University of Nevada, Reno

Reflective Cracking of Flexible Pavements: Literature

Review, Analysis Models, and Testing Methods

A thesis submitted in partial fulfillment of the

requirements for the degree of Master Science in
Civil Engineering

by

Luis Guillermo Loria-Salazar

Elie Y. Hajj, Ph.D.

Peter E. Sebaaly, Ph.D, P.E
Thesis Advisors

May, 2008
 i

THE GRADUATE SCHOOL

We recommend that the thesis

prepared under our supervision by

LUIS GUILLERMO LORIA-SALAZAR

entitled

Reflective Cracking Of Flexible Pavements: Literature Review, Analysis Models,

And Testing Methods

be accepted in partial fulfillment of the

requirements for the degree of

MASTER OF SCIENCE

Dr. Peter E. Sebaaly, P.E., Advisor

Dr. Elie Y. Hajj, Committee Member

Dr. Gary Norris, P.E., Committee Member

Dr. George G.C. Fernandez, Graduate School Representative

Marsha H. Read, Ph. D., Associate Dean, Graduate School

May, 2008
 ii

Abstract

Hot mixed asphalt (HMA) overlay is one of the commonly used methods for

rehabilitating deteriorated pavements. The Nevada Department of Transportation

(NDOT) uses HMA overlays as a rehabilitation technique for the majority of the state’s

flexible pavements. One major type of distress influencing the life of an overlay is

reflective cracking. In the past, NDOT has experimented with a number of techniques to

reduce the impact of reflective cracking on HMA overlays like cold in-place recycling,

reinforced fabrics, stress relief courses, mill and overlay, Portland Cement Concrete

(PCC) rubblization, and PCC crack and seat.

In 2006, the Nevada DOT initiated a three-phase research project to identify the

promising techniques to mitigate reflective cracking in HMA overlays: a) Phase I:

Review of literature and the performance of the various techniques in Nevada, b) Phase

II: Identify analysis models and laboratory tests, and c) Phase III: field verification of the

selected techniques.

A literature review was conducted for the current and previous efforts outside

Nevada on the reflective cracking mitigation techniques in HMA overlays. The standard

specifications on the reflective cracking mitigation techniques from all fifty state DOTs

were reviewed and summarized. Thirty two states out of fifty have specified a reflective

cracking control system in their current standard specifications.

Based on the review of the currently available analytical models to predict the

resistance of HMA overlays to reflective cracking, three design methods were identified

and summarized:
 iii

• Virginia Tech Simplified Overlay Design Model

• Rubber Pavements Association Overlay Design Model
• The New AASHTO model for Reflective Cracking

An overlay design was conducted for three different HMA overlay mixes

according to the three identified overlay design methods. In a summary, the Virginia

Tech method showed a thinner overlay thickness for the stiffer mix whereas, the Rubber

Pavements Association method, which considers both stiffness and fatigue characteristics

of the mix, the overlay thickness was dependent on the interaction between the two

material properties. On the other hand, a unique and thick overlay thickness was found

with the new AASHTO method as it does not consider the material properties of the

overlay mix as part of the design.

Additionally, a literature review was performed for the available laboratory tests

to evaluate the resistance of HMA mixtures to reflective cracking. None of the reviewed

laboratory test methods has undergone field validation except the Texas Transportation

Institute (TTI) Overlay Tester which showed consistency between the mixtures’ test

results and their corresponding field performance. The TTI Overlay Tester results on the

cores taken from different highway projects showed that asphalt mixtures performed very

well in the field when the reflective cracking life (from the overlay tester) is larger than

300.

Finally, based on the analysis of the various findings it was recommended to:

• Further evaluate the stress relief course as a reflective cracking mitigation

technique under Nevada’s conditions.
• Use the TTI Upgraded Overlay Tester to evaluate mixtures in the
Laboratory for reflective cracking resistance.
• Use the Rubber Pavements Association Overlay Design Model to design
the require overlay thickness.
 iv

Dedication

The Author dedicates this dissertation to his parents, Luis Guillermo Loria-
Meneses and Sandra Salazar-Morales, and to his beloved wife, Belkys Gonzalez
Deloria.
 v

Dedication

I would like to express my deep gratitude to Dr. Peter E. Sebaaly and Dr. Elie

Hajj for their endorsement, advice and assistance during the research work. Your

guidance and supervising are greatly appreciated.

Thanks to Dr. George Fernandez for all the help and cooperation in achieving the

statistical analysis.

Thanks to Dr. Norris for his participation in my thesis committee.

I would like to thank my two beloved sisters, Monica and Marcela, who always

have encouraged me in any moment of my life.

A special thanks and appreciation to the Nevada Department of Transportation for

making resources available and for their support and help.

I would like to acknowledge and thanks my friends at the Pavement Program for

all the good times we have had together, I will always have you in my heart: Haissam,

Jessa, Mano, Ziad, George, Raghu, Prateep, Thileepan, Bheem Raj, Shiva, Alvaro,

Wendy, Aaran, Nathan, Edward, Roy, Tusha and Andrew.

Finally, I would like to thank the good LORD, for always provide me of his

mercy and bless.

 vi

TABLE OF CONTENTS

Chapter 1- INTRODUCTION ........................................................................................................ 1

Chapter 2- LITERATURE REVIEW ............................................................................................. 7

2.1 Review of Current and Previous Efforts................................................................................7
2.1.1 Arizona Department of Transportation.......................................................................... 7
2.1.2 California Department of Transportation..................................................................... 10
2.1.3 Colorado Department of Transportation...................................................................... 11
2.1.4 Georgia Department of Transportation........................................................................ 13
2.1.5 Illinois Department of Transportation ......................................................................... 13
2.1.6 Indiana Department of Transportation......................................................................... 16
2.1.7 Louisiana Department of Transportation ..................................................................... 17
2.1.8 Michigan Department of Transportation ..................................................................... 18
2.1.9 Mississippi Department of Transportation................................................................... 19
2.1.10 Pennsylvania Department of Transportation ............................................................. 20
2.1.11 Texas Department of Transportation ......................................................................... 22
2.1.12 Virginia Department of Transportation ..................................................................... 26
2.1.13 Wisconsin Department of Transportation.................................................................. 27
2.1.14 International Practice ................................................................................................. 30
2.2 DOTs Standard Specifications for Highway Construction..................................................33

Chapter 3 – REVIEW AND ANALYSIS OF NDOT REFLECTIVE CRACKING

MITIGATIONS TECHNIQUES .................................................................................................. 34
3.1Review of Nevada DOT Experience.....................................................................................34
3.1.1 Performance Related Data ........................................................................................... 34
3.2 Performance of Selected NDOT Projects ............................................................................38
3.2.1 Cold In-Place Recycling (CIR) Projects ...................................................................... 39
3.2.2 Reinforced Fabrics (RF) Projects................................................................................. 44
3.2.3 Stress Relief Course (SRC) Projects............................................................................ 47
3.2.4 Mill and Overlay (MOL) Projects................................................................................ 50
3.2.5 Crack and Seat (CS) Projects....................................................................................... 54
3.2.6 Rubblization (RPCC) Projects ..................................................................................... 57
3.3 Overall Summary of the Performance of the Selected NDOT Projects...............................60
3.4 Statistical Analysis of the Surface Cracking Data – Principal Component Analysis ..........63
3.4.1 Overall Ranking of the Performance of Selected NDOT Projects .............................. 65
3.4.2 Fixed Qualitative and Qualitative Factorial Analysis.................................................. 71
3.5 Final Analysis from the NDOT Projects..............................................................................75
3.6 Summary of the Phase I Findings ........................................................................................76

Chapter 4 – REVIEW OF REFLECTIVE CRACKING ANALYSIS MODELS ........................ 79

4.1 Virginia Tech Simplified Overlay Design Model................................................................79
4.1.1 Fundamentals of Reflective Cracking................................................................... 79
4.1.2 Fracture Mechanics Analysis Using Finite Element Method ............................... 82
 vii

4.1.3 Development of Design Equations .............................................................................. 84

4.1.4 Service Life Prediction for Pavements with Potential Reflective Cracking ......... 84
4.1.5 Design Example .................................................................................................... 86
4.2 Asphalt Rubber Association Overlay Design Model...........................................................86
4.2.1 Proposed Reflective Cracking Design Method............................................................ 89
4.3 The New AASHTO model for Reflective Cracking............................................................94
4.3.1 HMA Overlay of Cracked HMA Pavement Surface ................................................... 95
4.3.2 HMA Overlay of Crack-free HMA Pavement Surface................................................ 96
4.4 Design Example Using the Various Analysis Models.........................................................97
4.4.1 Traffic Data.................................................................................................................. 97
4.4.2 HMA Overlay Material Properties............................................................................... 98
4.4.3 Overlay Design Using Virginia Tech Simplified Overlay Model ............................... 99
4.4.4 Overlay Design Using Asphalt Rubber Association Method ...................................... 99
4.4.5 Overlay Design Using the New AASHTO Model for Reflective Cracking.............. 100
4.4.6 Summary of Design Example .................................................................................... 101

Chapter 5 – REVIEW OF REFLECTIVE CRACKING TESTING METHODS ...................... 102

5.1 Cracow University of Technology, Cracow, Poland .........................................................102
5.2 Technion-Israel Institute of Technology, Haifa, Israel......................................................104
5.3 Geo-materials Laboratory, ENTPE, France.......................................................................106
5.4 Technical University of Vienna, Austria ...........................................................................108
5.5 Laboratory of Public Roads of Autum, France..................................................................111
5.6 University College of Dublin, Ireland ...............................................................................112
5.7 University of Illinois, United States...................................................................................114
5.8 Aeronautical Technological Institute (ATI), Brazil...........................................................116
5.9 Florida Atlantic University, United States.........................................................................117
5.10 Polytechnic University of Madrid, Spain.........................................................................120
5.11 Regional Laboratory of Ponts et Chausses, France .........................................................121
5.12 Texas Transportation Institute Overlay Tester, United States .........................................122
5.12.1 Variability of Upgraded Overlay Testing ................................................................ 125
5.12.2 Sensitivity of Upgraded Overlay Testing ................................................................ 126
5.12.3 Validation of Upgraded Overlay Tester................................................................... 128
5.13 Federal University of Rio Grande do Sul, Brazil.............................................................131
5.14 Summary of Laboratory Test Methods ............................................................................133

Chapter 6 – SUMMARIES AND RECOMMENDATIONS...................................................... 134

6.1 Summary of Phase I Findings ............................................................................................134
6.1.1 Recommendations of Phase I..........................................................................................134
6.2 Summary of Phase II Findings...........................................................................................135
6.2.1 Analysis Models......................................................................................................... 135
6.2.2 Laboratory Tests ........................................................................................................ 138
6.2.3. Recommendations of Phase II .......................................................................................139
6.3 Overall Recommendations.................................................................................................139
6.5 Proposed Plan for Phase III................................................................................................153
REFERENCES ........................................................................................................................... 155
 viii

TABLES ..................................................................................................................................... 161

FIGURES.................................................................................................................................... 191
APPENDIX A – Summary of Various Studies to Mitigate Reflective Cracking....................... 267
APPENDIX B – DOTs Standard Specifications for Highway Construction ............................. 275
APPENDIX C – PC1 Values Along with Measured Surface Cracks of NDOT Projects........... 282
 ix

LIST OF TABLES

Table 3.1 List of Selected NDOT Projects. ................................................................................ 162

Table 3.2a Pavement Distresses Summary of Selected NDOT Projects. ................................... 163
Table 3.3a Ranking of Selected NDOT Projects Based on Principal Component Analysis. ..... 165
Table 3.4 Statistical Analysis of the Various Treatments Based on the PCA Rankings at
One-year...................................................................................................................................... 167
Before Treatment Construction and One-year After Treatment Construction for Flexible
Pavements. .................................................................................................................................. 167
Table 3.5 Statistical Analysis of the Various Treatments Based on the PCA Rankings at
One-year...................................................................................................................................... 167
Before Treatment Construction and Three-years After Treatment Construction for
Flexible Pavements. .................................................................................................................... 167
Table 3.6 Statistical Analysis of the Various Treatments Based on the PCA Rankings at
One-year...................................................................................................................................... 168
Before Treatment Construction and Five-years After Treatment Construction for Flexible
Pavements. .................................................................................................................................. 168
Table 3.7 Fixed Qualitative Factorial Analysis Based on the PCA Rankings at One-year ........ 169
After Treatment Construction. .................................................................................................... 169
Table 3.8 Fixed Qualitative Factorial Analysis Based on the PCA Rankings at Three-
years After Treatment Construction............................................................................................ 170
Table 3.9 Fixed Qualitative Factorial Analysis Based on the PCA Rankings at Five-years ...... 171
After Treatment Construction. .................................................................................................... 171
Table 3.10 Fixed Qualitative-Quantitative Factorial Analysis Based on the PCA Rankings
at One-year After Treatment Construction. ................................................................................ 172
Table 3.11 Fixed Qualitative-Quantitative Factorial Analysis Based on the PCA Rankings
at Three-years After Treatment Construction. ............................................................................ 173
Table 3.12 Fixed Qualitative-Quantitative Factorial Analysis Based on the PCA Rankings
at Five-years After Treatment Construction. .............................................................................. 174
Table 4.1 Plane Strain Fracture Toughness (KIc) for Different HMA Mixtures........................ 175
Table 4.2 Statistical Coefficients for the εVM Model................................................................... 175
Table 4.3 Minimum and Maximum Values for the Pavement Thicknesses and Moduli. .......... 175
Table 4.4 Traffic Information. .................................................................................................... 175
Table 4.5 HMA Overlay Dynamic Modulus at 70°F and 10 Hz. ............................................... 176
Table 4.6 HMA Overlay Thickness Using the New AASHTO Model. ..................................... 176
Table 5.1 Results of the Bending Test under Repeated load. ..................................................... 177
Table 5.2 Results of the Shearing Test. ...................................................................................... 177
Table 5.3 Comparison of Different Mitigation Techniques........................................................ 177
Table 5.4 Geogrid effectiveness factor (FEG)............................................................................ 178
Table 5.5 Three Test Cells of MnRoad: Asphalt Mixture and Cracking Performance. ............. 178
Table 5.6 Summary of Laboratory Testing Methods for Reflective Cracking. .......................... 179
Table 6.1 Summary of Literature Review................................................................................... 183
Table 6.2 Summary of NDOT Reflective Cracking Mitigation Techniques Review................. 184
Table 6.3 Pavement Layers Material Properties. ........................................................................ 185
Table 6.4 Overlay Design Thicknesses for 20 years Design Period........................................... 185
 x

Table 6.5 ADOT Rubber Gradation for SAMI (Type A). .......................................................... 185
Table 6.6 Required Properties for the Asphalt-rubber................................................................ 186
Table 6.7 Required Properties for the Mineral Aggregates. ....................................................... 186
Table 6.8 Laboratory Asphalt Rubber Binder Design Data........................................................ 186
Table 6.9 Requirements for CALTRANS SAMI-R’s Construction........................................... 187
Table 6.10 Gradations for Ground Tire Rubber.......................................................................... 187
Table 6.11 FDOT Standard Specifications for Asphalt-rubber. ................................................. 187
Table 6.12 MDOT Standard Specifications for Asphalt-Rubber Gradation............................... 187
Table 6.13 Aggregate Quality Requirements.............................................................................. 188
Table 6.14 Gradation Requirements for Fine Aggregate............................................................ 188
Table 6.15 Master Gradation Bands and Volumetric Properties. ............................................... 188
Table 6.16 Mixture Design Properties. ....................................................................................... 189
Table 6.17 Compacted Lift Thickness and Required Core Height............................................. 189
Table 6.18 Minimum Pavement Surface Temperatures. ............................................................ 189
Table 6.19 UDOT Specifications for Asphalt Binder................................................................. 189
Table 6.20 Aggregate Gradation for SAMI According to UDOT Specifications. ..................... 190
Table 6.21 UDOT Specifications for SAMI. .............................................................................. 190
 xi

LIST OF FIGURES

Figure 3.1: Performance data of contract 2808a on US050 from mileposts 0 to 3..................... 192
Figure 3.3: Performance data of contract 2838 on SR396 from mileposts 1.13 to 7.7............... 194
Figure 3.4: Performance data of contract 2935 on SR360 from mileposts 0 to 23.25................ 195
Figure 3.5: Performance data of contract 2819 on US095 from mileposts 6.98 to 14.38 .......... 196
Figure 3.6: Performance data of contract 2873 on US095 from mileposts 56.24 to 72.00 ........ 197
Figure 3.7: Performance data of contract 2961 on US006 from mileposts 18.87 to 43.99 ........ 198
Figure 3.8: Performance data of contract 3013 on US095A from mileposts 24.55 to 44.18 ..... 199
Figure 3.9: Performance data of contract 3025a on US093 from mileposts 0 to 11................... 200
Figure 3.10: Performance data of contract 3025b on US093 from mileposts 109.33 to
131.55.......................................................................................................................................... 201
Figure 3.11: Performance data of contract 3025c on US093 from mileposts 147.45 to
172.4............................................................................................................................................ 202
Figure 3.12: Performance data of contract 2876 on SR208 from mileposts 2.33 to 8.86........... 203
Figure 3.13: Performance data of contract 2761 on SR443 from mileposts 0.00 to 0.87........... 204
Figure 3.14: Performance data of contract 2932 on US095 from mileposts 49.24 to 50.41 ..... 205
Figure 3.15: Performance data of contract 2980a on US050 from mileposts 19.4 to 21.14....... 206
Figure 3.16: Performance data of contract 2980b on US095 from mileposts 26.19 to
28.22............................................................................................................................................ 207
Figure 3.17: Performance data of contract 3006 on IR080 from mileposts 50.0 to 50.23 ......... 208
Figure 3.18: Performance data of contract 3008 on SR227 from mileposts 1.39 to 6.56........... 209
Figure 3.19: Performance data of contract 2723 on US095 from mileposts 96.89 to 117.5 ..... 210
Figure 3.20: Performance data of contract 3031 on US395 from mileposts 25.21 to 29.24 ...... 211
Figure 3.21: Performance data of contract 3048 on SR157 from mileposts 12.47 to 21.1......... 212
Figure 3.22: Performance data of contract 3045 on US050 from mileposts 36.81 to 37.38 ...... 213
Figure 3.23: Performance data of contract 3162on US395 from mileposts 17.37 to 19.4 ......... 214
Figure 3.24: Performance data of contract 2384a on US095 from mileposts 14.81 to 16.6....... 215
Figure 3.25: Performance data of contract 2384b on US095 from mileposts 0 to 2.74 ............. 216
Figure 3.26: Performance data of contract 2432 on SR157 from mileposts 3.41 to 12.18......... 217
Figure 3.27: Performance data of contract 2505 on US095 from mileposts 62.54 to 64.01 ...... 218
Figure 3.28: Performance data of contract 2651a on US095 from mileposts 12 to 15............... 219
Figure 3.29: Performance data of contract 2651b on US095 from mileposts 43 to 44.8 ........... 220
Figure 3.30: Performance data of contract 2651c on US095 from mileposts 32.88 to 39.......... 221
Figure 3.31: Performance data of contract 2679 on US095 from mileposts 19.96 to 32.88 ...... 222
Figure 3.32: Performance data of contract 3028 on SR512 from mileposts 0 to 1.38................ 223
Figure 3.33: Performance data of contract 3070 on SR160 from mileposts SR 160.................. 224
Figure 3.34: Performance data of contract 2886 on IR080 from mileposts 3.36 to 9.06 ........... 225
Figure 3.35: Performance data of contract 2889 on IR080 from mileposts 26.61 to 32 ............ 226
Figure 3.36: Performance data of contract 2962 on IR080 from mileposts 117.68 to
132.72.......................................................................................................................................... 227
Figure 3.37: Performance data of contract 2999 on IR080 from mileposts 69.02 to 74.92 ....... 228
Figure 3.38: Performance data of contract 3021 on IR080 from mileposts 26.77 to 41.99 ....... 229
Figure 3.39: Performance data of contract 2549 on IR080 from mileposts 11.08 to 20 ............ 230
Figure 3.40: Performance data of contract 2869 on IR080 from mileposts 1.13 to 7.51 ........... 231
 xii

Figure 3.41: Performance data of contract 2901 on IR080 from mileposts 11.68 to 17.36 ....... 232
Figure 3.42: Performance data of contract 3088 on IR080 from mileposts 18.59 to 26.21 ....... 233
Figure 3.43: Performance data of contract 3186a on IR080 from mileposts 60.32 to 61.38...... 234
Figure 3.44: Performance data of contract 3186b on IR080 from mileposts 85.25 to 26.21 ..... 235
Figure 3.45: Comparison between the ranking of the various treatments at 1-year before ........ 236
treatment construction and 1-year after treatment construction. ................................................ 236
Figure 3.46: Comparison between the ranking of the various treatments at 1-year before
treatment construction and 3-year after treatment construction. ................................................ 236
Figure 3.47: Comparison between the ranking of the various treatments at 1-year before
treatment construction and 5-year after treatment construction. ................................................ 237
Figure 4.1 Fracture Modes.......................................................................................................... 238
Figure 4.2 Schematic representation of reflective cracking device (RCD) ................................ 238
Figure 4.3 (a) Bottom platens of the reflective cracking device................................................. 238
(b) Reflective cracking device with both (top and bottom) plates.............................................. 238
Figure 4.4 Representation of the overlay zone subjected to the reflective cracking .................. 239
Figure 4.5 Flowchart of the Asphalt Rubber Association overlay design method ..................... 240
Figure 4.6 Existing HMA pavement structure............................................................................ 241
Figure 4.7 Required overlay thickness according to the Virginia tech method.......................... 241
Figure 4.8 T2C required overlay thickness according to the asphalt rubber method ................. 242
Figure 4.9 CT required overlay thickness according to the asphalt rubber method ................... 242
Figure 4.10 NRM required overlay thickness according to the asphalt rubber method ............. 243
Figure 4.11 Overlay thickness according to the asphalt rubber method for selected
percent cracking .......................................................................................................................... 244
Figure 4.12 Required overlay thickness according to the new AASHTO method..................... 244
Figure 4.13 Required overlay thickness according to the new AASHTO method for 100%
reflected cracks ........................................................................................................................... 245
Figure 4.14 Required T2C overlay thickness using various analysis models............................. 245
Figure 4.15 Required CT overlay thickness using various analysis models............................... 246
Figure 4.16 Required NRM overlay thickness using various analysis models........................... 246
Figure 5.1 Bending test under static load – Cracow University of Technology......................... 247
Figure 5.2 Results of the bending test under static load for HMA beams with and without
geotextile at 20°C........................................................................................................................ 247
Figure 5.3 Shearing test – Cracow University of Technology.................................................... 248
Figure 5.4 The laboratory wheel-tracking device ....................................................................... 248
Figure 5.5 Beam specimens for the laboratory wheel-tracking device....................................... 249
Figure 5.6 Summary of test results under the laboratory wheel-tracking device ....................... 250
Figure 5.7 Fissurometer apparatus description ........................................................................... 250
Figure 5.7 a. Principal of testing method (Tschegg 1986).......................................................... 251
b. Schematic load-deformation curve ......................................................................................... 251
Figure 5.8 Specimen shapes for the wedge splitting test method (Tschegg 1986)..................... 251
Figure 5.9 Typical results from the wedge splitting test method................................................ 251
Figure 5.10 Load-displacement curves for gravel and crushed HMA mixtures......................... 252
Figure 5.11 a. Maximum splitting force Fmax versus test temperature........................................ 252
b. Specific fracture energy Gf versus test temperature................................................................ 252
Figure 12 Shrinkage-bending test device for reflective cracking resistance .............................. 253
 xiii

Figure 5.13 Test results example for shrinkage-bending test device.......................................... 253
Figure 5.14 Mechanism of cracking of overlay .......................................................................... 253
Figure 5.16 Test set-up for top-down cracking tests................................................................... 254
Figure 5.17 Testing equipment for ISAC system ....................................................................... 254
Figure 5.18 Strain in HMA overlay as function of test cycles – control section........................ 255
Figure 5.19 Strain in HMA overlay as function of test cycles – ISAC section .......................... 255
Figure 5.20 Instrumentation of fatigue test................................................................................. 256
Figure 5.21 Cracking pattern at the end of test – beam without geogrid.................................... 256
Figure 5.22 Cracking pattern at the end of test – beam with geogrid......................................... 257
Figure 5.23 Schematic representation of the test setup............................................................... 257
Figure 5.24 Schematic representation of WRC test.................................................................... 258
Figure 5.25 Wheel Reflective Cracking (WRC) equipment....................................................... 258
Figure 5.26 Wheel Reflective Cracking device test specimen in plates of adherence ............... 259
Figure 5.27 Mechanism simulating the relative vertical movement in the WRC....................... 259
Figure 5.28 General scheme of the MEFISTO device................................................................ 259
Figure 5.29 Schematic of MEFISTO device and testing sample (one column) ......................... 260
Figure 5.30 Typical test results from the MEFISTO device....................................................... 260
Figure 5.31 Concept of TTI overlay tester (after Cleveland et al............................................... 260
Figure 5.32 Upgraded TTI overlay tester ................................................................................... 261
Figure 5.33 Typical TTI overlay tester result (each opening and closing is 10 sec.) ................. 261
Figure 5.34 Upgraded overlay tester SGC sample...................................................................... 261
Figure 5.35 Repeatability of overlay testing on TxDOT type D mixture................................... 262
Figure 5.36 Relationship between number of specimens and specified tolerance of
reflective cracking life for TxDOT type D mixture.................................................................... 262
Figure 5.37 Influence of temperature on reflective cracking life ............................................... 262
Figure 5.38 Influence of opening displacement on reflective cracking life ............................... 263
Figure 5.39 Influence of asphalt content on reflective cracking life .......................................... 263
Figure 5.40 Influence of air voids on reflective cracking life..................................................... 263
Figure 5.41 UFRGS-DAER traffic simulator from Brazil.......................................................... 264
Figure 5.42 Test pavement showing cracked areas and instrumentation location...................... 264
Figure 5.43 Test pavement sections............................................................................................ 264
Figure 5.44 Failed HMA layer cracking pattern......................................................................... 265
Figure 5.45 Transverse cracks painted in different colors according to their appearance.......... 265
Figure 5.46 Cracking severity index evolution in both overlays ................................................ 265
 1

Chapter 1- INTRODUCTION
Pavement rehabilitation is rapidly becoming one of the most important issues

facing many highway departments. Hot mixed asphalt (HMA) overlay is one of the

commonly used methods for rehabilitating deteriorated pavements. The Nevada

Department of Transportation (NDOT) uses HMA overlays as a rehabilitation technique

for the majority of the state’s flexible pavements.

One major type of distress influencing the life of an overlay is reflective cracking

(1). When asphalt overlays are placed over jointed or severely cracked existing rigid and

flexible pavements, cracks will reflect to the surface in a relatively short period of time.

Physical tearing of the overlay occurs because of movements under heavy wheel loads at

joints and cracks in the underlying pavement layer. Therefore, the long-term

performance of the HMA overlays will depend on their ability to resist reflective

cracking. Reflective cracking in the overlay allows water to percolate into the pavement

structure and weaken the HMA and the supporting layers, hence contributing to many

forms of pavement deterioration. Moisture can damage the HMA mix by promoting the

stripping of the asphalt binder from the aggregate. It can also significantly reduce the

strength of the base and subgrade materials, which would lead to the total failure of the

flexible pavement structure.

Numerous previous efforts have been exerted to reduce or prevent reflective

cracking of HMA overlays including the increase thickness of HMA overlay, the use of

stress absorbing membranes inter-layers, the use of fabrics and geotextiles membranes,

and the fracturing of the existing concrete slabs. The basic principle of reflective

cracking is that the tensile stresses at the interface of the crack and the new HMA overlay
 2

are significantly increased due to the discontinuity at the tip of the crack. The developed

tensile stresses rapidly exceed the tensile strength of the HMA overlay and the crack

forms at the interface and quickly propagates to the surface. Combating reflective

cracking can be achieved by either one of the two approaches: a) reduce the magnitude of

the tensile stresses at the crack-overlay interface or b) increase the tensile strength of the

HMA overlay.

The increase of the thickness of the HMA overlay as well as placing a stress

absorbing membrane inter-layer follow the approach of reducing the magnitude of the

tensile stresses at the crack-overlay interface. The stress absorbing membrane inter-layer

usually consists of a single or double chip seal. The chip seal is a highly flexible layer

which reduces the magnitude of the tensile stresses before they intersect with the new

HMA layer. The ability of the stress absorbing membrane inter-layer to reduce the

tensile stresses increases as its thickness increase (i.e. single versus double chip seal), as

its binder content increases, and as the flexibility of the binder increases. However,

having a thick, rich and highly flexible stress absorbing membrane inter-layer may cause

potential rutting and shoving problems under heavy traffic. An optimum design must be

established in order to effectively mitigate reflective cracking without negatively

impacting the long-term performance of the HMA overlay.

The fabric and geotextile technique follows the approach of increasing the tensile

strength of the HMA overlay. These materials have high tensile strengths and if they are

effectively bonded to the HMA layer, will improve the tensile strength of the overlay.

There are numerous brands of fabrics and geotextiles currently in the market covering a

wide range of cost and strength properties. Selecting the best type of fabric or geotextile
 3

requires an in-depth assessment of their relative properties and their long-term

performance.

The concept of slab fracturing before overlaying follows the approach of reducing

the magnitude of the tensile stresses and resulting strains at the crack-overlay interface by

reducing the movement of the cracked or broken concrete slabs beneath the overlay.

Methods used to fracture slabs are generally divided into three categories: rubblization,

cracking and seating, and breaking and seating. Rubblizing involves reducing the

concrete slab into fragments having textural and gradation characteristics similar to a

crushed aggregate material. It is usually accomplished with a resonant pavement breaker

and has been used on all types of concrete pavements (i.e., jointed plain concrete (JPC),

jointed reinforced concrete (JRC), and continuously reinforced concrete (CRC)

pavements). Crack/seat and break/seat are fracture techniques intended to produce very

short rigid slabs with effective lengths on the order of 12 to 48 inches. The process of

cracking or breaking the pavement is usually accomplished with modified pile drivers,

guillotines, or spring arm (whip) hammers. A significant distinction between the two

techniques exists. Crack/seat is associated with the fractured slab technique conducted

solely on JPC pavements. The intent of cracking is to create closely spaced pieces small

enough so that vertical and horizontal movement is reduced but full aggregate interlock is

maintained to permit load transfer across the crack with little loss of structural capacity.

Break/seat, on the other hand, is the process by which JRC slabs are more severely

fractured. A significantly higher level of energy is required to accomplish breaking of

JRC slabs because of the need to destroy the bond between reinforcing steel and concrete

(2).
 4

In addition to the development in construction methods to resist reflective

cracking, recent efforts also included the development of pavement design techniques to

account for reflective cracking. Three methods were identified and evaluated: The newly

developed AASHTO Mechanistic-empirical Pavement Design Guide (MEPDG) method

(3), the Rubber Pavements Association (RPA) Overlay Design method and the Virginia

Tech Simplified Overlay Design Model. The MEPDG method offers an analysis model

for reflective cracking but leaves the selection of the mitigation technique to the agency.

The Virginia Tech Simplified method is based on the linear elastic fracture mechanics

(LEFM) method and provides a regression equation to predict the number of cycles to

failure against the reflective cracking of an HMA overlay using the moduli and the

thicknesses of the new and the old pavement layers. The RPA Overlay Design method

provides a mathematical model based on statistical analysis and finite element analysis

(FEM). All these three methods are used to determine the required overlay thickness to

delay the reflective cracking in the new overlay.

Furthermore, recent efforts included the use of laboratory and field tests to

evaluate the resistance of HMA mixtures and pavements to reflective cracking.

Laboratory tests are typically used to evaluate the resistance of the HMA mixtures to

reflective cracking during the mix design stage. Filed test are used to evaluate the

performance of the rehabilitation technique used to minimize reflective cracking in HMA

pavements.

The research effort documented herein was directed toward identifying an

effective method to eliminate the propagation of the cracks from the old surface layer
 5

through the new HMA overlay. As discussed earlier, the propagation of the cracks

through the new HMA layer is impacted by various factors such as: treatment of the

cracks in the old pavement, the installation of a crack retarding layer, properties of the

new HMA mix, traffic loads, and environmental conditions. Traffic loads and

environmental conditions are primarily fixed for a given project location, which leaves

crack treatment, retarding layer, and the design of the new mix as the three factors that

can be optimized. NDOT recognizes that the reflection of these cracks through the newly

constructed HMA overlay must be effectively controlled in order to achieve good long-

term performance of the overlay.

In 2006, the Nevada Department of Transportation (NDOT) initiated a three-

phase research project to identify the most promising techniques to mitigate reflective

cracking in HMA overlays: a) Phase I: Review of literature and the performance of the

various techniques in Nevada, b) Phase II: Identify analysis models and laboratory tests,

and c) Phase III: Field verification of the selected techniques. The research effort

presented in this thesis includes the findings of Phase I and Phase II along with

recommendations for the Phase III study.

Under Phase I, the current and previous efforts on the mitigation of reflective

cracking in HMA overlays were reviewed. Additionally, the reflective cracking

techniques that have been used by NDOT were identified and analyzed to asses their

effectiveness in mitigating reflective cracking under Nevada’s conditions.

Under Phase II of this study, the currently available analytical models to predict

the resistance of HMA overlays to reflective cracking were identified and assessed based

on their technical merit and their ability to predict the performance of HMA overlays.
 6

Additionally, laboratory and field test methods used to evaluate the resistance of HMA

mixtures to reflective cracking were reviewed and summarized.

 7

Chapter 2- LITERATURE REVIEW

This first task of Phase I of the research was to review the current and previous

efforts used to mitigate reflective cracking in HMA overlays. An exerted effort was

made to identify all the case studies that have been completed to evaluate the

effectiveness of the various techniques under actual field conditions. The relevant

literature and data were collected from the various studies and analyzed to assess the

effectiveness of the various techniques that are currently available in the industry. The

following chapter describes the various studies conducted and are summarized in

Appendix A in the same order listed in the text.

2.1 REVIEW OF CURRENT AND PREVIOUS EFFORTS

2.1.1 Arizona Department of Transportation

The Arizona Department of Transportation (ADOT) has been distinguished as a

major leader in the use of Asphalt-Rubber (AR) overlays on Continuously Reinforced

Concrete Pavement (CRCP). In 1989, Scofield (4) documented in a research report the

history, development, and performance of asphalt rubber in Arizona. In that report the

following conclusion was stated, “asphalt rubber has successfully been used as an

encapsulating membrane to control pavement distortion due to expansive soils and to

reduce reflection cracking in overlays on both rigid and flexible pavements.”

In 1990, ADOT designed and constructed a large scale Asphalt-Rubber test

project in Flagstaff, Arizona on the very heavily trafficked Interstate 40 (5,6). The 1999

traffic exceeded 20,000 vehicles per day with 35% heavy trucks. The purpose of the test

project was to determine whether a relatively thin overlay with AR could reduce
 8

reflective cracking. AR is a mixture of 80% hot paving grade asphalt and 20% ground

tire rubber. This mixture is also commonly referred to as the Asphalt-Rubber wet process

or the McDonald process. The overlay project was built on top of a very badly cracked

concrete pavement that was in need of reconstruction. The design section included edge

drains, crack and seat the concrete pavement, a five inch overlay consisting of a 3-inch

conventional dense HMA overlaid by a two inch gap graded asphalt-rubber mix (AR-AC)

with a 6.5% binder and a one-half inch AR OGFC with a 9% binder content. It was

reported that the AR overlay has performed beyond original expectations. After nine

years of service the overlay was still nearly crack-free, with good ride, virtually no rutting

or maintenance, and good skid resistance. The benefits of using AR on this project

represent about $18 million in construction savings and four years less construction time.

The Strategic Highway Research Program SPS-6 test sections constructed in conjunction

with the project further illustrate the very good performance of AR. Results of this

project have led to widespread use of AR hot mixes throughout Arizona. On the basis of

this work, over 2,000 miles of successfully performing AR pavements have been

constructed since 1990 (7).

In 1989, Rahman, et al. (8) documented the results of the installation of three

commercial paving fabrics for the reduction of reflective cracking in asphalt overlays in

Arizona. The fabrics installed were Paveprep (by PavePrep Corporation), Glassgrid (by

Bay Mills Ltd.) and Tapecoat (by Tapecoat Company). The following recommendations

were made by the researchers:

 9

• The need for proper tack coat selection based on the expected construction

conditions and product selections when paving fabrics are used in pavement

rehabilitation.

• The need for additional field testing of Paveprep on milled surfaces.

• Caution regarding the use of Glassgrid on rough surfaces.

In summary, proper installation procedures are critical for optimum performance.

Installation of paving fabrics has become more sophisticated in recent years. But it is by

no means a closed science in the respect that everything has been learned (9).

In 1996, a study was conducted by the University of Arizona (7,10) to determine

the effectiveness of the stress absorbing membrane interlayer (SAMI). Various types of

fabric were analyzed and a simple ‘Pull-off’ test was used to measure the bitumen/fabric

bond strengths. Fabric structure and rate of spray of emulsion tack-coat were found to be

the variables that had the most significant effect. In addition, it was found that absorption

and retention of water from the tack-coat was a major factor. A series of samples

consisting of fabrics sandwiched between asphalt cores were manufactured and sheared

by a direct shear mechanism. Additional samples were tested in a creep shear test. The

variables considered included emulsion tack-coat rate, fabric type (i.e., geotextiles,

geogrids, geonets, geomembranes, clay liners, geopipes, geofoam, and geomcosites),

structure (i.e., woven and non-woven), and fabric orientation. The first two were found

to have considerable effect while the latter had virtually no effect.

Additionally, Woodside et al. (11) found that SAMIs can reduce the likelihood of

damage and the need for large reconstruction work. The researchers found that the

pavement life may be extended by 50% and that of an asphalt overlay by up to 200%.
 10

2.1.2 California Department of Transportation

The California Department of Transportation (Caltrans) evaluated the increase of

HMA thickness overlays to minimize reflective cracking. Research conducted by

Predoehl (1989) in California showed that 4.8 inches of overlay is required to reduce

reflective cracking for 10 years (12).

In 1992, Caltrans decided to evaluate the effectiveness of glassgrid in retarding

reflective cracking caused by thermal fatigue (13). From Caltrans’ past experience,

pavement reinforcing fabrics (PRF) have been minimally effective in areas of high

thermal stresses and moisture-related expansion and contraction. Caltrans applied the

glassgrid on a section of Highway 89 located in northeastern California through the

Sierra Nevada Mountains. The existing pavement consisted of 7 inches of asphalt

concrete (AC) over a granular base of volcanic origin. The pavement carries a high

volume of logging trucks (average annual daily truck traffic of 512) due to the timber

harvesting in the area. The existing asphalt pavement was experiencing bleeding, rutting,

alligator and transverse cracks. The maintenance strategy for this area consisted of a 1.5

inch AC overlay on top of the glassgrid mesh to be compared to a control section of a 1.5

inch AC overlay where no glassgrid was applied. Cracks wider than 1/4 inch were filled

with an asphalt mix prior to the placement of the glassgrid mesh.

The test site was visited in 1994 and 1995. In 1994, one partial transverse crack

was observed on the outer edge of the pavement over a culvert at the north end of the

glassgrid section, but no other pavement distress was present. There were no changes in

1995 from the 1994 evaluation. During the 1994 evaluation of the control section, three
 11

partial transverse cracks and approximately 15 feet of longitudinal cracks were observed.

In 1995, alligator cracking with associated pumping appeared in the control section.

Therefore, the glassgrid has reinforced the AC overlay and retarded reflective cracking in

an area of high tensile stress. Based on the findings of this study, Caltrans now considers

glassgrid application for the overlay strategy (13).

2.1.3 Colorado Department of Transportation

In 1982, FHWA reported a study on the performance of fabrics in asphalt

overlays from Colorado test sections (14). The report revealed that the control section

(1.5 inch AC overlay) developed 10% cracking in 1.25 years and 50% within 3.25 years,

while the fabric section (1.5 inch AC overlay and fabric) developed less than 2%

reflection cracking in 1.25 years and 30% within 3.25 years.

In 2004, an experiment to assess the economics of various types of reflective

cracking reduction techniques was conducted on I-25 near Colorado Springs, Colorado

(15). The project consisted of rehabilitating approximately four miles of I-25 by milling

the surface of the old pavement and replacing it with a new HMA overlay. The

responsibility for repairing the new surface course if reflection of the old cracks occurs

during the warranty period entailed the contractor to built eighteen test sections within

the project to evaluate the most cost effective solution to reduce reflective cracking.

Two experimental sections were constructed. Section 1 consisted of removing 1

inch of the old asphalt pavement surface in the passing lane and shoulders and 2.5 inches

in the driving lane by cold milling and replacing with 4 inches of new HMA in the

passing lane and shoulders and 5.5 inches in the driving lane. Section 2 was equivalent to
 12

Section 1 except the entire pavement width was milled 1 inch and replaced with 4 inches

of new HMA.

Eight different treatments and a control section were installed in each of Sections

1 and 2 to measure the effect on reflection cracking and consisted of:

• Treatment A: 90 lb Geotextile (Petromat).

• Treatment B: 120 lb Geotextile (Petromat).
• Treatment C: reinforced fabric (Petrotac).
• Treatment D: fiberglass tape (ProGuard), each 18 inches wide.
• Treatment E: crack sealer type ASTM D 3405, applied to original cracks after
routing to a ½-inch width and depth.
• Treatment F: crack sealer type ASTM D 3405, applied to original cracks without
routing.
• Treatment G: crack sealer polymer modified type ASTM D3405, applied to
original cracks after routing to a ½-inch width and depth.
• Treatment H: crack sealer polymer modified type ASTM D3405, applied to
original cracks without routing.
• Control treatment: consisted of applying the HMA according to the description of
Sections 1 and 2, directly over the milled surface.

The rate of crack reflections was measured each year since construction and

documented. After 5 years of pavement service the following conclusions were made

(15).

• Treatments A, B, C, F, G and H had a lower percentage of reflection cracks than

the control section in Test section 1.
• Treatment B, C, D, E, and H had a lower percentage of reflection cracks than the
control section in test section 2.
• No treatments performed better than the control in the passing lane.
• The control sections provided the least total cost of construction and repair over
the five years of analysis period.
• Treatments B, C, and H have performed better than the control with respect to
reduction in reflection cracking in both test sections.
 13

2.1.4 Georgia Department of Transportation

In 1984 and 1985, the evaluation of HMA overlays over deteriorated PCC

pavements in Georgia indicated that 20% of cracking area occurred in six years for a 6-

inch HMA overlay compared to two years for a 4-inch HMA overlay (16, 17). Reflective

cracking appeared almost immediately after construction for a 2-inch overlay.

In 1991, the Georgia Department of Transportation (GDOT) began to evaluate the

production and placement of crumb rubber HMA to reduce rutting and reflective cracking

in HMA overlays (18). The crumb rubber mix (CRM) used by GDOT was produced by

adding ground tire rubber to HMA using the wet process. A test section of CRM was

placed on I-75 in Henry County, just south of Atlanta, consisting of a surface mix

containing 6% crumb rubber by weight of asphalt cement (AC). The test section was

evaluated from 1991 to 1995. The test section indicated that the CRM became very

brittle over time, as indicated by a large increase in viscosity and decrease in penetration,

and by a large amount of transverse reflective cracking. Compared with the control mix,

the CRM did not reduce rutting and was more than twice as expensive to place (18).

2.1.5 Illinois Department of Transportation

In 1981, Mascunana summarized the application of four treatments to reduce

reflective cracking on 10 state and local roads resurfacing projects in Illinois (19).

Treatments included two commercially available engineering fabrics (Petromat by

Phillips Fibers Corporation and Mirafi 140 by Celanese Fibers Marketing Company), a

fabricated interlayer membrane (Heavy Duty Bituthene by W. R. Grace & Company),

and an asphalt-rubber membrane interlayer. The findings of this study have indicated

that the treatment methods were not effective in preventing the development of transverse
 14

reflective cracking on overlays with cement treated bases. However, they controlled

longitudinal reflective cracking. In addition, they were generally effective in reducing or

retarding both transverse and longitudinal reflective cracking on overlays with

bituminous base courses (19, 9).

In 1999, Buttlar et al. evaluated the cost-effectiveness of the Illinois Department

of Transportation (IDOT) reflective crack control system consisting of a nonwoven

polypropylene paving fabric (20). The study was limited to projects constructed

originally as rigid pavements and subsequently rehabilitated with one or more bituminous

overlays. The performance of 52 projects across Illinois was assessed through crack

mapping and from distress and serviceability data in IDOT’s condition rating survey

database. The performance monitoring indicated an increase in life spans by 1.1 and 3.6

years for paving fabric strip (over existing cracks) and area applications (over the entire

pavement), respectively. Life-cycle analysis indicated that the paving fabric reflective

control system was marginally cost-effective in Illinois. Cost saving ranged from a

break-even level for small projects (less than 1.6 lane-mile) to about 6.2% for large

projects (6 lane-miles). Greater savings were realized where large quantities of paving

fabric were used (20, 9).

In 1993, the University of Illinois completed a research project for IDOT on a

prototype Interlayer Stress Absorbing Composite (ISAC). A prototype test section was

placed on IL 38 near Rochelle, IL in 1993. Other ISAC test sections were placed on five

asphalt concrete overlay (ACOL) projects between 1997 and 2000. Some of these ACOL

sections contain other reflective cracking control methods, such as Sand Anti-Fracture

(SAF) layer, strip, and area-wide reflective cracking control fabric (21).
 15

The ISAC consists of a three-layer system. The top layer is a high strength woven

geotextile to resist stresses caused by underlying pavement movements. This layer has

the ability to, due to its weaving, expand like a chain link fence. This movement

dissipates the stress caused by the movement of the underlying pavement. Typically, this

geotextile has a tensile strength greater than 4,000 lb. /in. (700 N/mm) at 5% strain

(ASTM D 4595). High strength is needed to ensure that when the geotextile is expanded

to its full extent its strength is greater than the strength of the bituminous concrete

overlay. The bottom layer is a low strength nonwoven geotextile (meeting AASHTO M-

288-92). The middle layer is a modified rubberized asphalt layer to absorb the strain

energy and bond the two geotextiles together. The system bridges across the joint or

crack and dissipates stresses resulting from opening or closing movements. ISAC is

bonded to the existing pavement using a tack coat and then the overlay is placed.

The researchers concluded that the formation of reflective cracks and the

subsequent deterioration of these cracks were delayed at ISAC treated joints and cracks.

This statement was true for all five test sites. This delay ranged from over one year to

close to three years when compared to the untreated and other crack control methods. Of

special note, the ISAC areas consistently outperformed the System B products, PavePrep

and Roadtac. When compared with SAF, the ISAC delayed reflective cracks by about

two years. The two sections performed the same after the cracks were routed and sealed

prior to overlay. The cost analysis indicated that the higher the total cost of the project

the higher the number of cracks and joints that could be treated with ISAC. The present

cost of the ISAC strips being $10 to $14 per foot limits the conditions under which it
 16

would be cost effective to use. If asphalt costs are high or the cost of ISAC were to

decline more projects could benefit from using ISAC (21).

2.1.6 Indiana Department of Transportation

The Indiana Department of Transportation evaluated cracking and seating before

overlay and fiber reinforcement of the overlay mixture as potential methods for reducing

pavement cracking on asphalt overlays over concrete pavement (22, 9). The evaluated

project on I-74 in Indiana was constructed in 1984 and 1985 and divided into several

experimental and control sections. As a performance comparison with the experimental

sections, the control sections were overlaid by the conventional method. The study

results based on the 7-year pavement performance data indicated that the cracking and

seating technique was successful on this project: it delayed most of the transverse cracks

for 5 years. The majority of the transverse cracks on the cracked and seated sections

were thermal cracks, which were narrower and less severe than the reflective cracks on

the control section. It was also found that the type of hammers used for cracking the

concrete slabs had strong effects on pavement performance. The use of fibers in the

overlay mixture further reduced transverse cracks on cracked and seated sections but did

not improve the cracking resistance of the control sections. Fibers improved rutting

resistance on both control and cracked and seated sections. However, the sections with

fibers exhibited rapid decreases in pavement strength and rideability. Thicker overlays

increased the construction costs significantly but did not reduce the transverse crack

intensities. According to the pavement performance and the cost analyses, it was

recommended that the thickness of asphalt overlay be determined only by the pavement

strength requirement and not be increased as a means of cracking control (22, 9).
 17

2.1.7 Louisiana Department of Transportation

In 1991, a project was initiated to evaluate the pavement performance of an

alternative pavement design, referred to as stone interlayer or inverted pavement (23).

The test section consisted of a 3.5 inch HMA layer on top of a 4 inch stone base layer on

top of a 6 inch of in-place cement stabilized base. The control section consisted of a 3.5

inch HMA layer on top of 8.5 inches of cement-stabilized base layer on top of prepared

subgrade (standard design). The objective of the study was to evaluate reflective

cracking reduction through the asphaltic pavement as well as overall pavement

performance. The project took place in Acadia Parish, Route LA-97, near Jennings,

Louisiana. LA-97 is considered a low-volume rural highway, with average daily traffic of

2,000 vehicles. The pavement was monitored for 7 years after construction. During the

evaluation period, annual crack survey, ride, and deflection measurements were collected.

Additionally, as part of the Louisiana Transportation Research Center accelerated

pavement testing research program, the same pavement designs were tested to failure

using the Accelerated Loading Facility. Results of that investigation showed that the

stone interlayer had significantly reduced the amount of reflective cracking. The ride

characteristics and structural capacity of both sections were similar during the evaluation

period. Accelerated testing results also verified the superior performance of the stone

interlayer pavement system. The cost analysis showed initial construction costs for the

stone interlayer system may be as high as 20% more than standard design. However, the

life of the stone interlayer pavement system is increased to almost five times that of the

standard soil cement pavement as tested under accelerated loading (23).

