during this period, but water pressures increased
sharply at FRA-270-1716. This was attributed to the melting of the snow pack on site. It is also
noted that during winter and spring, the water pressure at upper locations (P1-5 and P1-6) was
higher than that at lower locations (P1-2, P1-3 and P1-4). In summer times, the trend was reversed.
This can be attributed to the relatively high evapotranspiration rate during the summer.
Figure 24b shows that the soil temperatures at the bottom (P1-2, P1-3 and P1-4) are relatively
consistent and are not affected much by the air temperature (P1-7 from the tiltmeter that is exposed
to the ambient environment). Since sensors P1-5 and P1-6 are close to the ground surface, their
temperatures are lower than sensors at the bottom and can change with air temperature.
66

Data from FRA-270-1716 (Figure 25) showed the same trend as that from FRA-270-1714. The
water pressure at FRA-270-1716 also fluctuated between –3 kPa and 3 kPa.
(a)
(b)
Figure 24. Piezometer data at FRA-270-1714. (a) Water pressure and precipitation.
(b) Temperature
67

(a)
(b)
Figure 25 Piezometer data at FRA-270-1716. (a) Water pressure and precipitation. (b) Temperature.
68

Close observations of the water pressure data (Figure 26) for a shorter period of time (from June
15 to July 1, 2017) revealed that:
1) After each precipitation event, pressures in the backfill increased. The pressure also seemed
to have daily variations, although small compared to the response due to precipitation. This
may be attributed to daily temperature change and thermally induced flow as well as the
daily change in the evapotranspiration rates. Infiltration may take several days, and the
change of pressure tended to lag behind the precipitation events.
2) At FRA-270-1714, the pressure at P1-6 was lowest; at FRA-270-1716, the pressure at

P2-6 was lowest. These two sensors were close to the ground surface. In summer,
evapotranspiration (evaporation and transpiration by the vegetative cover) is high.
Moisture was lost faster from the surface layers. The pressure at P2-6 was also lower than
that at P1-6, mainly because P2-6 was embedded at a shallower depth than P1-6, even
though they were installed at the same elevation.
3) The real seepage in the backfill seemed more complicated than is shown in Appendix E.
Based on the flow net in Appendix E, for locations at a same elevation, the water pressure
decreases in the direction toward the wall where the drainage layer is located. However,
from the measured pressure data, this is not the case. At the abutment bottom, P1-2 is
farther from the drainage board than P1-3 (Figure 20), yet P1-3 showed a slightly higher
pressure. P2-3 is farther away from the drainage backfill than P2-1 (Figure 26), yet P2-1
showed a slightly higher pressure. It is possible that local variations in the soil properties
resulted in a preferential flow that is not in the direction perpendicular to the wall. It is also
possible that the installation elevation of the sensors was not as precise as planned.
Overall, all the piezometer readings showed a similar trend, and the pressure changes were limited.
Both drainage systems worked in a satisfactory manner.
69

(a)
(b)
Figure 26. Piezometer data for a short time period (a) Water pressure and precipitation at FRA-270-1714.
(b) Water pressure and precipitation at FRA-270-1716.
70

5.5.3 Tiltmeter data
The variation of tilt angle with time at FRA-270-1714 and FRA-270-1716 are shown in Figure 27
and Figure 28, respectively.
At FRA-270-1714, the tilt angle fluctuated between –0.020° and 0.080°, while the initial reading
was 0.005° (not shown in figure). This means that since the installation of the sensor, the tilt angle
changed between –0.025° and 0.075°, where negative values indicate movement towards the
backfill. In terms of lateral movement, the top of the abutment back wall moved between –0.16
inch to 0.47 inch, which is much smaller than the specified 1.50 inches. The slight changes of
tilting correlated very well to the temperature change. When temperatures increased, the bridge
deck expanded; although expansion joints would accommodate most of the movement, the
abutment was slightly tilted towards the backfill, causing a decrease in the tilt angle. When
temperatures decreased, the bridge deck contracted, causing the abutment to move towards the
bridge deck and therefore increase the tilt angle. A plot for a shorter period of time shows more
clearly that even a small change in temperature would induce a very slight change in the tilt angle.
At FRA-270-1716, the tilt angle fluctuated between –0.290° and –0.180° by June 7, 2017, while
the initial reading was –0.250°. This means that since the installation of the sensor, the tilt angle
changed between –0.040° and 0.070°. In terms of lateral movement, the top of the abutment back
wall moved between –0.250 inch to 0.440 inch. On June 7th, there was a sudden drop of the tilt
angle to –1.740°. Compared to the initial position, the top of the back wall tilted –1.49°, which
corresponded to 9.36 inches if the wall had really moved. It was revealed that on June 6th, the ramp
at FRA-270-1716 was closed, and the temporary asphalt overlay that was placed on the deck after
a fire was milled off. Hydro-demolition on the deck was also performed. On June 7th, the new
concrete overlay on the deck was placed. It is highly possible that the tiltmeter itself was tilted
slightly due to construction activities on the bridge and approaches on June 6 and June 7h. The
data after July 7th showed only slight tilting, which was still correlated very well with the
temperature. The tilt angle range fluctuated between –1.745° to –1.700° (Figure 29). When using
the tilt angle from July 7th (–1.740°) as the new zero position, the abutment tilted between –0.005°
and 0.040°, corresponding to a movement between –0.031 inch and 0.25 inch.
An abutment wall moves in response to external loadings. The major loadings are lateral earth
pressure from the backfill soil and thermal expansion forces from the superstructure. Further, the
two external forces are interdependent. When a wall stands in a vertical position, the lateral earth
pressure is called the at-rest pressure. When the wall moves toward the backfill, the soil is
passively loaded and the reaction pressure is called passive pressure, which is greater than the at
rest pressure. When the wall moves away from the backfill, the backfill tends to move actively
towards the wall and applies an active earth pressure, which is lower than the at-rest value. On the
other hand, when the lateral earth pressure changes, it may also cause movement of the abutment.
For example, the lateral earth pressure may increase due to accumulation of water pressure or due
to heavy surcharge loading; if the change is substantial, the wall may move away from the backfill.
An extreme consequence is that a wall may move so much that damage will occur to the wall or
the superstructure. From the piezometer readings, the change of pore water pressure at the site is
slight (–3 kPa to 3 kPa). Furthermore, the positive pore water pressures were all temporary, since
they would be eventually dissipated by the drainage system. Therefore, it is unlikely that the
movement was due to changes in the pore pressure.
71

(a)
(b)
Figure 27. Tilt angle and temperature data from the tiltmeters (a) FRA-270-1714 and (b) FRA-270-1716.
72

(a)
(b)
Figure 28. Tilt angle and temperature data from the tiltmeters for a shorter period of time:
(a) FRA-270-1714 and (b) FRA-270-1716.
73

Figure 29. Tilt angle and temperature data from tiltmeters at FRA-270-1716 after June 7th.
74

5.5.4 Flume flow level and flow rate data
The flume water-level sensors experienced several freeze-thaw cycles before March 2017, and the
data showed unreasonable trends. When the sensor was frozen, the pressure acting on the sensor
diaphragm increased sharply and exceeded the operational range of the sensor. After recalibration
of the flume water-level sensors, the data was still found to be unreliable (Figure 30). It was finally
discovered that the sensors shipped from the vendor were not qualified, and the temperature
compensation function was missing. As shown in Figure 30, even when no precipitation events
occurred, daily variations of the water level in the flume were recorded. Those variations were due
to temperature changes, but no compensation and correction was made in the sensor output. The
different amplitudes for the two sensors were due to the intrinsic differences in sensor temperature
coefficients.
Figure 30. Flume water levels in the flume at abutments: (a) FRA-270-1714 and (b) FRA-270-1716.
75

It can be noticed that on top of the daily variations, sharp increases in water level were recorded
when there were precipitation events, indicating drainage activities at the flumes. However, it was
difficult to quantitatively calculate the drainage flow rate, since the exact temperature effect was
not quantified. This problem was eventually solved, and the sensors were reinstalled on July 25,
2017. Data has been collected since the reinstallation. Another type of sensor, e.g., ultrasonic depth
sensor, may work better for this application.
Nevertheless, the flumes were also observed visually during site visits. On January 17, 2017, it
was observed that the flume at FRA-270-1716 had water flowing in it, continuously draining at
around 11:30AM (Figure 31a). The water flow rate decreased gradually, and it was dripping at
around 3:00PM. There was no flowing water in the flume at FRA-270-1714, but sediments on the
flume bottom were obvious, indicating historical flow events (Figure 31b). In addition, as
discussed in Section 5.3.2, the location of the final drainage outlet was changed during construction.
The outlet location was far away from the abutment, and an additional drainage pipe of about 300
ft in length was used. Further, the grade was relatively flat (0.33%) for the new route of the
drainage pipes. It was suspected that the amount of drained water was too low to flow out of the
pipe due to the long distance and the flat grade.
Figure 31. Observation of water draining from the flumes on January 17, 2017:
(a) FRA-270-1716 and (b) FRA-270-1714.
76

(a) (b)
(c)
Figure 32. Observation of the flumes on June 18, 2017: (a) FRA-270-1716 and (b) FRA-270-171.
(c) improper termination of the drainage board edges.
Figure 32 shows the flumes on June 18, 2017, when no flow was observed at either of the two
flumes. However, sediments were found on the bottom of both flumes, and it is clear that the
77

sediment quantity was higher at FRA-270-1714. First, this may imply that soil particles intruded
into the drainage system, which underscores the importance of selection of the fabric layer of the
composite drainage board based on the backfill soil. Second, intrusion of the soil may also have
occurred during the installation process. Extra caution should be exercised when dealing with all
joints and edges of the boards. A site reconnaissance revealed that some portions of the top edges
of the composite drainage board were not sealed properly, which may have resulted in soil
intrusion. Third, the use of soils with high clayey content as backfills should be avoided.
5.6. Summary and conclusions
bridges (FRA-270-1714 and FRA-270-1716) were selected for the field monitoring and evaluation
of the installation process and field performance of two drainage system: the traditional two feet
of porous backfill with filter fabric and the prefabricated composite drainage system. The PCDS
at FRA-270-1714 was designed based on site conditions. Construction plan notes were also
developed to guide the installation of the PCDS. The forward abutments and wing walls at these
two bridges were instrumented with piezometers in the backfill, tiltmeters on the backwall, and
flow flumes with water level sensors at the drainage outlets. A third site, ATB-20-21.43, was
selected for additional observation of placing two feet of porous backfill. Major findings include:
• The number of workers required for installation of the two systems is similar. Both systems
require 2 or 3 workers.
• The installation of the traditional system with two feet of porous backfill is cumbersome,
and the system can only be installed in stages with the backfilling process. This results in
a longer overall time for the backfilling process. Further, the construction equipment (e.g.,
backhoe) is prone to damage the filter fabric during construction, and it is difficult to
control the thickness of the drainage layer.
in horizontal or vertical courses. Proper joining of the different courses is important, and
caution should be exercised to properly seal the edges.
• Feedback from the contractors suggested that the installation of the PCDS was much faster
and easier.
dissipated, and the pressure should remain at very low levels at all times. At both
FRA-270-1714 and FRA-270-1716, the soil water pressure generally varied between –3
kPa and 3 kPa, which was relatively low, indicating effective drainage. Soil water pressure
changes with precipitation: it increases gradually after precipitation and decreases
relatively fast after precipitation ceases. The actual drainage paths, though, are complex,
as indicated by the relative water pressure levels among sensors embedded at different
locations.
• Tiltmeters were installed to monitor the lateral movement of the top of the abutment. This
information can be used to predict the soil pressure status behind the wall (at-rest, active
or passive). This information also helps evaluate the health condition of the wall itself. At
78

of the abutment back wall moved between –0.16 inch to 0.47 inch, which is much less than
the specified 1.50 inches. The tilt angle of the abutment at FRA-270-1716 decreased
suddenly on June 7, 2017, in response to construction activities. Otherwise, the tilt angle
varied between –0.040° and 0.070°. The slight changes in tilt correlated very well to the
temperature changes, while no correlation between the pore water pressure change and the
tilt angle was observed.
• Two flumes were used to directly measure the flow rate of water coming out from the
cycles at the two sites and the malfunction of the sensors due to fire accidents and
manufacturer flaws. Nevertheless, visual observations of the flumes during multiple site
visits indicated that water was drained out from both of the abutments, as evidenced by the
sediment trace left by the draining water. Flowing water was observed at FRA-270-1716
on certain dates, but no flowing was observed at FRA-270-1714, where PCDS was installed.
This might be due to the fact that the outlet location at FRA-270-1714 was very far (>300
ft) from the abutment and the drainage pipes were installed at a very flat grade (0.33%),
while the outlet location was set very close (<50 ft) and the drainage pipes were installed
at a steeper grade. It was also realized that direct comparison of quantitative flow rate data
at the two bridge abutments should not be used to compare the performance of the two
drainage systems, since the two abutments had different geometry designs, backfill
conditions (coverage area, volume and final grades) and drainage pipe lengths and grades.
edges of the composite drainage board were not sealed properly, which may have resulted
in soil intrusion. This underscored the importance of quality control during the installation
of PCDS.
79

CHAPTER 6. COST ANALYSIS
The overall cost involved in construction of a drainage system includes material cost, labor,
equipment and other incidentals. In this chapter, three types of cost analysis were included. First,
the unit PCDS material costs were surveyed among different manufactures. Second, the cost
involved in a traditional system with two feet of porous backfill drainage as well as PCDS obtained
from a state-wide survey and ODOT’s historic bidding records were analyzed. Third, the actual
costs for drainage systems constructed at the two abutment sites were analyzed.
6.1 Unit material cost of commercially available PCDS
The unit price data obtained from the bidding history is the overall construction cost. The research
team also acquired material cost information from manufacturers/distributors of the PCDS
products. Table 27 summarizes the collected unit material cost data, and it will be updated when
additional responses are received.
The unit price of PCDS products ranges from $0.8/SF to $1.7/SF. The price varies from
manufacturer to manufacturer. For products from the same manufacturer, the price typically
increases based on the quality or the compressive strength of the product. It should be noted that
the material cost (Table 27) is much lower than the overall construction cost (Figure 34) for PCDS
systems. If the average unit material price ($1/SF) is used, the material cost only accounts for 6%
to 50% of the overall construction cost. The additional cost involves the handling, installation,
labor and excavation/refill. This is consistent with the survey result, as shown in Figure 7.
6.2 Cost analysis based on survey and bidding history
From the survey of Ohio county engineers, it was found that the typical unit price for porous
backfill with filters is $55 to $80 per cubic yard. ODOT also uses $70 per cubic yard in the budget
estimation process for retaining walls (ODOT, 2013). In the survey, only limited data was collected
for the PCDS systems, and it is necessary to conduct a more detailed cost analysis. In this section,
efforts are made to collect material cost for various PCDS products. Cost analysis is conducted
based on ODOT’s historical bidding records and awarded contracts.
The research team collected the bidding histories and awarded contracts for ODOT’s retaining
wall and abutments projects by searching the website of the ODOT Office of Planning
(http://www.dot.state.oh.us/Divisions/Planning/Estimating/Pages/default.aspx). By searching the
bidding history related to item 518E21200 (porous backfill with filter fabric) over the past several
years, it is found that the unit price of porous backfill with filter fabric typically varies from $55
to $100 per cubic yard. Figure 33 shows the histogram for the bidding average unit price of porous
backfill with filter fabric from ODOT bidding history in 2013. It is found that new projects are
always awarded to the lowest estimate for the entire project. But the awarded bidder does not
necessarily propose the lowest price on each specific item, such as item 518E62600. It is noted
that the unit price proposed by different bidders sometimes varies significantly. Therefore, the
bidding average is used in the following analysis.
80

Table 27. Unit Price of Commercially Available PCDS products
Manufacturer Product Name Price ($/SF)

®
DELTA -DRAIN 2000-4 0.982
DELTA®-DRAIN 2000-6 0.980
Cosella Dorken
DELTA®-DRAIN 6200 1.125
DELTA®-DRAIN 9000 HI-X 1.429
Panel Drains 0.900
Drain Away 20 0.750
Drainage Products, Inc.
Drain Away 50 0.800
Drain Away 65 1.080
Enkadrain 3611R 0.635
COLBIND
Enkadrain 3811R 0.840
J-Drain 300/302 n/a
J-Drain 400/420 n/a
J-Drain 400XL/420XL n/a
J-DRAIN J-Drain 700/720 n/a
J-Drain 740 n/a
J-Drain 760 n/a
J-Drain 780 n/a
SITEDRAIN Sheet 90 0.690
American Wick Drain, Inc.
SITEDRAIN Sheet 180 0.825
TERRADRAIN 200 n/a
TERRADRAIN 600 n/a
Terrafix Geosynthetics Products
TERRADRAIN 620 n/a
TERRADRAIN 900 n/a
US 200/220DB 0.925
US 400/420DB 1.000
US Fabrics
US 700/720DB 1.000
US 740DB 1.075
Mirafi G100N 0.990
TENCATE Mirafi Mirafi G100W 1.700
Mirafi G200N 1.520
n/a = not available
81

Figure 33. Histogram of bidding average unit price of porous backfill with filter fabric
from ODOT bidding history in 2013.
Item 518E62600 refers to PCDS. It is found that in the past 15 years, only 11 sections of
wall/abutments involve the use of PCDS as a structural drain. The history records for these projects
are attached in Appendix D. Among the eleven data points, five are from a recently awarded project
(ODOT project No. 150030), and the other six are from bidding histories from 2001, 2006, 2011,
and 2013.
18
PCDS
Historical
Porous Backfill with Fabrics (PBF)
16 bidding
Exponential fit for PCDS (y=2.32+22.7*exp(-0.00188*x)
data
Exponential fit for PBF (y=4.96+4.09*exp(-0.00239*x)
14
PCDS
This Study Porous Backfill with Fabrics
12
Unit Price ($)
10
0
0 2000 4000 6000 8000
Plan Quantity (SF)
Figure 34. Comparison of unit price between PCDS and porous backfill with filter fabric.
The data from this study is also included.
It is found that the unit price of the drainage systems generally varies with the total plan quantity.
Figure 34 compares the unit price of PCDS and porous backfill with filter fabric. Only eleven data
82

points are available for PCDS. About 200 data points are available for porous backfill, and the data
is from the bidding average unit price of porous backfill with filter fabric (item 518E21200) from
the ODOT bidding history in 2013 only. It should be noted that the methods of payment for these
two systems are different. The PCDS is measured by area (in SF) and the porous backfill with
fabric is measured by volume (in CY). Since the specified thickness of the porous backfill in CMS
518 is 2 ft, it is straightforward to express the unit price of porous backfill using the same unit as
that for PCDS. From Figure 34, it can be observed that the unit price of both systems decreases
with project scale. The data for the two systems can be fitted using exponential curves, as shown
in Figure 34. It can be concluded that porous backfill with fabrics are more cost-effective for small
projects (<1000 SF) and PCDS systems are more economical for large projects (>1000 SF). For
large projects, the cost of the PCDS system is approximately 47% of that of porous backfill system.
Hsuan and Koerner (2000) also compared the cost for PCDS systems and porous backfill systems
based on conceptual cost estimation. It was concluded that the PCDS system cost about 35% less
than the traditional porous backfill systems.
Since there are only 11 data points for the PCDS systems, the conclusions drawn above should be
viewed with caution. The reason for the decreasing unit cost with project size may be related to
the facts that (1) the material cost will decrease for large quantity, and (2) when the project is large,
the efficiency for installation will generally increase. The former may not be the dominant factor
since the material cost only accounts for a small fraction of the total cost and the discount level for
material cost is relatively low. Therefore, the increased efficiency or reduced unit price for other
cost components such as handling, installation, and excavation/refill may play a greater role.
Based on the limited bidding records, it seems that PCDS systems cost more than porous backfill
systems for small projects. Again, the unit material cost for the PCDS systems are generally lower
than that for the porous backfill. This may imply that the efficiency of installing PCDS systems
for small projects are lower than that of porous backfill systems. A possible reason for the low
efficiency is that contractors’ experience in installing PCDS systems, since they are less frequently
adopted and are not specified by ODOT. Specifications provided by ODOT can stimulate the use
of PCDS system in the future. With increasing use of PCDS systems, contractors may accumulate
related experience and improve their efficiency. Therefore, the authors envision that the cost for
PCDS systems will be lower in the future.
6.3. Cost analysis on the drainage systems at FRA-270-1714 and FRA-270-1716
The costs involved in installation of the two drainage systems are summarized and compared in
this section. It should be noted that the contractor was concerned about damaging the board when
compacting the natural soil backfill with compacting equipment such as “padfoot” rollers. It was
then proposed to place two feet of granular material (ODOT 703.16.C Type C, or ODOT 703.17
for ODOT 304) next to the drainage board. Note that these granular materials are different from
the currently specified “porous backfill” for drainage purposes. The former has a different
gradation and is less costly than the latter. ODOT approved the proposal, but this resulted in
additional cost for the granular material.
83

