Technical Report Documentation Page

1. Report No. 2. Government Accession No. 3. Recipient’s Catalog No.

FHWA-TX-1250-4
4. Title and Subtitle 5. Report Date
January 2000
DEVELOPMENT AND USE OF STATIC CREEP TEST TO
6. Performing Organization Code
EVALUATE RUT RESISTANCE OF SUPERPAVE MIXES
7. Author(s) 8. Performing Organization Report No.
Weng On Tam, Mansour Solaimanian, and Thomas W. Kennedy 1250-4

9. Performing Organization Name and Address 10. Work Unit No. (TRAIS)
Center for Transportation Research
The University of Texas at Austin
3208 Red River, Suite 200 11. Contract or Grant No.
Austin, TX 78705-2650 0-1250
12. Sponsoring Agency Name and Address 13. Type of Report and Period Covered
Texas Department of Transportation Research Report
Research and Technology Implementation Office (9/98 — 1/99)
P.O. Box 5080
Austin, TX 78763-5080 14. Sponsoring Agency Code

15. Supplementary Notes

Research performed in cooperation with the U.S. Department of Transportation, Federal Highway Administration,
and the Texas Department of Transportation
16. Abstract

This is the fourth report for research project 0-1250, The South Central Superpave Center. This study was initiated
and sponsored by the Texas Department of Transportation (TxDOT) to evaluate the uniaxial static creep test for
Superpave mixes. The research period for the project was scheduled for 1 year, from September 1998 through
August 1999. This report presents the results, findings, conclusions, and recommendations based on the laboratory
study and testing of asphalt mixtures.

17. Key Words 18. Distribution Statement

Superpave, asphalt pavements, gyratory compactors No restrictions. This document is available to the public
through the National Technical Information Service,
Springfield, Virginia 22161.
19. Security Classif. (of report) 20. Security Classif. (of this page) 21. No. of pages 22. Price
Unclassified Unclassified 82
Form DOT F 1700.7 (8-72) Reproduction of completed page authorized

DEVELOPMENT AND USE OF STATIC CREEP TEST TO
EVALUATE RUT RESISTANCE OF SUPERPAVE MIXES

by
Weng On Tam, Mansour Solaimanian, and Thomas W. Kennedy

Research Report Number 1250-4

Research Project 0-1250

The South Central Superpave Center

Conducted for the

TEXAS DEPARTMENT OF TRANSPORTATION
in cooperation with the
U.S. DEPARTMENT OF TRANSPORTATION
FEDERAL HIGHWAY ADMINISTRATION

by the

CENTER FOR TRANSPORTATION RESEARCH

Bureau of Engineering Research
THE UNIVERSITY OF TEXAS AT AUSTIN

January 2000
iv
IMPLEMENTATION RECOMMENDATIONS

Because of the lack of practical differences between mixes, no attempt was made to
develop acceptance criteria. However, it is believed that the work done in this project will
contribute to the growing knowledge and experience on static creep testing as a means of
evaluating rut resistance of asphalt concrete mixes.
In implementing the results of this research effort, adequate care must be taken to
identify the context of the recommendations and adapt them to each unique set of
circumstances. In order to extend the applicability of these conclusions, additional research
is required to verify the findings for a broader range of materials (asphalt and aggregates).
Only then can the findings be applied in a broader sense.
One underlying comment throughout the discussion of results and the conclusions is
the absence of practical differences in the results, even in the presence of statistical
differences. One concern in the current test setup is the fairly large variability mentioned in
Chapter 6. With a rather large variation between the three replicates, it is hard to say that one
mean is very different from another. One possible solution would be to increase the stress
level at which the test is run. Some other research efforts, such as the Asphalt Aggregate
Mixture Analysis System (AAMAS), have used a stress level of 207 kPa, and this stress level
should be investigated in the future.

DISCLAIMERS

The contents of this report reflect the views of the authors, who are responsible for
the facts and the accuracy of the data presented herein. The contents do not necessarily
reflect the official views or policies of either the Federal Highway Administration (FHWA)
or the Texas Department of Transportation (TxDOT). This report does not constitute a
standard, specification, or regulation.
There was no invention or discovery conceived or first actually reduced to practice in
the course of or under this contract, including any art, method, process, machine,
manufacture, design or composition of matter, or any new and useful improvement thereof,
or any variety of plant, which is or may be patentable under the patent laws of the United
States of America or any foreign country.

NOT INTENDED FOR CONSTRUCTION, BIDDING, OR PERMIT PURPOSES

Thomas W. Kennedy, P.E. (Texas No. 29596)

ACKNOWLEDGMENTS

The success of this project was made possible through the cooperation and assistance
of a number of dedicated people. Gratitude is expressed to all the people who made this
project possible.

v
Special thanks are extended to the director of the research project, Mr. Maghsoud
Tahmoressi (CSTM), who provided the research team with valuable guidance through the
course of the program. The Researchers would also like to thank the staff of the South
Central Superpave Center and the Bituminous Section of TxDOT’s Construction Division for
their support on this project.
Finally, the Researchers thank Center for Transportation Research staff members, in
particular Mr. Ray Donley and Ms. Clair LaVaye, for their support.

Research performed in cooperation with TxDOT and the U.S. Department of Transportation,
Federal Highway Administration.

PREFACE

This is the fourth report for research project 0-1250, The South Central Superpave
Center. This study was initiated and sponsored by the Texas Department of Transportation
(TxDOT) to evaluate the uniaxial static creep test for Superpave mixes. The research period
for the project was scheduled for one year, from September 1998 through August 1999. This
report presents the results, findings, conclusions, and recommendations based on the
laboratory study and testing of asphalt mixtures.

vi
TABLE OF CONTENTS

CHAPTER 1. INTRODUCTION ............................................................................................ 1

1.1. BACKGROUND.................................................................................................... 1
1.2 MOTIVATION2
1.3 OBJECTIVE AND METHODOLOGY.................................................................. 2
1.3.1 Phase I: Modifications to TxDOT’s Static Creep Test and
the Superpave Shear Tester ...................................................................... 3
1.3.2 Phase II: Evaluating the Effect of
Sample Diameters and Testing Machines ................................................ 3
1.3.3 Phase III: Evaluating the Effect of Sample Preparation Methods............ 4
1.3.4 Phase IV: Evaluating the Effect of Temperature and
Asphalt Content........................................................................................ 4
1.3.5 Phase V and VI: Evaluation of Good versus Poor Mixes ........................ 4
1.4 CONTRIBUTIONS TO INDUSTRY ..................................................................... 5
1.5 SCOPE .................................................................................................................... 5

CHAPTER 2. BACKGROUND ............................................................................................... 7

2.1 CREEP BEHAVIOR OF ASPHALT CONCRETE MIXES .................................. 7
2.1.1 Strain Rate Equation................................................................................. 7
2.1.2 Influence of Creep Stress Intensity on Strain Rate .................................. 9
2.1.3 Various Stages of Creep........................................................................... 9
2.2 ASPHALT CREEP TESTS................................................................................... 10
2.2.1 Evaluation of Creep Test Data ............................................................... 11
2.2.2 Comparison of Different Creep Modulus Criteria ................................. 11