 18

In 1994, the Louisiana Department of Transportation (LODOT) evaluated the

effect of GlasGrid in retarding reflective cracking on U.S. Highway 190, a secondary

arterial road located between Covington and Baton Rouge, Louisiana (24). A large

number of transverse and longitudinal cracks were present in the existing full-depth

asphalt and composite pavements. LODOT placed a 1.5-inch thick Type 8 binder course

followed by a 1.5-inch thick Type 8 wearing course. GlasGrid 8501 was placed between

the two courses, with one area left un-reinforced to serve as a control section. As a result

of the monitoring undertaken after construction, the GlasGrid System showed benefits in

retarding reflective cracking.

LODOT currently specifies an Asphaltic Surface Treatment (AST) as an

interlayer to delay or reduce the occurrence of reflective cracking (25). AST, sometimes

referred to as “chip seal”, consists of furnishing properly distributed asphalt material (a

specified emulsion applied “cold” or modified asphalt material applied “hot”) followed

by a uniform application of aggregate.

2.1.8 Michigan Department of Transportation

In 1989, the Michigan DOT evaluated the performance of the Petrotac Style 4591

which is a unique peel and sticks paving fabric/rubberized asphalt composite membrane

(26). Petrotac acts as a moisture barrier as well as a stress absorbing membrane interlayer.

It is used in pavement maintenance to treat joints and cracks before overlay placement.

Petrotac, used as a stress absorbing membrane interlayer, absorbs stresses from

underlying cracks or joints helping to insulate the overlay from reflective cracking.

Monitored field performance on high volume roadways demonstrated that pavements

with the Petrotac interlayer show considerably less recurrence of reflective cracking.
 19

2.1.9 Mississippi Department of Transportation

In 1990, The Mississippi Department of Transportation (MDOT) conducted a

study to evaluate paving fabrics and the asphalt rubber interlayer systems, and to compare

them with the use of a single bituminous surface treatment for the purpose of controlling

reflective cracking (27). The paving fabric included Amopave (by Amopave Fabrics

Company), Fibretex (Crown Zellerbach Corp.), Mirafi (by Mirafi Inc.), and Petromat (by

Phillips Fibers Corp.). The findings of this study indicated that the asphalt rubber

interlayer, in combination with a thin overlay (about 1.5 inches), reduced and/or delayed

reflective cracking over a period of about five years. The study recommended the use of

asphalt rubber interlayer to be seriously considered with all thin overlays (9).

In 2005, Amini (9) conducted a survey of the current paving fabric applications

(polypropylene, staple fiber, needlepunched, and nonwoven geotextiles) in the state of

Mississippi to determine the various practices and performances of the paving fabric

systems to reduce reflective cracking. Researcher found that the field performance of

overlays using fabric interlayers has generally been successful, although there have been

some cases where the paving fabric systems provided little or no improvements. In

particular, paving fabrics may not reduce cracking significantly with thin overlays. In

one case, at a 25% level of reflection cracking, the inclusion of fabric increased the

performances of the 1.5 inch overlays by a factor of 3 (4.5 years versus 1.5 years). In

general, fabric interlayers have been most effective when used for load related fatigue

distress and have not performed well when used to delay or retard thermal cracking. The

study also showed that paving fabrics have been beneficial in the reduction of water

entering the pavement; however, documentation showing the derived benefits needs to be
 20

developed through carefully planned field studies. Also, proper construction procedures

are critical for optimum performance of paving fabrics. Additionally, the researcher

found, in general, that the application of paving fabrics was most effective in warm

climates such as southern states. The inclusion of paving fabrics should not be expected

to stop the reflection of all thermal cracks. But, the waterproofing effects will help to

minimize the freeze/thaw damage and to improve the overall pavement service life (9).

2.1.10 Pennsylvania Department of Transportation

In 1989, Maurer and Malashekie (28, 9) reported the results of early performance

and evaluation of six treatments to control reflective cracking. All treatments were

compared with each other with a control section not receiving treatment. The treatments

included Reepave T-376 paving fabric (by Dupont), Amopave paving fabric (by Amoco),

Trevira 1115 paving fabric (by Hoechst Fibers Industries), Mirafi paving fabric (by

Mirafi), Fiber Pave reinforced asphalt membrane interlayer (by Hercules Inc.), and

Bonifiber reinforced asphalt concrete (by Kapejo Inc.). Treatment comparisons were

made based on the construction and maintenance costs, ease of placement, and the ability

to prevent or retard reflective cracking. Performance data were evaluated for surveys

conducted at 8, 26, and 44 months after construction. All treatments retarded cracks over

the evaluation period, although the amount and rate of reduction varied. The Trevira

paving fabric and the Bonifiber reinforced asphalt concrete had the highest crack

reduction ratios after the 44-month evaluation. On the basis of all factors considered in

the evaluation (i.e., cost, ease of construction, and performance relative to distress

treated) the Bonifiber reinforced asphalt concrete provided superior performance relative

to the other treatments. However, on the basis of the extent of cracking evident after the
 21

44-month survey, and considering current and proposed crack sealing costs in addition to

the documented construction costs, none of the treatments used on this project were

found to be cost-effective or recommended for use (28, 9).

The performance of 3 paving fabric/geogrid products were also evaluated by the

Pennsylvania Department of Transportation (29, 9). These products were evaluated in

three test sections with two control sections at two separate locations. The paving fabric

types included Petromat, Bit-U-Tex (combination of paving fabric and geogrid), and

Glassgrid. This study showed that none of the three paving fabric/geogrid types were

found to be effective in preventing or retarding reflective cracking.

In 2004, Morian et al. (30) evaluated the performance of cold in-place recycling

(CIR) of existing HMA materials on projects in northwestern Pennsylvania that has been

in-service for more than 20 years. A total of 44 pavement sections were available. A

subset of these projects was evaluated to determine performance characteristics and cost-

effectiveness of the treatment and the material. The treatment is used typically on

rehabilitation projects of roadways with 8,000 average daily traffic (ADT) or less but has

been used on projects with up to 13,000 ADT. A number of significant conclusions were

provided by the researchers.

• CIR provided an effective mean of extending the life of pavement rehabilitation

projects for highways with up to 13,000 ADT and 200,000 annual equivalent
single axle loads. Projects have been documented to provide service life of up to
160% of the ten-year design life typically provided by conventional mill and
resurface projects in the same geographical area.
• CIR provided resistance against reflective cracking between two and three times
that exhibited by conventionally resurfaced control sections. This form of
recycling has been shown to be very cost effective as a rehabilitation strategy.
The cost of CIR is one to two-thirds the cost of conventional HMA material while
providing superior performance.
 22

• When good construction process control is implemented, cold in-place recycled

material develops stiffness values comparable to those expected for conventional
HMA materials. This level of stiffness was identified in several 10-year old
projects and in one 15-year old project.
• CIR material appears to typically be a stress sensitive material, providing
increased stiffness in response to increased load. This “flexibility” is likely an
important factor in the observed good performance of these projects, and the
delay in the development of reflective cracking.

2.1.11 Texas Department of Transportation

In 2002, the Texas Transportation Institute (TTI) performed a research project in

cooperation with the Texas Department of Transportation (TxDOT) and FHWA to

investigate and develop information that will aid in the evaluation of the relative

effectiveness of commercially available geosynthetic materials in reducing the severity or

delaying the appearance of thermally induced reflective cracking in HMA overlays (31).

The researchers conducted an extensive review of literature on studies of geosynthetics

for reducing reflective cracking in HMA overlays, along with a formal survey of

engineers in each TxDOT district and an informal survey of certain other knowledgeable

individuals nationwide. Additionally, mechanistic laboratory testing and analyses were

conducted using the TTI Overlay Tester to examine the relative resistance of the three

major categories of geosynthetics (fabrics, grids, and composites) in resisting reflection

cracking in HMA mixtures. The overlay tester evaluates the laboratory resistance to

thermal cracking of a 3×6×20-inch HMA beam sample. Researchers then used fracture

mechanics to evaluate the laboratory results and develop a new evaluation methodology

termed the “reinforcing factor”. The “reinforcing factor” which is also termed “reduction

factor”, is developed to account for the differences between the laboratory measured test

value and the desired performance value, in order to convert the lab result into a site-
 23

specific allowable value. Finally, field tests were established in three locations in Texas

(Amarillo, Waco, and Pharr Districts) with widely differing climates. These planned

field tests will be constructed and all three will be evaluated for several years particularly

regarding reflective cracking.

Based on the review of literature and observations made during the fabrication

and testing of the HMA specimens the following conclusions were made (31).

• Performance of geosynthetics in addressing reflection cracking in HMA overlays

has ranged from highly successful to disastrous failure. Based on literature
review, the geosynthetics showed generally marginal cost effectiveness in
reducing reflection cracking. Many of the publications reviewed (particularly
those related to fabrics) were based on the cost of geosynthetics more than 10
years ago. In recent years, the in-place cost of geosynthetics has become more
favorable to paving agencies.
• Generally, the geosynthetics tested in the laboratory consistently increased the
number of cycles to failure in the TTI overlay tester.
• Quality assurance tests performed on selected laboratory test beams were
compared to the TxDOT job-mix formula (JMF). Extraction revealed asphalt
contents between 4.1 and 4.6 percent as compared to the optimum asphalt content
of 5.0 percent. Insufficient asphalt cement produces inadequate film thickness
around aggregate particles and decreases the tensile properties of the mixture.
The mixture for this investigation was sampled at a production plant and stored in
metal containers. Re-heating of the mixture for beam fabrication was necessary
which, of course, further oxidized these thin films. These findings are considered
the major causes for the relatively low number of cycles to failure recorded during
this investigation. The remainder of the quality assurance tests were within
acceptable ranges of the JMF.
• Control beams were fabricated with and without an asphalt tack coat (0.05
gal/yd2) between the overlay and leveling course. The comparison of the number
of load cycles to failure for these specimens indicated that the thin tack coat
increased the number of load cycles by 131 percent (from an average of 2.6 load
cycles to 6.0). Therefore, in typical overlay construction, researchers concluded
that the simple addition of a thin asphalt cement tack coat will increase the life of
the overlay.
• Limited experimentation indicated that the use of emulsified asphalt as a tack coat
for geosynthetics produced a plane of weak shear, which could promote slippage
during overlay construction and service.
 24

As a result of this study, researchers prepared comprehensive guidelines for using

geosynthetics with HMA overlays to reduce reflection cracking and made the following

recommendations (31):

• Emulsified asphalt should not normally be used as tack for geosynthetics installed
to address reflection cracking in HMA overlays. Proper construction methods
should be employed and care should be taken when emulsified asphalts are used
as a tack coat. Sufficient time should be allotted for breaking and curing of the
emulsion.
• When placing a self-adhesive grid to address reflective cracking in an HMA
overlay, a tack coat should be applied on top of the grid (i.e., after grid
application). The appropriate quantity of tack is that normally used without a
grid. Type of tack should be hot applied asphalt cement (not emulsion) of the
same grade as that determined for the HMA overlay. Placing any thin overlay
without a tack coat could invite delamination from the underlying pavement.
• When ordering geosynthetics, the contractor should specify the desired roll width
and length to minimize construction joints and maximize efficiency. The
contractor should also consider the maximum roll weight that his application
equipment can handle.
• During laboratory testing, cracking patterns occurred in an irregular fashion, due
to breaking of cohesive bonds within the HMA, and crack lengths at multiple
locations were often difficult to record. Video taping devices should be installed
near the specimen to record cracking. This information could be used to monitor
and review cracking patterns and could be stored as a permanent record for future
review and/or analysis.
• Volumetric properties of the unreinforced HMA mixture should be determined.
• Research needs to be performed to determine the ability of asphalt-impregnated
fabrics or composites to reduce intrusion of surface water into a pavement after
reflection cracks appear and quantify the benefits, if any, toward preserving
pavement structural integrity and smoothness.

In 2004, Darling et al. (32) conducted a study describing the performance of

geogrids used in two dissimilar North America climatic zones: Zone I – wet, no freeze;

Zone VI – Dry, hard freeze, spring thaw. The two study sites in Zone I are: US Federal

Highway 190 located in Hammond, Louisiana and US Federal Highway 96 located in

Lumberton, Texas. The study site in Zone VI is Highway 513:02 located in the

Municipal District of Taber, Alberta, Canada. On all three projects, varying design
 25

approaches and remedial repairs were carried out in conjunction with utilizing two

different strengths reinforcing geogrids. Using visual surveys and quantitative analyses,

the performances of the sites were evaluated after 2.5 and 6 years in service. The study

showed that the fiberglass reinforcements, which were used on these three sites, can

extend the life of the overlay by 2 to 3 times that of the same non-reinforced section of

roadway. On all three sites, the fiberglass reinforced pavement has out performed the

non-reinforced control sections in terms of retarding reflective cracking. Fiberglass

reinforcements, such as 8501 & 8502, are extremely strong and experience low

elongations at ultimate strength and are able to reduce the rate of crack reflections

significantly to that of non-reinforced overlays. Ultimately, fewer cracks can lead to

lower maintenance costs and extend the projected design life (32).

In 2006, Chen at al. (33) published a study on the performance of several different

rehabilitation strategies used in the past 10 years in Texas. Reflective cracking of HMA

overlays on jointed concrete pavements has been a persistent problem in Texas. The

long-term field performances of the various treatments are summarized as follows.

• The crack retarding grid (plastic geosynthetics) did not perform well in retarding
reflective cracks. Only one project realized a benefit of the crack retarding grid,
where the reflective cracking was delayed by about 1 year. The small openings in
the used crack retarding grid and the lack of an effective bond might have caused
the separation of the layer.
• A proprietary crack-retarding asphalt material (Strata) performed well over two
years of monitoring. Strata is an asphalt-rich, polymer modified binder and fine
aggregate hot mix. In one experimental project, 100% of the cracks reflected
through conventional overlay material in the first year. Both Petromat fabric
underseal and Strata have been performing satisfactorily to retard reflective
cracking. However, there is a large cost difference between these two treatments
(cost of strata 10 to 20 times higher than the Petromat fabric underseal).
• Seven experimental treatments on a major section of US 59 were evaluated. The
worst-performing section on the US 59 project was the break and seat method,
which failed due to weak subgrade support. The weak subgrade is unable to
 26

support the cracked concrete, which leads to a rocking action under traffic loads.
It was recommended that for future projects, the break and seat method should not
be applied on subgrade with a dynamic cone penetrometer penetration rate
exceeding 25 mm per blow.
• The flexible base overlay with thin asphalt surfacing has performed well. The
flexible base overlay involves placing high-quality crushed limestone directly
over the JCP followed by an underseal and thin asphalt overlay. Flexible base
overlays were able to absorb the joint movement and eliminate the reflective
cracking. The flexible base should be of top quality material with very low
moisture susceptibility. In Texas, this involves specifying a Texas triaxial Class 1
material with classification of “good” in the suction/dielectric test. It is also
critical to provide an effective seal for the flexible base. A chip seal followed by
a thin 75-mm hot mix layer is recommended.
• The Arkansas open graded large stone AC interlayer mix has performed well.
The Arkansas mix is the old TxDOT Type G mix.

2.1.12 Virginia Department of Transportation

Like other state highway agencies, the Virginia Department of Transportation

(VDOT) in recent years has more frequently used asphalt overlays to rehabilitate severely

distressed Portland cement concrete pavements (PCCP). Unfortunately, reflective cracks

that form above joints and cracks in the underlying PCCP drastically reduce the useful

lives of the HMA overlays. During the past 15 years, the fractured slab approach has

gained increased acceptance as a means of retarding the formation of reflective cracks.

In 2002, the Virginia Transportation Research Council conducted a research study

to evaluate the concrete slab fracturing technique as a mean of arresting or retarding

reflective cracking through asphalt overlays placed over severely distressed PCC

pavements in Virginia (34). The study involved monitoring the performance of five

pavement rehabilitation projects over a period of up to 8 years. Two of the projects were

jointed plain concrete pavements, and the other three were jointed reinforced concrete

pavements. The test sections were fractured with a guillotine drop hammer and then
 27

seated with a 50-ton pneumatic tire roller. For comparative purposes, control sections,

which were not fractured prior to placement of the asphalt overlay, were constructed just

beyond the bounds of three of the fractured test sections.

Detailed visual condition surveys were conducted annually on all sites. For each

survey, the number of occurrences of reflective cracks that formed in the fractured

sections was directly compared to the number of cracks observed in the control sections

to quantify the tendency of slab fracturing to retard or arrest the formation of reflective

cracks. The results of this study show that fracturing and seating distressed concrete

pavements appear to be an effective mean of retarding the formation of reflective

cracking through asphalt overlays on jointed plain concrete pavements. In the case of

reinforced concrete pavements, however, the fracturing technique was somewhat less

successful in that the formation of reflective cracks appeared to be delayed for only about

3 years. Beyond that point in time, the fractured reinforced sections exhibited

approximately the same amount of reflective cracking as the control section.

The researcher concluded that any observed benefit in terms of extended

pavement service life or enhanced ride quality resulting from even a slight delay in

reflective crack propagation would likely offset the rather nominal cost of the fracturing

and seating operation itself (34).

2.1.13 Wisconsin Department of Transportation

In 1989 the Wisconsin DOT initiated a project to evaluate the performance of

GlasGrid and a glass fiber mesh pavement reinforcement geotextile in prohibiting or

controlling the incidence of reflective cracking (35). In the summer of 1990, test sections

examining two different types of the product, a single strand grid and a double strand
 28

grid, were constructed on STH 57 in Sheboygan County. The glassgrid were placed

between the 1.5 inch bottom and top lifts of an HMA overlay over a jointed plain PCC

pavement that was constructed in 1957. The GlasGrid was only installed across the

transverse joints and cracks of the underlying PCC pavement.

Reflective cracks appeared in the test sections within six months of construction.

By the end of the fourth year, the percentage of reflective cracking that occurred in the

double strand test sections surpassed that of the control sections, which had no GlasGrid.

The reflective cracks in the GlasGrid test sections were wider than the reflective cracks in

the control section. Annual crack surveys were completed at the test site during the first

five years and after ten years. A final field review was completed in April of 2002. The

results of these surveys showed that both the single strand grid and double strand grid

variations of the GlasGrid product performed unsatisfactorily and were unable to prohibit

or control reflective cracking effectively. Based on the results of this study, it was

recommended that WisDOT not use GlasGrid fiberglass mesh reinforcement as a method

of reducing reflective cracking of an asphaltic concrete overlay or for extending the life

of an asphaltic concrete overlay placed on a PCC pavement (35).

In 2004, an experimental project was conducted to evaluate the Interlayer Stress-

Absorbing Composite (ISAC) (36). According to the manufacturer, the ISAC should

extend the life of a HMA overlay placed over a PCC pavement by retarding reflective

cracking. The scope of this project included milling 2 inches from the PCC pavement

and replacing it with a 2-inch HMA overlay. The ISAC material was placed over the

cleaned and sealed transverse joints prior to placement of the HMA overlay.

Immediately after construction a bump was visible in the HMA overlay along the
 29

transverse edge of the ISAC fabric. Due to the transverse bumps in the overlay, the

overlay and the ISAC fabric were removed and the test section was repaved without the

ISAC. The manufacturer explained that the likely cause of this problem was due to the

old age of the ISAC material used. The age of the ISAC was almost three years, which

caused the ISAC to wrinkle. According to the manufacturer, they usually don’t sell any

of the ISAC products beyond thirty days old. Due to the removal of the ISAC system, an

in-situ performance evaluation was not possible and this study was canceled (36).

The WisDOT and the City of Milwaukee tried a fine-aggregate, asphaltrich,

polymer-modified asphalt mix interlayer to absorb joint movements, delay reflective

cracking, and protect the existing pavement (37). Four Wisconsin projects were

constructed using this technique. In the first project, constructed in 1996, the interlayer

showed no impact on delaying reflection cracking within the first 3 years. Later projects,

however, included specifications for performance related design tests for flexural beam

fatigue and Hveem stability and were overlaid with improved mixtures to complement

the flexible interlayer. The later projects showed an average 42% improvement in the

time to the appearance of surface cracks compared with the control sections.

Furthermore, cores taken from the projects showed that even when the overlay cracked,

some of the interlayer samples did not, even under severe conditions, thus further

protecting the underlying pavement structure. Other major factors contributing to the

cracking delay included the type of concrete pavement, concrete patches, and climate

(37).
 30

2.1.14 International Practice

Belgium Experience

A study was conducted by the Belgian Road Research Center in Brussels, to

evaluate the performance of different interface systems used on various sites in Belgium

for preventing reflective cracking. Vanelstraete and Visscher described the results of the

long term evaluation of two experimental roads and five individual projects of HMA

overlays over PCC pavements where different interface systems were used for prevention

of reflective cracking (38). The following treatments were used on the experimental

roads:

• SAMI: 1.5 kg/m2 elastomeric binder, 9 kg/m2 7/10 pre-coated chippings, and
dense graded HMA overlay.
• Non-woven fabric: 1.2 kg/m2 elastomeric binder, Non-woven fabric, and dense
graded HMA overlay.
• Geogrid: 0.25 kg/m2 tack coat, geogrid, 1.2 kg/m2 elastomeric binder, 6 kg/m2
7/10 pre-coated chippings, and dense graded HMA overlay.
• Steel nettings: 0.15 kg/m2 tack coat, steel netting, 16 kg/m2 slurry seal with
elastomeric binder, 0.2 kg/m2 tack coat, and dense graded HMA overlay.
• No interface: 0.20 kg/m2 tack coat and dense graded HMA overlay.

The findings from the experimental roads are as follows.

• The long term evaluation of an experimental road for 8 years showed that crack
and seat and steel reinforcing nettings are both effective against reflective
cracking. Steel reinforcing nettings treatment consists of placing 0.15 kg/m2 tack
coat, steel netting, 16 kg/m2 slurry seal with elastomeric binder, 0.2 kg/m2 tack
coat, and overlaid by HMA.
• The long term evaluation of an experimental road for 5 years showed that less
reflective cracks were developed on sections with interface systems (SAMI, non-
woven, grid, steel reinforcing nettings) when compared to the control section
where no interface system was applied.

A follow up of several individual projects with the following interface systems

were constructed.
 31

• Nailed steel reinforcement netting, 3 inches dense graded HMA, and asphalt
surfacing.
• Nailed steel reinforcement netting, 3 inches dense graded HMA, and 1.5 inches
porous mix.
• Steel reinforcement netting and slurry, and 1.5 inches porous mix.
• Steel reinforcement netting and slurry, and 3.5 inches dense graded HMA mix.
• Non-woven polyester, 1.5 kg/m2 modified binder, 1.5 inches ultrathin HMA layer.

The field performances of the individual projects with the interface systems

confirmed the earlier observations and are summarized as follows.

• The overlay thickness remains one of the predominant factors affecting reflective
cracking, even with the use of an interface system.
• Crack and seat of the PCC pavement before placing the overlay system showed to
be highly efficient.
• The projects with steel reinforcing nettings performed very well even after more
than ten years of repair.

Canadian Experience

In 2003 Tighe et al. reported on the economic benefits of reducing and treating

reflection cracking of HMA overlays (39). In fact the Canadian Strategic Highway

Research Program’s (C-SHRP) Long Term Pavement Performance Study (LTPP) is

entirely directed to overlays. The overall goal of the C-LTPP project is to increase

pavement life and serviceability through the development of cost-effective rehabilitation

strategies, based on a systematic observation of in-service pavement performance.

Researchers found that measuring and treating cracking accordingly can yield significant

benefits. Proper and timely crack treatment (routing and sealing) can result in extending

life by 2 years and cost savings in the order of $7,000 per lane-km.

Finland Experience

Transverse thermal cracks are a difficult problem on the Finnish road network.

Very often, transverse cracks reflect through the HMA overlay in 2 years after
 32

construction. In July 2000 a field test was performed to identify the best method of

repairing cracks prior overlay (40). On the main road 1, altogether 49 transverse cracks

were repaired by eight different methods and the location of every crack was determined

carefully. Four of the methods were based on fiber grids (Glassgrid Geogrid 8511) and

the other four were more traditional methods. After the repair work a new SMA overlay

was constructed. The repair methods were as follows.

• Leveling the old surface, hot milling and asphalt surfacing.

• Milling the form of a box, 1.5 m wide grid, leveling the old surface, hot milling
and asphalt surfacing.
• Milling the form of a box, 0.75 m wide grid, leveling the old surface, hot milling
and asphalt surfacing.
• Sawing a crack, filling the crack with hot modified bitumen, leveling the old
surface, hot milling and asphalt surfacing.
• Sawing a crack, filling the crack with a band of modified bitumen, leveling the
old surface, hot milling and asphalt surfacing.
• Hot milling, apply 1.5 m wide grid, and asphalt surfacing.
• Hot milling, apply 0.75 m wide grid, and asphalt surfacing.
• Hot milling and asphalt surfacing.

As a result of the test area examination in the following spring 2001, not even one

of the cracks repaired with grids had occurred again. Traditional methods (crack filling

and HMA overlay or mill and HMA overlay) had not been as successful; some cracks

were already visible. After two winters, in spring 2002, the difference between grids and

traditional methods was even more significant. Then it seemed as if there is a difference

also whether the grid has been laid directly on heated remixed surface or on a "milled

box". The spring 2003 showed the final order of the tested repair methods. It was found

that the best promising method to prevent the reflection of transverse cracks is to lay a

grid on a "milled box" (40).

 33

2.2 DOTS STANDARD SPECIFICATIONS FOR HIGHWAY CONSTRUCTION

This section of the thesis summarizes the current standard specifications on the

reflective cracking mitigation techniques. The specifications from all fifty state DOTs

were reviewed and are summarized in Appendix B. Thirty two states out of fifty have

specified a reflective cracking control system in their current standard specifications.

The reflective cracking control systems varied between simple crack and joints sealing to

the application of geotextile and geosynthetics, geogrids, paving and reinforcing fabrics,

asphalt rubber stress absorbing membrane, and fiberized asphalt membrane.

 34

Chapter 3 – REVIEW AND ANALYSIS OF NDOT REFLECTIVE

CRACKING MITIGATIONS TECHNIQUES
This chapter describes in detail the NDOT’s experience applying various

techniques to delay the reflective cracking. Such techniques are cold in-place recycling,

reinforced fabrics, stress relief courses, mill and overlay for flexible pavements, and

crack and seat of PCCP and Rubblization of PCCP for rigid pavements. For this purpose,

forty-eight projects to mitigate reflective were selected from the NDOT’s Pavement

Management System (PMS) inventory. The performance of the various projects is

analyzed before and after the treatment applications.

3.1 REVIEW OF NEVADA DOT EXPERIENCE

In the past, NDOT has used several techniques to reduce the impact of reflective

cracking on HMA overlays. This second task of Phase I identified all field projects

where NDOT has implemented reflective cracking mitigation techniques. The design,

construction, and traffic details of these projects were collected along with their

corresponding long-term field performance. The information was analyzed to assess the

effectiveness of the various techniques in mitigating reflective cracking under Nevada’s

conditions.

3.1.1 Performance Related Data

Pavement performance data, collected over time, provides the basis for assessing

the actual performance of a pavement technology. Pavement roughness, rutting, and

cracking represent the major components of NDOT's pavement conditions survey

program.
 35

NDOT's philosophy on pavement performance can be summarized as: A "good"

pavement provides a comfort ride to its users, does not require extensive maintenance for

the repair of distresses, it is structurally adequate for the traffic loads, and provides

sufficient friction to avoid accidents.

The present serviceability index (PSI), surface cracking, and rut depth will be

used to assess the long term performance of the various reflective cracking techniques

that have been implemented by NDOT.

3.1.1.a Present Serviceability Index (PSI)

The PSI is based on the original AASHO Road Test Present Serviceability Rating

(PSR). The PSR relate the ride conditions of the road to the opinion of the user which

required a panel of observers to actually ride in an automobile over the pavement in

question. Since this type of rating is not practical for large-scale pavement networks, a

transition to a non-panel based system was needed.

To transition from a PSR serviceability measure (panel developed) to a PSI

serviceability measure (no panel required), a panel of raters from 1958 to 1960 rated

various roads in the states of Illinois, Minnesota, and Indiana for PSR. This information

was then correlated to various pavement measurements (such as slope variance or profile,

cracking, etc.) to develop PSI equations. Further, the raters were asked to provide an

opinion as to whether a specific pavement assessed for PSR was "acceptable" or

"unacceptable" as a primary highway. Thus, although PSI is based on the same 5-point

rating system as PSR it goes beyond a simple assessment of ride quality. About one-half

of the panel of raters found a PSR of 3.0 acceptable and a PSR of 2.5 unacceptable. Such
 36

information was useful in selecting a "terminal" (or failure) serviceability (PSI) design

input for empirical structural design equations.

The original PSI equation has been modified throughout the years by state

highway agencies in order to better describe the local conditions. Currently NDOT uses

the following PSI equation for flexible pavements:

PSI = 5e(-0.0041×IRI) – 1.38×RD2 – 0.03×(C+P)1/2 (3.1)

If PSI < 0 then PSI = 0.10

where, IRI = international roughness index (in/mile)

RD = rut depth (in)
C = cracking (ft2/1000ft2)
P = patching (ft2/1000ft2)

The international roughness index (IRI) is the first widely used profile index

where the analysis method is intended to work with different types of profilers. It is

defined as a property of the true profile, and therefore it can be measured with any valid

profilometer. The analysis equations were developed and tested to minimize the effects

of some profilometer measurement parameters such as sampling rate.

Prior to 1992, NDOT used the slope variance to measure road roughness, then the

ultrasonic profilometer was used until 2000 followed by the laser profilometer thereafter.

NDOT conduct roughness measurements on an annual basis for all the interstates,

many of the US routes and some of the state routes, and on a biannual basis for all others.

The profilometer measurement is continuous, but an average roughness value for every

mile is recorded unless otherwise specified.

 37

3.1.1.b Surface Cracking

For the purpose of this research, fatigue cracking, non-wheelpath longitudinal

cracking, transverse cracking, and block cracking monitored over the service life of the

pavement were utilized. The cracking classes, extent and severity are defined in the

NDOT Flexible Pavements Distress Identification Manual (41) and are summarized here.

• Fatigue Cracking: Fatigue cracking is caused by repeated traffic loading of the

pavement. These cracks initiate at the bottom of the HMA layer and slowly work
their way to the surface. Fatigue cracking usually starts as a longitudinal crack in
the wheelpath (Type A). Further weakening of the HMA and base layers coupled
with repeated traffic loading leads to the progression of the longitudinal crack and
the formation of interconnected cracks referred to as alligator cracking since they
resemble the shape of an alligator skin (Type B). An unstable base, inadequate
drainage, insufficient pavement thickness, degradation/stripping in the HMA
combined with traffic loadings will accelerate this type of distress. The extent of
type A fatigue cracking is measured as the total linear feet of this type of cracking
in the wheelpath of the pavement area being surveyed. The extent of type B
fatigue cracking is measured as the total area in square feet of this type of
cracking in the wheelpath of the pavement being surveyed (i.e., 10 by 100 foot
area at every milepost).

• Transverse Cracking: This type of cracking is primarily caused by the

contraction of the HMA layer due to temperature changes. Other causes include:
age hardening, reflection cracking from PCC pavement joints below, and
reflection cracking from transverse cracks below. The extent is measured as the
total linear feet of cracking throughout the pavement area being surveyed (i.e., 10
by 100 foot area at every milepost).

• Block Cracking: Block cracking starts as a combination of transverse and non-

wheelpath longitudinal cracking (Type A). It is caused by age hardening and
shrinkage of the HMA layer. Although traffic loading is not the primary cause of
this type of distress, continued loading on the brittle surface will accelerate this
distress and break the larger pieces into smaller pieces progressing to Type B, and
finally to Type C. The extent of type A is measured as the total linear feet of this
type of cracking throughout the pavement area being surveyed. The extent of
type B and C is measured as the total area in square feet of this type of cracking
throughout the pavement area being surveyed.
 38

3.1.1.c Rutting

Rutting is a load-related failure of the pavement. Ruts are described as

longitudinal surface depressions in the wheelpaths as a result of pavement densification

or wear. Pavement uplift may occur along the sides of the ruts. Anyone, or combination

of the following factors may cause rutting:

• Soft pavement due to poor quality HMA mix

• Insufficient pavement thickness
• Unstable HMA mix
• Insufficient compaction during construction
• Stripping of the HMA mix
• Pavement wear or loss due to abrasive action of traffic

3.2 PERFORMANCE OF SELECTED NDOT PROJECTS

This task documents Nevada’s experience with reflective cracking mitigation

techniques within the past 15 years. The following list represents a summary of the

various techniques that NDOT has evaluated.

• Cold in-place recycling (CIR)

• Reinforced fabrics (RF)
• Stress relief course (SRC)
• Mill and overlay (MOL)
• Crack and seat of PCCP (CS)
• Rubblization of PCCP (RPCC)

A number of projects were constructed under each category. A total of forty four

projects were reviewed in this research. The projects were located either in the northern

part or the southern part of the state of Nevada covering different environmental and

traffic conditions. Table 3.1 summarizes the projects contract number, projects location,

applied reflective cracking technique, date of construction (DOC), and the 2005 annual

average daily traffic (AADT). A detailed description of each of the projects along with
 39

their long term performance is provided in the following paragraphs. The projects DOC

varied between 1990 and 2005. No condition surveys of any of the sections were

conducted as part of or specifically for this study. All of the long-term field performance

characteristics were obtained from the NDOT pavement management system (PMS).

3.2.1 Cold In-Place Recycling (CIR) Projects

Cold in-place recycling is carried out using specialized recycling machines, the

heart of which is a milling drum equipped with a large number of hardened steel teeth.

The drum rotates upwards, milling the material in the existing road. As the milling

process is taking place, water from a water tanker that is pushed ahead of the recycler is

delivered through a flexible hose and is sprayed into the mixing chamber. The water,

which is metered accurately by the microprocessor controlled pumping system, is mixed

thoroughly with the milled material to bring the material up to its optimum compaction

moisture content (42).

In general, the CIR technology is used to build a strong flexible base course. The

CIR is believed to strengthen the existing pavement by treating many types and degrees

of distresses. Originally, the CIR projects were located on low traffic volume roads

ranging from 30 to 300 daily ESALs. However, some agencies, including NDOT,

currently apply this treatment to roads with traffic varying from 1000 to even 10000

AADT.

A total of twelve NDOT CIR projects constructed between 1990 and 2001 were

analyzed in this study. The following is a description of the various CIR projects.

Contract 2808a
 40

This contract was constructed in 1998 on US050 in White Pine county over 3.0

miles. The construction consisted of CIR the top 2.0” of the existing HMA layer and

overlaying it with 2.5” dense graded HMA and 0.75” OGFC.

Figure 3.1 shows the performance related data for contract 2808a. Prior to CIR

the pavement exhibited fair PSI (2.3) and moderate transverse cracks and Type A fatigue

cracking. After the CIR process, the pavement maintained a PSI value above 4.0 until

2005. Only minor transverse cracks showed up 4 years after construction.

Contract 2808b

This contract was constructed in 1997 on US050 in Eureka county over 9.35

miles. The construction consisted of CIR the top 2.0” of the existing HMA layer and

overlaying it with 2.5” dense graded HMA and 0.75” OGFC.

Figure 3.2 shows the performance related data for contract 2808b. Prior to CIR

the pavement exhibited a low PSI (1.2) and transverse and fatigue cracking. After the

CIR process, the pavement maintained a PSI value above 4.0 until 2005. Three years

after construction some transverse cracking start showing up.

Contract 2838

This contract was constructed in 1999 on SR396 in Pershing county over 6.57

miles. The construction consisted of CIR the top 2.0” of the existing HMA layer and

overlaying it with 2.5” dense graded HMA and 0.75” OGFC.

Figure 3.3 shows the performance related data for contract 2838. Prior to CIR the

pavement exhibited a low PSI (1.4), block cracking, and a rut depth of 0.25 inch. After

the CIR process, the pavement maintained a PSI value of 3.7 until 2005. Two years after

construction severe fatigue and block cracking showed up at the pavement surface.
 41

Contract 2935

This contract was constructed in 1999 on SR360 in Mineral county over 23.25

miles. The construction consisted of CIR the top 2.0” of the existing HMA layer and

overlaying it with 2.5” dense graded HMA and 0.75” OGFC.

Figure 3.4 shows the performance related data for contract 2935. Prior to CIR the

pavement exhibited a fair PSI (2.8), and minor transverse and block cracking. After the

CIR process, the pavement maintained a PSI value of 4.0 until 2005. No surface cracks

showed up after construction.

Contract 2819

This contract was constructed in 1998 on US095 in Nye county over 7.40 miles.

The construction consisted of CIR the top 3.0” of the existing HMA layer and overlaying

it with 3.0” dense graded HMA and 0.75” OGFC.

Figure 3.5 shows the performance related data for contract 2819. Prior to CIR the

pavement exhibited a moderate PSI (3.3), and minor transverse and fatigue cracking.

After the CIR process, the pavement maintained a PSI value of 4.0 until 2005. Two years

after construction moderate transverse cracks and minor Type A fatigue cracking showed

up at the pavement surface.

Contract 2873

This contract was constructed in 1999 on US095 in Nye county over 15.76 miles.

The construction consisted of CIR the top 3.0” of the existing HMA layer and overlaying

it with 3.0” dense graded HMA and 0.75” OGFC.

Figure 3.6 shows the performance related data for contract 2873. Prior to CIR the

pavement exhibited a fair PSI (2.7), minor transverse cracking, and moderate fatigue
 42

cracking. After the CIR process, the pavement maintained a PSI value of 4.0 until 2005.

No surface cracks showed up after construction until year of 2005 where minor

transverse cracks and Type A fatigue cracking start showing at the pavement surface.

Contract 2961

This contract was constructed in 1999 on US006 in Esmeralda county over 25.12

miles. The construction consisted of CIR the top 3.0” of the existing HMA layer and

overlaying it with 3.0” dense graded HMA and 0.75” OGFC.

Figure 3.7 shows the performance related data for contract 2961. Prior to CIR the

pavement exhibited a fair PSI (2.8), moderate transverse cracking, and severe Type B and

C block cracking. After the CIR process, the pavement maintained a PSI value of 4.0

until 2005. No surface cracks showed up after construction.

Contract 3013

This contract was constructed in 2003 on US095A in Lyon county over 19.63

miles. The construction consisted of CIR the top 3.0” of the existing HMA layer and

overlaying it with 3.0” dense graded HMA (AC-20P) and 0.75” OGFC.

Figure 3.8 shows the performance related data for contract 3013. Prior to CIR the

pavement exhibited a good PSI (3.5), and minor block cracking. Prior to 2003 the

pavement experienced some transverse cracks. After the CIR process, the pavement

maintained a PSI value of 4.3 until 2005. No surface cracks showed up after

construction. In 2005, very minor transverse cracks showed up at the pavement surface.

Contract 3025a

This contract was constructed in 2001 on US093 in White Pine county over 11.0

miles. The construction consisted of CIR the top 2.0” of the existing HMA layer and
 43

overlaying it with 2.0” dense graded HMA (AC-20P) followed by a chip seal surface

treatment.

Figure 3.9 shows the performance related data for contract 3025a. Prior to CIR

the pavement exhibited a fair PSI (2.5), and minor transverse and fatigue Type B

cracking. After the CIR process, the pavement maintained a PSI value of 3.9 until 2005.

The pavement started showing transverse cracks one year after construction.

Contract 3025b

This contract was constructed in 2001 on US093 in Lincoln county over 22.22

miles. The construction consisted of CIR the top 2.0” of the existing HMA layer and

overlaying it with 2.0” dense graded HMA (AC-20P) followed by a chip seal surface

treatment.

Figure 3.10 shows the performance related data for contract 3025b. Prior to CIR

the pavement exhibited a fair PSI (2.5), and some transverse cracking. After the CIR

process, the pavement maintained a PSI value of 3.7 until 2005. No surface cracks

showed up after construction.

Contract 3025c

This contract was constructed in 2001 on US093 in Lander county over 22.22

miles. The construction consisted of CIR the top 2.0” of the existing HMA layer and

overlaying it with 2.0” dense graded HMA (AC-20P) followed by a chip seal surface

treatment.

Figure 3.11 shows the performance related data for contract 3025c. Prior to CIR

the pavement exhibited a fair PSI (2.5), and some transverse cracking and block cracking.
 44

After the CIR process, the pavement maintained a PSI value of 3.8 until 2005. The

pavement started showing transverse cracks one year after construction.

Contract 2876

This contract was constructed in 2001 on SR208 in Lander county over 6.53

miles. The construction consisted of CIR the top 2.0” of the existing HMA layer and

overlaying it with 2.0” dense graded HMA and 0.75” OGFC.

Figure 3.12 shows the performance related data for contract 2876. Prior to CIR

the pavement exhibited a fair PSI (2.6), and transverse cracks. After the CIR process, the

pavement maintained a PSI value of 3.8 until 2005. The pavement started showing

transverse cracks one year after construction.

3.2.2 Reinforced Fabrics (RF) Projects

Paving fabrics are a special class of geosynthetic that provide the generally

acknowledged functions of a stress-absorbing interlayer and a waterproofing membrane.

The stress-related performance has been easily verified by the observed reductions of

cracking in pavement overlays. Paving fabric interlayer systems have been used in more

than 142,000 lane-miles of pavement in the U.S (31).

A total of six NDOT projects with paving fabrics constructed between 1999 and

2001 were analyzed in this study. The construction consisted of cold milling 2.0” of the

existing HMA layer, placing fiberglass yarns, and overlaying with 2.0” Type II (1 inch

maximum size) dense graded HMA and 0.75” OGFC.

Contract 2761

This contract was constructed in 1999 on SR443 in Washoe county over 0.87

miles. The construction consisted of cold milling 2.0” of the existing HMA layer,
 45

placing fiberglass yarns, and overlaying with 2.0” Type II dense graded HMA (AC-20P)

and 0.75” OGFC.

Figure 3.13 shows the performance related data for contract 2761. Prior to

construction the pavement was experiencing a PSI of 2.2 along with several surface

cracks. After construction, the pavement experienced an increase in PSI value to 3.5

which kept on decreasing with time to reach a value of 2.8 in 2005. The pavement

started showing minor transverse cracks one year after construction.

Contract 2932

This contract was constructed in 1999 on US095 in Mineral county over 1.17

miles. The construction consisted of cold milling 2.0” of the existing HMA layer,

placing fiberglass yarns, and overlaying with 2.0” Type II dense graded HMA (AC-20P)

and 0.75” OGFC.

Figure 3.14 shows the performance related data for contract 2932. Prior to

construction the pavement exhibited a PSI of 2.4 along with severe transverse cracks.

After construction, the pavement experienced an increase in PSI value to 3.9 and

maintained a constant PSI level for 3 years after which it started decreasing with time.

No surface cracks were observed since construction until 2005. In 2005, the pavement

showed minor rut depth of 0.1 inch.

Contract 2980a

This contract was constructed in 2000 on US050 in Churchill county over 1.74

miles. The construction consisted of cold milling 2.0” of the existing HMA layer,

placing fiberglass yarns, and overlaying with 2.0” Type II dense graded HMA (AC-20P)

and 0.75” OGFC.

 46

Figure 3.15 shows the performance related data for contract 2980a. Prior to

construction the pavement exhibited a PSI of 2.3 along with moderate transverse cracks.

After construction, the pavement experienced an increase in PSI value to 3.8 and

maintained a steady PSI level until 2005. No surface cracks were observed since

construction until 2005. In 2005, the pavement showed a rut depth of 0.14 inch.

Contract 2980b

This contract was constructed in 2000 on US095 in Churchill county over 2.03

miles. The construction consisted of cold milling 2.0” of the existing HMA layer,

placing fiberglass yarns, and overlaying with 2.0” Type II dense graded HMA (AC-20P)

and 0.75” OGFC.

Figure 3.16 shows the performance related data for contract 2980b. Prior to

construction the pavement exhibited a PSI of 2.8 along with moderate transverse cracks.

After construction, the pavement experienced a significant increase in PSI value to 4.2

and maintained a steady PSI level until 2005. Some minor transverse cracks started to

reflect 2 years after construction at which a maximum rut depth of 0.2 inch was also

observed. At the end of year 2005, the pavement showed a decrease in the rut depth to

0.03 inch.

Contract 3006

This contract was constructed in 2001 on IR080 in Humboldt county over 0.23

miles. The construction consisted of cold milling 2.0” of the existing HMA layer,

placing fiberglass yarns, and overlaying with 2.0” Type II dense graded HMA (AC-20P)

and 0.75” OGFC.

 47

Figure 3.17 shows the performance related data for contract 3006. Prior to

construction the pavement exhibited a PSI of 3.5 along with minor transverse and fatigue

cracking. After construction, the pavement experienced an increase in PSI value to 4.2

and maintained a steady PSI level until 2005. No surface cracks were observed until

2005. The pavement showed rut depths of 0.2 and 0.1 inch in years 2003 and 2005,

respectively.

Contract 3008

This contract was constructed in 2001 on IR080 in Elko county over 5.17 miles.

The construction consisted of cold milling 2.0” of the existing HMA layer, placing

fiberglass yarns, and overlaying with 2.0” Type II dense graded HMA (AC-20P) and

0.75” OGFC.

Figure 3.18 shows the performance related data for contract 3008. Prior to

construction the pavement exhibited a PSI of 3.8 along with minor transverse and fatigue

cracking. After construction, the pavement experienced an increase in PSI value to 4.1

and maintained a steady PSI level for 3 years after construction after which the PSI

dropped down to 3.9. Reflective transverse cracks were observed one year after

construction. The pavement showed rut depths of 0.2 and 0.1 inch in years 2003 and

2005, respectively.

3.2.3 Stress Relief Course (SRC) Projects

The stress relief course consists of a NDOT Type III (0.5 inch maximum size)

dense graded HMA layer. The stress relieve course is placed between the existing

pavement and the overlay layer and is intended to act as a separation layer between the

cracked surface and the overlay. A total of five NDOT projects with a stress relief course
 48

constructed between 1997 and 2003 were analyzed in this study. The SRC treatment

consisted of cold milling 2.0” of the existing HMA layer, placing a 1” stress relief course

and overlaying with 2.0” Type II (1 inch maximum size) dense graded HMA and 0.75”

OGFC.

Contract 2723

This contract was constructed in 1997 on US095 in Clark county over 20.61

miles. The construction consisted of cold milling 2.0” of the existing HMA layer,

placing a 1” stress relief course (AC-20P), and overlaying with 2.0” Type II dense graded

HMA (AC-30) and 0.375” OGFC.

Figure 3.19 shows the performance related data for contract 2723. Prior to

construction the pavement exhibited a fair PSI of 2.7 along with transverse cracking and

severe fatigue cracking. After construction, the pavement experienced an increase in PSI

value to 4.2 and maintained a steady PSI level. No surface cracks were observed since

construction until 2005. In 2005, the pavement showed a rut depth of 0.1 inch.

Contract 3031

This contract was constructed in 2000 on US395 in Washoe county over 4.03

miles. The construction consisted of cold milling 2.0” of the existing HMA layer,

placing a 1” stress relief course (AC-20P), and overlaying with 2.0” Type II dense graded

HMA and 0.375” OGFC.

Figure 3.20 shows the performance related data for contract 3031. Prior to

construction the pavement exhibited a moderate PSI of 3.1 along with moderate

transverse cracking and Type A block cracking. After construction, the pavement
 49

experienced an increase in PSI value to 3.8 and maintained a steady PSI level for 4 years

to drop to a level of 3.5 in 2005 when transverse cracks reflected to the pavement surface.

Contract 3048

This contract was constructed in 2000 on SR157 in Clark county over 8.63 miles.

The construction consisted of cold milling 2.0” of the existing HMA layer, placing a 1”

stress relief course (AC-20P), and overlaying with 2.0” Type II dense graded HMA (AC-

30) and 0.375” OGFC.

Figure 3.21 shows the performance related data for contract 3048. Prior to

construction the pavement exhibited a PSI of 3.4 along with moderate transverse cracking

and Type A fatigue cracking. After construction, the pavement experienced an increase

in PSI value to 4.3 and maintained a steady PSI level for 5 years. No serious surface

cracks were observed since construction until 2005.

Contract 3045

This contract was constructed in 2001 on US050 in Eureka county over 0.57

miles. The construction consisted of cold milling 2.0” of the existing HMA layer,

placing a 1” stress relief course (AC-20P), and overlaying with 2.0” Type II dense graded

HMA (AC-30) and 0.375” OGFC.

Figure 3.22 shows the performance related data for contract 3045. Prior to

construction the pavement exhibited a PSI of 3.8 along with minor transverse cracking

and major Type B fatigue cracking. After construction, the pavement experienced an

increase in PSI value to an average value of 4.2 and maintained a steady PSI level for 4

years. No surface cracks were observed since construction until 2005.

Contract 3162
 50

This contract was constructed in 2003 on US395 in Washoe county over 2.03

miles. The construction consisted of cold milling 2.0” of the existing HMA layer,

placing a 1” stress relief course (PG64-28NV), and overlaying with 2.0” Type II dense

graded HMA and 0.375” OGFC.

Figure 3.23 shows the performance related data for contract 3162. Prior to

construction the pavement exhibited a PSI of 3.4 along with minor transverse cracking

and a rut depth of 0.27 inch. After construction, the pavement did not experience an

increase in the PSI value and maintained a steady PSI level of 3.5 for 2 years. No surface

cracks were observed since construction.

3.2.4 Mill and Overlay (MOL) Projects

This technique consists of cold milling up to 2 inches of the existing HMA layer

and replacing it with a HMA overlay. The intention is to reduce reflective cracking by

eliminating surface cracks through cold milling and replacing with new HMA material.

A total of ten NDOT projects constructed between 1990 and 2003 were analyzed in this

study.

Contract 2384a

This contract was constructed in 1990 on US095 in Carson City over 1.79 miles.

The construction consisted of cold milling 1.0” of the existing HMA layer and overlaying

with 1.0” Type III dense graded HMA (AC-10) and 0.75” OGFC.