6.3.1 Abutment geometries
Table 27. Summary of Geometries of the select bridge abutment and the needed quantities of
backfill
Area (SF) /
Bridge No. Section Height* (ft) Width (ft)
volume (CY)
RAbut
30 74
5,859 SF/
FRA-270-1714 FAbut 30 75 (434 CY)
RWing (31.67, 18.42) 26
FWing (30.83, 16.5) 26
RAbut 30 47
FAbut (25.59, 29.53) 76.25
6,249 SF/
FRA-270-1716 RWing-1 (18.35, 30.86) 46
(460 CY)
RWing-2 (28.58, 17.7) 28
FWing (3, 25.58) 67
*If a section top has an even elevation, its height is indicated with a single number. If a section is sloped,
its height is indicated with two numbers: the first represents the height at the highest point and the second
represents the height at the lowest point.
6.3.2 Material cost
For the conventional drainage system with two feet of porous backfill:
According to the cost information provided by the contractor, the unit material cost for the porous
backfill aggregate was $16.25/ton, or $32.5/CY, or $2.4/SF of wall (considering the thickness of
two feet of the drainage layer)
For the drainage board system:
Materials for the drainage board system include: 1) drainage board, and 2) optional replacement
for the two feet void of non-perform backfill. The drainage board costs $1.05/SF and ODOT
703.17 for Item 304 granular material costs $11.75/ton, or $23.5/CY, or $1.74/SF of wall.
Since the replacement granular material is optional,

1) the upper bound for the unit material cost is $1.05/SF+$1.74/SF=$2.79/SF
2) the lower bound for the unit material cost is $1.05/SF
6.3.3 Labor and equipment cost
For the conventional drainage system with two feet of porous backfill:
The total unit cost for constructing the porous backfill is $62/CY, which includes the material cost,
as well as labor and equipment cost. Since the material cost is known ($32.5/CY), it can be back-
calculated that the unit labor and equipment cost for constructing conventional porous backfill
drainage layer is about $29.5/CY, or $2.2/SF of wall.
84

For the drainage board system:
In this particular project, since the original bid item was for porous backfill drainage system for
both abutments, the contractor priced the labor and equipment to install the drainage board the
same as the labor and equipment to install the porous backfill. But according to the feedback from
the contractor, it took significantly less time to construct the drainage board system than the
traditional system, and labor and equipment costs are also less for the drainage board system.
However, the contractors and the site engineers did not track the exact installation time. Based on
the feedback, a reasonable estimate of the time savings is 25%. Note that labor and equipment
costs should be less for the drainage board installation if it is a bid item.
Therefore, the unit labor and equipment cost for the drainage board system is
$2.2/SF×0.75=$1.65/SF
6.3.4 Unit total cost
For the conventional 2” porous backfill drainage system:
Unit total cost = Unit material cost + Unit labor and equipment cost
= $2.4/SF + $2.2/SF
= $4.6/SF
Alternatively, it can be calculated by
Unit total cost =$62/CY, consider a thickness of two feet, the unit total cost is equivalent to
Unit total cost =62×2/27=$4.6/SF
For drainage board system:
Upper bound Unit total cost = Unit material cost + Unit labor and equipment cost
= $2.79/SF + $1.65/SF
= $4.44/SF
Lower bound Unit total cost = Unit material cost + Unit labor and equipment cost
= $1.05/SF + $1.65/SF
= $2.7/SF
The cost analysis result based on the feedback from the contractor is summarized in Table 28.
85

Table 28. Cost comparison of different drainage systems
Drainage System Material Labor and Total Savings

equipment
Two feet porous backfill $2.4/SF $2.2/SF $4.6/SF -
Drainage board w/ backfill replacement) $2.79/SF $1.65/SF $4.44/SF 3.5%
Drainage board w/o backfill replacement) $1.05/SF $1.65/SF $2.7/SF 41%
If a layer of two feet of granular backfill (not the same porous backfill) was required for a
composite drainage board system, the cost savings compared to the traditional two feet of porous
backfill is not significant (3.5%). The only benefit is that the construction of the drainage board
system is less time-consuming and less cumbersome.
If a layer of two feet of granular backfill is not required for composite drainage board systems, a
savings of 41% can be achieved. This number is close to that from the cost analysis based on
ODOT historical bidding history (47%) (Section 6.2) and analysis from other studies (35%)
(Hsuan and Koerner, 2000).
6.4. Summary and conclusions
• Based on state-wide survey and ODOT historical bidding data, the typical unit total price
for porous backfill with filters is $55 to $80 per cubic yard, or $4/SF to $5.9/SF of wall.
The material cost typically accounts for about 50% of the total cost.
• Based on limited ODOT historical bidding data, the unit total price for PCDS varied in a
very wide range between $2/SF and $18/SF. The unit total price decreases with the increase
of the plan quantity. For relatively large plan quantity (>1,000 SF), the cost for PCDS is
more competitive than the porous backfill system.
• Direct comparison between the cost of drainage systems at FRA-270-1714 and
FRA-270-1716 showed that:
was not monitored, but feedback from the contractors suggested that the
material (ODOT CMS Item 703.17 for Item 304) between the drainage board and
the native soil backfill. Considering this replacement layer of granular material, the
upper bound of the unit total price for the PCDS was $4.44/SF, or a 3.5% cost
savings compared to the traditional system.
86

the PCDS of $2.7%, or a 41% savings compared to the traditional system.
• The cost savings by using PCDS without replacement of the granular layer at
FRA-270-1714 is consistent with the historical bidding record (47%) and findings from
other researchers (35%).
87

CHAPTER 7. SPECIFICATION FOR PREFABRICATED
COMPOSITE DRAINAGE SYSTEM
Currently, ODOT specifies “Drainage of Structures” in ODOT CMS Item 518, which is developed
for a porous backfill drainage system only. One of the objectives of this research is to develop
specifications for prefabricated composite drainage systems (PCDS) for bridge abutments and
retaining walls.
In this chapter, recommendations are made on the required properties of the PCDS, and a draft
specification is developed for the PCDS by referring to the specifications of other state DOTs, the
specifications from the PCDS manufacturers and the findings from this research.
7.1. Recommendations
7.1.1 Recommended specification for the core structure of PCDS
As discussed in Section 3.1.2, the minimum required core strength can be calculated based on the
lateral pressure theory. The required core strength increases with the wall height, sassuming that
the total surface and live load is 650 psf and the properties of backfill soils are as suggested in
ODOT BDM 204.6.2 (Table 29).
Table 29. Assumptions for Calculation of the Lateral Earth Pressure
Backfill Soil Type Unit Weight Friction Angle K0 Ka Kp Surcharge

Compacted sand, 120 pcf 30° 0.5 0.33 3 650 pcf
silty sand
It is recommended that different coefficients of earth pressure should be used for different types
of walls. For conventional walls and abutments with jointed superstructures, the at-rest coefficient
of earth pressure should be used. For integral or semi-integral abutments, coefficient of earth
pressure between passive and at-rest condition should be used. The coefficients can be determined
considering soil property (i.e., friction angle) and wall heights. The recommended specification
for the compressive strength of the core structure is obtained by applying a safety factor to the
allowable core strength.
The minimum ultimate core strengths for walls with different heights can be determined (results
are shown in Table 30).
88

Table 30. Core Strength Requirements for Walls with Different Heights
Wall Height H (ft) 10 20 30 40 50

Allowed wall movement Δ (in) 1.5 1.5 1.5 1.5 1.5
Δ/H 0.0125 0.00625 0.00417 0.00313 0.0025
K0 (Section 3.1.2) 0.5 0.5 0.5 0.5 0.5
Kmove (Section 3.1.2) 2.27 1.67 1.42 1.27 1.18
Allowable core strength for conventional 925 1,525 2,125 2,725 3,325
walls (psf) σ R = γ HK0 + σ q K0
Allowable core strength for integral 4,207 5,095 6,027 6,938 7,817
abutments (psf) σ R = γ HK move + σ K move
Proposed ultimate core strength for 4,625 7,625 10,625 13,625 16,625
conventional walls (FS=5), (psf)
Proposed ultimate core strength for integral 12,622 15,285 18,082 20,813 23,451
abutments (FS=3), (psf)
The required in-plane flow capacity/rate of the PCDS can be determined using the method
described in Section 3.1.2. It also changes with wall depth. Assuming a permeability of 5 × 10-5
m/s that is typical for silty sands, a safety factor of 3, and an overall reduction factor of 2.2, the
ultimate in-plane flow rate required for abutment/walls with different heights can be calculated
(Table 31).
Table 31. Required PCDS In-Plane Flow Rate for Walls with Different Heights
Wall Height (ft) 10 20 30 40 50

Recommended in-plane flow rate* (gal/min/ft) 5 10 15 20 25
*under 3,600 psf and i =1.0
7.1.2. Recommended specification for the geotextile fabrics of the PCDS
In ODOT CMS Item 518, geotextile fabrics, Type A, 712.09 is specified for the porous backfill
drainage system. Item 712.09 Type A geotextile fabric is originally specified for underdrains and
slope drains applications. Such drainage systems are typically embedded at a very shallow depth.
AASHTO M288 covers six geotextile applications: subsurface drainage, separation, stabilization,
permanent erosion control, sediment control and paving fabrics. When site conditions are unknown,
engineers can refer to AASHTO M288 Survivability Default Classes for guidance. Survivability
is divided into three classes, with Class 1 being the most severe and Class 3 being the least severe.
Each class is then subdivided according to fabric type (woven and nonwoven). This offers a choice
of non-woven geotextiles or woven geotextiles to be considered for each class. If an elongation is
not specified, keep in mind that the user will always pick the cheapest available option. For
subsurface drainage and erosion control, a woven fabric is typically more expensive
than the nonwoven option. For certain applications, hydraulic properties are included in AASHTO
M288.
89

When comparing the values specified by ODOT with those specified by other state DOTs and
AASHTO M288 Class 2, it was found that ODOT CMS 712.09 Type A geotextile has much lower
required values than AASHTO M288 Classes 2. The tensile strength of ODOT Type A is also
lower than that specified by most other DOTs. In abutment/wall drainage application, the
compaction equipment is usually heavyweight and the compaction energy is relatively high.
After consulting the AASHTO M288 requirements and the specifications of DOTs in other states,
the requirements from ODOT CMS 712.09 Type A is not recommended for abutment/wall
drainage application, since it was originally designed for underdrain and slope drain applications.
In abutment/wall drainage application, the compaction equipment is usually heavy and the
compaction energy is relatively high. AASHTO M288 specification Class 2 requirements are
recommended.
Table 32. Recommended requirements for geotextile of the PCDS
Requirement,
Property Test Method Unit Percent in situ soil passing 0.075 mm
<15 15 to 50 >50
Permittivity ASTM D4491 sec-1 ≥0.5 ≥0.2 ≥0.1
AOS ASTM D4751 mm <0.43 <0.25 <0.22
Grab Strength N (lb) ≥700 (157)

ASTM D4632/D4632M
Elongation % ≥50
Trapezoidal Tear
ASTM D4533/D4533M N (lb) ≥250 (56)
Strength
Puncture ASTM D6241 N (lb) ≥1375 (309)

Strength
ASTM D4833 N (lb) ≥260 (58)
Ultraviolet
stability
ASTM D4355/D4355M % 50% after 500 h of exposure
(retained
strength)
7.1.3. Recommended specifications for the construction of PCDS systems
Based on a comprehensive literature review, the specifications developed by various state DOTs
and manufacturers were synthesized. The following aspects during construction of PCDS systems
should be specified:
A. Geocomposite Packaging, Handling, and Storage

B. Product Acceptance
90

C. Surface Preparation
D. Placement
E. Joining
F. Connecting to Pipe Outlet
G. Material Backfill
H. Repair
7.2. Draft specification
In this section, draft specifications are developed for the PCDS in abutment/wall applications,
referencing the format of the ODOT CMS. There are two options to specify PCDS in ODOT CMS.
One is to develop a new designated Item XXX, similar to the existing CMS Item 518. The other
is to incorporate PCDS into the CMS Item 518, which is “Drainage system for structures”. In this
report, the former format is proposed.
Another new specification Item 712.XX is also developed to specify the PCDS materials. While
Item XXX specifies the general information and construction procedure. Item 712.XX specifies
the material requirement details for PCDS.
91

7.2.1. Draft Item 712.XX Prefabricated Composite Drainage Boards
712.XX.01 Description
712.XX.02 Materials
712.XX.03 Prequalification/basis of approval
712.XX.01 Description
A prefabricated composite drainage board is used to provide drainage behind bridge abutments,
retaining walls or other structural components. The composite drainage board consists of a flexible
plastic drainage core with a geotextile filter bonded to one side. The polymeric core provides
strength and serves as a secure flow channel for the collected water. The geotextile serves as a
filter and allows water to freely enter the drainage core while retaining soil particles and preventing
clogging. The wall drainage system shall meet or exceed the individual component requirements
for the core and geotextile, described in 712.XX.02 and Tables 712.XX-1 and 712.XX-2. All
numerical values listed in the required property tables shown below represent minimum average
roll values (MARV) per ASTM D 4759 unless indicated otherwise. Values for the weaker principal
direction should be used. Testing shall be performed in accordance with the methods referenced
in this specification. Sampling of lots shall be in accordance with ASTM D 4354.
712.XX.02 Materials
The drainage board shall be prefabricated type including a drainage core with geotextile bonded
to one side. The drainage core material shall consist of a preformed, stable, polymer plastic
material with a cuspated, nippled, or geonet structure. The drainage core shall provide support for
and shall be bonded to the geotextile filter fabric at intervals not exceeding 1-1/8 inches in any
direction. Its preformed structure shall be perforated to allow water to flow freely to the drainage
outlets. If not perforated during manufacture, the core shall be perforated in the field at the weep
hole or other drainage outlet locations. The core shall have at least 14 square inches per square
foot of flat area in contact with the geotextile fabric to support the fabric.
The geotextile fabric shall consist of over 85% of polyesters, polypropylene polyolefins or
polyamides by weight. The polymeric fibers shall be formed into a stable network to ensure the
performance during handling, installation and service life. The geotextile fabric shall be resistant
to chemical attach, rot and mildew. The geotextile fabric shall have no tears or defects, and free of
treatment or coating that would adversely change the hydraulic properties of geotextile after
installation. A minimum 3-inch wide flap/flange of geotextile fabric shall extend beyond both
longitudinal edges of the drainage core. The geotextile fabric shall cover the full length of the core.
The composite drain shall be furnished with an opaque, waterproof wrapping to protect against
extended moisture and ultraviolet exposure prior to installation. The drain shall be furnished with
fittings and an approved method for connecting with outlet pipes or weep holes as shown on the
plans. These fittings shall allow entry of water from the core, but shall not allow intrusion of
backfill material.
The drainage core and the filter fabric should meet or exceed the physical, mechanical and
hydraulic requirements as listed in Table 712.XX-1, and Table 712.XX-2, respectively. For
92

conventional abutment and retaining walls, use Type A drainage core. For integral or semi-integral
abutment, use Type B drainage core.
Table 712.XX-1. Physical, mechanical and hydraulic requirements for the core of the
prefabricated composite drainage board.
Type A: Drainage core for conventional abutments and retaining walls
Required value,
Test conventional abutment/wall height in m (ft)
Property Unit
Method
<3 (10) 3 (10) to 9 (30) 9 (30) to 15 (50)
ASTM
Thickness mm (in) 10 (0.4) to 25 (1.0)
D5199
Compressive ASTM
kPa (psf) 221 (4625) 508 (10625) 796 (16625)
Strength D1621
In-plane flow ASTM l/min/m
62 (5) 186 (15) 310 (25)
rate* D4716 (gla/min/ft )
* Tested under a confining pressure of 172kPa (3,600 psf) and a hydraulic gradient of 1.0.
Type B: Drainage core for integral/Semi-integral abutments
Required value,
Test integral abutment height in m (ft)
Property Unit
Method
<3 (10) 3 (10) to 9 (30) 9 (30) to 15 (50)
ASTM
Thickness mm (in) 10 (0.4) to 25 (1.0)
D5199
Compressive ASTM
kPa (psf) 604 (12,622) 866 (18,082) 1,123 (23,451)
Strength D1621
In-plane flow ASTM l/min/m
62 (5) 186 (15) 310 (25)
rate* D4716 (gla/min/ft )
* Tested under a confining pressure of 172kPa (3,600 psf) and a hydraulic gradient of 1.0.
93

Table 712.XX-2. Strength and hydraulic requirements for the fabric of the prefabricated
composite drainage board.
Requirement,
Property Test Method Unit percent in situ soil passing 0.075 mm
<15 15 to 50 >50
Permittivity ASTM D4491 sec-1 ≥0.5 ≥0.2 ≥0.1
AOS ASTM D4751 mm <0.43 <0.25 <0.22
Grab
ASTM N (lb) ≥700 (157)
Strength
D4632/D4632M
Elongation % ≥50
Trapezoidal
ASTM
Tear N (lb) ≥250 (56)
D4533/D4533M
Strength
Puncture ASTM D6241 N (lb) ≥1375 (309)

Strength
ASTM D4833 N (lb) ≥260 (58)
712.XX.03 Prequalification/basis of approval

For prequalification, supply samples of the finished product from production to the ODOT for
testing and evaluation. Submit separate samples of the composite material, the core material, and
the filter fabric material. All samples must be 10 feet long by nominal roll width. Submit a
manufacturer’s or independent laboratory test report addressing the properties as listed in this
specification.
When it becomes available, test results for the product will be accepted from the AASHTO
National Transportation Product Evaluation Program (NTPEP) without submitting samples.
Forward an official copy of the test report to ODOT for evaluation. Prequalification will be based
on satisfactory compliance of NTPEP results with this specification.
If the ODOT test or NTPEP results comply with 712.XX.02, the name of the product will be placed
on the approved PCDS product list. No PCDS will be used on ODOT projects unless it has been
prequalified.
Verification samples will be taken at the rate of 2 per source, per district, per year. Failure of any
verification sample to comply with this specification may be cause for removal of the manufacturer
from the approved product list.
94

7.2.2. Draft ITEM XXX Prefabricated Composite Drainage Systems for Structures
XXX.01 Description
XXX.02 Fabrication
XXX.03 Materials
XXX.04 General
XXX.05 Prefabricated Composite Drainage System
XXX.06 Pipe
XXX.07 Scuppers
XXX.08 Excavation
XXX.09 Method of Measurement
XXX.10 Basis of Payment
XXX.01 Description. This work consists of constructing prefabricated Composite Drainage