CHAPTER 3. TESTING EQUIPMENT: CURRENT SETUP AND MODIFICATIONS..... 13

3.1 TXDOT STATIC CREEP TEST (TEST METHOD TEX-231-F) ....................... 13
3.1.1 Equipment Requirements ....................................................................... 13
3.1.2 Sample Preparation ................................................................................ 15
3.1.3 Test Procedure........................................................................................ 16
3.1.4 Calculation of Material Properties ......................................................... 17

CHAPTER 4. MIX DESIGN AND MATERIAL SELECTION ............................................ 21

4.1 SELECTION OF BASE MIX DESIGNS ............................................................. 21
4.1.1 Aggregates and Asphalt Used in the Base Mixes .................................. 22
4.2 ADDITIONAL MIX DESIGNS ........................................................................... 23
4.2.1 Superpave 12.5 mm Mix in Waco, Texas .............................................. 23
4.2.2 CMHB-C Mix in Austin, Texas ............................................................. 25

vii
4.2.3 CMHB-F Mix in New Braunfels, Texas .............................................. 26
4.3 LIMITATIONS AND ASSUMPTIONS............................................................. 27

CHAPTER 5. EXPERIMENT DESIGN............................................................................... 29

5.1 PHASE II: EVALUATING THE EFFECT OF SAMPLE
DIAMETERS AND TESTING MACHINES..................................................... 29
5.2 PHASE III: EVALUATING THE EFFECT OF SAMPLE
PREPARATIONS………………………………………………………………29
5.3 PHASE IV: EVALUATING THE EFFECT OF ASPHALT
CONTENT AND TEMPERATURE .................................................................. 30
5.4 PHASE V: EVALUATING “GOOD” AND “POOR” MIXES.......................... 30
5.5 PHASE VI: DEVELOPMENT OF ACCEPTANCE CRITERIA....................... 31

CHAPTER 6. DISCUSSION OF RESULTS........................................................................ 33

6.1 RESULTS PHASE II: SAMPLE DIAMETER AND
TESTING MACHINE EVALUATION.............................................................. 33
6.2 PHASE III RESULTS: SAMPLE PREPARATION
AND TESTING MACHINE EVALUATION.................................................... 37
6.3 PHASE IV RESULTS: ASPHALT CONTENT
AND TEMPERATURE EVALUATION ........................................................... 39
6.4 PHASE V RESULTS: GOOD VERSUS POOR MIX EVALUATION............. 44
6.5 PHASE VI DEVELOPMENT OF ACCEPTANCE CRITERIA........................ 47

CHAPTER 7. CONCLUSIONS AND RECOMMENDATIONS ........................................ 49

7.1 CONCLUSIONS FROM PHASES II THROUGH V ........................................ 49
7.2 CONCLUSIONS AND RECOMMENDATIONS ............................................. 50
7.3 FUTURE RESEARCH ....................................................................................... 51

BIBLIOGRAPHY ................................................................................................................. 53
APPENDIX A………………………………………………………………………………57

viii
LIST OF FIGURES

Figure 1.1: Proportion of Hot-Mix Asphalt Projects Designed Using the Superpave System
(Dunn, 1999) ..................................................................................................................... 1
Figure 2.1: Influence of Creep Stress Intensity on Strain Rate................................................. 9
Figure 2.2: Stages of Creep ..................................................................................................... 10
Figure 3.1: Static Creep Testing Apparatus ............................................................................ 13
Figure 3.2: Schematic Diagram of the Static Creep Setup...................................................... 14
Figure 3.3: Static Creep Specimens with One Side Capped ................................................... 15
Figure 3.4: 150 mm and 100 mm Diameter Specimens after Capping ................................... 16
Figure 3.5: Static Creep Test Setup......................................................................................... 16
Figure 3.6: Loading Pattern for Static Creep Test .................................................................. 17
Figure 3.7: Sample Shape of the Response Measured by the Two LVDTs ........................... 18
Figure 3.8: Relationship Between Strain Versus Time for a Typical Static Creep Test ........ 19
Figure 4.1: Gradation of Three Base Mixes............................................................................ 22
Figure 4.2: Aggregate Gradation for Superpave 12.5 mm Mix in Waco, Texas .................... 24
Figure 4.3: Aggregate Gradation for CMHB-C Mix on IH-35 in Austin ............................... 25
Figure 4.4: Aggregate Gradation for CMHB-F Mix on FM 1103 in New Braunfels, Texas . 28
Figure 5.1: Example of a Method to Develop Acceptance Criteria ........................................ 32
Figure 6.1: Phase II Column Chart for Stiffness (ETS denotes TxDOT’s EnduraTEC Systems
[ETS] device) .................................................................................................................. 34
Figure 6.2: Phase II Line Graph for Stiffness (ETS denotes TxDOT’s EnduraTEC Systems
[ETS] device) .................................................................................................................. 34
Figure 6.3: Phase II Column Plot for Permanent Strain (ETS denotes TxDOT’s EnduraTEC
Systems [ETS] device) .................................................................................................... 35
Figure 6.4: Phase II Column Plot for Slope (ETS denotes TxDOT’s EnduraTEC Systems
[ETS] device) .................................................................................................................. 36
Figure 6.5: Phase II Interaction between Mix and Diameter for Permanent Strain (ETS
denotes TxDOT’s EnduraTEC Systems [ETS] device) .................................................. 36
Figure 6.6: Phase II Interaction between Mix and Diameter for Slope (ETS denotes TxDOT’s
EnduraTEC Systems [ETS] device)................................................................................ 37
Figure 6.7: Phase III Column Chart for Stiffness (ETS denotes TxDOT’s EnduraTEC
Systems [ETS] device) .................................................................................................... 38
Figure 6.8: Phase III Column Chart for Permanent Strain (ETS denotes TxDOT’s
EnduraTEC Systems [ETS] device)................................................................................ 38
Figure 6.9: Phase III Column Chart for Slope (ETS denotes TxDOT’s EnduraTEC Systems
[ETS] device) .................................................................................................................. 39
Figure 6.10: Phase IV Column Chart for Stiffness ................................................................. 41
Figure 6.11: Phase IV Column Chart for Permanent Strain.................................................... 41

ix
Figure 6.12: Phase IV Column Chart for Slope ...................................................................... 42
Figure 6.13: Phase IV Interaction between Mix and Temperature for Stiffness..................... 43
Figure 6.14: Phase IV Interaction between Mix and Temperature for Permanent Strain....... 44
Figure 6.15: Phase IV Interaction between Mix and Temperature for Slope ......................... 44
Figure 6.16: Phase V Column Chart for Stiffness................................................................... 45
Figure 6.17: Phase V Column Chart for Permanent Strain ..................................................... 46
Figure 6.18: Phase V Column Chart for Slope........................................................................ 46

