Instrumentation of a geosynthetically reinforced roadway
by Joseph Andrew Lapeyre
A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Civil
Engineering
Montana State University
© Copyright by Joseph Andrew Lapeyre (1996)
Abstract:
Reinforcing flexible pavement roadways with geosynthetics has been proposed to reduce the thickness
of the base course layer. Mechanisms of reinforcement have been identified, but not quantified. A
design procedure that incorporates the reinforcement provided by geo synthetics would provide more
efficient use of aggregate and geosynthetics than those currently available. Research at Montana State
University has been initiated in this area. Its goal is to quantify the reinforcing benefit of geosynthetics
leading to the development of a roadway design procedure.
This thesis, which is the first phase of the research, is the study of possible strain sensors and
installation techniques that can be used to monitor the performance of a geosynthetically reinforced
roadway. With suitable strain sensors identified, the follow on phases of research at Montana State
University will quantify the reinforcing benefit of geosynthetics.
Research was conducted in two areas. A full scale reinforced flexible pavement roadway was built and
instrumented with vibrating wire, foil strain gauge, and LVDT technologies. The instruments were
monitored over a four month period while the roadway was subjected to heavy truck traffic. The
evaluation of mounting techniques used to fasten the strain sensors to the geosynthetics was
accomplished with the use of a wide width uniaxial tension facility.
Results from the study show that all three of the technologies are viable candidates for use in further
research. The effect of mounting techniques used to fasten the strain sensors to the geosynthetics was
seen to have a major impact on the strain measured by the transducer. Calibration factors were
developed to convert the strain measured by a sensor to the global strain in the geosynthetic. The need
to account for temperature effects regarding thermal strain and signal distortion was also identified. INSTRUMENTATION OF A GEOSYNTHETICALLY REINFORCED ROADWAY
by
Joseph Andrew Lapeyre
A thesis submitted in partial fulfillment
o f the requirements for the degree
of
Master o f Science
in
Civil Engineering
MONTANA STATE UNIVERSITY
Bozeman, Montana
December 1996
APPROVAL
of a thesis submitted by
Joseph Andrew Lapeyre
This thesis has been read by each member o f the thesis committee and has been found
to be satisfactory regarding content, English usage, format, citations, bibliographic style, and
consistency, and is ready for submission to the College of Graduate Studies.
Dr. Steven W. Perkins
/UU
VJ
(Signature)
Date
Approved for the Department o f Civil Engineering
Dr. Donald A. Rabem
LA.
(Signature)
Date
Approved for the College o f Graduate Studies
Dr. Robert L. Brown
(Signature)
Date
c,
Ill
STATEMENT OF PERMISSION TO USE
In presenting this thesis in partial fulfillment of the requirements for a master’s
degree at Montana State University-Bozeman, I agree that the Library shall make it available
to borrowers under rules of the Library.
If I have indicated my intention to copyright this thesis by including a copyright
notice page, copying is allowable only for scholarly purposes, consistent with “fair use” as
prescribed in the U.S. Copyright Law. Requests for permission for extended quotation from
or reproduction o f this thesis in whole or in parts may be granted only by the copyright
holder.
Signature
TABLE OF CONTENTS
Page
1. INTRODUCTION.............................. : ................................................................... I
Background and P ro b lem .....................................
I
Scope of W o r k ................................................
3
Outline o f Thesis ............................................................................................... 4
2. LITERATURE R E V IE W ............................................................................
5
In tro d u ctio n ........................................................................................................ 5
Strain T h e o ry .....................
5
Roadway Instrumentation ............................ ’......................................■.........6
Strain Sensors for Cemented M a te ria l............................................... 7
H -G a u g es.................................................................................. 7
Foil Strain Gauges on Carrier B lo c k ................................. 8
Foil Strain Gauges Mounted on Cores ............................... 9'
Strain Gauges for Granular M a te ria l................................................10
Inductance C o ils..................................................................... 10
LVDU s .................................................................................. 11
Multidepth Deflectometers ..................................................11
Instrumented R oadw ays.................................................................................. 14
Full Scale Tests ............................
..14
M N /R O A D .............................................................................. 14
Center for Transportation Research, Univ o f Texas .. i . 16
North Carolina Instrumented Highway .............................. 16
Alberta Instrumented Highway ........................................... 17
TRRL Instrumented Highway ............................................. 17
Laboratory T e sts.....................................................................
18
. Virginia Polytechnical In s titu te ........................................... 18
Danish Road Testing M ach in e.....................................
19
U.S. Army Engineer Waterways S ta tio n ............................21
University o f W aterloo....................................................
21
Tension Testing of Geosynthetics .................................................................22
In tro d u ctio n .........................................................................................22
ASTM S ta n d a rd s................................................................................23
Laboratory T e stin g ..............................................................................24
Related Instrumented Geosynthetic Studies ..................................................27
3. RESEARCH M ETH O D O LO G Y ....................................................................... 30
Introduction...................................
30
Instrumented Roadway . . . *.............
30
Instrument ty p e s .................................................................................. 30
V
Page
Asphalt C o n c re te ...................................................................30
Instrumentation in the Base C o u rs e ........................
32
Instrumentation on the Geosynthetics ................................ 34
Data Logging S y ste m s....................................................................... 38
Long Term Data L o g g er........................................................ 39
Dynamic Data Logger .......................................................... 39
Roadway M o n ito rin g ........... .............................................................. 40
Long Term Monitoring P ro g ra m .........................................40
Dynamic Monitoring P ro g ra m ............................................. 41
Roadway Construction and Gauge Locations ................................45
In tro d u ctio n ............................................................................45
Roadway Construction Overview ...................................... 46
Instrumentation Layout ........................................................47
Instrumentation Installatio n ...............................................................50
Instruments Attached to the Geosynthetics ....................... 50
Instruments Embedded in Base C o u rs e ........... .................. 51
Instruments Embedded in Asphalt C o n c re te ..................... 53
Wide Width Tension T e stin g ......................................................................... 54
Loading F ra m e ..............................................................................
54
Sensors Used for Local Strain M easurem ent............................
58
Tests P e rfo rm e d .................................................................................. 61
4. RESULTS..................................................................................................................63
Introduction................................... : .................................................................63
Wide Width Tension T e stin g ...................................................................... 63
Accuracy and Repeatability ............................................................... 63
Comparison o f Global Strain to Manufacturer’s Specifications . 64
Strain Response of the Vibrating Wire Strain G auges................... 64
Strain Response o f the Vibrating Wire Displacement Gauges .. 66
Strain Response o f the LVDT G a u g es............................................. 68
Strain Response in the Foil Strain G au g es.......................................68
Summary .............................................................................................70
Roadway Results ..................................
72
Long Term R esu lts..............................................................................72
Truck Traffic Loading .......................................................... 72
Strain in the Geosynthetics ................................................. 72
Strain Gauges in the Base Course ..................... . . . . . . . . 7 6
Strain in the A C .................................................................. . 7 8
Dynamic Testing R e s u lts ........................................................
79
Truck Pass Tests ................................................................... 79
Road Rater Tests ................................................................... 81
Vl
Page
5. CONCLUSIONS AND REC O M M EN D A TIO N S............................................ 85
C onclusions............................
85
Recom m endations.................................................... : .......................................86
REFERENCES CITED ................................
89
A P P E N D IC E S ........... ...........................................
.9 2
A Wide Width Tension Testing Results .........................................
93
B Roadway Results ........................................................................................... 116
V ll
LIST OF TABLES
Table
Page
1. Chronological Order o f Events For Roadway
Construction and Testing ........................................................................... 42
2. Vehicles Used for Truck Pass Tests
.............................................................. 44
3. Instrumentation Specifications.......................
50
4. Geosynthetic Instrument Cover Options ...............................
52
V lll
LIST OF FIGURES
Figure
Page
1. Dynatest H-Gauge ..............................................................................................8
2. Foil Strain Gauge mounted in a Carrier B lo c k .............................................. 9
3. L V D T ...................................................................................................
12
4. Cross section of MDD after installation ......................................................... 13
5. The Danish Road Testing Machine ............................................................... 20
6. ASTM approved Sanders C la m p ............................................................... . .25
7. ASTM approved Wide Width Clamp ............................................................. 26
. 8. Vibrating Wire Embedment AC Strain Gage
(Geokon Model VCE-4200-HT) . . ........................................
31
9. Vibrating Wire Embedment Strain Gage
(Geokon Model VCE-4200) ......................................................................... 32 .
10. Vibrating Wire Embedment Displacement Gage'
(Geokon Model 4430) ................................................................................. 33
11. LVDT Embedment Displacement Gage
(RDP Electronics Model D 5/400W ).........................................
34
12. Vibrating Wire Strain Gage
(Geokon Model V SM -4000)...................
35
13. Vibrating Wire Displacement Gage
(Geokon Model 4 4 2 0 ).................................................................................... 36
14. LVDT Displacement Gage
(RDP Electronics Model D 5/200W )..................................................
37
15. Plan View o f Roadway Test Sight ...............
48
16. Roadway Instrumentation Layout .................................................................49
ix
Figure
Page
17. Schematic o f Wide Width Tension Testing Frame ...................................56
18. Dimensional Placement o f Instruments on a Geosynthetic Specimen . . . 60
19. Global Strain From Two Sets o f Celesco Gages: Geogrid,
Machine Direction .........................................................................................94
20. Global Strain From Two Sets o f Celesco Gages: Geogrid,
Transverse D irectio n ................................................................................
94
21. Global Strain From Two Sets o f Celesco Gages: Geotextile,
Machine Direction ......................................................................................... 95
22. Global Strain From Two Sets o f Celesco Gages:
Geotextile, Transverse Direction ...........................................
95
23. Global Strain From Two Tests:
' Geogrid, Machine Direction ............................ ; .......................................96
24. Global Strain From Two Tests:
Geogrid, Transverse D irection ................................................
25. Global Strain From Two Tests:
Geotextile, Machine D irection..........................................................
26. Global Strain From Two Tests:
Geotextile, Transverse D ire c tio n .................................................................97
27. Comparison o f Results, to M anufacturer’s Data:
Geogrid, Machine Direction ............... : ....................................................98
28. Comparison o f Results to Manufacturer’s Data:
'Geogrid, Transverse Direction . . ' ........... , ..................................................98
29. Comparison o f Results to Manufacturer’s Data:
Geotextile, Machine D irection.....................................................................99
30. Comparison o f Results to Manufacturer’s Data:
Geotextile, Transverse D ire c tio n ......................................... ...........■..........99
96
97
X
Figure
Page
3 1. Vibrating Wire Strain Gage:
Geogrid, Machine Direction ................................................................... 100
32. Vibrating Wire Strain Gage:
Geogrid, Transverse D irection................................................................. 100
33. Vibrating Wire Strain Gage:
Geotextile, Machine D irection........... ......................................................101
34. Vibrating Wire Strain Gage:
Geotextile, Transverse D ire c tio n .............................................................. 101
35. Calibrated Vibrating Wire Strain Gage:
Geogrid, Machine Direction ....................................................................102
36. Calibrated Vibrating Wire Strain Gage:
Geogrid, Transverse Direction ................................................................. 102
37. Calibrated Vibrating Wire Strain Gage:
Geotextile, Machine D ire c tio n ................................................................. 103
38. Calibrated Vibrating Wire Strain Gage:
Geotextile, Transverse Direction .............................................................103
39. Comparison o f Vibrating Wire Strain Gage
W ith 6, 4 and 2 Bolts F astened................................................................. 104
40. Vibrating Wire Displacement Gage:
Geogrid, Machine D ire c tio n .........■......................................................... 104
41. Vibrating Wire Displacement Gage:
Geogrid, Transverse Direction ................................................................. 105
42. Vibrating Wire Displacement Gage:
Geotextile, Machine D ire c tio n ....................................................
43. Vibrating Wire Displacement Gage:
Geotextile, Transverse Direction .................
.105
106
44. Calibrated Back-to-Back Vibrating Wire Displacement Gage:
Geogrid, Machine D ire c tio n ..................................................................... 106
xi
Figure
Page
45. Calibrated Back-to-Back Vibrating Wire Displacement Gage:
Geogrid, Transverse Direction ................................................................. 107
46. Calibrated Back-to-Back Vibrating Wire Displacement Gage:
Geotextile, Machine D ire c tio n ................................................................. 107
47. Calibrated Back-to-Back Vibrating Wire Displacement Gage:
Geotextile, Transverse Direction .............................................
108
48. LVDT Displacement Gage: Geogrid, Machine D ire c tio n ............. ..
108
49. LVDT Displacement Gage: Geogrid, Transverse Direction ...............
109
50. LVDT Displacement Gage: Geotextile, Machine Direction ................... 109
51. LVDT Displacement Gage: Geotextile, Transverse Direction ............. 110
52. Calibrated LVDT Displacement Gage:
Geogrid, Machine Direction .....................................
HO
53. Calibrated LVDT Displacement Gage:
Geotextile, Machine D ire c tio n ................................................................. I l l
54. Calibrated LVDT Displacement Gage:
Geotextile, Transverse Direction .............................................................I l l
55. Foil Strain Gage: Geogrid, Machine Direction ......................................... 112
56. Foil Strain Gage: Geogrid, Transverse D irection.......................................112
57. Calibrated Foil Strain Gage: Geogrid, Machine D irection........................113
58. Calibrated Foil StrainGage: Geogrid, Transverse D irectio n ................... .113
59. Unloading-Reloading Response From Foil Strain Gage:
Geogrid, Transverse Direction ................................................................. 114
60. Calibrated 1/4 Bridge Foil Strain Gage:
Geogrid, Machine Direction . ....................................................................1,14
xii
Figure
Page
61. Calibrated Foil Strain Gage With Environmental Protection:
Geogrid, Machine D ire c tio n ................................................................... 115
62. Calibrated Foil Strain Gage With Environmental Protection:
Geogrid, Transverse Direction .................................: ............................115
63. Daily Traffic Loading H is to ry .................................................................... 117
64. Weekly Truck Traffic Loading H is to r y .....................................................117
65. VW Displacement Gage #2 on Geogrid (on wheel-path) ........................ 118
66. VW Displacement Gage #1 on Geotextile (off wheel-path) ................... 118
67. VW Strain Gage #6 on Geogrid (off wheel-path) .........'......................... 119
68. VW Strain Gage #5 on Geotextile (on w heel-path).................................119
' 69. LVDT Displacement Gage # 31 on Geogrid (on wheel-path) ................. 120
70. LVDT Displacement Gage #32 on Geogrid (off wheel-path) ................. 121
71. LVDT Displacement Gage #29 on Geotextile (on wheel-path) . . . . . . . 1 2 2
72. LVDT Displacement Gage #30 on Geotextile (off wheel-path) .............123
73. VW Embedment Displacement Gage #3 in Base
Above Geogrid (off w heel-path)............................................................. 124
74. VW Embedment Displacement Gage #4 in Base
in Non-Reinforced Section (on wheel-path) ..........................................124
75. VW Embedment Strain Gage #8 in Base
Above Geogrid (on wheel-path) ..........'..................................................125
76. VW Embedment Strain Gage #7 in Base
Above Geotextile (on w heel-path).......................................................... 125
xiii
Figure
Page
77. LVDT Embedment Displacement Gage #27 in Base
Above Geogrid (off wheel-path) .............................. .............................. 126
78. LVDT Embedment Displacement Gage #28 in Base
Above Geogrid (on w heel-path)............................................................... 127
.79. LVDT Embedment Displacement Gage #25 in Base
Above Geotextile (on w heel-path)....................................... .'.................128
80. LVDT Embedment Displacement Gage #26 in Base
in Non-Reinforced Section (on wheel-path) ......................................... 129
8 1. VW Embedment Strain Gage #10 in AC
Above Geogrid (on wheel-path) ............. ............................................... 130
82. VW Embedment Strain Gage #11 in AC
Above Geogrid (off wheel-path) ............................................................. 131
83. VW Embedment Strain Gage #9 in AC Above Geotextile
(on w heel-path)........................................................................ •.................132'
84. VW Embedment Strain Gage #12 in AC
in Non-Reinfbrced Section (on wheel-path) ! .................................
133
85. Foil Strain Gage #36 on Geogrid (on wheel path):
Truck Pass Test I ...........: ........................................................ , ...............134
86. Foil Strain Gage #35 on Geogrid (below centerline):
Truck Pass Test I ....................................................................................... 134
87. Foil Strain Gage #36 on Geogrid (on wheel path):
Truck Pass Test 2 ....................................................................................... 135
88. Foil Strain Gage #35 on Geogrid (below centerline):
Truck Pass Test 2 ...................................................................... ’.................135
89. Foil Strain Gage #36 on Geogrid (on wheel path):
Truck Pass Test 3 .............................. :•...................................................... 136
90. Foil Strain Gage #35 on Geogrid (below centerline):
Truck Pass Test 3 ............................................ ...........................................136
XlV
Figure
Page
91. Foil Strain Gage #34 on Geogrid (on wheel path):
Truck Pass Test 3 .....................................................................
137
92. Foil Strain Gage #36 on Geogrid (on wheel path):
Truck Pass Test 4 ....................................................................
137
93. Foil Strain Gage #35 on Geogrid (below centerline):
Truck Pass Test 4 .......................
138
94. Foil Strain Gage #34 on Geogrid (on wheel path):
Truck Pass Test 4 .........................................................................................138
95. Foil Strain Gage #36 on Geogrid (on wheel path):
Truck Pass Test 5 .........................................................................................139
96. Foil Strain Gage #35 on Geogrid (below centerline):
Truck Pass Test 5 ......................................................
139
97. Foil Strain Gage #34 on Geogrid (on wheel path):
Truck Pass Test 5 ..............................................................
140
98. Foil Strain Gage #36 on Geogrid (on wheel path):
Truck Pass Test 6 .........................................................' .............................140
99. Foil Strain Gage #35 on Geogrid (below centerline):
Truck Pass Test 6 ...................
141
100. Foil Strain Gage #34 on Geogrid (on wheel path):
Truck Pass Test 6 .........................................................................................141
101. Foil Strain Gage #36 on Geogrid (on wheel path):
Truck Pass Test 7 ...........................................................■...........................142
102. Foil Strain Gage #35 on Geogrid (below centerline):
Truck Pass Test 7 .........................................................................................142
103. Foil Strain Gage #34 on Geogrid (on wheel path):
Truck Pass Test 7 .......................-................... ...........................................143
104. Foil Strain Gage #36 on Geogrid (on wheel path):
Truck Pass Test 8 .........................................................................................143
XV
Figure
Page
105. Foil Strain Gage #35 on Geogrid (below centerline):
Truck Pass Test 8 ......................................................................................... 144
106. Foil Strain Gage #36 on Geogrid (on wheel path):
Truck Pass Test 9 .........................................................................................144
107. Foil Strain Gage #35 on Geogrid (below centerline):
Truck Pass Test 9 ......................................................................................... 145
108. LVDT Displacement Gage #31 on Geogrid (on wheel path):
Truck Pass Test 4 .......................................................................................145
109. LVDT Displacement Gage #31 on Geogrid (on wheel path):
Truck Pass Test 5 .......................................................................................146
HO. LVDT Displacement Gage #32 on Geogrid (off wheel path):
Truck Pass Test 5 .....................................................................................146
111. LVDT Displacement Gage #29 on Geotextile (on wheel path):
Truck Pass Test 4 ................................................................................ . . 1 4 7
112. LVDT Displacement Gage #29 on Geotextile (on wheel path):
Truck Pass Test 5 .................................................................................. 147
113. LVDT Displacement Gage #29 on Geotextile.(on wheel path):
Truck Pass Test 6 .................................................................................. 148
114. LVDT Embedment Displacement Gage #27 in Base
Above Geogrid (off wheel path): Truck Pass Test 3 ........................ 148
115. LVDT Embedment Displacement Gage #27 in Base
Above Geogrid (off wheel path): Truck Pass Test 6 ........................ 149
116. LVDT Embedment Displacement Gage #27 in Base
Above Geogrid (off wheel path): Truck Pass Test 7 ........................149
117. LVDT Embedment Displacement Gage #28 in Base
Above Geogrid (on wheel path): Truck Pass Test 2 ........................ 150
118. LVDT Embedment Displacement Gage #28 in Base
Above Geogrid (on wheel path): Truck Pass Test 3 ..........................150
I
XVl
Figure
Page
119. LVDT Embedment.Displacement Gage #28 in Base
Above Geogrid (on wheel path): Truck Pass Test 4 ............... . . . . 1 5 1
120. LVDT Embedment Displacement Gage #28 in Base
Above Geogrid (on wheel path): Truck Pass Test 5 ............. ..