Systems for Structures
XXX.02 Fabrication. Fabricate scuppers according to Item 513. Select a fabricator that is at least
pre-qualified at level SF. The Department will base final acceptance of all fabricated members on
the Engineer’s approval that the fabricated items can be successfully incorporated into the
structures. Submit mill test reports for structural steel, steel castings, bronze, and sheet lead
certified according to 501.06.
XXX.03 Materials. Furnishing materials conforming to:
Scuppers, structural steel and cast steel ................................................ 513

Metal pipe ............................................................................................. 707
Plastic pipe .......................................................................... 707.33, 707.45
Other metals .......................................................................................... 711
Prefabricated Composite Drainage System ...................................... 712.XX
Reinforced thermosetting resin pipe ................................................ 707.80
Furnish pipe specials of a grade at least as high as the type of pipe specified.
XXX.04 General. As shown on the plans, connect all parts to new or existing sewers or other
outlets.
When installing to superstructure, take into account the deflection of spans under full dead load.
XXX.05 Furnishing and Installing Prefabricated Composite Drainage System
A. Furnishing. Furnish PCDS materials in rolls, or in another acceptable manner, wrapped

with a suitable protective covering to protect the fabric from mud, dirt, dust, debris, harmful
ultraviolet light, direct sunlight or temperature greater than 140˚F. Furnish tapes for the
sealing, seaming and connecting details referenced in this Item. Any tape designed for
underground use is acceptable provided it offers a strong bond that will not deteriorate over
time in typical or anticipated subsurface conditions. Furnish fittings or accessories
provided by the manufacturer if available.
95

B. Preparation. Clear ground site of sharp objects, cobbles, boulders and vegetation that will
prevent intimate contact between the surface and the drainage system. The concrete surface
of the walls or abutments against which the geocomposite drain is to be placed shall be free
of soil, debris and excessive irregularities that will prevent continuous contact between the
concrete surface and the drain material.
C. Placement. Unroll geocomposite directly onto the prepared surface. Dragging

geocomposite material on ground will not be allowed. Stretch geocomposite to remove any
creases or wrinkles. Do not expose geocomposite to the elements for longer than 5 days
between installation and backfilling. The single fabric surface shall be in contact with the
backfill material.
The geocomposite material shall be constructed in horizontal or vertical courses. The

geocomposite shall be in direct contact with the wall and secured with approved adhesives
or no less than 2-inch concrete nails along with approved washers or wood battens of not
less than 9 square inches. The adhesive or fasteners shall not affect the drainage area and
the downward flow within the core. The spacing of concrete nails shall be as directed by
the Engineer, but shall not be more than 3 feet apart, both horizontally and vertically. When
nails are utilized, there shall be at least one horizontal row of nails in each horizontal course
of geocomposite, or there shall be at least one vertical column of nails in each vertical
course of geocomposite.
D. Splicing and covering. Horizontal or vertical seams between courses shall be formed by
utilizing the flap of geotextile fabric extending from one course and lapping over the flap
on that of the next course. The overlapped fabric flaps shall be securely fastened by means
of a continuous strip of 3-inch-wide, waterproof plastic tape.
Where splices are necessary at the end of a geocomposite roller panel, a 12-inch-wide
continuous strip of geotextile fabric may be placed, centering over the seam and
continuously fastened on both sides with 3-inch-wide, waterproof plastic tape.
As an alternative method of splicing, either horizontally or vertically, rolls of geocomposite

drainage material may be joined together by turning back the fabric at the roll edges and
interlocking the cuspates approximately two inches. For overlapping in this manner, the
fabric is then lapped over and tightly taped beyond the seam with the 3-inch-wide,
waterproof tape. Shingle the core and fabric in the direction of water flow.
To prevent soil intrusion, all exposed edges of the geocomposite drainage core shall be
covered by tucking the fabric lap over and behind the core edge. Alternatively, a 12-inch-
wide strip of fabric may be utilized in the same manner, taping it to the exposed fabric
8 inches in from the edge with a continuous strip of 3-inch-wide, waterproof, plastic tape
and folding the remaining 4 inches over and behind the core edge. Caps (bottom, top or
end) provided by the manufacturers can also be used according to manufacturers’
instructions.
96

All seams, splices, bottom caps, top caps, and end caps shall be constructed so that backfill
material cannot enter the geocomposite during or after construction.
E. Connecting to Weep Holes and Pipe Outlet. Construct pipe outlets concurrently with the
geocomposite. Connect the geocomposite to the pipe outlets as shown on the plans or as
directed by the engineer. A positive outlet for the water in the geocomposite drain shall be
maintained at all locations.
Weep holes shall not be sealed or made ineffective by the wall drain material. Use weep
hole fittings provided by the manufacturer when available and following manufacturer’s
instructions. Or, if the core of the geocomposite wall drain is not perforated, perforations
shall be made in the core where the wall drain will lay against a weep hole or other drainage
outlet. Alternatively, a hole shall be made in the drainage core at the weep hole location.
Subsequently, an approved weep hole fabric cover extending at least 4 inches from the
edge(s) of the hole shall be securely fastened to the soil side of the wall drain by 3-inch-
wide plastic tape or contact adhesive applied continuously around its periphery. When
making perforations or holes on the core, the fabric shall not be damaged in any way.
Require the manufacturer to supply outlet fittings that transition between the drains and the
outlet pipe. Provide smooth-lined or corrugated outlet fittings. Outlet fittings shall be
fastened to the wall drains as directed by the manufacturer and so that backfill materials
cannot enter the system during or after construction. If necessary, to facilitate the rapid and
complete flow of water from the wall drain into the pipe outlet, a portion of the wall drain
core equal to the cross section at the outlet can be removed.
F. Repair. If rip, tear or damaged area on the deployed geocomposite is more than 50 percent
of the width of the roll, the damaged area shall be cut out and the two portions of the
geocomposite shall be spliced as explained above. Other rips, tears or damaged areas on
the deployed geocomposite shall be removed and patched by placing a patch extending 12
inches beyond the edges of the damaged area. The patch shall be secured to the original
geocomposite with cable ties.
If, in the judgment of the Engineer, the damage is not serious enough to warrant removal,
the damaged area may be repaired by overlaying with a piece of fabric, large enough to
cover the damaged area and provide a 6-inch overlap on all sides and taping it in place with
3-inch-wide strips of waterproof plastic tape. Damaged drainage core sections shall be
discarded and replaced. Any geocomposite wall drain damaged during installation shall be
replaced or repaired by the contractor at no additional cost.
G. Material Backfill. Structural backfill operations shall be started as soon as possible after
placing the geocomposite material, but in no case shall the geocomposite material be
exposed to sunlight for more than 14 days after installation. Care shall be taken during the
backfilling operation not to damage the geotextile surface of the drain and to avoid
excessive settlement of the backfill material. Any geocomposite drainage system
component that is damaged during the backfilling operation shall be replaced or repaired
as directed by the Engineer. When placing geocomposite drain on walls, use hand operated
97

compaction equipment or other proper equipment no closer than 1 foot away from the
geocomposite material to compact the backfills
XXX.06 Pipe. For drain pipe leading down from the superstructure, use either galvanized steel
pipe, 748.06, or plastic pipe, 707.45, or reinforced thermosetting resin pipe, 707.80. Provide
specials, elbows, tees, wyes, and other fittings essential for a complete and satisfactory installation
of the same material and quality as the pipe. Construct watertight joints of adequate strength. For
steel pipe, weld joints or use clamp-type couplings having a ring gasket. For plastic pipe, make
joints according to the applicable ASTM standard. For reinforced thermosetting resin pipe, make
joints according to manufacturer guidelines and procedures. Securely fasten the pipe to the
structure with hanger or clamp assemblies that are galvanized according to 711.02.
Place subsurface pipe as shown in the plans. If the plans require collector outlet pipe under the
drainage board, provide plastic pipe conforming to 707.33.
For corrugated metal pipe, perforated specials are not required and the Contractor may make bends
with adjustable elbows conforming to the thickness requirements of the pipe specifications.
XXX.07 Scuppers. Construct secure and watertight connections, including the connections to
adjacent concrete. Provide castings, true to form and dimension. Weld the joints of structural steel
scuppers. Galvanize scuppers according to 711.02.
XXX.08 Excavation. Excavate all material encountered to the dimensions necessary to provide
ample space at least to install pipe or other drainage facility behind abutments and for outlets.
XXX.09 Method of Measurement

The Department will measure Prefabricated Composite Drainage System by the number of square
feet (square meters). Specifically, the work amount is counted by multiplying plan neat line width
by the measured length in final position parallel to installation centerline. No allowance will be
made for overlap, whether at joints or patches. The department will measure pipe specials by the
same method as the pipe. If pipe is by the foot (meter), the Department will measure the pipe along
its centerline.
XXX.10 Basis of Payment

The cost to backfill, if not separately itemized in the Contract, and excavation is incidental to the
drainage facility that necessitates them. The Department will include Prefabricated Composite
Drainage boards for payment. The Department will pay for accepted quantities at the contract
prices as follows:
Item Unit Description
518 Square foot (Square Meter) or Lump Prefabricated Composite Drainage Boards
Sum
518 Foot (Meter) ___ inch (___ mm) ___ Pipe, Including Specials
518 Each Scuppers, Including Supports

518 Pound or Foot (Kilogram or Meter) Trough Horizontal Conductors
518 Pound or Foot (Kilogram or Meter) Pipe Horizontal Conductors
518 Foot (Meter) ___ inch (___ mm) Pipe Downspout Including
Specials
98

CHAPTER 8. SUMMARY, CONCLUSIONS AND
RECOMMENDATIONS
retaining walls and abutment walls. Current Ohio Department of Transportation (ODOT) practice
for drainage of structures (ODOT Construction and Material Specifications Item 518) specifies the
use of a two-foot porous backfill with filter fabric, which has a long performance history. ODOT
is seeking alternative drainage systems that are more cost- and time-effective, durable, and at the
same time, have comparable or superior drainage capability compared to current practice. A
prefabricated composite drainage system (PCDS) is proposed as an alternative drainage method
for bridge abutments and retaining walls.
The goal of this study is to compare the performance of different drainage systems for abutment
and retaining walls and to select the best system. These drainage systems include ODOT’s current
method of using two feet of porous backfill with filter fabric, the prefabricated composite drainage
system, and other promising drainage systems such as ground tire chips. The specific objectives
are to:
1) Develop specifications for the PCDS;
2) Select environmentally sound materials and procedures that are compliant with EPA
rules;
3) Quantify labor and materials savings of using an alternative abutment and retaining
wall drainage system, including the prefabricated composite drainage system; and
To achieve the goals and objectives, the following tasks were taken:
• Task 1: State wide survey among Ohio counties and contractors on the current state of the
practice for the structural drainage
• Task 2: Comprehensive literature review on other promising alternative drainage systems
used in other state DOTs, emphasizing on the commercially available PCDS products,
DOT specifications, and the potential of using tire chips as backfill materials
• Task 3: Laboratory testing and evaluation of select PCDS products to improve the
understanding of the various properties
• Task 4: Field evaluation and comparison of the performance of the two feet porous backfill
drainage system and PCDS
• Task 5: Cost analysis of comparison of the two drainage systems
• Task 6: Develop material and construction specifications for the PCDS
The findings from the above tasks are summarized in this chapter.
99

8.1. Current state of the practice for structure drainage in Ohio
• Based on the survey results, it can be concluded that porous backfill systems are the
predominant drainage systems used for retaining walls and bridge abutments in the state of
Ohio. Alternative drainage systems used in Ohio include low strength mortar fill and PCDS.
The owners (counties) are generally satisfied with the performance of the traditional
drainage system as well as the alternative systems. But the respondents also generally
advocate the usage of PCDS if it can bring benefits such as less cost, labor saving and
simpler construction process.
8.2 Current state of the practice for PCDS

systems were identified and their properties were summarized. The selection of PCDS
• The design methodology for PCDS were developed by referring to existing research
severity of the structures, survivability and the constructability.
state DOTs specified the construction procedures, and seven state DOTs provide approved
product lists
8.3 Laboratory evaluation of select PCDS products
Five PCDS products from different vendors, were acquired for index property testing in the
laboratory. Eight properties of the core or fabric were tested according to corresponding ASTM
standards.
• The eight properties and the standards selected for testing include,
1) In-plane flow at various gradients and pressures of the core, according to ASTM
D4716.
100

6) Puncture Strength of the fabric, according to ASTM D4833.
8) Apparent Opening Size and Equivalent Sieve of the fabric, according to ASTM
D4751.
Some of the methods were later found not ideal for PCDS and recommendations of
• Regarding the in-plane flow of the core, it was observed that:
o The presence of fabric tends to reduce the in-plane flow rate significantly. In this
tested case, the reduction is about 42%.
of 0.25 hours and 100 hours under the confining pressure levels in this study.
o Flow direction (machine direction or cross machine direction) has a moderate effect
on the in-plane flow of the core.
o In-plane flow increases nearly linearly with hydraulic gradient but does not change
much with the tested normal stress levels, which are not sufficient to significantly
deform the core.
o Compared to the manufacture listed values, the tested in-plane flows fall short for
all products when the fabric is bonded. When tested using the core only, the tested
value exceeded the listed value.
manufacturer’s values for the core alone. The differences varied from –25.5% to +12.6%. For
four of the five tested products, the tested values were lower than the manufacturer values. It
is latter realize this may due to the testing method and conditions used. It is recommended that
ASTM D6364 “Standard Test Method for Determining Short-Term Compression Behavior of
Geosynthetics” to be used to test the compressive strength of PCDS, in lieu of ASTM D1621.
under two different levels of normal stress were very close to each other. For the separate core,
three samples were received and the tested thicknesses were slightly less than the manufacturer
values. The thickness of a core is an indirect indicator for the drainage capacity.
• Regarding the strength properties of the fabric, only one product showed significantly lower
values than the listed values among the three products tested. It was found that different
manufacturers test the puncture strength of the fabric using different ASTM standards. ASTM
D6241 is actually more suitable than ASTM D4833 for geotextile materials.
• Regarding the hydraulic properties of the fabric, all the nonwoven products showed smaller
tested permittivity than the listed values. Since permittivity and flow rate of a geotextile
product is correlated and the permittivity is a value normalized by the differential head
specification. For every tested product, the tested AOS was significantly larger than the
listed value.
• The differences showed between the tested and listed values suggest that a factor of safety
should be used in design when selecting PCDS.
8.4 Field implementation and evaluation
bridges (FRA-270-1714 and FRA-270-1716) were selected for the field monitoring and evaluation
of the installation process and field performance of two drainage systems: the traditional two feet
101

of porous backfill with filter fabric and the prefabricated composite drainage system. The PCDS
at FRA-270-1714 was designed based on site conditions. Construction plan notes were also
developed to guide the installation of the PCDS. The forward abutments and wing walls at these
two bridges were instrumented with piezometers in the backfill, tiltmeters on the backwall, and
flow flumes with water level sensors at the drainage outlets. A third site, ATB-20-21.43, was
selected for additional observation of placing the two feet porous backfill. Major findings include:
• The number of workers required for installation of the two systems is similar. Both requires
2-3 workers.
• The installation of the traditional two feet porous backfill system is cumbersome and the
system can only be installed in stages with the backfilling process. This results in a longer
overall backfilling process. Further, the construction equipment (e.g., backhoe) is prone to
damage the filter fabric during construction. It is difficult to control the thickness of the
drainage layer.
in horizontal or vertical courses. The joining of different courses is of great importance,
and caution should be exercised to properly seal the edges.
• Feedback from the contractors suggested that the installation of the PCDS was much faster
and easier.
dissipated, and the pressure should remain at very low levels at all times. At both
FRA-270-1714 and FRA-270-1716, the soil water pressure generally varied between
–3 kPa and 3 kPa, which was relatively low, indicating effective drainage. The soil water
pressure changes with precipitation: it increases gradually after precipitation and decreases
relatively fast after precipitation stops. The actual drainage paths, though, are complex, as
indicated by the relative water pressure levels among sensors embedded at different
locations.
• Tiltmeters were installed to monitor the lateral movement at the top of the abutment. This
information can be used to predict the soil pressure status behind the wall (at-rest, active
or passive), and it can also help in evaluating the health condition of the wall itself. At
of the abutment back wall moved between –0.16 inch to 0.47 inch, much a smaller amount
than the specified 1.50 inches. The tilt angle of the abutment at FRA-270-1716 decreased
suddenly on June 7, 2017, in response to construction activities. Otherwise, the tilt angle
varied between –0.040° and 0.070°. The slight changes in tilting correlated very well to the
temperature changes, while no correlation was observed between pore water pressure
change and tilt angle.
• Two flumes were used to directly measure the flow rate of water coming out from the
cycles at the sites and the malfunction of the sensors due to fire accidents and manufacturer
flaws. Nevertheless, visual observation of the flumes during multiple site visits indicated
that water drained out from the abutments at both sites, evidenced by the sediment trace
left by the draining water. Flowing water was observed at FRA-270-1716 on certain dates,
but no flowing was observed at FRA-270-1714, where the PCDS was installed. This might
102

be due to the fact that the outlet location at FRA-270-1714 was very far (>300 ft) from the
abutment and the drainage pipes were installed at a very flat grade (0.33%), while the outlet
location was set very close (<50 ft) and the drainage pipes were installed at a steeper grade.
it was also realized that direct comparison of quantitative flow rate data at the two bridge
abutments should not be used to compare the performance of the two drainage systems,
since the two abutments had different geometry designs, the backfill conditions (coverage
area, volume and final grades) and drainage pipe lengths and grades.
edges of the composite drainage board were not sealed properly, which may have resulted
in soil intrusion. This underscored the importance of quality control during the installation
of PCDS.
8.5 Cost analysis and comparison
• Based on state-wide survey and the ODOT historical bidding data, the typical unit total
price for porous backfill with filters is $55 to $80 per cubic yard, or $4/SF to $5.9/SF of
wall. The material cost typically accounts for about 50% of the total cost.
• Based on limited ODOT historical bidding data, the unit total price for PCDS varied in a
very wide range between $2/SF and $18/SF. The unit total price decreases with the increase
of the plan quantity. For a relatively large plan quantity (>1,000 SF), the cost for PCDS is
more competitive than the porous backfill system.
• Direct comparison between the cost of drainage systems installed at FRA-270-1714 and
FRA-270-1716 showed that:
was not monitored, but the feedback from the contractors suggested that the
material (ODOT CMS Item 703.17 for Item 304) between the drainage board and
the native soil backfill. Considering this replacement layer of granular material, the
upper bound of the unit total price for the PCDS was $4.44/SF, or a 3.5% cost
savings compared to the traditional system.
the PCDS of $2.7%, or a 41% savings compared to the traditional system.
o The cost savings for using PCDS without replacement granular layer at
FRA-270-1714 is consistent with the historical bidding record (47%) and the
findings from other research studies (35%).
103

8.6 Development of the draft specification
Based on the proposed design methodology, recommendations are made on the required properties
of the PCDS. Draft specifications are developed for the PCDS by referring to the specifications of
other state DOTs, the specifications from the PCDS manufacturers and the findings from this
research.
• The recommended required PCDS properties depend on the abutment/wall type, height and
backfill and in-situ soil properties. Requirement tables for separate core and fabric are
developed.
• Two draft specifications for PCDS were developed, referring to the ODOT CMS format.
One specification discusses material requirements, including the following components:
o 712.XX.01 Description
o 712.XX.02 Materials
o 712.XX.03 Prequalification/basis of approval
The other specification discusses mainly the construction processes, including the
following components:
o XXX.01 Description
o XXX.02 Fabrication
o XXX.03 Materials
o XXX.04 General
o XXX.05 Prefabricated Composite Drainage System
o XXX.06 Pipe
o XXX.07 Scuppers
o XXX.08 Excavation
o XXX.09 Method of Measurement
o XXX.10 Basis of Payment
8.7 Recommendations
Based on the findings from this research, the following are recommended:
• Continue to monitor the performance of the two drainage systems at FRA-270-1714 and
FRA-270-1716 for a longer period of time, so as to evaluate their long-term performance.
• Incorporate the developed draft specifications for PCDS into ODOT CMS. There are two
options to specify PCDS in ODOT CMS. One is to develop a new designated Item XXX,
similar to the existing CMS Item 518. The other option is to incorporate PCDS into CMS
Item 518, “Drainage of Structures.”
104