x
LIST OF TABLES

Table 2.1: Minimum Creep Modulus Values from Mahboub and Little (1988)..................... 12
Table 2.2: Acceptable Creep Modulus from Kronfuss et al. (1984) ....................................... 12
Table 4.1: Aggregate Gradation for Colorado Materials Stockpiles....................................... 22
Table 4.2: TF&A AC-20 Asphalt Binder Properties............................................................... 23
Table 4.3: Aggregate Gradation of Superpave 12.5 mm Mix in Waco, Texas....................... 24
Table 4.4: TF&A PG 76-22 Asphalt Binder Properties .......................................................... 25
Table 4.5: Aggregate Gradation of CMHB-C Mix on IH 35 in Austin .................................. 26
Table 4.6: TF&A AC-45P Asphalt Binder Properties ............................................................ 26
Table 4.7: Aggregate Gradation of CMHB-F Mix on IH 35 in Austin................................... 27
Table 5.1: Experiment Design for Phase II ............................................................................. 29
Table 5.2: Experiment Design for Phase III............................................................................ 30
Table 5.3: Experiment Design for Phase IV ........................................................................... 30
Table 5.4: Experiment Design for Phase V............................................................................. 31
Table 6.1: Statistical Analysis for Phase II ............................................................................. 33
Table 6.2: Statistical Analysis for Phase III............................................................................ 37
Table 6.3: Statistical Analysis for Phase IV............................................................................ 40
Table 6.4: Statistical Analysis for Phase V ............................................................................. 45

xi
xii
SUMMARY

From October 1987 through March 1993, a $50 million strategic highway research
program (SHRP) project was conducted to develop new ways to specify, test, and design
asphalt paving materials. The results of this research effort are collectively referred to as
Superpave.
Currently, the Level I Superpave mix design procedure is used to design asphalt
concrete mixes. This procedure is a volumetric mix design procedure. Though the use of
Superpave mix design procedures is becoming more and more common, it has always been
felt that there was a need for a strength test to validate the volumetric mix design procedure.
A good strength test would serve to “calm the fears” of concerned agencies and contractors.
The static creep test is one such test that could be used to validate the Superpave
volumetric mix design procedures. The overall objective of this research effort is to
investigate the sensitivity of the static creep test to changes in various mix parameters. These
parameters are the specimen diameter, specimen preparation method (molded versus saw-
cut), asphalt content, and gradation. In addition, the Superpave shear tester (SST) will be
evaluated to determine if it can also be used to run this static creep test. If possible, some
guidelines will be presented to assist in the development of acceptance criteria for Superpave
mixes.
For the study on the effect of diameter and sample preparation, only the effect of
slope is of significance from a practical standpoint. However, there were practical
differences in the slope between the two material testing devices. There was also a trend of
decreasing stiffness, increasing permanent strain, and decreasing slope with increasing
asphalt content at 40 °C (104 °F). However, most of the differences were not of practical
significance. The results were less consistent at 64 °C (147 °F).
Because of the lack of practical differences between mixes, no attempt was made to
develop acceptance criteria. However, it is believed that the work done in this project will
contribute to the growing knowledge and experience on static creep testing as a means of
evaluating rut resistance of asphalt concrete mixes.

xiii
xiv
CHAPTER 1. INTRODUCTION

1.1. BACKGROUND

In 1987, Congress established the Strategic Highway Research Program (SHRP) to

sponsor several coordinated research projects that were directed at improving the
performance and durability of roads in the United States. From October 1987 through March
1993, one $50 million SHRP project was conducted to develop new ways to specify, test, and
design asphalt paving materials. The results of this research effort are collectively referred to
as Superpave (Kennedy et al. 1994). A product that resulted from this project is the
Superpave shear tester (SST). This material testing device is discussed in greater detail in
Chapter 3.
The past four years have seen a steady increase in the use of Superpave technology.
Forty-seven agencies expect to fully implement the Superpave binder specification by 2000,
with two planning to complete implementation after 2000 and three with undetermined plans.
Meanwhile, twenty-eight agencies are on target to fully implement the volumetric mix design
procedures by 2000. Of the remaining survey respondents, twelve plan to implement
Superpave mix designs after 2000, and twelve have undetermined plans (Dunn 1999).
The percentages of hot-mix asphalt projects designed using the Superpave system
over the past four paving seasons are shown in Figure 1.1. From 1996 to 1999, the percent of
projects designed using the Superpave system increased from 1 percent to 41 percent. Based
on the survey, the Superpave lead states team expects that by 2001, 82 percent of all hot-mix
asphalt projects will be designed using the Superpave procedures (Dunn 1999).

50
Percent of Projects
Designed Using

40
Superpave

30
20
10
0
1996 1997 1998 1999
Year

Figure 1.1: Proportion of Hot-Mix Asphalt Projects Designed Using the Superpave System
(Dunn, 1999)

1
1.2 MOTIVATION

Currently, Level I Superpave mix design procedures are used to design asphalt
concrete mixes. This level is a volumetric mix design procedure (Cominsky et al. 1994).
Although use of the Superpave mix design procedures is becoming more and more common,
it has always been felt that there is a need for a strength test to validate the effectiveness of
the volumetric mix design procedure. It has been suggested that volumetric properties alone
may not be adequate by themselves to ensure good performance for high-volume roads
(Brown et al. 1999). A strength test would serve to “calm the fears” of agencies and
contractors who are accustomed to the Marshall and Hveem stability tests from the Marshall
and Hveem mix design procedures (Roberts et al. 1994).
The Texas Department of Transportation (TxDOT) uses the static creep test as an
integral part of its mix design for a special asphalt mixture called coarse matrix high binder
(CMHB). The test results are compared with pass/fail criteria involving creep stiffness,
permanent strain, and slope of the creep curve. This static creep test is one test that could
possibly be used to validate the effectiveness of the Superpave volumetric mix design
procedures.
TxDOT is investigating the possibility of using the static creep test as a strength test
for Superpave mixes. To accomplish this objective, the South Central Superpave Center
conducted a research project to investigate the static creep behavior of Superpave mixes.
The Superpave shear tester (SST) was evaluated as well to determine if it could produce test
results similar to those of TxDOT’s static creep machine (an EnduraTEC Systems [ETS]
device) and also if the SST could be used for running this test.
1.3 OBJECTIVE AND METHODOLOGY

The overall objective of this research effort was to evaluate the possibility of using a
static creep test similar to that currently used by TxDOT to help differentiate between good-
performing and poor-performing Superpave mixes. To achieve this objective, TxDOT’s
current static creep test setup was modified to accommodate Superpave gyratory specimens.
In addition, the following factors were evaluated:

• the effect of specimen diameter and compaction method (Phase I),

• the effect of using two different machines, TxDOT’s current static creep tester
and the SST (Phase I),
• the effect of sample preparation method (molded to the correct height as
compared to saw-cutting a regular Superpave gyratory specimen) (Phase II), and
• the effect of temperature and asphalt content (Phase III).