Figure 3.24 shows the performance related data for contract 2384a. Prior to construction

the pavement exhibited a fair PSI of 2.8 along with minor transverse cracking and severe

Type B fatigue cracking. After construction, the pavement experienced a minor increase
 51

in the PSI value to 3.2 and kept on decreasing with time to reach a value of 2.8 in 1995.

Reflected transverse and fatigue cracking showed up one year after construction.

Contract 2384b

This contract was constructed in 1993 on US095 in Lyon county over 2.74 miles.

The construction consisted of cold milling 1.0” of the existing HMA layer and overlaying

with 1.0” Type III dense graded HMA (AC-10) and 0.75” OGFC.

Figure 3.25 shows the performance related data for contract 2384b. Prior to

construction the pavement exhibited a PSI of 2.4 along with transverse cracking and

fatigue cracking. After construction, the pavement experienced an increase in the PSI

value to 3.7 and kept steady for 2 years and then started decreasing with time to reach a

value of 3.3 in 2000. Reflected transverse cracks showed up two years after construction.

Contract 2432

This contract was constructed in 1993 on SR157 in Clark county over 8.77 miles.

The construction consisted of cold milling 1.0” of the existing HMA layer and overlaying

with 1.0” Type III dense graded HMA (AC-20) and 0.75” OGFC.

Figure 3.26 shows the performance related data for contract 2432. Prior to

construction the pavement exhibited a moderate PSI of 3.0 along with severe transverse

cracks. After construction, the pavement experienced an increase in the PSI value to 3.8

and started decreasing with time. Reflected transverse cracks showed up five years after

construction.
 52

Contract 2505

This contract was constructed in 1993 on US095 in Mineral county over 1.47 miles. The

construction consisted of cold milling 1.0” of the existing HMA layer and overlaying

with 1.0” Type III dense graded HMA (AC-20) and 0.75” OGFC.

Figure 3.27 shows the performance related data for contract 2505. Prior to

construction the pavement exhibited a fair PSI of 2.7 along with fatigue and block

cracking. After construction, the pavement experienced an increase in the PSI value to

3.7 and started decreasing with time. No surface cracks showed up after construction.

Contract 2651a

This contract was constructed in 1995 on US095 in Esmeralda county over 3.0

miles. The construction consisted of cold milling 1.5” of the existing HMA layer and

overlaying with 1.5” Type II dense graded HMA (AC-20P) and 0.75” OGFC.

Figure 3.28 shows the performance related data for contract 2651a. Prior to

construction the pavement exhibited a fair PSI of 2.8 along with moderate fatigue and

block cracking. After construction, the pavement experienced an increase in the PSI

value to 3.9 and started decreasing with time. Block cracking Type A was observed 5

years after construction.

Contract 2651b

This contract was constructed in 1996 on US095 in Esmeralda county over 6.12

miles. The construction consisted of cold milling 1.5” of the existing HMA layer and

overlaying with 1.5” Type II dense graded HMA (AC-20P) and 0.75” OGFC.

Figure 3.29 shows the performance related data for contract 2651b. Prior to

construction the pavement exhibited a moderate PSI of 3.0 along with severe block
 53

cracking. After construction, the pavement experienced an increase in the PSI value to

3.9 and started decreasing with time. Minor transverse cracks were observed 5 years

after construction.

Contract 2651c

This contract was constructed in 1996 on US095 in Esmeralda county over 3.0

miles. The construction consisted of cold milling 1.5” of the existing HMA layer and

overlaying with 1.5” Type II dense graded HMA (AC-20P) and 0.75” OGFC.

Figure 3.30 shows the performance related data for contract 2651c. Prior to

construction the pavement exhibited a moderate PSI of 3.1 along with moderate to severe

surface cracks. After construction, the pavement experienced an increase in the PSI

value to 4.0 and started decreasing with time. Minor transverse cracks and fatigue

cracking were observed 5 years and 3 years after construction, respectively.

Contract 2679

This contract was constructed in 1997 on US095 in Esmeralda county over 12.92

miles. The construction consisted of cold milling 1.5” of the existing HMA layer and

overlaying with 1.5” Type II dense graded HMA (AC-20P) and 0.75” OGFC.

Figure 3.31 shows the performance related data for contract 2679. Prior to

construction the pavement exhibited a fair PSI of 2.7 along with severe transverse cracks

and minor block and fatigue cracking. After construction, the pavement experienced an

increase in the PSI value to 3.9 and started decreasing with time. Reflected transverse

cracks were observed 5 years after construction.

 54

Contract 3028

This contract was constructed in 2000 on SR512 in Carson City over 1.38 miles.

The construction consisted of cold milling 1.0” of the existing HMA layer and overlaying

with 1.0” Type III dense graded HMA (AC-20P) and 0.75” OGFC.

Figure 3.32 shows the performance related data for contract 3028. Prior to

construction the pavement exhibited a moderate PSI of 3.1 along with moderate

transverse cracks. After construction, the pavement experienced an increase in the PSI

value to 3.8 and stayed steady for 2 years and started decreasing with time to reach a

value of 3.4 in 2005. Reflected transverse cracks were observed 3 years after

construction.

Contract 3070

This contract was constructed in 2003 on SR160 in Nye county over 3.56 miles.

The construction consisted of cold milling 1.5” of the existing HMA layer and overlaying

with 1.5” Type II dense graded HMA (AC-20P) and 0.75” OGFC.

Figure 3.33 shows the performance related data for contract 2651c. Prior to

construction the pavement exhibited a moderate PSI of 3.1 along with moderate surface

cracks. After construction, the pavement experienced an increase in the PSI value to 4.3

and maintained a steady value. No surface cracks were observed two years after

construction.

3.2.5 Crack and Seat (CS) Projects

The main concern with overlaying rigid pavements with HMA overlays is

reflective cracking at the joints and cracks propagating through the HMA layer. The

crack and seat technique is used in attempt to eliminate or retard reflective cracking in the
 55

HMA overlay over concrete pavements. Cracking the existing slab into large pieces of 1

to 3 feet in length is intended to produce tight cracks that permit load transfer with little

loss of structural value. Seating the cracked concrete slab is intended to reestablish full

contact with the foundation. A total of five NDOT projects with the crack and seat

technique constructed between 1998 and 2001 were analyzed in this study.

Contract 2886

This contract was constructed in 1998 on IR080 in Lander county over 5.70 miles.

The construction consisted of cracking and seating the existing 8-inch slabs, placing a

leveling course of 1.5”, and overlaying with 4.5” Type II dense graded HMA (AC-20P)

and 0.375” OGFC.

Figure 3.34 shows the performance related data for contract 2886. Prior to

construction the pavement exhibited a fair PSI of 2.8. No surface cracks were observed

before construction. After construction, the pavement experienced an increase in the PSI

value to 4.1 and maintained a steady value for 4 years after which reflected transverse

cracks start to show up resulting in a reduction of the PSI value as a function of time.

Contract 2889

This contract was constructed in 1999 on IR080 in Elko county over 5.39 miles.

The construction consisted of cracking and seating the existing 8-inch slabs, placing a

leveling course of 1.5”, and overlaying with 3.5” Type II dense graded HMA (AC-20P)

and 0.375” OGFC.

Figure 3.35 shows the performance related data for contract 2889. Prior to

construction the pavement exhibited a fair PSI of 2.9. No surface cracks were observed

before construction. After construction, the pavement experienced an increase in the PSI
 56

value to 4.0 and maintained a steady value. Minor fatigue cracking were observed 4

years after construction.

Contract 2962

This contract was constructed in 1999 on IR080 in Elko county over 15.04 miles.

The construction consisted of cracking and seating the existing 8-inch slabs, placing a

leveling course of 1.5”, and overlaying with 4.75” Type II dense graded HMA (AC-20P)

and 0.375” OGFC.

Figure 3.36 shows the performance related data for contract 2962. Prior to

construction the pavement exhibited a fair PSI of 2.7 along with minor transverse cracks.

After construction, the pavement experienced an increase in the PSI value to 4.3 and

maintained a steady value. Minor fatigue cracking were observed 6 years after

construction.

Contract 2999

This contract was constructed in 2001 on IR080 in Elko county over 5.90 miles.

The construction consisted of cracking and seating the existing 8-inch slabs, placing a

leveling course of 1.5”, and overlaying with 3.5” Type II dense graded HMA (AC-20P)

and 0.375” OGFC.

Figure 3.37 shows the performance related data for contract 2999. Prior to

construction the pavement exhibited a good PSI of 3.6. No surface cracks were observed

before construction. After construction, the pavement experienced an increase in the PSI

value to 4.1 and maintained a steady value. Minor fatigue cracking were observed 4 years

after construction.
 57

Contract 3021

This contract was constructed in 2001 on IR080 in Elko county over 14.72 miles.

The construction consisted of cracking and seating the existing 6-inch slabs, placing a

leveling course of 1.5”, and overlaying with 3.5” Type II dense graded HMA (AC-20P)

and 0.375” OGFC.

Figure 3.38 shows the performance related data for contract 3021. Prior to

construction the pavement exhibited a fair PSI of 2.9. No surface cracks were observed

before construction. After construction, the pavement experienced an increase in the PSI

value to 4.3 and started slightly decreasing with time. Minor Type A fatigue cracking

were observed 4 years after construction.

3.2.6 Rubblization (RPCC) Projects

The rubblization provides a total destruction of the existing slabs. A total of six

NDOT projects with the rubblization technique constructed between 1996 and 2005 were

analyzed in this study.

Contract 2549

This contract was constructed in 1996 on IR080 in Elko county over 8.92 miles.

The construction consisted of rubblizing the existing concrete slabs, placing a leveling

course of 1.5”, and overlaying with 5.0” Type II dense graded HMA and 0.375” OGFC.

Figure 3.39 shows the performance related data for contract 2549. Prior to

construction the pavement exhibited a fair PSI of 2.9 along with minor transverse cracks.

After construction, the pavement experienced an increase in the PSI value to 4.0 and

maintained a steady value. Minor fatigue cracking were observed 5 years after

construction.
 58

Contract 2869

This contract was constructed in 1999 on IR080 in Elko county over 6.38 miles.

The construction consisted of rubblizing the existing concrete slabs, placing a leveling

course of 1.5”, and overlaying with 5.0” Type II dense graded HMA and 0.375” OGFC.

Figure 3.40 shows the performance related data for contract 2869. Prior to

construction the pavement exhibited a low PSI of 1.9. No surface cracks were observed

before construction. After construction, the pavement experienced an increase in the PSI

value to 4.2 and started slightly decreasing with time. Fatigue cracking was observed 4

years after construction.

Contract 2901

This contract was constructed in 1999 on IR080 in Humboldt county over 5.68

miles. The construction consisted of rubblizing the existing concrete slabs, placing a

leveling course of 1.5”, and overlaying with 7.0” Type II dense graded HMA (AC-20P)

and 0.375” OGFC.

Figure 3.41 shows the performance related data for contract 2901. Prior to

construction the pavement exhibited a fair PSI of 2.7. No surface cracks were observed

before construction. After construction, the pavement experienced an increase in the PSI

value to 3.8 and maintained a steady value with time. Minor fatigue cracking were

observed 6 years after construction.

Contract 3088

This contract was constructed in 2002 on IR080 in Pershing county over 7.62

miles. The construction consisted of rubblizing the existing concrete slabs, placing a
 59

reinforced fabric and non-woven geotextile, placing a leveling course of 1.5”, and

overlaying with 5.0” Type II dense graded HMA (AC-20P) and 0.375” OGFC.

Figure 3.42 shows the performance related data for contract 3088. Prior to

construction the pavement exhibited a moderate PSI of 3.2 along with moderate

transverse cracks. After construction, the pavement experienced an increase in the PSI

value to 4.2 and maintained a steady value with time. No surface cracks were observed 3

years after construction.

Contract 3186a

This contract was constructed in 2005 on IR080 in Humboldt county over 1.06

miles. The construction consisted of rubblizing the existing concrete slabs, placing a

leveling course of 2.0”, and overlaying with 4.5” Type II dense graded HMA (AC-20P)

and 0.375” OGFC.

Figure 3.43 shows the performance related data for contract 3186a. Prior to

construction the pavement exhibited a PSI of 3.5. No surface cracks were observed

before construction. After construction, the pavement experienced an increase in the PSI

value to 4.2. No PMS data exist for this project at the time of writing this thesis.

Contract 3186b

This contract was constructed in 2005 on IR080 in Lander county over 1.06 miles.

The construction consisted of rubblizing the existing concrete slabs, placing a leveling

course of 2.0”, and overlaying with 4.5” Type II dense graded HMA (AC-20P) and

0.375” OGFC.

Figure 3.44 shows the performance related data for contract 3186b. Prior to

construction the pavement exhibited a PSI of 3.6. No surface cracks were observed
 60

before construction. After construction, the pavement experienced an increase in the PSI

value to 3.9. No PMS data exist for this project at the time of writing this thesis.

3.3 OVERALL SUMMARY OF THE PERFORMANCE OF THE SELECTED

NDOT PROJECTS

The pavement distresses in terms of surface cracks before and after the

application of the treatment are summarized in Table 3.2 for each project. Additionally,

Table 3.2 shows for each project, the severity of each type of distress considered along

with the number of years after construction that each type of cracks appeared on

pavement surface. The column labeled “treatment life” in the tables indicates the number

of years after construction when cracks first appeared on pavement surface. In general,

all projects experienced pre-rehabilitation surface cracks ranging from minor to severe.

Based on the review of the treatment types and applications, all forty four NDOT

projects were grouped into sub-treatments with similar characteristics.

CIR treatments

• CIR-A: Contracts 2808a, 2808b, 2838 and 2935. CIR the top 2.0” of the existing
HMA layer and overlaying it with 2.5” dense graded HMA and 0.75” open graded
friction course (OGFC).
• CIR-B: Contracts 2819, 2873, 2961, and 3013. CIR the top 3.0” of the existing
HMA layer and overlaying it with 3.0” dense graded HMA and 0.75” OGFC.
• CIR-C: Contracts 3025a, 3025b, 3025c and 2876. CIR the top 2.0” of the existing
HMA layer and overlaying it with 2.0” dense graded HMA and 0.75” OGFC.

RF treatments

• RF: Contracts 2761, 2932, 2980a, 2980b, 3006, and 3008. Cold milling the top
2.0” of the existing HMA layer, placing fiberglass yarns, and overlaying with 2.0”
Type II dense graded HMA and 0.75” OGFC.
 61

SRC treatments

• SRC: Contracts 2723, 3031, 3048, 3045, and 3162. Cold milling the top 2.0” of
the existing HMA layer, placing a 1” stress relief course and overlaying with 2.0”
Type II dense graded HMA and 0.75” OGFC.

MOL treatments

• MOL-A: Contracts 2384a, 2384b. Cold milling 1.0” of the existing HMA layer
and overlaying with 1.0” Type III dense graded HMA (AC-10) and 0.75” OGFC.
• MOL-B: Contracts 2432, 2505, and 3208. Cold milling 1.0” of the existing HMA
layer and overlaying with 1.0” Type III dense graded HMA (AC-20 or AC-20P)
and 0.75” OGFC.
• MOL-C: Contracts 2651a, 2651b, 2651c and 2679 and 3070. Cold milling 1.5” of
the existing HMA layer and overlaying with 1.5” Type II dense graded HMA
(AC-20P) and 0.75” OGFC.

CS treatments

• CS: Contracts 2886, 2889, 2962, 2999, and 3021. Cracking and seating the
existing slabs, placing a leveling course of 1.5”, and overlaying with Type II
dense graded HMA (AC-20P) and 0.375” OGFC.

RPCC treatments
• RPCC: Contracts 2549, 2869, 2901, 3088, 3186a, and 3186b. Rubblizing the
existing concrete slabs, placing a leveling course of 1.5 or 2.0”, and overlaying
with Type II dense graded HMA and 0.375” OGFC.

The CIR treatment was mainly used on roadways with an AADT less than 6,000

except one project on US95 with 14,500 AADT. The performance data in Table 3.2

indicate the following trends for the CIR projects:

• The CIR project with the lowest AADT (CIR-A-4) did not experience any
distresses after 6 years in service. On the other hand the CIR project with the
highest AADT (CIR-B-4) was among the best performers after 6 years in service.
• Reflective transverse cracking was the most common type of distress on CIR
projects. Seven CIR projects experienced reflective transverse cracking 1 to 2
years after construction. Three CIR projects experienced reflective transverse
cracking 6 to 7 years after construction.
• Two CIR projects with medium AADT (CIR-B-1 and CIR-B-2) experienced
fatigue cracking after 3 and 7 years in service, respectively.
• All the CIR-C projects (CIR 2.0”, HMA overlay 2”) experienced reflective
transverse cracking 1 year after construction.
 62

The reinforced fabric treatment was used on roadways with AADT between 1,000

and 10,000. The performance data in Table 3.2 indicate the following trends for the RF

projects:

• Three out of the six RF projects did not experience any distresses after 4 to 6
years in service.
• Reflective transverse cracking was the most common type of distress on the RF
projects. Three out of six RF projects experienced reflective transverse cracking
after 1 to 3 years in service.
• The two RF projects with the highest AADT (RF-1 and RF-6) were among the
worst performers.

The 1-inch stress relief course after cold milling and before application of overlay

was used on roadways with AADT between 1,900 and 40,000. The performance data in

Table 3.2 indicate the following trends for the SRC projects:

• Three out of the five SRC projects did not experience any distresses. One of these
projects is 8 years old and has the highest AADT (SRC-1) while the other two are
4 and 2 years old.
• Two out of the five SRC projects experienced reflective transverse cracking after
5 years in service.

The mill and overlay treatment was used on roadways with AADT between 1,700

and 40,000. The performance data in Table 3.2 indicate the following trends for the

HMA projects:

• One mill and overlay project did not experience any distresses after 12 years in
service (MOL-B-2).
• Reflective transverse cracking was the most common type of distress on the mill
and overlay projects. The length of time after construction for the transverse
cracking to reflect to the surface ranged from 1 to 5 years.
• The mill and overlay projects were the only ones that experienced block cracking
after 5 to 6 years in service.
• The worst performing mill and overlay project was the one that had the highest
AADT (MOL-A-1) which experienced moderate reflective fatigue and transverse
cracking after 1 year in service.
 63

The crack and seat treatment was used on PCC rehabilitation projects of roadways

with AADT between 4700 and 23000. The treatment existed for a period of four to seven

in-service years. The performance data in Table 3.2 indicate the following trends for the

crack and seat projects:

• Before rehabilitation, the PCC pavements did not show any type of surface cracks
for four CS projects out of five. Only project CS-3 had minor transverse cracking
before treatment application.
• Fatigue cracking was the most common type of distress on the crack and seat
projects. The crack and seat projects experienced fatigue cracking in the HMA
overlay after 4 to 6 years in service.
• Only project CS-1 out of the five CS projects experienced reflective transverse
cracking after 4 years in service.
• All CS projects with 3.5” overlay experienced fatigue cracking four years after
construction whereas CS projects with more than 4.5 inches overlay experienced
fatigue cracking six years after construction.

The rubblization treatment was used on PCC rehabilitation projects of roadways

with AADT between 6300 and 7700. The treatment existed for a period of maximum

nine in-service years. The performance data in Table 3.2 indicate the following trends for

the rubblization projects:

• Before rehabilitation, the PCC pavements did not show any type of surface cracks
for five RPCC projects out of six. Only project RPCC-1 had minor transverse
cracking before treatment application.
• Fatigue cracking was the type of distress on the rubblized projects. The rubblized
projects experienced fatigue cracking in the HMA overlay after 4 to 6 years in
service.
• The rubblized project RPCC-4 constructed in 2002 and after 3 years in service did
not experience any type of surface cracks.
• No performance data exists for the projects RPCC-5 and RPCC-6 that were
constructed in 2005.

3.4 STATISTICAL ANALYSIS OF THE SURFACE CRACKING DATA –

PRINCIPAL COMPONENT ANALYSIS

In order to objectively evaluate the overall performance of the various treatments,

and in order to relatively compare the performance of the treatments to each others the
 64

statistical technique called Principal Component Analysis (PCA) was used as a tool to

determine an overall performance indicator. The PCA indicator combines the

simultaneous effects of cracking occurrence and severity. Therefore, the PCA works as

an indicator of the ability of each treatment to delay the reflective cracking. A detailed

explanation of the PCA statistical tool is discussed in the following paragraphs.

The Principal Components Analysis (PCA) is a way of identifying patterns in

data, and expressing them in such a way as to highlight their similarities and differences.

A formal definition indicates that PCA is a method that reduces data dimensionality by

performing a covariance analysis among factors. As such, it is suitable for data sets in

multiple dimensions, such as a large data set of pavement distresses. The main

applications of PCA are: (1) to reduce the number of variables and (2) to detect structure

in the relationships among variables, that is to rank variables.

An example to understand the PCA analysis is provided here for combining two

variables into a single factor. Typically, the correlation between two variables can be

presented in a scatter-plot. A regression line can then be fitted that represents the "best"

summary of the linear relationship between the variables. If we could define a factor that

would approximate the regression line in such a plot, then that factor would capture most

of the "essence" of the two variables. Therefore, single scores on that new factor,

represented by the regression line, could then be used in future data analyses to represent

the essence of the two items. In a sense, we have reduced the two variables to one factor.

Note that the new factor is actually a linear combination of the two variables.

The example described above, combining two correlated variables into one factor,

illustrates the basic idea of PCA. If we extend the two-variable example to multiple
 65

variables, then the computations become more involved, but the basic principle of

expressing two or more variables by a single factor remains the same.

The most valuable product obtained from PCA is a linear parameter estimated

from a linear combination of the original variables. The objective of the PCA is to

reduce dimensionality by extracting the smallest number of variables (factors) that

account for most of the variation in the original multivariate data and to summarize the

data with little loss of information. In our case, the multivariate data corresponds to the

NDOT cracking surveys for each year while the dimensionality represents the extent,

severity, and years of occurrence of the various types of cracking. Therefore, in the case

of cracking, the PCA tries to provide a combined indicator or a factor that interprets the

simultaneous effects of fatigue, transverse, and block cracking in terms of their extent,

severity, and time of occurrence (years).

3.4.1 Overall Ranking of the Performance of Selected NDOT Projects

The PCA analysis was conducted on all projects combined at one year before

rehabilitation (-1) and at one (+1), three (+3), and five years (+5) after treatment

application based on their measured surface cracks (fatigue, transverse, and block

cracking). In other words, the PCA scores are used for ranking the combined projects

based on their performance at year -1 and +1, at year -1 and +3, and at year -1 and +5.

Conducting the PCA ranking of the projects at pre-rehabilitation and after construction at

the same time is a good indication of the relative effectiveness of the various treatments

taking into consideration pre-construction pavement conditions and allowing for a direct

comparison of the condition of the pavement after construction to its condition before

rehabilitation. The CIR, RF, SRC, and MOL projects are analyzed separately from the
 66

CS and RPCC projects which consisted of HMA overlays over PCC pavements. The

SAS Macro called “factor” was used to perform the PCA analysis (43). Such macro was

developed by Dr. George Fernandez. Appendix C shows the various PCA values

calculated for each year.

Table 3.3 shows the ranking of the various treatments based on the PCA analysis

of the cracking data of the various projects. As stated before, the analysis for the flexible

pavement techniques was conducted separately from the analysis for the rigid pavement

techniques and hence separate rankings were assigned. Additionally an “NA” was

assigned if the project life was younger than the indicated long-performance analysis

year.

In the case of the flexible pavement techniques (i.e., CIR, RF, SRC, and MOL),

first all thirty three projects at their pre-treatment conditions (-1) and their condition at

one year (+1) after treatment application were ranked between 1 and 66 with a rank of

“1” indicating the best conditions project and a rank of “66” indicating the worst

conditions project. However, the data in Table 3.3 show that the maximum ranking is

“38” instead of “66”, this occurred because some projects ranked the same indicating that

the conditions of the similarly ranked projects are statistically the same. For example

projects CIR-C-2, CIR-C-3, and CIR-C-4 were all ranked at (+1) as “5” indicating

statistically the same pavements’ condition.

Then, the projects were ranked at +3 and +5 years after treatment application by

re-conducting the PCA analysis at both the -1 and the analysis year (i.e., +3 or +5)

excluding the projects that were younger than the indicated long-performance year.

Subsequently, the ranking was based on the same total number of projects before (-1) and
 67

after treatment application (+3 or +5). At -1 and +3 years, a total of thirty projects were

ranked between 1 and 60 with a maximum ranking of “39”. At -1 and +5 years, a total of

24 projects were ranked between 1 and 48 with a maximum ranking of “36”.

In the case of the ranking of projects at the +1, +3, and +5 years, there are several

occasions where several projects are given the same rank. This occurred due to the

statistically similar performance of these projects. For example in the case of ranking

based on (-1) and (+1) years in Table 3.3, a ranking of “1” is assigned to each project that

did not experience any cracking 1 year after construction. On the other hand if two

projects showed cracking 1 year after construction, the extent and severity of the cracking

would then play a role in the ranking. For example, projects CIR-C-1 and CIR-C-2 both

showed minor transverse cracking in year 1 but the PCA analysis assigned a ranking of

“3” and “5”, respectively. The reason for the worse ranking of project CIR-C-2 is that

the extents of the minor transverse cracking of project CIR-C-2 is higher than the extent

of the minor transverse cracking on project CIR-C-1. In summary, the projects are first

ranked based on the year of cracking occurrence, followed by the severity of the

cracking, and finally by the extent of the cracking. In the cases where all three factors are

the same, then the projects are assigned similar ranks.

The same process was followed for ranking all eleven projects for the rigid

pavement techniques (i.e., CS and RPCC). All eleven projects at -1 and +1, and at -1 and

+3 years were ranked between 1 and 11 with a maximum ranking of “3.” All projects

ranked as “1” except the CS-3 and RPCC-1 projects that were ranked as “2” and “3” at

year -1 because of the minor transverse cracks that existed before the application of the

treatment.
 68

By examining the data presented in Table 3.3, the following observations can be

made:

• One year after construction a total of five CIR projects out of twelve performed
among the worst projects. Four out of these worst performing five CIR projects
(CIR-C-1, -2, -3, and -4) consisted of milling the top 2.0” of the existing layer and
overlaying it with 2.0” dense grade HMA. Only one RF project (RF-1) out of six
ranked among the worst projects. The RF-1 project experienced severe alligator
cracking (Type B) prior to the construction of the treatment. All of the SRC
projects ranked among the best performing projects. Two MOL projects (MOL-
A-1, C-4) ranked among the worst performing projects with MOL-C-4 pavement
condition 1 year after construction worse than its condition 1 year before
construction. In the case of rehabilitation techniques for rigid pavements all
treatments (i.e., CS and RPCC) ranked best with no difference in performance.
• Three years after construction, additional two CIR, RF, and MOL projects
performed among the worst projects whereas the SRC projects still outperformed
the other treatments. Evidently the MOL-C-4 pavement condition 3 year after
construction was worse than its condition 1 year before construction. In the case
of rehabilitation techniques for rigid pavements all treatments (i.e., CS and
RPCC) ranked best with no difference in performance.
• Five years after construction, five CIR projects out of eight performed among the
best projects. The five CIR projects included three out of the four CIR-B projects.
The CIR-B treatment consists of milling the top 3.0” of the existing HMA layer
and overlaying it with 3.0” dense graded HMA. Only two RF projects out of four,
two SRC projects out of three, and one MOL project out of nine ranked among
the best performing projects. In the case of rehabilitation techniques for rigid
pavements only one CS project out of five and one RPCC project out of three
ranked best with no surface cracks five years after construction. Two RPCC
treatments out of three ranked among the worst performing treatments.

A close examination of the performance data summarized in Table 3.3 show that

five out of the seven projects that experienced severe alligator cracking (Type B) prior to

the construction of the CIR, RF, SRC, and MOL treatments ranked among the worst

projects five years after construction. Out of these five projects, two are CIR projects

(CIR-A-2 and CIR-B-1), one is RF project (RF-1), and two are MOL projects (MOL-A-1

and MOL-A-2). This indicates the ineffectiveness of the CIR, RF, and MOL-A

treatments on a pavement with severe alligator cracking, and the effectiveness of the SRC
 69

treatment on a pavement with severe alligator cracking and high volume traffic (40,000

AADT).

Figures 3.45, 3.46, and 3.47 show a comparison between the ranking of the

various treatments at one year before construction versus the ranking at one, three and

five years after construction. The objective of these plots is to be able to assess the

effectiveness of the various treatments by comparing their historical performance as they

relate to the conditions of the pavement prior to rehabilitation. In other words, a

treatment that was applied to badly deteriorated pavements but still maintained good

performance would be more effective than a treatment that was applied to a less

deteriorated pavement that maintained good performance.

Additionally, Figures 3.45-3.47 show the mean, minimum, and maximum rank in

each treatment. This analysis was conducted on all projects having data on the analysis

performance year and excluding the projects considered as outliers ranked statistically

significantly different than the overall average rank of the same treatment. An outlier

rank in a treatment can considerably change the average rank and consequently skew and

bias the results. The statistical analysis tool called simple descriptive statistics and

exploratory graphical analysis using the SAS Macro called “univar” (43) was used to

identify the projects with an outlier rank in a specific treatment. These projects are

identified in Table 3.3. The total numbers of projects analyzed in each treatment are

shown in Tables 3.4-3.6.

By examining the data presented in Figures 3.45-3.47, the following observations

can be made:
 70

• The SRC and MOL-B and -C treatments were applied on flexible pavements that
experienced the widest range of pre-treatment (-1) performance. The CIR-C and
MOL-A were applied on flexible pavements that experienced the narrowest range
of pre-treatment performance.
• The CIR projects before treatment (-1) were among the worst performing projects.
• After one year (+1) from the treatment construction (Figure 3.45), all treatments
showed similar performance except the CIR-C and the MOL-A treatments
showed worse performance.
• After three years (+3) from the treatment construction (Figure 3.46), the CIR-A,
CIR-B, SRC, MOL-C treatments showed similar and best performance followed
by RF and MOL-B treatments. The CIR-C and the MOL-A treatments still
showed significantly worse performance.
• After five years (+5) from the treatments construction (Figure 3.47), the CIR-B
treatment performed the best followed by the CIR-A and RF treatments which
performed similarly. The performance of the SRC treatment became similar to
the MOL treatments. It should be noted that the worst performing CIR-C projects
based on the (+1) and (+3) analysis were not included in the (+5) analysis due to
the lack of the 5-years performance data from these projects.

Another way of looking at the effectiveness of the various treatments is to

examine the means and standard deviations of their relative rankings at the pre-treatment

and the 1, 3, and 5-years stages along with the change in their mean relative ranking.

Tables 3.4 through 3.6 summarize the means and standard deviation data of the various

projects along with the change in the mean relative ranking. An effective treatment

would be one that has high mean and low standard deviation for pre-treatment

performance and low mean and low standard deviation for the 1, 3, and 5 year

performances coupled with the highest change in the mean ranking. Based on this

criteria, the evaluated treatments for mitigating reflective cracking under Nevada’s

conditions are ranked from best to worst after 1, 3, and 5 year from the construction as

follows:
 71

After 1 year After 3 years After 5 years

from construction from construction from construction
CIR-A, CIR-B, MOL-C CIR-A, CIR-B, MOL-C CIR-A, CIR-B
SRC, MOL-B SRC, MOL-B RF
CIR-C CIR-C SRC, MOL-B
RF RF MOL-C
MOL-A MOL-A MOL-A

3.4.2 Fixed Qualitative and Qualitative Factorial Analysis

The objective of this analysis was to statistically determine the impact of traffic

level and project location on the long-term performance of the reflective cracking

mitigation techniques experienced by NDOT. The reflective cracking treatments (or

mitigation techniques) were: Cold in-place recycling, reinforced fabrics, stress-relief

course, mill and overlay, PCC rubblization, and PCC crack and seat. Three levels for

traffic (low: 0-3000 AADT; intermediate: 3000-7000 AADT; and high: >7000 AADT)

and two levels for project location (north of south) along with their interaction were

considered in this analysis. To this end, a fixed qualitative factorial analysis was applied

using the SAS macro-call [Link] written by Dr. G. Fernandez of the University of

Nevada (43). The analysis was based on the PCA values of the various projects after

accounting for traffic level and project location. The comparison for the various PCA

values was performed at +1, +3 and +5 years after treatment application. This factorial

treatment design is performed to investigate the relationships among the indicated factors

(reflective cracking mitigation technique, traffic, and location) and the response variable

(PCA value as a pavement performance indicator). This technique allows to provide

statistical information on the impact of the evaluated factors and the long-term pavement

performance. For example, for a given traffic level (i.e., low: 0-3000 AADT), the impact
 72

of project location (i.e., north vs south) on the long term performance of a given

reflective cracking treatment is evaluated by analyzing the PCA values a +1, +3 and +5

years. Additionally, the analysis will allow for detection of any interaction among the

various evaluated factors, thus, if any interaction exists between the traffic level and the

project location then both factors are dependant of one another and any conclusion that

needs to be drawn at the end will be a function of both traffic and project location levels

(44).

As with any analysis of variance (ANOVA), the assumptions that random

experimental errors are independent and normally distributed with zero mean and a

common variance for all treatments should not be violated.

All the analyses were performed at a 0.05 significance level (p-value = 0.05),

which mean that for each comparison reported as being significantly different, there is

only a 5% chance that this is not true. The significance level (i.e., p-value ≤ 0.05) of any

experiment is the probability that a given treatment (i.e., mitigation technique, traffic

level, and project location) has a significant effect on the response variable (i.e., PCA at

+1, +3 and +5 years. The treatment is considered as highly significant if the p-value is

equal or lower than 0.0001. A change in a highly significant treatment produces a highly

significant change in the response variable.

Tables 3.7, 3.8 and 3.9 show the general statistical analysis at +1, +3 and +5 years

after treatment construction. The statistical results indicate that the reflective cracking

mitigation technique, the traffic level, and the project location have highly significant

effects on the pavement performance at +1, +3 and +5 years after treatment construction.

In other words, there is enough statistic evidence to conclude that these factors are highly
 73

significant up to 5 years after treatment construction. It means that any change in one of

those factors is going to affect the long-term pavement performance. Additionally, the

interaction among these factors was found to be highly significant. Therefore, the

analysis of the performance of any project should consider simultaneously the

combination of the three factors evaluated: reflective cracking mitigation technique,

traffic level and project location. It implies that any consideration regarding the

pavement performance should include the type of treatment applied, the traffic level and

the location of the project because the combination of those factors will affect the

pavement condition. Unfortunately, although the three studied factors were found highly

significant as well as their interaction, it was not possible to find out any trend to

conclude under what conditions of traffic level and project location a reflective cracking

technique performed better. This may be related to the lack of number of replicates for

the various evaluated factors.

Additionally, a fixed quantitative factorial analysis was used to assess the effect

of the project condition before the treatment application on the long-term performance of

the applied reflective cracking treatment. The applied reflective cracking mitigation

technique was considered as the qualitative factor and the PCA value of the evaluated

projects before treatment application (PCA-1) was considered as quantitative factor. The

response variable selected was the current PCA value for the project at the corresponding

year. The quantitative factor (i.e., PCA-1) can have either a linear or a quadratic effect

on the response variable (i.e., PCA values at +1, +3, and +5 years). The analysis was

performed at years +1, +3 and +5 after treatment application. A fixed qualitative-

quantitative factorial analysis was applied using the SAS macro-call [Link] written by
 74

Dr. G. Fernandez of the University of Nevada (43). The significance p-value is also

applied here, in the same manner that it was explained before. Also, the interaction was

considered into the analysis. Additionally, the regression coefficient (R-square) was

included in the analysis. The multiple regression correlation coefficient, R-square, is a

measure of the proportion of the variability explained by the statistical model. It is a

number between zero and one and a value close to zero suggests a poor model. In

general, values of R-square higher than 0.7 are acceptable for field-related experiments

and indicate that the experiment is valid.

The results of this analysis are shown in Tables 3.10, 3.11 and 3.12. The data

show that the reflective cracking mitigation technique and the pavement condition before

the treatment application are highly significant at +1, +3 and +5 years after treatment

application. It means that any variation in those factors will significantly affect the

pavement performance. Therefore, the long-term performance of the project is highly

related to its previous condition before the treatment application and the type of reflective

cracking mitigation technique applied. Additionally, the interaction among such factors

was also found significant. Due to the interaction effect, the long-term performance of

the project will be affected by both the applied type of reflective cracking mitigation

technique and the pavement condition before treatment application. Unfortunately, due

to the lack of number of replicates, it was not possible to determine under what

conditions a better performance is expected.

The R-square for the three studied years was found to have an acceptable value

higher than 0.7. Therefore, the analyzed data and the performed statistical analysis
 75

accounts for most of the variability of the experiment and the conclusions can be

accepted as valid.

3.5 FINAL ANALYSIS FROM THE NDOT PROJECTS

The following general conclusions regarding the performance of reflective

cracking treatments in Nevada are based on the combined analyses of the distress data

and the PCA analysis.

• The PCA analysis was conducted to identify any reflective cracking treatment that
may be able to provide good long-term performance regardless of the conditions
of the pavement prior to its application. Based on the results of the PCA
presented in Figures 3.45 through 3.47, this goal was not achieved. In general,
the performance of the reflective cracking treatment is highly dependent on the
conditions of the pavement prior to the construction of the treatment. This is
supported by the data shown in Figure 3.47 where the CIR and MOL were applied
to pavements having lower performance than the pavement where the RF and
SRC were applied and showed worse performance after 5 years in service.

• The PCA ranking data in Table 3.3 showed that the CIR-A (CIR 2” and overlay
2.5”) and CIR-B (CIR 3.0” and overlay 3.0”) treatments regardless of the traffic
level are generally effective in stopping reflective cracking for 3 years and
retarding reflective cracking for 5 years as long as the existing pavement does not
show severe alligator cracking prior to the application of these treatments. On the
other hand the CIR-C (CIR 2” and overlay 2”) treatment is ineffective in resisting
reflective cracking for even a traffic level as low as 1000 AADT.

• The RF treatment showed marginal performance after 3 and 5 years of

construction. The RF treatment was able to retard reflective cracking for at least
3 years. The RF treatment was ineffective when applied on a pavement with
severe alligator cracking prior to the application of the treatment and/or a traffic
level above 3000 AADT.

• The SRC treatment showed excellent performance up to 3 years after construction

regardless of the traffic level and the existing pavement condition. However 5
years after construction, reflective transverse cracking showed up on the
pavement surface.

• The MOL-A (cold milling 1” and AC-10 overlay 1”) treatment was ineffective in
resisting reflective cracking of pavements with severe alligator cracking prior to
the construction of the treatment. The MOL-B and MOL-C treatments were
effective in stopping reflective cracking up to 3 years after construction for
 76

projects with AADT lower than 4000. After 5 years of construction the MOL-B
and MOL-C treatments showed minor reflected transverse cracks.

• The CS treatment showed excellent performance up to 3 years after construction

for rigid pavements with no or minor surface cracks prior construction. The CS
projects experienced fatigue cracking in the HMA overlay after 4 to 6 years in
service. The CS treatment with 3.5” HMA overlay retarded fatigue cracking for 4
years after construction whereas CS treatment with more than 4.5” HMA overlay
retarded fatigue cracking for 6 years after construction.

• The RPCC treatment showed excellent performance up to 3 years after

construction for rigid pavements with no or minor surface cracks prior
construction and AADT around 7000. The RPCC projects experienced fatigue
cracking in the HMA overlay after 4 to 6 years in service. The RPCC treatment
with 5” HMA overlay retarded fatigue cracking for 4 to 5 years after construction
whereas RPCC treatment with 7” HMA overlay retarded fatigue cracking for 6
years after construction.

• In general, the long-term effectiveness of the treatments was significantly reduced

by the existence of severe alligator cracking on the projects prior to the
application of these treatments. Therefore, it is recommended that projects
experiencing severe alligator cracking as classified by the NDOT pavement
distress manual should be subjected to either re-construction or full depth
reclamation.

3.6 SUMMARY OF THE PHASE I FINDINGS

The gathered information and conducted analyses in the various tasks of the Phase

I study of the research were used to identify the promising techniques to mitigate

reflective cracking in HMA overlays.

Based on the literature review of the current and previous efforts outside Nevada

on the mitigation of reflective cracking in HMA overlays, the following general

conclusions concerning the performance of reflective cracking treatments can be made.

• Performance of crumb rubber overlay mixtures produced by adding ground tire

rubber to HMA using the wet process in addressing reflection cracking in HMA
overlays has ranged from successful to devastating failures depending on the
percent of crumb rubber in the mix.
• Asphalt rubber interlayer with a combination of a thin overlay (about 1.5 inch)
reduced and/or delayed reflective cracking over a period of five years.
 77

• In general, performance of geosynthetics in addressing reflection cracking in

HMA overlays has ranged from highly successful to devastating failures.
• Fabric interlayers have been most effective when used for load-related fatigue
distress and have not performed well when used to delay or retard thermal
cracking. Proper construction procedures are critical for optimum performance of
paving fabrics.
• In general, glassgrid showed benefits in retarding reflective cracking and reducing
the rate of crack reflections specifically in an area of high tensile stress.
• Cold in-place recycling (CIR) of existing HMA materials provided an effective
mean of extending the life of pavement rehabilitation projects for highways with
up to 13,000 ADT and 200,000 annual equivalent single axle loads. CIR provided
resistance against reflective cracking between two and three times that exhibited
by conventionally resurfaced control sections.
• Interlayer stress-absorbing composite (ISAC) delayed reflective cracking at
treated joints and cracks from one year to three years when compared to the
untreated and other crack control methods. The ISAC consists of a three-layer
system. The top layer is a high strength woven geotextiles to resist stresses
caused by underlying pavement movements. The bottom layer is a low strength
nonwoven geotextile (meeting AASHTO M-288-92). The middle layer is a
modified rubberized asphalt layer to absorb the strain energy and bond the two
geotextiles together. Based on the local Illinois distributor of ISAC the typical
installation costs of ISAC range from $10 to $14 per foot depending on the
quantity purchased. The larger the quantity purchased, the lower the cost/foot of
ISAC.
• SAMI layers can reduce the likelihood of damage and the need for large
reconstruction work. They were successfully used to reduce the rate of reflective
cracking when the crack spacing and widths were small.
• In general, cracking and seating concrete pavements delayed reflected cracks for
five years. The use of fibers in the overlay mixture further reduced transverse
cracks.
• Fracturing technique of concrete pavement is somewhat less successful than the
crack and seat technique to retard reflective cracking in reinforced concrete
pavements. Formation of reflective cracks appeared to be delayed for only about
three years.

Based on the review of the long-term field performance of NDOT projects with

techniques to reduce the impact of reflective cracking on HMA overlays, the following

general conclusions can be made.

• Cold in-place recycling (CIR) of minimum 2” of the existing HMA materials and
overlaying it with a minimum 2.5” dense graded HMA mixture with a traffic level
up to 14,000 AADT provided an effective mean of stopping reflective cracking
 78

for 5 years after construction as long as the recycled pavement does not show any
severe alligator cracking.
• The reinforced fabrics (RF) treatment retarded reflective cracking for at least 3
years after construction and reduced the rate of reflected transverse cracks 5 years
after construction. The RF treatment was ineffective when applied on a pavement
with severe alligator cracking prior to the application of the treatment.
• The 1” stress relief course (SRC) treatment showed excellent performance up to 3
years after construction regardless of the traffic level (up to 40,000 AADT) and
the pre-rehabilitation condition of the pavement. However, the rate of the
reflected transverse cracks was accelerated 5 years after construction.
• The mill and overlay (MOL) treatment by cold milling 1” of the existing HMA
pavement and overlaying it by a 1” HMA mixture manufactured with an AC-10
asphalt binder was ineffective in resisting reflective cracking of pavements with
severe alligator cracking prior to the treatment construction.
• The mill and overlay (MOL) treatment by cold milling 1” of the existing HMA
pavement and overlaying it by a 1” HMA mixture manufactured with an AC-20P
polymer modified binder provided an effective mean of stopping reflective
cracking for 3 years after construction and a traffic level up to 4000 AADT. After
5 years of construction the treatment showed minor reflected transverse cracks.
• The mill and overlay (MOL) treatment by cold milling 1.5” of the existing HMA
pavement and overlaying it by a 1.5” HMA mixture manufactured with an AC-
20P polymer modified binder provided an effective mean of stopping reflective
cracking for 3 years after construction and a traffic level up to 2000 AADT. After
5 years of construction the treatment showed reflected transverse cracks. It
should be emphasized that this treatment was placed on pavements with a
condition worse than the condition of the pavements where the other two mill and
overlay treatments were applied.
• The crack and seat (CS) of concrete pavements with no or minor surface cracks
prior to overlay, showed excellent performance for 3 years after construction.
Fatigue cracking was experienced in the HMA overlay after 4 to 6 years in
service. The 3.5” HMA overlay retarded fatigue cracking for four years after
construction whereas the 4.5” and 4.75” HMA overlay retarded fatigue cracking
for six years after construction.
• The rubblization (RPCC) of concrete pavements with no or minor surface cracks
prior construction and a traffic level of 7000 AADT showed excellent
performance for 3 years after construction. Fatigue cracking was experienced in
the HMA overlay after 4 to 6 years in service. The 5” HMA overlay retarded
fatigue cracking for four to five years after construction whereas the 7” HMA
overlay retarded fatigue cracking for six years after construction.
• In general, the long-term effectiveness of the treatments was significantly
hampered by the existence of severe alligator cracking on the projects prior to the
application of these treatments. Therefore, it is recommended that projects
experiencing severe alligator cracking as classified by the NDOT pavement
distress manual should be subjected to either re-construction or full depth
reclamation.
 79

Chapter 4 – REVIEW OF REFLECTIVE CRACKING ANALYSIS

MODELS
The main objective of this chapter was to review and evaluate the currently

available analytical models that can be used to predict the resistance of HMA overlays to

reflective cracking. Based on the review of the available literatures, three design

methods were identified and are summarized in this chapter.

• Virginia Tech Simplified Overlay Design Model

• Asphalt Rubber Association Overlay Design Model
• The New AASHTO model for Reflective Cracking

The identified models are assessed based on their technical merit and their ability

to predict the performance of HMA overlays subjected to reflective cracking.

4.1 VIRGINIA TECH SIMPLIFIED OVERLAY DESIGN MODEL

In 2003, Elseifi and Al-Qadi (45) developed an overlay design procedure to

predict the service life of rehabilitated flexible pavement structures against reflective

cracking. The researchers used the linear elastic fracture mechanics (LEFM) principles

to derive a simple equation based on three-dimensional (3D) finite element (FE) models

that can be used to predict the number of cycles to failure against reflective cracking for

rehabilitated flexible pavements.

4.1.1 Fundamentals of Reflective Cracking

In 1996, Jacobs et al. (46) evaluated the application of fracture mechanics

principles to the discontinuous crack growth of HMA layers by comparing FE models

simulation (continuous crack growth) to experimental crack growth of HMA

(discontinuous crack growth). The researchers found that crack growth process in HMA
 80

might be accurately described using the fracture mechanics theory with a better accuracy

for mixes with small aggregate size.

In general, a cracked pavement system can be loaded in any one or a combination

of the three following fracture modes (45).

• Mode I loading (opening mode, KI) also called strain mode, results from loads
that are applied normally to the crack plane (thermal and traffic loading) (Figure
4.1.a).
• Mode II loading (sliding mode, KII) also called vertical shear strength mode,
results from in-plane shear loading, which leads to crack faces sliding against
each other normal to the leading edge of the crack (traffic loading) (Figure 4.1.b).
• Mode III loading (tearing mode, KIII) also called horizontal shear strength mode,
results from out-of-plane shear loading, which causes sliding of the crack faces
parallel to the crack leading edge (Figure 4.1.c). This mode of loading is
negligible for pavements.

By neglecting the ultimate failure stage in which the crack growth rate increases

rapidly as global instability is approached, two distinct phases are considered in the

cracking process of pavement systems and are described as follows (45).

The crack initiation phase: This phase consists of two distinct phases of

microcracking and formation of macrocracks, and is defined by the necessary number of

load applications to form a visible damaged zone at the bottom of the overlay (47). In

case of reflective cracking induced by Mode II loading, the number of cycles of a specific

load for crack initiation in the HMA layer may be determined using the Belgian Road

Research Center (BRCC) equation (48).

−4.73
N = 4.856 × 10 −16 ε zx (4.1)

where, N = number of cycles before crack initiation

εzx = shear strains 0.4 inch (10 mm) above the existing crack.
 81

The crack propagation phase: This phase represents the stage where the crack

propagates to the surface through the entire thickness of the HMA overlay. The Paris

power law may be used to describe the crack propagation phase in flexible pavements

(45, 49).

dC
= A(ΔK n ) (4.2)
dN

where, C = crack length

N = number of loading cycles
A and n = fracture parameters of the material
ΔK = stress intensity factor amplitude.

Elseifi and Al-Qadi (45) used the Paris’ law approach in their study which is only

valid under the following assumptions:

• Since no exact definition of the stress intensity factor for a multi-layer pavement
system was available, a regression analysis was performed to define the stress
intensity factor as a function of the crack length (C) for each considered design.
The developed regression models are dependent on the assumed stiffness for the
pavement layers and the crack propagation resistance of the mix.
• The computed number of cycles is highly sensitive to the assumed values of the
fracture parameters (i.e., A and n). The correct way to determine the fracture
parameters of a material is to examine the stable crack growth of HMA beam
samples under repeated loading conditions, which is a tedious and expensive
operation.
• Since no direct measurements of the fracture parameters (A and n) were feasible
in this study, and since such testing is not expected to be conducted in a routine
overlay design, theoretical relations between the fracture parameters of the
material and the mix properties were suggested. The first fracture parameter (n)
of the material was related to its creep properties and the second fracture
parameter (A) was determined by means of the volumetrics and modulus of the
mix (Equation 4.3).

2
n= (4.3a)
m

(4.3b)
 82

where, m = slope of the log creep compliance versus log time curve
AV = air-voids (%)
AC = asphalt content (%)
E = resilient modulus of the mixture (in psi).