REFERENCES
Ad Hoc Civil Engineering Committee. (1997). Design Guidelines to Minimize Internal Heating of Tire
Shred Fills. Paper presented at the Scrap Tire Management Council, Washington, DC.
Ahn, I.-S., & Cheng, L. (2014). Tire derived aggregate for retaining wall backfill under earthquake
loading. Construction and Building Materials, 57(0), 105-116. Available at:
http://dx.doi.org/10.1016/j.conbuildmat.2014.01.091
Alaska DOT. (2013). Juneau International Airport (JNU) Runway 8/26 Rehabilitation.
Arizona DOT. (2008). Standard Specifications for Road and Bridge Construction.
AASHTO (2007) LRFD Bridge Design Specifications.
ASTM D6270-08. (2008). Standard Practice for Use of Scrap Tires in Civil Engineering Applications,
ASTM D 6270 ASTM International.
ASTM D6270-98. (2004). Standard Practice for Use of Scrap Tires in Civil Engineering Applications,
ASTM D 6270 ASTM International.
Berg, R. R. (1993). Guidelines for Design, Specification & Contracting of Geosynthetic Mechanically
Stabilized Earth Slopes on Firm Foundation, Federal Highway Administration, Washington, D.C.
California DOT. (2010). Standard Specifications.
California Integrated Waste Management Board (1999) Tire Shreds: Solutions for Civil Engineers.
http://www.calrecycle.ca.gov/Publications/Documents/Tires/21299004.pdf Accessed on Nov. 3, 2017.
California Integrated Waste Management Board (2001) Waste tire marketing guide.
Cecich, V., Gonzales, L., Hoisaeter, A., Williams, J., & Reddy, K. (1996). USE OF SHREDDED TIRES
AS LIGHTWEIGHT BACKFILL MATERIAL FOR RETAINING STRUCTURES. Waste Management
& Research, 14(5), 433-451. Available at: http://dx.doi.org/10.1006/wmre.1996.0043
Cedergren, H. R. (1989). Seepage, Drainage, and Flow Nets, Third Edition, John Wiley & Sons, New
York.
Christopher, B. and A. Zhao (2001). Design Manual for Roadway Geocomposite Underdrain Systems,
Contech Construction Product Inc.
Clough, G. W., and J. M. Duncan. 1991. “Earth Pressures.” Foundation Engineering Handbook, 2nd Ed.
H. Y. Fang, ed. Van Nostrand Reinhold, New York, NY, Chap. 6.
Colorado DOT. Standard Specifications for Road and Bridge Construction.
Downs, L. A., Humphrey, D. N., Katz, L. E., & Rock, C. A. (1996). Water Quality Effects of Using Tire
Chips Below the Groundwater Table: University of Maine, Department of Civil and Environmental
Engineering, Orono, Maine.
Duncan, J.M. and Seed, R.B. 1986 Compaction-Induced Earth Pressures Under K0-Conditions. Journal of
Geotechnical Engineering 112: 1-22.
Duncan J.M., Williams, G.W., Sehn, A.L., and Seed, R.B. 1991 Estimation Earth Pressures due to
Compaction. Journal of Geotechnical Engineering 117: 1833-1847.
Drescher, A., & Newcomb, D. (1994). Development of design guidelines for use of shredded tires as a
lightweight fill in road subgrade and retaining walls: Department of Civil and Mineral Engineering,
University of Minnesota.
105

Geosyntec Consultants. (2008). Guidance Manual for Engineering uses of Scrap Tires: Maryland
Department of the Environment.
Hawaii DOT. (2005). Standard Specifications & Special Provisions.
Holtz, R. D., Christopher, B. D., and Berg, R. R. (1998). Geosynthetic Design and Construction
Guidelines, Ryan R. Berg & Associates, Inc.: 460.
Hoppe, M., and Oman, M. (2013). Use of Tire Derived Products (TDP) in Roadway Construction.
Minnesota Department of Transportation, Research Services Section.
Hsuan, Y.G. and Koerner, R.M. (2000) Prefabricated wall drain system for structures, Commonwealth of
Pennsylvania Department of Transportation Report No. FHWA-PA-2000-028+96-31 (3)
Humphrey. (2006a). Civil Engineering Applications Using Tire Derived Aggregates: Course Notes,
prepared for California Integrated Waste Management Board.
Humphrey. (2006b). Material specifications for use of tire derived aggregate (TDA) in civil engineering
applications - Revised: University of Maine, Orono, Maine.
Humphrey, & James, W. (2010). Lateral Pressures behind Retaining Walls Backfilled with Tire Derived
Aggregate. Paper presented at the Transportation Research Board 89th Annual Meeting, Washington, DC.
Humphrey, Sandford, T. C., Cribbs, M. M., & Manion, W. P. (1993). Shear strength and compressibility
of tire chips for use as retaining wall backfill. Transportation Research Board.
Humphrey, & Sanford, T. C. (1993). Tire Chips as Lightweight Sub-Grade Fill and Retaining Wall
Backfill. Paper presented at the Symposium on Recovery and Effective Reuse of Discarded Materials and
By-products for Construction of Highway Facilities, Denver, Colorado.
Humphrey, Whetten, N., Weaver, J., Recker, K., & Cosgrove, T. A. (1998). Tire Shreds as Lightweight
Fill for Embankment and Retaining Walls Recycled Materials in Geotechnical Applications (Vol.
Geotechnical Special Publication No. 79).
Illinois DOT. (2012). Standard Specifications for Road and Bridge Construction.
JDAA (1995). Engineering use of Geotextiles. Joint Departments of Army and Air Force.
Kansas DOT. (2007). Standard Specifications for State Road & Bridge Construction.
Koerner, R. M., Hwu, B.L. (1989) Geomembrane use in transportation systems. Public Works Magazine
102 (10), 135-139
Koerner, R.M, and Koerner, G.R. (2007) Reduction Factors (RFs) Used in Geosynthetic Design,
Geosynthetic Institute, GSI White Paper #4.
Lee, J., Salgado, R., Bernal, A., & Lovell, C. (1999). Shredded Tires and Rubber-Sand as Lightweight
Backfill. Journal of Geotechnical and Geoenvironmental Engineering, 125(2), 132-141.
doi:10.1061/(ASCE)1090-0241(1999)125:2(132).
Louisiana DOT. (2006). Standard Specifications for Roads and Bridges.
Maine DOT. (2003). experimental construction project ME 00-20, Third-year interim report on
experimental utilization of tire shreds to enhance highway drainage: Maine Department of Transportation
Transportation Research Division.
Moo-Young, H., Sellasie, K., Zeroka, D., & Sabnis, G. (2003). Physical and Chemical Properties of
Recycled Tire Shreds for Use in Construction. Journal of Environmental Engineering, 129(10), 921-929.
doi: doi:10.1061/(ASCE)0733-9372(2003)129:10(921)
Michigan DOT. (2012). Standard Specifications for Construction.
106

Missouri DOT. (1999). Missouri Standard Specifications for Highway Construction.
Mitoulis S., Argyroudis S. and Pitilakis K. (2014) Green Rubberised Compressible Inclusions to Enhance
the Longevity of Integral Abutment Bridges. Second European Conference on Earthquake Engineering
and Seismology, Istanbul Aug. 25-29 2014
Montana DOT. (2014). Standard Specifications for Road and Bridge Construction.
North Carolina DOT. (2012). Standard Specifications for Roads and Structures.
NYDOT (2008). New York State Department of Transportation, EI 08-006, Standard Specifications
Section 207 – Geosynthetics.
NYDOT (2012). New York State Department of Transportation, Geotechnical Design Manual, Chapter
18: Geosynthetic Design: pp 18-11 ~ 18-57.
ODOT (1995). Construction and Material Specifications. Ohio Department of Transportation (ODOT)
ODOT (2007). Bridge Design Manual. Ohio Department of Transportation, Columbus, Ohio. Available at
http://www.dot.state.oh.us/Divisions/Engineering/Structures/standard/Bridges/Pages/BDM2007.aspx
O'Shaughnessy, V., & Garga, V. K. (2000). Tire-reinforced earthfill. Part 3: environmental assessment.
Canadian Geotechnical Journal, Vol. 37(1).
Rixner, J.J., Kraemer, S.R., and Smith A.D. (1986), "Prefabricated Vertical Drains." U.S. Department of
Transportation, Federal Highway Administration, Research, Development, and Technology, Vol. I:
Engineering Guidelines, Report No. FHWA/RD-86/168. Available at:
www.fhwa.dot.gov/bridge/geopub.htm
Rowe, R., & McIsaac, R. (2005). Clogging of Tire Shreds and Gravel Permeated with Landfill Leachate.
Journal of Geotechnical and Geoenvironmental Engineering, 131(6), 682-693. doi:
doi:10.1061/(ASCE)1090-0241(2005)131:6(682)
Rubber Manufacturer Association (2015) 2015 U.S. Scrap Tire Management Summary
https://www.ustires.org/system/files/scraptire_summ_2015_05_2017_Final_USTMA.pdf Accessed on
Nov. 3, 2017.
Scrap Tire Management Council. (1997). Scrap Tire Use/Disposal Study: Rubber Manufactures
Assiciation, Washington, DC.
South Carolina DOT. (2000). Supplemental Specifications: Geocomposite Wall Drain.
South Carolina DOT. (2010). SCDOT Geotechnical Design Manual.
South Dakota DOT. (2004). Standard Specifications for Roads & Bridges.
Tennessee DOT. (2008). Special Provision Regarding Retaining Wall.
Tweedie, J., Humphrey, D., & Sandford, T. (1998). Tire Shreds as Lightweight Retaining Wall Backfill:
Active Conditions. Journal of Geotechnical and Geoenvironmental Engineering, 124(11), 1061-1070. doi:
doi:10.1061/(ASCE)1090-0241(1998)124:11(1061)
107

Twin City Testing. (1990). Environmental Study for the Use of Shredded Waste Tires for Roadway Sub-
grade Support: Waste Tire Management United, Minnesota Pollution Control.
Utah DOT. (2012). Standard Specifications for Road and Bridge Construction.
Van Dyke, Stacy (2014). Comparison of CBR and Pin Puncture Strength Testing Used in the Evaluation
of Geotextiles. Theses and Dissertations. Paper 773.
Virginia DOT. (2007). Road and Bridge Specifications.
VTrans (Vermont Department of Transportation) (2008), Integral Abutment Bridge Design Guidelines.
Montpelier, Vermont
White, D., S. Sritharan, M. Suleiman, M. Mekkawy and S. Chetlur (2005). Identification of the Best
Practices for Design, Construction, and Repair of Bridge Approaches Iowa State University: 379.
Xiao, M., Bowen, J., Graham, M., & Larralde, J. (2012). Comparison of Seismic Responses of
Geosynthetically Reinforced Walls with Tire-Derived Aggregates and Granular Backfills. Journal of
Materials in Civil Engineering, 24(11), 1368-1377. doi:10.1061/(ASCE)MT.1943-5533.0000514
108

Appendix A. Survey of Drainage System for Bridge Abutment and
Retaining Walls in Ohio
109

110

111

112

Appendix B. Commercially Available PCDS Products
Table 33. Properties of the core of PCDS products
Core Composite
Geocomposit
Supplier Core e Water
Manufacture Product Name Core
Location Compressiv Flow Rate@
Thickness Permittivity
e Strength hydr. grad.
(inch)
(psf) 1.0
(gal/min/ft)
DELTA®-DRAIN 2000 HI-
0.4 11000 18 1.70 sec-1
X
Canada (Has 0.4 15100 18 1.70 sec-1
X
Cosella Dorken distributors
in Ohio) 0.4 15100 18 1.70 sec-1
X
0.4 18000 18 2.10 sec-1
X
Panel Drains 0.625 4500 5 --
Drain Away 20 0.25 10800 9 --
Drainage --
Drain Away 50 Connecticut 0.4375 15000 16
Products, Inc.
Drain Away 65 0.4375 21000 18 --
Edge and Strip Drains 0.625 4500 5 --
Enkadrain 3611R North 0.45 >30000 22.5 1.80 sec-1
COLBIND
Enkadrain 3811R Carolina 0.45 >30000 20.2 1.80 sec-1
J-Drain 300/302 0.25 40000 8.5 --
J-Drain 400/420 0.4 15000 21 --
J-Drain 400XL/420XL 0.4 15000 21 --
J-DRAIN J-Drain 700/720 Georgia 0.4 21000 23 --
J-Drain 740 0.4 21000 23 --
J-Drain 760 0.4 21000 23 --
J-Drain 780 0.4 21000 23 --
SITEDRAIN Sheet 90 0.25 9000 12.5 2.80 sec-1
North
American Wick
SITEDRAIN Sheet 110 Carolina and 0.44 11000 18 2.80 sec-1
Drain, Inc.
Nevada
SITEDRAIN Sheet 114 0.44 11000 18 2.10 sec-1
113

Core Composite
Geocomposit
(inch)
(psf) 1.0
(gal/min/ft)
SITEDRAIN Sheet 184-T 0.44 18000 21 1.00 sec-1
SITEDRAIN Sheet 186-W 0.44 18000 21 2.30 sec-1
TERRADRAIN 200 0.25 10800 12.5 --
Terrafix TERRADRAIN 600 0.4 15000 17 --
Geosynthetics Maine
TERRADRAIN 620 0.4 15000 17 --
Products
TERRADRAIN 900 0.4 18000 21 --
US 200/220DB 0.4 11000 18 --
US 400/420DB Cincinnati, 0.4 15000 21 --
US Fabrics
US 700/720DB Ohio 0.4 21000 23 --
US 740DB 0.4 21000 23 --
CCW MIRADRAIN® 2000 0.25 10800 12.5 --
CCW MIRADRAIN® 6000 0.4 15000 17 --
CCW MIRADRAIN® Texas (Has --
CARLISLE distributors 0.4 16500 17
6000XL
in Ohio) --
CCW MIRADRAIN® 6200 0.4 15000 17
CCW MIRADRAIN® --
0.4 16500 17
6200XL
GSE PermaNet300 (6oz/yd2) 0.3 50000 -- 1.50 sec-1
GSE PermaNet300 --
0.3 50000 1.00 sec-1
(10oz/yd2)
GSE GSE TenDrain300 (6oz/yd2) Texas (Has 0.3 80000 -- 1.50 sec-1
distributors
Environmental GSE TenDrain300 (8oz/yd2) 0.3 80000 -- 1.30 sec-1
in Ohio)
GSE PermaNet330 --
0.33 60000 1.00 sec-1
(10oz/yd2)
Mirafi G100N 0.4 18000 12.5 1.90 sec-1
TENCATE
Mirafi G100W Georgia 0.4 18000 18 2.10 sec-1
Mirafi
Mirafi G200N 0.4 18000 12.5 1.90 sec-1
Hydroduct 200 Indiana, 0.44 15000 17 --
Hydroduct 220 Illinois, etc 0.44 15000 17 --
114

Core Composite
Geocomposit
(inch)
(psf) 1.0
(gal/min/ft)
Hydroduct 225 0.4 21000 21 --
GRACE
Construction Hydroduct 660 0.44 18000 17 --
Product --
Hydroduct Coil 600 0.44 1.0 9000 17 80
Aquadrain 10X 0.25 10800 11 --
Aquadrain 15X 0.43 15000 20 --
Aquadrain 15XP 0.43 15000 20 --
Aquadrain 18H 0.43 18000 16 --
CETGO Illinois
Aquadrain 20H 0.43 21000 16 --
Aquadrain 30H 0.25 30000 9 --
Aquadrain 100BD 1 10000 / --
Aquadrain G20 0.25 40000 7 --
SHEET DRAIN 0.38~0.42 15000 15 --
Greenstreak, St. Louis, 0.21~0.23 --

SHEET DRAIN HS 25000 5.7
Inc. MO 7
0.38~0.42
DECK DRAIN 20000 21.4
4
LINQ Industrial South
LINQ 140 EX 0.7 sec-1
Fabrics, Inc. Carolina
Polyguard
POLYFLOW® 10P TX 0.25 10000
Products, Inc.
Tenax data in
UVB 5065-2 Maryland 0.2 0.3cm/sec
Corporation process
5012 11000 12.5 2.3 sec-1
5012-B 11000 12.5 2.3 sec-1

5035* 15000 17 2.3 sec-1
5035-B* 15000 17 2.3 sec-1
W.R.Meadows
6012 Georgia 11000 12.5 2.1 sec-1
of Georgia
7012 11000 12.5 1.8 sec-1
7055 18000 21 1.8 sec-1
7555* 18000 21 2.3 sec-1
7572 30000 13 2.3 sec-1
9055 18000 21 1.3 sec-1
115

Core Composite
Geocomposit
(inch)
(psf) 1.0
(gal/min/ft)
4500 1.00 2.80 sec-

Typical Range 0.25 1 5 23
80000 1
Note: Highlighted products are not qualified according to the recommended specification
(Section 5.2.1)
116

Table 34 Properties of the fabric of PCDS products
Fabric
Fabric Fabric
Supplier Fabric
Manufacture Product Name CBR grab Filter Water
Location Elongation(AST Tear Apparent
Puncture Tensile Flow Rate
M D-4632-91) Strength Opening Size
Strength Strength (gal/min/ft)
(lbs)
(lbs) (lbs)
DELTA®-DRAIN 2000 HI-X 50% 310 120 50 70 US Sieve 135

DELTA®-DRAIN 6000 HI-X Canada (Has 50% 310 120 50 70 US Sieve 135
Cosella Dorken DELTA®-DRAIN 6200 HI-X distributors 50% 310 120 50 70 US Sieve 135
in Ohio)
24%(MD),10%(C
DELTA®-DRAIN 9000 HI-X 675 365 115 40 US Sieve 145
D)
Panel Drains 50% 70 120 -- -- 130
Drain Away 20 50% 70 115 -- -- 170
Drainage -- --
Drain Away 50 Connecticut 50% 70 115 170
Products, Inc.
Drain Away 65 31% 165 285 -- -- 26
Edge and Strip Drains 50% 70 120 -- -- 130
Enkadrain 3611R North 50% 70 120 50 70 US Sieve 120
COLBIND
Enkadrain 3811R Carolina 50% 70 120 50 70 US Sieve 120
J-Drain 300/302 -- 250 100 -- 70 US Sieve 140
J-Drain 400/420 -- 250 100 -- 70 US Sieve 140
J-Drain 400XL/420XL -- 410 160 -- 70 US Sieve 110
J-DRAIN J-Drain 700/720 Georgia -- 850 250 -- 40 US Sieve 60
J-Drain 740 -- 250 100 -- 70 US Sieve 140
J-Drain 760 -- 410 160 -- 70 US Sieve 110
J-Drain 780 -- 500 205 -- 80 US Sieve 95
117

Fabric
Fabric Fabric
Supplier Fabric
(lbs)
(lbs) (lbs)
SITEDRAIN Sheet 90 65% 35 90 -- 50 US Sieve 190