2
1.3.1 Phase I: Modifications to TxDOT’s Static Creep Test and the Superpave
Shear Tester
TxDOT’s static creep tester was designed to test 100 mm (3.9 in.) test specimens
compacted using the Texas gyratory compactor and needed to be modified to test specimens
compacted using the Superpave gyratory compactor (SGC). On the other hand, the SST was
designed to test 150 mm (5.9 in.) test specimens compacted using the SGC. However, the
built-in software in the SST was developed to run a range of Superpave performance tests,
but not a uniaxial static creep test. Accordingly, modifications were made to the current
software to develop the capability to run such a test on the SST. All these modifications
were carried out as Phase I of this research effort.
1.3.2 Phase II: Evaluating the Effect of Sample Diameters and Testing Machines
One concern with using the static creep test on Superpave gyratory-compacted
samples is the differences in compaction machines and specimen diameters. In this research
effort, the effect of diameter and the effect of compaction method were not considered
separately. Their individual effects were confounded. The reason for this decision lies in the
fact that production models of the Texas gyratory compactor (TGC) can compact only 100
mm (3.9 in.) diameter specimens, and production models of the Superpave gyratory
compactor (in its regular setup) can compact only 150 mm (5.9 in.) diameter specimens. It is
highly unlikely that either of these gyratory compactors will be used widely to compact
specimens to a different diameter on a production scale. The cost associated with acquiring
modified equipment to investigate these two factors separately was not justifiable.
Henceforth, whenever the effect of diameter is discussed, it should be noted that the 100 mm
(3.9 in.) diameter specimens were compacted using the TGC and that the 150 mm (5.9 in.)
specimens were compacted using the SGC.
However, it is important to recognize that there are differences in these two
compaction methods. Besides the difference in the specimen diameter, the pressure applied
during compaction is also different. In the Texas gyratory compactor, an initial pressure of
345 kPa (50 psi) is applied for most of the compaction process, but compaction ends with a
one-time final pressure of 17,238 kPa (2,500 psi) to level the specimen before the pressure is
released and the specimen extracted. The TGC uses a 5.8 degree angle of gyration. This
procedure is outlined in Texas Test Method Tex-206-F (Texas Department of Transportation
1991). The SGC uses a constant pressure of 600 kPa (87 psi), a 1.25 angle of gyration, and a
gyration rate of 30 gyrations per minute. This procedure is outlined in the AASHTO TP4,
PP19 procedure (American Association of State Highway and Transportation Officials
1998).
In addition, the static creep test was also run on the SST to evaluate the possibility of
running this static creep test on a similar materials testing device.

3
1.3.3 Phase III: Evaluating the Effect of Sample Preparation Methods
During SHRP, several test methods were developed for different performance tests.
Specimens tested on the SST and the IDT (indirect tension tester, another performance-
testing machine developed during SHRP) were saw-cut from the midsection of a regular
Superpave gyratory-compacted specimen. These saw-cut specimens were approximately 50
mm (2.0 in.) in height.
In recent years, some members of the asphalt industry have started to compact
specimens to a height of 50 mm (1.96 in.) in the SGC, instead of compacting specimens in
the traditional manner [to approximately 200 mm (7.9 in.)] and then saw-cutting out the
midsection of that specimen. This method of preparing specimens reduces material waste
and leads to substantial savings in sample preparation time (especially in the sieving of
aggregates). In this document, test specimens prepared in this manner are referred to as
molded specimens.
Currently, work is also underway at the Asphalt Institute to evaluate the effect of
sample preparation on the results of two commonly used performance tests run on the SST.
They are the frequency sweep and the repeated shear at constant height tests. There are no
available published results from this study at this time.
In order to conform with the specimen height used in TxDOT’s procedure, the saw-
cut specimens used in this research effort were cut to a height of approximately 51 mm (2.0
in.), instead of 50 mm (1.96), and henceforth are referred to as cut specimens.
1.3.4 Phase IV: Evaluating the Effect of Temperature and Asphalt Content
With the evolution of SHRP and the development of the Superpave system, there has
been a move towards testing asphalt mixes at temperatures that are likely to be experienced
in the field. This clearly can be seen in the new asphalt binder grading system —
performance grading. The Superpave performance grading system requires asphalt binders to
be tested at temperatures to which the binder will be exposed in the field. For example, in
central Texas, the average 7-day high temperature is 64 oC (147.2 oF) to 70 oC (158 oF).
Consequently, asphalt binders used in central Texas are tested at 64 oC or 70 oC to determine
if they are adequate for use in this region. However, many creep tests, such as the static
creep test, are run at 40 oC (104 oF) to determine the potential for rutting. In an effort to
evaluate the effect of running the mix tests at pavement temperatures, the static creep test
was run at both 40 oC and 64 oC.
In order to evaluate the effect of asphalt content, three mixes were compacted with
varying asphalt contents so as to determine the sensitivity of these mixes to asphalt content
changes.
1.3.5 Phases V and VI: Evaluation of Good versus Poor Mixes
In order to develop acceptance criteria for Superpave mixes, it is important to
evaluate a few good-performing and poor-performing pavements in the field. With the help

4
of TxDOT materials engineers in the Austin, San Antonio, New Braunfels, and Waco
Districts, the mix design of four good-performing pavements was selected to represent good
mix designs.
Mixes that were not performing well in the field were not selected to represent poor-
performing mixes. A pavement’s inadequate performance could be a result of poor
underlying layers, poor construction practices, or to a number of other factors. Therefore,
poor-performing pavements were not selected to represent poor mix designs. Instead, good-
performing mix designs were modified to create the poor-performing mixes. This
modification was achieved by compacting the laboratory specimens with 1 percent more
asphalt than optimum. Having 1 percent more asphalt over optimum is believed to lead to a
significantly greater susceptibility to rutting.
1.4 CONTRIBUTIONS TO INDUSTRY

The main contribution to the asphalt paving industry is the development of a static
creep test for Superpave mixes. In addition, evaluation of the effect of compaction method
and specimen diameter on static creep results will help TxDOT determine the applicability of
its CMHB acceptance criteria to Superpave mixtures. The evaluation of the SST will assist
state agencies and research institutions to evaluate the SST’s ability to conduct a comparable
static creep test. The evaluation of temperatures will also be useful in determining if a test at
64 oC (147.2 oF), instead of at 40 oC (104 oF), will better differentiate between a good-
performing and poor-performing mix. Last, but not least, the evaluation of the sample
preparation method (molded to the correct height as compared to saw-cutting a regular
Superpave gyratory specimen) will help determine if the time and materials savings in using
molded specimens accurately replicate results obtained from regular specimens that have
been saw-cut. It should be noted that while saw-cut Superpave gyratory specimens have
traditionally been tested on the SST, the practice does not mean that this sample preparation
method is superior to molding specimens to the correct height using the SGC.
1.5 SCOPE

This research project focused on evaluating various factors that could influence static
creep results, as mentioned in Section 1.3. This report is organized as follows:
Chapter 1 discusses issues surrounding the need for a strength test to validate
Superpave Level I mix design procedures. A brief overview of the objectives and the
research methodology is given. The research effort evaluated four factors considered in
developing acceptance criteria. They are diameter and compaction method, testing machines,
sample preparation method, and temperature and asphalt content.
Chapter 2 provides a theoretical overview of the creep behavior of asphalt concrete
mixes and discusses current static creep test criteria. First, the strain rate equation and the
generalized form used to illustrate creep in asphalt concrete mixes are discussed. Second,
material properties evaluated in creep tests and available creep test criteria are reviewed.