Based on this concept, and using results of conventional creep testing performed

on three types of HMA mixtures, three levels were established for the fracture

characteristics of the overlay:

• Type I: n = 3.40 and A = 1.37×10-6 (rich-binder mix)

• Type II: n = 3.85 and A = 1.67×10-6
• Type III: n = 4.50 and A = 2.33×10-6 (high air-voids mix)

The stress intensity factor (ΔK) was determined using FE for different locations of

the crack. The following section explains the stress intensity factor concept, and how the

stress intensity factor was determined using FE.

4.1.2 Fracture Mechanics Analysis Using Finite Element Method

The stress intensity factor (K) was utilized to measure the severity and stability of

a crack in a pavement layer. The stress intensity factor is a scale factor that is used to

define the magnitude of the crack tip stress field and is predicted using Equation 4.4.

K = σ π a × f (a ) (4.4)

where, σ = boundary stress

a = crack length
= geometry depending function.

A given material can resist propagation of a crack as long as the stress intensity

factor is below the fracture toughness (Kc) of the material. In contrast to the stress
 83

intensity factor, the fracture toughness is a quantity that is independent of the crack

geometry, and the loading imposed on the structure. Although the fracture toughness has

not been well defined for HMA, in 1997 Mobasher et al. (50) reported the plane strain

fracture toughness (KIc) for different HMA mixes and at different binder contents as

shown in Table 4.1.

Elseifi and Al-Qadi (45) used the commercial software ABAQUS 5.8-1 to

indirectly calculate the stress intensity factor using the path independent integral, called

J-integral. The J-integral is defined as the change in mechanical energy per unit area of

new crack surface. For plane strain condition, the J-integral is given by Equation 4.5

which is function of the Poisson’s ratio (ν), elastic modulus (E), and stress intensity

factor (K) of the HMA (51).

1 −ν 2
J= (K 2 ) (4.5)
E

The developed FE models represented a variety of three-layer systems that are

regularly encountered in typical HMA pavement overlay applications. A total of 216

different pavement designs were analyzed to develop the suggested design equation. The

pavement structure consisted of an HMA overlay with variable thicknesses (2 to 6

inches) applied on top of a cracked HMA layer, a base layer, and a subgrade. The crack

initiation and propagation phases were investigated using dynamic three-dimensional

models developed for different location of the cracks. Elastic foundations were used to

simulate the support provided by the subgrade (non-linear springs). The contact between

layers was assumed as fully-bonded. A crack was induced in the existing HMA layer and

the pavement structure was subjected to a quasi-static load with an amplitude function
 84

that was obtained based on actual vertical stress measurements in the flexible pavement

at the Virginia smart Road.

4.1.3 Development of Design Equations

The total number of load repetitions to produce the crack reflection to the

pavement surface was defined as the sum of the number of load repetitions for crack

initiation and the number of load repetitions for crack propagation. Global instability in

the HMA layer was assumed to be reached when the crack front is at 0.5 inch (12.7 mm)

from the pavement surface.

Ntotal = Ninitiation + Npropagation (4.6)

where, Ntotal = total number of cycles before the crack reach 0.5 inch (12.7 mm) from the
surface of the overlay
Ninitiation = Number of cycles for crack initiation at the bottom of the overlay.
Npropagation = Number of cycles for the crack to propagate from the bottom of the
overlay to 0.5 inch (12.7 mm) from the surface of the overlay.

The number of cycles for crack initiation (Ninitiation) was determined for each of

the pavement structures considered in this study using Equation 4.1 and FE to calculate

the maximum shear strain in the vicinity of the crack tip. A shift factor of nine was used

in all calculations to adjust the number of cycles for the difference between laboratory

and field conditions (e.g., load distribution in the wander area, state of stresses, and rest

periods).

The number of cycles for crack propagation (Npropagation) was determined using

Equation 4.2. The stress intensity factor (K) was computed using Equation 4.5 and the J-

inetgral that was obtained using FE for different locations of the crack in the HMA

overlay.

4.1.4 Service Life Prediction for Pavements with Potential Reflective Cracking
 85

In order to avoid the time consuming FE analyses, the researchers developed a

regression model to predict the number of cycles in ESALs as a function of the

significant variables. The developed design equation was based on the results of all the

considered cases in this study.

log Wt 80 =
1
(255H overlay + 2.08Eoverlay + 45.3H HMA + 8.73EHMA )
104
1.37 H Base + 6.93EBase + 1.49 Esubgrade ) (4.7)

where, Wt80 = total number of 80-kN single-axle load applications

Hoverlay = thickness of HMA overlay (mm)
Eoverlay = resilient modulus of HMA overlay (MPa)
HHMA = thickness of existing HMA layer (mm)
EHMA = resilient modulus of existing HMA layer (MPa)
Hbase = thickness of base layer (mm)
Ebase = resilient modulus of base layer (MPa)
Esubgrade = resilient modulus of subgrade (MPa).

The interaction between the different variables was also considered, but was

found statistically insignificant. It is interesting to notice from this equation that the

overlay thickness is undoubtedly the major factor in dictating the overlay performance

against reflective cracking failure, followed by the thickness of the existing HMA layer.

Additionally, it appears that the base thickness and subgrade modulus has the least effect

on the overlay performance.

Although this model was exclusively developed based on a variety of three-layer

pavement structures, it may be easily extended to cover a broader number of cases using

the Odemark’s method. For example, if a subbase was used, it may be converted to an

equivalent base thickness using Equation 4.8 as follows (52).

 86

( )
⎡ Esub 1 − ν 2base ⎤
he = hsub ⎢
1/ 3

( )
⎣ Ebase 1 − ν sub ⎦
2 ⎥ (4.8)

where, he = equivalent base thickness

hsub = original subbase thickness
νbase = base material Poisson’s ratio
νsubbase = base material Poisson’s ratio.

4.1.5 Design Example (45)

Consider a pavement structure consisting of a 150 mm HMA layer (EHMA = 2500

MPa, ν = 0.25), 150 mm base layer (Ebase = 275 MPa, ν = 0.35), 300 mm subbase layer

(Esubbase = 135 MPa, ν = 0.35), on top of a subgrade (Esubgrade = 68 MPa, ν = 0.40). Using

the Odemark’s method, the subbase layer may be transformed to an equivalent base layer

resulting in a total base thickness of 385 mm. The designer estimates a total

accumulative traffic (80-kN repetitions) of 9.0×105 over the service life of the overlay.

Using Equation 4.7, the required overlay thickness is 75 mm (Eoverlay = 4480 MPa, ν =

0.25) to sustain the estimated traffic. Depending on fund availability, the designer may

then decide to use a multi-stage construction strategy, a reinforcing interlayer system, or

any other rehabilitation alternatives to maximize the benefit-cost ratio.

4.2 ASPHALT RUBBER ASSOCIATION OVERLAY DESIGN MODEL

In 1999, the Rubber Pavements Association (RPA) contracted with Consulpav to

develop a mechanistic overlay design method for reflective cracking in HMA overlays

that are applied to existing cracked HMA pavements (53). The research project involved

the development of mathematical and statistical models based upon 3D finite element

method (FEM) to determine the stresses and strains in the HMA overlay above the crack.

The FEM approach was selected since it is the most sensible way to address the unusual
 87

stress and strain patterns and contours generated by a truck load moving over an overlay

placed upon a cracked pavement. The SAP2000 finite element program with linear

elastic ASOLID element under the plain strain mode was used in the analyses to model

the subgrade layer, the base layer, the cracked bituminous layer, the HMA overlay, and

the crack and the zone above the crack. The FEM modeled crack movements were

calibrated using actual field measurements with a crack activity meter (CAM) and a

Falling Weight Deflectometer (FWD) conducted in Portugal, Arizona and California.

The vast majority of this field testing was conducted on cracked Interstate pavements in

Arizona. The results of these experiments helped to calibrate the FEM statistically

derived model and showed that the CAM correlated very well crack movements and

deflections produced by the FWD in the pavement. The “Von Misses strain” was

recommended as the statistical factor to determine the development of the crack.

The next phase of this research involved laboratory testing that simulated the

observed field crack movements and measured stresses and strains on test specimens

similar to the actual field mixes. Two typical mixes using a good quality Portuguese

granite aggregate similar to Watsonville granite were prepared and tested in Portugal.

One mix consisted of a dense gradation like that used in Arizona and California. Five

percent PG70-10 asphalt binder was added to the dense graded aggregate and compacted

using the rolling wheel compactor to various levels of air voids (HMA-DG). The second

mix (AR-HMA-GG) consisted of an eight percent asphalt rubber (AR) gap graded mix

similar to that used in Arizona and California. The AR binder consisted of 80% PG64-16

(Penetration 35/50) base asphalt and 20% ground tire rubber.

These two HMA mixes represented the basis of the mechanistic design method.
 88

Cores and beams were sawed from the rolling wheel compacted slabs. The six inch (150

mm) diameter by two inch (50 mm) high cores were tested with a Reflective Crack

Testing device (RCD) specifically built for this research. The RCD is composed by two

U-shaped pieces where the specimen is fitted. The base of the device represents the

existing pavement and the opening between the U-shaped pieces represents the crack

width. The bottom of the specimen is glued to the base of the device and pressed by

pistons existing at the top of the device in such a way that either a horizontal or vertical

movement can be imposed on the core. Sensors on the core record stresses and strains of

the core immediately over the crack.

Figures 4.2 and 4.3 show the RCD with each of the U-shaped pieces connected to

a horizontal and a vertical actuator to apply displacements to the specimen. The device

was used to apply the displacements measured in the road with the crack activity meter.

The RCD simulates the zone above the existing crack of the old pavement which is the

zone of the overlay layer that is subjected to reflective cracking (Figure 4.4).

In addition, beams from both mixtures were tested with the four point bending

beam fatigue test developed during the Strategic Highway Research Program (SHRP).

Results of these tests indicated that the beam fatigue test could be used in place of the

reflective crack test (RCD) to derive the necessary input parameters to the FEM model.

The data and test results from the two phases of this research served in the

determination of the parameters that best fit the FEM reflective cracking statistical

simulation as a function of truck traffic. The model predicts how many truck loading and

their corresponding stresses and strains will be needed to bring about a reflective crack.

This mathematical-statistical model was converted into a practical pavement

 89

design method for reflective cracking by reviewing considerable actual field cracking

data and material layer properties. First, the estimated traffic to a reflective crack was

calculated from the layer thickness and layer modulus and then compared to the actual

observed number of truck loading and percent cracking. As long as the ratio of the

estimated traffic to the actual traffic stays below 1.0 no cracking would occur. For ratios

above 1.0 the different levels of percent cracking would be observed. Aging and

temperature adjustment factors were also introduced to this approach.

The final product of this research was an EXCEL spreadsheet that estimates the

thickness of a PG70-10 or an AR HMA overlay mix for the specified level of reflective

cracking for a wide range of truck traffic loading. The expected design level of cracking,

the thicknesses and the elastic moduli of the existing pavement layers, and the modulus

of the HMA overlay are inputs for the EXCEL spreadsheet. The moduli may be

backcalculated or determined in any reasonable manner, as long as they represent the in

situ conditions in the field.

The proposed solution is reliable for these two analyzed mixes (i.e., PG70-10

HMA and AR-HMA) and for climatic conditions similar to those encountered in the

(mainly) desert southwest (i.e., Arizona and California). With additional research, curves

could be developed for additional mixes, other climates and other field observed

historical reflective cracking levels.

4.2.1 Proposed Reflective Cracking Design Method

The method consists of the seven steps presented below (53). Currently the

model has been calibrated for only two materials: Dense graded mixes with PG70-10

binders (HMA-DG) and gap graded mixes with asphalt rubber modified binders (AR-
 90

HMA-GG). The asphalt rubber binder must be produced using the “wet” process and it

must contain approximately 19-20% crumb rubber.

Step 1: Determination of the moduli and thicknesses of the pavement section layers

This can be accomplished using FWD backcalculation methods or other forms of

estimating cracked pavement section moduli. Care must be taken in the selection of

modulus representative of the most damaged sections. As such, the 90th or 95th

percentile of deflections (or backcalculated moduli) should be selected. Coring for

determination of layer thicknesses should be carried out as close to the locations where

the 90-95th percentile FWD test points were selected.

Step 2: Determination of representative air temperatures

The maximum and minimum air temperature determined with the desired

reliability should be obtained for the location where the pavement is to be overlaid.

Furthermore, it is necessary to compute the weighted mean annual air temperature (w-

MAAT) as proposed by the Shell design method and provided by Equation 4.9 (54).

w − MAAT = 7.7068 × ln(w − factor ) + 20.275 (4.9)

where the w-factor is a function of the mean monthly air temperature (MMAT) and can

be determined using Shell Equation 4.10. The w-factor in Equation 4.9 is the average of

w-factors calculated for all 12 months of the year.

w − factor = 0.0723 × e 0.1296 xMMAT (4.10)

Step 3: Selection of design cracking percentage

The design cracking percentage value selected should be in keeping with an

 91

agencies overlay policy. ADOT generally has observed less than five percent cracking

over a period of ten years when an asphalt rubber surface mix is used.

Step 4: Determination of adjustment factors

Several adjustment factors must be calculated for the location where the overlay

will be placed and for the desired cracking level at the end of the overlay’s design life.

Aging Adjustment Factor

The Aging Adjustment Factor (AAF) is determined from the maximum air

temperature (Tmax) using Equation 4.11 for conventional PG70-10 mixes (i.e., HMA-DG),

and using Equation 4.12 for asphalt rubber mixed with the gap-graded wet process at 19 –

20% crumb binder (i.e., AR-HMA-GG).

AAF = 0.0363 × T(max air ) + 0.3000 (4.11)

AAF = 0.0088 × T(max air ) + 0.8800 (4.12)

The AAF was introduced to capture the effect of aging in the overlay as a function

of the maximum air temperature. Aging will stiffen the HMA overlay as a function of

time and temperature. The above relationships are applicable for maximum air

temperatures (Tmax) between 35 and 50°C. Further improvements in the relationships are

still needed.

Temperature Adjustment Factor

The temperature adjustment factor was included in the analysis to take into

consideration the combined action of the two important variables in reflective cracking:

the wheel loads on a daily basis above (or near) the crack and the overlay material above
 92

the crack being under tension due to rapidly decreasing or low temperatures. Those

factors have been identified as the most likely causes of high states of stress and strain

above the crack.

The Temperature Adjustment Factor (TAF) is determined from Equation 4.13 for

HMA-DG mixes and Equation 4.14 for AR-HMA-GG mixes (wet process only). This

method is applicable for Reflective Cracking Temperatures (RCTs) between –10°C and

+10°C. The RCTs corresponds to the typical range of temperatures when the reflective

cracking is expected to occur.

TAF = −0.09 × RCT + 2.55 (4.13)

TAF = −0.072 × RCT + 1.7448 (4.14)

RCT = T(min air ) + 0.5 × (T(average air mean monthly ) − T(min air ) ) (4.15)

Field Adjustment Factor

The Field Adjustment Factor (FAF) was introduced to relate the predictions

obtained using the empirical-mechanistic reflective cracking model with the actual

(reported and observed) field performance. The FAF is computed from the actual percent

cracking (PC) using Equation 4.16. Cracking is only expected when FAF is greater than

1.

FAF = e 0.2303×PC (4.16)

Step 5: Selection of overlay material modulus

Two types of materials for the overlay may be selected: Conventional HMA-DG

or AR-HMA-GG, with the rubberized binder prepared through the wet process at 19-20%
 93

crumb rubber in the binder. For these materials, the modulus and flexural fatigue life are

obtained through flexural fatigue tests (AASHTO T321). Other moduli can be computed

and introduced in the method based on actual tests performed on other types of materials.

However, it must be assumed that the Temperature Adjustment Factor (TAF) or the

Aging Adjustment Factor (AAF) will either be identical to that of the HMA-DG material

or the AR-HMA-GG material.

Step 6: Determination of the design value “Von Mises” strain

The modulus of the overlay must be multiplied by the computed Aging

Adjustment Factor. Then, the modulus of the existing and overlay HMA must be

adjusted to the w-MAAT temperature. With the modulus and thickness for each layer, the

“Von Mises” strain (εVM) value for the overlay is determined using Equations 4.17, 4.18,

and 4.19. The εVM value obtained through these equations must be multiplied by 86/132

to convert εVM from 130 kN to 80 kN axle loads, and also by the Temperature Adjustment

Factor (TAF). Such factor was included because the “Von Mises” Strain was originally

computed for 130 kN axles, and it was then adjusted proportionally for the 86 kN axles.

The value obtained in this process is thus the design εVM.

ε VM (1x10 −6 ) = a[Overlay thickness(m)]b (4.17)

6
a = ∏ [a1i * ln( X i ) + a2i ] (4.18)
i =6

6
b = ∏ [b1i * ln( X i ) + b2i ] (4.19)
i =6

where εVM is in 1×10-6 m/m, overlay thickness in meter, a1i and b1i are coefficients given
 94

in Table 4.2, and the variables Xi represent the pavement properties of thicknesses and

moduli with i = 1 for thickness of the cracked HMA; i = 2 for thickness of the granular

base; i = 3 for HMA overlay modulus; i = 4 for cracked HMA modulus; i = 5 for granular

base modulus; and i = 6 for subgrade modulus. Table 4.3 shows the order introduced for

the Xi values along with the minimum and maximum limits for each variable used in the

definition of the model.

Step 7: Determination of design equivalent single axle load (ESAL’s)

The last step of this process, determines the number of ESALs that can be

withstood by the overlay prior to the onset of reflective cracking using the flexural

fatigue Equation 4.20 for the HMA-DG mix with PG70-10 binder or Equation 4.21 for

the AR-HMA-GG mix with the asphalt rubber binder binders derived through the wet

process, with a 19-20% crumb rubber. εVM is in 1×10-6 m/m.

ESALs = 4.1245 × 1019 × ε VM (1 × 10 −6 ) 4.9761 (4.20)

ESALs = 6.4467 × 1019 × ε VM (1 × 10 −6 ) −5.93 (4.21)

The design ESALs need to be multiplied by the FAF computed in Step 4. The

resulting number should represent the number of ESALs required for the overlay to reach

the selected percentage of cracking. Other fatigue curves can be determined and used by

this method, based on actual flexural fatigue tests performed on the specific asphalt

(whether conventional or modified) material type proposed with due consideration to all

adjustment factors. Figure 4.5 shows a flowchart of the proposed design method.

4.3 THE NEW AASHTO MODEL FOR REFLECTIVE CRACKING

In the new AASHTO Mechanistic-Empirical Pavement Design Guide (MEPDG),

 95

the design procedure for HMA overlays of existing HMA surfaced pavements considers

distresses developing in the overlay as well as the continuation of damage in the existing

pavement structure. However, it should be noted that the reflective cracking models

incorporated in the MEPDG were based strictly on empirical observations and were not a

result of rigorous mechanistic-empirical analysis (55).

An active NCHRP 01-41 research study is undergoing to address the issues

associated with reflection cracking and to develop mechanistic-based models for use in

mechanistic-empirical procedures for the analysis and design of HMA overlays. The

objective of the research is to identify or develop mechanistic-based models for

predicting reflection cracking in HMA overlays of flexible and rigid pavements and

associated computational software for use in mechanistic-empirical procedures for

overlay design and analysis.

The proposed MEPDG overlay design procedure allows the designer to consider

two types of reflective cracks: a) reflective cracks that exist on the surface prior to

overlay placement and b) those that develop in the existing surface after overlay

placement.

4.3.1 HMA Overlay of Cracked HMA Pavement Surface

The percentage of reflective cracks through the overlay is predicted as a function

of time using the sigmoidal function shown in Equation 4.21.

100
RC = (4.21)
1 + e a + bt

where, RC = Percent of cracks reflected

t = Time in years
 96

a = 3.5 + 0.75×hac
b = – 0.688584 – 3.37302×(hac)-0.915469
hac = HMA overlay thickness in inches.

The a and b fitting parameters are function of the HMA overlay thickness and are

hard coded in the MEPDG software. The designer cannot directly alter these parameters

as inputs, but can change them in the software. It is recommended that an agency use

historical data to develop a local reflection cracking model.

Additionally, the MEPDG approach assumes that a properly installed fabric is

equivalent to 2 inch of HMA overlay. This is purely based on empirical considerations.

A Minimum of 2 inches is recommended for the HMA overlay thickness (hac) of flexible

pavements.

4.3.2 HMA Overlay of Crack-free HMA Pavement Surface

Even after overlay placement, the underlying bound layers (including all asphalt

bound and chemically stabilized layers) undergo additional fatigue damage with

continued traffic loading, and will eventually crack. The continual fatigue damage

accumulation of these layers is considered in the overlay analysis procedures of the

MEPDG. For any given month m, the total fatigue damage is estimated by Equation

4.22.

m
Dm = ∑ ΔDi (4.22)
i =1

where, Dm = Damage for month m

ΔDi = Increment of damage in month i.

The area of fatigue cracking for the underlying layer at month m (CAm) is given
 97

by Equation 4.23.

100
CAm = (4.23)
1 + e 6−6× Dm

For each month i, there will be an increment of damage ΔDi which will cause an

increment of cracking area ΔCAi to the stabilized layer. To estimate the amount of

cracking reflected from the stabilized layer to the surface of the pavement for month m,

the reflective cracking model is applied incrementally, as follows:

m
TRAm = ∑ RC m −i × ΔCAi (4.24)
i =1

where, TRA = Total reflected area for month m

RCm-i = Percent cracking reflected for age m – i (age in years)
ΔCAi = Increment of fatigue cracking for month i.

The reflective cracking model is applied to each increment of fatigue cracking

area because the time elapsed for each of these increments is different. The model

included in the MEPDG is based on engineering judgment and a limited amount of

published data.

4.4 DESIGN EXAMPLE USING THE VARIOUS ANALYSIS MODELS

In this example, an HMA overlay was designed for a typical flexible pavement

section (Figure 4.6) using the various analysis models described before. The existing

pavement structure consists of a 4.0 inch (100 mm) HMA layer with a modulus of 360

ksi (EHMA = 2500 MPa) and a 10 inch (250 mm) base layer with a modulus of 30 ksi

(Ebase = 210 MPa) on top of a subgrade with a modulus of 12 ksi (Esubgrade = 83 MPa).

4.4.1 Traffic Data

 98

The traffic ESALs are estimated as function of years for an annual average daily

traffic (AADT) of 30,000 and a truck percentage of 3.84 percent. The traffic information

was obtained from the NDOT 2006 annual traffic report for the McCarran Boulevard in

Reno Nevada area. Table 4.4 summarizes the design ESALs over the 20 years analysis

period.

4.4.2 HMA Overlay Material Properties

The analysis was conducted for three different HMA mixes that were designed using

different aggregate gradations from the Sloan pit in Southern Nevada and a PG76-22NV

polymer modified asphalt binder. The three mixtures consisted of:

• NDOT Type 2C gradation, designated as T2C.

• Caltrans gradation used for intersection, designated as CT.
• No Rut Mixture gradation, designated as NRM.

All three mixes were designed according to the NDOT Hveem Mix Design Method as

outlined in the NDOT Testing Manual. It should be noted that all mixtures were treated

with 1.5% of hydrated lime following the NDOT specifications. The dynamic modulus

test (AASHTO TP62) was used to develop the dynamic modulus master curve of the

various HMA mixtures. Table 4.5 shows the dynamic modulus (Eoverlay) of the various

overlay mixtures at a temperature of 70°F and a loading frequency of 10 Hz.

The fatigue characteristics of the HMA mixtures were evaluated using the flexural

beam fatigue test “AASHTO T321-03: Determining the Fatigue Life of Compacted Hot-

Mix Asphalt Subjected to Repeated Flexural Bending”. The following equations show

the fatigue relationship for each of the evaluated mixtures.

 99

2.0374
⎛1⎞
4.4142
⎛ 1 ⎞
−5
T2C: N f = 1.3740 × 10 × ⎜ ⎟ ⎜ ⎟ (4.25)
⎝ε ⎠ ⎜E ⎟
⎝ overlay ⎠

2.0598
⎛1⎞
4.3172
⎛ 1 ⎞
CT: −5
N f = 6.5815 × 10 × ⎜ ⎟ ⎜ ⎟ (4.26)
⎝ε ⎠ ⎜E ⎟
⎝ overlay ⎠

2.1470
⎛1⎞
4.2395
⎛ 1 ⎞
−5
NRM: N f = 6.8745 × 10 × ⎜ ⎟ ⎜ ⎟ 4.27)
⎝ε ⎠ ⎜E ⎟
⎝ overlay ⎠

where, ε is the flexural strain in microns and EOverlay is the stiffness of the HMA mixture

in ksi.

4.4.3 Overlay Design Using Virginia Tech Simplified Overlay Model

Using Equation 4.7, the required HMA overlay thickness as a function of the

predicted traffic (ESALs) is shown in Figure 4.7 for the various mixtures. Based on the

Virginia Tech Simplified Overlay Design method an overlay thickness of 3.25, 4.0, and

4.5 inch is required for the NRM, CT, and T2C mixes to sustain the predicted 20 years

design ESAls, respectively. Therefore, the T2C mix exhibited relatively the best

resistance to reflective cracking, followed by the CT mix, followed by the NRM mix.

4.4.4 Overlay Design Using Asphalt Rubber Association Method

Currently the AR model has been calibrated for only two materials: dense graded

mixes with PG70-10 binders (HMA-DG) or gap graded mixes with asphalt rubber

modified binders (AR-HMA-GG). However, still the AR model can be used with the

overlay mixture specific fatigue characteristics. Therefore, the various mixtures fatigue

curves shown in Equations 4.25-4.27 are used in this method to estimate the traffic in

ESALs required for the overlay to reach a selected percentage of cracking.

 100

The maximum, minimum, and mean monthly air temperatures are determined for

the Reno area using the LTPPBind (version 3.1) software. A temperature of -16°F (-

26.5°C) and 103°F (39.3°C) were determined for the minimum and maximum air

temperature where the existing pavement is to be overlaid, respectively. The mean

monthly air temperature was found to be 50°F (10°C).

The analysis was conducted for 5, 10, and 15 percent of reflective cracking at the

end of the design life. The required HMA overlay thickness as a function of the

predicted traffic (ESALs) is shown in Figures 4.8 to 4.10 for the various mixtures.

Figures 4.8-4.10 show that, for the same design ESALs, a thinner overlay

thickness is required for the T2C mix followed by the CT mix followed by the NRM mix

to reach the same selected percentage of cracking.

An additional analysis was conducted to determine the percentage of cracking in

each mix using the previously determined corresponding overlay thickness after 20 years

using the Virginia tech method. In other words, the percentage of cracking in the overlay

mix was obtained for an overlay thickness of 4.5, 4.0, and 3.25 inches for the T2C, CT,

and NRM mixes, respectively.

Figure 4.11 shows that, the 4.5 inch T2C overlay mix exhibits no cracks after 20

years, the 4.0 inch CT overlay mix exhibits 3.5 percent cracking after 20 years, and the

3.25 inch NRM overlay mix exhibits 17 percent cracking after 20 years.

4.4.5 Overlay Design Using the New AASHTO Model for Reflective Cracking

The percentage of reflected cracks (RC) from a cracked HMA pavement surface

through the overlay is predicted as a function of time using the sigmoidal function shown

in Equation 4.21. The AASHTO method was applied for a 5, 10, 15, and 20 percent of
 101

reflected cracks. Table 4.6 and Figure 4.12 show the required overlay thickness as a

function of predicted traffic ESALs. It should be noted that the new AASHTO method is

independent of the overlay material properties; hence the same overlay thickness is

required for all three mixes (i.e., T2C, CT, and NRM).

It should be noted that RC is the percent of cracks reflected from the existing

cracked pavement surface. Therefore, the time for the cracks in the existing pavement to

be 100% reflected in the overlay was also determined. Figure 4.13 shows the required

overlay thickness as a function of traffic for a 100% reflection of the cracks in the

existing pavement. An overlay thickness of 11.25 inch is required to sustain the

predicted 20 years design ESAls (i.e., 7,074,390). It should be noted that, the new

AASHTO design manual recommend the development of a local reflection cracking

model using historical data.

4.4.6 Summary of Design Example

Figures 4.14-4.16 compares the required overlay thickness determined from the

Virginia Tech, asphalt rubber, and the new AASHTO analysis methods. The Virginia

tech and the asphalt rubber analysis models resulted in a relatively comparable overlay

design thicknesses with the Virginia tech method being more conservative. On the other

hand, the AASHTO method overestimated the overlay thickness compared to the

Virginia Tech and the asphalt rubber methods.

 102

Chapter 5 – REVIEW OF REFLECTIVE CRACKING TESTING

METHODS
This chapter reviews all the laboratory and field tests that are being used to

evaluate the resistance of HMA mixtures and pavements to reflective cracking.

Laboratory tests are typically used to evaluate the resistance of the HMA mixtures to

reflective cracking during the mix design stage. Field tests are used to evaluate the

performance of the rehabilitation technique used to minimize reflective cracking in HMA

pavements. In this chapter, the appropriate laboratory and field tests will be identified

and assess their potential effectiveness in NDOT’s mix design and field evaluation

processes. Based on these evaluations, the most promising tests will be recommended.

5.1 CRACOW UNIVERSITY OF TECHNOLOGY, CRACOW, POLAND

In 1993, the University of Technology of Cracow, Poland, evaluated the

effectiveness of geosynthetics in HMA overlays for preventing reflective cracking (56).

The reflective cracking resistance of an HMA overlay reinforced with a geogrid was

assessed with the following three laboratory tests: bending test under a static load,

bending test under a repeated load, and the shearing test.

The bending test under a static load was performed to study the resistance of an

HMA overlay with and without geotextile reinforcement at the bottom of the HMA under

a static load at a constant rate of application. The setup and dimensions of the tested

HMA beams are presented in Figure 5.1. The test samples consist of a 12×3×3 inch

(30×7.5×7.5 cm) HMA beams compacted in a static manner. The geotextile is placed at

1/3 of the beam height from the bottom. The beam is simply supported at both ends and
 103

loaded uniaxially at the center at a rate of 0.47 in/min (12 mm/min). The samples were

tested at a temperature of 20°C.

The test results for the bending test under static load are shown in Figure 5.2. The

maximum measured applied force along with the corresponding bending deflection of the

beam are used to calculate the bending strength. It was concluded that the bending

strength of HMA mixtures was not substantially affected by the presence of the

geotextile. The geotextile influenced the behavior of the samples after cracking of the

HMA mix by prolonging the integrity of the beams.

The bending test under repeated load was performed to analyze a loading pattern

similar to field loading. Tests were performed at 20°C and on beams with the same setup

and dimensions of the bending under static load test. The samples were simply supported

at both ends and subjected to the vertical repeated loading using a haversine function with

a frequency of 5 Hz and an amplitude of 450 lbs (2 kN) resulting in a bending stress at

the outer fiber of 193 psi (1.33 MPa). Measurements of the applied force versus the

beam deflection were conduced until the cracking of the HMA mix. Table 5.1

summarizes the number of load repetitions to visible crack in the HMA beams with and

without geotextile. Researchers concluded that the HMA samples with geotextile

exhibited a substantially greater resistance to the crack development than the samples

without geotextile (56). Due to the presence of the geotextile, the HMA layer sustained

nearly twice the number of loading cycles before the crack appears.

The shearing tests was performed to study the role of geotextile in HMA overlays

as a stress-relieving medium diminishing the stress concentrations above the crack in the

lower part of a pavement structure. This approach required the examination and
 104

qualitative estimation of the adhesion between the geotextile and the HMA layers. To

this end, a method of direct shear test was developed at Cracow University of

Technology (56). The samples, in the form of beams were cut into pieces of a length 2.7

to 3.1 inch (68 to 80 mm) and subjected to shear stresses as shown in Figure 5.3. This

test consisted of applying a static force on the sample at a loading rate of 4×10-2 in/min (1

mm/min). The test was conducted at a temperature of 30°C. At failure, the maximum

shear force is determined and used to calculate the maximum shear stress. Table 5.2

summarizes the maximum shear force and stress in the HMA beams with and without

geotextile.

The results for the shearing test showed that the presence of the geotextile

diminished by more than two times the maximum shear force and the maximum shear

stress acting over the HMA sample. This factor is highly significant in reducing the

susceptibility to reflective cracking of the HMA overlay.

The results for the shearing test showed that the presence of a geotextile interlayer

diminishes by more than 2 times the adhesion between asphalt layers. This phenomenon

is advantageous for the prevention of reflective cracking because the geotextile absorbs

the part of the crack energy from the lower cracked layers and does not transfer it

upwards.

5.2 TECHNION-ISRAEL INSTITUTE OF TECHNOLOGY, HAIFA, ISRAEL

In 1993, Livneh et al. (57) used the wheel tracking device to predict the behavior

of asphaltic pre-coated geotextile felts in retarding reflective cracking. The wheel

tracking device consisted of a loaded wheel that travels back and forth on top of an HMA

beam that is placed over an elastic rubber base simulating the mode I type of fracture.
 105

The Mode I loading (opening mode, KI) is also called strain mode and results from loads

that are applied normally to the crack plane (thermal and traffic loading). The HMA

beam had a rectangular groove with a starter notch at the bottom of the groove. Figure

5.4 shows a schematic diagram of the wheel tracking device. The crack length along

both sides of the HMA slab is visually determined. The device was located in a

temperature-controlled room at a constant temperature of 25°C. The tested HMA beam is

27.6 inch long, 4 inch wide and 4 inch thick (70×10×10 cm).

Three different types of beams were prepared for all tests and are shown in Figure

5.5: 1) beam composed of two HMA layers with a welded fabric in between, 2) monolith

HMA beam, and 3) beam composed of two HMA layers bonded with a tack-coat. An

artificial groove of 1.2 to 2.0 inch (30 to 50 mm) deep and 0.16 inch (4 mm) wide is

made at the center of the bottom of the beam samples. The edge of the groove is

rounded. Neither the magnitude of the applied load nor the frequency was reported by

the Researchers.

For this study, three types of fabrics were examined (3M, 3/250 and 4/180) along

with a monolithic HMA sample and an HMA sample compacted in two layers bonded

with a tack coat at a rate of 250 g/yd2. For all the cases, the HMA was dense graded with

an asphalt grade of 60/70.

In addition to the number of wheel loading repetitions to failure, a phenomenon of

dual cracking was observed in some of the tests. The phenomenon is characterized by the

propagation of a crack which begins at the edge of the artificial groove and the

propagation of a crack which begins on the upper surface. This phenomenon was
 106

observed in half of all beams which did not contain fabric and took place in all beams

which included membranes of all of the various types

Figure 5.6 shows the fatigue test results under the wheel-tracking device. The

results showed that the geotextile fabric type 3/250 significantly outperformed the rest of

the treatments. Its resistance to reflective cracking was about 4 times greater than the

other treatments. However, it was shown that the resistance to reflective cracking of the

monolithic HMA beam and the HMA sample compacted in two layers was greater than

the rest of the geotextile felts (the 3M and the 4/180).

Despite its limitations, the researchers concluded that the wheel-tracking testing

device adopted and improved in this study can serve as a reliable method for diagnosing

the efficiency of the various treatments for reflection cracking retardation.

5.3 GEO-MATERIALS LABORATORY, ENTPE, FRANCE

In 1992, the Geo-materials Laboratory of France conducted a research to study

the reflective cracking in a flexible pavement composed of an HMA overlay placed over

a cement treated base course (58). A prototype apparatus called “Fissurometer” was

proposed to simulate the reflective cracking phenomenon in the laboratory. The

fissurometer allows only for the Mode I type of fracture mechanics failure by simulating

reflective cracking due to thermal shrinkages of the pavement. The test is run in two

steps: monotonic and dynamic modes.

The fissurometer apparatus shown in Figure 5.7 consists of an upper HMA layer

placed on top of two steel plates. One of the steel plates is fixed and the other is moved

apart by two force transducers that apply a fixed opening and/or closing rate and

subjecting the HMA layer to shear loads. The applied load can be either monotonic or
 107

dynamic cycles. The shearing cycles mimic the typical thermal gradients in France. The

shear is provoked by the horizontal displacement of the moving part. The opening and

closing speed of the moving part varies between 0.05 and 0.22 in/hr (1.3 to 5.5 mm/hr).

The measuring system is composed of an ultrasonic transmission (US) system,

force transducers, strain gauges and temperature probes. The force transducers can vary

the speed of opening and closing of the gap below the HMA overlay. The amplitude of

the opening plate can be varied for different test conditions. The US system can measure

the energy transmitted by an ultrasonic wave train applied through the test sample. This

wave train goes through the area where the cracking should appear (near the vertical

section crossing the pre-cracking gap). The force transducers and the US provide a

damage characterization of the HMA overlay. It can be used as an indication of the

cracking resistance of the material.

The proposed apparatus was validated by comparing the results given by the

fissurometer for mixtures with different cracking mitigation techniques to their

performance from test sections built on the French highway RN 20. As a result, the

researchers concluded that the four evaluated techniques were classified with the

fissurometer in reverse order for the top two techniques when compared to their field

observations at the experimental test sections on RN 20. The researchers relate the

difference in resistance to reflective cracking between laboratory and field back to the

basis of the fissurometer apparatus that simulates only thermal shrinkage without

considering the effect of the traffic.

 108

5.4 TECHNICAL UNIVERSITY OF VIENNA, AUSTRIA

In 1993, Tschegg et al. from the Technical University of Vienna, Austria

developed a new testing device to characterize the mode I type of fracture in HMA

mixtures (59). The device is called the wedge splitting method. The test device

measures the load-displacement curve of a cubical or cylindrical HMA sample and

provides information to completely characterize the fracture behavior of the tested

specimens. Crushed and natural gravel HMA mixtures were tested to initiate the

validation of this test method.

The principle of the testing method is shown schematically in Figure 5.7.a.

Cubical (or prismatic) specimens are placed on a narrow linear support in a compression

testing machine. The specimen has a rectangular groove with a starter notch at the

bottom of the groove. Two load transmission pieces are placed in the groove. A wedge

split is formed by inserting a wedge. Wedge, starter notch and linear support area are in

the same vertical plane. The wedge transmits a force FM from testing machine to the

specimen. The slender wedge exerts a large horizontal force component FH and a small

vertical force component FV on the specimen. The horizontal component splits the

specimen similar as in a bending test. The force FM is determined with a load cell in the

testing machine. Knowing the wedge angle (approximately 5-10°), the force FH in the

horizontal direction which causes splitting of the specimen is calculated. The vertical

force component FV is small and does not influence the result if the wedge angle is small

enough. The load application rate is 0.05 in/min (1.3 mm/min). Figure 5.7b shows the

schematic load-deformation curve for a brittle and a ductile material.

 109

Splitting must take place during stable crack propagation until complete

separation of the specimen. This is possible with a stiff loading system only (wedge-

loading system – testing machine). The wedge loading part is extremely stiff, i.e., it

stores little deformation energy. Therefore, such experiments can be performed with

usual mechanical spindle driven machines or hydraulic machines under strain or stroke

controls together with the described loading device.

In order to reduce friction between wedged and the load transmission pieces, roll

bodies (e.g. roller bearings) have been introduced in between. Displacement gauges are

mounted on both ends of the starter notch in order to measure the displacement on the

line of force application (crack opening displacement).

Different specimen shapes may be used in the above described method and are shown

schematically in Figure 5.8. The Force FM and the displacement δ can be registered

during the test using an electronic device. The area under the load-displacement curve

corresponds to the necessary energy to split the sample. Dividing this energy by the

fracture area (plain projection of the area under the applied load), yields the specific

fracture energy of the material (GF). The GF is a property of the material and does not

depend on the specimen shape and size. Figure 5.9 shows the typical three charts

obtained through this test:

• Horizontal force versus displacement curve at any give temperature

• Maximum vertical force versus temperature curve
• Fracture energy versus temperature curve

Characteristic load-deformation diagrams were performed for crushed and natural

gravel HMA mixtures and the results are shown in Figure 5.10. The tests were

performed at three temperatures: 8°C, -0.5°C, and -21°C. The six diagrams clearly show
 110

that the deformation and the fracture behavior of the materials is changing with

temperature. The slope of the curves is determined by the elasto-plastic behavior of the

material. A steep and linear curve implies a high modulus of elasticity and little plastic

deformation. On the other hand, a pronounced deviation from the linear behavior

indicates a low modulus of elasticity and high deformation. The post-peak behavior

characterized the fracture process. If the curve decreases quickly as shown in Figure

5.10.c, little energy is consumed for the crack propagation, and brittle fracture is

observed. Contrary to this behavior, Figure 5.10.a shows a ductile fracture. In this case,

the horizontal force approaches zero only at high values of the displacement. It is also

shown that at low temperatures the crushed gravel withstand higher horizontal forces than

the natural gravel.

Figure 5.11 shows the test results for the maximum splitting force and the specific

fracture energy as a function of the testing temperatures. Figure 5.11.a shows that for

temperatures between -21°C and +8°C the crushed stone has higher values of maximum

splitting force. Figure 5.11.b shows that the specific fracture energy varies between 200

and 1100 N/m. At -21°C there is no significant difference between both mixtures.

Between -5°C and +5°C, the specific fracture energy for the crushed gravel mixture

exhibit a pronounced maximum which is about 2.5 times higher than that of natural

gravel. At +8°C the crushed gravel specific fracture energy is about twice of that of

natural gravel.

The researchers concluded that the maximum splitting force is not an appropriate

parameter to differentiate between HMA mixes since two different mixes can have the

same maximum splitting force and different fracture behavior. On the other hand, the
 111

specific fracture energy was recommended as a more reliable testing parameter to

differentiate between various mixes. This parameter can better describe the fracture

behavior of the material.

5.5 LABORATORY OF PUBLIC ROADS OF AUTUM, FRANCE

In 1993, Dumas and Vecoven (60) summarized the results and findings of 210

shrinkage-bending tests performed on various kinds of reflective cracking mitigation

techniques. The shrinkage bending test used in this study was developed in 1988 to

assess the efficiency of anticracks systems. The test simulates at the same time the

pavement thermal contraction and the heavy traffic loads at a constant temperature (5°C).

The shrinkage-bending test shown in Figure 12 consists of an HMA overlay or a

combination of HMA overlay and a cement treated base that are placed on top of two

steel plates. One plate is fixed and the other is forced to move at a constant speed. The

traffic loads are simulated by applying a vertical sinusoidal cycling load with a frequency

of 1 Hz and a vertical movement of 7.9×10-3 inch (0.2 mm). Consequently, the test

simulates concurrently both modes I and II type of fracture mechanics failures. The test

is performed as follows.

• The thermal shrinkage of pre-cracked pavement is simulated by the opening of the

movable plate (average speed 0.024 inch/hr).
• The action of a traffic axle is simulated by a cyclic loading at a frequency of 1 Hz
monitored by the deflection set to 0.079 in (0.2 mm).
• The test sample is 23.6×2.8×2.8 inch (60×7.0×7.0 cm).
• The speed at which a crack open and close is 0.04 inch/hr (1 mm/hr).

The measurements during the test consist basically of a visual determination of

the crack initiation time and length, crack propagation time and length, and breaking time

of the sample. Such information is used to determine the crack speed at any of these
 112

stages. The breaking time is considered as the efficiency of the reflective cracking

mitigation technique.

Figure 5.13 shows the initiation time of the crack, its propagation speed in the

first centimeter and the breaking time for two different HMA overlay mixtures.

Table 5.3 shows the test results of a 2.4 inch (6 cm) AC 0/10 overlay on top of a

fabric interlayer, a rich-binder HMA mix, and a Stress Absorbing Membrane Interlayer

(SAMI). It was concluded that the paving fabrics delays the crack initiation time, while

the rich HMA mix slows down the crack propagation. The behavior of the SAMI is more

similar to that of the paving fabrics.

5.6 UNIVERSITY COLLEGE OF DUBLIN, IRELAND

In 2002, Gibney et al. (61) assessed the resistance to reflective cracking of a

variety of HMA mixtures that are used in Ireland using an accelerated simulative wheel

tracking test in accordance with the British Standard (BS) 598 (62). The device is

capable to simulate top-down and bottom-up cracking in the HMA layer on top of

concrete slabs. This test simulates the modes I and II fracture mechanics type of failures.

Figure 5.14 shows the two mechanisms of reflective cracking that can be

examined with the device: 1) overall deterioration and cracking of the underlying

concrete slab on either side of a joint position and 2) differential movement of the

underlying slab. The effect on the overlay in the first mechanism is such that cracks will

start at the bottom of the overlay and propagate upwards to the surface. This will be

referred to as “bottom-up” cracking. Conversely the effect of the second mechanism on

the overlay will be that cracks will first appear at the point of highest tension force, on the

surface and propagate downwards.

 113

The test set-up to examine bottom-up cracking is of the form of a simply

supported beam as shown in Figure 5.15. The ends of the specimen are supported on

narrow timber blocks while the span of the specimen is underlain by foam (representing a

weak foundation). Metal plates of 0.5 inch (10 mm) thick are placed above the foam

support to mimic the concrete layer and a joint of 0.5 inch (10 mm) is provided under the

centre of the specimen. The set-up is constrained within a metal mold which is

12×12×5.1 inch (305×305×130 mm) deep. The testing specimens are 5.5×11×2.0 inches

(140×280×50 mm). This was chosen to allow a reasonable span of test duration for the

different mixes (i.e. between approximately 20 minutes and 8 hours). The length of the

specimen allows clearance for the specimen to deform naturally and avoid constraint

against the ends of the mold. Foam spacers are placed at each end to prevent horizontal

“slip” of the specimen (61).

The test set-up to examine top-down cracking is of the form of a cantilever beam

and is shown in Figure 5.16. The testing specimens are 5.5×10.2×2.0 inches

(140×260×50 mm) and are supported on one side on a rigid timber block with a timber

clamping mechanism above. This mechanism is omitted from the photograph in Figure

5.16 for clarity but is shown in the line diagram. The cantilever end is supported on

foam. Steel plates, 0.5 inch (10 mm) thick, are again used to represent the concrete layer

and a joint of 0.5 inch (10 mm) is provided between the timber and foam supports. The

timber clamping mechanism prevents uplift during the test and a screw is fixed through

the clamping mechanism to prevent horizontal movement of the specimen (61).

In both test set-ups the travel speed of the wheel is fixed at 21 cycles per minute

in accordance with the requirement of BS 598 (62). Thus the rest period between wheel
 114

passes is short and the effect of variation in rest period on the mixtures is not examined.

A cantilever arm is used to apply a normal load of 520 ± 5 N to the specimen. A test

temperature of 25°C was used. During testing the loaded wheel travels over the

specimen until failure occurs. Failure is taken as being the instant when a crack

progresses through the full depth of the specimen. Depending on the type of test

performed, the test result can be the number of wheel passes to produce either the

bottom-up or the top-down failure. Throughout the test the deformation of the slabs over

the central 200 mm is recorded at 10 mm intervals.

5.7 UNIVERSITY OF ILLINOIS, UNITED STATES

In 2001, Dempsey developed and evaluated an Interlayer Stress Absorbing

Composite (ISAC) for mitigating the reflective cracking in HMA overlays (63). The

ISAC system consists of a low stiffness geotextile as the bottom layer, a viscoelastic

membrane layer as the core, and a very high stiffness geotextile for the upper layer. The

effectiveness of the ISAC layer to control reflective cracking was evaluated in a

laboratory pavement section with an HMA overlay placed on a jointed PCC slab

constructed and tested in an environmental chamber. A mechanical device was used to

simulate thermal strain in the PCC slab. The testing was conducted at -1.1 °C and strain

in the overlay was monitored using a sensitive LVDT device. Performance of ISAC was

evaluated by comparing the cycles to failure of an ISAC treated overlay with a control

section without ISAC and with two commercially available products. The base isolation

properties of the ISAC system were demonstrated in the laboratory evaluation studies.

The laboratory evaluation studies indicated that the ISAC system vastly outperformed the

control section and the two commercial products tested. Several years of field
 115

performance data have shown that the ISAC system is highly effective for mitigating

reflective cracking in HMA overlays used on both airport and highway pavement

systems.

The laboratory testing equipment consisted of a one fixed box section and a

second horizontally movable box section on roller as shown in Figure 5.17. A

mechanical device is used to simulate thermal strain in the slab by opening and closing

the joint at an extremely slow rate. The force required to pull and push the slab back to

its original position is monitored using a load cell placed between the slab and a

hydraulic ram. Propagation of cracking in the overlay is also visually monitored. This

test simulates the modes I and II fracture mechanics type of failures.

The test sample is a 6 inch wide and 90 inch long (15×225 cm) pavement section

and 2.7 inch (6.8 cm) thick PCC slab, ISAC layer, and 2.5 inch (6.35 cm) of HMA was

placed on top of the two box sections. The movable box section is attached to a hydraulic

ram that can open and close the PCC slab joint very slowly to simulate seasonal or daily

temperature variations (load frequency 0.041 mm/min). A load cell placed between the

movable box section and the hydraulic ram indicates the force exerted by the hydraulic

ram as it opens and closes the PCC slab joint. An LVDT device is located between the

fixed and movable box sections to indicate their relative movement. The testing device is

placed into an environmental chamber that is held constant during testing. Figure 5.18

shows the strain in an HMA overlay as a function of test cycles in pavement control

section (HMA overlay on top of a PCC) and Figure 5.19 shows the strain in the HMA

overlay as a function of test cycles in an ISAC pavement surface. The HMA overlay

without ISAC completely split apart after 7 cycles.

 116

5.8 AERONAUTICAL TECHNOLOGICAL INSTITUTE (ATI), BRAZIL

In 2003, Montestruque et al. presented an innovative laboratory test method to

study the effect of a polyester geogrid interlayer on the performance of a non-cracked

HMA layer on top of a cracked HMA layer (64). The laboratory evaluation was carried

out using a dynamic fatigue test that was performed on prismatic beams resting on an

elastic foundation (steel plates). This system was conceived to simulate a cracked

pavement after rehabilitation with the load applied at two critical conditions: on bending

and on shearing. Figure 5.20 shows the testing device.