SITEDRAIN Sheet 116 North 70% 450 160 -- 70 US Sieve 110
American Wick
Carolina and
Drain, Inc. SITEDRAIN Sheet 118 70% 600 205 -- 80 US Sieve 90
Nevada
SITEDRAIN Sheet 184-T 60% 50 145 -- 80 US Sieve 80
SITEDRAIN Sheet 186-W 15% 105 220 -- 45 US Sieve 160
TERRADRAIN 200 -- 65 100 -- 70 US Sieve 140
Terrafix TERRADRAIN 600 -- 65 100 -- 70 US Sieve 140
Geosynthetics Maine
Products
US 200/220DB -- 250 100 -- 70 US Sieve 140
Cincinnati, -- --
US Fabrics US 400/420DB 250 100 70 US Sieve 140
Ohio
US 700/720DB -- 850 250 -- 40 US Sieve 60
118

Fabric
Fabric Fabric
Supplier Fabric
(lbs)
(lbs) (lbs)
US 740DB -- 250 100 -- 70 US Sieve 140

CCW MIRADRAIN® 2000 60% 250 100 -- 70 US Sieve 140
CCW MIRADRAIN® 6000 Texas (Has 60% / 80 / 40 US Sieve 200
CARLISLE CCW MIRADRAIN® 6000XL distributors 50% 410 160 / 70 US Sieve 110
CCW MIRADRAIN® 6200 in Ohio) 60% / 80 / 40 US Sieve 200
CCW MIRADRAIN® 6200XL 50% 410 160 / 70 US Sieve 110
GSE PermaNet300 (6oz/yd2) 50% 435 160 65 70 US Sieve 110
GSE TenDrain300 (6oz/yd2) Texas (Has 50% 435 160 65 70 US Sieve 110
GSE
distributors
Environmental GSE TenDrain300 (8oz/yd2) 50% 575 220 90 80 US Sieve 95
in Ohio)
Mirafi G100N -- 337 111 45 70 US Sieve 140
TENCATE --
Mirafi G100W Georgia 675 200 75 40 US Sieve 145
Mirafi
Mirafi G200N -- 337 111 45 70 US Sieve 140
Hydroduct 200 -- 275 100 -- 100 US Sieve 165
GRACE Hydroduct 220 -- 275 100 -- 70 US Sieve 165
Indiana,
Construction
Hydroduct 225 Illinois, etc -- 725 385 -- 45 US Sieve 160
Product
Hydroduct 660 -- 600 205 -- 80 US Sieve 90
119

Fabric
Fabric Fabric
Supplier Fabric
(lbs)
(lbs) (lbs)
Hydroduct Coil 600 -- 320 115 -- 70 US Sieve 150

70 100 US
Aquadrain 10X 50% 65 100 40 150
Sieve
70 100 US
Aquadrain 15X 50% 65 100 40 150
Sieve
70 100 US
Aquadrain 15XP 50% 65 100 40 150
Sieve
70 100 US
CETGO Aquadrain 18H Illinois 50% 150 200 105 110
Sieve
Aquadrain 20H 16% 135 370 100 70 US Sieve 110
Aquadrain 30H 16% 135 370 100 70 US Sieve 110
70 100 US
Aquadrain 100BD 50% 90 160 65 110
Sieve
70 100 US
Aquadrain G20 50% 65 100 40 150
Sieve
SHEET DRAIN -- -- -- -- -- --
SHEET DRAIN HS -- -- -- -- -- --
DECK DRAIN -- -- -- -- -- --
St. Louis,
Greenstreak, Inc. NEEDLEPUNCHED
MO 60% 60 100 45 70 145
POLYPROPYLENE
Warp
Warp 30% Fill Warp 75
WOVEN POLYPROPYLENE 120 300 Fill 30~50 50
23% Fill 65
200
120

Fabric
Fabric Fabric
Supplier Fabric
(lbs)
(lbs) (lbs)
LINQ Industrial South

LINQ 140 EX 50% 30 120 40 60
Fabrics, Inc. Carolina
Polyguard
POLYFLOW® 10P TX 60% 65 110 100 150
Products, Inc.
Tenax
UVB 5065-2 Maryland -- 85 170 65 70 135
Corporation
5012 65% 65 100 -- 70 165
5012-B 65% 65 100 -- 70 165
5035* 65% 65 100 -- 70 165
5035-B* 65% 65 100 -- 70 165
W.R.Meadows of 6012 70% 75 130 -- 70 150
Georgia
Georgia 7012 70% 90 160 -- 70 110
7055 70% 90 160 -- 70 110
7555* 15% 105 410×220 -- 45 160
7572 15% 105 410×220 -- 45 160
9055 70% 120 205 -- 80 90
40 100 US
Typical Range 15% 70 35 850 80 350 40 115 26 200
Sieve
Note: Highlighted products are not qualified according to the recommended specification (Section 5.2.1)
121

Appendix C. Synthesis on specifications for PCDS systems
1. DESCRIPTION
The prefabricated composite drainage system (PCDS) shall consist of a non-woven geotextile fabric and a
supporting drainage core with the geotextile permanently bonded to the core on one side facing the
backfill (Arizona DOT, 2008; Utah DOT, 2012) or both sides(Alaska DOT, 2013). The geotextile fabric
shall be sufficiently bonded with the core to prevent separation between the two components and thus to
prevent the penetration of the backfill material during installation. The composite polymer core is bonded
to the geotextile at intervals not exceeding 1 1/8 inch (Kansas DOT, 2007) or 1.5 inches (South Carolina
DOT, 2000) in any direction. The geotextile fabric is thermal (heat) bonded or fungicide glue bonded to
the polymeric core. The geocomposite shall be resistant to commonly encountered chemicals and
hydrocarbons, and resistant to ultraviolet exposure. PCDS can be employed in many applications, such as
drainage for landfills, highway pavement, abutment, retaining structures, and so on. The geocomposite
product sheets or rolls shall have a minimum width of 3 feet with a minimum coverage area of 40 square
feet (South Carolina DOT, 2000). The PCDS products shall conform to the departmental approved PCDS
product list (to be developed).
The following covers the material requirements, construction requirements and methods of testing for
geocomposite system primarily used for subsurface drainage behind wall structures. This specification is
less focusing on the subsurface drainage (e.g. roadway edge drain) in pavement structure.
2. MATERIAL
The wall drainage system shall meet the following geocomposite system properties in addition to the
individual component properties of the polymeric drainage core and the non-woven geotextile. All
numerical values listed in the required property tables shown below represent minimum average roll
values (MARV) per ASTM D 4759 (South Carolina DOT, 2000) unless indicated otherwise. Values for
the weaker principal direction should be used. Testing shall be performed in accordance with the methods
referenced in this specification. Sampling of lots shall be in accordance with ASTM D 4354.
A. Geotextile
(a). General Requirements
The geotextile fabric shall consist of over 85% (New York DOT, 2014; South Dakota DOT, 2004) or
95% (Hawaii DOT, 2005) of polyesters, polypropylene polyolefins or polyamides by weight. The
polymeric fibers shall be formed into a stable network to ensure the performance during handling,
installation and service life. The geotextile fabric shall be resistant to chemical attach, rot and mildew.
The geotextile fabric shall have no tears or defects, and free of treatment or coating that would adversely
change the hydraulic properties of geotextile after installation. Geotextile fabric rolls shall be furnished
with an opaque, waterproof wrapping to protect against extended moisture and ultraviolet exposure prior
to installation. A minimum three-inch wide flap of geotextile fabric shall extend beyond both longitudinal
edges of the drainage core. The geotextile fabric shall cover the full length of the core.
(b). Detailed Requirements
The geotextile fabric shall conform to the departmental approved product list (to be developed). The
geotextile fabric shall additionally meet the minimum average roll values (to be developed, example
122

values can be found in Section 5.2.1 and Table 3 ). (Minimum average roll values represent the average
test results for a lot in the weaker direct ion when sampled according to ASTM D 4354 and tested
according to the test method specified.)
B. Drainage Core
The drain core material shall consist of a preformed, stable, polymer plastic material with a cuspated,
nippled, or geonet structure. The drainage core shall provide support for and shall be bonded to the
geotextile filter fabric at intervals not exceeding 1-1/8 inches in any direction (Arizona DOT, 2008). Its
preformed structure shall be perforated to allow water to flow freely to the weep hole drainage outlets. If
not perforated during manufacture, the core shall be perforated in the field at the weep hole drainage
outlet locations. The core shall have at least 14 square inches per square foot of flat area in contact with
the geotextile fabric to support the fabric (Arizona DOT, 2008).
The geocomposite core shall be furnished with an approved method for connecting with outlet pipes or
weep holes as shown on the plans. These fittings shall allow entry of water from the core, but shall not
allow intrusion of backfill material into the core. The core material shall additionally conform to the
following minimum physical and hydraulic requirements (to be developed, example values can be found
in Tables 2, 6 and 7).
3. PREQUALIFICATION/BASIS OF APPROVAL (Based on Kansas DOT, 2007)

For prequalification, supply samples of the finished product from production to the ODOT for testing and
evaluation. Submit separate samples of the core material and the filter fabric. All samples must be 10 feet
long by nominal roll width. Submit a manufacturer’s or independent laboratory test report addressing the
properties in subsection ‘MATERIAL’.
When it becomes available, test results for the product will be accepted from the AASHTO National
Transportation Product Evaluation Program (NTPEP) without submitting samples. Forward an official
copy of the test report to the Bureau Chief of Materials and Research for evaluation. Prequalification will
be based on satisfactory compliance of NTPEP results with this specification.
If the ODOT test or NTPEP results comply with subsection ‘MATERIAL’, the name of the product will
be placed on the approved PCDS product list. No geocomposite drainage system will be used on ODOT
projects unless it has been prequalified.
Verification samples will be taken at the rate of 2 per source, per district, per year. Failure of any
verification sample to comply with this specification may be cause for removal of the manufacturer from
the approved product list.
4. CONSTRUCTION REQUIREMENTS
A. Geocomposite Packaging, Handling, and Storage
The identification, packaging, handling, and storage of the geocomposite wall drain material shall be in
accordance with ASTM D 4873(Arizona DOT, 2008; South Carolina DOT, 2000) and also in accordance
with manufacturer’s recommendations. Geocomposite wall drain shall be furnished in rolls, or in another
acceptable manner, wrapped with a suitable protective covering to protect the fabric from mud, dirt, dust,
debris, harmful ultraviolet light, direct sunlight or temperature greater than 140˚F(Kansas DOT, 2007) /
71˚C(South Carolina DOT, 2000).
Each roll or package shall be labeled or tagged to provide product identification sufficient to determine
the product type, manufacturer, quantity, lot number, roll number, date of manufacture, shipping date, and
the project number and name to which it is assigned. Geocomposite wall drain materials shall be stored on
123

the site or at another location approved by the Engineer in a manner which protects them from the
elements. If stored outdoors, the materials shall be elevated and protected with a light colored, opaque,
waterproof cover. At no time shall the geocomposite wall drain materials be exposed to direct sunlight for
a period exceeding 14 days(Arizona DOT, 2008), or 10 days(South Dakota DOT, 2004).
All material delivered to the project must meet or exceed the physical requirements based on minimum
average roll values that are specified in the material requirement section. The wall drain material shall be
free of defects or flaws which significantly affect its physical properties at the time of delivery and
installation. The geocomposite wall drain will be rejected at the time of installation if it has defects, tears,
punctures, flaws, deterioration, or damage incurred during manufacture, shipment, or storage. Torn or
punctured sections shall be removed or repaired as directed by the Engineer. Any geocomposite wall
drain damaged during manufacture, shipment, or storage shall be replaced by the Contractor at no
additional cost.
B. Product Acceptance
Geotextile shall be packaged in rolls of length and width to meet requirement. Each roll shall be packaged
individually in a sheath, wrapper, or container. Each geotextile fabric roll shall be labeled with the
manufacture’s name, date of manufacture, name of product, product number, geotextile polymer types,
recommended geotextile application, recommended stapling pattern and machine direction. The
manufacture’s certificate of compliance and certified test results on the product, tested within six
months(New York DOT, 2014) or one year(South Carolina DOT, 2000) of submission, shall be
submitted. Acceptance of geotextile fabric shall also conform to the material requirements in this section.
Sampling shall be in accordance with ASTM D 4354 and this section. Samples shall be cut from the
geotextile roll with sharp tools such that a smooth edge is made and tearing of the material is avoided. If
the geotextile seams are to be field sewn, a section of sewn seam shall be provided by the contractor
before installation. The equipment and procedure for the sewn seam sample shall be the same as that will
be used to sew production seams. If the geotextile seams are to be factory sewn, samples of factory seam
shall be provided randomly from the rolls.
C. Surface Preparation
Before placing geocomposite material, clear ground site of sharp objects, cobbles, boulders and
vegetation that will prevent intimate contact between the surface and the drainage system. Excavate rock
protruding above surfaces to neat lines. Backfill voids and over-excavations with structural backfill
material, as required to construct neat lines. Finished surface should be smooth and conform to the design
grade.
For geocomposite applied to subsurface drainage behind wall, the concrete surface of the structure against
which the geocomposite drain is to be placed shall be free of soil, debris and excessive irregularities that
will prevent continuous contact between the concrete surface and the drain material.
D. Placement
When placing geocomposite drains, unroll geocomposite directly onto the prepared surface. Dragging
geocomposite material on ground will not be allowed. Stretch geocomposite to remove any creases or
wrinkles. Do not expose geocomposite to the elements for longer than 5 days (Alaska DOT, 2013; Hawaii
DOT, 2005) between laydown and cover. For single-sided geocomposite drain, the single fabric surface
shall be in contact with the backfill material.
124

When the core of the geocomposite wall drain is not perforated during manufacture, perforations shall be
made in the core where the wall drain will lay against a weep hole or other drainage outlet. When making
these perforations, the fabric shall not be damaged in any way.
To prevent infiltration of the backfill material, the geocomposite material shall be constructed in
horizontal courses with the first course resting on the top of the footing. The geocomposite shall be in
direct contact with the wall and secured with adhesive or 1.5 to 2 inch long(Arizona DOT, 2008), or no
less than 2 inch(Illinois DOT, 2012) concrete nails along with approved washers or wood battens of not
less than 4 square inches in area(Arizona DOT, 2008), or not less than 9 square inches(Illinois DOT,
2012). The adhesive or alignment of the core shall not affect the drainage area or downward flow within
the core. The spacing of concrete nails shall be as directed by the Engineer, but shall not be more than 4
feet(Arizona DOT, 2008), or 3 feet(Illinois DOT, 2012) apart, both horizontally and vertically. When
nails are utilized, there shall be at least one horizontal row of nails in each course of geocomposite.
E. Joining
Horizontal seams shall be formed by utilizing the 3-inches(Alaska DOT, 2013; Hawaii DOT, 2005), or 4-
inch(Arizona DOT, 2008; Illinois DOT, 2012) flap, or 1 foot overlap(Tennessee DOT, 2008) of geotextile
fabric extending from the upper geocomposite course and lapping over the top of the fabric on the next
lower course. The fabric flap shall be securely fastened to the lower fabric by means of a continuous strip
of 3-inch wide, waterproof plastic tape(Arizona DOT, 2008; Illinois DOT, 2012; South Carolina DOT,
2000).
Where vertical splices are necessary at the end of a geocomposite roller panel, an 8-inch wide(Arizona
DOT, 2008), or 12-inch wide(Illinois DOT, 2012) continuous strip of geotextile fabric may be placed,
centering over the seam and continuously fastened on both sides with the 3-inch wide, waterproof plastic
tape.
As an alternative method of splicing, either horizontally or vertically, rolls of geocomposite drain material
may be joined together by turning back the fabric at the roll edges and interlocking the cuspations
approximately two inches. For overlapping in this manner, the fabric is then lapped over and tightly taped
beyond the seam with the three-inch wide, waterproof tape. Inter locking of the core shall always be in the
direct ion of water flow.
To prevent soil intrusion, all exposed edges of the geocomposite drainage core shall be covered by
tucking the 4-inch(Arizona DOT, 2008; Kansas DOT, 2007; South Carolina DOT, 2000) fabric lap over
and behind the core edge. Alternatively, a 12-inch wide strip of fabric may be utilized in the same
manner, taping it to the exposed fabric 8 inches in from the edge with a continuous strip of 3-inch wide,
waterproof, plastic tape and folding the remaining four inches over and behind the core edge.
The bottom, side, and top edges of the geocomposite shall be covered with a suitable cap formed by
folding a 6 in. (150 mm) flap or a 12 in. (300 mms) wide(Illinois DOT, 2012) strip of geotextile over the
edge and securing it in place with a continuous application of contact adhesive or 3 in. (75 mm) wide
plastic tape. All seams, splices, bottom caps, top caps, and end caps shall be constructed so that backfill
material cannot enter the geocomposite during or after construction.
F. Connecting to Pipe Outlet

Connection to pipe outlet systems shall be as shown on the plans or as directed by the engineer. A
positive outlet for the water in the geocomposite drain shall be maintained at all locations. Weep holes
shall not be sealed or made ineffective by the wall drain material. This may involve making a hole in the
drainage core at the weep hole location. Typically separate the fabric from the core, then wrap it around
the circumference of the perforated underdrain pipe and tuck 6 inches behind the core(Kansas DOT,
125

2007; South Carolina DOT, 2000) and overlapped over the geocomposite drain a distance of 12 inches
and continuously secured with 3 inch wide waterproof plastic tape. Outlet fittings shall be fastened to the
wall drains as directed by the manufacturer and so that backfill materials cannot enter the system during
or after construction. If necessary, to facilitate the rapid and complete flow of water from the wall drain
into the pipe outlet, a portion of the wall drain core equal to the cross section at the outlet shall be
removed. Weep holes shall be accommodated by cutting a matching hole through the wall drain. An
approved weep hole cover extending at least 4 in. (100 mm)(Illinois DOT, 2012) from the edge(s) of the
hole shall be securely fastened to the soil side of the wall drain by 3 in. (75 mm) wide plastic tape or
contact adhesive applied continuously around its periphery.
G. Material Backfill
Structural backfill operations shall be started as soon as possible after placing the geocomposite material,
but in no case shall the geocomposite material be exposed to sunlight for more than 14 days(Arizona
DOT, 2008) or 7 days(South Carolina DOT, 2000) after installation. Care shall be taken during the back
fill operation not to damage the geotextile surface of the drain and to avoid excessive settlement of the
backfill material. Any geocomposite drainage system component that is damaged during the backfilling
operation shall be replaced or repaired as directed by the Engineer. Kansas DOT (2007) requires that the
liquid limit of backfill soil should not exceed 50.
When placing geocomposite drain on walls, use hand operated compaction equipment no closer than 1
foot away from the geocomposite material to compact the backfills.
Structure backfill shall be free of vegetable matter and other deleterious substance. Grading of the backfill
material shall conform to the following table(Hawaii DOT, 2005):
H. Repair
If rip, tear or damaged area on the deployed geocomposite is more than 50 percent of the width of the roll,
the damaged area shall be cut out and the two portions of the geocomposite shall be joined as explained
above. Other rips, tears or damaged areas on the deployed geocomposite shall be removed and patched by
placing a patch extending 12 inches(Alaska DOT, 2013) beyond the edges of the damaged area. The patch
shall be secured to the original geocomposite with cable ties.
If, in the judgment of the Engineer, the damage is not serious enough to warrant removal, the damaged
area may be repaired by overlaying with a piece of fabric, large enough to cover the damaged area and
provide a 4-inch (Arizona DOT, 2008; Kansas DOT, 2007; South Carolina DOT, 2000) overlap on all
sides (6-inch overlapping(California DOT, 2010), or extending 12 inches beyond the edges of the
damaged area (Alaska DOT, 2013)), and taping it in place with 3-inch wide strips of waterproof, plastic
tape. Damaged drainage core sections shall be discarded and replaced. Any geocomposite wall drain
damaged during installation shall be replaced or repaired by the contractor at no additional cost.
126

5. METHOD OF MEASUREMENT
The work will be measured for pavement in place and the area computed in square yards/square meters
(Alaska DOT, 2013; Illinois DOT, 2012; Kansas DOT, 2007. Montana DOT, 2014).s Specifically, the
work amount is counted by multiplying plan neat line width by the measured length in final position
parallel to installation centerline. No allowance will be made for overlap, whether at joints or patches.
6. METHOD OF PAYMENT
Engineer will pay for accepted quantities of geocomposite drain and perforated plastic pipe. Payment will
be made at the contract unit price per square yard (Alaska DOT, 2013; Illinois DOT, 2012). Material used
to fill ruts and holes will be paid for at the unit price for the type of material used.
Meanwhile, Hawaii DOT (2005) will pay for the accepted quantities of geocomposite drain and
perforated pipe on a lump sum basis.
South Carolina DOT (2000) does not measure Geocomposite wall drains for payment. The cost of the
geocomposite wall drain is included in the cost of the retaining wall or wall system where the drainage
system is being installed.
127

Appendix D. Data Source for Cost Analysis
(a). Geocomposite Unit Price from ODOT Project No. 150030
128

(b). Geocomposite Unit Price from ODOT bidding history in year 2001, 2006, 2011 and 2013
Construction Item Survey - 2001

Items 502-605
YrProj Cnty Ref Bidr Quan Bid.Avg. Aw.Bid. Low-Bid. Hi.Bid.