5
Chapter 3 discusses the development, current test procedures, and acceptance criteria
for TxDOT’s static creep test. It also discusses modifications made to TxDOT’s static creep
machine and the SST in order to achieve the objectives of this research effort.
Chapter 4 explains why the mix designs and materials used in this research effort
were chosen. This explanation includes a presentation of the material properties and how the
mixes were selected for this study.
Chapter 5 shows the experiment designs for each part of this research effort. It also
discusses the proposed approach to developing static creep acceptance criteria. Chapter 6
discusses the results from Phase II through Phase V of this research effort. Finally, Chapter 7
presents the conclusions from this research effort and makes recommendations for
developing acceptance criteria for Superpave mixes.

6
CHAPTER 2. BACKGROUND

2.1 CREEP BEHAVIOR OF ASPHALT CONCRETE MIXES

Creep is the time-dependent deformation resulting from a constant applied stress.

The deformation of soil involves time-dependent rearrangement of materials. Therefore creep
in soil and other particulate matter can be explained as a rate process. The basis of the rate
process theory is that atoms, molecules, and particles participating in a time-dependent flow
process, called flow units, are constrained from relative movement by energy barriers
separating equilibrium positions. The displacement of flow units to new positions requires
the introduction of sufficient activation energy to overcome the barrier (Mitchell 1976).
2.1.1 Strain Rate Equation
Of the total number of activated flow units at any instant, some may have sufficient
energy to overcome the energy barrier and pass over to the next equilibrium position.
Others, with insufficient activation energy to surmount the barrier, return to their original
position. For each unit that successfully crosses the barrier, there will be a displacement.
The component of this displacement in a given direction, multiplied by the number of
successful jumps per unit time, gives the rate of movement per unit time. Strain rate is this
movement expressed as a displacement per unit time. Mitchell (1976) explains that this
shear rate in a particular medium is influenced by a number of factors, as explained by the
equation:

ε = 2X (kT/h) exp(-∆F/RT) sinh(fλ/2kT) (Eq 2.1)

where ε = strain rate

∆F = activation energy,
T = absolute temperature (oK),
k = Boltzman constant,
h = Planck constant,
f = force,
λ = distance between successive equilibrium positions,
X = proportion of successful barrier crossings, and
R = universal gas constant.

7
Equation 2.1 shows that the strain rate is directly proportional to temperature. As temperature
increases, the rate of strain increases. If (fλ/2kT) < 1, then sinh (fλ/2kT) ≈ (fλ/2kT), and rate
is directly proportional to force, f. This is the case for ordinary Newtonian fluids. If (fλ/2kT)
> 1, Equation 2.1 is a reasonable first approximation of the rate process that explains creep of
asphalt concrete mixtures.
Hill (1973) presented a pseudo-theoretical physical model that attempted to explain
the relative displacement of a pair of adjacent aggregate particles in a viscoelastic matrix. He
assumed:

• a thin binder film and no hydrostatic effects in the binder,

• deformation is solely due to shear in the binder and volume change is negligible,
and
• under the influence of a constant stress, the rate of relative displacement of a pair
of adjacent particles is determined by the decreasing thickness of the binder film
and by the viscosity of the binder.

Based on these assumptions, Hill characterized the strain in asphalt mixtures using (Equation
2.2),

εmix/Fy = 2q [(1+ σt/3ηFx)1/2q – 1] (Eq 2.2)

where 1/Fx and 1/Fy are constants, q is an integer greater than 1 corresponding to the number
of “Chinese boxes” used in the model determined by the aggregate gradation, and σt/3ηFx is
the viscous component of strain.
Other work by Shell researchers (Van der Loo 1974 and 1978) has led to the
development of a formula (Equation 2.3) that related rut depth (∆H), thickness of asphalt
layer (H), stiffness of the mix (Sm), average compressive stress within the asphalt layer (σtire),
dynamic versus static deformation correction factor (Cm), and the stress distribution factor
(Z). For asphalt, Cm is in the range of 1.2 and 1.6.

∆H = H ⋅ σtire/Sm ⋅ Cm ⋅ Z (Eq 2.3)

Alterations were made to the Shell rutting prediction model based on research
conducted for Texas Department of Transportation (TxDOT) Research Project 7-2474
(Mahboub and Little 1988). The original Shell model for rutting prediction assumes a linear
relationship between the applied stress and the accumulated strain. However, this relationship
need not be linear. As such, a modified Shell equation was proposed (Equation 2.4),

8
h = H ⋅ (Zσtire/σlab)1.61 εVP (t) (Eq 2.4)

where h is the calculated rut depth, H is the asphalt layer thickness, Z is the vertical stress
distribution factor, σtire is the average contact pressure, σlab is the stress level at which the
creep test is run, and εVP (t) is the viscoplastic trend in the mixture measured in the creep test.
Yandell (1971), Kinder (1986), Lai and Anderson (1973), and Tseng and Litton
(1986) have proposed using some form of the power law to approximate the accumulation of
permanent deformation as a function of time or number of load cycles.
2.1.2 Influence of Creep Stress Intensity on Strain Rate
A schematic representation of the influence of creep stress intensity on strain rate at
some given time after stress application is shown in Figure 2.1. At low stresses, creep rates
are small and of little practical importance. The curve shape is compatible with the
hyperbolic sine function predicted by the rate process in Equation 2.1. In the midrange of
stresses, a nearly linear relationship is found between logarithm of strain rate and stress. This
is also predicted in Equation 2.1, when the argument of the hyperbolic sine is greater than 1.
At stresses approaching the strength of the material, the strain rates become very large,
indicating the onset of failure (Mitchell 1976).
Log Strain Rate (ε)

Failure

Creep Stress Intensity (D)

Figure 2.1: Influence of Creep Stress Intensity on Strain Rate

2.1.3 Various Stages of Creep

Figure 2.2 illustrates another generalized form used to show the various stages of
creep. This figure provides the generalized shape for the strain (or deformation) as a function
of time. The first stage, transient or primary creep, consists of a relatively high initial strain
rate that decreases with time. The second stage, steady state or secondary creep, consists of a

9
constant strain rate (Finnie and Heller 1959). The third state, tertiary creep, consists of
acceleration in strain rate leading to failure.