The test specimen consists of an HMA beam of 18×6×3 inch (460×150×75 mm)

resting on a steel base plate with a pre-crack gap. The beam is simply supported and the

load actuator applies the loading at the center of the beam. The test is conducted for a

crack opening (gap) of 0.078, 0.24 and 0.35 inch (2, 6 and 9 mm). The geogrid is placed

on the crack tip between the HMA sample and the steel plate. A sinusoidal load is

applied in the vertical axis with a loading frequency of 20 Hz. The position of the load is

changed in relation to the crack to produce either bending or shearing as shown in

Figures 5.21 and 5.22. The sinusoidal load is applied by hydraulic equipment through a

steel plate with dimensions of 1.6×3.0 inch (40×75 mm), generating pressures of 80, 62

and 47 psi. Between the steel and the asphalt beam, a rubber was installed in order to

minimize the concentration of stresses related to the stiffness of the steel plate.

The horizontal movements of the reflective cracks opening and the plastic

deformation during the load application cycles are measured using the displacement

Crack Activity Meter (CAM) which is installed in the steel plate. CAM is fixed by

screws embedded within the asphalt concrete. A clip gage is used to record the pre-crack
 117

opening during the test. The central part of the beam is white painted for an easier visual

observation of the crack propagation.

The factor of effectiveness of geogrid ( FEG = N f ( with geogrid ) N f ( without geogrid ) )

represents the beneficial effect of the geogrid and is calculated as follows:

1
Nf = (5.1)
cf

1 2
cf = + (5.2)
N f ( B) N f (s)

where, cf is the fatigue consumption, Nf(B) represents the fatigue life of the beam with the

load in the bend mode and Nf(s) the fatigue life in the shear mode. Table 5.4 shows the

calculated FEG values for the various beams evaluated by Montestruque et al. (64).

5.9 FLORIDA ATLANTIC UNIVERSITY, UNITED STATES

In 2004, Sobhan et al. (65) evaluated the effects of reinforcing an HMA overlay

with stiff geogrids as a means to mitigate reflective cracking when placed on PCC joints.

The primary objectives of this study were: 1) to evaluate the growth propagation of the

reflective cracks under cycling loads, and 2) to assess the effects of placement location of

the geogrid within the overlay on the propagation of reflective cracks. A new test was

proposed by researchers to achieve the objectives and it is described as follows.

Figure 5.23 shows the schematic representation of the proposed test setup which

simulates the mode I type of failure. HMA slab specimens of 18×6×7.5 inch

(450×150×190 mm) representing the overlay were prepared in a detachable steel mold

using roller compaction techniques developed in the laboratory. A cylindrical steel roller

provided a kneading compaction to the HMA overlay and simulated the field compaction
 118

practices as closely as possible. Such compaction technique allowed for compacting

HMA beams reinforced with grogrids or other materials. After de-molding, and 24 hours

before testing, two plywood pieces are attached at the bottom of the specimen (with 0.4

inch gap in between) by means of a tack coat to present the joints of the underlying

concrete pavement. Then, the specimen was placed on the rubber foundation for testing.

Two types of test were conducted in the study:

• Static tests: these included series of test on unreinforced specimens with

monotonically increasing loads in order to evaluate the ability of the test set up to
successfully simulate the initiation and propagation of the reflection cracking, to
determine the static load bearing capacity, and to develop various failure criteria.
The load is applied in the center of the specimen. The HMA overlay is placed
over a neoprene rubber foundation having a pre-crack or joint of 10 mm. Such
pre-crack caused the cracking initiation.
• Cyclic tests: these included series of cyclic/fatigue tests on both unreinforced and
geogrid reinforced specimens, conducted using a sinusoidal loading waveform at
various constant amplitudes of 222, 444, 888, 1110 and 1332 N, and at a constant
frequency of 2 Hz. The load is applied in the center of the specimen. The HMA
overlay is placed over a neoprene rubber foundation having a pre-crack or joint of
10 mm. Such pre-crack caused the cracking initiation.

The data acquisition included time and load and deformation which were

continuously recorded for all loading cycles to failure. In addition, continuous digital

video data focused on the zone of the simulated concrete joint during the progress of the

test was recorded for analyzing crack propagation.

The test setup was used by Sobhan et al. (65) to evaluate the performance of

HMA overlays with two geosynthetic materials: the Tensar Biaxial Geogrid (BX 1500)

and the Amoco Petrogrid (4582). Three failure criteria based on the observed length of

crack propagation were established.

• Initial Crack, Cif : the load or number of cycles at which a reflective crack is first
visible.
 119

• Mature Crack, Cmf : the load or number of load cycles it takes for the reflection
crack to propagate half of the overly depth.
• Terminal Cracking, Ctf : The load or number of load cycles it takes for the crack
to advance to the top of the overlay.

These observed criteria for seven static tests conducted on the HMA overlay

system revealed that the average Cif value is about 1,110 N. The results showed that for

the unreinforced specimens, the formation of reflection cracks during the early stages and

additional secondary cracks (cracks developing near the joint of the loaded area on both

sides) near failure stages are clearly observed. In the case of reinforced specimens,

although the crack initiation was clearly evident, the presence of geogrids restricted or

diverted the cracks remarkably well. The secondary cracks for this case appeared near

failure stages. For these cases, failure was defined when the specimens reached one

million cycles. For geogrids placed at the middle, the cracks initiated early, but after

reaching approximately the middle of the overlay, the rate of propagation was severely

restricted due to the presence of the geogrids.

The beneficial effect of the geogrid was quantified by calculating a Fabric

Effectiveness Factor (FEF) that is defined as a function of the number of cycles to crack

corresponding to any particular load ratio as follows.

C1 f ( re inf orced )
FEF = (5.3)
C1 f (unre inf orced )

It was found, at the same load ratio, the slabs having geogrids embedded at the

bottom showed better resistance to reflection cracking compared to specimens in which

the geogrids was simply attached to the bottom with a tack coat. Additionally, it was
 120

found that geogrid embedded at mid-height was more effective than geogrid embedded at

the bottom of overlay.

5.10 POLYTECHNIC UNIVERSITY OF MADRID, SPAIN

In 2006, Gallego and Prieto (66) presented the Wheel Reflective Cracking (WRC)

device for testing the reflective cracking resistance of HMA overlays. A schematic

representation and a picture of the WRC are shown in Figures 5.24 and 5.25,

respectively.

The base temperature of 5±1°C was chosen to simulate the low-temperature

distress mechanism and for running the test. However, the test can be performed in a

range of temperatures between 0°C and 20°C.

As shown in Figures 5.24 and 5.25, the WRC test comprises a supporting system

of four trapezoidal frames and two chasses. Each chassis is supported at one end by the

trapezoidal frames by means of a horizontal shaft allowing it to rock and, at the other end,

it rests on a rocker that simulates the relative vertical movement between the borders of

the crack or the pavement joint. Both chasses are set 0.4 inch (10 mm) apart, a distance

simulating the joint or crack in the deteriorated pavement on which the asphalt overlay is

installed.

The temperature-change distress mechanism is simulated using a system of

sliding plates. The specimen is stuck by a synthetic adhesive to two plates (Figure 5.26),

then placed inside the apparatus and screwed down fast. During the test, one of the plates

remains immobile while the other is displaced horizontally by the action of the tensile

force applied to one end at a speed ranging from of 0.001 to 50 μm/hr. This method is

chosen to simulate the progressive widening of the crack as a result of the effect of the
 121

temperature changes causing the concrete pavement slabs to contract, making joints and

cracks open up.

Deflection is simulated by placing a prism-shaped rubber block under the rocker

support (Figure 5.27). As a result, independently of the effect of the rocker on the centre

of the specimen, a vertical displacement occurs caused by the strain in the rubber block.

This equipment allows different magnitudes of deflection to be simulated by varying the

hardness or thickness of the rubber blocks used, to which end different heights of rocker

support are available.

During the test, the vertical deflection in the center of the test specimen, the

vertical length of the crack, and the relative movement between the borders of the crack

are measured. Also, a measure of the time to failure can be performed. The researchers

suggested the failure criteria when a relative movement of 0.2 mm occurs between the

crack edges.

5.11 REGIONAL LABORATORY OF PONTS ET CHAUSSES, FRANCE

In 2004, Tamagny et al. (67) evaluated the capabilities and effectiveness of the

MEFISTO device designed to determine the anti-reflective cracking behavior of various

materials. The MEFISTO device is used to study the efficiency of anti-reflective cracking

methods by simulating fatigue cracking in HMA overlays.

The MEFISTO test consists of subjecting the sample to a sinusoidal loading with

a frequency of 10 Hz, applied with one or two columns close to the center of the beam,

plus a monotonous horizontal loading (Figures 5.28 and 5.29). However, since the

machine is still under development, it has not been yet possible to test such combined

modes of loading, imposing vertical and longitudinal dynamic displacements at the same
 122

time. Instead only vertical loading or horizontal loading modes have been tested

independently.

The MEFISTO has two side connections of metal places which allow the device

frames to act as knee-joints with negligible flexural stiffness and zero imposed horizontal

displacements. The test beam is 2×2×26 inch (50×50×650 mm). During the specimen

compaction a 0.4 inch (10 mm) wide pre-crack is made in the center of the beam. The

test is conducted at 5°C and the applied vertical load is 1.9 kips (8.5 kN).

The device has the capability of measuring the applied vertical and horizontal

forces and the displacement in the center of the beam as shown in Figures 5.28. During

the test, the dissipated energy as a function of the number of cycles until failure is

calculated. Failure is defined as the moment when the crack has completely crossed the

beam thickness. It is also possible to visually measure the crack length as a function of

number of load cycles until the specimen fails.

The MEFISTO test produces a fatigue crack in the central part of the HMA

sample, starting from the notch end that is submitted to traction and then growing

upwards with time. Figure 5.30 shows a typical test results from the MEFISTO test. The

amplitude of the vertical load is monitored with respect to the number of loading cycles

as well as the crack expansion as a percentage of the sample thickness.

5.12 TEXAS TRANSPORTATION INSTITUTE OVERLAY TESTER, UNITED

STATES

In 2003, Zhou et al. from the Texas Transportation Institute (TTI) upgraded the

TTI overlay tester that has been widely used to evaluate the effectiveness of different

geosynthetic materials since it was designed by Lytton et al. in late 1970’s (68).
 123

The apparatus is shown in Figure 5.31 consists of two steel plates, one fixed and

the other movable horizontally to simulate the opening and closing of joint or crack in the

old pavements beneath an overlay. A beam sample is spanned across the crack and

epoxied to the horizontal surface platens with half of the length of the specimen resting

on each platen. The moveable platen is opened and closed within a preset amount

(between 0.0001 and 2 mm). For most applications this value is computed based on

project specific information such as the slab length and anticipated temperature variation.

Each sample is painted white in the areas where cracking was most likely to occur. This

technique enhanced the visibility and measurement of cracks during the testing phase.

The TTI overlay tester was upgraded to be fully computer-controlled system with

special programs. The test data including time, displacement, and force, are

automatically recorded and saved as Excel file. The sample size of the upgraded TTI

overlay tester has been reduced to 150 mm long by 75 mm wide by 38 through 50 mm

high, making the overlay tester more practical and easier to handle samples from the

Suprepave Gyratory Compactor (SGC) or field cores. Figures 5.32 shows the upgraded

TTI overlay tester. The upgraded system can be conducted in controlled displacement

mode under the following conditions.

• Temperature: 0 – 35 ºC
• Opening displacement: 0 – 2 mm
• Loading rate: 24 hours (or more) per cycle – 10 seconds per cycle
• Loading type: the loading is applied in a cyclic triangular waveform with constant
magnitude.

The main output of the overlay tester is the reflective cracking life of an HMA

mix which is the number of cycles to completely break the sample. Researchers found
 124

that this value is a good indicator of reflective cracking resistance of asphalt mixtures and

has a good correlation with the change of load (68, 69).

A typical relationship between load and time (number of load cycles) during

testing is illustrated in Figure 5.33. The observation of results from many overlay tests

showed that this plot has three distinct phases for the crack propagation (69).

• Phase I: Crack initiation and steady propagation. In this phase the load and
displacement have similar shapes. As the displacement increases, the load
increases too. For the first cycle, the load reaches its maximum value before the
displacement arrives at the maximum displacement indicating the crack initiation
at the bottom. After the first cycle, the load decreases rapidly as the crack starts
to propagate through the specimen. However, both load and displacement reach
the maximum values at the same time. In this stage, the cracking is steadily and
slowly propagated to the top surface.
• Phase II: Late crack propagation. Phase II is the late stage of crack propagation,
which is monitored as a saddle-shaped load. The saddle-shaped load indicates
that the crack has partially gone through the whole cross section of the specimen.
In fact, the first peak load is associated with the minor adhesion as the specimen
gap is closed and the two halves of the specimen bond together. Then, the load
rapidly decreases just after breaking the weak adhesion bonds. With the
increasing opening displacement, more loading is needed to break the remaining
parts of the specimen. Corresponding to the maximum displacement, there is
another peak load. With the continuing cyclic loading, the crack will totally break
the specimen and the second peak load will disappear. This indicates the onset of
Phase III.
• Phase III: Specimen failure. As described above, the crack has propagated
completely through the specimen in this phase. The maximum load induced by
the minor adhesion occurs well before the maximum displacement.

Based on the above discussion, the reflective cracking life of asphalt mixtures,

thus, can be defined by the number of cycles corresponding to the onset of Phase II or

Phase III. From the conservative point of view, the onset of Phase II should be used to

define the reflective cracking life. Using the evaluation scheme described above, the

reflective cracking life of the specimen was determined to be four cycles (10 sec/cycle).
 125

5.12.1 Variability of Upgraded Overlay Testing

The first step in evaluating the overlay tester concept, especially with the

recommended small sample size (Superpave gyratory compacted – SGC cores instead of

HMA slabs), was to determine the repeatability of the test. In general, the smaller the

specimen, the more variable the test results can be. Since the upgraded overlay tester was

using a small specimen, there was concern about its repeatability. Thus, two types of

TxDOT mixtures, Type D and CMHB-C using PG64-22 asphalt binder, were selected to

make six identical specimens (6 in [150 mm] diameter by 2.25 in [57 mm] high) for each

mixture. All the specimens were molded using the SGC. Then, the specimens were cut to

be 1.5 in (38 mm) high using a double blade saw; after that, 1.5 in (38 mm) was trimmed

from each side of the specimens (Figure 5.34). The air void content of each specimen

was controlled within 7±0.5 percent after trimming the specimens. Finally, six overlay

tester specimens for each mixture were glued to the overlay tester plates. The testing was

conducted at room temperature (77 ºF [25 ºC]) and the opening displacement was set to

0.025 in (0.63 mm).

Figure 5.35 shows the reflective cracking lives of six identical Type D specimens.

An average reflective cracking life of 140 cycles was found with a corresponding

standard deviation and coefficient of variation of 11.7 and 8.3 percent, respectively.

Generally speaking, the coefficient variation of asphalt mixtures is around 10 to 25

percent. These results clearly indicated that the overlay testing is a repeatable test.

On the other hand, The CMHB-C mixture showed a poor resistance to reflective

cracking under the overlay tester with each of the specimens failing after only two cycles.

Therefore, this mixture was excluded from the repeatability analysis.

 126

The test results for the TxDOT type D mixture were used to provide an estimation

of the minimum number of replicates needed to perform the test. Figure 5.36 shows the

relationship between the number of specimens and the specified tolerance. It can be seen

that the average reflective cracking life of two specimens, for Type D mix, will be within

±12 percent of the “true” reflective cracking life of asphalt mixture with 95 percent

reliability.

Based on the results of this analysis, it was recommended to use at least three

replicates of the HMA mixture to get an error of less than 10 percent.

5.12.2 Sensitivity of Upgraded Overlay Testing

The sensitivity of the upgraded overlay tester to material properties and test

conditions were also investigated. The parameters investigated in this project included

test temperature, opening displacement, air voids, asphalt performance grade, and asphalt

content. The TxDOT type D mixture from US281 in the Fort Worth District was used for

this analysis. The optimum asphalt content was 5.1 percent. It should be noted that only

one parameter was variable in this sensitivity test and the others were kept the same. The

detailed results are presented as follows.

Influence of temperature on reflective cracking life

A PG76-22 SBS modified binder was used to mold six identical specimens at 4

percent air void content. Overlay testing was conducted at 77 ºF (25 ºC) and 50 ºF (10

ºC) and for an opening displacement of 0.025 inch (0.63 mm). At each temperature, three

replicates were used. The averaged reflective cracking life presented in Figure 5.37

showed a significant influence for the temperature on the reflective cracking life of the
 127

HMA mix. Therefore, it was concluded that the overlay testing is sensitive to

temperature.

Influence of opening displacement on reflective cracking life

Similar to the previous test, a PG76-22 SBS modified binder was used to mold six

identical specimens at 4 percent air void content. Overlay testing was conducted at 77 ºF

(25 ºC) and for two opening displacements: 0.025 inch (0.63 mm) and 0.035 in (0.89

mm). At each opening displacement, three replicates were used. The averaged reflective

cracking life presented in Figure 5.38 indicated that overlay testing results are sensitive to

the opening displacement. The reflective cracking life of HMA mixes decreased with the

increase of opening displacement.

Influence of asphalt content on reflective cracking life

A PG64-22 binder was used to mold three replicate specimens at each of the three

asphalt binder contents: 4.2, 5.1 (optimum), and 6.1 percent. The overlay testing was

conducted at 77 ºF (25 ºC) and 0.025 inch (0.63 mm) opening displacement. The

averaged reflective cracking life presented in Figure 5.39 showed a significant increase in

the reflective cracking life of the HMA mix with the increase of asphalt content. The

results were consistent with the traditional flexural beam fatigue test results.

Influence of asphalt performance grade on reflective cracking life

The data from Figures 5.38 and 5.39 showed that an increase from a PG of 64 to

76 resulted in a drop in reflective cracking life from 90 to 33 indicating a poorer

reflective cracking resistance for the stiffer asphalt binder.

 128

Influence of air voids on reflective cracking life

Figure 5.40 shows the influence of air voids content on reflective cracking life

with a better reflective cracking resistance for higher air voids. The researchers related

this behavior to most possibility the generation of denser and stronger specimens when

reducing air voids from 7.4 to 4.2 percent. Therefore, specimens with lower air voids

content would have higher stiffness and higher strength as well. However, thermal

reflective cracking simulated by the overlay tester is a different scenario. If temperature

dropping is kept constant, the denser mixture with higher modulus will suffer a higher

thermal stress. Inversely, although its strength is lower, the thermal stress induced within

the specimen with higher air void content will be lower, too. When the thermal stress

induced within a specimen is higher than its strength, a crack will occur. Whether or not

a specimen with lower air void content is resistant to thermal reflective cracking depends

on both its stiffness and strength.

5.12.3 Validation of Upgraded Overlay Tester

The validation of the upgraded overlay tester was composed of three steps. First,

the effectiveness of the overlay tester on characterizing reflective cracking resistance of

asphalt mixtures was discussed. Field cores with known reflective cracking performance

were used for this validation. Then, the potential application of the overlay tester to

evaluate the fatigue cracking resistance of asphalt mixtures was investigated. Finally,

cores taken from MnRoad were tested to check the potential of the overlay tester on

characterizing the low temperature cracking resistance of asphalt mixtures.

The validation of the TTI Overlay Tester was performed based on the

performance of various projects around Texas. Since 2000, the TTI small overlay tester
 129

has been successfully employed to characterize the reflective cracking resistance of

different asphalt mixtures with known reflective cracking performance in the field.

Reflective cracks quickly appeared in new overlays placed on US175, US84, SH3, SH6,

and IH10 throughout the state. Cores taken from these poorly performing pavements

were tested in the overlay tester and the results were compared with those of cores from

the Special Pavement Studies 5 (SPS5) section on US175 near Dallas. This overlay was

placed over a stabilized base and had no reflective cracks after 10 years in service. All of

these good and poor performing cores were used to validate the TTI overlay tester

concept. The HMA mixes tested by the overlay tester covered TxDOT type C mixes with

PG76-22 tire-rubber, type D mixes with PG64-22, and type D mixes with 30 and 75

percent recycled asphalt concrete. Several cores were taken from the various projects and

tested in the overlay tester. Also, a detailed condition survey was conducted in each of

the project to assess the condition of the pavement. Therefore, the mechanisms of failure

as well as the layer subjected to either reflective cracking or to a good performance were

identified. This information was used to validate the capabilities of the TTI overlay tester

to assess the reflective cracking performance of the HMA overlay.

In summary, the test results of the overlay tester were consistent with the field

performance, and the poor crack resistant HMA mixes were differentiated from the good

crack resistant mixes. The evaluated case studies also confirmed the overlay tester as a

rapid performance-related tool to evaluate the reflective cracking resistance of asphalt

mixtures. The overlay tester results on the known field performance cores taken from

different highways showed that asphalt mixtures performed very well when the reflective

cracking life (from the overlay tester) is larger than 300. Thus, the researchers proposed
 130

the preliminary pass/fail criterion on reflective cracking resistance to be 300 cycles at 77

ºF (25 ºC) and 0.025 in (0.64 mm) opening displacement. When a rich bottom layer is

used, the proposed reflective cracking life in the overlay tester should be at least 750

cycles.

The TTI OT was also validated using MnRoad Cores. For this purpose, it was

used to evaluate the low temperature cracking resistance of asphalt mixtures from

MnRoad. Three representative test cells (15, 18, and 20) at MnRoad were selected for

evaluation. Table 5.5 presents the asphalt mixture information and field performance of

the three cells. Two 6-inch (150 mm) diameter cores from each cell were taken from the

mid-lane of the driving lane (6 feet offset) and tested under the overlay testing.

Although the cores are composed of different layers, only the top HMA layer was

tested under the overlay tester since it is the critical layer for low temperature cracking.

The overlay tester was conducted at a temperature of 77°F (25 °C) with an opening

displacement of 0.025 in. The overlay tester results (Table 5.5) are found to be consistent

with the observed field cracking performance of asphalt mixtures. The results also

indicated that both asphalt content (cells 15 and 18) and asphalt binder PG (cel15, 18, and

20) had influence on crack resistance, which is consistent with the results of the

conducted sensitivity study.

 131

5.13 FEDERAL UNIVERSITY OF RIO GRANDE DO SUL, BRAZIL

In 1994 the State Roads Department (DAER/RS) constructed the traffic simulator

called UFRGS-DAER/RS (Figure 5.41) that was designed by the Mechanical

Engineering Department at the Federal University of Rio Grande do Sul (UFRGS) in

Brazil (70).

The traffic simulator dimensions are 49.2 ft (15 m) long, 8.2 ft (2.5 m) wide and

14.1 ft (4.5 m) high with a weight around 22 tons. The tires are mounted on a dual

wheel system on the loading carriage. The system is controlled with a computer that

controls the travel speed and wandering. The path of the tires can be varied in range of

±1.3 ft (0.4 m) from centerline. The loads are hydraulically applied and can vary from

9.2 to 22.5 kips (63 to 155 MPa). The tire pressures can be varied from 81 to 100 psi

(0.56 to 0.73 MPa). Wheel load and tire pressure are controlled within a tolerance of

±2%. The test wheels travel at 3.7 mph (6 km/h) over a 23 ft (7.0 m) pavement. Loads

are applied in one direction and normally distributed around the wheel path.

From March 2003 to February 2005, the traffic simulator was used on two full-

scale pavement sections to evaluate the resistance to reflective cracking of HMA overlays

(70). The experiment was carried out in an outdoor full-scale pavement testing facility in

the UFRGS Campus.

The test pavement was 53 feet long (16.2 m) by 10.5 feet (3.2 m) wide. The

pavement was longitudinally divided in two sections. In one of the sections the overlay

consisted of HMA with conventional binder (Brazilian designation CAP 20), while in the

other the overlay was made of asphalt rubber (AR). Figure 5.42 shows the cracked areas
 132

in both sections and the location of instrumentation embedded during pavement

construction.

Figure 5.43 shows the pavement structures. It may be seen that the sole

difference between both sections is the type of HMA mix (CAP 20 or AR) used in the 2

inch (50 mm) thick overlay. The HMA overlays were placed on top of a 1.6 inch (4.0

cm) pre-cracked HMA layer, on top of an 11.8 inch (30 cm) granular base course, on top

of a 20 inch (50.0 cm) clayey soil subgrade. The cracked layer consisted of a new dense

graded HMA layer which was laid and compacted in situ.

An already failed old HMA layer was represented by sawing cracks in the newly

compacted HMA layer on top of the granular base according to the pattern shown in

Figure 5.44. Cracks were 1.6 inch (4.0 cm) deep and 0.2 inch (0.5 cm) wide simulating

severe fatigue cracking. Four cracking areas of 4 feet (1.2 m) long by 2.7 feet (0.8 m)

wide each were sawn, alternating with uncracked areas, in each test section. Cracks were

filled with clayey soil in order to prevent crack sealing by the tack coat applied before

overlay setting. An axle load of 22.5 kip with a tire pressure of 91.3 psi was applied

using the UFRGS-DAER/RS traffic simulator.

During the testing process three types of measurements were conducted:

• Environmental measures: rainfall, air temperature and sun radiation were

continuously recorded in a meteorological station built in the pavement test
facility. Pavement temperatures were also recorded every time that deflections
and strains were measured.
• Structural response: surface deflections were measured at seven locations along
each test section, before and during trafficking. Deflections were measured under
axle loads of 18, 22, and 26 kips (82, 100, and 120 kN) with tire inflation
pressures of 80, 100 and 120 psi (0.56, 0.69 and 0.84 MPa), respectively.
Deflection basins were also measured with a road surface deflectometer, in order
to obtain back-calculated pavement and subgrade moduli.
 133

• Pavement performance measurements: included cracking and rutting data.

Rutting was measured using a profilograph. To eliminate initial surface
irregularities from the rut depth data, profiles obtained before trafficking were
used as references. A manual procedure was used to record and measure
cracking. Different color sprays were used each time a crack survey was
performed, as shown in Figure 5.45. Then, cracks were mapped using a square
metal grid, 1.0 by 1.0 m, with a mesh opening of 0.10 m. Afterwards, a cracking
severity index, given by the ratio of cracking total length to trafficked area was
computed.

Figure 5.46 shows the cracking evolution in both the AR and the CAP 20 HMA

sections. The asphalt rubber overlay outperformed the conventional asphalt concrete by

delaying reflection cracking. During testing, the first cracks appeared in the conventional

asphalt concrete (AC) overlay after 14,000 cycles of the 22 kip axle load. The cracks

sawn in the HMA layer underneath the neat binder section propagated through the HMA

overlay, and after 98,000 loadings the crack density was 220 cm/m2. On the other hand,

the asphalt-rubber HMA overlay showed its first crack after 123,000 loading cycles of the

22 kip axle load. The reflected cracks in the asphalt-rubber HMA showed up 5 to 6 times

loading cycles after the cracks of the conventional dense-graded HMA overlay.

5.14 SUMMARY OF LABORATORY TEST METHODS

Table 5.6 summarizes the review of the various laboratory test methods used to

evaluate the resistance of HMA mixtures to reflective cracking.

 134

Chapter 6 – SUMMARIES AND RECOMMENDATIONS

In 2006, the Nevada DOT initiated a three-phase research study to identify the

promising techniques to mitigate reflective cracking in HMA overlays: a) Phase I:

Review of literature and the performance of the various techniques in Nevada, b) Phase

II: Identify analysis models and laboratory tests, and c) Phase III: field verification of the

selected techniques. This chapter summarizes the findings of Phase I and Phase II of this

research and presents the recommendations for Phase III.

6.1 SUMMARY OF PHASE I FINDINGS

Table 6.1 summarizes the literature review of the current and previous efforts

outside Nevada on the mitigation of reflective cracking in HMA overlays.

Table 6.2 summarizes the general results of the review of the long-term field

performance of NDOT projects with different techniques to reduce the impact of

reflective cracking on HMA overlays. It should be noted that all treatments had either

0.375” or 0.75” open graded friction course on top. In general, the long-term

effectiveness of the treatments experienced by NDOT was significantly hampered by the

existence of severe alligator cracking on the projects prior to the application of these

treatments. Therefore, it is recommended that projects experiencing severe alligator

cracking as classified by the NDOT pavement distress manual should be subjected to

either re-construction or full depth reclamation.

6.1.1 Recommendations of Phase I

The research effort documented in the Phase I of the study was directed toward

identifying an effective method to eliminate the propagation of the cracks from the old

surface layer through the new HMA overlay. Several techniques showed promising
 135

results but none of the reviewed methods was able to completely stop reflective cracking.

Even though, under Nevada’s conditions the cold in-place recycling of a minimum 2.0

inches and overlaying with a minimum 2.5 inches of dense grade HMA mix was proven

to be effective in stopping reflective cracking for at least 5 years, it may not be the most

cost effective rehabilitation technique for every highway pavement. Therefore, based on

the literature and performance reviews, it is recommended to further investigate the stress

relief course technique. It should be noted that the stress relief course experienced by

NDOT was not specifically designed or checked for its reflective cracking resistance and

only consisted of a 1-inch of the typical Type II (1” max size) dense graded HMA mix

placed between the existing HMA and the overlay. This study revealed promising

performance for the stress relief course when specifically designed to resist reflective

cracking. Additionally, for a good overall performance, the stress relief course needs to

be coupled with an overlay mix with good resistance to reflective cracking.

6.2 SUMMARY OF PHASE II FINIDINGS

The research effort of Phase II was directed toward identifying an analytical

model(s) and laboratory or field test(s) that can be used to predict the resistance of HMA

overlays to reflective cracking and predict their long-term performance.

6.2.1 Analysis Models

Based on the review of the currently available analytical models to predict the

resistance of HMA overlays to reflective cracking, three design methods were identified

and summarized.

• Virginia Tech Simplified Overlay Design Model

• Asphalt Rubber Association Overlay Design Model
• The New AASHTO model for Reflective Cracking
 136

The Virginia Tech Simplified Overlay Design Model (45) consist of a simple

regression equation for predicting the number of cycles in ESALs (Wt80) to produce the

crack reflection to the pavement surface as a function of: thickness and resilient modulus

of HMA overlay (Hoverlay [mm], Eoverlay [MPa]), thickness and resilient modulus of

existing HMA layer (HHMA [mm], EHMA [MPa]), thickness and resilient modulus of base

layer (Hbase [mm], Ebase [MPa]), and resilient modulus of subgrade layer (Esubgrade [MPa]).

The total number of load repetitions is defined as the sum of the number of load

repetitions for crack initiation and the number of load repetitions for crack propagation.

log Wt 80 =
1
(255H overlay + 2.08Eoverlay + 45.3H HMA + 8.73EHMA )
104
1.37 H Base + 6.93EBase + 1.49 Esubgrade ) (6.1)

Equation 6.1 clearly shows that the overlay mix with a higher stiffness will

withstand a higher number of load repetitions. However, it should be noted that a stiffer

mix may be brittle thus more susceptible to reflective cracking.

The Asphalt Rubber Association Overlay Design Model (53) consists of a

mechanistic relationships and statistically based equations for designing HMA overlays

on top of HMA pavements. The proposed models are based on a finite element model

that closely approximates actual field phenomena of dense-graded HMA and gap-graded

asphalt rubber-AR (wet process) overlay mixes. However, other HMA mixes used for

overlays may also be calibrated and used through the proposed method using the relevant

mix properties. The overlay design program is available from the Rubber Pavements

Association in the form of an EXCEL spreadsheet that estimates the thickness of a PG70-

10 or an AR HMA overlay mix for the specified level of reflective cracking for a wide

range of traffic loading. The expected design level of cracking, the thicknesses and the
 137

elastic moduli of the existing pavement layers, and the modulus of the HMA overlay are

inputs for the EXCEL spreadsheet. The moduli may be backcalculated or determined in

any reasonable manner, as long as they represent the in situ conditions in the field.

The reflective cracking models incorporated in the new AASHTO Mechanistic-

Empirical Pavement Design Guide (MEPDG) are strictly based on empirical observations

without any rigorous mechanistic-empirical analysis (55). The empirical models consider

the development of distresses in the overlay as well as the continuation of damage in the

existing pavement structure. The proposed MEPDG overlay design procedure allows for

two types of reflective cracks: a) reflective cracks that exist on the surface prior to

overlay placement and b) those that develop in the existing surface after overlay

placement.

An overlay design was conducted for three different HMA overlay mixes using

the three identified overlay design methods. The overlay mixes were manufactured using

different aggregate gradations from the Sloan pit in Southern Nevada and a PG76-22NV

polymer modified asphalt binder. The overlay was designed for 7,075,000 ESALs over

the 20 years analysis period. Table 6.3 shows the material properties of the various

pavement layers used in the analysis. It should be noted that the fatigue characteristics of

the various mixes can only be incorporated in the Asphalt Rubber Association Overlay

Design Method. Table 6.4 shows the required overlay thicknesses for the design ESALs

according to all three design methods.

The data in Table 6.4 show that for the same design ESALs, a thicker overlay

thickness is required for the T2C mix followed by the CT mix followed by the NRM mix

when designing using the Virginia Tech method. On the other hand, the opposite was
 138

found when designing using the Asphalt Rubber Association method where a thinner

overlay thickness is required for the T2C mix followed by the CT mix followed by the

NRM mix to reach the same selected percentage of cracking. The AASHTO MEPDG

design method resulted in a 12 inch overlay thickness to reach 100% reflected cracking

after 20 years design period regardless of the type of the overlay mix.

In a summary when only the stiffness of the overlay mix is considered, a thinner

overlay thickness was found for the stiffer mix whereas, when both the stiffness and the

fatigue characteristic of the mix are considered, the overlay thickness was depended on

the interaction between the two material properties. On the other hand, when the material

properties of the overlay mix are not considered a unique and thick overlay thickness was

found.

6.2.2 Laboratory Tests

This task reviewed the laboratory tests that have been used to evaluate the

resistance of HMA mixtures to reflective cracking. Laboratory tests are typically used to

evaluate the resistance of the HMA mixtures to reflective cracking during the mix design

stage.

None of the reviewed laboratory test methods has undergone field validation

except the TTI Overlay Tester which showed consistency between the mixtures’ test

results and their corresponding field performance. The Overlay Tester was able to

differentiate between the poor crack resistant and the good crack resistant HMA mixes.

Additionally, the overlay tester results on the known field performance cores taken from

different highways showed that asphalt mixtures performed very well when the reflective

cracking life (from the overlay tester) is larger than 300. Thus, the researchers proposed
 139

the preliminary pass/fail criterion on reflective cracking resistance to be 300 cycles at

77ºF (25ºC) and 0.025 inch (0.64 mm) opening displacement. When a rich bottom layer

is used, it was proposed that the reflective cracking life in the overlay tester should be at

least 750 cycles.

6.2.3. Recommendations of Phase II

The assessment of the three identified design models illustrated the following

advantages for Asphalt Rubber Association method:

• Consider the overlay mixture material properties such as the stiffness as

determined by the dynamic modulus test and the fatigue characteristic as
determined by the flexural beam fatigue test at constant strains and various
temperatures.
• Allow for the selection of the desired percent of cracking at the end of the design
life.

Consequently, it is recommended to validate the Asphalt Rubber Association

Overlay Design Model under Nevada’s mixes and conditions.

It is recommended to use the TTI Upgraded Overlay Tester to evaluate typical

NDOT mixtures along with their actual field performance to develop performance criteria

that can be used at the design stage of the overlay mixture.

6.3 OVERALL RECOMMENDATIONS

Based on the findings of the Phase I and Phase II of the study it is recommended

to evaluate the stress relieve courses under Nevada’s conditions. Therefore the following

recommendations are made:

• Review the current state highway agencies’ (SHA) specifications for stress relieve
courses (SRC).
• Use the TTI Upgraded Overlay Tester to evaluate mixtures in the Laboratory for
reflective cracking resistance.
• Use the Asphalt Rubber Association Overlay Design Model to design the require
overlay thickness.
 140

6.4 REVIEW OF STRESS RELIEF COURSES (SRC) SPECIFICATIONS

Following the overall recommendations, the SHA specifications for SRC were

reviewed and are summarized in the following section.

6.4.1 Arizona Department of Transportation

The Arizona Department of Transportation (ADOT) defines the Stress absorbing

membrane interlayer (SAMI) as a furnishing of asphalt-rubber, tack coat, and a cover

material of mineral aggregate (71).

6.4.1.a Asphalt Rubber

The asphalt rubber shall be a mixture of asphalt cement and rubber. The asphalt

binder should have a performance grade (PG) conforming AASHTO Provisional

Standard MP1.

The rubber gradation for the SAMI shall meet the requirements shown in Table

6.5 when tested in accordance with Arizona Test Method 714.

The rubber shall have a specific gravity of 1.15 ± 0.05. Additionally, the rubber

shall be free of wire or other contaminating materials and shall contain not more than 0.1

percent fabric. Calcium carbonate, up to 4 percent by weight of the granulated rubber,

may be added to prevent the particles from sticking together. The asphalt rubber shall

contain a minimum of 20 percent ground rubber by the weight of the asphalt cement and

shall conform to the requirements in Table 6.6.

6.4.1.b Mineral Aggregates

Aggregates for cover material shall be of clean sand, gravel or crushed rock and

shall be free from lumps or balls of clay and shall not contain calcareous or clay coatings,
 141

caliche, synthetic materials, organic matter or foreign substances. The aggregates shall

meet the requirements in Table 6.7.

6.4.1.c Construction Requirements

The asphalt rubber shall be placed on a previously cleaned surface. After

cleaning and prior to the application of the SAMI, the existing pavement surface shall be

treated with a tack coat with the following conditions.

• Ambient air temperature and pavement surface temperature are both above 65°F.
• Pavement is dry.
• Wind conditions are such that a satisfactory SAMI can be achieved.
• All construction equipment such as asphalt rubber distributor, aggregate spreader,
haul trucks with aggregate material, and rollers are in position and ready to start
placement operations.

Distributor trucks shall be so designed, equipped, maintained and operated that

bituminous material at even heat may be applied uniformly on variable widths of surface

up to 15 feet at readily determined and controlled rates of 0.03 to one gallon per square

yard, with uniform pressure, and with an allowable transverse variation from any

specified rate not to exceed 10 percent or 0.02 gallons per square yard, whichever is less.

The maximum deviation from the specified rate shall not exceed 0.05 gallons per square

yard.

The hot asphalt-rubber mixture shall be applied at the rate of approximately 0.55

± 0.05 gallons per square yard (based on a unit weight of 7.75 pounds per gallon of hot

asphalt-rubber). Cover material (aggregate) shall be immediately and uniformly spread

over the freshly applied asphalt-rubber at the rate of approximately 0.014 cubic yards per

square yard. Cover material shall be precoated with 0.40 to 0.60 percent asphalt cement,
 142

by weight of the aggregate, and shall have a minimum temperature of 250 °F at the time

of application.

At least three pneumatic rollers shall be provided to accomplish the required

rolling. The first pass shall be made immediately behind the spreader and if the

spreading is stopped for any reason, the spreader shall be moved ahead so that all cover

material may be immediately rolled. The rolling shall continue until a minimum of four

complete coverages have been made.

Traffic of all types shall be kept off the stress-absorbing membrane until it has

had time to set properly. The minimum traffic free period shall be three hours. Sweeping

shall be completed and all excess cover material removed prior to the placement of any

subsequent layers of asphaltic concrete. If the asphalt-rubber membrane has been

subjected to traffic, a tack coat shall be applied at the rate of approximately 0.06 gallons

per square yard prior to placement of the asphaltic concrete.

6.4.2 California Department of Transportation

The California Department of Transportation (CALTRANS) defines an Asphalt-

rubber membrane interlayer (SAMI-R) as an asphalt rubber chip seal that is overlaid with

conventional dense graded asphalt mixture (AC) or a gap graded hot mix (RAC-G) (72).

SAMI-R is a low modulus (nonstructural) layer that is used to retard and minimize

reflective cracking in overlays placed on it, and to minimize further infiltration of surface

water through the pavement structure.

No fog seal or sand should be applied over a SAMI-R because this could interfere

with bonding of the overlay. SAMI-R may be applied to any type of rigid (PCC) or

asphalt pavement, and have proved very effective at minimizing reflection of PCC joints.
 143

However, according to the Caltrans Maintenance Manual, if the surface

irregularities (rutting in AC or faulting of PCC) exceed 12.5 mm then either a leveling

course should be placed or grinding and crack filling are required prior to placing SAMI-

R.

SAMIs are used under corrective maintenance overlays and are a pavement

rehabilitation tool. A SAMI-R would not be included as part of new construction.

Design of the asphalt rubber binder is the same as for chip seal. Determination of

appropriate binder and cover aggregate application rates is also the same.

6.4.2.a Asphalt Rubber

Typically the AR-4000 is used as a base asphalt binder. The asphalt rubber shall

conform to the requirements in Table 6.8.

6.4.2.b Mineral Aggregates

The standard chip size for Caltrans asphalt rubber seals is 9.5 mm. The 12.5 mm

chips are used by Caltrans only where ADT is less than 5,000 per lane.

6.4.2.c Construction Requirements

To construct a chip seal, the hot asphalt rubber binder is sprayed on the roadway

surface at a rate determined by the Engineer. The binder is immediately covered with a

layer of hot pre-coated chips that must be quickly embedded into the binder by rolling

before the membrane cools. Best results are achieved with clean nominal 9.5 to 12.5 mm

single-sized chips. Lightweight aggregates may be substituted to minimize windshield

breakage by loose chips in areas where traffic is heavy or fast.

Pre-coating the aggregate with asphalt cement improves adhesion by removing

surface dust and “wetting” the chips. Caltrans requires that the aggregate chips be
 144

delivered to the job site precoated and hot. To further aid chip retention after the chips

have been embedded and swept, a fog seal of asphalt emulsion (diluted 1:1 with water) is

sprayed over the chips at a typical rate of 0.14 to 0.27 liter/m2. A light dusting of sand, 1

to 2 kg/m2 is then applied as blotter as directed by the Engineer. According to Caltrans

standard special provisions for asphalt rubber seal coat, the application rates for asphalt

rubber chip seals are shown in Table 6.9.

However, the exact rate is to be determined by the Engineer. There are a number of

factors that can affect the asphalt rubber binder and cover aggregate application rates

including:

• Surface texture of the existing pavement: severely aged, oxidized and open-
textured surfaces will absorb more binder than newer tighter surfaces.
• Traffic volumes: typically use smaller chips for higher volumes to reduce
potential for vehicle damage by loose chips. Binder application rates can be
increased for low traffic volume areas.
• Seasonal temperature ranges: thicker membranes may be used in areas with cool
climates.
• Aggregate size: large stone requires more asphalt rubber binder (thicker
membrane) to achieve 50 to 70 percent embedment.
• Aggregate gradation: single-sized materials require more asphalt.

6.4.3 Florida Department of Transportation

The Florida Department of Transportation (FDOT) specification indicates that an

asphalt rubber membrane interlayer is composed of a separate application of asphalt

rubber binder covered with a single application of aggregate (73).

6.4.3.a Asphalt Rubber

The asphalt-rubber binder is a mixture of a PG-graded asphalt binder and ground

tire rubber. The asphalt binder shall be graded in accordance to the AASHTO M-320.

Three types of binders are identified: PG 64-22, PG 67-22 and PG 76-22.

 145

The ground tire rubber shall be produced from tires by an ambient grinding

method. The entire process shall be at or above ordinary room temperature. The rubber

shall be sufficiently dry so as to be free flowing and to prevent foaming when mixed with

asphalt cement. Also, it should be substantially free from contaminants including fabric,

metal, mineral, and other non-rubber substances. Up to 4% (by weight of rubber) of talc

or other inert dusting agent, may be added to prevent sticking and caking of the particles.

The physical properties of the ground tire rubber shall meet the following requirements:

• Specific Gravity: 1.10 ± 0.06.

• Moisture Content: Maximum 0.75%.
• Metal Contaminants: Maximum 0.01%.
• Gradation: Table 6.10.

Additionally, the asphalt-rubber should meet the specification shown in Table

6.11. The chemical composition of the ground tire rubber shall be determined in

accordance with ASTM D297 and shall meet the following requirements:

• Acetone Extract: Maximum 25%.

• Rubber Hydrocarbon Content: 40 to 55%.
• Ash Content: Maximum 8% (10% for Type A).
• Carbon Black Content: 20 to 40%.
• Natural Rubber: 16 to 45%.

6.4.3.b Mineral Aggregates

The aggregate size No. 6 (0.75 inch aggregate size) shall be used and shall meet

the requirements of Section 901 of the FDOT Standard Specifications.

6.4.3.c Construction Requirements

The existing pavement should be cleaned prior to the application of the asphalt rubber

binder. The asphalt rubber binder should be applied only under the following conditions:
 146

• Air temperature is above 50ºF and rising.

• Pavement is absolutely dry.
• Wind conditions are such that cooling of the asphalt rubber binder will not be so
rapid as to prevent good bonding of the aggregate.

The asphalt rubber binder shall be applied at the rate of 0.6 to 0.8 gal/yd2 or as

directed by the Engineer. Immediately after application of the asphalt rubber binder, the

cover material (aggregate) should be spread uniformly at a rate of 0.26 and 0.33 ft3/yd2. The

application of the asphalt rubber binder and the application of the cover material should not

be separated by more than 300 feet, unless approved by the Engineer.

In order to ensure maximum embedment of the aggregate, the entire width of the mat

should be covered immediately by the traffic rollers. For the first coverage, a minimum of

three traffic rollers should be provided in order to accomplish simultaneous rolling in echelon

of the entire width of the spread.

6.4.4 Iowa Department of Transportation

The Iowa Department of Transportation (IDOT) only specifies the SAMI’s for

being placed under an unbounded Portland cement concrete (PCC). Such SAMI consists

of a 1-inch nominal aggregate size HMA overlay (74).

The asphalt binder shall be a PG58-28. The mixture shall meet the following

characteristics:

• 300,000 design ESAL.

• 3/8 inch (9.5 mm) nominal aggregate size gradation.
• The air voids target is 3.0%.
• No maximum film thickness restriction.
• No minimum filler/bitumen ratio restriction.
• Aggregate shall be Type B with no percent crushed particle requirements and
gradation shall fall below the restricted zone.
 147

The only performance test specified to the mixture is the Tensile strength ratio (TSR).

The contractor shall run AASHTO T 283 during production. The test results shall satisfy

80% TSR when compared to the dry strength of specimens prepared with asphalt binder

containing the anti-strip additive.

6.4.5 Massachusetts Department of Transportation

The Massachusetts Department of Transportation (MDOT) specifies the SAMI as

the application of hot, rubberized asphalt and immediately embedding aggregate therein

by spreading and rolling according with these specifications (75).

6.4.5.a Asphalt Rubber

The asphalt cement for the asphalt-rubber mixture shall be AC-10 or AC-20. If

AC-10 is used, the SAMI shall be overlaid within ten days. The granulated rubber shall

be a vulcanize rubber product from the ambient temperature processing of pneumatic

tires. The granulated rubber type shall meet the gradation in Table 6.12.

The percent of asphalt-rubber shall be 23 ± 2% by the total weight of asphalt

cement plus granulated rubber. The temperature of the asphalt shall be between 347°F

and 428°F at the time of addition of the vulcanized rubber. The asphalt and the rubber

shall be mixed together in a blender unit and reacted in a distributor for a period of time

determined as required by the engineer. The temperature of the rubberized asphalt

mixture shall be above 320 °C during the reaction period.

After the reaction between the asphalt and rubber has occurred, the viscosity of

the hot rubberized asphalt mixture may be adjusted for spraying and “wetting” of the

cover material by the addition of a diluent. The diluent shall comply with the
 148

requirements of ASTM D369, Grade #1 fuel oil and shall not exceed 7.5 percent by

volume of the hot asphalt rubber mixture.

Viscosities shall be run on each blended load of rubberized asphalt rubber using a

Haake field viscometer. One viscosity prior to the induction of the diluent and one after

the induction of the diluent blended into the asphalt and rubber mixture. The viscosity of

the final product shall be in the range of 2 to 3 Pascal-seconds.

6.4.5.b Mineral Aggregates

The aggregate shall conform the requirements of M2.01.0 of the standard

specifications for crushed stone. The percentage of wear as determined by the Los

Angeles Abrasion Test (AASHTO T 96) shall be a maximum of 30.

6.4.5.c Construction Requirements

Prior to the application of the rubberized asphalt, the entire pavement surface to

be treated shall be cleaned by sweeping, blowing and other methods until free of dirt and

loose particles.

The rubberized asphalt mixture shall be applied at a temperature of 284°F to

338°F at a rate of 2.75 ± 0.25 liters/m2. Transverse joints shall be constructed by placing

building paper across and over the end of the previous rubberized asphalt application.

Longitudinal joints shall be overlapped from 100 to 150 mm.

The application of the aggregate shall follow as close as possible behind the

application of the hot rubberized asphalt which shall not be spread further in advance of

the aggregate spread that can be immediately covered.

The dry aggregate should be pre-coated with 0.5 to 1.0% of AC-20 and shall be

spread uniformly by a self-propelled spreader at a rate directed by the engineer, generally

 149

between 15 and 20 kg per square meter. Prior to application the aggregate shall be pre-

heated to a temperature between 248°F to 302°F.

Rolling shall be immediately following the spread of aggregate. There shall be at

least three complete passes by the pneumatic tired rollers to embed the aggregates

particles firmly into the rubberized asphalt, followed by an additional pass of the steel

roller.

The rubberized asphalt surface should be overlaid immediately following

completion of sweeping. If traffic must travel over the SAMI, it shall be allowed to cool

and speed controlled as not to exceed 25 miles per hour.

6.4.6 Texas Department of Transportation

The Crack attenuating mixture (CAM) is a pavement layer composed of a hot

compacted mixture of aggregate and asphalt binder mixed in a mix plant. It is not

allowed to use reclaimed asphalt pavement (RAP) in the CAM mixture (76).

6.4.6.a Asphalt Binder

It is specified to use a conventional PG-graded asphalt binder based on AASHTO

M-320 according to the contract requirements.

6.4.6.b Mineral Aggregates

The mineral aggregate shall meet the properties specified in Table 6.13. The

coarse aggregate stockpiles must have no more than 20% material passing the No. 8

sieve. The fine aggregate consist of manufacture sands. Natural sands are not allowed.