151
121. LVDT Embedment Displacement Gage #28 in Base
Above Geogrid (on wheel path): Truck Pass Test 6 ........................ 152
122. LVDT Embedment Displacement Gage #25 in Base
Above Geotextile (on wheel path): Truck Pass Test I ................... 152
123. LVDT Embedment Displacement Gage #25 in Base
Above Geotextile (on wheel path): Truck Pass Test 2 ....................153
124. LVDT Embedment Displacement Gage #25 in Base
Above Geotextile (on wheel path): Truck Pass Test 3 .........i . . . . 153
125. LVDT Embedment Displacement Gage #25 in Base
Above Geotextile (on wheel path): Truck Pass Test 4 ............... . . 1 5 4
126. LVDT Embedment Displacement Gage #25 in Base
Above Geotextile (on wheel path): Truck Pass Test 5 ....................154
127. LVDT Embedment Displacement Gage #25 in Base
Above Geotextile (on wheel path): Truck Pass Test 6 ....................155
128. LVDT Embedment Displacement Gage #25 in Base
Above Geotextile (on wheel path): Truck Pass Test 7 ....................155
129. LVDT Embedment Displacement Gage #26 in Base
o f Non-Reinforced Section (on wheel path): Truck Pass Test I . . . 156
130. LVDT Embedment Displacement Gage #26 in Base
o f Non-Reinforced Section (on wheel path): Truck Pass Test 2 . . . 156
131. LVDT Embedment Displacement Gage #26 in Base
of Non-Reinforced Section (on wheel path): Truck Pass Test 3 . . . 157
132. LVDT Embedment Displacement Gage #26 in Base
of Non-Reinforced Section (on wheel path): Truck Pass Test 4 . . . 157
xvii
Figure
Page
133. LVDT Embedment Displacement Gage #26 in Base
o f Non-Reinforced Section (on wheel path): Truck Pass Test 5 . . . 158
134. LVDT Embedment Displacement Gage #26 in Base
o f Non-Reinforced Section (on wheel path): Truck Pass Test 6 . . . 158
135. Foil Strain Gage #34 on Geogrid,
July 31st T e s t ...........................159
136. Foil Strain Gage #35 on Geogrid,
July 31st Test ...........................159
137. Foil Strain Gage #36 on Geogrid,
July 31st T e s t ...........................160
138. Foil Strain Gage #34 on Geogrid,
September 21st T e s t.............
139. Foil Strain Gage #35 on Geogrid,
September 21st T e s t............... 161
140. Foil Strain Gage #36 on Geogrid,
September 21st T e s t................ 161
160
141. VW Displacement Gage #2 on Geogrid, September 21st Test ...........162
142. VW Strain Gage #5 on Geotextile, September 21st T e s t......................162
143. VW Embedment Displacement Gage #3 in Base
Above Geogrid, September 21st Test .................................................163
144. VW Embedment Strain Gage #7 in Base
Above Geotextile, September 21st T e s t..............................................163
145. VW Embedment Strain Gage #9 in AC
Above Geotextile, September 21st T e s t ..............................................164
146. Resilient Modulus Values From July 21 Road Rater T e s t....................164
147. Average Resilient Modulus Values
Frbm July 21 Road Rater T e s t ............................................................. 165
148. Resilient Modulus o f AC Layer
From September 21 Rpad Rater T e s t.................................................. 165
149. Resilient Modulus o f Base and Subgrade Layers
From September 21 Road Rater T e s t.................................................. 166
XVlll
Figure
Page
150. Average Resilient Modulus o f AC Layer
From September 21 Road Rater T e s t.................................................. 166
151. Average Resilient Modulus of Base and Subgrade Layers
From September 21 Road Rater T e s t.................................................. 167
4
XlX
ABSTRACT
Reinforcing flexible pavement roadways with geosynthetics has been proposed to reduce
the thickness o f the base course layer. Mechanisms o f reinforcement have been identified,
but not quantified. A design procedure that incorporates the reinforcement provided by
geo synthetics would provide more efficient use o f aggregate and geosynthetics than those
currently available. Research at Montana State University has been initiated in this area. Its
goal is to quantify the reinforcing benefit o f geosynthetics leading to the development o f a
roadway design procedure.
This thesis, which is the first phase o f the research, is the study o f possible strain sensors
and installation techniques that can be used to monitor the performance o f a geosynthetically
reinforced roadway. With suitable strain sensors identified, the follow on phases of research
at Montana State University will quantify the reinforcing benefit o f geosynthetics.
Research was conducted in two areas. A full scale reinforced flexible pavement roadway
was built and instrumented with vibrating wire, foil strain gauge, and LVDT technologies.
The instruments were monitored over a four month period while the roadway was subjected
to heavy truck traffic. The evaluation of mounting techniques used to fasten the strain
sensors to the geosynthetics was accomplished with the use o f a wide width uniaxial tension
facility.
Results from the study show that all three o f the technologies are viable candidates for
use in further research. The effect o f mounting techniques used to fasten the strain sensors
to the geosynthetics was seen to have a major impact on the strain measured by the
transducer. Calibration factors were developed to convert the strain measured by a sensor
to the global strain in the geosynthetic. The need to account for temperature effects
regarding thermal strain and signal distortion was also identified.
I
CHAPTER I
INTRODUCTION
Background and Problem
The use o f geosynthetics (geotextiles and geogrids) has become prevalent in
geotechnical and transportation projects in recent years. It has been proposed that it may be
possible to reinforce the base course layer in paved roadways with geosynthetics. If true, this
would be o f great benefit in certain projects. Some areas, such as Eastern Montana, have
limited natural sources o f quality aggregate for base course construction. It often is cost
prohibitive to transport aggregate to these areas. For these areas, reducing base course
thickness with reinforcement through the use o f geosynthetics may be a cost effective
alternative to bringing in outside aggregate.
Currently, the improvement in performance gained by adding geosynthetics to the
base layer reinforcement is poorly understood. Studies have shown conflicting results with
regard to geosynthetic reinforcement o f base courses in paved roadways. This situation is
partially due to the lack of understanding o f the mechanical performance o f geosynthetically
reinforced roadways. Research at Montana State University has been initiated in this area
with the goal o f describing the reinforcement performance o f geosynthetics in paved
roadways and providing a design tool for geosynthetically reinforced flexible pavements.
The research program at Montana State University involves several areas of study.
To examine the interaction between geosynthetics and base course, a confined geosynthetic
2
pullout facility has been constructed. The facility is a 6 walled steel box in which a
geosynthetic is sandwiched in base material. An air bladder provides confining pressure to
the top o f the base course material. Once the desired level o f confinement is achieved, the
geosynthetic is pulled through the box by means o f a low geared electric motor. Strain
instrumentation on the geosynthetic and a load cell in the pull out mechanism provides
measurements o f the geosynthetic performance. In addition to providing information on the
interaction between the geosynthetic and base course, creep and stress relaxation effects in
geosynthetic may be studied.
To identify conditions under which geosynthetic reinforcement is seen to offer
improvement, a cyclic load plate facility is being constructed. This facility is a reinforced
concrete box in which a roadway structure is built and instrumented. A pneumatically
activated circular plate mounted on a steel H beam is used to apply loads to the roadway.
Through the use o f rollers, the steel plate will be able to move over the surface o f the
roadway to apply the load at specific locations. The performance o f the roadway structure
will help identify types o f test sections that are necessary to advance the understanding of
geosynthetic reinforcement.
W ith knowledge gained from the two laboratory tests a design procedure will be
established. Finite element analysis will be important in developing the design procedure.
A finite element model o f the cyclic load plate facility will be created and verified against
the measured performance. The model will then be used to develop the design procedure.
Once the design procedure is established, it will be verified by constructing a full scale test
road.
3
To study the mechanisms of reinforcement in a geosynthetically reinforced roadway
the mechanical performance o f the roadway must be measured. Parameters such as stress,
strain, pore water pressure, moisture content, and temperature need to be measured at select
locations in the roadway cross section. For many geotechnical structures involving only
granular materials and long term static loading, measurement o f these parameters is well
established. For geosynthetics and roadway studies however, the methodology is not as
clearly established. In addition to the granular base course o f the roadway, measurements
need to be made in the asphalt concrete (AC) pavement, and on the geosynthetic. Long term
measurements as well as short term dynamic responses from individual axle loadings need
to be made. During the initial research at Montana State University, particular concerns
existed regarding strain measurements. The appropriate strain gauges for use in roadways
and attendant installation techniques had to be determined before the research could
continue. Therefore the topic o f this study is examination o f the proposed strain measuring
instrumentation and installation techniques for use in follow on research.
Scope o f Work
The study consists o f two areas. First, a fully instrumented geosynthetic reinforced
roadway was constructed and monitored. The roadway study provided information in two
areas. Gauge performance, including both strain measurement and reliability was evaluated.
Second, installation techniques for the strain gauges were evaluated. The roadway carried
over 7000 passes o f heavy truck traffic during its 4 month life. It consisted o f a geogrid
reinforced section, a geotextile reinforced section, and a control section. The roadway
4
contained a total o f 24 strain gauges located in the AC, base course, and mounted on the
geosynthetics. Two types o f strain measurements were made. Static long term cumulative
strain measurements were recorded over the life time o f the roadway as well as dynamic
short term measurements o f specific events, such as the loading from a single axle.
To support the in-field study, a wide width tension testing facility was constructed
in the laboratory.
This facility was used to study the strain gauges mounted on the
geosynthetics. The facility was able to test geosynthetic specimens up to 1.83 m wide and
3 m long. Tests in the facility covered three objectives. First, it allowed comparison o f the
local strain in a geosynthetic (measured by a strain gauge) to the global strain experienced
by the geosynthetic when subjected to a uniaxial load. Second, the mounting techniques used
for the strain gauges in the field were evaluated. Third, the stress strain response o f the
geosynthetic provided by the manufacturer was verified.
Outline o f the Thesis
The thesis is organized into 5 chapters. Chapter 2 is a literature review to provide
background information concerning instrumentation o f roadways and geosynthetics. Its main
topics are strain gauge types, previous instrumented roadway studies, and tension testing o f
geosynthetics. Chapter 3 provides a detailed discussion of the experimental work performed
by the author. Topics include instrumentation used in the study, roadway construction, and
the wide width tension testing facility. Chapter 4 provides results o f the study. Chapter 5
is devoted to conclusions and recommendations for further work.
5
CHAPTER 2
LITERATURE REVIEW
Introduction
The literature review consists of five sections. In section one, a brief presentation o f
the calculation of strain is made. Section two presents the types o f strain sensors used in
roadway studies. Previous instrumented roadway studies from both the field and laboratory
are discussed in section three. Section four looks at the ASTM tension test for geosynthetics
and the significant parameters for tension tests in general. Section five discusses other
geosynthetic instrumented projects with regards to bonding foil strain gauges to geogrids.
Strain Theory
Strain (e) is a measurement o f the relative change in length o f an object. It is defined
as the ratio o f the change in length (8L) o f an object to its original length (L). This
relationship can be written as:
■
- T
The relationship of the strain o f an object to the force causing the strain is in general non­
linear. For many engineering situations however, the magnitude o f the strains is small,
covering only the linear portion o f the material’s stress-strain diagram. Hooke’s Law
governs the stress-strain relationship of a material in this region. It states that the stress is
6
directly proportional to the strain. It is written as the following:
o = eE
'
(2)
.
■
The importance of this relationship is that it allows the calculation o f stress (d) based on the
measured strain and the material property E, the modulus o f elasticity.
Several technologies are used to measure strain. Electrical and mechanical are most
common but optical, acoustical, and pneumatic strain sensors exist.
Regardless o f the
phenomena, all strain gauges rely on the measurement o f the displacement between two
points a known distance apart and then the computation o f strain using equation I .
Roadway Instrumentation
Due to the different materials in a roadway structure, there are several types o f strain
sensors typically used. In granular material the technology used is different from that used
in cemented material such as AC or concrete. The profile o f a strain sensor for granular
material is large in order to produce an accurate measurement. Strain sensors in cemented
material do not always have this luxury due to the small thickness o f the surface material.
Additionally, the installation o f strain sensors in cemented material typically occurs when
the hot mix is placed. This requires high temperature resistance o f both the instrument and
its electrical cables.
7
Strain Sensors for Cemented Material
H-Gauges. H-gauges are given their name from their appearance (See figure I).
They consist o f a strip o f material onto which a foil strain gauge is bonded. On each end o f
the strip, metal bars are fixed at their midpoint. The metal bars serve as anchors in the
cemented material. Thus when the material is strained the metal bars move relative to each
other and the resulting strain in the strip material is measured by the foil strain gauge. The
length o f the gauge should depend on the size o f the aggregate in the AC. For pavement
materials it is commonly believed that the gauge length should be at least 4 to 5 times the
maximum particle diameter (Van Deusen, 1992). To ensure that accurate strain readings are
being made it is important that the stiffness o f the strip be the same or less than that o f the
cemented material. If it is greater, the H-gauge will under measure the actual strain o f the
cemented material.
The first H-gauges were designed by the Transportation Road Research Laboratory
(TRRL) in Great Britain (Sebaaly, 1989). Important aspects of their evolution include better
matching of gauge and material stiff nesses, and survivability of the foil strain gauge. To this
end current H-gauges manufactured by Dynatest Corporation use epoxy reinforced fiberglass
as the strip material (Van Deusen, 1992). Fiberglass has the desired properties o f low
stiffness and high flexibility. To increase survivability, the foil strain gauge is embedded in
the fiberglass. Additionally, various layers o f waterproofing compounds and aluminum
plates are placed around the fiberglass strip.
8
PFT-sleeve
ZO
Titanium plate
Silicone
stainless steel Anchor
Teflon
Antiosmotic base
^^f^ingauge embed-
f ir e g l a s s rein-
Antiosmotic base"
Silicone rubber
Titanium plate
Asphalt Coating
Figure I. Dynatest H-Gauge (from Van Deusen et al., 1992)
Foil Strain Gauges on Carrier Block. Carrier blocks are laboratory compacted AC
bricks to which foil strain gauges have been attached. When the hot mix lift is being
constructed, the carrier block is placed on the base course in the proper location and
orientation. As the hot mix surrounds the carrier block the heat softens the carrier block and
the hot mix and carrier block bond forming a monolithic layer of AC.
9
A drawback o f earner blocks is protection of the foil strain gauge when the hot mix
softens the block.
Debonding and mechanical damage of the gauge are possible.
A
technique to overcome this dilemma is to cut the carrier block in the lab and mount the foil
strain gauge on the cut surface as seen in figure 2. The carrier block is then glued back
together. The location o f the foil strain gauge in the middle o f the carrier block gives it
protection from aggregate in the hot mix. The location also minimizes temperature change
around the gauge, which can be detrimental to the bonding agent.
t
4 cm
I
4 era
I
I.Scm
I
Figure 2. Foil Strain Gauge mounted in a Carrier Block (from Sebaaly et al., 1989)
Foil Strain Gauges Mounted on Cores. This technique is similar to a carrier block
except that the core is cut out o f the existing roadway. Bonding the core back to the AC is
a significant problem. The stiffness of the bonding agent and that o f the AC must be closely
10
matched. Ifthe bonding agent is soft relative to the AC the core tends to act as a rigid body
immersed in a monolithic material. If the bonding agent is stiff relative to the AC, stress
j
concentrations will occur in the core, which induce cracking and premature failure o f the
core. A major advantage o f using cores is that the instruments can be easily placed after the
roadway is constructed.
Strain Gauges for Granular Material
Inductance Coils. Inductance coils are two disc shaped coils that produce an
electromagnetic output proportional to the distance between them. They are available
commercially from Bison Instruments, Inc. The discs may be oriented coplanar, orthogonal,
or most commonly coaxial. Typical distance between the disks is I to 4 times the diameter
o f a single disk (Selig, 1975). One o f the disks acts as a transmitter and the other as a
receiver. As the distance between them changes, the response o f the coil in the receiver
changes which is converted to a displacement. Inductance coils have a unique advantage in
that the two discs are not mechanically connected. Therefore disturbance and altering o f the
tested material is not an issue. They have good stability for long term measurements, but are
subject to error when used for certain dynamic measurements. Poor performance in dynamic
measurements can be caused by the movement o f the vehicle applying the load through the
gauge’s electromagnetic field. The metal and electrical system o f the vehicle induces
changes in the field o f the gauge.
The strain resolution for long-term and dynamic
measurements is 0.003 % and 0.1 % respectively (Brown, 1977). Inductance coils appear
to have excellent durability: Selig (1975) notes a roadway study in which less than 1% of
11
408 gauges were inoperative after one year in service.
LVDT. Linear Variable Differential Transformers (LVDT) use electro-mechanical
effects to measure displacement. The gauge operates by moving a rod shaped core through
a cylinder containing three symmetrically spaced electrical coils. The center coil, called the
primary coil, creates an electrical field with power from an external source. As the core
moves through the electrical field it provides a pathway for magnetic flux to induce a voltage
in the two secondary coils. The induced voltages are directly proportional to the position o f
the core allowing displacement and subsequent strain measurements to be made.
A typical LVDT is shown in figure 3. The gauges are composed o f a rod shaped core
with an attached disk and a cylinder with a similar attached disk. As the granular material
shifts, the discs are forced to move which causes displacement o f the core relative to the
coils.
LVDT’s are available in both DC and AC models. AC models are more common in
situations where environmental protection is an issue because the electronic circuitry can be
housed separate from the gauge. Separating the electronics from the gauge also allows a
physically smaller unit to be used.
Multidepth Deflectometer.
Multidepth deflectometers (MDD) are a series of
connected LVDTs used to measure deflections at various depths in a roadway structure
(Pilson, 1995). A roadway cross section instrumented with an MDD is shown in figure 4.
3 core screened cable
I
PT.FE. bearing
PT. FE.
covered piston
Actuating rod
Lld securing
screws
120 mm
Figure 3. LVDT (from Potter et al, 1969)
6
13
CONNECTOR
CONNECTOR CABLE.'
POLY-URETHANE CASTING COMPOUND
FLEXIBLE SURFACE CAP
MULTIDEPTH DEFLECTOMETER MODULES
SNAP CONNECTOR
SNAP HEAD (POSITIVE LOCKING)
■
ANCHOR EXTENSION
12 IN.
N o t to S c a le
ANCHOR
Figure 4. Cross section of a MDD after installation (from Pilson, 1995)
14
A MDD is installed by the following procedure, After drilling a core hole, a rubber
wall liner is placed. Then an anchor is cemented into the bottom o f the core hole. The
lowest MDD module is then lowered into the core and tightened against the core wall at the
desired depth. An inter-connecting rod is fastened from the module to the anchor at the base
o f the core. Successive modules are then lowered into the core hole, fixed against the wall
and connected to the previous module. Ducting in each module allows for the wiring to run
up the core hole. This technique allows measurements to be taken at any depth in the
roadway. The MDD can measure either relative elastic deformations between points or total
deformations from the anchor depth.
Instrumented Roadways
Full Scale Tests
Mn/ROAD. Mn/Road is the largest and most extensive instrumented roadway in the
world (WWW, 1996). Located 64 km northwest o f Minneapolis, Minnesota on 1-94, the
facility has 8.8 km o f roadway instrumented with over 4500 sensors. Opened in 1994, the
project is expected to have a life time of 20 years. The primary objective o f the facility is to
provide findings which will help to develop better fundamental approaches to pavement
design and maintenance (Newcomb, 1989). The facility incorporates a variety of testing
options. A 4.8 km long high-volume traffic section exists on the west bound lane of 1-94.
The design average daily traffic for the roadway is 24,200 vehicles, 3200 o f which are
■
•
classified as heavy commercial vehicles. A weigh in motion (WIM) device is located at the
15
start o f the high-volume traffic section to record the actual count o f heavy vehicle traffic.
The low-volume traffic roadway is a 4 km closed loop road north o f the interstate. Loading
for it consists o f a known number o f truck passes at the legal weight limit (355 kN) and
above the legal limit (454 kN).
The Mn/Road facility uses AC and portland cement concrete over 6 types o f
aggregate bases. The roadway structure does not contain any geosynthetics. The existing
subgrade in the area is a loam with an AASHTO classification of A-6 and an R value on the
order o f 15. In addition to the native subgrade, the low-volume road will import a heavy clay
with an R value of 5 for 3 of its test sections. Ground water levels over the entire site vary
from 7.62 cm above the grade to a depth o f 4.26 m.
Strain instrumentation o f Mn/Road is located in the cemented surfaces o f the
roadway. Six types o f strain gauges are utilized. One type is the Alberta Research Council
biaxial gauges, which consist of 4 foil strain gauges embedded in asphalt mastic mix. Two
o f the foil strain gauges are oriented in the longitudinal direction while the other two are
rotated 90° to measure strain in the transverse direction. A total o f forty o f these gauges were
installed. H-gauges consist o f 230 Dynatest PAST-2AC/CC gauges. The most numerous
gauge is the Tokyo Sokki PML-60 of which 436 were installed. This gauge is similar to an
H-gauge with the removal o f the two metal bars on each end. A coarse grit coats the surface
o f the gauge to help bond the gauge to the cemented material. Vertical displacements are
measured by Schaevitz HCD-500 L V D T S . One-hundred and sixty two Geokon VCE 4200
vibrating wire strain gauges are used in the rigid pavement sections. Their purpose is to
measure warp in the slabs, which can be correlated to temperature, moisture, and shrinkage.