Transient Steady State

Tertiary
or Primary or Secondary
Strain

Time

Figure 2.2: Stages of Creep

The relationship between creep strain and the logarithm of time may be linear,
concave upward, or concave downward. However, a linear relationship is often assumed as a
practical approximation because of its simplicity in analysis and ease of use. There is no
fundamental law of behavior that dictates one or the other.
2.2 ASPHALT CREEP TESTS

Uniaxial creep tests have been used to evaluate rutting susceptibility of asphalt
concrete mixes for a long time. The use of these tests is a result of the test’s relative
simplicity and because of its logical connection with permanent deformation in asphalt
mixes. However, as with all other laboratory tests, one major problem with the laboratory
creep tests is the difficulty in relating laboratory results with actual field performance.
Van der Loo (1989) analyzed the relationship between rutting in the field and creep
(static and dynamic) testing in the laboratory. After analyzing the use of results from
laboratory-prepared specimens to predict rutting behavior, Van der Loo concluded “it may be
that the main purpose of laboratory test methods must be limited to the ranking of materials
rather than the prediction of rut depth.”
Though creep tests on laboratory-prepared specimens may not be able to accurately
predict rut depth, some work has been done to develop acceptance criteria for asphalt
concrete mixes based on creep results.

10
2.2.1 Evaluation of Creep Test Data
The total strain at failure after a period of loading is often used as a criterion for
accepting or rejecting mixes. Typically, the strain after 1 hour of loading is used. Such was
the time period used in the asphalt aggregate mixture analysis system (AAMAS, a mixture
analysis system developed prior to, and which was to serve as a starting point for, the
Strategic Highway Research Program [SHRP]). A minimum creep modulus after 1 hour of
loading was recommended in AAMAS (Von Quintus et al. 1991). Another commonly used
criterion is the creep stiffness. This is calculated by dividing the applied stress by the
resulting strain at a specific time period. This is often also calculated after 1 hour of loading.
Two other criteria are the permanent strain and the slope. The permanent strain is the
residual strain remaining after a sample is allowed to recover the elastic portion of the strain.
For permanent strain criteria, a fixed-loading period is followed by a fixed-recovery period.
TxDOT’s test method Tex-231-F is one such example (Texas Department of Transportation
1995). Slope is defined as the gradient of the strain versus time curve. A maximum slope is
specified for the end of the loading period to determine whether the sample is in the
secondary creep phase or in the tertiary creep phase of creep. The secondary creep phase has
a relatively small slope compared to the slope in the tertiary creep phase. Tertiary creep is an
indication that a sample is about to fail and may not be acceptable.
2.2.2 Comparison of Different Creep Modulus Criteria
The AAMAS creep test has a loading period of 1 hour and a recovery time of 1 hour.
The 1-hour test period is relatively popular because it is long enough to be applicable to the
loading conditions during which rutting occurs, yet short enough to be practical.
The minimum creep modulus values at 1 hour of loading from the creep modulus
charts developed by Mahboub and Little (1988) are as shown in Table 2.1. These values are
based on testing conducted at 40 oC (104 oF) under uniaxial compressive stresses that are
representative of actual pavement conditions.
Researchers have also developed similar criteria from other creep tests. Khedr (1986)
recommends a minimum creep modulus of 137.9 MPa (20 ksi) after 1 hour of testing under
an applied pressure of 207 kPa (30 psi). Table 2.2 shows the range of acceptable creep
modulus after 1 hour of loading at 40 oC (104 oF) under applied pressures of 103 kPa (14.9
psi) suggested by Kronfuss et al. (1984). Kronfuss et al. also established an upper limit for
creep modulus at 46.5 Mpa (6.7 ksi) because they believe that higher creep moduli could lead
to mixes susceptible to load-induced fatigue or thermal effects. However, this criterion was
established for cooler climates and should be interpreted with caution before applying it to
other environmental conditions. Viljoen and Meadows (1981) developed a minimum creep
modulus criterion of 82.7 MPa (12 ksi) after 100 minutes of loading at 40 oC (104 oF) under
an applied pressure of 207 kPa (30 psi).

11
Table 2.1: Minimum Creep Modulus Values from Mahboub and Little (1988)
Pavement Category Rut Potential Min. Creep Modulus (MPa)
Low 69
Asphalt Concrete Over Rigid Base
Moderate 34.5
Full-Depth Asphalt Concrete (Intermediate Low 55.2
Layers) Moderate 20.7
Low 27.6
Full-Depth Asphalt Concrete (Lower Layers)
Moderate 17.2
Low 55.2
Surface Asphalt Concrete Layers
Moderate 27.6

Table 2.2: Acceptable Creep Modulus from Kronfuss et al. (1984)

Traffic Intensity Level Range of Acceptable Creep Modulus (MPa)
Low ≥ 20.7
Moderate 20.7 – 31.0
High 31.0 – 45.6

Sousa et al. (1991) documented one of the most comprehensive studies on the effects
of mix variables on compressive creep properties of asphalt concrete mixtures. The test
temperature was 40 oC (104 oF), the applied vertical pressure was 207 kPa (30 psi), and the
loading time was 1 hour. The parameters considered in this evaluation included the creep
modulus or permanent strain following a specific period of loading, time to reach a critical
level of strain (time to rupture), and slope of the strain-versus-time curve in a designated
region, such as the steady-state region. Mixture variables considered in this study include
asphalt cement type, asphalt content, aggregate type, air void content, compaction
temperature, and stress level. A review of this data indicates that compressive creep is
sensitive to all of these factors.
However, the levels selected for each of the variables need to be evaluated carefully
before a general conclusion for all possible levels for each factor is reached. Overall, a creep
modulus greater than 69 Mpa (10 ksi) under realistic conditions indicates a mix that has low
sensitivity to rutting. Creep moduli in the range of 41.4 (6 ksi) to 69 MPa (10 ksi) indicate
that a mix is moderately to highly sensitive to rutting.
Little et al. (1993) also conducted a study on 100 mixtures to develop critical creep
values of slope for the steady-state portion of the creep-versus-time curve and permanent
strain after 1 hour. The researcher concluded that creep test parameters that have been shown
to relate rationally to permanent deformation potential are strain at 1 hour of loading,
concomitant creep stiffness at 1 hour, and the log-log slope of the steady-state portion of the
creep strain versus time curve.

12
CHAPTER 3. TESTING EQUIPMENT: CURRENT SETUP AND MODIFICATIONS

3.1 TXDOT STATIC CREEP TEST (TEST METHOD TEX-231-F)

The Texas static creep test was initially developed as a strength test for crumb rubber
modified (CRM) asphalt hot mix. This work was completed prior to 1993 (Tahmoressi
1993). TxDOT uses the static creep test to determine resistance to permanent deformation of
bituminous mixtures at temperatures and loads similar to those experienced by these
materials in the field (Texas Department of Transportation 1994).
3.1.1 Equipment Requirements
The loading press must be capable of applying a constant axial load, up to 2224 N,
over a specified time period. It must also be capable of measuring the axial load to an
accuracy of plus or minus 1 percent of the applied axial load. The temperature control system
must be capable of controlling temperature within the range of 40 oC (104 oF) to 60 oC (140
o
F). The temperature must be held to within 1 oC (33.8 oF) of the specified test temperature.
Vertical deformations must be measured with a linear variable displacement
transducer (LVDT). Two LVDTs, placed diametrically opposite each other, must be used to
measure this deformation. The resolution on each LVDT must be better than 0.0025 mm
(0.000098 in.). Smooth-loading platens must be used to minimize the effects of friction on
the ends of the sample. The upper-load platen must be of the same diameter as the sample
being tested to provide for positive centering of the specimen under load. The upper platen
must be of the floating compression type to account for minor deviations in a specimen’s
surface (Texas Department of Transportation 1994). Figure 3.1 shows a picture of the test
setup, and Figure 3.2 shows a schematic diagram of the test setup.