Fine aggregate stockpiles must meet the gradation requirements in Table 6.14. The fine

aggregate shall be supplied free from organic impurities. If 10% or more of the fine

aggregate stockpile is retained on the No.4 sieve, test the stockpile and verify that it
 150

meets the requirements in Table 6.13 for coarse aggregate angularity and flat and

elongated particles. Lime shall be added as mineral filler at a rate of 1% by weight of the

total dry aggregate.

6.4.6.c Tack Coat

A base PG-graded binder should be used with a minimum high-temperature grade

of PG58.

6.4.6.d Mixture Design

The mixture shall be designed according to the Superpave design procedure given

in Tex-204-F, Part IV. The target laboratory density should be 98.0% at Ndes = 50. The

engineer will approve the target asphalt percentage based on acceptable results from the

Hamburg Wheel and Overlay tests. The mixture gradation specification and volumetric

properties of the mixture are shown in Table 6.15. The mixture design properties are

shown in Table 6.16. Additives such lime or antistrip liquid can be added to the mixture

if required.

6.4.6.e Construction Requirements

Prepare the surface by removing raised pavement markers and objectionable

material such as moisture, dirt, leaves and other loose materials. Remove vegetation

from pavement edges. Place the mixture to meet the typical section requirements and

produce a smooth, finished surface with a uniform appearance and texture. Offset

longitudinal joints of successive courses of hot mix by at least 6 inch. Place mixture so

longitudinal joints on the surface coincide with lane lines, or as directed. Ensure all

finished surface will drain properly. Place mixture within the compacted lift thickness

shown in Table 6.17 unless otherwise shown on the plans.

 151

Place the mixture when the roadway surface temperature is equal or higher than

temperatures listed in Table 6.18 unless otherwise approved. The surface shall be

cleaned before applying the tack coat. The tack coat rate shall be between 0.04 to 0.10

gal. of residual asphalt per square yard of surface area. Apply a thin, uniform tack coat to

all contact surfaces of curbs, gutter, and structures. The mixture shall be compacted to

achieve a maximum in-place air voids value of 4.0%.

6.4.7 Utah Department of Transportation

The Utah Department of Transportation (UDOT) specification covers the

materials and construction requirements for producing and placing a reflective cracking

relief bituminous mixture. Such stress absorbing membrane interface is a highly elastic,

impermeable hot mix interlayer that is designed to reduce reflective cracking for

underlying pavements. The SAMI should be placed in one lift and covered with an HMA

overlay (77).

The reflective crack relief bituminous mixture shall meet the general requirements

specified for a HMA mixture, except as modified herein.

6.4.7.a Asphalt Binder

The asphalt binder should meet the requirements of AASHTO MP-1 with a PG

high temperature of 64 °C or higher and a PG low temperature of -34 °C or lower as

required to meet the Hveem stability and the flexural beam fatigue mix requirements, in

addition to the following requirements in Table 6.19.

6.4.7.b Blended Aggregate

The blended aggregate shall consist of natural sands and crushed fines. The

aggregate gradation should meet the ranges specified in Table 6.20. The blended
 152

aggregate sand equivalent should have a minimum value of 45% as determined by

AASHTO T 176.

6.4.7.c Mixture Design

The mixture design should use fifty gyrations (Nmax = 50) for gyratory

compaction. The mixture should be aged for beam testing for 4 hours at 135 °C in

accordance to AASHTO PP2-99 Section 7.2 (mechanical property testing), prior to

compact the beams. The volumetric properties of the SAMI should meet the

specifications in Table 6.21.

6.4.7.d Construction Requirements

Immediately prior to applying the SAMI, thoroughly clean the surface of all

vegetation, loss materials, dirt, mud, visible moisture and other objectionable materials.

Fill the joints that are larger than 0.5 inch wide as determined by the engineer. Prior to

the placement of the SAMI, fill large surface deformities (greater than 3 inch deep and 4

feet in diameter) with approved mix. During the placement of the SAMI, fill smaller

pavement deformities, with the reflective crack bituminous mixture.

A tack coat should be applied between the HMA layer and the SAMI at typical

rates of 0.02 to 0.04 gallons per square yard (undiluted tack). The SAMI should not be

placed when the temperature is below 50°F. To reduce the occurrence of blisters, do not

place the SAMI on a wet surface or within 24 hours of rain.

The SAMI should be placed in an average thickness of 1 inch with a tolerance of

± 1/4 inch. The longitudinal joints should be overlapped by at least 6 inch to eliminate

construction joints over the existing longitudinal joints. The SAMI should not be heated

above 350°F.
 153

The compaction operations should start promptly after placement of the SAMI.

Only steel wheel rollers in static mode are allowed for compaction of the mixture. The

density of the SAMI should be within 97 ± 2% of the maximum specific gravity as

determined by AASHTO T 209.

The SAMI should be covered with the binder course within five days after

placement. It should be opened to traffic or covered with the HMA overlay after cooling

to less than 140°F.

6.5 PROPOSED PLAN FOR PHASE III

Based on NDOT’s positive experience with stress relief courses and the review of

the states’ specifications for stress relief courses, it is recommended to conduct an

extensive laboratory evaluation for the Texas DOT and the UTAH DOT SRC designs

using Nevada’s materials during 2008. The following summarizes the major steps

recommended to be completed in the Phase III of the NDOT study:

• Select two aggregate sources with different mineralogy and absorption: one in
northern Nevada and one in southern Nevada.
• Identify three categories of binders to be used with each of the aggregate sources.
All binders must meet the PG64-28 grade for the northern part and the PG76-22
grade for the southern part.
o Polymer modified asphalt binder
o Tire rubber terminal blend asphalt binder
o Crumb rubber asphalt binder
• For each combination of aggregate source and asphalt binder conduct a mix
design according to the TxDOT and UDOT stress relief courses.
• Evaluate the mechanical properties of the SRC mixes in the laboratory in terms of
their dynamic modulus, fatigue resistance, rutting resistance using the RLT,
thermal cracking resistance using the TSRST, reflective cracking resistance using
the TTI Upgraded Overlay tester, and moisture sensitivity.
• Conduct a Type 2C NDOT Hveem Mix design for each of the aggregate sources
and the polymer modified asphalt binder.
• Evaluate the mechanical properties of the NDOT T2C mixes in the laboratory in
terms of their dynamic modulus, fatigue resistance, rutting resistance using the
 154

RLT, thermal cracking resistance using the TSRST, reflective cracking resistance
using the TTI Upgraded Overlay tester, and moisture sensitivity.
• Optimize the gradation of the NDOT T2C HMA mixes.

Based on the results of the laboratory evaluation, recommendations will be made

for constructing field test sections with SRC mixes in 2009. The overlay thickness will

be designed using the Rubber Pavements Association Overlay Design Model. Field

mixtures from the SRC and overlay mixes will be collected during construction from

behind the paver and evaluated for asphalt binder content and gradation. Additionally,

the field mixtures will be evaluated in terms of their dynamic modulus, fatigue resistance,

rutting resistance using the RLT, thermal cracking resistance using the TSRST, reflective

cracking resistance using the TTI Upgraded Overlay tester, and moisture sensitivity.

The performance of the field test sections will be monitored and field cores will

be sampled for evaluation in the TTI Overlay Tester. Based on the laboratory evaluation

and the field performance the specifications for reflective cracking resistance will be

adjusted and the most effective technique will be selected.

 155

REFERENCES
1. Dempsey, B. J. (2002). “Development and Performance of Interlayer Stress-
Absorbing Composite in Asphalt Concrete Overlays,” In Transportation Research
Record 1809, TRB, National Research Council, pp. 175-183.
2. Freeman T. E. (2002). “Evaluation of Concrete Slab Fracturing Techniques in
Mitigating Reflective Cracking Through Asphalt Overlays,” Report VTRC 03-R3,
Virginia Transportation Research Council, Charlottesville, Virginia.
3. NCHRP Research Project 1-37A, Development of the 2002 Guide for the Design of
New and Rehabilitated Pavement Structures, Currently underway.
4. Scofield, L. A. (1989). “The History, Development, and Performance of Asphalt
Rubber at ADOT,” Report Number AZ-SP-8902, Arizona DOT.
5. Way, G. B. (1999). “Flagstaff I-40 Asphalt-Rubber Overlay-Nine Years of success,”
Presented at the 78th Annual Meeting of the Transportation Research Board.
6. The I-40 in Flagstaff, Arizona, The successful Asphalt-Rubber project that inspires
engineers from around the world. [Link]/newsletter/fall2003/.
7. Trevino, M., Dossey, T., McCullough, F., and Yildirim, Y. (2003). “Applicability of
Asphalt Concrete Overlays on Continuously Reinforced Concrete Pavements.” CTR
Research Report 4398-1, Center for Transportation Research, Bureau of Engineering
Research, The University of Texas at Austin.
8. Rahman, M., Scofield, L., and Wolf, T. (1989). “Paving Fabrics for Reducing
Reflective Cracking,” Report No. FHWA-AZ-8801, Arizona Transportation Research
Center, Prepared for Arizona Department of Transportation.
9. Amini, F. (2005). “Potential Applications of Paving Fabrics to Reduce Reflective
Cracking,” Report No. FHWA/MS-DOT-RD-05-174, Jackson State University,
Prepared for Mississipi Department of Transportation.
10. Woodside, A. R., McIlhagger, R., Woodward, W. D. H., and Clements, H. W. (1996).
”Inhibiting Reflective Cracking,” University of Arizona, Department of Civil
Engineering and Engineering Mechanics, Tucson, AZ.
11. Woodside, A. R., Woodward, W. D. H., Clements, H. W., Sikich, J., and Russell, T.
(1997). “The Use of Fabrics to Inhibit Reflective Cracking in Porous Asphalt,”
European Conference on Porous Asphalt, Madrid, Spain, Volume I, pp. 271-83.
12. Predoehl, N. H. (1989). “Use of Paving Fabric Test Installations in California,” Final
Report, California Department of Transportation, Translab, 1989.
13. Banasiak, D. (1997). “Fiberglass Mesh Reduces Reflective Cracking on California
Highway,” Road & Bridges, Vol. 35, Issue 3, pp.54.
14. FHWA (1982). “Report on Performance of Fabrics in Asphalt Overlays,” Office of
Highway Operation, Experimental Application and Evaluation Branch.
15. Shuler, S., and Hermelink, D. (2004). “Reducing Reflection Cracking in Asphalt
Pavements,” Cracking in Pavements, Mitigation, Risk Assessment and Prevention,
Proceedings of the Fifth International RILEM Conference, France, pp. 451-458.
16. Gulden, W., and Brown, D. (1984). “Overlays for Plain Jointed Concrete Pavements,”
Research Project No. 7502, Office of Materials and Research, Georgia Department of
Transportation.
 156

17. Gulden, W., and Brown, D. (1985). “Treatments for Reduction of Reflective
Cracking of Asphalt Overlays on Jointed-Concrete Pavements in Georgia,” In
Transportation Research Record 1007, TRB, National Research Council,
Washington, D.C., pp. 26-36.
18. Brown, D. R., Jared, D., Jones, C., and Watson, D. (1997). “Georgia’s Experience
with Crumb Rubber in Hot-Mix Asphalt,” In Transportation Research Record 1583,
TRB, National Research Council pp. 45-51.
19. Mascunana, I. (1981). “An Evaluation of Engineering Fabric in Pavement
Rehabilitation (IHD-21),” Final Report, Physical Research Report No. 88, Illinois
Department of Transportation.
20. Buttlar, W. G., Bozkurt, D. and Dempsey, B. J. (1999). “Evaluation of Reflective
Crack Control Policy,” Final Report, Report No. ITRC FR 95/96-4, Illinois
Transportation Research Center, Edwardsville, IL, p. 184.
21. Vespa, J. (2005). “An Evaluation of Interlayer Stress Absorbing Composite (ISAC)
Reflective Crack Relief System,” Final Report No. FHWA/IL/PRR 150, Illinois
Department of Transportation, Bureau of Materials & Physical Research, IL.
22. Jiang, Y., and McDaniel, R. S. (1993). “Application of Cracking and Seating and Use
of Fibers to Control Reflective Cracking,” In Transportation Research Record 1388,
TRB, National Research Council pp. 150-159.
23. Rasoulian, M., Becnel, B., and Keel, G. (2000). “Stone Interlayer Pavement Design,”
In Transportation Research Record 1709, TRB, National Research Council, pp. 60-
68.
24. Glasgrid Pavement Reinforcement System. “System Overview,” Tensar International
Corporation.
25. 2006 Standard Specifications for Roads & Bridges Manual, Section 507 Asphaltic
Surface Treatment, Louisiana Department of Transportation.
26. V.T. Barnhart (1989). “Field Evaluation of Experimental Fabrics to Prevent
Reflective Cracking in Bituminous Resurfacing,” Report No. R-1300. Materials &
Technology Division, Michigan Transportation Commission, Lansing, Michigan.
27. Kidd, S. Q. (1990). “Paving Fabric and Asphalt Stress Absorbing Membrane
Interlayers (SAMI),” Final Report, Report No. MSHD-RD-90-67-6, Mississippi State
Highway Department Research and Development Division, Jackson, MS.
28. Maurer, D. A. and Malashekie G. J. (1989). “Field Performance of Fabrics and Fibers
to Retard Reflective Cracking,” In Transportation Research Record 1248, TRB,
National Research Council, pp. 13-23.
29. Hughes, J. J., and Somers, E. (2000). “Geogrid Mesh for Reflective Crack Control in
Bituminous Overlays,” Final Report, Report No. PA 200-013-86-001, Pennsylvania
Department of Transportation.
30. Morian, D. A., Oswalt, J., and Deodhar A. (2004). “Experience with Cold In-Place
Recycling as a Reflective Crack Control Technique: Twenty Years Later,” In
Transportation Research Record 1869, TRB, National Research Council, pp. 47-55.
31. Cleveland G. S., Button J. W., and Lytton R. (2002). “Geosynthetics in Flexible and
Rigid Pavement Overlay Systems to Reduce Reflection Cracking,” Research Report
1777-1, Project Number 0-1777, Texas Transportation Institute.
 157

32. Darling, J.R., and Woolstencroft, J.H. (2004). “Fiberglass Pavement Reinforcement
Used in Dissimilar Climatic Zones for Retarding Reflective Cracking in Asphalt
Overlays,” Cracking in Pavements, Mitigation, Risk Assessment and Prevention,
Proceedings of the Fifth International RILEM Conference, France, pp. 435-442.
33. Chen, D. H., Scullion, T., and Bilyeu, J. (2006). “Lessons Learned on Jointed
Concrete Pavement Rehabilitation Strategies in Texas,” In Journal of Transportation
Engineering, ASCE, Volume 132, Issue 3, pp. 257-265.
34. Freeman, T. E. (2002). “Evaluation of Concrete Slab Fracturing Techniques in
Mitigating Reflective Cracking Through Asphalt Overlays,” Report VTRC 03-R3,
Virginia Transportation Research Council.
35. Bischoff, D., and Toepel, A. (2003). “GlassGrid Pavement Reinforcement Product
Evaluation,” Final Report FEP-03-03, WisDOT Research study # WI-89-03,
Wisconsin Department of Transportation.
36. Abu Al-eis, K. (2004). “Evaluation of the Interlayer Stress-Absorbing Composite
(ISAC) to Mitigate Reflective Cracking in Asphaltic Concrete Ovelarys,”
Construction Report WI-09-04, Wisconsin Department of Transportation.
37. Makowski, L., Bischoff, D., Blankenship, P., Sobczak, D., and Haulter F. (2005).
“Wisconsin Experiences with Reflective Crack Relief Projects,” In Transportation
Research Record 1905, TRB, National Research Council, pp. 44-55.
38. Vanelstraete A., and De Visscher J. (2004). “Long Term Performance on Site of
Interface Systems,” Cracking in Pavements, Mitigation, Risk Assessment and
Prevention, Proceedings of the Fifth International RILEM Conference, France, pp.
699-706.
39. Tighe, S., Haas, R., and Ponniah, J.. (2003). “Life Cycle Cost Analysis of Mitigating
Pavement Rehabilitation Reflection Cracking,” In Transportation Research Record
1823, TRB, National Research Council, pp. 73-79.
40. Valtonen, J., and Hyyppa, I. (2004). “Field Study of Repair Methods for Transversal
Cracks,” Cracking in Pavements, Mitigation, Risk Assessment and Prevention,
Proceedings of the Fifth International RILEM Conference, France, pp. 555-562.
41. Flexible Pavements Distress Identification Manual, Nevada DOT, Materials Division,
Carson City, Nevada.
42. Lewis, A., and Collins, D. (1999). “Cold In-Place Recycling: A Relevant Process for
Road Rehabilitation and Upgrading,” 7th Conference on Asphalt Pavements for
Southern Africa.
43. Fernandez, George. Data Mining using SAS Applications. Chapman & HALL/CRC,
Washington, D.C, 2003.
44. Kuelh, R., Design of Experiments: statistical Principles of Research Design and
Analysis. Second Edition, 2000. pp175-222.
45. Elseifi, M., and Al-Qadi, I. (2003). “A Simplified Overlay Design Model Against
Reflective Cracking Utilizing Service Life Prediction,” Paper No. 03-3285 presented
at the TRB 82nd Annual Meeting, Washington, D.C.
46. Jacobs, M. M., Hopman, P. C., and Molenaar, A. A. A. (1996). “Application of
Fracture Mechanics Principles to Analyze Cracking in Asphalt Concrete.” In
Proceeding Annual Meeting of the Association of Asphalt Paving Technologists,
AAPT, Vol. 65, Baltimore, MD, pp. 1–39.
 158

47. Uzan, J (1997). “Evaluation of Fatigue Cracking,” In Transportation Research

Record 1570, TRB, National Research Council, Washington, D.C., pp. 89-95.
48. BRRC, Belgian Road Research Center (1998). “Design of Overlaid Cement Concrete
Pavements Reinforced with Bitufor® traffic loading.” Research report EP5035/3544,
Brussels, Belgium.
49. Paris, P. C., and Erdogan, F. A. (1963). “Critical Analysis of Crack Propagation
Laws,” In Transactions of the ASME, Journal of Basic Engineering, Series D, No. 3.
50. Mobasher, B., Mamlouk, M. S., and Lin, H-M (1997). “Evaluation of Crack
Propagation Properties of Asphalt Mixtures,” In Journal of Transportation
Engineering, ASCE, Vol. 123, No. 5, pp. 405-413.
51. Kanninen, M. F., and Popelar, C. H. (1985). “Advanced Fracture Mechanics,” Oxford
University Press, Inc., NY.
52. Ullidtz, P. (1987). Pavement Analysis. New York: Elsevier Science.
53. Sousa, J., Pais, J., Saim, R., Way, G., and Stubstad, R. (2001). “Development of a
Mechanistic Overlay Design Method Based on Reflective Cracking Concepts,” Final
Report for Rubber Pavements Association, Consulpav International.
54. Claussen, A.I.M., Edwards, J.M., Sommer, P., and Ugé, P. (1977). “Asphalt
Pavement Design. The Shell Method,” Proceedings, Fourth International Conference
on the Structural Design of Asphalt Pavements, Vol. 1, Ann Arbor, pp. 39-74.
55. Guide for Mechanistic-Empirical Design of New and Rehabilitated Pavement
Structures (2004). Final Report, Part 3 Design Analysis, Ch 6, National Cooperative
Highway Research Program, Transportation Research Board, National Research
Council.
56. Grzybowska, W., Wojtowicz, J., and Fonferko, L. C. (1993). “Application of Geo-
Synthetics to Overlays in Cracow Region of Poland,” Reflective Cracking in
Pavements, State of the Art and Design Recommendations, Proceedings of the Second
International RILEM Conference, Liege, Belgium, pp. 290-298.
57. Livneh, M., Ishai, I., and Kief, O. (1993). “Bituminous Pre-Coated Geotextile Felts
for Retarding Reflection Cracks,” Reflective Cracking in Pavements, State of the Art
and Design Recommendations, Proceedings of the Second International RILEM
Conference, Liege, Belgium, pp. 343-350.
58. Di Benedetto, H, Neji, J, Antoine, J. P, and Pasquier, M. (1993). “Apparatus for
Laboratory Study of Cracking Resistance,” Reflective Cracking in Pavements, State
of the Art and Design Recommendations, Proceedings of the Second International
RILEM Conference, Liege, Belgium, pp. 179-186.
59. Tschegg, E. K, Stanzl- Tschegg, S. E., and Litzka, J. (1993). “New Testing Method to
Characterize Mode 1 Fracturing of Asphalt Aggregate Mixtures,” Reflective Cracking
in Pavements, State of the Art and Design Recommendations, Proceedings of the
Second International RILEM Conference, Liege, Belgium, pp. 263-270.
60. Dumas, Ph., and Vecoven, J. (1993). “Processes Reducing Reflective Cracking;
Synthesis of Laboratory Tests,” Reflective Cracking in Pavements, State of the Art
and Design Recommendations, Proceedings of the Second International RILEM
Conference, Liege, Belgium, pp. 246-253.
 159

61. Gibney, A., Lohan, G., and Moore, V. (2002). “Laboratory Study of Resistance of
Bituminous Overlays to Reflective Cracking,” In Transportation Research Record
1809, TRB, National Research Council pp. 184-190.
62. British Standard Institution (1998). “Sampling and Examination of Bituminous
Mixtures for Road and Other Paved Areas: Methods of Test for the Determination of
Wheel Tracking Rate,” British Standard 598, Part 110, London, England.
63. Dempsey, B. J. (2002). “Development and Performance of Interlayer Stress
Absorbing Composite (ISAC) in AC Overlays”. In Transportation Research Record
1809, TRB, National Research Council pp. 175-183.
64. Montestruque, G., Rodrigues, R., Nods, M., and Elsing, A. (2004). “Stop of
Reflection Crack Propagation with the Use of Pet Geogrid as Asphalt Overlay
Reinforcement,” Cracking in Pavements, Mitigation, Risk Assessment and
Prevention, Proceedings of the Fifth International RILEM Conference, Lomoges,
France, pp. 231-238.
65. Sobhan, T., Crooks, T., Tandon, T., and Mattingly, S. (2004). “Laboratory Simulation
of the Growth and Propagation of Reflection Cracks in Geogrid Reinforced Asphalt
Overlays,” Cracking in Pavements, Mitigation, Risk Assessment and Prevention,
Proceedings of the Fifth International RILEM Conference, Lomoges, France, pp. 589-
596.
66. Gallego, J., and Prieto, J. N. (2006). “Development of New Laboratory Equipment for
the Study of Reflective Cracking in Asphalt Overlays,” In Asphalt Overlays and
Infrastructure Distress. CD-ROM. Transportation Research Board of the National
Academies, Washington, D.C.
67. Tamagny, P., Wendling, L., and Piau, J. M. (2004). “A New Explanation of Pavement
Cracking from Top to Bottom: The Visco-elasticity of Asphalt Materials,” Cracking
in Pavements, Mitigation, Risk Assessment and Prevention, Proceedings of the Fifth
International RILEM Conference, Lomoges, France, pp. 425-432.
68. Zhou, F., Scullion, T., and Williammee, R. (2004). “Upgraded TTI Overlay Tester: A
Simple Reflective Cracking Simulation Test,” In Pavement Management, Design, and
Testing. CD-ROM. Transportation Research Board of the National Academies,
Washington, D.C.
69. Zhou, F., and Scullion, T. (2005). “Overlay Tester: A Rapid Performance Related
Crack Resistance Test,” Texas Transportation Institute, the Texas A&M University
System, Report FHWA/TX-05/0-4667-2, Austin, Texas.
70. Núñez, P. W., Ceratti, J. A. P., Theisen, K. M., and Morhila, A. (2006). “Using APT
to Compare the Efficiency of Overlay Materials Used in Brazil,” In Bituminous
Materials, Pavement Management, International. CD-ROM. Transportation Research
Board of the National Academies, Washington, D.C.
71. Standard Specifications for Highway Construction (2000). Arizona Department of
Transportation, ADOT, Sections 404, 410 and 1009.
[Link]
72. Asphalt Rubber Usage Guide Division of Engineering Services (2003). California
Department of Transportation, Caltrans,
[Link]
.pdf.
 160

73. Standard Specifications for Highway Construction (2005). Florida Department of

Transportation, FDOT, Sections 336, 341 and 916,
[Link]
74. Standard Specifications for Highway Construction (2006). Iowa Department of
Transportation, IDOT, Sections 2303, 2310 and 4127.
[Link]
75. Standard Specifications for Highway Construction (1995). Massachusetts Department
of Transportation (MDOT), Section 466,
[Link]
76. Standard Specifications for Highway Construction (2004). Texas Department of
Transportation TxDOT, CSJ 0353-04-090, Section 3141.
77. Standard Specifications for Highway Construction (2002). Utah Department of
Transportation, UDOT, Section 02744S,
[Link]
 161

TABLES
 162

Table 3.1 List of Selected NDOT Projects.

Contract Route County Location Treatment DOC AADT

2191 US095 NY Southern NV CIR April 1990 5,500
2385 US093 EL Northern NV CIR April 1990 5,250
2385 US093 EL Northern NV CIR April 1990 5,250
2428 US050 EU Northern NV CIR April 1991 1,900
2808 US050 WP Northern NV CIR April 1997 2,950
2808 US050 EU Northern NV CIR April 1997 2,950
2838 SR396 PE Northern NV CIR April 1997 1,100
2935 SR360 MI Southern NV CIR March 1998 800
2819 US095 NY Southern NV CIR May 1997 5,550
2873 US095 NY Southern NV CIR May 1998 5,550
2961 US006 ES Southern NV CIR March 1999 2,000
3013 US095A LY Northern NV CIR May 2001 14,500
3025 US093 WP Northern NV CIR June 2000 5,000
3025 US093 LN Northern NV CIR June 2000 1,000
3025 SR208 LN Northern NV CIR June 2000 1,000
2876 SR208 LY Northern NV CIR June 2000 1,500
2761 SR443 WA Northern NV Reinforced Fabrics June 1996 9600
2932 US095 MI Southern NV Reinforced Fabrics Feb. 1999 2850
2980 US050 CH Northern NV Reinforced Fabrics June 1999 1000
2980 US095 CH Northern NV Reinforced Fabrics June 1999 4400
3006 IR080 HU Northern NV Reinforced Fabrics March 2000 7200
3008 SR227 EL Northern NV Reinforced Fabrics May 2000 10000
2723 US095 CL Southern NV Stress Relief course May 1997 40000
3031 US395 WA Northern NV Stress Relief course June 2000 15600
3048 SR157 CL Southern NV Stress Relief course Nov. 2000 2700
3045 US050 EU Northern NV Stress Relief course June 2001 1900
3162 US395 WA Northern NV Stress Relief course June 2003 29100
2384 US095 CC Northern NV Mill & Overlay April 1990 40,000
2384 US095 LY Northern NV Mill & Overlay April 1990 5,500
2432 SR157 CL Southern NV Mill & Overlay May 1991 2,700
2505 US095 MI Southern NV Mill & Overlay June 1992 3,350
2651 US095 ES Southern NV Mill & Overlay July 1994 2,000
2651 US095 ES Southern NV Mill & Overlay July 1994 1,700
2651 US095 ES Southern NV Mill & Overlay July 1994 1,700
2679 US095 ES Southern NV Mill & Overlay Feb. 1996 6,750
3028 SR512 CC Northern NV Mill & Overlay June 2000 7250
2070 SR160 NY Southern NV Mill & Overlay May 2005 7500
2886 IR080 LA Northern NV Crack & Seat June 1998 6,750
2889 IR080 EL Northern NV Crack & Seat July 1998 10,800
2962 IR080 EL Northern NV Crack & Seat May 1999 4,700
2999 IR080 EL Northern NV Crack & Seat Dec. 1999 6,200
3021 IR080 WA Northern NV Crack & Seat June 2000 23,000
2544 IR080 EL Northern NV Rubblized PCCP June 1993 7000
2869 IR080 EL Northern NV Rubblized PCCP April 1998 6550
2901 IR080 HU Northern NV Rubblized PCCP Sep. 1998 7200
3088 IR080 PE Northern NV Rubblized PCCP Sep. 1998 7700
3186 IR080 HU Northern NV Rubblized PCCP Oct. 2003 6300
3186 IR080 LA Northern NV Rubblized PCCP Dec. 2001 6300
 163

Table 3.2a Pavement Distresses Summary of Selected NDOT Projects.

Pavement Distresses Before Treatment Pavement Distresses After Treatment Application

Date of Construction
Application (Severity) (Severity/Time in Years to Develop)

Treatment Life
(North/South)
Fatigue
Treatment

Contract Fatigue cracking

Location
County
cracking Block Cracking Type Block Cracking Type

AADT
Route

Transverse

Transverse
Type
Type

Cracks

Cracks
A B A B C A B A B C

CIR-A-1 2808a WP US050 N 1998 7 2,950 Sev - Mod Min - - - - Min/7 - - -

Cold In-Place Recycling (CIR)

CIR-A-2 2808b EU US050 N 1997 8 2,950 Min Sev Mod - - Min - - Min/2 - - -
CIR-A-3 2838 PE SR396 N 1999 6 1,100 - - Sev Sev Sev Sev - - Min/1 - - -
CIR-A-4 2935 MI SR360 S 1999 6 800 Sev Mod Min Min Mod Mod - - - - - -
CIR-B-1 2819 NY US095 S 1998 7 5,550 - Sev Mod - Min Min Min/3 - Min/2 - - -
CIR-B-2 2873 NY US095 S 1998 7 5,550 Min - Min Min Sev Min Min/7 - Min/7 - - -
CIR-B-3 2961 ES US006 S 1999 6 2,000 Min - Min Sev Sev Sev - - - - - -
CIR-B-4 3013 LY US095 N 1999 6 14,500 Min - Min Mod - - - - Min/6 - - -
CIR-C-1 3025a WP US093 N 2003 2 5,000 Min Min Min - - - - - Min/1 - - -
CIR-C-2 3025b LN US093 N 2001 4 1,000 Min Mod Min Mod Mod Min - - Min/1 - - -
CIR-C-3 3025c LN SR208 N 2001 4 1,000 Min Min Min Mod Mod Min - - Min/1 - - -
CIR-C-4 2876 LY SR208 N 2001 4 1,500 Min - Sev Mod - Min - - Min/1 - - -
RF-1 2761 WA SR443 N 1999 6 9600 Min Sev Mod Sev Mod - Min/5 - Min/1 - - -
Fabric (RF)
Reinforced

RF-2 2932 MI US095 S 1999 6 2850 Min Min Sev - - - - - - - - -

RF-3 2980a CH US050 N 2000 5 1000 Min Min Sev - Min Min - - - - - -
RF-4 2980b CH US095 N 2000 5 4400 Min Min Min - - - - - Min/3 - - -
RF-5 3006 HU IR080 N 2001 4 7200 Min Min Min - - - - - - - - -
RF-6 3008 EL SR227 N 2001 4 10000 Min Min Min - - - - - Min/2 - - -
SRC-1 2723 CL US095 S 1997 8 40000 Mod Sev Mod Mod Min - - - - - - -
Stress Relief

SRC-2 3031 WA US395 N 2000 5 15600 Min Min Mod Sev - - - - Min/5 - - -
Course
(SRC)

SRC-3 3048 CL SR157 S 2000 5 2700 Mod - Mod - - - - - Min/5 - - -

SRC-4 3045 EU US050 N 2001 4 1900 Min Mod Mod - - - - - - - - -
SRC-5 3162 WA US395 N 2003 2 29100 - - Min - - - - - - - - -

Min Denotes “Minor”, Mod Denotes “Moderate”, Sev Denotes “Severe”

 164

Table 3.2b Pavement Distresses Summary of Selected NDOT Projects (Continued).

Pavement Distresses Before Treatment Pavement Distresses After Treatment Application

Date of Construction
Application (Severity) (Severity/Time in Years to Develop)

Treatment Life
(North/South)
Fatigue
Treatment

Contract Fatigue cracking

Location
County
cracking Block Cracking Type Block Cracking Type

AADT
Route

Transverse

Transverse
Type
Type

Cracks

Cracks
A B A B C A B A B C

MOL-A-1 2384a CC US095 N 1990 15 40,000 - Sev Min Mod - - Mod/1 - Mod/1 - - -
Mill and Overlay (MOL)

MOL-A-2 2384b LY US095 N 1993 12 5,500 Mod Sev Mod Min - - Min/4 - Min/2 - - -
MOL-B-1 2432 CL SR157 S 1993 12 2,700 Mod - Mod - - - - - Min/5 - - -
MOL-B-2 2505 MI US095 S 1993 12 3,350 - Sev - Mod - Sev - - - - - -
MOL-C-1 2651a ES US095 S 1995 10 2,000 - Mod - Mod - Min - - - Min/5 - -
MOL-C-2 2651b ES US095 S 1996 9 1,700 - - - - Sev Sev - - Min/5 - - -
MOL-C-3 2651c ES US095 S 1996 9 1,700 Mod - Mod Mod Sev Sev Min/5 - Min/5 - - -
MOL-C-4 2679 ES US095 S 1997 8 6,750 Mod - Min Mod Min - - - Mod/1 Min/6 - -
MOL-B-3 3028 CC SR512 N 2000 5 7250 - - Mod - - - - - Min/3 - - -
MOL-C-5 3070 NY SR160 S 2003 2 7500 Min Min Min Mod Mod - - - - - - -
CS-1 2886 LA IR080 N 1998 7 6,750 - - - - - - Min/6 - Min/4 - - -
Seat PCCP
Crack &

CS-2 2889 EL IR080 N 1999 6 10,800 - - - - - - Min/4 - - - - -

(CS)

CS-3 2962 EL IR080 N 1999 6 4,700 - - Min - - - Min/6 - - - - -

CS-4 2999 EL IR080 N 2001 4 6,200 - - - - - - Min/4 - - - - -
CS-5 3021 WA IR080 N 2001 4 23,000 - - - - - - Min/4 - - - - -
RPCC-1 2549 EL IR080 N 1996 9 7000 - - Min - - - Min/5 - - - - -
RPCC-2 2869 EL IR080 N 1999 6 6550 - - - - - - Mod/4 - - - - -
Rubblized
PCCP (R)

RPCC-3 2901 HU IR080 N 1999 6 7200 - - - - - - Min/6 - - - - -

RPCC-4 3088 PE IR080 N 2002 3 7700 - - - - - - - - - - - -
RPCC-5 3186a HU IR080 N 2005 0 6300 - - - - - - - - - - - -
RPCC-6 3186b LA IR080 N 2005 0 6300 - - - - - - - - - - - -

Min Denotes “Minor”, Mod Denotes “Moderate”, Sev Denotes “Severe”

 165

Table 3.3a Ranking of Selected NDOT Projects Based on Principal Component Analysis.

Ranking+
Trt. Cont. ID Description of Treatment AADT DOC Life Loc.
-1# +1# -1# +3# -1# +5#
2808a CIR-A-1 CIR 2.0” + 2.5” DGHMA* + 0.75” OGFC 2,950 1998 7 N 27 1 26 1 20 1
2808b CIR-A-2 CIR 2.0” + 2.5” DGHMA + 0.75” OGFC 2,950 1997 8 N 17! 1 19! 14! 21 12!
2838 CIR-A-3 CIR 2.0” + 2.5” DGHMA + 0.75” OGFC 1,100 1999 6 N 33 2 35 2 32 5
2935 CIR-A-4 CIR 2.0” + 2.5” DGHMA + 0.75” OGFC 800 1999 6 S 31 1 33 1 29 1
2819 CIR-B-1 CIR 3.0” + 3.0” DGHMA + 0.75” OGFC 5,550 1998 7 S 24 1 27 13! 26 13!
2873 CIR-B-2 CIR 3.0” + 3.0” DGHMA + 0.75” OGFC 5,550 1998 7 S 30 1 31 1 30 1
CIR

2961 CIR-B-3 CIR 3.0” + 3.0” DGHMA + 0.75” OGFC 2,000 1999 6 S 36 1 38 1 35 1
!
3013 CIR-B-4 CIR 3.0” + 3.0” DGHMA (AC-20P) + 0.75” OGFC 14,500 1999 6 N 16 1 18 1 18 1
3025a CIR-C-1 CIR 2.0” + 2.0” DGHMA (AC-20P) + chip seal 5,000 2003 2 N 8! 3 NA NA$ NA NA
3025b CIR-C-2 CIR 2.0” + 2.0” DGHMA (AC-20P) + chip seal 1,000 2001 4 N 23 5 25 1 NA NA
3025c CIR-C-3 CIR 2.0” + 2.0” DGHMA (AC-20P) + chip seal 1,000 2001 4 N 25 5 28 8 NA NA
2876 CIR-C-4 CIR 2.0” + 2.0” DGHMA + 0.75” OGFC 1,500 2001 4 N 28 5 29 8 NA NA
! ! ! !
2761 RF-1 Cold milling 2.0” + fiberglass + 2.0” Type II DGHMA (AC-20P) + 0.75” OGFC 9,600 1999 6 N 37 6 37 8 34 7
2932 RF-2 Cold milling 2.0” + fiberglass + 2.0” Type II DGHMA (AC-20P) + 0.75” OGFC 2,850 1999 6 S 21 1 22 1 23 1
2980a RF-3 Cold milling 2.0” + fiberglass + 2.0” Type II DGHMA (AC-20P) + 0.75” OGFC 1,000 2000 5 N 13 1 15 1 17 1
RF

2980b RF-4 Cold milling 2.0” + fiberglass + 2.0” Type II DGHMA (AC-20P) + 0.75” OGFC 4,400 2000 5 N 9 1 9 2 8 3
3006 RF-5 Cold milling 2.0” + fiberglass + 2.0” Type II DGHMA (AC-20P) + 0.75” OGFC 7,200 2001 4 N 10 1 10 1 NA NA
3008 RF-6 Cold milling 2.0” + fiberglass + 2.0” Type II DGHMA (AC-20P) + 0.75” OGFC 10,000 2001 4 N 7 1 7 4 NA NA
2723 SRC-1 Cold milling 2.0” + 1” SRC + 2.0” Type II DGHMA (AC-30) + 0.375” OGFC 40,000 1997 8 S 32 1 32 1 28 1
3031 SRC-2 Cold milling 2.0” + 1” SRC + 2.0” Type II DGHMA (AC-30) + 0.375” OGFC 15,600 2000 5 N 29 1 30 1 27 10
SRC

3048 SRC-3 Cold milling 2.0” + 1” SRC + 2.0” Type II DGHMA (AC-30) + 0.375” OGFC 2,700 2000 5 S 14 1 16 1 16 2
3045 SRC-4 Cold milling 2.0” + 1” SRC + 2.0” Type II DGHMA (AC-30) + 0.375” OGFC 1,900 2001 4 N 15 1 17 1 NA NA
3162 SRC-5 Cold milling 2.0” + 1” SRC + 2.0” Type II DGHMA + 0.375” OGFC 29,100 2003 2 N 4 1 NA NA NA NA

+
Ranking from best to worst; i.e., the projects ranked as “1” had the best performance.
#
-1: Previous year to construction, +1, +3, +5: one, three, and five years after construction.
$
NA: Project is younger than the indicated long-performance year.
* Denotes “dense graded HMA”
!
Considered as an outlier according to the “univar” statistical analysis. Therefore, not included in the calculation of the mean ranking of each treatment.
 166

Table 3.3b Ranking of Selected NDOT Projects Based on Principal Component Analysis (Continued).

Ranking+
Trt. Cont. ID Description of Treatment AADT DOC Life Loc.
-1# +1# -1# +3# -1# +5#
2384a MOL-A-1 Cold milling 1.0” + 1.0” Type III DGHMA* (AC-10) + 0.75” OGFC 40,000 1990 15 N 19 12 21 12 24 12
2384b MOL-A-2 Cold milling 1.0” + 1.0” Type III DGHMA (AC-10) + 0.75” OGFC 5,500 1993 12 N 22 1 24 5 22 9
2432 MOL-B-1 Cold milling 1.0” + 1.0” Type III DGHMA (AC-20) + 0.75” OGFC 2,700 1993 12 S 20 1 20 1 19 12
2505 MOL-B-2 Cold milling 1.0” + 1.0” Type III DGHMA (AC-20) + 0.75” OGFC 3,350 1993 12 S 35 1 34 1 31 1
2651a MOL-C-1 Cold milling 1.5” + 1.5” Type II DGHMA (AC-20P) + 0.75” OGFC 2,000 1995 10 S 18 1 23 1 25 11
MOL

2651b MOL-C-2 Cold milling 1.5” + 1.5” Type II DGHMA (AC-20P) + 0.75” OGFC 1,700 1996 9 S 34 1 36 1 33 6
2651c MOL-C-3 Cold milling 1.5” + 1.5” Type II DGHMA (AC-20P) + 0.75” OGFC 1,700 1996 9 S 38 1 39 1 36 15
2679 MOL-C-4 Cold milling 1.5” + 1.5” Type II DGHMA (AC-20P) + 0.75” OGFC 6,750 1997 8 S 3! 5! 3! 6! 4! 6!
3028 MOL-B-3 Cold milling 1.0” + 1.0” Type III DGHMA (AC-20P) + 0.75” OGFC 7,250 2000 5 N 11 1 11 3 14 4
$
3070 MOL-C-5 Cold milling 1.5” + 1.5” Type II DGHMA (AC-20P) + 0.75” OGFC 7,500 2003 2 S 26 1 NA NA NA NA
Ranking for rigid techniques
2886 CS-1 CS 8” slabs + 1.5” leveling course + 4.5” Type II DGHMA (AC-20P) + 0.375” OGFC 6,750 1998 7 N 1 1 1 1 1 3
2889 CS-2 CS 8” slabs + 1.5” leveling course + 3.5” Type II DGHMA (AC-20P) + 0.375” OGFC 10,800 1999 6 N 1 1 1 1 1 2
CS

2962 CS-3 CS 8” slabs + 1.5” leveling course +4.75” Type II DGHMA (AC-20P) + 0.375” OGFC 4,700 1999 6 N 2 1 2 1 3 1
2999 CS-4 CS 8” slabs + 1.5” leveling course + 3.5” Type II DGHMA (AC-20P) + 0.375” OGFC 6,200 2001 4 N 1 1 1 1 1 2
3021 CS-5 CS 6” slabs + 1.5” leveling course + 3.5” Type II DGHMA (AC-20P) + 0.375” OGFC 23,000 2001 4 N 1 1 1 1 1 2
2549 RPCC-1 Rubblizing + 1.5” leveling course + 5.0” Type II DGHMA + 0.375” OGFC 7,000 1996 9 N 3 1 3 1 7 4
2869 RPCC-2 Rubblizing + 1.5” leveling course + 5.0” Type II DGHMA + 0.375” OGFC 6,550 1999 6 N 1 1 1 1 1 6
2901 RPCC-3 Rubblizing + 1.5” leveling course + 7.0” Type II DGHMA (AC-20P) + 0.375” OGFC 7,200 1999 6 N 1 1 1 1 1 1
RPCC

Rubblizing + reinforced fabric & non-woven geotextile + 1.5” leveling course + 5.0”
3088 RPCC-4 7,700 2002 3 N 1 1 1 1 NA NA
Type II DGHMA (AC-20P) + 0.375” OGFC
3186a RPCC-5 Rubblizing + 2.0” leveling course + 4.5” Type II DGHMA (AC-20P) + 0.375” OGFC 6,300 2005 0 N 1 1 1 1 NA NA
3186b RPCC-6 Rubblizing + 2.0" leveling course + 4.5” Type II DGHMA (AC-20P) + 0.375” OGFC 6,300 2005 0 N 1 1 1 1 NA NA

+
Ranking from best to worst; i.e., the projects ranked as “1” had the best performance.
#
-1: Previous year to construction, +1, +3, +5: one, three, and five years after construction.
$
NA: Project is younger than the indicated long-performance year.
* Denotes “dense graded HMA”
!
Considered as an outlier according to the “univar” statistical analysis. Therefore, not included in the calculation of the mean ranking of each treatment.
 167

Table 3.4 Statistical Analysis of the Various Treatments Based on the PCA Rankings at One-year
Before Treatment Construction and One-year After Treatment Construction for Flexible Pavements.

1-year Pre-construction 1-year after

No. of Projects Change in Relative
Treatment (-1) Construction (+1)
Analyzed Ranking
Average STD* Average STD*
CIR-A 3 30 3 1 1 29
CIR-B 3 30 6 1 0 29
CIR-C 3 25 3 5 0 20
RF 5 12 5 1 0 11
SRC 5 19 12 1 0 18
MOL-A 2 21 2 7 8 14
MOL-B 3 22 12 1 0 21
MOL-C 4 29 9 1 0 28
* Standard deviation

Table 3.5 Statistical Analysis of the Various Treatments Based on the PCA Rankings at One-year
Before Treatment Construction and Three-years After Treatment Construction for Flexible Pavements.

1-year Pre-construction 3-year after

No. of Projects Change in Relative
Treatment (-1) Construction (+3)
Analyzed Ranking
Average STD* Average STD*
CIR-A 3 31 5 1 1 30
CIR-B 3 29 10 1 0 28
CIR-C 3 27 2 6 4 22
RF 5 13 6 2 1 11
SRC 4 24 8 1 0 23
MOL-A 2 23 2 9 5 14
MOL-B 3 22 12 2 1 20
MOL-C 3 33 9 1 0 32
* Standard deviation
 168

Table 3.6 Statistical Analysis of the Various Treatments Based on the PCA Rankings at One-year
Before Treatment Construction and Five-years After Treatment Construction for Flexible Pavements.

1-year Pre-construction 5-year after

No. of Projects Change in Relative
Treatment (-1) Construction (+5)
Analyzed Ranking
Average STD* Average STD*
CIR-A 3 27 6 2 2 25
CIR-B 3 28 9 1 0 27
CIR-C 0 NA NA NA NA NA
RF 3 16 8 2 1 14
SRC 3 24 7 4 5 19
MOL-A 2 23 1 11 2 13
MOL-B 3 21 9 6 6 16
MOL-C 3 31 6 11 5 21

* Standard deviation
 169

Table 3.7 Fixed Qualitative Factorial Analysis Based on the PCA Rankings at One-year
After Treatment Construction.

Treatment p-value (!) Comment

(*) (#)
Reflective Cracking Treatment (RCT) <0.0001 Highly Significant
Location (LOC) <0.0001 Highly Significant
RCT*Location <0.0001 Highly Significant
Traffic (Tr) <0.0001 Highly Significant
RCT × Tr <0.0001 Highly Significant
LOC × Tr <0.0001 Highly Significant
RCT × LOC × Tr <0.0001 Highly Significant
(*) The reflective cracking treatments are: Cold in-place recycling, reinforced fabrics, stress-relief course, mill and overlay, PCC rubblization and PCC crack
and seat.
(!) A p-value lower than 0.05 indicates a significant treatment.
(#) If the probability value is <0.0001, it indicates a highly significant treatment.
 170

Table 3.8 Fixed Qualitative Factorial Analysis Based on the PCA Rankings at Three-years After Treatment Construction.

Treatment p-value (!) Comment

Reflective Cracking Treatment (RCT)(*) <0.0001 (#) Highly Significant
Location (LOC) <0.0001 Highly Significant
RCT*Location <0.0001 Highly Significant
Traffic (Tr) <0.0001 Highly Significant
RCT × Tr <0.0001 Highly Significant
LOC × Tr <0.0001 Highly Significant
RCT × LOC × Tr <0.0001 Highly Significant
(*) The reflective cracking treatments are: Cold in-place recycling, reinforced fabrics, stress-relief course, mill and overlay, PCC rubblization and PCC crack
and seat.
(!) A p-value lower than 0.05 indicates a significant treatment.
(#) If the probability value is <0.0001, it indicates a highly significant treatment.
 171

Table 3.9 Fixed Qualitative Factorial Analysis Based on the PCA Rankings at Five-years
After Treatment Construction.

Treatment p-value (!) Comment

(*) (#)
Reflective Cracking Treatment (RCT) <0.0001 Highly Significant
Location (LOC) <0.0001 Highly Significant
RCT*Location <0.0001 Highly Significant
Traffic (Tr) <0.0001 Highly Significant
RCT × Tr <0.0001 Highly Significant
LOC × Tr <0.0001 Highly Significant
RCT × LOC × Tr <0.0001 Highly Significant
(*) The reflective cracking treatments are: Cold in-place recycling, reinforced fabrics, stress-relief course, mill and overlay, PCC rubblization and PCC crack
and seat.
(!) A p-value lower than 0.05 indicates a significant treatment.
(#) If the probability value is <0.0001, it indicates a highly significant treatment.
 172

Table 3.10 Fixed Qualitative-Quantitative Factorial Analysis Based on the PCA Rankings at One-year After Treatment Construction.

Effect p-value (!) Comment

Reflective Cracking Treatment (RCT)(*) <0.0001 (#) Highly Significant
PCA-1 0.1776 Not Significant
2
(PCA-1) <0.0001 Highly Significant
RCT × PCA-1 <0.0001 Highly Significant
RCT × (PCA-1)2 <0.0001 Highly Significant
R-Square 0.77
(*) The reflective cracking treatments are: Cold in-place recycling, reinforced fabrics, stress-relief course, mill and overlay, PCC rubblization and PCC crack
and seat.
(!) A p-value lower than 0.05 indicates a significant treatment. If the probability value is <0.0001, it indicates a highly significant treatment.
(#) If the probability value is <0.0001, it indicates a highly significant treatment.
 173

Table 3.11 Fixed Qualitative-Quantitative Factorial Analysis Based on the PCA Rankings at Three-years After Treatment Construction.

Effect p-value (!) Comment

Reflective Cracking Treatment (RCT)(*) <0.0001 (#) Highly Significant
PCA-1 0.8130 Not Significant
(PCA-1)2 <0.0001 Highly Significant
RCT × PCA-1 0.0002 Highly Significant
RCT × (PCA-1)2 <0.0001 Highly Significant
R-Square 0.71
(*) The reflective cracking treatments are: Cold in-place recycling, reinforced fabrics, stress-relief course, mill and overlay, PCC rubblization and PCC crack
and seat.
(!) A p-value lower than 0.05 indicates a significant treatment. If the probability value is <0.0001, it indicates a highly significant treatment.
(#) If the probability value is <0.0001, it indicates a highly significant treatment.
 174

Table 3.12 Fixed Qualitative-Quantitative Factorial Analysis Based on the PCA Rankings at Five-years After Treatment Construction.