16
Center for Transportation Research. University o f Texas. Austin. Conducted as a
preliminary study, Pilson (1995) instrumented a flexible pavement roadway to investigate
the best instruments and locations for follow on studies with the Texas Mobile Load
Simulator. The roadway used was located in Victoria, Texas and consisted o f thin and thick
test sections. The thin section had a 5.08 cm AC surface over a 30 cm base course. The
thick section was identical to the thin section, but used 19.8 cm o f AC. Loading was
provided by a dump truck. Statistical analysis was performed on the data to determine the
precision o f each gauge.
Five types o f H-gauges and a durable foil strain gauge capable o f being placed
directly into the AC without the aid of a core or carrier block were tested. All measurements
taken were dynamic. The study concluded that the foil strain gauge, a Hottinger Baldwin
Measurements model DA3, gave the most precise and accurate measurements. The
recommended sampling rate was at least 100 Hz.
North Carolina Instrumented Highway. This study consists o f a 12 km instrumented
section o f State Route #421 near Silver City, North Carolina. Begun in 1988, the main
objective o f the research project is to develop mechanistic relationships between stress and
strain levels imposed by traffic for a variety o f typical flexible pavements (Stubstad, 1989).
The roadway structure is AC over cement treated base course (CTBC) or aggregate base
course (ABC). Geosynthetics are not included in the roadway. Strain instrumentation for
the study consists o f 72 Dynatest H-gauges placed at the bottom o f the AC layer. Although
stress measurements are taken in the base course, no strain measurements are taken there.
17
Alberta Instrumented Highway. The Alberta Transportation Department operated
an instrumented roadway northwest o f Edmonton from 1973 to 1976. The purpose of the test
facility was to provide detailed measurements, o f pavement structural responses for
evaluating the effects of axle loads and loading configurations o f pavements (Christison,
1978). The two lane roadway was built on a subgrade o f highly plastic (CH) clay. Asphalt
concrete surfaces o f 28 cm and 18 cm were used. Instrumentation consisted o f foil strain
gauges embedded in carrier blocks, surface deflection gauges, and diaphragm pressure cells.
The study centered on the effects of vehicle velocity, surface temperature, and service
age on surface strains and deflections.
Loadings from a variety o f axle weights and
configurations were compared to a standard 80 kN single axle dual wheel configuration.
Results show that surface strains and deflections from the standard 80 kN axle increased with
service life, surface temperature, and decreasing vehicle velocity. Load equivalency factors
calculated are in close agreement from those based on the AASHO Road Test.
TRRL Instrumentated Highway. Constructed in 1968, this was the first study in
which vertical, longitudinal, and transverse measurements of stress and strain were taken in
a full scale roadway (Potter, 1969). The study was conducted by the Transportation Road
Research Laboratory (TRRL) on the A l trunk road at Conington, Huntingdonshire (UK).
The primary use o f strain gauges was in the bottom o f the AC surface layer. Eighty-nine
early generation H-gauges were used of which 36 were damaged during construction.
Twenty-two foil strain gauge imbedded carrier blocks were also used, o f which only 5
survived construction. Additionally, 5 modified variable reluctance transducers, (similar to
18
an LVDT) measured strain in the base course. Dynamic measurements were made with
loadings from a single unit truck every six months while normal commercial traffic provided
the loading for long term measurements.
Laboratory Tests
VPL 1995. Conducted at the Virginia Polytechnic Institute, this study investigated
the potential benefits of geogrid and geotextile reinforcement in flexible pavements, (Smith,
1995). Geosynthetic reinforced roadway structures were constructed in a reinforced concrete
box of dimensions 3 m by 1.82 m by 2.13 m wide. A pneumatically driven circular loading
plate applied a 40 IcN force at a frequency o f .5 Hz. Fifteen flexible pavement test sections
were studied.
Each highway structure was composed of four materials. The subgrade consisted of
1.22 m o f Yatesville Silty Sand (YSS) with an AASHTO classification of A-4.
A
geosynthetic was placed at the interface between the subgrade and the base course.
Geosynthetics consisted o f the Amoco 2002 and 2016 geotextiles, and Tensar SS2 geogrid.
The aggregate used had a USCS classification o f GW and a maximum dry density o f 23
kN/m3 at an optimum water content of 5.8 % measured by a Modified Proctor test. The hotmix asphalt (HMA) wearing surface had a resilient modulus o f 1.72 x 106kPa.
Measurements o f the roadway consisted o f surface deformations o f the pavement.
There was no internal strain instrumentation. The surface deformation was measured with
an array o f 8 LVDT’S attached to the support structure o f the loading plate. Two o f the
LVDT’S were attached to the loading plate itself and the other 6 were placed at 15.2 cm
19
intervals. Loading took place until failure occurred, which was defined as 2.54 cm of
permanent deformation in the section surface. Depending on the roadway structure, this
could take over 6000 cycles.
The study found that two to three times the number o f ESAL repetitions could be
applied to geotextile reinforced pavement sections compared to unreinforced sections. Use
o f geogrid reinforcement did not alter the number o f ESALs the roadway was capable o f
supporting. Researchers concluded that the separation function provided by geosynthetics
is a key component for improving the performance o f the pavement structure.
Danish Road Testing Machine. Designed and built in the 1970s, the Road Testing
Machine (RTM) allows fatigue testing o f pavement structures under controlled climatic
conditions (Ullidtz, 1979). The facility is enclosed in a building 27.4 m in length, 4 m wide,
and 3.8 m in height. The RTM itself consists o f a pit 9 m long, 2.5 m wide, and 2 m deep.
Hydraulic wheel loading consists of single or dual wheels. Maximum wheel load is 650 kN
and maximum velocity is 25 kph. A schematic is provided in figure 5. The RTM is capable
of applying 10,000 wheel loads in a 24 hour period. Because the RTM is enclosed, specific
climatic conditions can be controlled.
Temperatures range from -10° C to +40 C.
Groundwater level in the pit is also controllable.
As o f 1989, the RTM had been used in 5 major test programs. During this time theDanish RTM has witnessed the development o f H-gauges. The gauges in the initial test
program failed quickly. Upon excavation of the gauges extensive corrosion was determined
as the cause o f failure. Foil strain gauges mounted on cores were then tried. Initially these
20
gauges worked satisfactorily but eventually failed. Due to improvements in weatherproofing,
survivability o f H-gauges was not reported as an issue by Krarup (1992). Strains in the
granular material are made using LVDT’S. Ullidtz (1979) performed finite element analysis
to determine the effect o f the LVDT’S on strain in the surrounding soil. It was found that
results depended heavily on modeling assumptions of friction between the LVDT and the
soil. With full friction, the LVDT would underestimate strain by 30%, while no friction
would cause overestimation o f 40%. Bison coils were also used, but were found not to be
suitable for the dynamic measurements due to signal interference from the wheel loading
apparatus.
South
30 feet
Test track - 9.0 m
North
m
W
Plan
L 9 ##2- A
■I
-f
2.0m
I
Cross section
West L__ 2T-_J East
Figure 5. The Danish Road Testing Machine (from Krarup, 1992)
21
U-S. Army Engineer Waterways Experiment Station. Webster (1992) performed a
study on the performance o f geogrid reinforced base courses in flexible pavements for light
aircraft. Sixteen different runway structures were built. Subgrade for the project was
classified as a CH according to the Unified Soil Classification System. Base course material
met FAA Item P-208 for Aggregate Base Course. The surface course o f asphaltic concrete
was not a test variable and was in all cases 10.2 cm thick. Six types o f geogrid were used
at varying depths in the roadway. Loading was provided by a 133 kN single-wheel-assembly
cart. The cart was built on the frame o f a cargo track and used the wheel from a C -130
aircraft to apply the load.
Instrumentation consisted o f four sets ofM D D ’s. Modules were placed at 5 cm, 30.5
cm, and 61 cm below the surface, and the entire unit was anchored at a depth of 2.44 m.
Although failure was defined as 2.54 cm o f surface deformation, loading was usually
continued up to 7.63 cm o f deformation. Results showed that the amount o f base course
could be reduced through the use o f geogrid reinforcement. Geogrid performance was
determined to be a function o f depth and for subgrade strengths greater than 1.5 CBR, the
optimal reinforcement performance took place with the geogrid at the base course- subgrade
interface.
University o f Waterloo. Work performed at the University o f Waterloo in 1985
studied the possibility o f developing structural equivalency factors for geogrid reinforced
base sections (Penner et al., 1985). The study was conducted by constructing numerous
roadway structures in a laboratory and subjecting them to cyclic plate loading. The roadway
22
cross sections were built in a 4.5 by 1.8 by 0.9 m plywood box reinforced by a steel frame.
An MTS function generator and servo-hydralic controller were used to drive a 300 mm
diameter loading plate onto the roadway surface. The loading force was 40 kN applied at a
rate o f 8 Hz followed by a single static load at predetermined cycle counts. Instrumentation
consisted o f dial gauges to .measure the deformation o f the AC surface and foil strain gauges
bonded to the geogrid.
Various depths and types of materials went into the roadway resulting in 24 unique
cross sections. The subgrade consisted o f a very fine grained beach sand either pure or
mixed with peat. The base was o f quality well graded aggregate. The AC surface was a
dense graded material with 15 mm maximum particle size mixed with 85/100 penetration
grade asphalt cement. The geogrid used was biaxial polypropylene Tensar Corporation SSI.
The conclusion seemed to indicate that the geogrid reinforcement was significant in
cases where inadequate base course thicknesses existed. The AASHTO design method was
modified by the addition of granular layer coefficients. The coefficients measured the ratio
o f reinforced to unreinforcd base layers and ranged from 1.4 to 1.8. Rutting in the subgrade
decreased with the use o f reinforcement.
Tension Testing o f Geosynthetics
Introduction
The goal o f tension tests on geosynthetics is too measure material indicies that are
accepted throughout the industry. At this point, the difference between material indicies and
material properties should be made clear. Material indicies are dependent on the testing
23
procedure used. An example for geosynthetics is the index o f puncture resistance. A material
property is intrinsic to the material, such as the modulus o f elasticity. Because rate o f strain
and coupon size have been shown to change results in geosynthetic tension testing, results
from these tests are classified as indicies.
To perform useful tension tests on geosynthetics, a test procedure must meet two
criteria. The test must accurately resemble the state o f the geosynthetic in use and be simple
enough to enable comparative testing. If the test is complicated and difficult to reproduce
it will not gain general acceptance. This section o f the thesis reviews a widely accepted
tensile test (ASTM Standard D 4595-86) and then considers the affect o f test parameters on
measured tensile properties o f a geosynthetic.
ASTM Standards
Standard Test Method for Tensile Properties o f Geotextiles by the Wide-Width Strip
Method (D 4595-86) covers unconfined tensile properties o f geotextiles.
The test is
characterized by a constant rate o f extension (CRE) and 200 mm wide specimen. The
specified strain rate is 10 %/min with a 3%/min tolerance. If the testing machine is not
capable o f running at a constant rate the standard will allow for an approximate strain rate
of 2 %/min to be applied manually. In addition to being 200 mm wide, the specimen must
be at least 100 mm long. The test is run until rupture o f the specimen occurs.
Strain is obtained from jaw to jaw measurements. Strain gauges may be placed in the
specimen center and through comparison with the jaw to jaw measurements slippage in the
clamping system identified. The two clamp designs permitted are shown in figures 6 and
24
7. Both types o f clamps operate on the principle o f constricting the geosynthetic between a
wedge and a fixed surface. As tension is applied by the testing machine, the geotextile pulls
the wedge into a continually constricting space, thereby providing a tighter grip.
Laboratory Testing
Three testing parameters have been found to significantly affect the measured tensile
properties o f geosynthetics. They are the grip system, specimen dimensions, and rate of
strain. Gripping systems have been broken down into 3 main techniques. The mechanical
wedge (used by ASTM) uses increasing friction o f a wedge being pulled into a confined slot
to hold the geosynthetic. This is the best technique and allows the least amount of slip. It
is also typically more expensive to manufacture and difficult to mount the specimen in. As
a consequence, materials testers have looked to other gripping techniques (as this study did).
An alternative gripping technique is the roller grip or capstan. W ith this technique the
geosynthetic is wrapped several times around a cylinder, As tension is applied, the outside
layers o f geosynthetic press against the interior rolls holding the geosynthetic in place.
Although simpler, significant slippage occurs with this technique as higher tensile loads are
applied. The third technique is to encapsulate the specimen to the grips using epoxy. This
technique is good for low strength geosyntheics, particularly geotextiles which have large
surface areas as compared to geogrids.
The varieties o f geosynthetics and gripping techniques makes selecting a universal
solution difficult. Myles (1986) found that there is no single gripping method suitable for
all geotextiles. High strength geotextiles are difficult to grip in mechanical compressive jaws
1 / 2 ' DlA. HOLE
CONNECTING PIECE
MACHINE BOLTS
1 /2 ' x 2 1 /2 '
REACTION BAR
LOCATV G PIWS
CASE HAPCCNCD STEEL
SERRATED JA V S
SIDE VlEV
FRONT VIEV
Figure 6. ASTM approved Sanders Clamp (from ASTM, 1993)
26
2 00 "
9 00"
2 00"
900"
0 S O 'O IA
13 T h f t o d s / " '
200'
3 30
2 00
■TO CONNECT TO
TENSILE TESTING
_______
MACHI NE
^NOT
EOR
TO
SCALE
CL A RI T Y
ONLY
Figure 7. ASTM approved Wide Width Clamp (from ASTM, 1993)
27
and low strength, highly extendible non-woven geotextiles are often difficult to use in
capstan type jaws.
Specimen dimensions have been shown to affect the measured strength o f a material.
The aspect ratio o f the specimen is the ratio o f width to length. Rowe studied a variety of
geosyntheics and specimen dimensions. His results show that the modulus is dependant on
aspect ratio and recommends specimens with gauge lengths o f at least 100 mm and aspect
ratios o f 5 or greater be used. De Groot (1990) and Leshchinsky (1990) found similar results
but do not recommend specific specimen dimensions.
It has been suggested that the rate o f strain in geosynthetic tension testing is too high
and new rates need be applied. Myles (1987) feels the rates o f strain in existing “textile”
tension tests are too high for geotextiles and a lower rate would be more realistic. Rowe
(1986) showed the ASTM rate of 10 0ZoZmin to be too high and suggests a rate of 2 %/min.
The lower loading rates are recommended to more accurately replicate loading in actual
geotechnical structures.
Related Instrumented Geosynthetic Studies
A significant number o f studies involving instrumented geosynthetic reinforced
retaining walls and embankments have been made. Electrical resistance strain gauges have
been used almost exclusively to measure strains developed in the geosynthetics of these
projects. In order to make accurate and reliable measurements, the inherently difficult task
o f bonding a foil gauge to the geosynthetic must be accomplished. Review o f existing
literature indicates that there have been many different techniques used to bond foil strain
28
gauges to geosynthetics. This section is a review o f those techniques.
Bathurst (1990) has performed studies on large scale geosynthetic reinforced soil
walls. After much experimentation, he developed a mounting technique for applying Showa
Measuring Instruments gauge type Y l l-FA-5-120 to polypropylene geogrids. The geogrid
surface preparation includes abrading with light sand paper, cleaning, and application o f a
neutralizing agent.
The bonding agent used is a RTC two-part epoxy resin cement.
Environmental protection is provided by a bubble o f silicon and outer wrapping o f flexible
plastic tubing. This procedure allowed strain measurements o f up to 3% and survival rates
of 100% to be common during his testing routine.
Prior to his study o f a geogrid reinforced test wall, Simac (1990) also conducted an
extensive laboratory calibration program to identify high elongation foil resistance strain.
gauges that were compatible with the stress/strain characteristics o f the geogrid in his study.
A Kyowa Dengyo gauge type KFE-5-C1 bonded with high elongation cyancoacrylate
adhesive type CN from Tokyo Sokki Kenkyujo Company was ultimately selected.
Weatherproofing was provided by application o f an elastic asphaltic emulsion: The study
reported strain levels up to 2 % but does not discuss failure rates.
Rowe (1994) measured geotextile strain in a full scale embankment in order to
monitor the structure’s performance through construction and deliberate overloading to
induce failure.
The study used Micro-measurements gauge type EP-08-40CBY-120
protected by Dow Coming 3145 RTV adhesive/sealant to monitor strains in a geotextile. O f
the 38 gauges planned for this study 5 did not survive installation and 4 others failed during
the course o f the study. This mounting technique was adequate to strain levels of about 5%.
29
Leshchinsky (1990) conducted studies to determine techniques for using foil strain
gauges in laboratory tension tests o f geotextiles. He used elastic silicon from Teroson GmbH
called Terostat 33 Silicone Sealant to mount Micro-Measurements EP-08-40CBY -120
gauges to geotextiles. Although strain rates o f up to 10% were possible with this method it
is not clear whether the procedure could be weatherproofed for use in a working
environment.
Oglesby (1992) studied techniques for,using foil strain gauges to measure high strains
(over .5 %) during unconfined tension tests o f geogrids. This was a preliminary study to
develop methods for confined tests and eventually full scale field applications. Variables
included surface preparation techniques, adhesives, clamping techniques, and strain gauges.
For high density polyethelyne, the recommended procedure is to prepare the surface with
steel wool and sand paper. After cleansing with alcohol, Micro-Measurements EP-08250BG-120 gauges are bonded with A-12 adhesive. Uniform pressure on the gauges is
provided by clamping the area between neoprene pads. The geogrid is then oven cured at
125° Fahrenheit for at least 4 hours.
30
CHAPTER 3
RESEARCH METHODOLOGY
Introduction
The purpose o f this chapter is to present the experimental methods and equipment
used in the study. The chapter is divided into sections on the instrumented roadway and the
wide width testing facility. The instrumented roadway section contains descriptions of strain
sensors, data logging systems, roadway monitoring, roadway construction, and emplacement
o f sensors.
The second section describes the wide width tension testing facility and
experiments performed with the apparatus.
Instrumented Roadway
Instrument Types
Four different types o f sensors were evaluated in the study. They were selected such
that dynamic and static measurements could be taken. Three sensor types were placed in the
base course and on the geosynthetics, while only one type was used in the asphalt concrete.
All the sensors were placed in a horizontal plane oriented to measure strain perpendicular to
the direction o f traffic (the transverse direction in the case o f the geosynthetics).
Asphalt Concrete. Geokon (Lebanon, NH) model VCE-4200-HT (high temperature)
vibrating wire embedment strain gauges were used in the asphalt concrete. The gauge is
shown in figure 8. Vibrating wire technology was selected because it is excellent for long
31
term measurements. They measure strain by determining the frequency o f a vibrating wire
that runs in the shaft o f the gauge. A mechanical device plucks the wire and the resonant
frequency is measured. Knowledge of the frequency allows calculation o f the length o f the
wire and hence the length o f the gauge. Knowing this information allows the strain to be
easily calculated. Vibrating wire strain gauges can be used with long cable lengths because
their output signal is a frequency which is not susceptible to degradation caused by changes
in cable resistance. This is an advantage compared to foil strain gauges, which use voltage
for signals and are susceptible to changes in cable resistance. The major disadvantage of
vibrating wire technology is their slow response. Several seconds are required to pluck the
wire and obtain the resonant frequency. This makes dynamic measurements impractical with
the gauges.
The active gauge length of the gauge is 150 mm. Range o f the gauge is 0.3 % strain
(3000 micro strain) with a sensitivity of 0.0001 % strain (I micro strain). High temperature
resistant Tefton cable is used with the gauges to withstand placement in the asphalt concrete.
A thermister is also included in the gauge. Four o f the gauges were used in the roadway.
Figure 8. Vibrating Wire Embedment AC Strain Gauge
32
Instrumentation in the Base Course. Geokon model VCE-4200 vibrating wire
embedment type strain gauges (shown in figure 9) were one o f three gauge types used in the
base course.
The instrument is identical to the HT model used in the AC with two
exceptions. High temperature Teflon cable is not used and the end discs are larger (76 mm
in diameter) to produce better interaction between the base material and the sensor. A
thermistor is also included in the gauge. Two o f the gauges were used in the roadway.
Figure 9. Vibrating Wire Embedment Strain Gauge
For measuring larger displacements in the base course, vibrating wire embedment
displacement gauges were used. Originally, two Geokon model 4430 vibrating wire borehole
deformation meters were to be used. During installation however, one o f the sensors was
found to be defective and in its place a model 4420 vibrating wire crack meter was used. The
two sensors have the same physical dimensions and operate in the same manner.
33
The vibrating wire displacement gauges operate on the same principle as the vibrating
wire strain gauges.
The major difference between the two sensors is the range of
displacement measured. The embedment displacement gauges have a range o f 8 % while the
strain gauges had a range o f only 0.3 %. A vibrating wire displacement gauge is shown in
figure 10. The end discs are 101 mm in diameter and the length o f the instrument is 317 mm.
The sensitivity o f the gauge is 0.008 %. Thermisters are included in each gauge. One each
o f the model 4430 and 4420 were used.