STRUCTURE
ITEM 518E62600 SF DRAINAGE, MISC.:
010059 LUCAS 242 6 6,400 SF $1.79/SF $1.10/SF $1.00/SF $3.00/SF

Items 502-605

STRUCTURE
ITEM 518E62600 SF DRAINAGE, MISC.:
060045 HAMIL 234 3 2,385 SF $2.42/SF $3.00/SF $1.80/SF $3.00/SF

Items 453-605

STRUCTURE
ITEM 518E62600 SF DRAINAGE, MISC.
110154 CUYAH 0118 7 447 SF $8.54/SF $10.00/SF $2.25/SF $14.00/SF

Items 453-605
STRUCTURE
ITEM 518E62600 SF DRAINAGE, MISC.
129

(c). Porous backfill Unit Price from ODOT bidding history in year 2013

ITEM 518E21200 CY POROUS BACKFILL WITH FILTER FABRIC
130002 ALLEN 0538 6 2 $110.69 $75.00 $43.00 $175.00
130002 ALLEN 0545 6 2 $93.74 $68.00 $43.00 $126.45
130002 ALLEN 0552 6 3 $94.47 $70.00 $43.00 $128.79
130309 COLUM 0050 6 3 $101.67 $75.00 $50.00 $250.00
130458 CLINT 0084 1 4 $184.04 $184.04 $184.04 $184.04
130309 COLUM 0066 6 5 $100.83 $75.00 $50.00 $250.00
130383 LORAI 0225 2 6 $70.00 $60.00 $60.00 $80.00
130544 MEDIN 0065 3 8 $183.53 $250.00 $125.00 $250.00
130213 TUSCA 0076 9 8 $76.11 $75.00 $30.00 $100.00
130365 COLUM 0045 4 9 $207.54 $55.00 $55.00 $550.17
130545 MEDIN 0068 7 10 $91.43 $100.00 $50.00 $140.00
130013 MONRO 0049 7 10 $122.14 $100.00 $60.00 $210.00
130213 TUSCA 0065 9 10 $72.44 $75.00 $30.00 $92.00
138007 WILLI 0059 1 10 $75.26 $75.26 $75.26 $75.26
130060 ASHTA 0069 7 11 $120.71 $100.00 $45.00 $245.00
130284 FRANK 0495 3 11 $121.67 $100.00 $100.00 $150.00
130575 BUTLE 0063 5 11 $98.80 $62.00 $62.00 $150.00
130583 SCIOT 0075 5 11 $151.60 $90.00 $90.00 $220.00
130284 FRANK 0483 3 12 $118.33 $100.00 $100.00 $150.00
130325 HURON 0084 1 12 $2,000.00 $2,000.00 $2,000.00 $2,000.00
130100 MADIS 0062 12 13 $110.20 $272.00 $36.00 $272.00
130308 WOOD 0081 3 13 $83.91 $65.00 $65.00 $121.73
130002 ALLEN 0564 6 14 $102.53 $140.00 $59.00 $166.17
130253 CLERM 0053 4 14 $107.18 $117.73 $85.00 $117.73
138014 CUYAH 0158 4 14 $112.88 $161.50 $65.00 $161.50
130035 CUYAH 0159 5 14 $93.00 $150.00 $50.00 $150.00
130137 STARK 0096 3 14 $78.33 $60.00 $60.00 $100.00
130226 BELMO 0050 10 15 $100.50 $80.00 $60.00 $210.00
130575 BUTLE 0085 5 15 $98.50 $63.00 $63.00 $131.00
130421 FRANK 0049 3 16 $80.00 $85.00 $77.00 $85.00
130157 MEIGS 0040 8 16.2 $106.76 $74.08 $60.00 $198.00
130185 ADAMS 0067 8 17 $98.34 $109.18 $50.00 $200.00
130174 WOOD 0157 2 17 $95.00 $90.00 $90.00 $100.00
130512 BROWN 0056 9 18 $97.04 $80.00 $47.95 $175.00
130093 HAMIL 0058 8 18 $85.56 $48.44 $48.44 $124.00
130100 MADIS 0077 12 18 $101.59 $204.00 $36.00 $204.00
130275 NOBLE 0057 8 18 $109.63 $100.00 $60.00 $180.00
130174 WOOD 0103 2 18 $95.00 $90.00 $90.00 $100.00
130229 COSHO 0058 9 19 $88.89 $85.00 $70.00 $145.00
130257 FRANK 0060 5 20 $102.20 $120.00 $70.00 $125.00
130568 ALLEN 0334 3 21 $70.00 $60.00 $60.00 $80.00
130340 BELMO 0060 6 23 $108.83 $100.00 $60.00 $158.00
138041 ATHEN 0049 6 25 $119.17 $115.00 $65.00 $200.00
130061 COSHO 0057 9 25 $97.26 $100.00 $60.00 $140.00
130251 LUCAS 0169 6 25 $59.67 $55.00 $48.00 $70.00
130522 GREEN 0246 4 26 $101.25 $75.00 $75.00 $150.00
130522 GREEN 0278 4 26 $96.25 $75.00 $75.00 $150.00
130251 LUCAS 0124 6 26 $60.33 $55.00 $47.00 $70.00
130185 ADAMS 0090 8 27 $90.39 $72.24 $50.00 $200.00
130332 MUSKI 0057 2 27 $107.50 $95.00 $95.00 $120.00
138034 HARRI 0046 5 27 $93.60 $70.00 $70.00 $125.00
130002 ALLEN 0506 6 28 $74.57 $65.00 $43.00 $100.00
138048 BROWN 0054 11 29 $81.33 $60.00 $60.00 $125.00
130348 FAIRF 0112 4 30 $82.00 $80.00 $60.00 $108.00
130

130330 MAHON 0100 8 30 $70.67 $50.00 $45.00 $100.00
130557 HENRY 0046 4 31 $64.36 $75.00 $50.00 $75.00
130618 MERCE 0046 7 32 $78.96 $40.00 $40.00 $175.00
130433 MUSKI 0131 6 32 $86.17 $105.00 $60.00 $105.00
130313 CLERM 0708 2 33 $55.00 $55.00 $55.00 $55.00
130123 GREEN 0086 7 33 $94.29 $57.00 $57.00 $158.54
130127 KNOX 0056 7 33 $73.14 $100.00 $50.00 $100.00
130223 FRANK 0288 4 34 $92.50 $100.00 $75.00 $110.00
138048 BROWN 0071 11 35 $84.70 $60.00 $58.17 $125.00
130223 FRANK 0312 4 35 $92.50 $100.00 $75.00 $110.00
130042 LUCAS 0116 3 35 $60.00 $70.00 $55.00 $70.00
130139 TUSCA 0049 6 36 $72.67 $111.00 $35.00 $111.00
130546 MIAMI 0056 6 39 $63.26 $60.00 $44.00 $99.00
130511 ATHEN 0037 7 40 $81.56 $107.90 $50.00 $107.90
130314 CUYAH 0295 5 40 $90.61 $150.00 $62.03 $150.00
130131 MONTG 0107 6 40 $68.32 $71.89 $65.00 $75.00
130002 ALLEN 0127 6 42 $67.13 $53.00 $53.00 $90.00
130313 CLERM 0727 2 42 $55.00 $55.00 $55.00 $55.00
120680 MAHON 0176 6 42 $70.04 $70.00 $45.00 $100.00
130073 MEIGS 0058 4 42 $74.46 $99.00 $40.50 $99.00
130016 TRUMB 0073 8 42 $72.81 $65.00 $45.00 $99.00
130584 SENEC 0062 5 42 $64.49 $60.00 $55.00 $75.00
130284 FRANK 0535 3 43 $72.33 $75.00 $60.00 $82.00
130359 ATHEN 0060 5 45 $93.20 $84.00 $84.00 $110.00
130106 PUTNA 0064 7 45 $65.71 $70.00 $50.00 $100.00
130348 FAIRF 0055 4 46 $81.25 $85.00 $60.00 $100.00
130156 MEIGS 0062 6 46 $80.67 $49.00 $49.00 $125.00
130353 MUSKI 0055 5 46 $82.80 $80.00 $60.00 $105.00
130535 BELMO 0109 2 47 $111.50 $113.00 $110.00 $113.00
130011 HOLME 0080 7 47 $84.86 $65.00 $40.00 $144.00
130535 BELMO 0131 2 48 $111.00 $112.00 $110.00 $112.00
130379 LICKI 0056 3 48 $106.00 $117.00 $100.00 $117.00
130616 ASHTA 0059 10 48 $94.22 $55.00 $49.52 $162.43
138033 SUMMI 0468 3 49 $73.08 $84.24 $60.00 $84.24
130147 SUMMI 0465 5 49 $76.80 $60.00 $60.00 $99.00
138047 ATHEN 0043 9 50 $93.84 $125.00 $50.00 $177.55
138047 ATHEN 0083 9 50 $94.84 $125.00 $50.00 $177.55
130431 MONTG 0062 7 51 $75.14 $60.00 $56.00 $110.00
130613 TRUMB 0151 7 51 $78.29 $66.00 $50.00 $125.00
130205 MAHON 0117 5 52 $58.00 $55.00 $45.00 $80.00
130075 ROSS 0053 11 55 $68.19 $57.00 $47.00 $100.00
130433 MUSKI 0060 6 56 $83.50 $110.00 $59.00 $110.00
130032 CLINT 0154 2 58 $80.00 $95.00 $65.00 $95.00
130062 CUYAH 0096 5 58 $85.60 $55.00 $55.00 $168.00
130326 JEFFE 0072 5 59 $101.80 $75.00 $74.00 $150.00
133024 MONTG 0071 3 59 $86.67 $100.00 $75.00 $100.00
133024 MONTG 0108 3 60 $86.67 $100.00 $75.00 $100.00
130597 TUSCA 0093 4 61 $106.75 $80.00 $75.00 $175.00
130597 TUSCA 0119 4 61 $106.75 $80.00 $75.00 $175.00
130234 KNOX 0066 6 65 $71.33 $90.00 $45.00 $90.00
130526 LORAI 0050 9 65 $65.33 $40.00 $40.00 $100.00
130267 JEFFE 0106 2 66 $87.50 $100.00 $75.00 $100.00
130568 ALLEN 0384 3 67 $57.00 $56.00 $47.00 $68.00
130265 HIGHL 0057 10 67 $67.61 $50.30 $40.00 $85.79
138001 MAHON 0090 5 68 $66.00 $45.00 $45.00 $85.00
138033 SUMMI 0431 3 68 $79.47 $88.40 $70.00 $88.40
130147 SUMMI 0428 5 68 $74.80 $65.00 $60.00 $99.00
130563 SUMMI 0143 4 68 $67.00 $68.00 $50.00 $80.00
131

130266 HOCKI 0062 5 69 $82.60 $90.00 $58.00 $105.00
130563 SUMMI 0112 4 70 $61.25 $60.00 $50.00 $70.00
130002 ALLEN 0664 6 71 $65.83 $65.00 $55.00 $80.00
130002 ALLEN 0699 6 71 $65.83 $65.00 $55.00 $80.00
130568 ALLEN 0413 3 72 $56.67 $56.00 $45.00 $69.00
130361 BELMO 0064 4 72 $95.50 $70.00 $70.00 $120.00
130031 AUGLA 0115 4 74 $81.25 $60.00 $60.00 $125.00
130174 WOOD 0134 2 75 $85.00 $70.00 $70.00 $100.00
130284 FRANK 0517 3 77 $65.00 $75.00 $50.00 $75.00
130199 HIGHL 0055 9 78 $72.57 $41.03 $41.03 $94.40
130313 CLERM 0325 2 79 $55.00 $55.00 $55.00 $55.00
130525 FRANK 0071 5 80 $71.40 $48.00 $48.00 $86.00
130522 GREEN 0211 4 82 $91.25 $75.00 $75.00 $125.00
130112 UNION 0069 8 82 $74.83 $75.00 $60.00 $104.00
130102 MIAMI 0058 9 85 $61.41 $38.00 $30.00 $120.00
130174 WOOD 0199 2 87 $70.00 $65.00 $65.00 $75.00
130131 MONTG 0096 6 88 $68.24 $69.86 $50.00 $85.00
130402 FRANK 0518 4 90 $70.63 $66.00 $61.50 $80.00
130313 CLERM 0682 2 92 $55.00 $55.00 $55.00 $55.00
138033 SUMMI 0617 3 92 $74.40 $83.20 $60.00 $83.20
130147 SUMMI 0614 5 92 $73.14 $60.00 $60.00 $95.70
130402 FRANK 0574 4 93 $68.50 $64.00 $60.00 $75.00
130002 ALLEN 0798 6 97 $61.83 $56.00 $50.00 $70.00
130002 ALLEN 0843 6 97 $61.83 $56.00 $50.00 $70.00
130586 HANCO 0264 6 98 $63.83 $60.00 $55.00 $75.00
138033 SUMMI 0581 3 98 $76.56 $89.44 $60.00 $89.44
130147 SUMMI 0578 5 98 $74.14 $60.00 $60.00 $95.70
130078 TRUMB 0094 4 98 $69.00 $66.00 $45.00 $95.00
138033 SUMMI 0542 3 99 $76.48 $89.44 $60.00 $89.44
130147 SUMMI 0539 5 99 $75.14 $60.00 $60.00 $95.70
138046 DELAW 0023 9 100 $68.11 $50.00 $48.00 $125.00
120680 MAHON 0137 6 100 $60.00 $65.00 $45.00 $75.00
130002 ALLEN 0591 6 102 $65.00 $65.00 $45.00 $90.00
130002 ALLEN 0628 6 102 $63.33 $55.00 $45.00 $90.00
130313 CLERM 0352 2 102 $55.00 $55.00 $55.00 $55.00
130089 COLUM 0074 7 107 $75.14 $60.00 $45.00 $120.00
130592 HIGHL 0059 11 107 $60.17 $50.00 $40.00 $95.00
138033 SUMMI 0690 3 108 $75.44 $86.32 $60.00 $86.32
130147 SUMMI 0688 5 108 $73.58 $60.00 $60.00 $97.90
130377 HURON 0061 5 109 $64.80 $115.00 $42.00 $115.00
130186 ASHTA 0091 6 115 $86.67 $95.00 $50.00 $140.00
130186 ASHTA 0127 6 115 $77.67 $95.00 $45.00 $140.00
130151 CUYAH 0255 4 121 $94.71 $150.00 $73.85 $150.00
130362 BELMO 0060 3 122 $90.00 $65.00 $65.00 $110.00
130568 ALLEN 0587 3 124 $53.67 $50.00 $49.00 $62.00
130568 ALLEN 0517 3 126 $53.67 $50.00 $49.00 $62.00
130568 ALLEN 0553 3 126 $54.67 $52.00 $49.00 $63.00
130314 CUYAH 0306 5 127 $78.94 $100.00 $62.72 $100.00
138033 SUMMI 0503 3 135 $73.43 $85.28 $60.00 $85.28
130147 SUMMI 0500 5 135 $74.58 $60.00 $60.00 $97.90
138017 MAHON 0250 4 136 $57.13 $55.00 $45.00 $68.50
130189 BELMO 0042 7 138 $97.06 $108.00 $65.00 $135.00
130185 ADAMS 0074 8 139 $66.77 $26.23 $26.23 $110.00
130568 ALLEN 0622 3 142 $55.67 $50.00 $50.00 $63.00
130534 AUGLA 0121 6 144 $63.50 $55.00 $50.00 $95.00
130484 MIAMI 0059 8 148 $63.11 $64.47 $50.00 $84.97
130314 CUYAH 0335 5 151 $73.27 $80.00 $56.34 $90.00
130313 CLERM 0275 2 154 $55.00 $55.00 $55.00 $55.00
132

130313 CLERM 0301 2 154 $55.00 $55.00 $55.00 $55.00
130026 LICKI 0174 6 161 $72.50 $65.00 $50.00 $90.00
130002 ALLEN 0730 6 163 $61.50 $54.00 $50.00 $70.00
130002 ALLEN 0761 6 163 $61.83 $56.00 $50.00 $70.00
130568 ALLEN 0481 3 171 $52.67 $53.00 $48.00 $57.00
138033 SUMMI 0652 3 172 $72.75 $78.00 $60.00 $80.25
130147 SUMMI 0650 5 172 $71.36 $60.00 $60.00 $96.80
138012 LICKI 0123 5 177 $74.60 $85.00 $65.00 $85.00
130103 MIAMI 0071 8 182 $68.41 $90.00 $28.54 $159.49
130568 ALLEN 0654 3 189 $51.00 $50.00 $45.00 $58.00
130063 CUYAH 0119 4 198 $64.75 $65.00 $60.00 $70.00
130577 BROWN 0039 5 200 $84.40 $110.00 $50.00 $110.00
130402 FRANK 0530 4 209 $69.75 $66.00 $61.00 $77.00
130376 HURON 0103 7 210 $61.71 $68.00 $50.00 $68.00
130402 FRANK 0661 4 214 $69.75 $65.00 $60.00 $81.00
130151 CUYAH 0291 4 237 $78.46 $85.00 $73.85 $85.00
130567 TRUMB 0172 5 278 $65.60 $60.00 $50.00 $86.00
130284 FRANK 0596 3 282 $57.33 $58.00 $50.00 $64.00
130568 ALLEN 0446 3 291 $50.67 $44.00 $44.00 $61.00
130284 FRANK 0568 3 291 $55.00 $60.00 $45.00 $60.00
130184 CUYAH 0257 4 310 $78.75 $70.00 $65.00 $100.00
130567 TRUMB 0208 5 324 $65.80 $60.00 $50.00 $87.00
130001 LORAI 0218 8 340 $59.18 $52.00 $52.00 $70.00
130268 JEFFE 0096 4 400 $65.00 $75.00 $40.00 $85.00
130086 CLERM 0250 5 485 $55.44 $38.21 $38.21 $74.00
130375 HARRI 0053 11 505 $87.45 $65.00 $60.00 $135.00
130313 CLERM 0744 2 572 $55.13 $52.25 $52.25 $58.00
130284 FRANK 0546 3 756 $56.67 $55.00 $45.00 $70.00
130284 FRANK 0507 3 971 $78.33 $55.00 $55.00 $115.00
130285 HAMIL 0165 4 1116 $52.50 $65.00 $45.00 $65.00
130001 LORAI 0188 8 1259 $56.05 $45.00 $45.00 $65.00
133