Effect p-value (!) Comment

Reflective Cracking Treatment (RCT)(*) <0.0001 (#) Highly Significant
PCA-1 0.2576 Not Significant
(PCA-1)2 < 0.0001 Significant
RCT × PCA-1 <0.0001 Highly Significant
RCT × (PCA-1)2 <0.0001 Highly Significant
R-Square 0.71
(*) The reflective cracking treatments are: Cold in-place recycling, reinforced fabrics, stress-relief course, mill and overlay, PCC rubblization and PCC crack
and seat.
(!) A p-value lower than 0.05 indicates a significant treatment. If the probability value is <0.0001, it indicates a highly significant treatment.
(#) If the probability value is <0.0001, it indicates a highly significant treatment.
 175

Table 4.1 Plane Strain Fracture Toughness (KIc) for Different HMA Mixtures (50).

Binder type Temperature (°C) Binder content (%) Modulus (GPa) KIc (MPa/m0.5)
4 12.48 1.00
–7 5 17.65 1.17
Asphalt
6 11.93 1.04
cement (AC-
4 7.10 0.77
20)
–1 5 9.44 0.91
6 6.62 0.90
Asphalt 7 4.65 0.67
rubber (80% –7 8 3.85 0.80
AC-20 + 20% 9 5.99 0.74
reclaimed 7 3.87 0.60
rubber by –1 8 4.21 0.79
weight) 9 4.79 1.06

Table 4.2 Statistical Coefficients for the εVM Model.

i a1i a2i b1i b2i

1 -1.038E-04 -1.446E-01 7.169E-03 1.314E-01
2 2.777E-01 -4.022E+00 9.773E-05 -6.36E-01
3 -1.173E+00 1.212E+01 -4.946E-01 7.069E00
4 1.281E+00 5.070E-01 3.923E-02 2.641E00
5 -5.160E-01 6.964E+00 3.265E-02 -1.287E00
6 -1.775E-01 2.385E+00 1.875E-03 -8.167E-01

Table 4.3 Minimum and Maximum Values for the Pavement Thicknesses and Moduli.

i Xi Minimum Maximum
1 Thickness of the existing cracked layer (m) 0.10 0.25
2 Thickness of the granular layer (m) 0.2 0.40
3 Modulus of the overlay layer (MPa) 2000 10000
4 Modulus of the existing cracked layer (MPa) 2000 3500
5 Modulus of the granular layer (MPa) 150 450
6 Modulus of the subgrade layer (MPa) 50 150

Table 4.4 Traffic Information.

Year ESALs*
1 172,565
3 554,780
5 992,375
10 2,384,235
15 4,336,390
20 7,074,390
* Truck factor = 0.912, annual growth rate = 7%, directional distribution factor = 50%, lane distribution factor =
 176
90%

Table 4.5 HMA Overlay Dynamic Modulus at 70°F and 10 Hz.

Type of Mix Eoverlay, ksi (MPa)

T2C 790 (5,455)
CT 1,045 (7,205)
NRM 1,375 (9,485)

Table 4.6 HMA Overlay Thickness Using the New AASHTO Model.

HMA overlay
RC (%)* a+ b+ Time (years) Traffic (ESALs)
thickness (inch)
2 5.00 -2.48 0.83 142,382
4 6.50 -1.64 2.17 390,329
5.0
8 9.50 -1.19 5.50 1,112,109
10 11.00 -1.10 7.33 1,583,897
12 12.50 -1.04 9.23 2,137,788
2 5.00 -2.48 1.13 196,153
10.0
4 6.50 -1.64 2.63 479,910
8 9.50 -1.19 6.13 1,267,196
10 11.00 -1.10 8.01 1,774,625
12 12.50 -1.04 9.95 2,368,123
15.0
2 5.00 -2.48 1.32 229,999
4 6.50 -1.64 2.91 536,776
8 9.50 -1.19 6.52 1,366,568
10 11.00 -1.10 8.44 1,897,188
20.0
12 12.50 -1.04 10.40 2,516,472
2 5.00 -2.48 1.46 255,765

100
* Percent of cracks reflected, RC =
1 + e a +bt
+
a = 3.5 + 0.75×hac, b = – 0.688584 – 3.37302×(hac)-0.915469, hac HMA overlay thickness.
 177
Table 5.1 Results of the Bending Test under Repeated load.

Structure Sample ID Number of Average Standard Coefficient

cycles to crack deviation of variation
1.A 890
Without 2.A 990
952 57 6%
geotextile 3.A 1,010
4.A 920
1.B 2,100
With 2.B 2,740
1,810 752 42%
geotextile 3.B 1,170
4.B 1,230

Table 5.2 Results of the Shearing Test.

Structure Sample ID Maximum Shear Maximum Shear

Force (N) Stress (MN/m2)
1.C 1,237.4 0.21
2.C 1,202.1 0.21
Without 3.C 1,661.7 0.28
geotextile 4.C 2,262.7 0.37
5.C 2,086.0 0.37
6.C 1,909.2 0.33
Average 1,726.5 0.30
Standard Deviation 401.8 0.068
Coefficient of Variation 23% 23%
1.D 1,173.8 0.20
2.D 735.4 0.13
With 3.D 827.3 0.15
geotextile 4.D 824.4 0.14
5.D 586.9 0.11
6.D 756.6 0.13
Average 819.1 0.14
Standard Deviation 178.4 0.029
Coefficient of Variation 22% 20%

Table 5.3 Comparison of Different Mitigation Techniques.

Crack speed in the

Anti-reflective Crack initiation Breaking time
first centimeter
cracking system time (minutes) (minutes)
( μm/min)
Paving fabric 170 90 520
Rich binder mix 95 65 535
SAMI 155 85 580
 178
Table 5.4 Geogrid effectiveness factor (FEG).

Pre-crack Nf(B) Nf(s)

Beam Cf (Cycles-1) Nf (Cycles) FEG
opening (Cycles) (Cycles)
-5
Without geogrid 79,884 93,290 3.40×10 29,450
3 mm 6.14
With geogrid 490,491 573,560 5.53×10-6 180,970
Without geogrid 68,690 77,710 4.03×10-5 24,820
6 mm 4.60
With geogrid 329,393 346,400 8.81×10-6 113,510
Without geogrid 63,020 72,920 4.33×10-5 23,100
9 mm 5.11
With geogrid 340,702 364,530 8.42×10-6 118,740

Table 5.5 Three Test Cells of MnRoad: Asphalt Mixture and Cracking Performance.

Test Cell Asphalt Type Mix Design Linear Feet of Overlay

(Marshall) Cracking in Tester Results
Field
15 PG 64-22 75 blow 475 91
18 PG 64-22 50 blow 315 153
20 PG58-28 35 blow 100 500
 179

Table 5.6 Summary of Laboratory Testing Methods for Reflective Cracking.

Geometry
Type of Failure
Testing device of Type of load Test results Findings Schematic
test mode
specimens
Cracow Bending I & II Beams: - Static bending load: - Static bending: The bending test under
University of or 12×3×3 inch loading rate 0.47 cracking time, max repeated load indicates that
Technology, Shearing in/min. force & bending HMA overlays reinforced
Poland (56) - Repeated haversine strength. with geotextiles exhibited a
bending load: 5 Hz. - Dynamic bending: greater resistance to the crack
- Static shearing: loading number of repetitions. development. The shearing
rate 4×10-2 in/min. - Static shear: max shear test showed that the presence
force & stress. of a geotextile diminished
more than two times the
adhesion between the asphalt
layers.
Technion-Israel Wheel I Beams: - Cyclic wheel load. - Number of wheel The evaluated geotextile
Institute of tracking 28×4×4 inch loading repetitions to fabric had a resistance to
Technology, device failure. reflective cracking 4 times
Israel (57) - Crack length versus greater than other techniques.
number of repetitions
and testing time.
Geo-materials Uniaxial I Slabs - Static: rate of 0.05 to - Measure of energy A comparision between the
Laboratory, tension 0.22 in/hr. transmitted by an fissurometer and field results
ENTPE, France - Cyclic uniaxial. ultrasonic wave train. from different cracking
- Fissurometer mitigation techniques showed
(58) that the device classified
them in reverse order
(opposite to field
performance). It might be
because the fissurometer only
simulates thermal shrinkage.
 180

Table 5.6 Summary of Laboratory Testing Methods for Reflective Cracking (continued).

Geometry
Type of Failure
Testing device of Type of load Test results Findings Schematic
test mode
specimens
Technical Splitting I Cubical or - Static: loading rate 0.05 - Horizontal force versus The researchers concluded
University of prismatic in/min. displacement. that the maximum splitting
Vienna, Austria - Maximum vertical force is not an appropriate
- Wedge force versus parameter to differentiate
splitting (59) temperature. between HMA mixes since
- Fracture energy versus two different mixes can have
temperature. the same maximum splitting
force and different fracture
behavior. On the other hand,
the specific fracture energy
was recommended as a more
reliable testing parameter to
differentiate between various
mixes.
Laboratory of Biaxial I & II Beams: - Cyclic vertical load: 1 - Crack initiation time & The test results of a 2.4 inch
Public Roads, 24×2.8×2.8 Hz. length. (6 cm) AC 0/10 overlay on
France (60) inch - Static horizontal load: - Crack propagation time top of a fabric interlayer, a
0.024 in/hr. & length. rich-binder HMA mix, and a
- Breaking time. Stress Absorbing Membrane
Interlayer (SAMI). It was
concluded that the paving
fabrics delays the crack
initiation time, while the rich
HMA mix slows down the
crack propagation.
University Wheel I & II Beams - Cyclic wheel load: 21 - Number of wheel No tests were still performed.
College of tracking - Bottom-up cycles/min. loading repetitions
Dublin - device 5.5×11×2.0 versus crack length.
accelerated inch. - Deformation of the
simulative - Top-down slabs over the central 8
wheel tracking 5.5×10.2×2.0 inch throughout the test. Bottom-up cracking Top-down cracking
(61) inch.
 181

Table 5.6 Summary of Laboratory Testing Methods for Reflective Cracking (continued).

Type of Failure Geometry of Findings

Testing device Type of load Test results Schematic
test mode specimens
University of Uniaxial I HMA layer - Cyclic uniaxial load: - Strain in HMA overlay The Interlayer Stress
Illinois, U.S.A tension on top of a frequency 0.0016 in/min as function of test Absorbing Composite
(63) PCC slab of (triangular). cycles. (ISAC) had a much better
6×90×2.7 - Crack length versus performance than other
inch time commercial products when
was tested in the proposed
test device.
Aeronautical Bending I & II Beams: - Sinusoidal load: loading - Permanent strain versus The HMA overlay reinforced
Technological or 18×6×3 inch frequency 20 Hz. number of load cycles. with geogrid had a life up to
N.A.
Institute, ATI, Shearing - Tensile stress versus 6 times higher than a HMA
Brazil (64) crack length. overlay without it.
Florida Bending I Beams: - Static - The load value or At the same load ratio, the
Atlantic (single 18×6×7.5 - Cyclic (Sinusoidal) load: number of repetitions to slabs having geogrids
University, point of inch 2 Hz. first reflected crack. embedded at the bottom
U.S.A (65) loading) - The load value or showed better resistance to
number of repetitions reflection cracking compared
for crack propagation to to specimens in which the
half way of overlay. geogrids was simply attached
- The load value or to the bottom with a tack
number of repetitions coat. Additionally, it was
for crack propagation to found that geogrid embedded
top of overlay. at mid-height was more
effective than geogrid
embedded at the bottom of
overlay.
Polytechnic Biaxial I & II Beams: - Cyclic wheel load. - Vertical length of the Three treatments were
University of 12×12×2.4 - Static traction force: crack with time. studied: HMA overlay
Madrid, Spain - inch 0.001 to 50 μm/hr. - Vertical displacement without geotextile, and
Wheel with time reinforced with two different
Reflective - Relative movement types of geotextile The
Cracking between crack edges. overlay without reinforcing
device (66) showed the worst
performance.
 182

Table 5.6 Summary of Laboratory Testing Methods for Reflective Cracking (continued).

Testing device Type of Failure Geometry of Type of load Test results Findings
Schematic
test mode specimens
Regional Biaxial I & II Beams: - Static (horizontal load) - Number of repetitions No tests were performed.
Laboratory of 2×2×26 inch - Cyclic (vertical load): versus vertical force or
Pont et sinusoidal 10 Hz. dissipated energy.
Chausses, - Number of repetitions
France - versus crack length.
MEFISTO (67)
Texas Biaxial I & II Cores: 6-inch - Triangular cyclic load: - Number of repetitions The major findings of the
Transportation diameter 10 seconds/cycle. versus crack length. extensive experimental work
Institute Beams: - Number of repetitions indicated that a very good
Overlay Tester 6×3×2 inch versus testing time. repeatability for the device.
(68) It was found to be sensitive
to the testing temperature,
opening displacement,
asphalt binder content and
grade, and air voids. Also it
showed consistency between
the mixtures’ test results and
their corresponding field
performance. The pass/fail
criterion on reflective
cracking resistance is 300
cycles at 77 ºF (25 ºC) and
0.025 in (0.64 mm) opening
displacement. When a rich
bottom layer is used, the
reflective cracking life in the
overlay tester should be at
least 750 cycles.
 183

Table 6.1 Summary of Literature Review.

Treatment Description Performance

Cold in-place Remove and mill the upper layers of the Promising performance for roads with up to
recycling existing pavement with specialized recycling 13,000 ADT and 200,000 annual equivalent
equipment then mix with virgin materials to single axle loads.
produce a strong flexible base course.
Glassgrid Geosynthetic material consisting of connected Benefits in retarding or preventing reflective
parallel sets of intersecting ribs with openings cracking are not clear. Field performance has
of sufficient size. varied from excellent to very poor. Concerns
when used on rough surfaces.
Fabric interlayer Geosynthetic comprised solely of textiles. A Effective when used for load-related fatigue
paving fabric interlayer provides the generally distress. It did not performed well when used to
acknowledged functions of a stress-absorbing delay or retard thermal cracking.
interlayer and a waterproofing membrane. The Optimum performance highly associated with
stress-related performance has been easily proper construction procedures. The key factor
verified by the observed reductions of is proper tack-coat installation. In general,
cracking in pavement overlays. overlays reinforced with fabrics have shown
better performance than unreinforced overlays
under same conditions.
Asphalt rubber Asphalt rubber chip seal overlaid with Reduce and/or delay reflective cracking for a
interlayer + thin conventional dense graded HMA or gap period of 5 years.
overlay (about graded HMA.
1.5”)
Stress absorbing A thin layer placed between an underlying Successful in reducing the rate of reflective
membrane pavement and an HMA overlay for the cracking.
interlayer purpose of dissipating movements and stresses
at a crack in the underlying pavement before
they create stresses in the overlay. SAMIs
consist of a spray application of rubber or
polymer-modified asphalt as the stress-
relieving material, followed by placing and
seating aggregate chips.
Crumb rubber Produced by adding ground tire rubber to Ranged from successful to devastating failures
overlay HMA using the wet process. depending on percent of crumb rubber in mix.
 184

Table 6.2 Summary of NDOT Reflective Cracking Mitigation Techniques Review.

Treatment Description Application Conditions Performance

Traffic Pre-rehabilitation
Pavement Condition
Cold in-place CIR of minimum top 2.0” of Up to No severe alligator Stopped reflective cracking
recycling existing HMA materials and 14,000 cracking. for 5 years after
(CIR) overlaying it with a minimum AADT. construction
of 2.5” dense graded HMA
mixture.
Reinforced Cold milling 2.0” of existing Between No severe alligator Retarded reflective cracking
fabric (RF) HMA layer, placing fiberglass 1,000 and cracking. for at least 3 years after
yarns, and overlaying with 10,000 construction and reduced
2.0” Type II (1 inch maximum AADT. the rate of reflected
size) dense graded HMA. transverse cracks 5 years
after construction.
Stress relief Cold milling 2.0” of existing Up to NA Stopped reflective cracking
course (SRC) HMA layer, placing a 1” 40,000 for 3 years after
stress relieve course and AADT. construction.
overlaying with 2.0” Type II Rate of reflected transverse
(1” max size) dense graded cracks accelerated 5 years
HMA. after construction.
Mill and Cold milling 1.0” of existing Up to NA Reflected fatigue and
overlay HMA pavement and 40,000 transverse cracks 1 to 2
(MOL) overlaying it by 1.0” HMA AADT. years after construction.
mixture manufactured with an
AC-10 asphalt binder.
Cold milling 1.0” of existing Up to 4,000 NA Stopped reflective cracking
HMA pavement and AADT. for 3 years after
overlaying it by 1.0” HMA construction. Minor
mixture manufactured with an reflected transverse cracks 5
AC-20P asphalt binder. years after construction.
Cold milling 1.5” of existing Up to 2,000 NA Stopped reflective cracking
HMA pavement and AADT. for 3 years after
overlaying it by 1.5” HMA construction. Minor
mixture manufactured with an reflected transverse cracks 5
AC-20P asphalt binder. (*) years after construction. (*)

* This treatment was placed on pavements with a condition worse than the condition of the pavements
where the other two mill and overlay treatments were applied.
 185

Table 6.3 Pavement Layers Material Properties.

Modulus
Thickness
Layers at 70°F Fatigue characteristics*
(inch)
(ksi)
2.0374
NDOT ⎛1⎞
4.4142
⎛ 1 ⎞
-- #
790 N f = 1.3740 × 10 × ⎜ ⎟ −5 ⎜ ⎟
T2C ⎝ε ⎠ ⎜E ⎟
⎝ overlay ⎠
2.0598
HMA ⎛1⎞
4.3172
⎛ 1 ⎞
CT -- #
1,045 N f = 6.5815 × 10 × ⎜ ⎟ −5 ⎜ ⎟
overlay ⎝ε ⎠ ⎜E ⎟
⎝ overlay ⎠
2.1470
⎛1⎞
4.2395
⎛ 1 ⎞
NRM -- #
1,375 N f = 6.8745 × 10 −5 × ⎜ ⎟ ⎜ ⎟
⎝ε ⎠ ⎜E ⎟
⎝ overlay ⎠
Cracked HMA 4.0 360 N.A.

Unbound base 10.0 30 N.A.

Subgrade -- 12 N.A.

*Nf is the number of repetitions to failure, ε is the flexural strain in microns

#
to be designed according to all three reflective cracking design methods

Table 6.4 Overlay Design Thicknesses for 20 years Design Period.

Overlay thickness (inches)

HMA Asphalt Rubber Association
Virginia AASHTO MEPDG
overlay mix % reflected cracking
Tech (100% reflected cracks)
0% 2% 5% 15%
NDOT T2C 4.50 3.60 2.40 2.00 0.75 12.00
CT 4.00 4.90 4.00 3.00 1.00 12.00
NRM 3.25 15.25 12.50 9.25 3.25 12.00

Table 6.5 ADOT Rubber Gradation for SAMI (Type A).

Sieve size Percent Passing

No. 8 100
No. 10 95-100
No. 16 0-10
No. 30 --
No. 50 --
No. 200 --
 186

Table 6.6 Required Properties for the Asphalt-rubber.

Specifications
Property
Type 1 Type 2 Type 3
Grade of base asphalt cement PG 64-16 PG 58-22 PG 52-28
Rotational viscosity: 350 °F, Pa.s 1.5 – 4.0 1.5 – 4.0 1.5 – 4.0
Penetration at 4°C, 200 g, 60 s (ASTM D5),
10 15 25
minimum
Softening point (ASTM D 36), °F, minimum 135 130 125
Resilience, 77 °F (ASTM D 5329), % minimum 30 25 15

Table 6.7 Required Properties for the Mineral Aggregates.

Property Test Method Specifications

Los on Abrasion AASHTO T96 9% for 100 revolutions
40% for 500
revolutions
Arizona Test Method
% of carbonates Max. 30
238
% fractured coarse aggregate Arizona Test Method
Min. 70
particles 212
Arizona Test Method
Flakiness Index Max. 25
233
Arizona Test Method
Bulk Oven Dry Specific Gravity 2.30 – 2.85
210
Gradation Arizona Test Method Sieve Size % Passing
201 3/8 inch 100
¼ inch 0-15
No. 8 0-5
No. 200 0-2

Table 6.8 Laboratory Asphalt Rubber Binder Design Data.

Minutes of Reaction 45 minutes

Property Specifications
45 90 240 360 1,440
Limits
Viscosity, Haake at 190°C, Pa.s,
2,400 2,800 2,800 2,800 2,100 1,500-4,000
(10-3), or cP (*)
Resilience at 25°C, % Rebound
27 -- 33 -- 23 18 Minimum
(ASTM D 3407)
Ring & Ball Softening Point, °C
59.0 59.5 59.5 60.0 58.5 52-74
(ASTM D 36)
Cone Pen. at 25°C, 150g, 5 sec.,
39 -- 46 -- 50 25-70
1/10 mm (ASTM D217)

(*) The viscosity test shall be conducted using a hand-held Haake viscometer or equivalent.
 187

Table 6.9 Requirements for CALTRANS SAMI-R’s Construction.

Chip Size (mm) Asphalt Rubber Binder (l/m2) rate Aggregate rate (kg/m2)
9.0, 12.5 2.5 – 3.0 15-22

Table 6.10 Gradations for Ground Tire Rubber.

% Passing
Sieve size
Type A Type B Type C
No. 16 - - 100
No. 30 - 100 70-100
No. 50 100 40-60 20-40
No. 100 50-80 - -

Table 6.11 FDOT Standard Specifications for Asphalt-rubber.

Binder Type ARB 5 ARB 12 ARB 20

Rubber Type A or B (*) B or A (**) C (or A or B) (**)
Min Ground Tire Rubber (by weight of asphalt binder) 5% 12% 20%
Binder Grade PG 67-22 PG 67-22 PG 64-22
Minimum Temperature, °F 300 300 300
Maximum Temperature, °F 335 350 375
Minimum Reaction Time, minutes 10 15 (Type B) 30 (Type C)
Unit Weight at 60°F, lb/gal (***) 8.6 8.7 8.8
Minimum viscosity (****), Poise at 300 °F 4.0 10.0 15.0
* Use of Type B rubber may require an increase in the mix temperature in order to offset higher viscosity
values.
** Use of finer rubber could result in the reduction of the minimum reaction time.
*** Conversions to standard 60ºF are as specified in 300-9.3.
**** FM5-548, Viscosity of Asphalt Rubber Binder by use of the Rotational Viscometer.

Table 6.12 MDOT Standard Specifications for Asphalt-Rubber Gradation.

Sieve Percent Passing

2.36 mm 100
2.00 mm 95-100
1.18 mm -
600 μm 0-10
300 μm 0-5
 188

Table 6.13 Aggregate Quality Requirements.

Property Test Method Requirement

Coarse Aggregate
Deleterious material, %, max Tex-217-F, part I 1.0
Decantation, % max Tex-217-F, part II 1.5
1
Micro-Deval abrasion, % max Tex-431-A
Los Angeles abrasion, % max Tex-410-A 30
Magnesium sulfate soundness, 5 cycles, %, max Tex-411-A 20
Coarse aggregate angularity, 2 crushed faces, %, min Tex-460-A, Part I 952
Flat end elongated particles @5:1, %, max Tex-280-F 10
Fine Aggregate
Linear shrinkage, % max Tex-107-E 3
Combined Aggregate3
Sand equivalent, %, min Tex-203-F 45
1
Not used for acceptance purposes. Used by the engineer as an indicator of the need for further investigation.
2
Only applies to crushed grave.
3
Aggregates, without mineral fillers, or additives, combined as used in the job-mix formula (JMF)

Table 6.14 Gradation Requirements for Fine Aggregate.

Sieve Size % Passing by Weight or Volume

3/8” 100
#8 70-100
#200 0-30

Table 6.15 Master Gradation Bands and Volumetric Properties.

Sieve Size Fine Mixture (% passing by weight or volume)

2” -
1 ½” -
¾” -
½” -
3/8” 98.0-100.0
#4 70.0-90.0
#8 40.0-65.0
#16 20.0-45.0
#30 10.0-30.0
#50 10.0-20.0
#200 2.0-10.0
Property Requirement
Binder Content 6.5% minimum
Design VMA2, % Minimum 16.0
Plant-Produced VMA, % Minimum 15.5
 189

Table 6.16 Mixture Design Properties.

Mixture Property Test Method Requirement

Target laboratory-molded density, % Tex-207-F 98.0
Tensile strength (dry), psi Tex-226-F 85-2001
Dust/asphalt ratio2 - 0.0-1.6 max
Boil test3 Tex-530-C -
Hamburg Wheel Test Requirements
High-Temperature Binder Grade Test Method Minimum # of passes @0.5” Rut Depth,
Tested @ 122°F4
PG 64 or lower 7,000
PG 70 Tex-242-F 15,000
PG 76 or higher 20,000
Overlay Tester Requirements
Cycles to failure Tex-248-F 750 cycles minimum
1
May exceed 200 psi when approved and may be waived when approved.
2
Defined as % passing #200 sieve divided by asphalt content.
3
Used to establish baseline for comparison to production results. May be waived when approved.
4
May be decreased or waived when shown on the plans or when directed.

Table 6.17 Compacted Lift Thickness and Required Core Height.

Compacted Lift thickness Minimum Untrimmed

Mixture Type (inch) Core Height (inch)
Minimum Maximum Eligible for Testing
Crack attenuation mixture (CAM) 1.0 2.0 0.75

Table 6.18 Minimum Pavement Surface Temperatures.

Minimum Pavement Surface Temperatures (°F)

High Temperature
Binder Grade Subsurface Layers or Night Surface Layers Placed in
Paving Operations daylight Operations
PG 64 45 50
PG 70 551 601
PG 66 601 601

Table 6.19 UDOT Specifications for Asphalt Binder.

Property Test Method Criteria

RTFO elastic recovery ASTM D6084 Section 6.2 Min. 45% @ 25°C
Separation Test, ASTM D5976 Section 6.1 Max. 6°C difference after 48 hr
 190

Table 6.20 Aggregate Gradation for SAMI According to UDOT Specifications.

Sieve % Passing
3/8 inch 100
No. 4 80-100
No. 8 60-85
No. 16 40-70
No. 30 25-55
No. 50 15-35
No. 100 8-20
No. 200 6-14

Table 6.21 UDOT Specifications for SAMI.

Property Criteria
Air voids, % 0.5-2.5
Voids in the mineral aggregate (VMA) Min 16.0 min
Hveem stability (AASHTO T 246) at 140°F, 100 mm molds, 50 gyrations Min. 18.0
Flexural beam fatigue-AASHTO T321, 2000 μstrain, 10 Hz, 3±1 air voids at 15°C Min. 100,000 cycles
 191

FIGURES
 192

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1994 1996 1998 2000 2002 2004 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

.

800 800
Block Cracking Type A, ft

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1994 1996 1998 2000 2002 2004 2006 1994 1996 1998 2000 2002 2004 2006
Year Year

1000 100

900 90
.

800 80
Transvers Cracking, ft
Block Cracking Type C, ft^2

700 70

600 60

500 50

400 40

300 30
20
200
10
100
0
0
1994 1996 1998 2000 2002 2004 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

0.500 4.5

4.0
0.400
3.5
Rut Depth, inch

0.300 3.0

2.5
PSI

0.200
2.0

0.100 1.5

1.0
0.000
0.5
1994 1996 1998 2000 2002 2004 2006
0.0
-0.100
1994 1996 1998 2000 2002 2004 2006
Year
Year

Figure 3.1: Performance data of contract 2808a on US050 from mileposts 0 to 3

 193

400 400

350 350

300 300

.
Fatigue Type A, ft

Fatigue Type B, ft^2

250 250

200 200

150 150

100 100

50
50
0
0
1994 1996 1998 2000 2002 2004 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

1000 1000

900 900

.
.

800 800

Block Cracking Type B, ft^2

Block Cracking Type A, ft

700 700

600 600

500 500
400 400

300 300
200 200
100 100
0 0
1994 1996 1998 2000 2002 2004 2006 1994 1996 1998 2000 2002 2004 2006
Year Year

1000 100

900 90
.

800 80
Transvers Cracking, ft
Block Cracking Type C, ft^2

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1994 1996 1998 2000 2002 2004 2006 1994 1996 1998 2000 2002 2004 2006

Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
0.200
2.0

0.150 1.5

0.100 1.0
0.050 0.5
0.000
0.0
1994 1996 1998 2000 2002 2004 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

Figure 3.2: Performance data of contract 2808b on US050 from mileposts 38.04 to 47.39
 194

400 1000
900
350

.
800
300

Block Cracking Type A, ft

700
Fatigue Type A, ft

250 600

200 500
400
150
300
100
200
50 100

0 0
1994 1996 1998 2000 2002 2004 2006 1994 1996 1998 2000 2002 2004 2006
Year Year

1000 1000

900 900
.

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1994 1996 1998 2000 2002 2004 2006 1994 1996 1998 2000 2002 2004 2006
Year Year

1000

900 500
.

800
Transvers Cracking, ft .
Block Cracking Type C, ft^2

700 400

600
300
500
400
200
300

200
100
100

0 0
1994 1996 1998 2000 2002 2004 2006 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
0.200
2.0

0.150 1.5

0.100 1.0
0.050 0.5
0.000
0.0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

Figure 3.3: Performance data of contract 2838 on SR396 from mileposts 1.13 to 7.7
 195

400 400

350 350

.
300 300
Fatigue Type A, ft

250 250

Fatigue Type B, ft^2

200 200

150 150

100 100

50
50
0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

.
FBlock Cracking Type B, ft .

800 800

Block Cracking Type B, ft^2

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1994 1996 1998 2000 2002 2004 2006
Year Year

1000
90
900
.

80
800
.

70
Transvers Cracking, ft
Block Cracking Type C, ft^2

700
60
600
50
500
400 40

300 30

200 20

100 10

0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
0.200
2.0

0.150 1.5

0.100 1.0
0.050 0.5
0.000
0.0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.4: Performance data of contract 2935 on SR360 from mileposts 0 to 23.25
 196

400 600

350
500

.
300
400
Fatigue Type A, ft

Fatigue Type B, ft^2

250

200 300

150 200

100
100
50
0
0
1994 1996 1998 2000 2002 2004 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

1000 1000

900 900

.
.

800 800

Block Cracking Type B, ft^2

Block Cracking Type A, ft

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1994 1996 1998 2000 2002 2004 2006 1994 1996 1998 2000 2002 2004 2006
Year Year

1000

900 100
.

90
800
.

80
Block Cracking Type C, ft^2

700
Transvers Cracking, ft

70
600 60
500 50
400 40

300 30
20
200
10
100
0
0 1994 1996 1998 2000 2002 2004 2006
1994 1996 1998 2000 2002 2004 2006 Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1994 1996 1998 2000 2002 2004 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

Figure 3.5: Performance data of contract 2819 on US095 from mileposts 6.98 to 14.38
 197

400 400

350 350

.
300 300
Fatigue Type A, ft

250 250

Fatigue Type B, ft^2

200 200

150 150

100 100

50
50

0
0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

1000 1000

900 900

.
800 800
.
Block Cracking Type A, ft

700 700

Block Cracking Type B, ft^2

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1994 1996 1998 2000 2002 2004 2006 1994 1996 1998 2000 2002 2004 2006
Year Year

1000 100

900 90
.

80
.

800

700 70
Block Cracking Type C, ft^2

Transvers Cracking, ft

600 60

500 50

400 40

300 30

20
200
100 10

0
0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

Figure 3.6: Performance data of contract 2873 on US095 from mileposts 56.24 to 72.00
 198

400 400

350 350

.
300 300
Fatigue Type A, ft
250

Fatigue Type B, ft^2

250

200 200

150 150

100 100

50
50

0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

.
.

800 800

Block Cracking Type B, ft^2

700 700
Block Cracking Type A, ft

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
.

80
.

800

700 70
Block Cracking Type C, ft^2

Transvers Cracking, ft

600 60

500 50

400 40

300 30

20
200
100 10

0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.7: Performance data of contract 2961 on US006 from mileposts 18.87 to 43.99
 199

400 400

.
350 350

300 300

Fatigue Type B, ft^2

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1998 1999 2000 2001 2002 2003 2004 2005 2006
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

.
800 800
Block Cracking Type A, ft

Block Cracking Type B, ft^2

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1998 1999 2000 2001 2002 2003 2004 2005 2006 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100
.

900 90
.
Block Cracking Type C, ft^2

800 80

700 70
Transvers Cracking, ft

600 60

500 50

400 40

300 30
200 20

100 10
0 0
1998 1999 2000 2001 2002 2003 2004 2005 2006 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1998 1999 2000 2001 2002 2003 2004 2005 2006
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.8: Performance data of contract 3013 on US095A from mileposts 24.55 to 44.18
 200

400 400

350 350

.
300 300

Fatigue Type A, ft

Fatigue Type B, ft^2

250 250

200 200

150 150

100 100

50
50

0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
.

800 80
Block Cracking Type C, ft^2

700 70
Transvers Cracking, ft

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, incj

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.9: Performance data of contract 3025a on US093 from mileposts 0 to 11

 201

400 400

350 350

.
300 300
Fatigue Type A, ft

Fatigue Type B, ft^2

250 250

200 200

150 150

100 100

50
50

0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

.
800 800
Block Cracking Type A, ft

Block Cracking Type B, ft^2

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100
900 90
.

800 80
Block Cracking Type C, ft^2

Transvers Cracking, ft

700 70
600 60
500 50
400 40
300 30
200 20
100 10
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.10: Performance data of contract 3025b on US093 from mileposts 109.33 to
131.55
 202

400 400

350 350

300

.
300
Fatigue Type A, ft

Fatigue Type B, ft^2

250 250

200 200

150 150

100 100

50
50

0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

.
800 800
Block Cracking Type A, ft

Block Cracking Type B, ft^2

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000 0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.11: Performance data of contract 3025c on US093 from mileposts 147.45 to
172.4
 203

400 400

350 350

.
300 300
Fatigue Type A, ft
250

Fatigue Type B, ft^2

250

200 200

150 150

100 100

50
50

0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

.
800 800

Block Cracking Type B, ft^2

700 700
Fatigue Type A, ft

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100
900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft .

700 70
600 60
500 50
400 40
300 30

200 20

100 10

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.12: Performance data of contract 2876 on SR208 from mileposts 2.33 to 8.86
 204

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1992 1994 1996 1998 2000 2002 2004 2006
1992 1994 1996 1998 2000 2002 2004 2006
Year
Year

1000 1000

900 900

.
800 800
Block Cracking Type A, ft

Block Cracking Type B, ft^2

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1992 1994 1996 1998 2000 2002 2004 2006 1992 1994 1996 1998 2000 2002 2004 2006
Year Year

1000 100
900 90
800
Block Cracking Type C, ft^2 .

80
Transvers Cracking, ft

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1992 1994 1996 1998 2000 2002 2004 2006 1992 1994 1996 1998 2000 2002 2004 2006

Year Year

Present Serviciability Index (PSI)

0.500 4.5

0.450 4.0
0.400
3.5
0.350
Rut Depth, inch

0.300 3.0

0.250 2.5
PSI

0.200
2.0
0.150
1.5
0.100
0.050 1.0

0.000
0.5
1992 1994 1996 1998 2000 2002 2004 2006
Year 0.0
1992 1994 1996 1998 2000 2002 2004 2006
Year

Figure 3.13: Performance data of contract 2761 on SR443 from mileposts 0.00 to 0.87
 205

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft^2 .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100
900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft

700 70
600 60
500 50
400
40
300
30
200
20
100
10
0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, in

0.300
2.5
PSI

0.250
2.0
0.200
0.150 1.5

0.100 1.0

0.050 0.5
0.000 0.0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.14: Performance data of contract 2932 on US095 from mileposts 49.24 to 50.41
 206

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft^2

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
80
Block Cracking Type C, ft^2 .

800
.

70
Transvers Cracking, ft

700

600 60
50
500
40
400
30
300
20
200
10
100
0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200
0.150 1.5

0.100 1.0

0.050 0.5
0.000 0.0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.15: Performance data of contract 2980a on US050 from mileposts 19.4 to 21.14
 207

100 400

90 350
80
300

Fatigue Type B, ft^2 .

70
Fatigue Type A, ft
250
60

50 200

40 150
30
100
20
50
10
0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

Block Cracking Type A, ft^2 .

800 800

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft .

700 70

60
600
50
500
40
400
30
300
20
200
10
100
0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5

0.450 4.0
0.400 3.5
0.350
Rut Depth, inch

3.0
0.300
2.5
PSI

0.250
2.0
0.200
0.150 1.5

0.100 1.0

0.050 0.5
0.000 0.0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.16: Performance data of contract 2980b on US095 from mileposts 26.19 to
28.22
 208

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900
Block Cracking Type A, ft^2 .

Block Cracking Type B, ft^2 .

800 800

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100
900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft .

700 70

600 60

500 50
400 40

300 30

200 20

100 10

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, in

0.300
2.5
PSI

0.250
2.0
0.200
0.150 1.5

0.100 1.0

0.050 0.5
0.000 0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.17: Performance data of contract 3006 on IR080 from mileposts 50.0 to 50.23
 209

400 400

350 350

300 300

.
Fatigue Type A, ft

Fatigue Type B, ft^2

250 250

200 200

150 150

100 100

50
50
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900
Block Cracking Type A, ft^2 .

800 800

Block Cracking Type B, ft^2 .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90

800 80
Block Cracking Type C, ft^2 .

Transvers Cracking, ft .

700 70

600 60

50
500
40
400
30
300
20
200
10
100
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5
0.450
4.4
0.400

0.350 4.3
Rut Depth, inch

0.300
4.2
PSI

0.250
0.200 4.1

0.150
4.0
0.100
3.9
0.050
0.000 3.8
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.18: Performance data of contract 3008 on SR227 from mileposts 1.39 to 6.56
 210

140 500

450
120
400

Fatigue Type B, ft^2 .

Fatigue Type A, ft 100 350

300
80
250
60 200

150
40
100
20 50
0
0
1994 1996 1998 2000 2002 2004 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1994 1996 1998 2000 2002 2004 2006 1994 1996 1998 2000 2002 2004 2006
Year Year

1000 100

900 90

80
Block Cracking Type C, ft^2 .

800
.

700 70
Transvers Cracking, ft

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1994 1996 1998 2000 2002 2004 2006 1994 1996 1998 2000 2002 2004 2006
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1994 1996 1998 2000 2002 2004 2006
1994 1996 1998 2000 2002 2004 2006
Year
Year

Figure 3.19: Performance data of contract 2723 on US095 from mileposts 96.89 to 117.5
 211

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

g
1000 100

900 90
Block Cracking Type C, ft^2 .

800 80
.

700
70
Transvers Cracking, ft

600
60
500
50
400
40
300

200 30

100 20

0 10
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
0
Year
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, in

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.20: Performance data of contract 3031 on US395 from mileposts 25.21 to 29.24
 212

100 400

90 350
80
300

Fatigue Type B, ft^2 .

Fatigue Type A, ft 70
250
60

50 200

40 150
30
100
20
50
10
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

800 800

Block Cracking Type B, ft^2 .

Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, in

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.21: Performance data of contract 3048 on SR157 from mileposts 12.47 to 21.1
 213

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

800 800

Block Cracking Type B, ft^2 .

Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90

80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5
0.450
4.4
0.400
4.3
0.350
Rut Depth, inch

0.300 4.2
PSI

0.250 4.1
0.200
4.0
0.150
3.9
0.100

0.050 3.8
0.000
3.7
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.22: Performance data of contract 3045 on US050 from mileposts 36.81 to 37.38
 214

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft
250 250

200 200

150 150

100 100

50
50
0
0
2000 2001 2002 2003 2004 2005 2006
2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
.
Block Cracking Type A, ft

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
2000 2001 2002 2003 2004 2005 2006 2000.5 2001 2001.5 2002 2002.5 2003 2003.5 2004 2004.5 2005 2005.5
Year Year

1000 100

900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0
0
2000.5 2001 2001.5 2002 2002.5 2003 2003.5 2004 2004.5 2005 2005.5
2000.5 2001 2001.5 2002 2002.5 2003 2003.5 2004 2004.5 2005 2005.5
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
2000.5 2001 2001.5 2002 2002.5 2003 2003.5 2004 2004.5 2005 2005.5
2000.5 2001 2001.5 2002 2002.5 2003 2003.5 2004 2004.5 2005 2005.5
Year
Year

Figure 3.23: Performance data of contract 3162on US395 from mileposts 17.37 to 19.4
 215

400 600

350
500
300

Fatigue Type B, ft^2 .

400
Fatigue Type A, ft
250

200 300

150 200

100
100
50
0
0
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Year Year

1000 100

900 90

80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft

700 70

600 60

50
500
40
400
30
300
20
200
10
100
0
0
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
1987 1988 1989 1990 1991 1992 1993 1994 1995 1996
Year
Year

Figure 3.24: Performance data of contract 2384a on US095 from mileposts 14.81 to 16.6
 216

400 400

350 350

300 300

.
Fatigue Type A, ft

Fatigue Type B, ft^2

250 250

200 200

150 150

100 100

50
50
0
0
1986 1988 1990 1992 1994 1996 1998 2000 2002
1986 1988 1990 1992 1994 1996 1998 2000 2002
Year
Year

1000 1000

900 900

800 800

Block Cracking Type B, ft^2 .

Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1986 1988 1990 1992 1994 1996 1998 2000 2002 1986 1988 1990 1992 1994 1996 1998 2000 2002
Year Year

1000 100

900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1986 1988 1990 1992 1994 1996 1998 2000 2002 1986 1988 1990 1992 1994 1996 1998 2000 2002

Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1986 1988 1990 1992 1994 1996 1998 2000 2002
1986 1988 1990 1992 1994 1996 1998 2000 2002
Year
Year

Figure 3.25: Performance data of contract 2384b on US095 from mileposts 0 to 2.74
 217

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1988 1990 1992 1994 1996 1998 2000 2002
1988 1990 1992 1994 1996 1998 2000 2002
Year
Year

1000 1000

900 900

800 800

Block Cracking Type B, ft^2 .

Block Cracking Type A, ft .

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
1988 1990 1992 1994 1996 1998 2000 2002 1988 1990 1992 1994 1996 1998 2000 2002
Year Year

1000 100

900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft

700 70

600 60

500 50
400 40
300 30
200 20
100
10
0
0
1988 1990 1992 1994 1996 1998 2000 2002 1988 1990 1992 1994 1996 1998 2000 2002
Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
Rut Depth, inch

3.0
0.300

0.250
2.5
PSI

0.200 2.0
0.150 1.5
0.100
1.0
0.050
0.5
0.000
1988 1990 1992 1994 1996 1998 2000 2002 0.0
Year 1988 1990 1992 1994 1996 1998 2000 2002
Year

Figure 3.26: Performance data of contract 2432 on SR157 from mileposts 3.41 to 12.18
 218

400 600

350
500
300

Fatigue Type B, ft^2 .

400
Fatigue Type A, ft

250

200 300

150 200

100
100
50
0
0
1988 1990 1992 1994 1996 1998 2000 2002 2004
1988 1990 1992 1994 1996 1998 2000 2002 2004
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1988 1990 1992 1994 1996 1998 2000 2002 2004 1988 1990 1992 1994 1996 1998 2000 2002 2004
Year Year

1000 100

900 90

80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft

70
700
60
600
50
500
40
400
30
300
20
200
10
100
0
0 1988 1990 1992 1994 1996 1998 2000 2002 2004
1988 1990 1992 1994 1996 1998 2000 2002 2004 Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
Rut Depth, inch

3.0
0.300

0.250
2.5
PSI

0.200 2.0
0.150 1.5
0.100
1.0
0.050
0.5
0.000
1988 1990 1992 1994 1996 1998 2000 2002 2004 0.0
Year 1988 1990 1992 1994 1996 1998 2000 2002 2004
Year

Figure 3.27: Performance data of contract 2505 on US095 from mileposts 62.54 to 64.01
 219

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1990 1992 1994 1996 1998 2000 2002 2004
1990 1992 1994 1996 1998 2000 2002 2004
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1990 1992 1994 1996 1998 2000 2002 2004 1990 1992 1994 1996 1998 2000 2002 2004
Year Year

1000 100

900 90

80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft .

700 70

600 60

500 50

400 40
30
300
20
200
10
100
0
0
1990 1992 1994 1996 1998 2000 2002 2004
1990 1992 1994 1996 1998 2000 2002 2004
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1990 1992 1994 1996 1998 2000 2002 2004
1990 1992 1994 1996 1998 2000 2002 2004
Year
Year

Figure 3.28: Performance data of contract 2651a on US095 from mileposts 12 to 15

 220

400 400

350 350

300 300

.
Fatigue Type A, ft

Fatigue Type B, ft^2

250 250

200 200

150 150

100 100

50
50
0
0
1990 1992 1994 1996 1998 2000 2002 2004 2006
1990 1992 1994 1996 1998 2000 2002 2004 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1990 1992 1994 1996 1998 2000 2002 2004 2006 1990 1992 1994 1996 1998 2000 2002 2004 2006
Year Year

1000 100

900 90

80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft

700 70

60
600
50
500
40
400
30
300
20
200
10
100
0
0
1990 1992 1994 1996 1998 2000 2002 2004 2006
1990 1992 1994 1996 1998 2000 2002 2004 2006
Year
Year

0.500
4.5
0.450
4.0
0.400
3.5
0.350
3.0
Prut Depth, in

0.300
2.5
PSI

0.250

0.200
2.0

0.150 1.5

0.100 1.0

0.050 0.5

0.000 0.0
1990 1992 1994 1996 1998 2000 2002 2004 2006
1990 1992 1994 1996 1998 2000 2002 2004 2006
Year
Year

Figure 3.29: Performance data of contract 2651b on US095 from mileposts 43 to 44.8
 221

400
350
350
300
300

.
Fatigue Type A, ft
250

Fatigue Type B, ft^2

250

200 200

150 150

100
100
50
50
0
0 1990 1992 1994 1996 1998 2000 2002 2004 2006
1990 1992 1994 1996 1998 2000 2002 2004 2006 Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1990 1992 1994 1996 1998 2000 2002 2004 2006 1990 1992 1994 1996 1998 2000 2002 2004 2006
Year Year

1000 100

900 90

80
Block Cracking Type C, ft^2 .

800
.

70
Transvers Cracking, ft

700

600 60

500 50

400 40
30
300
20
200
10
100
0
0
1990 1992 1994 1996 1998 2000 2002 2004 2006
1990 1992 1994 1996 1998 2000 2002 2004 2006
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, in

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1990 1992 1994 1996 1998 2000 2002 2004 2006
1990 1992 1994 1996 1998 2000 2002 2004 2006
Year
Year

Figure 3.30: Performance data of contract 2651c on US095 from mileposts 32.88 to 39
 222

400 400

350 350

300 300

.
Fatigue Type A, ft

Fatigue Type B, ft^2

250 250

200 200

150 150

100 100

50
50
0
0
1992 1994 1996 1998 2000 2002 2004 2006
1992 1994 1996 1998 2000 2002 2004 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1992 1994 1996 1998 2000 2002 2004 2006 1992 1994 1996 1998 2000 2002 2004 2006
Year Year

1000 100

900 90
.

800 80
Block Cracking Type C, ft^2

Transvers Cracking, ft

700 70
600 60
500 50
400 40
300 30
200 20
100 10
0 0
1992 1994 1996 1998 2000 2002 2004 2006 1992 1994 1996 1998 2000 2002 2004 2006
Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1992 1994 1996 1998 2000 2002 2004 2006
1992 1994 1996 1998 2000 2002 2004 2006
Year
Year

Figure 3.31: Performance data of contract 2679 on US095 from mileposts 19.96 to 32.88
 223

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft

700 70

600 60

500 50
400 40
300 30
200 20
100 10
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, in

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.32: Performance data of contract 3028 on SR512 from mileposts 0 to 1.38
 224

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1999 2000 2001 2002 2003 2004 2005 2006
1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1999 2000 2001 2002 2003 2004 2005 2006 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
80
Block Cracking Type C, ft^2 .

800
.

700 70
Transvers Cracking, ft

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1999 2000 2001 2002 2003 2004 2005 2006 1999 2000 2001 2002 2003 2004 2005 2006

Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
Ruth Depht, inch

3.0
0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000
0.0
1999 2000 2001 2002 2003 2004 2005 2006
1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.33: Performance data of contract 3070 on SR160 from mileposts SR 160
 225

400 400

350 350

300 300

.
Fatigue Type A, ft

Fatigue Type B, ft^2

250 250

200 200

150 150

100 100

50
50
0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200
0.150 1.5

0.100 1.0

0.050 0.5
0.000 0.0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.34: Performance data of contract 2886 on IR080 from mileposts 3.36 to 9.06
 226

400 400

350 350

300 300

.
Fatigue Type A, ft

Fatigue Type B, ft^2

250 250

200 200

150 150

100 100

50 50

0
0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700

600 600

500 500

400 400

300 300
200 200
100 100

0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft .

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5

0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5

0.000 0.0
1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.35: Performance data of contract 2889 on IR080 from mileposts 26.61 to 32
 227

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50 50

0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700

600 600

500 500

400 400

300 300
200 200
100 100

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90

80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft .

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.450 4.5

0.400 4.0

0.350 3.5
Rut Depth, inch

0.300 3.0

2.5
0.250
PSI

0.200 2.0

1.5
0.150
1.0
0.100
0.5
0.050
0.0
0.000
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

Figure 3.36: Performance data of contract 2962 on IR080 from mileposts 117.68 to
132.72
 228

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50 50

0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

800 800

Block Cracking Type B, ft^2 .

Block Cracking Type A, ft .

700 700

600 600

500 500

400 400

300 300
200 200
100 100

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90

80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft .

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5

0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5

0.000 0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.37: Performance data of contract 2999 on IR080 from mileposts 69.02 to 74.92
 229

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700

600 600

500 500

400 400

300 300
200 200
100 100

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft

700 70

600 60

500 50

400 40

300 30

200 20

100 10

0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5

0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5

0.000 0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.38: Performance data of contract 3021 on IR080 from mileposts 26.77 to 41.99
 230

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1992 1994 1996 1998 2000 2002 2004
1992 1994 1996 1998 2000 2002 2004
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1992 1994 1996 1998 2000 2002 2004 1992 1994 1996 1998 2000 2002 2004
Year Year

1000 100
900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft

700 70

600 60

500 50

400 40
30
300
20
200
10
100
0
0
1992 1994 1996 1998 2000 2002 2004
1992 1994 1996 1998 2000 2002 2004
Year
Year

0.500 4.5
0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200
0.150 1.5

0.100 1.0

0.050 0.5
0.000 0.0
1992 1994 1996 1998 2000 2002 2004 1992 1994 1996 1998 2000 2002 2004
Year Year

Figure 3.39: Performance data of contract 2549 on IR080 from mileposts 11.08 to 20
 231

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

800 800

Block Cracking Type B, ft^2 .

Block Cracking Type A, ft .

700 700
600 600

500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000
100
900
90
Block Cracking Type C, ft^2 ..

800
80
.

700
70
Transvers Cracking, ft

600
60
500
50
400
40
300
30
200
20
100
10
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5

0.450 4.0
0.400 3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5

0.000 0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.40: Performance data of contract 2869 on IR080 from mileposts 1.13 to 7.51
 232

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000
900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700
600 600
500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

1000 100

900 90
Block Cracking Type C, ft^2 .

800 80
Transvers Cracking, ft

700 70

600 60
500 50
400 40
300 30
200 20
100 10
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5

0.450 4.0
0.400 3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5

0.000 0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.41: Performance data of contract 2901 on IR080 from mileposts 11.68 to 17.36
 233

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft
250 250

200 200

150 150

100 100

50
50
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000
900 900
800 800

Block Cracking Type B, ft^2 .

Block Cracking Type A, ft .

700 700
600 600
500 500
400 400
300 300
200 200
100 100
0 0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

100
1000
90
900
80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft

700 70

600 60

500 50

400 40
300 30
200 20
100
10
0
0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5

0.450 4.0
0.400 3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5

0.000 0.0
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.42: Performance data of contract 3088 on IR080 from mileposts 18.59 to 26.21
 234

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100 100

50
50
0
0
2000 2001 2002 2003 2004 2005 2006
2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

Block Cracking Type B, ft^2 .

800 800
Block Cracking Type A, ft .

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
2000 2001 2002 2003 2004 2005 2006 2000 2001 2002 2003 2004 2005 2006
Year Year

1000
100
900
90
800
Block Cracking Type C, ft^2 .

80
Transvers Cracking, ft .

700
70
600
60
500 50
400 40
300 30
200 20
100 10
0 0
2000 2001 2002 2003 2004 2005 2006 2000 2001 2002 2003 2004 2005 2006
Year Year

0.500 4.5

0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000 0.0
2000 2001 2002 2003 2004 2005 2006 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.43: Performance data of contract 3186a on IR080 from mileposts 60.32 to 61.38
 235

400 400

350 350

300 300

Fatigue Type B, ft^2 .

Fatigue Type A, ft

250 250

200 200

150 150

100
100
50
50
0
0
2000 2001 2002 2003 2004 2005 2006
2000 2001 2002 2003 2004 2005 2006
Year
Year

1000 1000

900 900

800 800

Block Cracking Type B, ft^2 .

Block Cracking Type A, ft .

700 700

600 600

500 500
400 400
300 300
200 200
100 100
0 0
2000 2001 2002 2003 2004 2005 2006 2000 2001 2002 2003 2004 2005 2006
Year Year

1000
90
900
80
Block Cracking Type C, ft^2 .

800
Transvers Cracking, ft .

70
700
60
600

500 50

400 40

300 30

200 20
100 10
0
0
2000 2001 2002 2003 2004 2005 2006
2000 2001 2002 2003 2004 2005 2006
Year
Year

0.500 4.5

0.450 4.0
0.400
3.5
0.350
3.0
Rut Depth, inch

0.300
2.5
PSI

0.250
2.0
0.200

0.150 1.5

0.100 1.0

0.050 0.5
0.000 0.0
2000 2001 2002 2003 2004 2005 2006 2000 2001 2002 2003 2004 2005 2006
Year Year

Figure 3.44: Performance data of contract 3186b on IR080 from mileposts 85.25 to 26.21
 236

Flexible treatments
(Total number of projects = 28)
41
Max
36
Ranking of various projects
31
Average
26

21
Min
16

11

SRC-1

SRC+1
CIR-A-1

CIR-A+1

CIR-B-1

CIR-B+1

CIR-C-1

CIR-C+1

RF-1

RF+1

MOL-A+1

MOL-B-1

MOL-B+1

MOL-C-1

MOL-C+1
MOL-A-1
3 projects 3 projects 3 projects 5 projects 5 projects 2 projects 3 projects 4 projects
Treatments

Figure 3.45: Comparison between the ranking of the various treatments at 1-year before
treatment construction and 1-year after treatment construction.
Flexible treatments
(Total number of projects = 26)
41
Max
36

Average
Ranking of various projects

31

26
Min
21

16

11

1
SRC-1

SRC+3
CIR-A-1

CIR-A+3

CIR-B-1

CIR-B+3

CIR-C-1

CIR-C+3

RF-1

RF+3

MOL-A+3

MOL-B-1

MOL-B+3

MOL-C-1

MOL-C+3
MOL-A-1

3 projects 3 projects 3 projects 5 projects 4 projects 2 projects 3 projects 3 projects

Treatments

Figure 3.46: Comparison between the ranking of the various treatments at 1-year before
treatment construction and 3-year after treatment construction.
 237

Flexible treatments
(Total number of projects = 20)
41

36 Max
Ranking of various projects
31 Average

26 Min
21

16

11

SRC-1

SRC+5
CIR-A-1

CIR-A+5

CIR-B-1

CIR-B+5

CIR-C-1

CIR-C+5

RF-1

RF+5

MOL-A+5

MOL-B-1

MOL-B+5

MOL-C-1

MOL-C+5
MOL-A-1
3 projects 3 projects 0 projects 3 projects 3 projects 2 projects 3 projects 3 projects
Treatments

Figure 3.47: Comparison between the ranking of the various treatments at 1-year before
treatment construction and 5-year after treatment construction.
 238

(a) Opening Mode (b) Sliding Mode (c) Tearing Mode

Figure 4.1 Fracture Modes

Figure 4.2 Schematic representation of reflective cracking device (RCD)

(a) (b)

Figure 4.3 (a) Bottom platens of the reflective cracking device

(b) Reflective cracking device with both (top and bottom) plates
 239

Figure 4.4 Representation of the overlay zone subjected to the reflective cracking
 240

Figure 4.5 Flowchart of the Asphalt Rubber Association overlay design method
 241

Cracked HMA layer

4 inch
EHMA = 360 ksi (2500 MPa), ν = 0.25

Base layer
EBase = 30 ksi (210 MPa), ν = 0.35
10 inch

Subgrade
ESubgrade = 12 ksi (83 MPa), ν = 0.40

Figure 4.6 Existing HMA pavement structure

Figure 4.7 Required overlay thickness according to the Virginia tech method
 242

Figure 4.8 T2C required overlay thickness according to the asphalt rubber method

Figure 4.9 CT required overlay thickness according to the asphalt rubber method
 243

Figure 4.10 NRM required overlay thickness according to the asphalt rubber method
 244

Figure 4.11 Overlay thickness according to the asphalt rubber method for selected
percent cracking

Figure 4.12 Required overlay thickness according to the new AASHTO method
 245

Figure 4.13 Required overlay thickness according to the new AASHTO method for 100%
reflected cracks

Figure 4.14 Required T2C overlay thickness using various analysis models
 246

Figure 4.15 Required CT overlay thickness using various analysis models

Figure 4.16 Required NRM overlay thickness using various analysis models
 247

Figure 5.1 Bending test under static load – Cracow University of Technology

Figure 5.2 Results of the bending test under static load for HMA beams with and without
geotextile at 20°C
 248

Figure 5.3 Shearing test – Cracow University of Technology

Figure 5.4 The laboratory wheel-tracking device

 249

Figure 5.5 Beam specimens for the laboratory wheel-tracking device

 250

Figure 5.6 Summary of test results under the laboratory wheel-tracking device

Figure 5.7 Fissurometer apparatus description

 251

Figure 5.7 a. Principal of testing method (Tschegg 1986)

b. Schematic load-deformation curve

Figure 5.8 Specimen shapes for the wedge splitting test method (Tschegg 1986)

FMAX (N) Gf (N/m)

FH (KN) 5000 1200
T = -0.5 °C
4000 1000

3000 800

2000 600

1000 200

δ (mm) -25 -20 -15 -10 -5 0 5 10 -25 -20 -15 -10 -5 0 5 10

Temperature (°C) Temperature (°C)

Figure 5.9 Typical results from the wedge splitting test method
 252

Figure 5.10 Load-displacement curves for gravel and crushed HMA mixtures

a. b.

Figure 5.11 a. Maximum splitting force Fmax versus test temperature

b. Specific fracture energy Gf versus test temperature
 253

Figure 12 Shrinkage-bending test device for reflective cracking resistance

--- AC 0/10 Mix Rich Sand with 80/100 + HMA Mix

Figure 5.13 Test results example for shrinkage-bending test device

Figure 5.14 Mechanism of cracking of overlay

 254

Figure 5.15 Test set-up for bottom-up cracking tests

Figure 5.16 Test set-up for top-down cracking tests

Figure 5.17 Testing equipment for ISAC system

 255

Figure 5.18 Strain in HMA overlay as function of test cycles – control section

Figure 5.19 Strain in HMA overlay as function of test cycles – ISAC section
 256

Load Actuator

Figure 5.20 Instrumentation of fatigue test

Beam center line

Beam center line

a. Bending mode b. Shearing mode

Figure 5.21 Cracking pattern at the end of test – beam without geogrid
 257

a. Bending mode b. Shearing mode

Figure 5.22 Cracking pattern at the end of test – beam with geogrid

Figure 5.23 Schematic representation of the test setup

 258

Figure 5.24 Schematic representation of WRC test

Figure 5.25 Wheel Reflective Cracking (WRC) equipment

 259

Figure 5.26 Wheel Reflective Cracking device test specimen in plates of adherence

Figure 5.27 Mechanism simulating the relative vertical movement in the WRC

Environmental chamber

HMA Sample
Measure of horizontal
force Displacement gauge

System supports

Vertical displacement gauges

Measure of vertical forces

System engine

Figure 5.28 General scheme of the MEFISTO device

 260

Figure 5.29 Schematic of MEFISTO device and testing sample (one column)

10 100
Force Amplitude
9 90
8 80
Amplitud of the vertical load (kN)

7 70

Crack expasion (%)

6 Crack Expansion 60
5 50
4 40
3 30
2 20
1 10
0 0
0

20000

40000

60000

80000

100000

120000

140000

160000

180000

Number of load cycles 200000

Figure 5.30 Typical test results from the MEFISTO device

Figure 5.31 Concept of TTI overlay tester (after Cleveland et al. (68))
 261

Figure 5.32 Upgraded TTI overlay tester

Figure 5.33 Typical TTI overlay tester result (each opening and closing is 10 sec.)

Figure 5.34 Upgraded overlay tester SGC sample

 262

Figure 5.35 Repeatability of overlay testing on TxDOT type D mixture.

Figure 5.36 Relationship between number of specimens and specified tolerance of

reflective cracking life for TxDOT type D mixture

Figure 5.37 Influence of temperature on reflective cracking life

 263

Figure 5.38 Influence of opening displacement on reflective cracking life

Figure 5.39 Influence of asphalt content on reflective cracking life

Figure 5.40 Influence of air voids on reflective cracking life

 264

Figure 5.41 UFRGS-DAER traffic simulator from Brazil

Figure 5.42 Test pavement showing cracked areas and instrumentation location

Figure 5.43 Test pavement sections

 265

Figure 5.44 Failed HMA layer cracking pattern

Figure 5.45 Transverse cracks painted in different colors according to their appearance

Figure 5.46 Cracking severity index evolution in both overlays

 266

APPENDICES
 267

APPENDIX A – Summary of Various Studies to Mitigate Reflective Cracking

Study Title Researchers/Agency Reference Year Evaluated Techniques Findings
The History, Scofield, L. 5 1989 Asphalt rubber (AR) AR has successfully been used as an encapsulating
Development, A./Arizona DOT membrane to control pavement distortion due to
and expansive soils and to reduce reflective cracking in
Performance overlays on both rigid and flexible pavements.
of Asphalt
Rubber at
ADOT
Flagstaff I-40 Way, G. B./ Arizona 6 1999 Asphalt rubber (AR) After 9 years of service, overlay was still nearly crack-
Asphalt- DOT free, with good ride, virtually no rutting or
Rubber maintenance, and good skid resistance.
Overlay-Nine
Years of
success
Paving Rahman, M., 9 1989 Paving fabrics Need for proper tack coat selection
Fabrics for Scofield, L., and Caution regarding the use of Glassgrid on rough
Reducing Wolf, T/ Arizona surfaces.
Reflective Transportation
Cracking Research Center
Potential Amini, F/ Jackson 10 2005 Paving fabrics Proper installation procedures are critical for optimum
Applications State University performance.
of Paving Field performance of overlays using fabric interlayers
Fabrics to has generally been successful.
Reduce Most effective in warm climates-southern states.
Reflective
Cracking
Inhibiting Woodside, A. R., 11/12 1996/ Paving fabrics Fabric structure and rate of spray of emulsion tack-coat
Reflective McIlhagger, R., 1997 Stress absorbing membrane interlayer had the most significant effect in the bitumen/fabric
Cracking and Woodward, W. D. (SAMI). bond strength.
Use of Fabrics H., and Clements, H. SAMIs can reduce the likelihood of damage and the
to Inhibit W./ Arizona need for large reconstruction work.
Reflective University Extend pavement fatigue life.
Cracking in
Porous
 268

Study Title Researchers/Agency Reference Year Evaluated Techniques Findings

Asphalt
Use of Paving Predoehl, N. H./ 13 1989 HMA overlays 4.8 inches of overlay is required to reduce reflective
Fabric Test Caltrans cracking for 10 years.
Installations in
California
Fiberglass Banasiak, D./ 14 1997 Glassgrids + 7 inch HMA overlay After 3 years, glassgrid reinforced the HMA overlay
Mesh Reduces Caltrans and retarded reflective cracking in an area of high
Reflective tensile stress.
Cracking on
California
Highway
Report on FHWA 15 1982 Paving fabrics in HMA overlays Fabric section (1.5 inch AC overlay + fabric)
Performance developed less than 2% reflection cracking in 1.25
of Fabrics in years and 30% within 3.25 years while control section
Asphalt developed 10% cracking in 1.25 years and 50% within
Overlays 3.25 years.
Reducing Shuler, S., and 16 2004 Geotextile 90 lb No treatments performed better than the control (HMA
Reflective Hermelink, Geotextile 120 lb over milled surface directly) in the passing lane.
Cracking in D./Colorado State Reinforcing fabric Geotextile, paving fabric and crack sealer polymer
Asphalt University Fiberglass tape modified ASTM D3405 performed better than control
Pavements Crack sealer type ASTM D 3405 with to reduce reflective cracking.
routing original cracks
Crack sealer type ASTM D 3405 without
routing original cracks
Crack sealer type ASTM D 3405 polymer
modified
Overlays for Gulden, W., and 17/18 1984/ HMA overlays over deteriorated PCC 20% of cracking area occurred in 6 years for a 6-inch
Plain Jointed Brown, D/ Georgia 1985 HMA overlay compared to 2 years for a 4-inch HMA
Concrete DOT overlay.
Pavements Reflective cracking appeared almost immediately after
and construction for a 2-inch overlay.
Treatments for
Reduction of
Reflective
Cracking of
Asphalt
 269

Study Title Researchers/Agency Reference Year Evaluated Techniques Findings

Overlays on
Jointed-
Concrete
Pavements in
Georgia
Georgia’s Brown, D. R., Jared, 19 1997 Crumb rubber mix (CRM) After 4 years the section showed a large amount of
Experience D., Jones, C., and transverse reflective cracking.
with Crumb Watson, D./ Georgia Compared with the control mix, the CRM did not
Rubber in DOT reduce rutting and was more than twice as expensive to
Hot-Mix place.
Asphalt
An Evaluation Mascunana, I/ 20 1981 Paving fabrics Treatments not effective in preventing the development
of Illinois DOT Fabricated interlayer membrane of transverse reflective cracking on overlays with
Engineering Asphalt rubber membrane interlayer cement treated bases but longitudinal reflective
Fabric in cracking.
Pavement Treatments effective in reducing transverse and
Rehabilitation longitudinal reflective cracking on overlays with
(IHD-21) bituminous base courses.
Evaluation of Buttlar, W. G., 21 1999 Non-woven paving fabrics Performance monitoring indicated an increase in life
Reflective Bozkurt, D. and spans by 1.1 and 3.6 years for paving fabric strip.
Crack Control Dempsey, B. J/
Policy Illinois
Transportation
Research Center
An Evaluation Vespa, J./ Illinois 22 2000 Interlayer Stress Absorbing Composite The formation of reflective cracks and the subsequent
of Interlayer DOT (ISAC) deterioration of these cracks were delayed at ISAC
Stress treated joints and cracks.
Absorbing
Composite
(ISAC)
Reflective
Crack Relief
System
Application of Jiang, Y., and 23 1993 Cracking and seating before overlay Cracking and seating was successful to delay most of
Cracking and McDaniel, R. Fiber reinforcement of the overlay mixture the reflected cracks for 5 years.
Seating and S./Indiana DOT The use of fibers in the overlay mixture further reduced
 270

Study Title Researchers/Agency Reference Year Evaluated Techniques Findings

Use of Fibers transverse cracks on cracked and seated sections.
to Control
Reflective
Cracking
Stone Rasoulian, M., 24 2000 Stone interlayer or inverted pavement Stone interlayer had significantly reduced the amount
Interlayer Becnel, B., and Keel, of reflective cracking.
Pavement G./ Louisiana DOT The life of the stone interlayer pavement system is
Design increased to almost 5 times that of the standard soil
cement pavement.
Glasgrid Tensar International 25 1994 Glassgrid GlasGrid System showed benefits in retarding
Pavement Corporation reflective cracking.
Reinforcement
System.
“System
Overview
Field V.T. 27 1989 Paving fabric/rubberized asphalt Pavements with the Petrotac interlayer show
Evaluation of Barnhart/Michigan composite membrane considerably less recurrence of reflective cracking.
Experimental Transportation
Fabrics to Commission
Prevent
Reflective
Cracking in
Bituminous
Resurfacing
Paving Fabric Kidd, S. Q./ 28 1990 Paving fabrics Asphalt rubber interlayer, in combination with a thin
and Asphalt Mississippi DOT Asphalt rubber interlayer system overlay (about 1 ½ inch), reduced and/or delayed
Stress Single bituminous surface treatment reflective cracking over a lengthy period of time (about
Absorbing 5 years).
Membrane
Interlayers
(SAMI)
Field Maurer, D. A. and 29 1989 Paving fabrics All treatments retarded cracks over the evaluation
Performance Malashekie G. J./ Fiber Pave reinforced asphalt membrane period, although the amount and rate of reduction
of Fabrics and Pennsylvania DOT interlayer varied.
Fibers to Bonifiber reinforced asphalt concrete
Retard
 271

Study Title Researchers/Agency Reference Year Evaluated Techniques Findings

Reflective
Cracking
Geogrid Mesh Hughes, J. J., and 30 2000 Fabric/geogrid None of the 3 paving fabric/geogrid types were found
for Reflective Somers, E./ to be effective in preventing or retarding reflective
Crack Control Pennsylvania DOT cracking.
in Bituminous
Overlays
Experience Morian, D. A., 31 2004 Cold in-place recycling (CIR) CIR provided resistance against reflective cracking
with Cold In- Oswalt, J., and between 2 and 3 times that exhibited by conventionally
Place Deodhar A./ resurfaced control sections.
Recycling as a Pennsylvania DOT The cost of CIR is one to two-thirds the cost of
Reflective conventional HMA material while providing superior
Crack Control performance.
Technique:
Twenty Years
Later
Geosynthetics Cleveland G. S., 32 2002 Geosynthetics Performance of geosynthetics in addressing reflection
in Flexible Button J. W., and cracking in HMA overlays has ranged from highly
and Rigid Lytton R./ Texas successful to disastrous failures.
Pavement Transportation
Overlay Institute
Systems to
Reduce
Reflection
Cracking
Fiberglass Darling, J.R., and 33 2004 Fiberglass Fiberglass reinforcements are extremely strong and
Pavement Woolstencroft, experience low elongations at ultimate strength and are
Reinforcement J.H./Saint-Gobain able to reduce the rate of crack reflections significantly
Used in Technical Fabrics, to that of non-reinforced overlay.
Dissimilar Canada
Climatic
Zones for
Retarding
Reflective
Cracking in
Asphalt
 272

Study Title Researchers/Agency Reference Year Evaluated Techniques Findings

Overlays
Lessons Chen, D. H., 34 2006 Crack-retarding grid Crack-retarding grid (plastic geosynthetics) did not
Learned on Scullion, T., and Crack-retarding asphalt material perform well in retarding reflective cracks.
Jointed Bilyeu, J./Texas Flexible base overlay with thin asphalt Crack-retarding asphalt material (Strata) performed
Concrete DOT surfacing well over 2 years of monitoring.
Pavement Arkansas open graded large stone AC Flexible base overlay with thin asphalt surfacing
Rehabilitation interlayer performed well.
Strategies in Arkansas open graded large stone AC interlayer mix
Texas performed well.
Evaluation of Freeman, T. E/ 35 2002 Concrete slab fracturing Fracturing and seating distressed concrete pavements
Concrete Slab Virginia effective in retarding the formation of reflective
Fracturing Transportation cracking through asphalt overlays on jointed plain
Techniques in Research Council concrete pavements.
Mitigating Fracturing technique was somewhat less successful in
Reflective reinforced concrete pavements. Formation of reflective
Cracking cracks appeared to be delayed for only about 3 years.
Through
Asphalt
Overlays
GlassGrid Bischoff, D., and 36 2003 GlasGrid Both products performed unsatisfactorily and were
Pavement Toepel, A/ Glass fiber mesh unable to prohibit or control reflective cracking
Reinforcement Wisconsin DOT effectively.
Product
Evaluation
Evaluation of Abu Al-eis, K/ 37 2004 Interlayer Stress-Absorbing Composite Due to the transverse bumps in the overlay, the overlay
the Interlayer Wisconsin DOT (ISAC) and the ISAC fabric were removed and the test section
Stress- was repaved without the ISAC
Absorbing
Composite
(ISAC) to
Mitigate
Reflective
Cracking in
Asphaltic
Concrete
Ovelarys
 273

Study Title Researchers/Agency Reference Year Evaluated Techniques Findings

Wisconsin Makowski, L., 38 2005 Modified asphalt mix interlayer In one project, interlayer showed no impact on
Experiences Bischoff, D., delaying reflection cracking within the first 3 years.
with Blankenship, P., Other projects showed an average 42% improvement in
Reflective Sobczak, D., and the time to the appearance of surface cracks compared
Crack Relief Haulter F/Wisconsin with the control sections.
Projects DOT
Long Term Vanelstraete A., and 39 2004 2 SAMI 6 Crack and seat and steel reinforcing nettings are
Performance De Visscher 3 Non-woven fabric both effective against reflective cracking.
on Site of J./Belgium Road 4 Geogrid 7 Less reflective cracks were developed on sections
Interface Research Center, 5 Steel nettings with interface systems (SAMI, non-woven, grid,
Systems Belgium steel reinforcing nettings).
8 Projects with steel reinforcing nettings performed
very well even after more than 10 years of repair.
Life Cycle Tighe, S., Haas, R., 40 2003 9 Routing and sealing 10 Proper and timely crack treatment (routing and
Cost Analysis and Ponniah, J../ sealing) can result in extending life by 2 years and
of Mitigating University of cost savings in the order of $7,000/lane-km.
Pavement Waterloo, Canada
Rehabilitation
Reflection
Cracking
Field Study of Valtonen, J., and 41 2004 Milling the form of a box, 1.5 m wide After one year, traditional methods (crack fillling and
Repair Hyyppa, I./Helsinki grid, leveling the old surface, hot milling HMA overlay or mill and HMA overlay) were not
Methods for University of and asphalt surfacing. successful.
Transversal Technology, Finland Milling the form of a box, 0.75 m wide The best promising method to prevent the reflection of
Cracks grid, leveling the old surface, hot milling transverse cracks is to lay a grid on a "milled box.
and asphalt surfacing.
Sawing a crack, filling the crack with hot
modified bitumen, leveling the old
surface, hot milling and asphalt surfacing.
Sawing a crack, filling the crack with a
band of modified bitumen, leveling the old
surface, hot milling and asphalt surfacing.
Hot milling, apply 1.5 m wide grid, and
asphalt surfacing.
Hot milling, apply 0.75 m wide grid, and
asphalt surfacing.
 274

Study Title Researchers/Agency Reference Year Evaluated Techniques Findings

Hot milling and asphalt surfacing.
 275

APPENDIX B – DOTs Standard Specifications for Highway Construction

Agency Std Specs for Highway Reflective Cracking Description
Construction Edition Control System
Alabama 2006 Geotextiles Stress relieving membrane within pavement structure
Used as strips over transverse or longitudinal joints
Cracks exceeding 1/8” in width shall be filled with a rubberized joint sealer meeting
the requirements of ASTM D 6690, for Type II sealant. Potholes shall be properly
repaired as directed by Engineer.
Alaska 2004 Paving fabrics Fill all potholes and cracks wider than 1/4 inch with an approved asphalt emulsion
and sand slurry.
Overlap transverse joints in the direction of paving. Apply 0.20 gallons/yd2 of
additional asphalt sealant beneath all fabric joints.
Arizona 2000 Asphalt rubber stress Cover material shall be precoated with 0.40 to 0.60% AC by weight of aggregate.
absorbing membrane The AC shall meet the requirements of Section 1005. The end result shall be a dust-
Crack sealing free material.
Joint sealing Crack sealing: a. mixture of asphalt and 100% vulcanized granulated rubber. b.
Premixed block material consisting of asphalt and 100% vulcanized granulated
rubber.
Joint sealant: a. mixture of asphalt and 100% vulcanized granulated rubber. b
.mixture of asphalt, extender oil, and reclaimed high natural and ground vulcanized
rubber. c. Premixed block material consisting of asphalt and 100% vulcanized
rubber. d. Premixed block material consisting of asphalt and extender oil, reclaimed
high natural and ground vulcanized rubber.
Arkansas 2003 NA NA
California 2006 Reinforcing fabric Shall be 100% polypropylene staple fiber fabric material, needle-punched, thermally
bonded on one side
Colorado 2006 Joint and crack sealant Joint and crack sealant: fill cracks with a width greater than 1/8” and less than 1”.
Geosynthetics Cracks shall be cleaned of loose and foreign matter to a depth approximately twice
the crack width before applying hot poured sealant.
Pavement surface shall be broomed clean immediately prior to beginning the crack
reduction geotextile treatment using a self-propelled power broom. AC binder shall
be applied to pavement surface at the rate of approximately 0.25 gallon/yd2.
Geosynthetic rolls shall be furnished with suitable wrapping to protect against
moisture and extended ultraviolet exposure prior to placement.
Connecticut 2005 NA NA
 276

Agency Std Specs for Highway Reflective Cracking Description

Construction Edition Control System
Delaware 2005 NA NA
District of 2005 Replace of bituminous Removing and replacing the binder and surface courses in defective areas by: a.
Columbia materials and base Cutting defective areas, b. Repair of cuts, c. Cutting to a neat line: The perimeter of
all cuts and/or defective area repairs in asphalt roadways, alleys, sidewalks, gutters
and other miscellaneous pavements which become part of the permanent roadway
surface shall be cut to a neat line by means of power saw, and d. Compaction of
bituminous materials to the specified compaction.
Florida 2005 Asphalt rubber membrane Constructing an asphalt rubber membrane interlayer composed of a separate
interlayer application of asphalt rubber binder covered with a single application of aggregate.
Use Asphalt Rubber Binder ARB-20. Cover Material: Use Size No. 6 stone, slag, or
gravel. Combine the materials as rapidly as possible for such a time and at such a
temperature that the consistency of the binder approaches that of a semi-fluid
material.
Georgia 2006 Asphalt rubber joint and Filling (Type M) or sealing (Type S) joints and cracks in existing pavements with
crack seal rubber asphalt mixtures. A polymer-modified asphalt rubber (PMAR) blend may be
Hot Asphalt-Rubber Seal used in lieu of both Type M and Type S. The mixture pours readily and penetrates a
Treatment for Stress- 1/4” pavement joint or crack to a depth of at least 1”.
Relieving Interlayer For Hot Asphalt-Rubber Seal Treatment for Stress-Relieving Interlayer follow
Geogrid reinforcement specifications as included in the contract.
Pavement reinforcement For geogrid reinforcement follow specifications as included in the contract.
fabric Pavement reinforcement fabric properties: a. non-woven, heat-resistant material
composed of polypropylene or polyester fibers, b. Can be saturated with asphalt
cement, c. Can be placed smooth with mechanical devices and be without wrinkles,
d. Can withstand the heat of asphaltic concrete mixes during paving operations,
d. Can withstand normal field handling and construction operations without damage.
Hawaii 2006 Paving fabric Surface Preparation: clean surface before applying AC, and cracks and joints with
compressed air. Seal cracks and joints wider than 3/8” with sand slurry consisting of
20% CSS-1 emulsified asphalt, approximately 2% portland cement, and water.
Clean potholes and other surface defects and fill with HMA pavement. Apply AC to
cover area of paving fabric plus 3” on each side.
Place fabric onto AC before it has cooled and lost its tackiness, with heat-bonded
side up and minimum wrinkling/folding. Overlap fabric 6” at joints.
Idaho 2004 Geotextile Shall be composed only of long chain polymeric filaments or yarns oriented into a
stable network, which retains its relative structure, including selvedges, during
handling, placement, and design service life.
 277

Agency Std Specs for Highway Reflective Cracking Description

Construction Edition Control System
Only nonwoven geotextiles are acceptable
Illinois 2002 Reflective cracking control Surface on which reflective crack control system is to be constructed shall be clean
treatment and dry. All base failures shall be repaired and all cracks, spalls, potholes or other
depressions shall be sealed with an approved crack sealer or filled with mixture for
cracks, joints and flange ways.
Area reflective crack control treatment System A: area to be covered with fabric
shall be sprayed uniformly with asphalt binder at a rate of 0.25 to 0.30 gal/yd2 as
directed by Engineer. Binder application shall be accomplished with a pressure
distributor for all surfaces except, where the distributor does not have room to
operate, hand spraying will be allowed.
Area reflective crack control treatment System B: primer to be used with the
waterproofing membrane shall be supplied by the manufacturer of the membrane
and shall be compatible with the membrane.
Area reflective crack control treatment System C: immediately prior to application
of a tack coat, the surface shall be thoroughly cleaned by sweeping. When placed as
a strip treatment, the strip shall be 24” wide. Also when placed as a strip treatment, a
self-propelled distributor will not be required for applying the tack coat nor the
asphalt-rubber, nor will a self-propelled spreader be required to place the cover
aggregate.
Indiana 2006 Seal cracking and joints Sealing longitudinal and transverse cracks and joints in existing asphalt pavement.
The steps to do that are: a. Routing and Filling Cracks and Joints, b. Sealing Cracks
and Joints Cracks and joints shall be cleaned by blowing with compressed air or by
other suitable means. Asphalt material shall be placed utilizing a “V” shaped wand
tip. The cracks and joints shall be completely filled or over banded not to exceed 5”.
Iowa 2006 Asphaltic overlay fabric Fabric placed under asphalt mixtures to provide waterproofing and delay reflective
cracking. Shall be capable of withstanding installation stresses and shall not be
damaged by temperatures common to asphalt mixtures
Asphalt absorption shall be sufficient to produce good bond between overlay and
overlaid surface when a tack coat of 0.20 to 0.25 gallon/yd2 is used.
Kansas 1990 NA NA
Kentucky 2004 NA NA
Louisiana 2004 Asphaltic surface treatment Asphaltic surface treatment consist of a specified emulsion applied "cold" or
Paving fabric modified asphalt material applied "hot", at the temperature range specified.
Paving fabric shall conform to requirements in AASHTO M 288.
Maine 2002 Geotextiles Shall have property values expressed as Minimum Average Roll Value (MARV) in
 278

Agency Std Specs for Highway Reflective Cracking Description

Construction Edition Control System
the weakest principal direction, which meet or exceed given values. Sampling and
conformance testing shall be in accordance with ASTM D4354. Geotextile product
acceptance shall be based on ASTM D4759. Geotextile storage and handling
requirements shall be based on ASTM D4873.
Maryland 2001 Filling cracks in HMA Cleaning and filling cracks 1/8 to 1-3/4” wide in HMA pavement as specified in the
pavements Contract Documents or as directed by the Engineer. Distressed areas shall be
repaired. Cracks more than 1-3/4” wide; and map, edge or alligator cracks requiring
major repairs are not included in Specification
Massachusetts 1995 NA NA
Michigan 2003 NA NA
Minnesota 2005 NA NA
Mississippi 2004 Cleaning and sealing of Joints and cracks to be sealed shall be cleaned by routing, sawing and/or sand
cracks blasting to the minimum dimensions specified. Other cleaning methods shall be
approved by Engineer. Surface which is to receive the new joint sealing material
shall be dry and free of all lubricants, tar, asphalt, discoloration and stain as well as
all other forms of contamination.
Missouri 2004 Bituminous cracking crack This work consists of preparing and sealing all working transverse and longitudinal
sealing cracks in bituminous pavement as shown on the plans or as directed by engineer.
The sealant shall be a single-component material in accordance with AASHTO M
301, except as herein modified.
Montana 2006 NA NA
Nebraska 2007 Fabric reinforcement Fabric reinforcement shall be applied immediately before the placement of the
Crack sealing bituminous bituminous overlay. The materials and application method shall resist shoving and
surface lifting during placement of the bituminous overlay.
Sealant shall be a mixture of paving grade asphalt, vulcanized recycled rubber, and
polymer modifier
Nevada 2007 Engineering fabrics Pavement Reinforcing Fabric manufactured from polyester, polypropylene, or
polypropylene-nylon material. Fabric shall be nonwoven.
New 2006 Hot poured crack sealing Crack sealing material to be covered by a 1” or less overlay shall cure a minimum of
Hampshire 45 days prior to the placement of bituminous pavement.
Material covered by an asphalt pavement overlay shall be low modulus conforming
to ASTM D 3405, modified.
All cracks greater than 1/8” up to 3/4” in width shall be shaped with a power router
to a dimension of 3/4” + 1/8” wide by 5/8” deep rectangular shape and treated
unless otherwise directed. Cracks greater than 3/4” shall be treated but not routed.
 279

Agency Std Specs for Highway Reflective Cracking Description

Construction Edition Control System
New Jersey 2001 Sawing and sealing of Sawing and Sealing Joints in HMA Overlays. Existing transverse joints that are
joints offset at the longitudinal joint by more than 1”, measured between the centers of the
joints, require separate saw cuts terminating at the longitudinal joint. Overlays shall
be saw cut over transverse cracks that are reasonably straight, at least 1/8” wide, and
extend one full lane width.
For overlays whose total thickness is 2 inches or less, the saw cut shall be 3/8 to
1/2" wide by 5/8” deep. For overlays whose total thickness is greater than 2”, the
saw cut shall be a T-shaped cut consisting of the saw cut specified in Item a above
plus a 1/8” wide saw cut at the center.
New Mexico 2002 NA NA
New York 2006 NA NA
N. Carolina 2006 NA NA
N. Dakota 2006 Geotextile fabric Shall be a fabric consisting of polymeric filament or yarns such as polypropylene,
polyethylene, polyester, polyamide, or polyvinylidene chloride. The filaments or
yarns shall be formed into a stable network so they retain their relative position to
each other.
Ohio 2005 Sawing and sealing of Saw cutting and sealing finished surface of the asphalt concrete pavement and
asphalt concrete pavement shoulders directly over and in line with transverse joints in the underlying portland
joints cement concrete pavement.
Oklahoma 1999 Fabric reinforcement for Surface should be cleaned and free of any material.
asphalt concrete pavement Binder should be placed at no less than 290°F and a rate of 0.5 gal/yd2.
Fabric should be placed after application of the bituminous binder.
Oregon 1999 Crack sealing flexible All sealant materials for crack repair of flexible pavements shall be approved by the
pavements Engineer before being incorporated into the work.
Geotextile Clean all cracks designated for sealing of loose and foreign matter.
Seal the cracks from the bottom up in a neat manner, so that upon completion of the
work, the surface of the sealant material is flush to 3/16” below adjacent pavement
surface.
Geotextile can be use under an overlay and should follow the requirements of table
02320-1.
Pennsylvania 1994 Fiberized asphalt Fiberized asphalt membrane: cleaning and sealing of longitudinal and transverse
membrane joints and cracks in existing pavement surfaces with a fiberized asphalt membrane
Polymer modified asphalt prior to overlaying. Provide membrane having a width of 5” + 1” on concrete and
joint and cracking bituminous pavement surfaces.
Polymer modified asphalt joint and cracking: cleaning and sealing of longitudinal
 280

Agency Std Specs for Highway Reflective Cracking Description

Construction Edition Control System
and transverse joints and cracks in existing pavement surfaces with polymer
modified asphalt prior to overlaying.
Rhode Island 1999 Cleaning, routing, and Cracks Less than 1” in width. Poured joint seal material shall be an asphalt rubber
sealing cracks in compound, hot poured type, conforming to the requirements of AASHTO M173 and
bituminous concrete approved by the Engineer. Cracks 1” and over in width, bituminous concrete, Class
pavement I-2.
South 2000 NA NA
Carolina
South Dakota 2004 Asphalt concrete crack Routing equipment shall be mechanical, power driven, and capable of cutting a
sealing reservoir to the required dimensions. Equipment designed to plow the cracks to
dimension will not be permitted.
Cracks which are less than 3/4” in width or depth will require routing to a width and
depth of 3/4 to 7/8”.
Cracks which are 3/4” or greater in width and depth will not require routing, but
shall be thoroughly cleaned of foreign material to a depth equal to the width of the
crack.
Tennessee 2006 NA NA
Texas 2004 Fabric underseal Furnish and place fabric underseal in a longitudinal, full-road-width application or
over pavement joints.
Prepare the surface by cleaning off dirt, dust, or other debris.
Apply asphalt binder immediately after asphalt binder application, align the fabric
and broom or roll it in place. Overlap transverse joints by minimum of 6”.
Overlap longitudinal joints by a minimum of 4”.
Utah 2002 NA NA
Vermont 2001 Bituminous crack filling This work shall consist of furnishing and placing sealing compound in cracks of
existing bituminous concrete pavement.
Cracks 1/4” and wider shall be shaped with a power router and flame cleaned of all
dirt, foreign material, and loose edges to a minimum depth of 3/4”. All cracks may
be cleaned with a hot compressed air lance instead of routing, if approved by the
Engineer.
Virginia 2006 NA NA
Washington 2002 NA NA
West Virginia 2000 NA NA
Wisconsin 2006 NA NA
 281

Agency Std Specs for Highway Reflective Cracking Description

Construction Edition Control System
Wyoming 2003 Crack sealing Crack sealing consist of: a. Removing of existing material, b. routing, c.
Paving fabric cleaning/drying, d. Sealing.
Paving fabric applies to fabric membranes used for full coverage of the pavement or
as strips over transverse and longitudinal pavement joints
 282

APPENDIX C – PC1 Values Along with Measured Surface Cracks of NDOT

Projects

Table C1 PC1 Values 1-year Before Rehabilitation and 1-year After Treatment
Construction for Flexible Pavements.

Fatigue cracking Transverse Block cracking

Project
Contract Type A Type B Cracking Type A Type B Type C PC1 at -1 PC1 at +1
ID
(ft) (ft2) (ft) (ft) (ft2) (ft2)
2808a CIR-A-1 300 0 20 100 0 0 0.841196 -0.7828
2808b CIR-A-2 50 300 35 110 0 0 0.313319 -0.7828
2838 CIR-A-3 0 0 0 400 750 700 1.728775 -0.67827
2935 CIR-A-4 150 300 35 100 300 300 1.464507 -0.7828
2819 CIR-B-1 0 530 75 0 150 0 0.79174 -0.7828
2873 CIR-B-2 50 0 40 190 675 190 1.291939 -0.7828
2961 CIR-B-3 40 0 40 100 1000 1000 2.485455 -0.7828
3013 CIR-B-4 75 0 30 225 0 0 0.227509 -0.7828
3025a CIR-C-1 10 50 25 0 0 0 -0.4243 -0.62601
3025b CIR-C-2 35 150 35 125 300 110 0.639205 -0.52148
3025c CIR-C-3 50 150 10 300 400 50 0.802542 -0.52148
2876 CIR-D-1 50 0 75 400 0 50 0.953418 -0.52148
2761 RF-1 25 100 70 1000 500 0 2.545001 -0.46921
2932 RF-2 25 100 100 0 0 0 0.477578 -0.7828
2980a RF-3 25 100 30 0 150 10 -0.0425 -0.7828
2980b RF-4 25 100 20 0 0 0 -0.35866 -0.7828
3006 RF-5 25 100 21 0 0 0 -0.34821 -0.7828
3008 RF-6 25 100 10 0 0 0 -0.46319 -0.7828
2723 SCR-1 120 445 60 210 210 0 1.477647 -0.7828
3031 SCR-2 75 75 50 500 0 0 0.991769 -0.7828
3048 SCR-3 50 0 55 0 0 0 -0.00065 -0.7828
3045 SCR-4 50 300 25 0 0 0 0.020147 -0.7828
3162 SCR-5 0 0 20 0 0 0 -0.57374 -0.7828
2384a MOL-A-1 0 600 20 200 0 0 0.43801 -0.15926
2384b MOL-A-2 100 325 42 60 0 0 0.535849 -0.7828
2432 MOL-B-1 110 0 75 0 0 0 0.457102 -0.7828
2505 MOL-B-2 0 600 0 400 0 1000 1.752618 -0.7828
2651a MOL-C-1 0 300 0 300 0 300 0.420265 -0.7828
2651b MOL-C-2 0 0 0 0 1000 1000 1.730051 -0.7828
2651c MOL-C-3 100 0 95 200 1000 1000 3.48055 -0.7828
2679 MOL-C-4 0 0 15 0 0 0 -0.62601 -0.52148
3028 MOL-B-3 0 0 48 0 0 0 -0.28106 -0.7828
3070 MOL-C-5 35 50 35 325 375 0 0.840768 -0.7828
 283

Table C2 PC1 Values 1-year Before Rehabilitation and 3-year After Treatment
Construction for Flexible Pavements.

Fatigue cracking Transverse Block cracking

Project
Contract Type A Type B Cracking Type A Type B Type C PC1 at -1 PC1 at +3
ID
(ft) (ft2) (ft) (ft) (ft2) (ft2)
2808a CIR-A-1 300 0 20 100 0 0 0.58561 -0.81830
2808b CIR-A-2 50 300 35 110 0 0 0.19113 -0.24700
2838 CIR-A-3 0 0 0 400 750 700 1.77037 -0.72010
2935 CIR-A-4 150 300 35 100 300 300 1.30582 -0.81830
2819 CIR-B-1 0 530 75 0 150 0 0.66354 -0.26733
2873 CIR-B-2 50 0 40 190 675 190 1.23997 -0.81830
2961 CIR-B-3 40 0 40 100 1000 1000 2.50515 -0.81830
3013 CIR-B-4 75 0 30 225 0 0 0.12057 -0.81830
3025a CIR-C-1 -- -- -- -- -- -- -- --
3025b CIR-C-2 35 150 35 125 300 110 0.56324 -0.81830
3025c CIR-C-3 50 150 10 300 400 50 0.73270 -0.52369
2876 CIR-D-1 50 0 75 400 0 50 0.83777 -0.52369
2761 RF-1 25 100 70 1000 500 0 2.45912 -0.52369
2932 RF-2 25 100 100 0 0 0 0.35166 -0.81830
2980a RF-3 25 100 30 0 150 10 -0.11632 -0.81830
2980b RF-4 25 100 20 0 0 0 -0.43397 -0.72010
3006 RF-5 25 100 21 0 0 0 -0.42415 -0.81830
3008 RF-6 25 100 10 0 0 0 -0.53218 -0.64154
2723 SCR-1 120 445 60 210 210 0 1.28675 -0.81830
3031 SCR-2 75 75 50 500 0 0 0.86394 -0.81830
3048 SCR-3 50 0 55 0 0 0 -0.10546 -0.81830
3045 SCR-4 50 300 25 0 0 0 -0.09539 -0.81830
3162 SCR-5 -- -- -- -- -- -- -- --
2384a MOL-A-1 0 600 20 200 0 0 0.32986 -0.27646
2384b MOL-A-2 100 325 42 60 0 0 0.37238 -0.62189
2432 MOL-B-1 110 0 75 0 0 0 0.29821 -0.81830
2505 MOL-B-2 0 600 0 400 0 1000 1.71684 -0.81830
2651a MOL-C-1 0 300 0 300 0 300 0.37226 -0.81830
2651b MOL-C-2 0 0 0 0 1000 1000 1.80296 -0.81830
2651c MOL-C-3 100 0 95 200 1000 1000 3.42373 -0.81830
2679 MOL-C-4 0 0 15 0 0 0 -0.67100 -0.60225
3028 MOL-B-3 0 0 48 0 0 0 -0.34692 -0.67100
3070 MOL-C-5 -- -- -- -- -- -- -- --
 284

Table C3 PC1 Values 1-year Before Rehabilitation and 5-year After Treatment
Construction for Flexible Pavements.

Fatigue cracking Transverse Block cracking

Project
Contract Type A Type B Cracking Type A Type B Type C PC1 at -1 PC1 at +5
ID
(ft) (ft2) (ft) (ft) (ft2) (ft2)
2808a CIR-A-1 300 0 20 100 0 0 -0.06549 -0.80584
2808b CIR-A-2 50 300 35 110 0 0 -0.02962 -0.47097
2838 CIR-A-3 0 0 0 400 750 700 1.87241 -0.65080
2935 CIR-A-4 150 300 35 100 300 300 0.93381 -0.80584
2819 CIR-B-1 0 530 75 0 150 0 0.44757 -0.44988
2873 CIR-B-2 50 0 40 190 675 190 1.12401 -0.80584
2961 CIR-B-3 40 0 40 100 1000 1000 2.51798 -0.80584
3013 CIR-B-4 75 0 30 225 0 0 -0.10277 -0.80584
3025a CIR-C-1 -- -- -- -- -- -- -- --
3025b CIR-C-2 -- -- -- -- -- -- -- --
3025c CIR-C-3 -- -- -- -- -- -- -- --
2876 CIR-D-1 -- -- -- -- -- -- -- --
2761 RF-1 25 100 70 1000 500 0 2.21048 -0.54500
2932 RF-2 25 100 100 0 0 0 0.08997 -0.80584
2980a RF-3 25 100 30 0 150 10 -0.21971 -0.80584
2980b RF-4 25 100 20 0 0 0 -0.53018 -0.72832
3006 RF-5 -- -- -- -- -- -- -- --
3008 RF-6 -- -- -- -- -- -- -- --
2723 SCR-1 120 445 60 210 210 0 0.85628 -0.80584
3031 SCR-2 75 75 50 500 0 0 0.56204 -0.51127
3048 SCR-3 50 0 55 0 0 0 -0.30895 -0.76708
3045 SCR-4 -- -- -- -- -- -- -- --
3162 SCR-5 -- -- -- -- -- -- -- --
2384a MOL-A-1 0 600 20 200 0 0 0.18552 -0.47097
2384b MOL-A-2 100 325 42 60 0 0 0.03547 -0.52035
2432 MOL-B-1 110 0 75 0 0 0 -0.06927 -0.47097
2505 MOL-B-2 0 600 0 400 0 1000 1.68564 -0.80584
2651a MOL-C-1 0 300 0 300 0 300 0.33580 -0.48165
2651b MOL-C-2 0 0 0 0 1000 1000 1.98938 -0.57328
2651c MOL-C-3 100 0 95 200 1000 1000 3.19106 -0.43221
2679 MOL-C-4 0 0 15 0 0 0 -0.68956 -0.57328
3028 MOL-B-3 0 0 48 0 0 0 -0.43375 -0.68956
3070 MOL-C-5 -- -- -- -- -- -- -- --
 285

Table C4 PC1 Values 1-year Before Rehabilitation and 5-year After Treatment
Construction for Rigid Pavements.

Fatigue cracking Transverse Block cracking

Project
Contract Type A Type B Cracking Type A Type B Type C PC1 at -1 PC1 at +5
ID
(ft) (ft2) (ft) (ft) (ft2) (ft2)
2886 CS-1 0 0 0 0 0 0 -0.75771 0.17659
2889 CS-2 0 0 0 0 0 0 -0.75771 -0.20605
2962 CS-3 0 0 10 0 0 0 0.17659 -0.75771
2999 CS-4 0 0 0 0 0 0 -0.75771 0.62144
3021 CS-5 0 0 0 0 0 0 -0.75771 0.62144
2549 RPCC-1 0 0 33 0 0 0 2.32550 0.34561
2869 RPCC-2 0 0 0 0 0 0 -0.75771 2.00059
2901 RPCC-3 0 0 0 0 0 0 -0.75771 -0.75771
3088 RPCC-4 -- -- -- -- -- -- -- --
3186a RPCC-5 -- -- -- -- -- -- -- --
3186b RPCC-6 -- -- -- -- -- -- -- --