Figure 10. Vibrating Wire Embedment Displacement Gauge
To measure both the static and dynamic performance of the base course, embedment
LVDTs were utilized.
Shown in figure 11, the models chosen were RDP Electronics
(Pottstown, PA), submersible, miniature AC, model D5/400W LVDTs with a range o f +/-
34
10.2 mm and a sensitivity o f +/- 0.03 mm. For the gauge length o f the instruments, this
corresponds to strain measurements o f +/- 13 % and a sensitivity o f +/- 0.04 %. The end
discs o f the LV D T s have diameters o f 39 mm and the length o f the sensor is 8 1 mm. Four
o f these instruments were used in the roadway.
M
Figure 11. LVDT Embedment Strain Gauge
Instrumentation on the Geosvnthetics. Vibrating wire, LVDTs, and foil strain gauges
were used on the geosynthetics. Figure 12 shows a vibrating wire strain gauge used for long
term measurements. The sensor was attached by sandwiching the geosynthetic between two
rectangular mounting plates measuring 89 by 57 mm. The plates were held together with 6
bolts. The strain gauge itself was mounted to the steel plate by means o f a circular collar
and held in place with a set screw.
When mounted, the axis o f the strain gauge was
35
approximately 14 mm above the surface of the geosynthetic. Two o f these gauges were used
Figure 12. Vibrating Wire Strain Gauge
To measure large static strains in the geosynthetics, vibrating wire displacement
gauges were used. The model selected was a Geokon model 4420 vibrating wire crackmeter
shown in figure 13. The gauge has a range of 25.4 mm and a sensitivity o f 0.025 mm. The
nominal gauge length of the instrument is 280 mm, which allows a strain range of 9 % with
a sensitivity of 0.009 %.
The method used to mount the vibrating wire displacement gauge to the geosynthetic
was similar to that used for the vibrating wire strain gauge. Each end o f the gauge was
connected to a rectangular mounting plate measuring 89 by 57 mm. The geosynthetic was
36
sandwiched between the plate and an identical mounting plate. Six bolts were used to hold
the plates together. A ball pivot point connection was utilized between the gauge and the
plates which allowed any orientation of the mounting plate. The axis o f the gauge was
approximately 25.4 mm above the surface of the geosynthetic. Two o f the gauges were used
in the roadway.
Figure 13. Vibrating Wire Displacement Gauge
The LVDT used on the geosynthetic was an RDP Electronics submersible, miniature
AC modelD5/200W LVDT with a range of +/-5.1 mm and a sensitivity o f +/- 0.015 mm.
Shown in figure 14, the gauge measured both dynamic and static strains. Given the 50 mm
nominal gauge length o f the instrument, this device has a strain range o f +/- 10 % and a
sensitivity o f +/- 0.03 %.
37
The mounting brackets for the LVDT were small compared to those used by the VW
technology. They measured 30 by 12 mm and were fixed in orientation. The axis of the
gauge was approximately 14 mm above the surface o f the geosynthetic. Four of the gauges
were used in the roadway.
Figure 14. LVDT Displacement Gauge
Kyowa (Soltec Corp., San Fernando, CA) high elongation foil strain gauges (model
KFE-5-120-C1) were used directly on the geogrid. Due to their small size (approximately
3.4 by 11 mm) they could be mounted on an individual rib of the geogrid. Foil strain gauges
were not used on the geotextile. The foil strain gauges have excellent response for dynamic
measurements, but are unsuitable for long term measurements due to drift in their signal as
discussed previously.
38
The following procedure was used to bond the foil strain gauges to the geogrid. The
surface was prepared by sanding with 600 grit sand paper. The rib was then etched with
MicroMeasurements Tetra-Etch (TEC-1). Next, the surface was rinsed with water and
allowed to dry.
A single gauge was attached to the upper side o f the geogrid using
MicroMeasurements M-Bond AE-10. An additional gauge was attached to a small, detached
sample o f the geogrid. The two gauges were attached by a full wheatstone bridge circuit and
connected to 2-pair, twisted pairs, overall shielded, 22 gauge cable. The gauge attached to
the unstressed sample o f geogrid provided for temperature compensation. This bonding
procedure appeared to be adequate for the small strains that were ultimately observed in the
field.
Additional laboratory tension testing, however, indicated that gauge debonding
occurred at 1-3 % strain.
Environmental protection o f the gauges was provided by sandwiching the geogrid
area between sheets o f MicroMeasurements M-Coat F Butyl rubber sealant and neoprene.
Additional protection was provided by sandwiching this area between sheets o f clear plastic
with liberal amounts of silicone use between the sheets, particularly around the cable entry.
Finally,, the entire area was sandwiched between sheets o f geotextile, again with liberal
amounts of silicon applied. Four o f the foil gauges were used in the roadway.
Data logging Systems
Two data logging systems were used in the study. One was designed to handle the
long term recordings and the other dynamic recordings.
39
Long Term Data Logger. The long term data logger was a Campbell Scientific
(Logan, UT) data logger. Model CR-10. Supporting equipment included 2 AM-416 16
channel multiplexers (all 16 channels o f a multiplexer were routed into one channel o f the
logger), one A V W l Vibrating Wire Interface (provides excitation for all VW gauges),
PS12LA Power Supply (12VDC and battery), SC32A RS-232 Interface (allows
communication directly through a PC serial port), and a DCl 12 Modem (1200 baud modem
for remote access to the data). The CR-10 system is designed for stand-alone field data
acquisition and storage at relatively slow rates. It can interface to all types o f sensors
including VW gauges, sensors with DC outputs, and all types o f temperature sensors.
The Campbell data logger was programmed to record data every hour, 24 hours a day.
The data recorded for each hour was data averaged over a 10 minute period prior to the time
o f recording. The logger is capable of cycling through the 24 channels every 30 seconds,
meaning that for the 10 minute data collection period, 20 samples per instrument were taken
and averaged. The data was stored in the data logger and could be downloaded from a
remote computer using the telephone modem or collected on site as needed.
Dynamic Data Logger. An IOTech (Cleveland, OH) Daqbook 100 was used as the
dynamic data logger. A disadvantage o f the unit was that it could not monitor the vibrating
wire instruments. It could however, be readily adjusted to monitor only specific instruments
or desired sampling rates. The logger included a DBK-40 BNC Analog Interface (16 channel
box that allows BNC connections to be made). The Daqbook is basically an AZD converter
that can communicate with a PC. It allows data to be collected and stored at various speeds.
40
Supporting electronic equipment was used by LV D T s and foil strain gauges. Signal
conditioning for L V D T s was provided through a RDP Electrosense S7-AC signal
conditioner, which provides an AC excitation and a +/- IOV DC output. This output could
be read by both the Campbell logger and Daqbook. The foil strain gauges used a RDP
Electrosense S7-DC signal conditioner with an excitation voltage o f 5 V DC and the gain set
to 260.
' Roadway Monitoring
Long Term Monitoring Program. The data logging began with the Campbell data
logger on July 22, 1995 at 2:00 P.M. At this point in time the roadway was still under
construction. The base course had been placed and spread but not compacted. The only
transducers recording meaningful data were those attached to the geosynthetics. The first
compaction o f the base course took place on July 24th at approximately 7:00 A.M. The
Campbell logger was turned off at this point such that the transducers could be monitored
real time with the dynamic data logger to ensure that damage was not occurring during
compaction. The next day, July 25th, the aggregate base course was compacted around the
embedment transducers.
Following successful compaction of the base course, a series o f tests were conducted
on the unsurfaced roadway.
On July 25th, four truck pass tests were conducted. A t
completion o f the tests the Campbell data logger was turned back on. On July 31st, road
rater tests and two more truck pass tests were performed on the unsurfaced roadway. At this
point, the only transducers not providing meaningful data were the embedment strain gauges
41
for the asphalt concrete.
On August 1st the asphalt concrete running surface was placed. At approximately
7:00 A.M. the base course layer was leveled and compacted for a second time. The asphalt
concrete was placed, shortly thereafter with the embedment gauges placed in the AC. The
asphalt was allowed to cure for several hours with the section being opened to truck traffic
at 1:00 P.M. It is noted that the unsurfaced section was closed to traffic up until the time that
the asphalt was placed.
The Campbell data logger was inoperable between August 3, 12:00 P.M. to
September 7,12:00 P.M. Any stored data was lost during this period. During the period, the
logger was returned to the manufacturer for service. The malfunction was due to a bad diode
which was replaced. During this time, dynamic data could still be collected. A truck pass
test was performed on August 29. Additional Road rater tests were performed on the
surfaced roadway on September 21st. The final two truck pass tests were performed on the
AC surfaced roadway on October 19.
The logger was permanently disconnected on October 26. The roadway itself, was
dismantled in early November and all instrumentes, with the exception o f the AC vibrating
wire embedment strain gauges were retrieved. A history o f the significant events concerning
the roadway is shown in Table I . The truck pass tests and road rater tests are discussed in
the next section.
Dynamic Monitoring Program. The dynamic testing program recorded the response
o f the instruments to individual loadings on the roadway. Truck pass tests were conducted
by a variety o f different sized trucks operating in the yard. Road Rater tests were conducted
42
Table I. Chronological Order o f Events For Roadway Construction and Testing •
Date
Time
Event
July 22
2:00 P.M.
Campbell data logger turned on,
Reference Times (hours)
—
uncompacted base in place.
j July 24
7:00 A.M.
Base course compacted.
I July 25
7 A.M.
Base course transducers set.
- 2 P.M.
truck pass tests 1-4.
July 25
3:00 P.M.
Logger turned back on.
31
July 3 1
10:00 A.M.
Road rater and truck pass
170
0
22-29
tests 5 & 6 on unsurfaced section.
Aug. I
7:00 A.M.
Base leveled & recompacted,
191
AC placed and compacted.
Aug. I
1:00 P.M.
Section open to traffic.
197
Aug. 3-
12:00 P.M.
Campbell data logger
242-1083
Sept. 6
12:00 P.M.
inoperable
Aug, 29
8:00 A.M.
Truck pass test 7
864
Sept. 21
9:00 A.M.
Road rater tests
1441
Oct. 19
8:00 A.M.
Truck pass tests 8 and 9
2112
Oct. 26
5:00 P.M.
Logger disconnected
2289
by the Non-Destructive Testing Unit o f the Montana Department o f Transportation (MDT).
The Daqbook, described earlier in this report, was used to collect data from the non-vibrating
wire instruments during these tests. During the second Road Rater test performed on
September 21st, monitoring programs were downloaded to the Campbell logger to enable it
43
to cycle through one channel at a time, such that the Road Rater could be placed in the
vicinity o f a vibrating wire instrument with data from this instrument being taken at a rate
o f approximately I sample every second.
For the truck pass tests, the non-vibrating wire instruments were monitored as trucks
were allowed to slowly pass over the test section. The trucks typically traveled at a speed
o f 8 kph. Tests were performed prior to and after the placement o f the AC layer. Four types
o f trucks were used for the nine truck pass tests listed in Table I . Truck pass tests I and 2
used a 2-axle pickup truck weighing approximately 15 kN. An empty belly dump truck
weighing 156 kN was used for truck pass 3. A single unit dump truck with a pup trailer was
used for truck pass 4. Both the truck and pup were full with gravel. A weight of each unit
was not obtained. A full single unit dump truck with a pup trailer Was used for truck pass
tests 5 and 6. The entire vehicle weighed 423 and 406 kN for tests 5 and 6, respectively.
Tests 1-6 were performed on the unsurfaced test section. The record o f the truck used for
test 7 was not available. A full belly dump truck weighing 414 and 429 kN was used for
truck pass tests 8 and 9, respectively. Tests 7-9 were performed on the surfaced test section.
A summary o f the truck data for the different tests is given in Table 2.
Road Rater tests were performed for the purpose of evaluating the ability o f the
instruments to respond to the dynamic surface load. The tests were also conducted for the
purpose o f determining whether this was a suitable means o f providing dynamic response
data for use in follow on research. Although not a primary objective, the Road Rater tests
were also used to back calculate elastic moduli for the various pavement layers.
44
Table 2. Vehicles Used For Truck Pass Tests
I
T est#
Date
Truck Type
Weight, kN
Surfacing
I
July 25
Pickup
15
Unsurfaced
July 25
Pickup
15
Unsurfaced
3
July 25
Belly Dump (empty)
156
Unsurfaced
4
July 25
Single Unit With Pup
not known
Unsurfaced
5
July 3 1
Single Unit With Pup
423
Unsurfaced
6
July 3 1
Single Unit With Pup
406
Unsurfaced
7
Aug. 29
Unknown
not known
Surfaced
j
8
Oct. 19
Belly Dump
414
Surfaced
j
9
Oct. 19
Belly Dump
429
Surfaced
jj
|
The first Road Rater tests on July 31st were performed on the unsurfaced test section
after the base course was initially placed, compacted, and after base material was compacted
around the embedment transducers. A model 2000 Road Rater (Foundation Mechanics, Inc.,
El Segundo, CA) machine was used for these tests and consisted o f a circular plate 30.5 cm
in diameter through which a dynamic load was applied over a period o f time at a given
frequency. Dynamic displacement transducers monitored the vertical deformation o f the
ground surface at points extending radially from the applied load. Dynamic displacement
was measured at points 0,20.3, 30.5, 61, 91, and 122 cm from the center o f the applied load.
The load frequency was 25 Hz with the instrument sampling frequency set at 20 Hz. For the
tests performed on July 31st, five multi-level force values were applied within a give load
cycle, with this load cycle repeated 5 times for the majority o f the test locations. The
dynamic force levels were 6.67, 8.9, 11.1, and 13.3 kN. Each load in the load cycle was
45
applied for approximately 3-5 seconds, with the next load in the cycle applied immediately
thereafter. For each test, the Road Rater loading plates were set on the ground surface
directly above the instruments to be monitored.
The second series o f Road Rater tests performed on September 21st were similar to
the first series. The loading sequence consisted o f four cycles o f incrementing loads, where
the loads were set at 11.1, 13.3, 15.6, and 17.8 kN. The frequency o f loading was reduced
from 25 Hz to 20 Hz with the instrument sampling rate increased from 20 Hz to 60 Hz in an
attempt to capture the response for a given load application.
For the September 21st Road Rater tests, the Road Rater was also used to apply load
atop the vibrating wire instruments. This was done to assess the possibility o f using these
instruments for dynamic measurements, provided the sampling rate could be sufficiently
high. Individual programs were downloaded to the Campbell logger allowing for one
vibrating wire instrument to be sampled and monitored.
This approach increased the
sampling time to approximately I Hz. The road rater was configured to apply four cycles o f
the same load, 17.8 kN, such that a near constant dynamic load could be applied td the
instrument. The loading frequency was increased to 50 Hz in an attempt to make this load
appear even more constant. Six of the twelve vibrating wire instruments were successfully
tested. The 17.8 kN load was applied for approximately 100 seconds to each instrument.
Roadway Construction and Gauge Locations
Introduction. The roadway was constructed in a local contractor’s gravel quarry
located between Bozeman and Belgrade.
The contractor, JTL Inc., allowed for the
46
construction o f the roadway along a track travel path leading from the gravel pit to the weigh
scale. The travel path was covered by a layer o f deteriorating asphalt. JTL Inc. donated the
raw materials (gravel base and asphalt concrete) and the equipment and labor necessary to
construct the roadway. Construction o f the roadway began on 13 July and was completed
on !August, 1995.
The purpose o f the roadway was to evaluate instruments that were thought to be
suitable for use in the follow on research and whose likelihood for success was uncertain.
The roadway was not chosen with the aim o f duplicating the types o f subgrade and layer
thickness and preparation conditions anticipated in follow on research. While this would
have been a desirable feature to incorporate in the roadway, cost and time limitations did not
allow for such action.
Roadway Construction Overview. The roadway was constructed by excavating the
existing asphalt and subgrade along a straight section measuring 42 m in length by 4.5 m in
width. The existing asphalt was less than 5 cm thick. The existing subgrade consisted o f a
dry to moist, silty, sandy gravel with cobbles. The subgrade appeared to be at a relative
density of 75-85 %. The excavation extended'to approximately 26 cm below the top o f the
existing asphalt grade. Approximately 10 cm o f the subgrade was scarified and recompacted
with a vibratory steel drum roller. A thin layer o f base course material was placed and spread
to level the bottom o f the excavation. This layer was not compacted and ranged in thickness
from 1.27 cm to 5 cm. When completed, tracks entered the pit along the existing travel way
to the west of the test section and returned by entering onto the test section when coming out
47
o f the pit and traveling towards the weigh scale. In this way, the test section saw only loaded
trucks whose weights were recorded at the scales.
A 16.7 m section o f geogrid (Tensar BX 1100) was placed within the extreme
Northeast end o f the test section. A 13.7 m section o f geotextile (Amoco 2006) was placed
immediately to the Southwest o f the geogrid section. The remaining 12.2 m o f the test
section to the Southwest was left as an unreinforced control section. A plan view o f the test
section is shown in figure 15.
The instrumentation for the geosynthetics was then attached. Instrument installation
is detailed later in the thesis. Approximately 20.3 cm o f base course was then placed and
spread throughout the test section. It was a gravel with sand and silt, with no particle larger
than 7.6 cm. Instruments were installed in the base as it was placed. The base course was
placed with a large front-end loader and spread with a small skid-steer loader. It had to be
hand-spread in locations where instruments were to be embedded. Then the base course was
compacted with a vibratory steel-drum roller. Additional base course was placed and spread
level with a grader and recompacted with the same roller. Approximately 6.35 cm of asphalt
concrete was then placed with a belly-dump truck, spread level with a grader and then
compacted with a steel-drum vibratory roller.
Instrumentation Layout. A total of 24 instruments were placed in the roadway. The
locations are shown in figure 16. The dashed lines running the length o f the roadway show
where the anticipated truck wheel paths were to be located. The embedment strain gages, in
the base and asphalt concrete were placed as close as possible to the bottom o f the respective
layers. A summary o f the gauges used in the roadway is given in Table 3.
48
I yard exit
point
North
A
weigh scale
A
Arrows indicate direction
of truck traffic
geogrid section
geotextile section
control section
From Gravel Pit
Figure 15. Plan view o f Roadway Test Site
49
■
VW Embedment Strain Gage in Base
•
VW Strain Gage on Geosynthetic
D
VW Embedment Displacement Gage in Base
X
VW Displacement Gage on Geosynlhetic
O
LVDT Embedment Gage in Base
®
LVDT on Geosynthetic
B
VW Embedment Strain Gage in AC
A
Foil Strain Gage on Geogrid
NonReinforced
Geotextile
Geogrid
□OB)
64mm
203mm
Subgrade
Geotextile
or
Geogrid
Figure 16. Roadway Instrumentation Layout
50
Table 3. Instrumentation Specifications
Instrument
Instrument Type
Location
Gage Length
Strain
Strain
mm
R an g e(%)
Sensitivity (%)
Geosyn.
VW Strain Gage
150
0.3
0.0001
Geosyn.
VW Displacement
280
9
0.009
Geosyn.
LVDT Displacement
50
+ / - 10
+/- 0.03
Geosyn.
Foil Strain Gage
5
10
———
Base
VW Embedment Strain
161
0.3
0.0001
Base
VW Embedment Displ.
317
8
0.008
Base
LVDT Embed. Displ.
81
+/-13
+/- 0.04
AC
VW Embed. Strain
153
0.3
0.0001
Instrumentation Installation
Instruments Attached to the Geosynthetics.
Instruments were installed by
' sandwiching the geosynthetic between the two mounting plates o f the sensor and tightening
the connecting bolts. Holes were predrilled in the geotextile to allow the bolts to pass
through. A protective layer of medium sand, approximately 0.5 cm in thickness, was placed
below each o f the instruments attached to the geogrid. In addition, a small piece o f the
geotextile was placed above the footprint o f the instrument to prevent the instrument from
binding with the aggregate base course material. The protective layer o f sand beneath the
geogrid was continued along the length o f the cables attached to the instruments and
deepened to approximately I cm to prevent base course aggregate form puncturing the
cables. The cables were also fastened to the geogrid ribs with plastic zip strips. The sand
51
pack was not used beneath the geotextile as it was with the geogrid.
Three strategies were employed for permanently covering and protecting the
transducers located on the geosynthetics prior to the placement o f base course aggregate.
The first option consisted o f covering the instrument with a half section o f PVC pipe. The
ends of the pipe were packed with sand to prevent base course from entering the pipe. The
second option consisted o f covering the transducer with medium sand and then covering the
sand mound with a layer of geotextile. The third option consisted o f covering the instrument
with a medium sand and placing a half section of PVC pipe over the sand pack such that the
entire pipe cavity was filled with sand. The options used for the instruments attached to the
geosynthetics are listed in table 4. Due to limited number of vibrating wire displacement and
strain gauges, not all cover options could be used for these sensors. Note that the PVC pipe
used for instrument #1 was significantly more flexible than the other PVC pipes. All cables
extending across the top surface o f the geosynthetics were also covered with a sand pack to
prevent cable damage from the aggregate base.