Appendix E. Flow Net and Coefficients of Earth Pressure
This appendix discusses the flow net behind walls with a vertical drainage layer and the selection
of coefficient of earth pressures.
Flow net:
A schematic flow net behind an abutment was shown in Figure E-1. The number of flow lines F=5
and the number of potential drops N=5.
Drainage layer
Equipotential line
Flow line
Figure 35. Raw flow nets behind a typical bridge abutment.
Coefficient of Earth Pressure: If the wall is not allowed to move or deflect, the at-rest lateral earth
pressure coefficient K0 should be used. For normally consolidated soils, vertical wall and level
ground, K0 = 1-sinϕ, where ϕ is the friction angle of backfill soil.
Walls which can move away from the soil mass should be designed for pressures between active
and at-rest conditions, depending on the magnitude of the tolerable movements. This applies to
most walls and abutments with jointed superstructures, where the joints can accommodate the
lateral movement. Integral and semi-integral abutments shall be designed to resist and/or absorb
creep, shrinkage and thermal deformations of the superstructure, and earth pressure between the
at-rest and passive condition can be used.
Active and passive earth pressure can be calculated based on Coulomb earth pressure theories,
which is suggested by ODOT BDM Item 204.6.2. For vertical smooth wall and level backfill
ground, the active and passive earth pressure may be taken as follows.
134

Coefficient of active earth pressure:
1− sin φ '
Ka =
1+ sin φ '
Coefficient of passive earth pressure:
1+ sin φ '
Ka =
1− sin φ '
Movement required to reach the minimum active pressure or the maximum passive pressure is a
function of the wall height and the soil type, as shown in Table 35. The variation of earth pressure
coefficient with wall movement is shown in Figure 36. It should be noted that the movement
required to mobilize passive pressure is approximately 10 times as large as the movement required
for active pressure. Consequently, retaining structures are often designed for the full active state
when earth retaining wall can move outward, but only for a fraction of the full passive state when
the retaining wall moves inward.
Table 35. Approximate Values of Relative Movements Required to Reach Active or Passive Earth
Pressure Conditions (Clough and Duncan, 1991)
135

Figure 36. Relationship between wall movement and earth pressure. (After Clough and Duncan 1991)
Based on surveys of performance of bridges conducted by various researchers (Bozozuk, 1978;

Walkinshaw, 1978; Moulton et al., 1985; Wahls, 1990), a horizontal abutment movements of less
than 1.5 in. can usually be tolerated by bridge superstructures without significant damage.
Therefore, earth pressures used in the design of abutments should be selected to be consistent with
the requirement that the abutment should not move more than 1.5 in. laterally. If the design height
of the wall is known, the active and passive coefficients for design purposes can be determined
using the relationship shown in Figure 36.
Effect of Compaction: ODOT CMS specifies that the backfill behind abutments and walls should
be compacted by a 10-ton roller. In general, when soil is compacted in layers by rollers, vibrating
plates or rammers, the horizontal earth pressures within the compacted soil mass are increased.
The compaction effect is ignored in the design of the walls as suggested by ODOT BDM: S3.11.2.
However, as suggested in AASHTO 3.11.2 (AASHTO, 2007), the effect of additional earth
pressure induced by compaction shall be taken into account where activity by mechanical
compaction equipment is anticipated within a distance of one-half the height of the wall. Duncan
and Seed (1986) evaluated compaction-induced earth pressures under at-rest conditions and
proposed a hysteretic method for calculation. Duncan et al. (1991) later developed charts and
manual calculation procedures to estimate earth pressures due to compaction. Field measurements
indicate that the compaction induced pressures of sands do not change appreciably with time unless
the wall on which they act moves toward or away from the backfill. For the design of PCDS
systems, it is suggested to include the effect of compaction when determining the required core
strength. AASHTO C3.11.2 (AASHTO, 2007) suggests that compaction-induced earth pressures
may be estimated using the procedures described by Clough and Duncan (1991). Alternatively, the
modified K-Δ/H curve considering the compaction effects can be used (Figure 37).
136

Figure 37. Relationship between wall movement and earth pressure for compacted soil.
(After Clough and Duncan 1991).flat
137

Appendix F Installation plan as shown in the marked plan sheets
138

139

140

141

142

143

144

Appendix G Detailed Laboratory Procedures and Results
G1. Thickness of the Core and Fabric
Nominal thickness of the geosynthetics was determined following ASTM D5199-12 by measuring
the distance between two parallel surfaces of test specimens confined under specified normal stress
(20 kPa or 2 kPa). A thickness gauge incorporating a rigid base plate and circular presser foot (2.22
inch and 0.25 inch in diameter for geotextile and core sample in diameter, respectively) oriented
parallel to the base plate was used (Fig. G1).
Thickness measurements were obtained for samples of composite geosynthetic products supplied
by the manufacturers (core and fabric bonded together), as well as for separated samples of core
and fabric supplied in select cases. Ten replicate specimens (2.5 inch by 2.5 inch) were sampled
from the rolls for each measurement. Samples were obtained from locations representing the full
width of the roll.
Table G1 summarizes measured thickness of the composite geosynthetics (bonded fabric and core)
under applied stress of 20 kPa and 2 kPa. Table G2 summarizes measured thickness of the
separated fabric and core.
Fig. G1. Testing apparatus for thickness measurement
145

Table G1. Thickness of composite geosynthetics (bonded fabric and core).
Table G2. Thickness of separated fabric and core.
146

G2. Permittivity of the Fabric
Permittivity of separated fabric samples was measured using constant-head test procedures
following ASTM D4491-99a. Permittivity (Ψ) is the volumetric flow rate of water per unit cross
sectional area per unit head under laminar flow conditions. A photograph of the testing apparatus
is provided as Figure G2. Four replicate specimens from a 1-m long section of each full-width
fabric roll were taken (Fig G2a). Specimen diameter was 2 in (51 mm). Hydraulic head was
maintained in the range from 10 mm to 75 mm and five readings of flow rate were obtained at
each hydraulic head. Permittivity was calculated using the slope of a plot of flow rate versus head.
Table G3 summarizes average permittivity for the five sets of readings at a hydraulic head 50 mm
along with manufacturer-reported values for comparison. Detailed permittivity calculations for
two fabrics are provided in Tables G4 and Table G5.
(a) (b)
Fig. G2. (a) Constant-head permittivity apparatus; (b) example locations of specimens sampled
from fabric roll.
Table G3. Average permittivity and flow rate at 50 mm head.
147

Table G4. Detailed permittivity calculations (Product A).
Permittivity Calculation for J-Drain 400 XL
Specimen A Specimen B Specimen C Specimen D
Head mm Q/At (mm/sec) Ψ sec^-1 Head mm Q/At (mm/sec) Ψ sec^-1 Head mm Q/At (mm/sec) Ψ sec^-1 Head mm Q/At (mm/sec) Ψ sec^-1
50 55.58 1.11 50 51.34 1.03 50 58.41 1.17 50 54.51 1.09
10 14.03 1.40 10 15.41 1.54 10 16.16 1.62 10 13.75 1.37
15 22.14 1.48 15 18.88 1.26 15 26.42 1.76 15 20.35 1.36
20 27.66 1.38 20 22.20 1.11 20 34.21 1.71 20 29.65 1.48
25 32.31 1.29 25 29.27 1.17 25 39.41 1.58 25 34.89 1.40
30 38.17 1.27 30 34.42 1.15 30 45.80 1.53 30 41.70 1.39
35 43.02 1.23 35 37.18 1.06 35 49.52 1.41 35 46.79 1.34
40 45.47 1.14 40 41.12 1.03 40 51.52 1.29 40 52.53 1.31
45 48.51 1.08 45 43.59 0.97 45 55.85 1.24 45 51.95 1.15
50 52.67 1.05 50 45.46 0.91 50 57.08 1.14 50 56.07 1.12
55 54.67 0.99 55 46.04 0.84 55 64.07 1.16 55 58.04 1.06
60 54.82 0.91 60 48.43 0.81 60 65.45 1.09 60 57.86 0.96
65 63.88 0.98 65 49.88 0.77 65 64.24 0.99 65 57.77 0.89
70 58.51 0.84 70 56.30 0.80 70 65.81 0.94 70 57.36 0.82
75 75 57.62 0.77 75 67.91 0.91 75 60.54 0.81
Avg. Permittivity, Ψ 1.24 1.12 1.44 1.30
Standart Deviation 0.15 0.18 0.23 0.13
Coefficient of Variance % 11.82 15.93 15.64 10.18
Permittivity of each
1.08 0.97 1.15 1.11
specimen at 50 mm head
Avg. Permittivity of four
1.08
specimens at 50 mm head
Flow rate at 50 mm
3247.68 2903.74 3464.59 3317.52
l/min/m2
Avg. Flow rate at 50 mm
3233
l/min/m2
148

Table G5. Detailed permittivity calculations (Product B).

Permittivity Calculation for Mira G100W
Specimen A Specimen B Specimen C Specimen D
Head mm Q/At (mm/sec) Ψ sec^-1 Head mm Q/At (mm/sec) Ψ sec^-1 Head mm Q/At (mm/sec) Ψ sec^-1 Head mm Q/At (mm/sec) Ψ sec^-1
50 103.44 2.07 50 99.03 1.98 50 93.41 1.87 50 103.36 2.07
10 32.34 3.23 10 28.95 2.90 10 30.42 3.04 10 34.89 3.49
15 34.83 2.32 15 42.89 2.86 15 39.80 2.65 15 49.45 3.30
20 45.26 2.26 20 59.36 2.97 20 46.20 2.31 20 59.23 2.96
25 60.55 2.42 25 62.18 2.49 25 62.66 2.51 25 67.59 2.70
30 76.97 2.57 30 70.86 2.36 30 67.66 2.26 30 73.69 2.46
35 82.61 2.36 35 82.63 2.36 35 64.34 1.84 35 82.96 2.37
40 92.58 2.31 40 97.01 2.43 40 89.07 2.23 40 86.34 2.16
45 98.28 2.18 45 97.05 2.16 45 78.97 1.75 45 100.27 2.23
50 104.38 2.09 50 103.24 2.06 50 96.97 1.94 50 112.63 2.25
55 115.00 2.09 55 112.61 2.05 55 103.66 1.88 55 110.99 2.02
60 110.86 1.85 60 105.09 1.75 60 101.34 1.69 60 116.29 1.94
65 98.76 1.52 65 112.79 1.74 65 100.48 1.55 65 100.48 1.55
70 115.67 1.65 70 112.71 1.61 70 89.56 1.28 70 na na
75 128.77 1.72 75 106.61 1.42 75 88.83 1.18 75 na na
Avg. Permittivity, Ψ 2.38 2.46 2.24 2.60
Standart Deviation 0.33 0.35 0.41 0.50

Coefficient of Variance % 14.05 14.27 18.33 19.15
Permittivity of each
2.08 2.02 1.90 2.16
specimen at 50 mm head
Avg. Permittivity of four
2.04
specimens at 50 mm
Flow rate at 50 mm
6234.754 6068.103 5711.353 6479.793
l/min/m2
Avg. Flow rate at 50 mm
6124
l/min/m2
149

G3. Compressive Strength of the Composite (Core + Fabric)
Compressive strength of the composite geosynthetics (bonded fabric and core) was determined
following procedures specified in ASTM D1621-10. Fig. G3a is a photograph of the test apparatus,
which included a stepper-motor controlled load frame (GeoTac, Houston, TX) with integrated
force transducer and linear variable differential transformer (LVDT). Five replicate specimens
were prepared and tested for each geosynthetic. Specimens were square in shape with dimensions
of 4.5 in X 4.5 in (10 x 10 mm). Compressive force was applied using a constant displacement rate
of 0.1 in/min (0.25 mm/min). Table G6 is a summary of average compressive strength from five
replicate measurements along with manufacturer-supplied values for comparison. Figures G4
through A8 are summary plots and statistics for each test.
Specimen in
two circular
plates
Data
acquisition Load and
speed control
(a) (b)
Fig. G3. (a) Apparatus for measuring compressive strength of composite geosynthetics (bonded
fabric and core); (b) five replicate specimens with dimensions of 4.5 in X 4.5 in (10 x 10 mm)
were sampled from each manufacturer-supplied roll.
Table G6. Summary of compressive strength of composite geosynthetics (bonded fabric and
core)
150

Product A (Fabric +Core)

24000
Test-1 Test-2 Test-3 Test-4 Test-5
Compressive Stress (psf) 20000
16000
12000
8000
4000
0
0 2 4 6 8 10 12 14 16 18 20 22 24
Strain (%)
Nominal Dimension Max. Strain at

Compressive Max.
Test No
Length Width Thickness Strength Strength
(in) (in) (in) (psf) (%)
1 4.5 4.5 0.39 13,890 11.2
2 4.5 4.5 0.39 15,285 10.6
3 4.5 4.5 0.39 14,651 13.0
4 4.5 4.5 0.39 13,002 11.3
5 4.5 4.5 0.39 12,495 13.5
Mean 13,865 11.9
SD 1,025.4 1.1
COV % 7.4 9.3
Fig. G4. Compressive stress-strain plots and summary statistics (Product A).
151

Product B (Fabric+Core)
24000
16000
12000
8000
4000
0
0 5 10 15 20 25
Strain (%)

Compressive Max.
Test No
(in) (in) (in) (psf) (%)
1 4.5 4.5 0.38 18,266 9.9
2 4.5 4.5 0.38 17,758 10.8
3 4.5 4.5 0.38 17,505 10.0
4 4.5 4.5 0.38 17,505 10.2
5 4.5 4.5 0.38 16,490 9.7
Mean 17,505 10.1
SD 578.5 0.4
COV % 3.3 3.7
Fig. G5. Compressive stress-strain plots and summary statistics (Product B).
152

Product C
24000
20000
Compressive Stress (psf)
16000
12000
8000
4000
0
0 5 10 15 20 25 30
Strain (%)

Compressive Max.
Test No
(in) (in) (in) (psf) (%)
1 4.5 4.5 0.39 15,983 11.1
2 4.5 4.5 0.39 13,763 12.8
3 4.5 4.5 0.39 13,192 13.6
4 4.5 4.5 0.39 13,382 9.3
5 4.5 4.5 0.39 14,017 10.3
Mean 14,067 11.4
SD 999.9 1.6
COV % 7.1 13.9
Fig. G6. Compressive stress-strain plots and summary statistics (Product C).
153

Product D (Fabric+Core)
24000
16000
12000
8000
4000
0
0 5 10 15 20 25 30
Strain (%)

Compressive Max.
Test No
(in) (in) (in) (psf) (%)
1 4.5 4.5 0.42 17,061 8.2
2 4.5 4.5 0.42 17,188 10.8
3 4.5 4.5 0.42 16,110 9.6
4 4.5 4.5 0.42 17,061 10.8
5 4.5 4.5 0.42 17,061 10.3
Mean 16,896 10
SD 396.3 1.0
COV % 2.3 9.9
Fig. G7. Compressive stress-strain plots and summary statistics (Product D).
154

Product E (Fabric+Core)
20000
Compressive Stress (psf)
15000
10000
5000
0
0 5 10 15 20
Strain (%)

Compressive Max.
Test No
(in) (in) (in) (psf) (%)
1 4.5 4.5 0.39 17,315 10.28
2 4.5 4.5 0.39 17,632 13.52
3 4.5 4.5 0.39 17,695 10.90
4 4.5 4.5 0.39 17,378 12.59
5 4.5 4.5 0.39 17,632 12.05
Mean 17,530 11.9
STD 153.3 1.2
CV % 0.9 9.8
Fig. G8. Compressive stress-strain plots and summary statistics (Product E).
155

G4. Grab Tensile Strength and Elongation of the Fabric
Tests were performed according to ASTM Standard D4632/D4632M-15a to quantify breaking

load (grab strength) and elongation (grab elongation) of fabric samples separated from the
composite geosynthetics. A tension testing machine (Fig. G9a) was used to perform the grab
tensile/elongation tests. Specimens from the fabric rolls were cut to sizes of 4 inches wide by 5
inches long along a diagonal line on the roll in two directions: parallel to machine direction (MD)
and cross to the machine direction (XMD) (Fig. G9b). The number of specimens taken from each
roll ranged from 7 to 10 depending on the width of the roll supplied by the manufacturer. Tests
were conducted on dry samples at room temperature. The length of the specimen between the
clamps at the start of test was 3 inches. Displacement rate was 12 in/min.
Average grab tensile strength and elongation at the breaking load is summarized in Table G7.
Maximum load and elongation for replicate specimens are presented in Tables G8 and G9.
Corresponding plots of load versus strain are presented in Fig. G10 and Fig G11. Since the fabric
used in both Product A and Product C is the same type (Tencate Mirafi 160N), identical test results
are reported for these products.
Tensile testing
machine
Specimen in
clamps
Data Acquisition
Width
(a) (b)
Fig. G9. (a) Tensile testing machine used for grab strength and elongation tests; (b) specimens
were obtained along a diagonal line across the entire width of the fabric roll and in the machine
direction (MD) and cross machine direction (XMD).
156

Table G7. Summary of grab tensile strength and elongation.
Table G8. Grab tensile strength and elongation (Product A)
157

Table G9. Grab tensile strength and elongation (Product E).
158

a)
b)
Fig. G10. Grab tensile load vs. elongation of Product A fabric in (a) cross machine direction
(XMD) and (b) parallel machine direction (MD).
159

a)
b)
Fig. G11. Grab tensile load vs. elongation of Product E fabric in (a) cross machine direction
(XMD) and (b) parallel machine direction (MD).
160

G5. Trapezoidal Tear Strength of the Fabric
Tests were conducted according to ASTM D4533/D4533M-15 to measure the force required to
continue to propagate a tear in woven and non-woven geotextiles using the trapezoidal method.
Apparatus and specimen preparation were the same as the grab tensile strength test described in
the previous section. Specimens from each fabric roll were cut in size of 3 inches by 8 inches along
a diagonal line on the roll in two directions: parallel to the machine direction (MD) and cross to
the machine direction (XMD). The number of specimen number varied from 7 to 10 depending on
width of the fabric roll. Tests were conducted on dry samples at room temperature. The length of
the specimen between the clamps at the start of test was 1 inch and the machine was operated at a
displacement rate of 12 in/min.
Results for maximum trapezoidal tensile strength at the breaking load are summarized in Table
G10. Maximum breaking loads for replicate specimens are summarized in Tables G11 and G12.
Corresponding plots of load versus strain are presented in Fig. G12 and Fig G13. Since the fabric
used in both Product A and Product C is the same type (Tencate Mirafi 160N), identical test results
are reported for these products.
Table G10. Summary of trapezoidal tear strength results.
161

Table G11. Trapezoidal tensile strength (Product A).
Table G12. Trapezoidal tensile strength (Product E).
162

a)
b)
Fig. G12. Trapezoidal tear strength of Product A: (a) machine direction (MD) and (b) cross-
machine direction (XMD).
163

a)
b)
Fig. G13. Trapezoidal tear strength of Product E: (a) machine direction (MD) and (b) cross-
machine direction (XMD).
164

G6. In-Plane Flow Rate
Tests were conducted following ASTM D4716/D4716M-14 to measure flow rate per unit width
within the manufactured plane of geosynthetics under varying normal compressive stresses and a
constant head. The flow rate per unit width was determined by measuring the quantity of water
that passed through a test specimen in a specific time interval under three different compressive
stresses (720 psf, 2000 psf, and 3600 psf) at hydraulic gradients of 0.1, 0.5 and 1. Fig. G14 is a
photograph of the constant head (in-plane) flow testing device used for these tests.
Fig. G14. Constant head (in-plane) flow rate testing device
Samples from each product were cut into square specimen sizes of 12 by 14 inches from the roll
in two directions; parallel to machine direction (MD) and cross to machine direction (XMD) as
marked on fabric (Fig. G15). Actual specimen size is 12 x 12 inches after the specimen is placed
into the specimen box. Six specimens from each product corresponding to three different
compressive stresses were used for this test (i.e., one MD specimen at each stress and one XMD
specimen at each stress). Tests were conducted using two seating periods (0.25 h and 100 h) to
assess time-dependent structural stability of the specimen. Tests were conducted on fully saturated
specimens at room temperature.
165