Figure 3.1: Static Creep Testing Apparatus

13
Figure 3.2: Schematic Diagram of the Static Creep Setup
(Texas Department of Transportation 1994)

14
3.1.2 Sample Preparation
TxDOT static creep equipment is designed to test specimens compacted with the
Texas gyratory compactor. This compaction method is described in Test Method Tex-205-F,
“Laboratory Method of Mixing Bituminous Mixtures” (Texas Department of Transportation
1994).
Texas gyratory specimens are compacted to a diameter of 101 mm (3.98 in.). For the
static creep test, they are compacted to a height of 51 (2 in.) ± 2 mm (0.08 in.). Prior to
testing, the specimens are leveled with hydrocal, a capping compound that is a combination
of gypsum cement and hydrostone, each contributing 50 percent by volume. The two
components are mixed with water to an appropriate consistency, with the specimens then
placed on top of the compound and leveled with a bubble level. Figure 3.3 shows three test
specimens after one side has been capped. Once the hydrocal has dried, the specimen is
turned over and the process is repeated to cap the other end of the specimen (Texas
Department of Transportation 1994). Figure 3.4 shows a 100 mm (3.9 in.) specimen and a
150 mm (5.9 in.) specimen after both ends have been capped.

Figure 3.3: Static Creep Specimens with One Side Capped

15
Figure 3.4: 150 mm and 100 mm Diameter Specimens after Capping

3.1.3 Test Procedure

The test specimens are placed into a controlled temperature chamber maintained at
40 C (104 oF) for 3 to 5 hours prior to the start of the test to bring the specimens to the test
o

temperature. Following this preparation, the specimen is mounted as shown in Figure 3.5.

Figure 3.5: Static Creep Test Setup

16
To achieve more uniform contact between the loading platens and the specimen, three
preloading cycles are applied prior to the test. Each cycle consists of a 556 N load applied
for 1 minute, followed by a 1-minute rest period. After the preloading cycles, a 556 N load is
applied to the specimen for 1 hour. At the end of 1 hour, the load is removed and the
specimen allowed to rebound for 10 minutes (Texas Department of Transportation 1994).
Studies conducted by TxDOT in 1995 indicated that applying a 22 N load for the 1-
minute rest period eliminated the possibility of a dynamic load hitting the specimen during
loading because the loading head would no longer float during the 1-minute rest period. In
addition, applying a 22 N load during the recovery phase of the test (when the 556 N load is
removed) improved the repeatability of the permanent strain readings. As such, 1 N loads
are applied during these rest or recovery periods of the test. Figure 3.6 shows the load
applied during the test (Texas Department of Transportation 1995).

6x1 60 10
minutes minutes minutes

556
Load, N

22

6 66 76
Time, minutes

Figure 3.6: Loading Pattern for Static Creep Test

3.1.4 Calculation of Material Properties

Three material properties are measured from the static creep test: stiffness, permanent
strain, and slope of the steady-state curve. The procedure to calculate required material
properties is described in Texas Test Method Tex-231-F (Texas Department of

17
Transportation 1994). To begin, the average deformation is calculated by averaging the two
LVDT readings. The strain is then calculated by dividing the average deformation by the
specimen thickness (height). This is shown in Equation 3.1.

Strain (mm/mm) = Average Deformation / Thickness (Eq 3.1)

Next, strain is plotted against time for each of the three specimens. Figure 3.7 shows
an example of the typical shape of the measured responses from each of the two LVDTs.
Figure 3.8 shows the shape of a typical strain (averaged from the two LVDTs) versus time
plot for one specimen. From this plot, the total strain (mm/mm), the permanent strain
(mm/mm), and the slope of the steady-state curve (mm/mm/sec) are obtained. The creep
stiffness is calculated using Equation 3.2.

Creep Stiffness = Applied Pressure / Total Strain (Eq 3.2)

Time, minutes

Figure 3.7: Sample Shape of the Response Measured by the Two LVDTs (Texas Department
of Transportation 1994)

18
Figure 3.8: Relationship between strain versus time for a typical static creep test (Texas
Department of Transportation 1994)

19
20
CHAPTER 4. MIX DESIGN AND MATERIAL SELECTION

In this research effort, several sources for aggregates and asphalts were used. Among
the aggregates used were limestone rock from Colorado Materials and Capitol Aggregates,
sandstone from Delta Materials, and field sand from Odell-Greer Construction.
Coincidentally, the three types of asphalts used on all of these projects were supplied by
Texas Fuel and Asphalt (TF&A) in Corpus Christi, Texas.
4.1 SELECTION OF BASE MIX DESIGNS

BIH-35 (New Braunfels) mix design was chosen as the primary mix design. This
mix was selected for a number of reasons. For one, the BIH-35 mix design is a coarse
(aggregate gradation passes below the restricted zone) Superpave mix that is performing well
in the field. Choosing BIH-35 as the primary mix design will shed more light on the behavior
of Superpave mixes. In addition, the mix design was conducted at the South Central
Superpave Center (SCSC). The aggregates (from Colorado Materials) were readily available
at the SCSC, and the staff had a fair amount of experience with this aggregate source because
Colorado Materials aggregates are used as the laboratory standard. The AC-20 asphalt that
was used was also readily available from TF&A.
Figure 4.1 shows the aggregate gradation for BIH-35 (Mix A), as well as for two
other finer mixes. The intermediate mix (Mix B) has a gradation that passes through the
restricted zone. The fine mix (Mix C) passes over the restricted zone. Mixes B and C were
developed to evaluate the effect of having additional fines in a mix. These three mixes will
henceforth be referred to as the base mixes. All three base mixes are composed of aggregates
from Colorado Materials and asphalt binder from TF&A. In Figure 4.1, the black solid
squares represent boundary limits that the gradations must meet to satisfy Superpave 19 mm
(0.75 in.) mix requirements.
It should be noted that the restricted zone resides along the maximum density
gradation between the intermediate size [either 4.75 mm (0.19 in.) or 2.36 mm (.09 in.)] and
the 0.3 mm (0.01 in.) size. It forms a area through which aggregate gradations are not
permitted to pass under the Superpave system. Gradations that pass through this restricted
zone, violating Superpave design requirements, are often called humped gradations, a term
that denotes the noticeable hump in the gradation curve. In most cases, a humped gradation
indicates a mixture that possesses too much fine sand in relation to total sand. This type of
gradation can result in tender mix behavior that manifests itself by being difficult to compact
during construction and more susceptible to permanent deformation. Gradations that violate
this zone also possess weak aggregate skeletons that depend too much on asphalt binder
stiffness to achieve shear strength and can be sensitive to asphalt content (McGennis et al.
1995).