Due to the extensive environmental protection around the foil gauges, additional
covering in the field consisted only o f a thin layer o f sand mounded atop the gauge locations.
With the above precautions taken, the instruments on the geosynthetics were ready for the
base course to be placed.
Instruments Embedded in Base Course. Prior to the placement o f the aggregate base
course, temporary covers were placed around the gauges to be embedded in the base to
protect them from damage during compaction.
Three types o f covers were used to
accommodate the different sizes o f the instruments and to explore different methods. A large
52
Table 4. Geosynthetic Instrument Cover Options
I Instrument Number
Geosynthetic Instrument Type
Cover Option
|
I '
VW Displacement Gage
3
I
2
VW Displacement Gage
2
' 5
VW Strain Gage
I
9
VW Strain Gage
2
29
LVDT Displacement Gage
2
30
LVDT Displacement Gage
3
31
LVDT Displacement Gage
I
32
LVDT Displacement Gage
‘ 3
irrigation control box with the bottom removed was placed around the vibrating wire
embedment displacement gauges. A layer o f sand was placed in the bottom of the container
and along the length o f the cable extending from the container. A similar arrangement was
used for the LVDT embedment gauges with a sand pack also being used. Both o f these
containers were rigid. A flexible tube was used to encase the vibrating wire embedment
strain gauges. With these containers in place, base course material was placed and spread
as previously described. Hand spreading was necessary in the vicinity o f the embedment
gauges.
Aggregate base was mounded around the outside of the two rigid containers such that
the base course extended above the top of the container. This step was necessary to prevent
the container from being pushed down by forthcoming compaction. The flexible container
was allowed to extend above the base course grade such that the base course could be spread
directly adjacent to the container. The flexibility o f the container allowed for the compaction
53
equipment to compress both the container and the base as the compacter rolled over the
location. Observing the compaction operation showed that both techniques work quite well
and that no physical differences in the base course material existed between material adjacent
to the containers and that lying far away.
Once the base course was compacted, the containers were pulled from the ground.
The flexible container had to be slit to be removed while the other containers pulled from the
ground more easily due to their sloped sides. Aggregate base material was then compacted
around the instruments to backfill the cavities left by the containers. This was accomplished
with the use o f a short length o f wooden stake and a hammer. The stake allowed for the base
to be compacted well within areas interior to the discs o f the gauges.
The base was
compacted to the greatest density possible. This technique also allowed for the gauges to be
positioned within the middle o f their range. This was accomplished by monitoring the
instruments in real time while gently tapping the base course either in the interior or exterior
areas o f the end discs, The remaining base necessary to fill the cavity was placed in the same
fashion. As described previously, additional base course was added to level the test section
and additional compaction was provided. This compaction took place without any protection
o f the instruments.
Instruments Embedded in Asphalt Concrete. The embedment strain gauges in the
asphalt concrete were placed using carrier blocks. This was accomplished by placing the f
gauge in an open wooden mold and compacting asphalt around the gauge with a wooden
stake and hammer. For one o f the gauges (# 10, over the geogrid under a wheel path), the
54
asphalt was allowed to cool and set-up over night. For the other three gauges, the asphalt
was compacted approximately an hour before placement o f the asphalt layer. These two
approaches were taken to examine the effect o f curing time on the compatibility o f the carrier
block with the surrounding AC. Using the carrier block compacted a day in advance, it was
feared the AC constrained in the brick would not be sufficiently heated by the surrounding
AC, with the result being that the block would not bond well with the adjacent material. It
was speculated that this problem would be avoided by compacting the brick immediately
before AC placement. The carrier blocks were oriented in the transverse direction o f the
roadway. The asphalt was placed as described previously with care taken to avoid driving
over the gauges during the initial dumping o f the asphalt.
Wide Width Tension Testing
Loading Frame
The loading frame for in-air wide width tensile testing o f geosynthetics consists o f
an upper and lower assembly. The frame is illustrated in figure 17. The loading frame
accommodates a specimen with maximum dimensions of 1.8 m by 0.91 m. This size
specimen is much larger than that used in most wide-width tensile tests, where specimens
measure 200 mm by 100 mm. The large width is necessary to accommodate the different
instruments placed on the specimen, while preserving.a 2 :1 ratio o f specimen width to height.
The 2:1 ratio is generally recommended to ensure that a condition o f zero lateral strain is
achieved over the majority o f the central region o f the specimen.
A modified roller grip system was used for the loading frame. The upper and lower
55
assemblies use 178 mm diameter steel cylinders which act as drums around which the
geosynthetic is wrapped and gripped. The upper drum is rigidly attached to two vertical
supports which connect to an I-beam which in turn rests on top o f the load cell. The steel
cylinder in the lower assembly contains a central axle, which it is rotated about during
loading and unloading o f specimens. The lower cylinder is attached to an I-beam via short
vertical supports. The I-beam itself, is bolted to the bottom load platen o f a Baldwin testing
machine load frame. The Baldwin operates by driving the top load platen, and therefore the
upper assembly, upwards.
The loading o f a specimen is as follows. The geosynthetic is initially placed on the
upper drum giving extreme attention to proper alignment. To prevent excessive slipping o f
the geosynthetic around the cylinder, a metal retaining strip measuring 20 mm by 5 mm in
cross section is placed atop the leading edge o f the geosynthetic and bolted to the cylinder.
The bolt hole spacing is 10 cm. The geosynthetic is then wrapped around the cylinder 2 1A
times such that the geosynthetic leaves the drum on the opposite side o f the retaining strip.
The geosynthetic is then wrapped a half-turn around the lower drum with a retaining strip
fastening the specimen to the drum in the same manner as that used for the upper drum. The
cylinder is then rotated to provide 2 1A wraps o f specimen around the cylinder, with the
cylinder then bolted to the vertical supports to prevent further rotation. The lower assembly
is then moved by the Baldwin machine to remove slack in the specimen."
During loading of the specimen, the geosynthetic will slip a small amount around the
drum. In addition, the material contained in the wrap will strain and deform. This makes it
difficult to define a gauge length necessary to calculate an average strain for the specimen.
56
I-Beam
uuu
Load Cell
Hydraulic
Actuator
Roller Dmm
/ »
0.914m
Geosyntlheilic
Test
Specimen
Roller Dram
178mm
I-Beam
Figure 17. Schematic of Wide Width Tension Testing Frame
57
To overcome this problem, it was necessary to monitor the absolute displacement o f two
extreme points on the specimen, one close to the top o f the specimen and one close to the
bottom. This allowed the relative displacement between these two points to be calculated.
Knowing the length between the two points allowed for the average strain across the
specimen to be determined. This procedure was provided for two lateral positions along the
specimen, such that the average strain could be determined along two vertical lines. Celesco
(Canoga Park, CA) position transducers (PT-IOl) with a 25.4 cm range were used to monitor
the absolute displacement o f each o f the four points. These transducers operate by a cable
which extends from a spool within the transducer body. Additional cable was attached to
this cable to allow for connections to the top o f the specimen. The position transducers were
clamped to the I-beam o f the lower assembly. The mounts for the cable ends consisted o f
a bolt and washer arrangement with a cable attachment point being held as closely as
possible to the face o f the geotextile.
A Computer Boards (Mansfield, MA) data collection board (CIO-DAS08-PGM) was
installed in a 386 PC with Computer Boards Control-CB software used to operate the board.
A terminal board (CIO-MINI37) and IOV DC power supply were used to power and connect
to the load cell and the four Celesco position transducers. The Control-CB software allowed
for the data (as well as mathematically manipulated data) to be monitored and plotted real
time on the screen o f the PC. This feature was used to monitor the strain rate associated with
the applied load during a test.
Several tests were performed on the geotextile and the geogrid with each material
placed in both the machine and transverse directions while manually measuring the applied
58
load with an electronic readout scale. The machine direction corresponds to the direction in
which the geosynthetic emerges from the textile mill, while the transverse direction
corresponds to the orthogonal direction. This step was performed to assess the performance
o f the load frame system. Approximately 36 IcN (24 kN/m o f material) was applied to the
geotextile while strained in it’s machine direction. During this loading application, the
geotextile strained appreciably, but did riot rupture. It was also observed that the geotextile
did not tear in the vicinity o f the steel retaining strip used to fasten the geotextile to the drum.
Five tests on different geogrid specimens were performed. Three tests were performed in the
machine direction and two tests were performed in the transverse direction. The geogrid was
loaded to the point o f rupture. The geogrid was seen to rupture by splitting through the mid­
cord of a longitudinal fiber. For the material oriented in the machine direction, the rupture
load ranged from 14.5 to 11 kN/m, with the material oriented in the transverse direction
ranging from 21.5 to 19.7 kN/m.
Sensors Used for Local Stain Measurement
The sensors used in the roadway were mounted on geosynthetic specimens tested in
the load frame described above. The sensor types used are listed below.1234
1. Vibrating wire strain gauges
2. Vibrating wire displacement gauges
3. LVDT displacement gauges
4. Foil strain gauges
59
A description o f these instruments was given in the previous section Instrumented Roadway,
Instrument Types. Holes were pre-drilled in the geotextile to allow the bolts for the
mounting plates to pass. The typical dimensional positions o f the instruments is shown in
figure 18. The dimensional placement of the instruments on the geogrid was similar to that
o f the geotextile.
A different procedure was used to mount the foil gauges in the lab than in the
roadway. The change was made because the new mounting technique was felt to allow
higher strain levels to be attained in the lab tests. The geogrid rib was sanded with 320-A
sand paper along both its length and width. Care was taken to dull and roughen the originally
shiny surface without raising large strands o f fiber from the rib. W hen the sanding was
judged to be adequate acetone soaked cotton swabs were used to clean the area.
A surface treatment agent (Kyowa S-8) was then applied to the rib. The agent was
applied with a cotton swab and then dried with a blow dryer for 15 seconds. The surface
treatment agent was then applied a second time and allowed to dry for approximately 5
minutes.
Strain gauge cement (Kyowa CC-33A Strain Gauge Cement) was applied to the rib.
The foil strain gauge was grasped with tweezers and placed on the rib. Once the gauge was
properly oriented on the rib, a polyethylene sheet (included with the foil strain gauges) was
placed over the foil strain gauge.
Firm finger pressure was applied to the sheet for
approximately 60 seconds. At this point the gauge was firmly attached to the geogrid.
MicroMeasurements (Raleigh, NC) M-Coat A was applied as a protector over the length of
the foil strain gauge. The coating dries in approximately 1A hour and provides a physical
60
E
E
Lm
S
a
Q
WV-SG-5:
VW Strain Gage #5
LVDT-DG-30:
VW-DG-2:
LVDT Displacement Gage #30
VW Displacement Gage #2
Q
Cable-DG (common): Cable Displacement
Gages (same for
all four shown)
Figure 18. Dimensional Placement of Instruments on a Geosynthetic Specimen
61
protection o f the foil strain gauge.
The majority o f the specimens tested contained a
duplicate gauge mounted on the opposite side o f the rib. To mount the second gauge the
procedure was simply repeated on the other side o f the rib. The two gauges were connected
through a half-bridge arrangement. This arrangement eliminated any bending effects from
the measured results. Specimens with only one gauge used a quarter bridge arrangement.
Tests Performed
In general, experiments were performed with VW strain gauges, vibrating wire
gauges, and LVDTs attached to the geotextile specimens. For the geogrid specimens, the
same set of gauges was used with the addition o f foil strain gauges. Celesco position
transducers were mounted to the specimens to measure global strain.
The following variables were included in the test program:
1. Geosynthetic type: Tensar B X l 100 geogrid (Atlanta, GA) and Amoco 2006 geotextile
(Atlanta, GA).
2. Direction of load with respect to geosynthetic orientation: Machine or transverse
direction.
3. Arrangement o f mounting brackets.
The two geosynthetics tested were mounted in the test head such that the load was
applied in either the machine or in the transverse direction. In general, the strength and
stiffness o f geosynthetics is different in the two directions. In addition, the manner in which
the mounting plates grip the geogrid specimen is different in the two directions due to the
difference in aperture opening size. The majority o f the tests were performed by applying
62
two unloading-reloading cycles during the course o f loading.
Various mounting techniques were examined for the vibrating wire strain and
displacement gauges. The purpose o f these tests was to examine the effect o f the mounting
brackets on the strain measured by the gauge. Securing the mounting brackets in different
orientations was expected to influence the effective gauge length o f the sensor, thereby
influencing the calculated strain measurement. For the vibrating wire strain gauges, tests
were performed with all six bolts on a given mounting plate secured, with four o f the six
bolts on the extreme ends of the plates secured, and with two o f the six bolts on the extreme
ends secured. The end plates for the vibrating wire displacement gauges were oriented with
the long dimension of the plate In the direction o f the strain measurement and with the short
dimension o f the plate in the direction o f the strain measurement.
Vibrating wire
displacement gauges were mounted back-to-back on a geogrid and geotextile specimen to
evaluate the influence o f bending in the material. Tests were performed with the foil strain
gauges to evaluate the influence of the strain gauge cement and the environmental protection
layers used to protect the gauges in the field application.
»
Z
63
CHAPTER 4
RESULTS
Introduction
The results o f the study are presented in two parts. The wide width testing facility
is presented first, followed by the roadway study. The reason for this is that calibration
factors developed in the tension testing were used in processing the strain data collected by
the sensors in the roadway. Figures from the tension testing and roadway results are
presented in Appendix A and Appendix B, respectively.
Wide Width Tension Testing Facility
Accuracy and Repeatability
The accuracy o f the uniform load applied by the loading frame was investigated by
comparing the results o f the Celesco position transducers. The right and left side strain
measurements were compared to see if they were consistent, indicating a uniform strain in
the geosynthetic. Figures 19-26 show the results from the two sets o f Celesco position
transducers for geogrid and geotextile specimens in the machine and transverse directions.
From the figures it is seen that only minor differences exist in the measured strain from one
side o f the geosynthetic to the other, indicating that the loading frame provided a uniform
load and strain over the geotextile. For each test conducted, the comparison o f the global
strain response from each set o f Celesco position transducers indicated the quality of the test.
64
Ifthe two responses were significantly different, the test results were discarded. Figures 2324 show the global strain, averaged from the two sets o f Celesco position transducer
readings, from two different specimens for the geogrid and geotextile specimens in the
machine and transverse directions. The figures show that the measured results had good
repeatability.
Comparison o f Global Strain to Manufacturer’s Specifications
The global strain measurements from the Celesco position transducers were
compared to data published by the manufacturers. Figures 27-30 show the average global
strain for the geosynthetics compared to the manufacturer’s data. The behavior measured in
wide width tension tests is seen to be similar to that reported by the manufacturer.
Strain Response o f the Vibrating Wire Strain Gauges
Tests were performed with two vibrating wire strain gauges attached to the
geosynthetics. The loading level for the tests is small compared to the other sensors due to
the small strain range of the vibrating wire strain gauges (i.e. 0.3 % compared to at least 9.0
% for other instruments). Figures 31-34 show the comparison o f the strain measured by the
vibrating wire strain gauges compared to the global strain measured by the Celesco
transducers. The responses shown are the average o f both transducers, for each type o f
transducer.
From the figures it is seen that the apparent strain measured by the vibrating wire
strain gauges is lower than that from the Celesco transducers.' This is explained by the
mounting plates used for the vibrating wire strain gauges. The relatively large size of the two
65
plates left only'6.20 cm o f unconstrained geosynthetic between them. Due to the flush
interface between the mounting plate and the geotextile, it is expected that the material
confined between the plates would not be allowed to strain, which leaves only 6.20 cm o f
unconfined material to strain. To calculate strain, the measured displacement is divided by
the gauge length o f the sensor (15 cm) to produce the output plotted in figures 31-34. The
true strain experienced by the undam ped material, should then be equal to the measured
strain multiplied by the ratio o f 15 cm to 6.20 cm (a multiplication factor o f 2.4). It is this
true strain which needs to be compared to global strain measured by the Celesco transducers.
The affect o f the mounting plates on the geogrid is more complicated. The geogrid has
discrete ribs and the mounting plates contact the geogrid only at the rib intersections. For
the geogrid, it is expected that the multiplication factor would be a function o f the number
o f ribs between the mounting plates.
The measured data was experimentally calibrated such that it matches the global
strain measured. This was accomplished by finding multiplication factors (the ratio o f the
global strain to the local strain from the vibrating wire strain gauge) at different load levels.
Based on the above discussion, it is expected that the multiplication factor would be 2.4 for
all levels of load. This was not found to be the case. For the geogrid the value was found
to be 2.8 for both the machine and transverse directions. Figures 35-36 show the data with
the multiplication factors applied, plotted with the global strain measurements for the
geogrid. Multiplication factors o f 5 and 2.4 were found for the geotextile in the machine and
transverse direction respectively.
Figures 37-38 show the geotextile data with the
multiplication factors applied plotted with the global strain measurements.
66
The affect o f the mounting plates altering the strain measurement is also seen in the
results when only 4 and 2 bolts were used in each plate, (the normal is 6 bolts per plate).
Theses results are plotted in figure 39. As the number o f bolts is decreased (i.e. the
unclamped length between the plates increases) the measured strain approaches the global
strain.
Strain Response o f the Vibrating Wire Displacement Gauges
In theory, it is expected that the vibrating wire displacement gauges would exhibit
the same behavior as the vibrating wire strain gauges due to the similar clamping systems.
... For the vibrating wire displacement gauges, the nominal gauge length is 27.94 cm and the
unclamped material between the clamps is 18.5 cm. Using the rational presented above, the
expected multiplication factor is 1.5. The vibrating wire displacement gauge mounting
system is complicated however, by the ball joint connection between the plates and the
gauge. At the start o f the test bowing in the material occurred due to the tension in the
instrument’s spring causing the mounting plates to rotate. The effect was seen in both the
geotextile and the geogrid. As load was applied to the specimen, the bow was seen to
dissipate. Physical observation o f the tests indicated that the bow was not completely
removed until nearly 4 kN/m o f load was applied to the specimen. The result o f the bowing
effect was that the strain versus load curve was concave until the bow was removed. This
behavior caused the strains from the gauge to be typically larger than the global strain. This
does not seem to be possible when compared with the other gauges where the size o f the
mounting plates reduced the amount of fabric strained, resulting in local strain readings less
67
than global strain readings.
A certain amount o f the initial concave response was removed from the data
presented in figures 40-43, which shows the measured response from the vibrating wire
displacement gauges compared to the global strain for the geosynthetics in the machine and
transverse directions. The figures show that the majority o f the time the strain from the
vibrating wire displacement gauges exceeds the global strain. With the mounting plates
rotated 90° such that the undam ped area between the plates is increased, the measured strain
was seen to be closer to the global strain. This improvement occurred because the effect o f
bowing was less with the plates in this orientation.
The third set of experiments run with the gauges was to place them back to back on
each side of the geosynthetic. This was done originally to investigate any effects o f bending
in the geosynthetic. When performed however, it was also seen to reduce the bowing effect
created in the geosynthetic when a gauge was placed on one side o f the geosynthetic only.
The strain measured in both gauges was seen to be nearly identical, so the effect o f bending
was not considered to be significant in the tests. The flat orientation o f the gauge on the face
o f the geosynthetic was thought to better replicate the in-field situation where the confining
effect o f the soil surrounding the gauge would prevent bowing o f the geosynthetic. Figures
44-47 show the calibrated results o f tests performed on the geosynthetics. The vibrating wire
data plotted is the average measurement from both gauges with the appropriate multiplication
factor applied to match the global strain. With the exception o f the test with the geotextile
strained in the transverse direction, the raw strain measured by the gauges was less than the
global strain.
This coincides with the trend seen in the vibrating wire strain gauges.
68
Calibration factors used for the gauges on the geogrid were L I and 1.05 for the machine and
transverse directions respectively. For the geotextile the factors were 1.08 and I in the
machine and transverse directions.
Strain Response o f the LVDT gauges
The response o f the LVDTs is similar to that o f the vibrating wire strain gauges. The
length o f the undam ped material between the mounting plates is 3.91 cm and the nominal
length o f the gauges is 5 cm. This results in an expected multiplication factor o f 1.28. The
results o f the tests are shown in figures 48-51. With one exception (geogrid in the transverse
direction) the response from the LVDTs was less than the global strain. For geogrid strained
in the machine direction the multiplication factor used was 1.15 (see figure 52). It was found
that no multiplication factor was needed for the geogrid in the transverse direction. Figure
53 shows the results of geotextile in the machine direction, where a calibration factor o f L I 1
was used. For the geotextile in the transverse direction, a best-fit was achieved by using the
following power expression:
M.F. = 1.2 e"0'138
where M.F. is the multiplication factor and e is the strain in percent. The power expression
gives values from 1.4, at 0.2 percent strain, to 1.05 for 2.5 percent strain. The corrected
response is shown in figure 54.