XMD
Fig. G15. Specimen cut into two directions and used for in-plane flow rate testing
The following equation was used to calculate the in-plane flow rate:
$%
! =
& '
Where q is the flow rate per unit width, m3/s-m [gpm/ft], Qt is a measured quantity of water
collected during collection time t [s], and W is width of specimen [0.304 m or 1 ft]. Hydraulic
transmissivity, which is the volumetric flow rate per unit width of specimen per unit gradient in a
direction parallel to the plane of the specimen was calculated from:
$% )
( =
'*
where θ is the hydraulic transmissivity [m2/s] Qt is a measured quantity of fluid discharged per
unit time [m3/s] L is the length of specimen subjected to the normal load [m], W is the width of
the specimen [m], and H is the head difference.
Results are presented in two different sections according to time period used for the specimen
seating period under the normal load (short term for 0.25 h and long term for 100 h).
Short Term In-Plane Flow Rate (0.25-hour seating period)
Average in-plane flow rate measured in two directions (MD and XMD) for the short term (0.25-h)
seating load are summarized in Table G13. The table includes values reported by the manufacturers
for the core only. Results from the in-plane flow rate test conducted here with the composite fabric
166

and core do not match the flow rates reported for the core only. To assess these differences, an
additional test was conducted using the Product A using only the core. These results are
summarized in Table G14, where it may be seen that the results for the core only measured here
and reported by the manufacturer are in closer agreement.
Table G13. Summary of in-plane flow results for short term (0.25-h) seating time.
Tables G14 through G18 summarize results from replicate specimens in the MD and XMD
directions at each applied stress. Figures A16 through A20 are corresponding plots of measured
flow rate versus gradient.
Table G14. In-plane flow rate and hydraulic transmissivity (Product A)
Seating Specimen Hydraulic

Flow Hydraulic Flow Rate
Normal Test No. Time size (width x Transmissivity
Direction gradient (l/min-m ) (gpm/ft)
Stress (min) length) (in) (m2 /sec)
1 MD 0.1 8.07E-03 48.4 3.9
2 MD 0.5 4.04E-03 121.3 9.8
3 MD 1 2.94E-03 176.5 14.2
720 psf
4 Cross MD 0.1 9.36E-03 56.2 4.5
5 Cross MD 0.5 4.35E-03 130.5 10.5
6 Cross MD 1 3.06E-03 183.8 14.8
7 MD 0.1 8.57E-03 51.4 4.1
8 MD 0.5 4.35E-03 130.4 10.5
9 MD 1 2.94E-03 176.5 14.2
2000 psf 15 12 x 12
10 Cross MD 0.1 8.22E-03 49.3 4.0
11 Cross MD 0.5 4.16E-03 124.9 10.1
12 Cross MD 1 3.05E-03 183.3 14.8
13 MD 0.1 8.19E-03 49.2 4.0
14 MD 0.5 4.23E-03 126.9 10.2
15 MD 1 2.97E-03 178.1 14.3
3600 psf
16 Cross MD 0.1 8.31E-03 49.8 4.0
17 Cross MD 0.5 4.47E-03 134.2 10.8
18 Cross MD 1 3.04E-03 182.4 14.7
167

Table G15. In-plane flow rate and hydraulic transmissivity (Product B)
Seating Specimen Hydraulic

Test Flow Hydraulic Flow Rate
Normal Time size (width x Transmissivity
No. Direction gradient (l/min-m ) (gpm/ft)
Stress (min) length) (in) (m2/sec)
1 MD 0.1 1.10E-02 66.1 5.3
2 MD 0.5 4.91E-03 147.2 11.9
3 MD 1 3.46E-03 207.5 16.7
720 psf
4 Cross MD 0.1 9.35E-03 56.1 4.5
5 Cross MD 0.5 4.66E-03 139.9 11.3
6 Cross MD 1 3.36E-03 201.6 16.2
7 MD 0.1 1.14E-02 68.2 5.5
8 MD 0.5 5.42E-03 162.5 13.1
9 MD 1 3.96E-03 237.4 19.1
2000 psf 15 12 x 12
10 Cross MD 0.1 8.73E-03 52.4 4.2
11 Cross MD 0.5 4.50E-03 134.9 10.9
12 Cross MD 1 3.34E-03 200.4 16.1
13 MD 0.1 1.13E-02 67.9 5.5
14 MD 0.5 5.49E-03 164.7 13.3
15 MD 1 4.04E-03 242.5 19.5
3600 psf
16 Cross MD 0.1 8.01E-03 48.1 3.9
17 Cross MD 0.5 4.52E-03 135.5 10.9
18 Cross MD 1 3.21E-03 192.3 15.5
Table G16. In-plane flow rate and hydraulic transmissivity (Product C)

Specimen Hydraulic
Test Flow Hydraulic Seating Time Flow Rate
Normal size (width x Transmissivity
No. Direction gradient (min) (l/min-m ) (gpm/ft)
Stress length) (in) (m2/sec)
1 MD 0.1 1.01E-02 60.7 4.9
2 MD 0.5 4.46E-03 133.8 10.8
3 MD 1 3.25E-03 195.1 15.7
720 psf
4 Cross MD 0.1 6.59E-03 39.5 3.2
5 Cross MD 0.5 3.10E-03 92.9 7.5
6 Cross MD 1 1.89E-03 113.4 9.1
7 MD 0.1 8.65E-03 51.9 4.2
8 MD 0.5 4.27E-03 128.1 10.3
9 MD 1 3.21E-03 192.4 15.5
2000 psf 15 12 x 12
10 Cross MD 0.1 5.23E-03 31.4 2.5
11 Cross MD 0.5 2.63E-03 79.0 6.4
12 Cross MD 1 1.85E-03 111.0 8.9
13 MD 0.1 8.88E-03 53.3 4.3
14 MD 0.5 4.23E-03 126.9 10.2
15 MD 1 3.02E-03 181.4 14.6
3600 psf
16 Cross MD 0.1 5.28E-03 31.7 2.6
17 Cross MD 0.5 2.57E-03 77.2 6.2
18 Cross MD 1 1.85E-03 111.2 9.0
168

Table G17. In-plane flow rate and hydraulic transmissivity (Product D)
Specimen Hydraulic
Flow Hydraulic Seating Time Flow Rate
Normal Test No. size (width x Transmissivity
Direction gradient (min) (l/min-m ) (gpm/ft)
1 MD 0.1 1.02E-02 60.9 4.9
2 MD 0.5 5.12E-03 153.5 12.4
3 MD 1 3.72E-03 223.0 18.0
720 psf
4 Cross MD 0.1 9.40E-03 56.4 4.5
5 Cross MD 0.5 3.92E-03 117.6 9.5
6 Cross MD 1 2.71E-03 162.8 13.1
7 MD 0.1 9.57E-03 57.4 4.6
8 MD 0.5 4.74E-03 142.3 11.5
9 MD 1 3.54E-03 212.7 17.1
2000 psf 15 12 x 12
10 Cross MD 0.1 7.50E-03 45.0 3.6
11 Cross MD 0.5 3.41E-03 102.2 8.2
12 Cross MD 1 2.67E-03 159.9 12.9
13 MD 0.1 8.51E-03 51.1 4.1
14 MD 0.5 4.60E-03 137.9 11.1
15 MD 1 3.40E-03 203.7 16.4
3600 psf
16 Cross MD 0.1 6.73E-03 40.4 3.3
17 Cross MD 0.5 3.38E-03 101.3 8.2
18 Cross MD 1 2.60E-03 155.9 12.6
Table G18. In-plane flow rate per width and hydraulic transmissivity of Product E
Specimen Hydraulic
Test Flow Hydraulic Seating Time Flow Rate
Normal size (width x Transmissivity
No. Direction gradient (min) (l/min-m ) (gpm/ft)
1 MD 0.1 1.11E-02 66.6 5.4
2 MD 0.5 5.18E-03 155.5 12.5
3 MD 1 3.83E-03 230.0 18.5
720 psf
4 Cross MD 0.1 8.65E-03 51.9 4.2
5 Cross MD 0.5 4.84E-03 145.2 11.7
6 Cross MD 1 3.35E-03 200.9 16.2
7 MD 0.1 1.20E-02 72.3 5.8
8 MD 0.5 6.39E-03 191.8 15.5
9 MD 1 4.21E-03 252.9 20.4
2000 psf 15 12 x 12
10 Cross MD 0.1 7.95E-03 47.7 3.8
11 Cross MD 0.5 4.62E-03 138.5 11.2
12 Cross MD 1 3.39E-03 203.1 16.4
13 MD 0.1 9.78E-03 58.7 4.7
14 MD 0.5 4.94E-03 148.3 11.9
15 MD 1 3.77E-03 226.4 18.2
3600 psf
16 Cross MD 0.1 7.81E-03 46.9 3.8
17 Cross MD 0.5 4.58E-03 137.3 11.1
18 Cross MD 1 3.49E-03 209.3 16.9
169

Product A (MD)
20.0
15.0
Flow rate gpm/ft
10.0 720 psf
2000 psf
5.0
3600 psf
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Product A (XMD)
20.0
15.0
Flow rate gpm/ft
10.0 720 psf
2000
5.0 psf
0.0
0 0.5 1 1.5
Gradient, i
Fig. G16. In-plane flow rate per width of Product A under normal stress of 720, 2000 and 3600
psf at gradient of 0.1, 0.5 and 1. (a) Specimen parallel to machine direction (MD), (b) specimen
cross to machine direction (XMD).
170

Product B (MD)
20.0
15.0
Flow rate gpm/ft
10.0
720 psf
5.0 2000 psf
3600 psf
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Product B (XMD)
20.0
15.0
Flow rate gpm/ft
10.0
720 psf
2000 psf
5.0
3600 psf
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Fig. G17. In-plane flow rate per width of Product B under normal stress of 720, 2000 and 3600 psf
at gradient of 0.1, 0.5 and 1. (a) Specimen parallel to machine direction (MD), (b) specimen cross
to machine direction (XMD)
171

Product C (MD)
20.0
15.0
Flow rate gpm/ft
10.0
720 psf
2000 psf
5.0 3600 psf
0.0
0 0.5 1 1.5
Gradient, i
Product C (XMD)
20.0
720 psf
15.0 2000 psf

Flow rate gpm/ft
3600 psf
10.0
5.0
0.0
0 0.5 1 1.5
Gradient, i
Fig. G18. In-plane flow rate per width of Product C under normal stress of 720, 2000 and 3600 psf
172

Product D (MD)
20.0
15.0
Flow rate gpm/ft
10.0
720 psf
2000 psf
5.0
3600 psf
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Product D (XMD)
20.0
15.0
Flow rate gpm/ft
10.0
720 psf
5.0
2000 psf
3600 psf
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Fig. G19. In-plane flow rate per width of Product D under normal stress of 720, 2000 and 3600
psf at gradient of 0.1, 0.5 and 1. (a) Specimen parallel to machine direction (MD), (b) specimen
cross to machine direction (XMD)
173

Product E (MD)
25.0
20.0
Flow rate gpm/ft

15.0
10.0 720 psf

2000 psf
5.0 3600 psf
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Product E (XMD)
25.0
20.0
Flow rate gpm/ft
15.0
10.0 720 psf

2000 psf
5.0 3600 psf
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Fig. G20. In-plane flow rate per width of Product E under normal stress of 720, 2000 and 3600 psf
174

Long Term In-Plane Flow Rate (100-hour seating period)
In plane flow rate was measured under a compressive stress of 3600 psf at a hydraulic gradient of
0.1, 0.5 and 1.0 for specimens subjected to compressive stress for 100 hours before taking the
measurement. Composite specimens (core + fabric) cut parallel to the machine direction (MD)
were used for these tests.
Results are summarized in Table G19. Plots of flow rate versus hydraulic gradient for the short
term (0.25 h) and long-term (100 h) seating period are plotted for comparison in Figures A21
through A25. There is no significant difference in short-term and long-term flow rates for the
Product A, Product D and Product E products. However, flow rates for the Product B and Product
C are not well matched for the 0.25 h and 100 h seating conditions.
Table G19. In-plane flow rate and hydraulic transmissivity (100-h seating period)
175

Product A
20.0
Flow rate gpm/ft 15.0
10.0
0.25 hr seating
5.0 time
100 hr seating time
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Fig. G21. Comparison of short- and long-term in-plane flow rates for Product A (3600 psf)
Product B
20.0
15.0
Flow rate gpm/ft
10.0
0.25 hr seating
time
5.0 100 hr seating
time
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Fig. G22. Comparison of short- and long-term in-plane flow rates for Product B (3600 psf)
176

Product C
20.0
10.0
0.25 hr seating
time
5.0 100 hr seating
time
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Fig. G23. Comparison of short- and long-term in-plane flow rates for Product C (3600 psf)
Product D
20.0
15.0
Flow rate gpm/ft
10.0 100 hr seating

time
5.0 0.25 hr seating
time
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Fig. G24. Comparison of short- and long-term in-plane flow rates for Product D (3600 psf)
177

Product E
20.0
10.0
0.25 hr seating time

5.0
100 hr seating time
0.0
0 0.2 0.4 0.6 0.8 1 1.2
Gradient, i
Fig. G25. Comparison of short- and long-term in-plane flow rates for Product E (3600 psf)
178

G7. Puncture Strength of Fabric
Tests were conducted following ASTM D4833/D4833M to measure index puncture resistance of
the fabric. Fabric samples were cut into a diameter of 10 cm (4 in) and clamped without tension
between circular plates of a ring clamp attachment secured in a compressive loading machine (Fig.
G26). Force was applied with a solid steel rod and recorded with a load indicator until rupture of
the specimen occurred. The maximum force was recorded as puncture strength of the specimen.
Measured puncture strengths for replicate specimens are summarized in Table G20.
Fig. G26. Testing equipment for puncture strength testing.
Table G20. Summary of puncture strength for replicate specimens
179

G8. Apparent Opening Size of Fabric
Tests were conducted following ASTM D 4751-12 to determine the apparent opening size (AOS)
of the geotextile fabrics. AOS indicates the largest particle that would effectively pass through the
geotextile. Five replicate circular specimens were cut to fit in a sieve (20 cm; 8 in diameter) along
a diagonal line on the fabric rolls. Specimens were placed in a sieve frame and size-fractioned
glass beads were placed on the geotextile surface. Using a mechanical sieve shaker, the specimen
and frame were shaken laterally for 15 minutes such that the jarring motion induced the beads to
pass through the specimen (Fig. G27). To prevent the accumulation of static electricity on the
surface of the specimen when the beads were shaken, static eliminators were attached to the cover
of the pan. Table G21 summarizes the results.
Fig. G27. Apparatus for apparent opening size: (a) specimen placed in sieve, (b and c) specimen
with glass beads, (d) mechanical sieve shaker.
Table G21. Summary of apparent opening size results.
180

Standards followed:
ASTM-DESIGNATION: D4491. “Standard Test Methods for Water Permeability of Geotextiles

by Permittivity.” Annual Book of ASTM Standards.
AST- DESIGNATION: D5199 -Determination of Nominal Thickness of Geosynthetics” Annual

Book of ASTM Standards.
ASTM-DESIGNATION: D4632. “Standard Test Method for Grab Breaking Load and Elongation
of Geotextile" Annual Book of ASTM Standards.
ASTM- DESIGNATION: D1621 “Standard Test Method for Compressive Properties of Rigid
Cellular Plastics” Annual Book of ASTM Standards.
ASTM- DESIGNATION: D4533 “Standard Test Method for Trapezoidal Tearing Strength of
Geotextile” Annual Book of ASTM Standards.
ASTM- DESIGNATION: D4716 “Standard Test Method for Determining In-Plane flow Rate of
Geosynthetics” Annual Book of ASTM Standards.
ASTM- DESIGNATION: D4833/D4833M “Standard Test Method for Index Puncture Resistance
of Geomembrane and Related Products” Annual Book of ASTM Standards.
ASTM- DESIGNATION: D4751. "Standard Test Method for Determining Apparent Opening Size
of Geotextile." Annual Book of ASTM Standards.
181

FR SJN135025 Clean1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

FR SJN135025 Clean1

Uploaded by

Copyright:

Available Formats

Performance Comparison of Abutment and

Retaining Wall Drainage Systems

State Job Number 135025

4. Title and Subtitle 5. Report Date (Month and Year)

7. Author(s) 8. Performing Organization Report No.

The University of Akron

15. Supplementary Notes

Unclassified Unclassified 180

Form DOT F 1700.7 (8-72) Reproduction of completed pages authorized

Junliang Tao, Ph.D.

Robert Liang, Ph.D., P.E.

Department of Civil Engineering

Prepared in cooperation with the Ohio Department of Transportation

I. Current state of the practice for structure drainage in Ohio

II. Current state of the practice for PCDS

• An online search was conducted to collect information on commercially available

III. Laboratory evaluation of select PCDS products

IV. Field implementation and evaluation

V. Cost analysis and comparison

1.1. Problem Statement

A prefabricated composite drainage system (PCDS) is proposed as an alternative drainage method

1) Develop specifications for prefabricated composite drainage systems.

1.3. Research Context

1.3.2. Criteria for a proper drainage system

1.3.3. Overview of drainage systems

Figure 2. Drainage method Type 3: prefabricated geocomposite drainage system

Core High Profile Core

Low Profile Sheet Core

Hydraulic Requirements: Hydraulic requirements ensure the drainage performance of the

Constructability Requirements: Constructability requirements prevent material damage or failure

1.4. Organization of the Report

The remaining portion of this report is organized as follows:

2.2. Modifications to CMS 518

For the porous material, modifications include:

2.3. Survey of County Engineers and Contractors

Figure 6. Cost of porous backfill with or without filter fabric.

Figure 7. Averaged weight of cost components for different drainage systems.

Figure 8. Performance of porous backfill with or without filter fabric.

Figure 9. Reported issues for porous backfill with filter fabric.

Figure 10. Maintenance work conducted for wall drainage systems.

2.3.4 Implementation of CMS 518

Figure 11. Averaged challenge rating in implementing CMS 518.

Figure 12. Benefits typically leading to the adoption of PCDS.

2.3.6 Survey of contractors

3.1. State-of-the-Practice of PCDS as Structural Drains

3.1.1 Commercial products

3.1.2 Design methodology

For vertical walls (Appendix E),

To choose a drainage product, the ultimate transmissivity should be calculated by considering a

θ ult = θ allow ( RFIN ⋅ RFCF ⋅ RFCC ⋅ RFBC )

Table 1. Recommended drainage reduction factors (Koerner and Koerner, 2007)

RF = RFIN ⋅ RFCF ⋅ RFCC ⋅ RFBC = 1.4 ⋅1.2 ⋅1.1⋅1.2 = 2.2

More discussions on selection of K is detailed in Appendix E.

3.1.3 DOT survey

Table 3. State DOT specifications on the core structure of PCDS

ASTM D ASTM D ASTM D ASTM D ARIZ 730/ASTM D ASTM D

South Carolina DOT -- 80 -- 50 150 40 50 -- 1 70

Virginia DOT -- 180 -- -- -- -- -- 50 0.5 --

3.2. Tire Derived Aggregates as Backfill and Drainage Layer

General applications and advantages

Cost and availability

Special considerations and potential disadvantages

4.1. Laboratory testing program and methods