Strain Response in the Foil Strain Gauges
The response measured by the foil strain gauges was found to exceed the global strain
69
(see figures 55-56)., The measurement is from two strain gauges mounted on each side o f a
rib on a geogrid specimen. Calibration factors o f 0.8 for the machine direction and 0.625 for
the transverse direction were applied and are used in the graphs shown in figures 57 and 58.
From figures 55-58, it is seen that the application o f the calibration factors does not
create a perfect match between the foil strain gauges and the global strain measurements.
The measured strain exceeds the global strain for monotonically increasing loads by a
constant factor, but during unloading and reloading a different behavior is observed. Here
it appears that the global and local response is nearly one to one. This situation is illustrated
in figure 59 where the second unloading-reloading cycle form the transverse loading test
(figure 56) is plotted by simply shifting the local response curve back on top o f the global
response curve. Here the local strain was not multiplied by the factor o f 0.625. From figure
59, it is seen that the raw local response matches the global response nearly one for one.
These results suggest that to assess the global response o f geogrids in field applications, it
is necessary to duplicate the field loadings in the laboratory. As with the other instruments,
results from the foil strain gauges show a difference between the machine and transverse
directions. This implies that the calibration factor will also have to be specific to the type
o f geogrid and procedure used to manufacture the material.
,
Comparison o f a test with a single foil strain gauge to that o f a test with foil strain
gauges mounted on each side o f one rib was also conducted. The results from a single foil
strain gauge test for geogrid in the machine direction are shown in figure 60. The local strain
in the figure has a multiplication factor of 0.8 applied to reproduce the local response seen
in figure 57. Comparing the two figures it is seen that a single strain gauge in a quarter-
70
bridge circuit produces nearly the same type o f response as two strain gauges in a half-bridge
circuit. This observation suggests two conclusions. First, there does not appear to be
significant bending during tension loading. Second, the strain gauge bonding procedure does
little to affect stiffening o f the geogrid rib.
The affect o f the environmental protection used in the roadway was the last area
studied with the wide width tension testing facility. Butyl rubber and neoprene were applied
over a single foil strain gauge. The results from these tests are shown in figures 61-62. The
response o f the foil strain gauge is quite close to the global strain when the geogrid is
strained in the machine direction. Recall that without environmental protection, response
from the foil strain gauge exceeded the global response . This implies that the environmental
protection provides some stiffening o f the geogrid in the area o f the gauge or acts to
distribute the strain in the area surrounding the gauge. The same result was seen in the
transverse direction. In figure 62, a correction factor o f 0.87 was used, while a factor o f
0. 625.was required without environmental protection. The results from the field application
have used calibration factors o f 1.00 and 0.87.
Summary
The following conclusions were made from the wide width tension tests:
1.
2.
The test head provides a uniform, uniaxial strain in the geosynthetic specimen.
The results are similar to the stress strain parameters published by the
manufacturer.
3.
The amount o f geosynthetic confined between the mounting plates has an effect
71
on the strain reading o f the instrument.
As the ratio o f undam ped length o f
geosynthetic between the mounting plates to nominal gauge length o f an instrument
decreases, so does the apparent strain measured by the instrument compared to the
global strain. A summary o f the multiplication factors used to calibrate the measured
strain to the true strain is given in table 5. The calibration factors listed in table 5
correspond to the situation most likely to. be seen in field applications and are the
factors that are used to correct the field data.
4.
The local strain in a rib o f geogrid measured by a foil strain gauge was more than
the global strain measured. A suitable calibration factor could not be used for
both monotonic loading and unloading-reloading for the foil gauges. During
monotonic loading a constant calibration factor was needed, but for unloading­
reloading ho calibration constant was needed. While the procedure used to bond
the foil gauges to the geosynthetic did not affect the response o f the geogrid, the
environmental protection placed over the foil strain gauge did.
Table 5. Multiplication Factors Applied To Sensors On The Geosynthetics
Sensor
Type
Geosynthetic
Direction
VW Strain
VW Displ.
LVDT
Foil Strain
Geogrid
Machine
2.8
1.1
1.15
1.00
Geogrid
Transverse
2.8
1.05
I
0.87
Geotextile
Machine
5
1.08
1.1
Geotextile
Transverse
2.4
1.00
i.2s-*;"»
72
Roadway Results
Long term results are listed first, followed by the dynamic results. The calibration
factors determined in the tension testing and listed in Table 5 are applied to the strain
measurements made by roadway instruments.
Long Term Results
Truck Traffic Loading. During the 14 weeks o f operation, 7550 trucks passed over
the section with a total weight o f 1490 MN. The average daily loading was 82 trucks at
approximately 200 kN per truck. Figures 63 and 64 show the daily and weekly truck traffic
on the roadway. From inspection o f the figure it is seen that daily traffic can fluctuate
greatly, by a factor o f 5 in some cases, but weekly traffic is more consistent.
Strain in the Geosynthetics.
Results from the instruments mounted to the
geosynthetics are shown in figures 65-72. The figures show the cumulative strain measured
by each instrument plotted against time. The starting point for time is July 24th, 7:00 A.M.
when the base course had been spread on the roadway, but not compacted, (see table I).
Initial instrument readings were made before the base was spread and compared to the
readings when the Campbell data logger was turned on. It was seen that very little strain was
induced in the geosynthetics prior to compaction o f the base course. Strains are positive for
tension in the geosynthetics.
The vibrating wire displacement gauges are shown in figures 65 and 66. For
vibrating wire displacement gauge #2 located on the geogrid under the proposed truck wheel
73
path, approximately 0.4 % strain was induced due to the initial compaction o f the base
course. The deformation o f the geogrid appeared to be permanent. An additional 0.15 %
strain appeared to be induced in the geogrid after completion o f the roadway and during its
first 50 hours o f traffic. At this point the Campbell data logger became inoperable and was
not able to take readings again until hour 1083. During this time the strain increased by
approximately 0.025 %. After this point spikes o f up to 0.5 % strain are seen in the data.
These are believed to be erroneous and not due to traffic loading. The behavior is seen only
in the vibrating wire gauges and has been learned to be a problem with this type o f
instrument. From hour 1083 until the end o f the study, strain is seen to increase by 0.05 %
During periods around 1400,1750, and 2000 hours, increased cumulative strain can be seen.
The second road rater test occurred about hour 1400. Increased strain was seen in other
instruments about hour 1400 and 1750. Examination o f figure 65 does not show any
increased vehicle loading during these time periods.
The results for gauge #1 (the vibrating wire displacement gauge on the geotextile off
wheel path) are presented in figure 66. The gauge did not measure a jum p in the tension o f
the geotextile during construction o f the roadway. The spike seen before the data logger
became inoperable is seen as erroneous. In general the gauge measured strain that was lower
than that measured in the geogrid, however, gauge #2 was located under an anticipated wheel
path.
The results from the vibrating wire strain gauges are shown in figures 67 and 68. The
gauge on the geogrid has strain readings well beyond its range (0.3 %), indicating that the
instrument went out o f range during construction o f the roadway and produced erroneous
74
readings.
The results o f the gauge on the geotextile located below the wheel path are shown in
figure 68. It shows more strain than the vibrating wire displacement gauge mounted on the
geotextile, but the displacement gauge was not under a wheel path. It also shows more strain
than the vibrating wire displacement gauge mounted on the geogrid located under a wheel
path.
The measurement trend of the gauge is similar to that o f the vibrating wire
displacement gauge oft the geogrid. During initial compaction 0.325 % strain was induced.
An additional 0.5 % strain occurred during completion o f the roadway and opening to traffic.
Unlike the vibrating wire displacement gauge on the geogrid however, the strain is seen to
decrease while the data logger was inoperable. Over the lifetime o f the roadway the strain
is seen to increase again by 0.03 %. Periods o f increased strain are seen about hours 1400
and 1750, like those observed in the geogrid. The increased strain about hour 2000 was not
seen as it was for the geogrid. The strain levels seen in figure 68 are greater than 0.3 %, the
maximum strain measured by the instrument. This occurrence is explained by the calibration
factor o f 2.4 applied to the instrument readings.
Results from the L V D T s mounted on the geogrid are shown in figures 69 and 70.
Each of the figures contain three graphs. The top graph is the plot o f the strain measured by
the gauge with calibration factors applied. The bottom two graphs contain temperature
compensation effects and are explained in detail below. LVDT 3 1 is more in line with the
anticipated wheel path than LVDT 32.
Both gauges indicate a response during the
construction of the roadway similar to the vibrating wire displacement gauge (figure 65).
Spikes are seen in the data when the roadway was first opened to traffic, but do not continue
75
throughout the lifetime o f the roadway. The L V D T s did not show accumulated strain over
time as the vibrating wire displacement gauges did. It could be possible that the relatively
small mounts did not have sufficient gripping ability to firmly attach to the geogrid ribs and
consequently relaxed over time.
The LVDT under the wheel path did show more
accumulated permanent strain than did the one off the wheel path.
The raw strain data showed fluctuations over 24 hour periods which led to a suspicion
that temperature effects were altering the output signal o f the LVDT’s. Periods of minimal
traffic and large temperature changes were used to investigate the effect. This usually
occurred from about 8:00 P.M. to 6:00 A.M. From the data in these periods it was seen that
there was a change in strain readings with changes in temperature. The LVDT’s did not
contain thermistors, so temperature readings from the closest vibrating wire instrument were
used. Typically, these gauges were only 1.5 m from the LVDT. The data showed that the
strain decreased as temperature decreased with the peak in the strain leading the peak in the
temperature anywhere from 2 to 10 hours.
The following procedure was used in an attempt to remove the temperature effects.
A strain versus temperature factor was generated, from the different gauges. Coefficients
ranging from 0.0108 to 0.0144 percent strain per degree Celsius were used for the LVDT’s.
The strain correction record is the middle graph o f each LVDT figure. Due to the uncertainty
with the time shift existing between the peak, strain and peak temperature with each cycle,
no correction was attempted by shifting the strain correction record.
Once the strain
correction record was generated, it was subtracted from the raw strain readings to obtain the
strain record corrected for temperature effects. This is the lower graph in the diagram. The
76
temperature corrected strain record shows a gradual increase in strain over time, which is not
obvious from the uncorrected graph. For gauge #32, a moderate increase in strain is seen
around the 1400 and 1750 hour marks as with the vibrating wire gauges. The temperature
corrected record still exhibits daily fluxuations however, which is attributed to the time shift
between the gauge response and the calculated temperature induced strain response.
Figures 71 and 72 show the response o f the LV D T s mounted on the geotextile. Like
the geogrid, the LVDT under the wheel path (#29) showed more strain than the gauge off the
wheel path (#30). The corrected strain gradually increases with time. Increases in strain are
seen around the 1400 and 1750 hour periods. It is not clear whether the temperature has an
effect on the roadway which is monitored by the instruments, or if the temperature has a
universal effect on the instruments themselves.
The long term results from the foil strain gauges were seen to be o f poor quality. This
was expected because o f drift and temperature effects in their output signals. The gauges
were primarily employed for short term dynamic use.
Strain Gauges in the Base Course. The results from the embedment gauges are
shown in figures 73-80. The zero time for the gauges is the same as for the instruments on
the geosynthetics (i.e. July 24, 7:00 A.M.) Strains are positive for gauge extension.
The ,two vibrating wire embedment displacement gauge responses are shown in
figures 73 and 74. Gauge #3 (above geogrid, off wheel path) exhibits a number of excessive
spikes which may indicate a faulty gauge. During the lifetime o f the roadway 0.3 %
compressive strain was measured. Gauge #4 (non-reinforced section, on wheel path) showed
77
that approximately 1.5% strain (extension) occurred during construction o f the roadway. The
cumulative strain increases slightly over time. Unlike gauge #3 there are much more
moderate spikes seen in the response of gauge #4.
' Figure 75 shows the response o f the vibrating wire embedment strain gauge located
above the geogrid along a wheel path. Both compression and extension were seen during the
construction o f the roadway, with strain spikes no greater than +0.2 %. A small cumulative
extension is seen in the later data. The daily response o f the gauge fluctuates between
compression and tension. This may be due to slight changes in the wheel path of the trucks.
Gauge #7 (above geotextile, on wheel path) shows permanent strain o f 0.02% from the
compaction process. There appears to be an overall increase in accumulated strain over time.
The graph for gauge #7 is shown in figure 76.
Figures 77 and 78 show the response o f LVDT embedment displacement gauges #27
and #28. The gauges are located in the base above the geogrid. The three graphs in each
figure correspond to the strain with calibration factor applied, a correction record accounting
for strain changes with temperature, and the strain record corrected for temperature changes.
The strain correction record was determined by the same technique described above for the
L V D T s on the geosynthetic.
Gauge 27 is an off-path location while gauge 28 is below the wheel path. Both
gauges show extensional strains during construction in the range o f 0.15 to 1.7 % and are
seen to result in permanent deformations. A small amount o f extensional strain (0.1-0.2 %)
was seen to accumulate after the roadway was open to traffic. The corrected record shows
a gradual increase in extensional strain with increased time.
78
Figure 79 shows the response o f the LVDT located in the base above the geotextile.
This gauge shows a similar response to the LVDT gauges located in the base above the
geogrid. The response o f the gauge located in the non-reinforced section is shown in figure
80. This gauge shows a 2.25 % strain during base leveling and recompaction.
The significance o f correcting for temperature effects is evident from figures 77-80.
W ithout taking temperature effects into account the strain records show little cumulative
effect. With the temperature corrections, significant cumulative effects are seen. The
method used to develop the temperature record involves a number o f assumptions. In light
o f these assumptions, it is not clear if the corrected record is an accurate representation o f the
material response.
Strain in the A C . Figures 81-84 show the response o f the four vibrating wire
embedment strain gauges in the asphalt concrete. For the first 48 hours after their placement,
the gauges’ response was very dramatic. At this point the data logger broke and when it was
repaired the instruments’ responses showed daily cycles of strain. The figures plot data only
after the logger was repaired due to the erratic nature o f the initial data. Vibrating wire
technology is designed to be very stable over the long term due to the nature of its output
signal as discussed previously. Due to the nature o f the instrument, it was believed that the
daily cycles o f strain were due either to expansion and contraction of the asphalt concrete or
due to mechanical loading.
To determine the contribution due to thermal expansion and contraction of the asphalt
concrete, a strain versus time response due to thermal expansion/contraction was predicted
79
using a representative coefficient of thermal expansion for asphalt concrete and the
temperature record from the instrument. The coefficient o f thermal expansion was estimated
by examining the records o f strain versus temperature for the four AC sensors during times
when traffic was not present, (i.e. approximately 8:00 P.M. to 6:00 A.M.). The records
produced coefficients ranging from 5.4x1 O'4 to I . IxlO '3 percent strain per degree Celsius.
An average coefficient of 7.2x10"4 percent strain per degree Celsius was used to predict the
strain versus time records. This strain record was then subtracted from the raw strain
readings to produce a temperature corrected plot. In the figures, the top graph is the raw
strain, the middle graph is the cyclic thermal effects only plot, and the bottom plot represents
the strain in the asphalt concrete due to mechanical loading and. any permanent effects o f
thermal cycles. From these figures it is seen that the mechanical strain also appears to be
periodic with an amplitude o f approximately 0.01 %.
Dynamic Testing Results
Truck Pass Tests. Figures 85-107 give the results from the foil strain gauges for truck
pass tests 1-9.
One o f the foil gauges was found to be inoperable when it was first
monitored, leaving only three foil strain gauges. Foil strain gauge #36 was located along a
wheel path on the west side o f the roadway.- Gauge #35 was located in the centerline o f the
roadway, and gauge #34 was located along a wheel path on the east side o f the roadway. The
strain response was zeroed with respect to the reading at the beginning o f the truck pass test.
Negative values of strain on the graph do not indicate that the geogrid was in compression.
They indicate less strain in the geogrid than at the beginning of the test. Only significant
80
responses are plottejd in the figures. During truck passes I and 2, it is seen that only gauge
#36, test 2 showed increased tension in the geogrid due to the truck pass. For truck pass 3,
all six axles o f the truck were recorded by gauge #36 with the strains positive for the first 3
axles and negative for the last three. Negative strains are seen for other gauges as well.
Similar results are seen for test 4 in figures 92-94, however the pup trailer is not seen. From
these figures it is seen that the maximum positive strains increase from approximately 0.007
% to 0.025 % and maximum negative strains decrease form -0.002 % to -0.05 % as the truck
weight increases.
The results from the tests on the surfaced roadway, tests 7-9, are presented in figures
101 to 107. The results are less defined and dramatic, but the general trends remain the
same. Maximum positive and negative strains from the test's are approximately 0.007 % and
- 0.01 % .
As discussed previously, the, roadway had only a single foil strain gauge at each
location. W ith this setup, it is not possible to compensate for the effect offending in the
geogrid rib. W ith the small deflections seen however, it is not thought that the effects o f
bending are significant.
The responses seen from the LVDT5s on the geosynthetics are not as defined as those
from the foil strain gauges. The best responses were seen in tests 4,5, and 6 which were the
heaviest loadings on the unsurfaced roadway. Figures 108-113 show the results, and as with
the foil strain gauges, only instruments showing ,a significant response are plotted. Figure
112 is the LVDT which best shows each individual truck axle. Besides the lack o f defined
responses, there are permanent strains measured with the LVDT5s that are not seen with the
81
foil gauges. The results are most likely resulting from the high profile and small mounting
area associated, with the gauge. The results cannot be explained from experience in the
tension testing facility. The importance o f close examination o f such an instrument before
use in a field application is apparent from these tests.
The results from the L V D T s embedded in the base course above the geogrid are
shown in figures 114-121. The response from these gauges was less defined than for the
LVDT5s on the geosynthetics. In only one case was a response observed for a test conducted
on the surfaced roadway.
In general, the gauges located off the wheel path show
compressional strains, while those under the wheel path show tensile strains. Compressive
strains were measured up to 0.25 % and tensile strains were as great as 0.35 %.
Figures 122-128 show the results from gauge #25, above the geotextile and below a
wheel path. The majority o f the strains are tensile, with values as great as 1.2 %. Like the
gauges above the geogrid, only one response was seen for a surfaced roadway test.
Figures 129-134 show the results for the LVDT embedment displacement gauge
located in the non-reinforced section below a wheel path. Both compressional and tensile
strains are observed for these results. The magnitude o f these strains vary from -0.15 % to
1.2 % .
Road Rater Tests. Results from the first road rater test are shown in figures 135-137.
These tests were conducted when the roadway was unsurfaced and dynamic force levels o f
6.67, 8.9, 11.1, and 13.3 IcN were applied at a frequency of 25 Hz. The Daqbook was
sampling at a rate of 20 Hz. Five distinct load cycles are seen for gauge #34 and #35. The
82
response o f #36 is not as clear. The results indicate that a tensile strain o f about 0.002 % was
induced in the geogrid due to the Road Rater loading. The results do not show the effect o f
increasing load within a cycle, but there is relaxation seen in the geogrid as the number o f
cycles increases. There is very little permanent strain in the geogrid due to the Road Rater.
The LVDTs attached to the geogrid and in the base course showed strain values well below
the sensitivity o f the instrument and were not included in the figures.
The second Road Rater test results are presented in figures 138-145. Four loading
cycles were used in the tests (11.1, 13.3, 15.6, and 17.8 kN) at a frequency o f 20 Hz and
Daqbook sampling rates from 20 to 60 Hz. The results are similar to the unsurfaced
roadway, but not as distinct. The foil strain gauge responses were plotted, but the LVDT
responses were not. The addition of asphalt on the section and the higher load levels appears
to have induced a strain in the geogrid which is comparable to the previous tests
(approximately 0.002%). These curves do tend to show the effect o f increasing load level
during a given load cycle.
Sampling the vibrating wire instruments was also performed during the second Road
Rater test on September 21st. The Campbell logger sampled one instrument at a time at a
frequency of about I Hz. The Road Rater was configured to apply 4 cycles o f the same load
such that a near constant load was applied to the pavement. The results are shown if Figures
141-145. Figure 141 shows the results from vibrating wire displacement gauge #2 located
on the geogrid. The results show a compressive strain o f approximately 0.03 % with
approximately 0.01 % o f this strain appeared to be permanent at the end o f loading. Gauge
#5 is shown in figure 142. The vibrating wire strain gauge on the geotextile shows a tensile
83
strain o f 0.002 % with about half the strain being permanent at the end o f loading.
Figure 143 shows the results from the vibrating wire embedment displacement gauge
#3 located in the base above the geogrid. The response shows that the base experienced 0.06
% of compressional strain, all o f which was permanent. Figure 144 shows the response o f
vibrating wire strain gauge #7 located in the base above the geotextile. Approximately 0.008
% tensile strain was induced in the base with 0.003 % being permanent. Figure 145 shows
the response o f vibrating wire embedment strain gauge #9 located in the asphalt concrete
above the geotextile. Approximately 0.0002 % tensile strain was induced in the asphalt
concrete with approximately half o f this strain being permanent at the end o f loading.
■ The M DT NDT unit used the Road Rater to determine the elastic moduli o f the
various roadway layers. This is the primary use o f the unit at MDT. For the July 21st tests
the largest moduli value of the base course was seen in the non-reinforced section, followed
by the geogrid section, and finally the geotextile section. In the subgrade the largest value
was seen in the geogrid, followed by the non-reinforced, then the geotextile section. These
results are shown in figure 146. Each data point on the graph was averaged from the Road
Rater tests at one location for four different force application values. Average values o f
resilient modulus for each section are given in figure 147. Results from the 21 September
test (after the asphalt concrete was placed) are shown in figures 148-149. Figure 148 shows
the resilient modulus values for the asphalt concrete layer while the base and subgrade layers
are shown in figure 149. Average values for each test section are give in figures 150 and
151. The results show somewhat lower values for the base and subgrade in the geotextile
section compared to the geogrid. Due to only one test being performed in the control section
84
it is difficult to compare to the geosynthetically reinforced sections. Due to the lack of
quality control regarding the thicknesses o f the sections it is not possible to make definitive
conclusions regarding the support characteristics o f the various sections due to the inclusion
o f geosynthetics.
85
CHAPTER 5
CONCLUSIONS AND RECOMMENDATIONS
Conclusions
The range o f the vibrating wire strain gauge is too small for this type o f work.
Studies on geosynthetic reinforced embankments and retaining walls have shown strains o f
up to 2 % in the geosynthetics. The roadway in this study did not develop large deformations
and was not near failure at the end o f the study. This however, will not be the case for follow
on research where rutting and high strain levels are anticipated. It is felt that the 0.3 % range
o f the vibrating wire strain gauges will be to small. The strain range o f the vibrating wire
displacement gauges appeared to be better suited for the range o f anticipated strains.
The effect o f stram gauge mounting plates on the geosynthetics was seen to be very
important. The only way to evaluate the effect is through tension testing. As the plates
became bigger and the unconfined length o f geosynthetic decreased, the multiplication factor
needed to convert the instrument’s measured strain to global strain increased. For this reason
the wide width tension testing facility was judged to be vital to the study.
The techniques for emplacing the embedment instruments were judged to be
successful. The irrigation control boxes allowed the base course to be placed and compacted
without damaging the instruments. Inspection o f the AC during recovery o f the gauges
showed that the carrier blocks used by the vibrating wire AC gauges appeared to form a
monolithic structure with the surrounding AC.
86
There is significant strain occurring in a roadway during construction. The vibrating
wire strain gauge on the geogrid measured 0.5 % strain by the time the roadway was finished.
Vibrating wire strain gauge # 7 in the base course was out o f range until compaction o f the
AC occurred, which knocked it into range.
The reliability o f the gauges was adequate. One foil gauge and one vibrating wire
strain gauge became inoperable. Given the total instrument count o f 24, this is a reliability
o f more than 91 %. The loss o f the Campbell data logger could have been disastrous if it had
occurred during the construction process or if large deformations occurred during the study.
Recommendations
The expanded use o f foil strain gauges would be cost effective for future research if
a suitable bonding technique can be established for geosynthetics. After conducting the
research, the author learned that there are foil strain gauges available which are much less
susceptible to drift. This would make the instruments more attractive for long term
measurements. Even though the bonding procedure used in this study was judged to be
adequate, it is not clear that a suitable bonding procedure exists for large strain measurements
in a field environment. Additionally, the issue o f bonding foil strain gauges to geotextiles
was not addressed in this study. Provided these problems could be solved, the use o f foil
strain gauges instead o f other technologies would allow more instruments to be used for the
same amount of money (the cost of foil strain gauges is in the tens o f dollars while other
technologies are in the hundreds of dollars).
87
Foil strain gauges have additional advantages over the other technologies. They are
superior to instruments with mounting plates in measuring bending. The ability to place foil
gauges on each side of a single geogrid rib allows bending as well as strain to be measured.
With large deformations this could be a significant effect. The profile o f the gauges on the
geosynthetics is not an issue as it can be for other gauges.
The quantity of strain gauges should be increased in further studies. This study was
limited in many situations to only one instrument for a measurement. Future studies should
include multiple instruments making the same measurement, such as LVDT in base course,
above geogrid, under wheel path. The redundancy o f instruments will allow more accurate
conclusions to be drawn than was possible with the present study.
Accurately determining the effect o f temperature on the gauges must be made. To
do this one alternative is to set up the long term data logger and.instruments in a strain free
environment and subject them to daily temperature cycles. Monitoring the instruments over
time would allow accurate temperature calibration constants to be generated. A second and
more costly alternative is to place dummy gauges next to the actual gauges in the roadway.
A dummy gauge is a gauge identical to the strain gauge placed at the same location, but not
measuring strain. The signal from the gauge reflects only thermal and electrical changes in
the instrument. This allows the effect of temperature and electrical drift to be compensated
for by subtracting the dummy gauge signal from the signal o f the actual strain gauge.
Placement o f a dummy gauge in the AC at a location where there is no traffic would also be
useful. This would allow measurement o f purely thermal strain to be made.
Critical equipment whose malfunction could jeopardize the entire experiment need
88
to be identified. Alternative procedures or extra equipment should be available to ensure that
data can be recorded.
The effect o f creep in geosynthetics must also be. determined. From readings on other
geosynthecially reinforced projects, it appears that the effect becomes more significant as
strain level increases. Although this study did not address this issue, it could be significant
over the lifetime o f a roadway as larger deformations occur.
Further work should concentrate on making appropriate measurements in a roadway
from which a finite element model can be made. To do this I recommend a road testing
machine similar to the Danish Road Testing Machine be constructed. Strain and pressure
measurements should be made in the transverse horizontal and vertical directions.
Appropriate strain gauges would include LVDTs and foil strain gauges due to their ability
to make long term and dynamic measurements, thus saving money that would have to be
spent on vibrating wire technology. The advantage of a road testing machine over a full scale
roadway is that the loading can be controlled more easily and set up/ tear down is easier due
to having no actual traffic and dependence on outside agencies. Only after a design method
is developed and verified in the road testing machine should a full scale instrumented
roadway be designed and constructed to verify the design method.
89
REFERENCES CITED
ASTM (1993). “Standard Test Method for Tensile Properties o f Geotextiles by the Wide
Width Strip Method,” Designation D 4595-86, 1993 Annual Book o f ASTM Standards,
Section 4, Volume 04.08 Soil and Rock; Dimension Stone; Geosynthetics, pp. 887-897.
Bathurst, R.J. (1990). “Instrumentation of Geogrid-Reinforced Soil W alls,” Transportation
Research Record, No. 1277, pp. 102-111.
Brown, S.F. (1977). “State-of-the-Art Report on Field Instrumentation for Pavement
Experiments,” Transportation Research Board, No. 640, pp. 13-28.
Christison, J.T., Anderson, K.O., and Shields, B.P. (1978). “In Situ Measurements of Strains
and Deflections in a Full-Depth Asphaltic Concrete Pavement,” Proceedings o f the
Association o f Asphalt Paving Technologists, pp. 398-433.
De Groot, M. (1990). “Towards a Standard Tensile Test: Evaluated Results o f Dutch Interlab
Test Programme,” Proceedings o f the Fourth International Conference on Geotextiles,
Geomembranes, and Related Products, Vol. 2, The Hague, The Netherlands, pp. 771-776.
Kr amp, J. (1992). “Bearing Capacity and Water, Part I: Materials, Construction, and
Instrumentation,” Institute o f Roads, Transport & Town Planning, Technical University o f
Denmark, Report No. 68, Danish Road Institute Note 238.
Leshchinsky, G.M. and Fowler, J. (1990). “Laboratory Measurement o f Load-Elongation
Relationship o f High-Strength Geotextiles,” Geotextiles and Geomembranes, Vol. 9, pp.
146-164.
Myles, B. and Carswell, LG. (1986). “Tensile Testing o f Geotextiles,” Proceedings: Third
International Conference on Geotextiles,” Vol. 13, pp. 713-718.
Myles, B. (1987). “A Review of Existing Geotextile Tension Testing Methods,” Geotextile
Testing and The Design Engineer, ASTM STP 952, pp. 57-68.
Newcomb, D.E., Wolters, R.O., and Lund, S. (1989). “Minnesota Cold Regions Pavement
Test Facility,” State o f the Art o f Pavement Response Monitoring Systems for Roads and
Airfields, U.S. Army Cold Regions Research and Engineering Laboratory, Special Report
89-23, pp. 360-369.
Oglesby, J.W., Mahmoodsadegan, B., and Griffin, P.M. (1992). “Evaluation o f Methods and
Materials Used to Attach Strain Gauges to Polymer Grids for High Strain Conditions,”
Louisiana Transportation Research Center.
90
Penner5 R. et al. (1985). “Geogrid Reinforcement o f Granular Bases,” Roads and
Transportation Association o f Canada Annual Conference, Vancouver, Canada.
Pilson, C.C., Hudson, W.R., and Anderson, V. (1995). “Experimental Design, Planning, and
Analysis o f Pavement Test Sections for the Texas Mobile Load Simulator,” Center for
Transportation Research, Report 2921-IF.
Potter, J.F., Mayhew, H.C., and Mayo, A.P. (1969). “Instrumentation o f the Full-Scale
Experiment on A l Trunk Road at Connington, Huntingdonshire,” Road Research
Laboratory, Report LR 296.
Rowe, R.K. (1986). “Determination of Geotextile Stress-Strain Characteristics Using a Wide
Width Strip Test,” Proceedings: Third International Conference on Geotextiles, Vol. 3,
Vienna, Austria, pp„ 885-890.
Rowe, R.K. and Gnanendran, C.T. (1994). “Geotextile Strain in a Full Scale Reinforced Test
Embankment,” Geotextiles and Geomembranes, Vol. 13, pp. 781-806.
Sebaaly, P., et al. (1989). “Instrumentation For Flexible Pavements,” Federal Highway
Administration, Report FHWA-RD-89-084.
Selig, E.T. (1975). “Soil Strain Measurement Using Inductance Coil Method,” Performance
Monitoring for Geotechnical Construction, American Society for Testing Materials, ASTM
STP 584, pp. 141-158.
Simac, M.R. et al. (1990). “Instrumented Field Performance o f a 6 m Geogrid Soil W all,”
Proceedings o f the 4th International Conference on Geotextiles, Geomembranes and Related
Products, pp. 53-69.
Smith, T.E., et al. (1995). “Laboratory Behaviour o f Geogrid and Geotextile Reinforced
Flexible Pavements,” Virginia Polytechnic Institute and State University, Blacksburg, VA.
Stubstad, R.N., Khosla, N.P., and Wynn, W.W. (1989). “Construction o f Fully Instrumented
Test Pavements in North Carolina,” State o f the Art o f Pavement Response Monitoring
Systems for Roads and Airfields, U.S. Army Cold Regions Research and Engineering
Laboratory, Special Report 89-23, pp. 394-401.
Ullidtz, P. and Busch, C. (1979). “Laboratory Testing of a Full-Scale Pavement: The Danish
Road-Testing Machine,” Transportation Research Bulletin, No 715, pp. 52-62.
Van Deusen, D.A., et al. (1992). “A Review o f Instrumentation Technology for the
Minnesota Road Research Project,” FHWA/MN/RC-92/10.
91
Webster, S.L. (1992). “Geogrid Reinforced Base Courses For Flexible Pavements For Light
Aircraft: Test Section Construction, Behaviour under Traffic, Laboratory Tests, and Design
Criteria,” U.S. Army Watenvays Experiment Station, Corps o f Engineers, Paper
DOT/FAA/RD-92/25.
WWW, (1996).. Home page o f the Minnesota Road Research Project on the world wide web,
http://mnroad.dot.state.mn.us/
92
APPENDICES
93
APPENDIX A
WIDE WIDTH TENSION TESTING RESULTS
Load (kN/m)
94
2
Strain (%)
Load (kN/m)
Geogrid, Machine Direction
Strain (%)
Figure 20. Global Strain From Two Sets of Celesco Gages:
Geogrid, I ransverse Direction
95
Strain (%)
Figure 21. Global Strain From Two Sets o f Celesco Gages:
Geotextile, Machine Direction
Strain (%)
Figure 22. Global Strain From Two Sets of Celesco Gages:
Geotextile, Transverse Direction
96
10.0
Load (kN/m)
8.0
6.0
4.0
2.0
0.0
0.0
1.0
2.0
3.0
4.0
5.0
s 23. Global Strain From Two Tests: Geogrid, Machine Direction
Load (kN/m)
c
Strain (%)
Strain (%)
Figure 24. Global Strain From Two Tests: Geogrid. Transverse Direction
97
Strain (%)
Figure 25. Global Strain From Two Tests: Geotextile, Machine Direction
Strain (%)
Figure 26. Global Strain From Two Tests: Geotextile, Transverse Direction
4
98
Strain (%)
Figure 27. Comparison of Results to Manufacturer’s Data: Geogrid, Machine Direction
Strain (%)
Figure 28. Comparison of Results to Manufacturer’s Data:
Geogrid, Transverse Direction
99
Strain (%)
Figure 29. Comparison o f Results to Manufacturer’s Data:
Geotextile, Machine Direction
Strain (%)
Figure 30. Comparison of Results to Manufacturer’s Data:
Geotextile, Transverse Direction
100
VW Strain Gase
Global Strain
Strain (%)
Figure 3 1. Vibrating Wire Strain Gage: Geogrid, Machine Direction
VW Strain Gage
Global Strain
Strain (%)
Figure 32. Vibrating Wire Strain Gage: Geogrid, Transverse Direction
101
VW Strain Gage
Global Strain
Strain (%)
Figure 33. Vibrating Wire Strain Gage: Geotextile, Machine Direction
VW Strain Gage
Global Strain
Strain (%)
Figure 34. Vibrating Wire Strain Gage: Geotextile, Transverse Direction
4
102
Load (kN/m)
Global Strain
VW Strain Gage
H
Strain (%)
. Calibrated Vibrating Wire Strain Gage: Geogrid, Machine Direction
Load (kN/tn)
Global Strain
VW Strain Gage
Strain (%)
Figure 36. Calibrated Vibrating Wire Strain Gage: Geogrid, Transverse Direction
If
103
VW Strain Gage
Global Strain
Strain (%)
Figure 37. Calibrated Vibrating Wire Strain Gage: Geotextile, Machine Direction
Global Strain
VW Strain Gage
Strain (%)
Figure 38. Calibrated Vibrating Wire Strain Gage: Geotextile, Transverse Direction
104
VW-6 Bolt:
VW-2 Bolts
Global Strain
VW-4 Bolts
Strain (%)
Figure 39. Comparison of Vibrating Wire Strain Gage With 6, 4 and 2 Bolts Fastened
Global Strain
/
/
/
/
/
Z
Z VW Displ. Gage
Strain (%)
Figure 40. Vibrating Wire Displacement Gage: Geogrid, Machine Direction
105
Global Strain
VW Displ. Gage
Strain (%)
Figure 41. Vibrating Wire Displacement Gage: Geogrid, Transverse Direction
Global Strain
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
Strain (%)
Figure 42. Vibrating Wire Displacement Gage: Geotextile, Machine Direction
106
Load (kN/m)
Global Strain
VW Displ. Gage
. Vibrating Wire Displacement Gage: Geotextile, Transverse Direction
Load (kN/m)
%
Strain (%)
Global Strain
Strain (%)
Figure 44. Calibrated Back-to-Back Vibrating Wire Displacement Gage:
Geogrid, Direction
■e
107
Global Strain
VWD Displ. Gage
Strain (%)
Figure 45. Calibrated Back-to-Back Vibrating Wire Displacement Gage:
Geogrid, Transverse Direction
Global Strain
Strain (%)
Figure 46. Calibrated Back-to-Back Vibrating Wire Displacement Gage:
Geotextile, Machine Direction
108
Global Strain
VW Displ. Gage
Strain (%)
Figure 47. Calibrated Back-to-Back Vibrating Wire Displacement Gage:
Geotextile, Transverse Direction
LVDT Displ. Gage
Global Strain
Strain (%)
Figure 48. LVDT Displacement Gage: Geogrid, Machine Direction
109
LVDT Displ. Gage
Global Strain
I
I
Strain (%)
Figure 49. LVDT Displacement Gage: Geogrid, Transverse Direction
LVDT Displ. Gage
Global Strain
Strain (%)
Figure 50. LVDT Displacement Gage: Geotextile, Machine Direction
■f
no
LVDT Displ. Gage
Global Strain
Strain (%)
Figure 5 1. LVDT Displacement Gage: Geotextile, Transverse Direction
Global Strain
LVDT Displ. Gage
Strain (%)
Figure 52. Calibrated LVDT Displacement Gage: Geogrid, Machine Direction
I l l
Global Strain —
LVDT Displ. Gage
Strain (%)
Figure 53. Calibrated LVDT Displacement Gage: Geotextile, Machine Direction
LVDT Displ. Gage
Global Strain
Strain (%)
Figure 54. Calibrated LVDT Displacement Gage: Geotextile, Transverse Direction
4
112
Global Strain
Foil Strain Gage
Strain (%)
Figure 55. Foil Strain Gage: Geogrid, Machine Direction
Global Strain
Foil Strain Gage
Strain (%)
Figure 56. Foil Strain Gage: Geogrid, Transverse Direction
4
113
Foil Strain Gage
Global Strain
Strain (%)
Figure 57. Calibrated Foil Strain Gage: Geogrid, Machine Direction
Foil Strain Gage
Global Strain
0.0
1.0
2.0
3.0
4.0
5.0
Strain (%)
Figure 58. Calibrated Foil Strain Gage: Geogrid, Transverse Direction
6.0
I*
114
Global Strain
Foil Strain Gage
Strain (%)
Figure 59. Unloading-Reloading Response Foil Strain Gage:
Geogrid, Transverse Direction
_
Calibrated Foil Gage: 1/4 Bridge
Global Strain
Strain (%)
Figure 60. Calibrated 1/4 Bridge Foil Strain Gage: Geogrid, Machine Direction
i#
115
Foil Strain Gase
Global Strain
Strain (%)
Figure 61. Calibrated Foil Strain Gage With Environmental Protection:
Geogrid, Machine Direction
Foil Strain Gage
10.0
-
Global Strain
Strain (%)
Figure 62. Calibrated Foil Strain Gage With Environmental Protection:
Geogrid, Transverse Direction
116
APPENDIX B
ROADWAY RESULTS
•#
117
Z
50000
g 40000
k-
D
CL
30000
-C
'S
20000
U
2
H 10000
5
o
H
D ay s
T o tal T ru ck W eig h t p e r W eek (kN )
Figure 63. Daily Traffic Loading History
0 I 2 3 4 5 6 7 8 9 10 11 12 13 14
W eek s
Figure 64. Weekly Truck Traffic Loading History
118
base course compacted
0
500
1000
1500
Time (hours)
2000
2500
Figure 65. VW Displacement Gage #2 on Geogrid (on wheel-path)
base course compacted
ain (%)
logger inoperable
logger repaired
base course recompacted and opened to traffic
-0.05
1000
150C
Time (hours)
Figure 66. VW Displacement Gage #1 on Geotextile (off wheel-path)
119
O
500
1000
1500
Time (Hours)
2000
2500
Figure 67. VW Strain Gage #6 on Geogrid (off wheel-path)
logger inoperable
.E 0.2
logger repaired
base course compacted
0
500
1000
1500
Time (Hours)
2000
Figure 68. VW Strain Gage #5 on Geotextile (on wheel-path)
2500
4
Strain (%)
120
-0.4
0
500
1000
1500
2000
2500
2000
2500
2000
2500
T im e (H rs)
0.2
------
_
_ Strain (%)
0.4 - -
0
500
1000
1500
Strain (%)
T im e (H rs)
0
500
1000
1500
T im e (H rs)
Figure 69. LVDT Displacement Gage #31 on Geogrid (on wheel-path)
4
121
0.6
Strain (%)
0.4
0.2
0
0.2
-
-0.4
0
500
1000
1500 2000 2500
T im e (H rs)
0.6
Strain (%)
0.4
0.2
0
-0.2
-0.4
0
500
1000
1500 2000 2500
T im e (H rs)
0.6
Strain (%)
0.4
0.2
0
-
0.2
-0.4
0
500
1000
1500 2000 2500
T im e (H rs)
Figure 70. LVDT Displacement Gage #32 on Geogrid (off wheel-path)
•»
122
Strain (%)
0.2 i
-0.2
Strain (%)
-0.3 -------------------------------------------0
500 1000 1500 2000 2500
Time (Hrs)
500
1000 1500
Time (Hrs)
2000
2500
0
500
1000 1500
Time (Hrs)
2000
2500
ain (%)
0
Figure 71. LVDT Displacement Gage #29 on Geotextile (on wheel-path)
4
strain (%
123
1000 1500
Time (Hrs)
2000
2500
0
500
1000 1500
Time (Hrs)
2000
2500
0
500
1000 1500
Time (Hrs)
2000
2500
Strain (%)
Strain (%)
500
Figure 72. LVDT Displacement Gage #30 on Geotextile (off wheel-path)
124
base course compacted
logger repaired
logger inoperable
base course recompacted and open to traffic
Time (hours)
Figure 73. VW Embedment Displacement Gage #3 in Base Above Geogrid
(off wheel-path)
2
I
________ I_______ _______ _______ _______ ______ ^
I
1.5
^
O
-0.5
!
:
!logger inoperable
I
I«
I
I '
1 . .
OPPrP r r e n a i r e n
I
\
base course recompacted and open to traffic
_____ II_____ :I_____ :I_____ I_____ TI_____ iI_____ -,_____
I
'_____ I_____ I_______ !_______ I_________
base course compacted
_______________ _____
500
I
. __________•__________i_________
!_________________ :
1000
1500
Time (Hours)
2000
2500
Figure 74. VW Embedment Displacement Gage #4 in Base in Non-Reinforced Section
(on wheel-path)
4
125
base course compacted
logger inoperable
“ base course recompacted and open to traffic
1000
1500
Time (Hours)
Figure 75. VW Embedment Strain Gage #8 in Base Above Geogrid (on wheel-path)
base course compacted
0.05
, logger inoperable
0 I
^ -0.05
a
£
-0.1
-0.15
3
X
logger repaired
I
I
base course recompacted and open to traffic
—
-0.2
0
500
1000
1500
Time (Hours)
2000
2500
Figure 76. VW Embedment Strain Gage #7 in Base Above Geotextile (on wheel-path)
•>
126
Strain (%)
2.5
-0.5
0
500
1000
1500
2000
2500
2000
2500
2000
2500
Strain (%)
Time (Hrs)
0
500
1000
1500
ain (%)
Time (Hrs)
uo 0.5
0
500
1000
1500
Time (Hrs)
Figure 77. LVDT Embedment Displacement Gage #27 in Base Above Geogrid
(off wheel-path)
127
0.6
Strain (%)
0.4
0.2
0
-0.2
-0.4
0
500
1000
1500
2000
2500
Time (Hrs)
%
0.6
Strain (%)
0.4
0.2
0
-0.2
-0.4
0
500
1000
1500
2000
2500
2000
2500
ain (%)
Time (Hrs)
500
1000
1500
Time (Hrs)
Figure 78. LVDT Embedment Displacement Gage #28 in Base Above Geogrid
(on wheel-path)
Strain (%)
128
500
1000
1500
2000
2500
2000
2500
2000
2500
Strain (%)
Time (Hrs)
-0.2
500
1000
1500
ain (%)
Time (Hrs)
0
500
1000
1500
Time (Hrs)
Figure 79. LVDT Embedment Displacement Gage #25 in Base Above Geotextile
(on wheel-path)
129
on 0.5
0
500
1000
1500
T im e (H rs)
2000
2500
-0.5 J
0
500
1000
1500
T im e (H rs)
2000
2500
2.5
2 qxr 1.5
O
•C
•—
«
II
5
0.5
_______ I
I
—
-
J
0
-0.5 J
0
;
'
________
I
________ _________
I
I
;
____u_
1
I
500
'
:I
I
;
'
!
!
1
;___ L_
I
i
I
1000
1500
T im e (H rs)
-
I
!
2000
2500
Figure 80. LVDT Embedment Displacement Gage #26 in Base in Non-Reinforced
Section (on wheel-path)
130
0.02
Strain (%)
0.015
0.01
0.005
0
-0.005
-
0.01
0
500
1000
1500
2000
2500
Time (Hrs)
0.02
Strain (%)
0.015
0.01
0.005
0
■0.005
-
0.01
0
500
1000
1500
2000
2500
2000
2500
Time (Hrs)
0.02
ain (%)
0.015
0.01
0.005
0
-0.005
-
0.01
0
500
1000
1500
Time (Hrs)
Figure 81. VW Embedment Strain Gage #10 in AC Above Geogrid (on wheel-path)
0.02
S train (% )
0.01
0
-
0.01
-
0.02
0
500
1000
1500
T im e (H rs)
2000
2500
0
500
1000
1500
T im e (H rs)
2000
2500
0
500
1000
1500
T im e (H rs)
2000
2500
0.02
S train (% )
0.01
0
■
0.01
0.02
■
0.02
S train (% )
0.01
0
-
0.01
-
0.02
Figure 82. VW Embedment Strain Gage #11 in AC Above Geogrid (off wheel-path)
132
0.02
i
Strain (%)
0.01
0
-
0.01
-
0.02
0
500
1000
1500
2000
2500
2000
2500
Time (Hrs)
0.02
Strain (%)
0.01
0
-
0.01
-
0.02
0
500
1000
1500
Time (Hrs)
0.02
Strain (%)
0.01
0
-
0.01
-
0.02
0
500
1000
1500
2 0 0 0 2500
Time (Hrs)
Figure 83. VW Embedment Strain Gage #9 in AC Above Geotextile (on wheel-path)
133
0.01
0.005
-0.015
500
1000
1500
T im e (H rs)
2000
2500
0.01
-0.015
500
1000
1500
T im e (H rs)
2000
2500
500
1000
1500
T im e (H rs)
2000
2500
0.01
0.005
Figure 84. VW Embedment Strain Gage #12 in AC in Non-Reinforced Section
(on wheel-path)
4
134
0.0004
0.0002
F irti
Strain (%)
0
-
0.0002
-
0.0004
-
0.0006
-
0.0008
-
-
Il I
-I____ L
—
0.001
0.0012
0
2
4
6
8
10
12
14
16
18 20
22
24
Time (Sec)
Figure 85. Foil Strain Gage #36 on Geogrid (on wheel path): Truck Pass Test I
ain (%)
0.0005 i
0.0005
- 0.001
£
-
0.0015
-
0.002
Time (Sec)
Figure 86. Foil Strain Gage #35 on Geogrid (below centerline): Truck Pass Test I
135
0.006
0.005
0.004
•
0.003
&
_I
____ _____I
____ I
____
■
I
I
_I
____ i
____ I
____ I
____ I
____
I
‘J 0.002
I
;
I
CZD
;
i
^
i
:
10
12
i
____ :___
I
,
J ____
I____
____ I___
0.001
0
-
0.001
0
2
4
6
8
________ ,____i____ _____
I
14
16
18 20
22
24
T im e (S ec )
Figure 87. Foil Strain Gage #36 on Geogrid (on wheel path): Truck Pass Test 2
0.0004
0.0002
&
J= -0 .0002
s
^
-
0.0004
-
0.0006
-
0.0008
8
10
12
14
16
18 20
22
24
T im e (S ec )
Figure 88. Foil Strain Gage #35 on Geogrid (below centerline): Truck Pass Test 2
136
0.025
0.015
.E
0.01
0.005
-0.005
0
2
4
6
8
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 89. Foil Strain Gage #36 on Geogrid (on wheel path): Truck Pass Test 3
0 .0 0 0 4
S train (% )
0.0002
0
-
0.0002
-0 .0 0 0 4
-0 .0 0 0 6
-0 .0 0 0 8
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 90. Foil Strain Gage #35 on Geogrid (below centerline): Truck Pass Test 3
137
S train (% )
0.001
-
0.001
-
0.002
-
0.003
-
0.004
-
0.005
-
0.006
-
0.007
0
2
4
6
8
10
12
14
16
18 20
22
24
^
T im e (S ec )
igure 91. Foil Strain Gage #34 on Geogrid (on wheel path): Truck Pass Test 3
S train (% )
0.015
0.005
-
0.005
0
2
4
6
8
10
12
14
16
18 20
22
24
T im e (S ec)
Figure 92. Foil Strain Gage #36 on Geogrid (on wheel path): Truck Pass Test 4
■f
138
0.001
I
I
I
M
I
i,
I
^
i
i
j
M
-
!
!
:
■
1
Strain (%)
0
'
1
'
1
I
=
■ 1
0.001
-
-
0.002
-
0.003
;
:
;
; I
1
I
t f f t a y p r f c w d W h n iiw r iy
I
I
I
I
18 20
22
I
^
"
I
0.004
-
0
2
4
6
8
10
12
14
16
24
3
T im e (S ec)
^ure 93. Foil Strain Gage #35 on Geogrid (below centerline): Truck Pass Test 4
0.002
0
Strain (%)
0.002
0.004
-
0.006
-
0.008
-
0.01
0
2
4
6
8
10
12
14
16
18 20
22
24
T im e (S ec)
Figure 94. Foil Strain Gage #34 on Geogrid (on wheel path): Truck Pass Test 4
4
139
0.02 i
S train (% )
0.015
0.005
-
0.005
0
2
4
6
8
10
12
14
16
18 20
22
24
T im e (S ec)
Figure 95. Foil Strain Gage #36 on Geogrid (on wheel path): Truck Pass Test 5
ain (% )
0.002 i
-
0.002
-
0.004
-
0.008
-
0.012
0
2
4
6
8
10
12
14
16
18 20
22
24
T im e (S ec)
Figure 96. Foil Strain Gage #35 on Geogrid (below centerline): Truck Pass Test 5
4
140
2 -0.03
-0 .0 4
-0.06
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 97. Foil Strain Gage #34 on Geogrid (on wheel path): Truck Pass Test 5
0.025
0.02
0.015
.E
0.01
0.005
-0.005
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 98. Foil Strain Gage #36 on Geogrid (on wheel path): Truck Pass Test 6
141
0.001
-
0.001
-
0.002
.S -0.003
oo -0.004
-0.005
-0 .0 0 6
-0 .0 0 7
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 99. Foil Strain Gage #35 on Geogrid (below centerline): Truck Pass Test 6
0.002 i
^
-
0.002
£ -0 .0 0 4
-0 .0 0 6
-0.008
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 100. Foil Strain Gage #34 on Geogrid (on wheel path): Truck Pass Test 6
142
0.006
0.004
Strain (%)
0.002 ------:
-
0.002
-
0.004
-
0.006
-
0.008
0
2
4
6
8
10
12
14
16
18
20
22
24
^
Time (Sec)
igure 101. Foil Strain Gage #36 on Geogrid (on wheel path): Truck Pass Test 7
ain (%)
0.002
£
-
0.002
-
0.004
-
0.006
-
0.008
0
2
4
6
8
10
12
14
16
18
20
22
24
Time (Sec)
Figure 102. Foil Strain Gage #35 on Geogrid (below centerline): Truck Pass Test 7
143
0.0005
0
^
0.0005
-
&
.H -0.001
C ti
t:
^
0.0015
-
-
0.002
0.0025
-
0
2
4
6
8
10
12
14
16
18 20
22
24
Time (Sec)
Figure 103. Foil Strain Gage #34 on Geogrid (on wheel path): Truck Pass Test 7
0.0005 i
^
-
£
0.0005
-
-
0.001
0.0015
-
0.002
Time (Sec)
Figure 104. Foil Strain Gage #36 on Geogrid (on wheel path): Truck Pass Test 8
144
0.006
0.005
0.004
^
E
s
.
'
0.003
S
^ 0.002
0.001
0
0
2
4
6
8
10
12
14
16
18
20
22
24
Time (Sec)
Figure 105. Foil Strain Gage #35 on Geogrid (below centerline): Truck Pass Test 8
0.006 i
0.005
0.004
.S
0.003
0.002
0.001
Time (Sec)
Figure 106. Foil Strain Gage #36 on Geogrid (on wheel path): Truck Pass Test 9
145
^c -0.0004
-0.0006
^ -0.0008
-
-
0.001
0.0012
-0.0014
8 10 12 14 16 18 20 22 24
Time (Sec)
Figure 107. Foil Strain Gage #35 on Geogrid (below centerline): Truck Pass Test 9
0
2
4
6
8
10 12 14 16 18 20 22 24
Time (Sec)
Figure 108. LVDT Displacement Gage #31 on Geogrid (on wheel path):
Truck Pass Test 4
146
n
2
O
4
6
8
10
12
14
16
18 20
22
24
Time (Sec)
Figure 109. LVDT Displacement Gage # 3 1 on Geogrid (on wheel path):
Truck Pass Test 5
0.15
'
I
i
i L
I
ain (%
;
4 ____ L -
0.1
I
I
I
J ________ L _ j _ .
0.05
1
.b
4
, m
m
.
oo
J ------- -L
-
0.05
0
2
4
6
8
10
12
14
16
18 20
22
24
Time (Sec)
Figure 110. LVDT Displacement Gage #32 on Geogrid (off wheel path):
Truck Pass Test 5
147
-0 .0 6
--------------------------------------------------------------------------------0
2
4
6
8
10 12 14
T im e (S e c )
16
18
20
22
24
Figure 111. LVDT Displacement Gage #29 on Geotextile (on wheel path):
Truck Pass Test 4
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 112. LVD I' Displacement Gage #29 on Geotextile (on wheel path):
Truck Pass Test 5
148
0.15
0.1
&
. ¥ 0.05
2
0
-0.05
0
2
4
6
8
10 12 14 16 18 2 0 22 24
T im e (S ec )
Figure 113. LVDT Displacement Gage #29 on Geotextile (on wheel path):
T r u c k Pass Test 6
0
2
4
6
8
0.01
0
-
0.01
3 - 0 .0 2
Ox
S -0.03
2
oo -0 .0 4
-0.05
-0 .0 6
-0 .0 7
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 114. LVDT Embedment Displacement Gage #27 in Base Above Geogrid
(off wheel path): Truck Pass Test 3
149
-0 .0 5 - —
.S
-0.1
0
2
4
6
8
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 115. LVDT Embedment Displacement Gage #27 in Base Above Geogrid
(off wheel path):Truck Pass Test 6
0
2
4
6
8
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 116. LVDT Embedment Displacement Gage #27 in Base Above Geogrid
(off wheel path):Truck Pass Test 7
150
0.1
0.08
_
0 .0 6
&
.S
0 .0 4
S
^
0.02
0
-
0.02
0
2
4
6
8
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 117. LVDT Embedment Displacement Gage #28 in Base Above Geogrid
(on wheel path):Truck Pass Test 2
^
0.2
0
2
4
6
8
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 118. LVDT Embedment Displacement Gage #28 in Base Above Geogrid
(on wheel path)Truck Pass Test 3
151
0 .1 5 - -
.S
0.05
0
2
4
6
8
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 119. LVDT Embedment Displacement Gage #28 in Base Above Geogrid
(on wheel path):Truck Pass Test 4
0.1
0.05
0
£ - 0 .0 5
I -0,
tZ)
-0 .1 5
-
0.2
-0 .2 5
0
2
4
6
8
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 120. LVDT Embedment Displacement Gage #28 in Base Above Geogrid
(on wheel path):Truck Pass Test 5
152
O
2
4
6
8
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 121. LVDT Embedment Displacement Gage #28 in Base Above Geogrid
(on wheel path) Truck Pass Test 6
C o-0 .0 4
Ox
E -0 .0 6
£) -0.08
0
2
4
6
8
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 122. LVDT Embedment Displacement Gage #25 in Base Above Geotextile
(on wheel path): Truck Pass Test I
153
0.2
J
l
l
i
I
,
0.15
- . - L
?
0.1
-
_____________________ __
.
T
I
I
I
0.05
T
' n
r
r
i
L
H r-M rw uW dH ii
r
A
.
0
I
I
■
I
'
I
;
-0.05
0
2
4
6
8
10 12 14
T im e (S ec)
16
18
20
22
24
Figure 123. LVDT Embedment Displacement Gage #25 in Base Above Geotextile
(on wheel path): Truck Pass Test 2
.5 0.4
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 124. LVDT Embedment Displacement Gage #25 in Base Above Geotextile
(on wheel path): Truck Pass Test 3
154
^
0.4
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 125. LVDT Embedment Displacement Gage #25 in Base Above Geotextile
(on wheel path): Truck Pass Test 4
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 126. LVDT Embedment Displacement Gage #25 in Base Above Geotextile
(on wheel path): Truck Pass Test 5
155
0.8
—
.E 0.6
cz) 0.4
-
0.2
0
2
4
6
8
10 12 14 16 18 20 22 24
Time (Sec)
Figure 127. LVDT Embedment Displacem ent Gage #25 in Base A bove G eotextile
(on wheel path): Truck Pass Test 6
.E 0.02
a
0.01
-0.02 J
0
2
4
6
8
10 12 14 16 18 20 22 24
Time (Sec)
Figure 128. LVDT Embedment Displacement Gage #25 in Base Above Geotextile
(on wheel path): Truck Pass Test 7
156
0.08
-----------------------------------------------------------------------------------
—
0.06
S train (% )
0.04 6
>
i
I
:
.
!
10 12 14
T im e (S ec )
16
18
20
22
i
0.02
I 0
T im
.
-#
-#
— —._ .
_ —— _
.
—-
—
-0.02
-0 .0 4 -
I
-0.06
0
2
4
6
8
24
Figure 129. LVDT Embedment Displacement Gage #26 in Base of Non-Reinforced
Section (on wheel path): Truck Pass Test I
0.02
-0.08
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 130 LVDT Embedment Displacement Gage #26 in Base of Non-Reinforced
Section (on wheel path): Truck Pass Test 2
157
O
2
4
6
8
10
12
14
16
18
20
22
24
Time (Sec)
Figure 131. LVDT Embedment Displacement Gage #26 in Base o f Non-Reinforced
Section (on wheel path): Truck Pass Test 3
0
2
8
10
12
14
16
18
20
22
24
Time (Sec)
Figure 132. LVDT Embedment Displacement Gage #26 in Base of Non-Reinforced
Section (on wheel path): Truck Pass Test 4
158
.5
0.6
czD 0.4
0
2
4
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 133. LVDT Embedment Displacement Gage #26 in Base o f Non-Reinforced
Section (on wheel path): Truck Pass Test 5
0
2
4
6
8
10 12 14
T im e (S ec )
16
18
20
22
24
Figure 134. LVDT Embedment Displacement Gage #26 in Base of Non-Reinforced
Section (on wheel path): Truck Pass Test 6
159
.S 0.0 0 0 8
2
£ 0 .0 0 0 6
30
40
T im e (S ec)
Figure 135. Foil Strain Gage #34 on Geogrid, July 31st Test
0.002
S train (% )
0.0 0 1 5
0.001
0.0 0 0 5
0
-0 .0005
0
20
40
T im e (sec)
60
Figure 136. Foil Strain Gage #35 on Geogrid, July 31st Test
80
160
0 .0 0 2 -
S train (% ) .
0.0015
0.001
0.0005
0
-0 .0005
_
0
20
40
60
80
100
T im e (S ec)
120
140
160
'igure 137. Foil Strain Gage #36 on Geogrid, July 31st Test
0.0008
■ain (% )
0.0006
0.0004
0.0002
~
0
-
0.0002
0
20
40
Tim e (Sec)
60
Figure 138. Foil Strain Gage #34 on Geogrid. September 21 st Test
80
161
0.0004
0.0002
^
0
•I
^
-
0.0002
-0.0004
-0.0006
0
10
20
30
Tim e (Sec)
40
50
60
Figure 139. Foil Strain Gage #35 on Geogrid, September 21st Test
0.0004
n(% )
0.0002
2
0
-
0.0002
^ -0.0004
-0.0006
-0.0008
0
10
20
30
Tim e (Sec)
40
50
Figure 140. Foil Strain Gage #36 on Geogrid, September 21st Test
60
162
O
20
40
60
T im e (Sec)
80
100
120
Figure 141. VW Displacement Gage #2 on Geogrid, September 21st Test
-0.005
Tim e (Sec)
Figure 142. VW Strain Gage #5 on Geotextile, September 21st Test
163
-0.005
-0.015
£
-
- 0.02
-0.025
-0.035
Tim e (Sec)
Figure 143. VW Embedment Displacement Gage #3 in Base Above Geogrid,
September 21st Test
0.008
^ 0.006
£ 0.004
0.002
Tim e (Sec)
Figure 144. VW Embedment Strain Gage #7 in Base Above Geotextile,
September 21st Test
164
0.0004
0.0003
^ 0.0002
0.0001
~
-
0.0001
Tim e (Sec)
Figure 145. VW Embedment Strain Gage #9 in AC Above Geotextile,
September 21st Test
?
Geotextile
200
Subgrade
% 100
Geogrid Section
Unreinforced
2
3
4
5
6
7
8
9
10 11
12 13
Test Number
Figure 146. Resilient Modulus Values From July 21 Road Rater Test
14
165
?
200
# Subgrade
100
-
Section
Figure 147. Average Resilient Modulus Values From July 21 Road Rater Test
?
4000
c2
1000
Unreinforced
Geogrid Section
Geotextile
16
18
20
Test Number
Figure 148. Resilient Modulus of AC Layer From September 21 Road Rater Test
166
Geogrid Section
Subgrade
I
■
I
I
Test Number
Figure 149. Resilient Modulus of Base and Subgrade Layers from
September 2 1 Road Rater Test
Section
Figure 150. Average Resilient Modulus o f AC Layer
From September 2 1 Road Rater Test
.
I
167
Subgrade
Section
Figure 151. Average Resilient Modulus of Base and Subgrade Layers From
September 21 Road Rater Test
3 1762
I
I
I
!
I