21
100
90
80

Percent Passing
70
60
50
40 Blend A
30 Blend B
20 Blend C
10
0
0.075 0.300 2.36 4.75 9.5 12.5 19 25
Sieve Size (mm) to 0.45 Power

Figure 4.1: Gradation of Three Base Mixes

4.1.1 Aggregates and Asphalt Used in the Base Mixes

The contractor of the BIH-35 project, Colorado Materials, provided the aggregates
used in the three base mixes. The component aggregate stockpiles were C-Rock, D-Rock, F-
Rock, and washed screenings. Table 4.1 shows the gradation of these four component
aggregate stockpiles.
Table 4.1: Aggregate Gradation for Colorado Materials Stockpiles
Percent Passing
Colo. Matls. Colo. Matls. Colo. Matls. Colo. Matls.
Size, mm
C-Rock D-Rock F-Rock Washed Scrn.
25 100.0 100.0 100.0 100.0
19 98.7 100.0 100.0 100.0
12.5 70.0 99.3 100.0 100.0
9.5 7.9 70.1 100.0 100.0
4.75 2.7 6.9 73.6 98.4
2.36 2.0 3.9 13.0 81.5
1.18 1.9 3.0 4.6 53.0
0.6 1.7 2.8 4.0 34.0
0.3 1.6 2.5 3.5 21.5
0.15 1.5 2.3 3.3 12.0
0.0075 1.4 1.9 3.0 10.0

22
In trying to replicate the mix that is in place on IH-35 in New Braunfels, the research
team developed the design of Mix A based on aggregate gradations and asphalt content from
a QC/QA testing summary provided by the New Braunfels District office. The aggregate
gradation is shown in Figure 4.1; the asphalt content is 5.1 percent of the total mass of the
mix. As per the original design for BIH-35, Mix A was designed based on a design gyration
level (Ndesign) of 96 gyrations. The two other mixes (Mix B and Mix C) were designed based
on a design gyration level (Ndesign) of 68 gyrations. Mixes B and C both have an asphalt
content of 4.8 percent by total mass of the mix.
The asphalt binder used in the BIH-35 project is an AC-20 supplied by TF&A. Table
4.2 shows the asphalt binder properties for this AC-20. This asphalt was graded as a PG 64-
22 using the Superpave performance graded classification system. The same asphalt was
used in Mixes B and C, so that the effect of having greater quantities of fines would not be
confounded with the type of asphalt.

Table 4.2: TF&A AC-20 Asphalt Binder Properties

Property Test Result Specification
Unaged Properties:
G*/sin # oC 1.56 kPa 1.00 kPa min.
RTFO-Aged Properties:
G*/sin # oC 3.28 kPa 2.20 kPa min.
PAV-Aged Properties:
G*sin # oC 3677 kPa 5000 kPa max.
S @ -12 oC 228 MPa 300 MPa max.
m @ -12 oC 0.357 0.300 min.

4.2 ADDITIONAL MIX DESIGNS

In addition to the three base mix designs, the mix designs for three other mixes that
are performing well in the field were selected for evaluation. Once again, the TxDOT district
offices assisted in the selection of these mixes. One of the mixes is a Superpave 12.5 mm
(0.49 in.) mix laid down on IH-35 in Waco, Texas. The other two are coarse matrix high
binder (CMHB) mixes. The first one is a CMHB-C (coarse CMHB) laid down on IH-35 in
Austin, Texas. The second one is a CMHB-F (fine CMHB) laid down on FM 1103 in New
Braunfels, Texas.
4.2.1 Superpave 12.5 mm Mix in Waco, Texas
The Superpave 12.5 mm (0.49 in.) mix was laid down on IH-35 in Waco, Texas. The
field gradation and the laboratory gradation used in this research effort for this mix are
shown in Figure 4.2; the asphalt content is 4.9 percent of the total mass of the mix. In Figure

23
4.2, the black solid squares represent boundary limits that the gradations must meet to satisfy
Superpave 12.5 mm (0.49 in.) mix requirements.

100
90
80
70
Percent Passing

60
Lab
50
Field
40
30
20
10
0
0.075 0.300 2.36 4.75 2.36 12.5 19
Sieve Size (mm) to 0.45 Power

Figure 4.2: Aggregate Gradation for Superpave 12.5 mm (0.49 in.) Mix in Waco, Texas

Delta Materials, Capitol Aggregates, and Odell-Greer supplied the aggregates for this
mix. The gradations for the aggregate stockpiles are shown in Table 4.3. TF&A supplied the
PG 76-22 asphalt binder. Table 4.4 shows the properties of this asphalt binder.

Table 4.3: Aggregate Gradation of Superpave 12.5 mm (0.49 in.) Mix in Waco, Texas
Percent Passing
Delta Matls. Cap. Aggr. Cap. Aggr. Odell-Greer
Size, mm
D-Rock F-Rock Screenings Field Sand
19 100.0 100.0 100.0 100.0
12.5 99.0 100.0 100.0 100.0
9.5 77.0 100.0 100.0 100.0
4.75 14.0 69.0 100.0 100.0
2.36 5.0 16.0 87.0 100.0
1.18 3.0 6.0 59.0 100.0
0.6 3.0 4.0 39.0 100.0
0.3 3.0 3.0 23.0 89.0
0.15 3.0 3.0 11.0 34.0
0.0075 3.0 3.0 6.0 11.0

24
Table 4.4: TF&A PG 76-22 Asphalt Binder Properties
Property Test Result Specification
Unaged Properties:
G*/sin # oC 1.32 kPa 1.00 kPa min.
RTFO-Aged Properties:
G*/sin # oC 2.80 kPa 2.20 kPa min.
PAV-Aged Properties:
G*sin # oC 1676 kPa 5000 kPa max.
S @ -12 oC 202 MPa 300 MPa max.
m @ -12 oC 0.301 0.300 min.

4.2.2 CMHB-C Mix in Austin, Texas

The CMHB-C mix was laid down on IH-35 in Austin, Texas. The field gradation and
the laboratory gradation replicated in this research effort for this mix are shown in Figure 4.3;
the asphalt content is 5.3 percent of the total mass of the mix. In Figure 4.3, the black circles
represent boundary limits that the gradations must meet to satisfy CMHB-C requirements.

100
90
80
Percent Passing

70
60
Lab
50
Field
40
30
20
10
0
0.0075 0.6 2.36 4.75 9.5 19 25
Sieve Size (mm) to 0.45 Power

Figure 4.3: Aggregate Gradation for CMHB-C Mix on IH 35 in Austin

Delta Materials and Capitol Aggregates supplied the aggregates for this mix. The
gradations for the aggregate stockpiles are shown in Table 4.5. TF&A supplied the AC-45P
asphalt binder. Table 4.6 shows the properties of this asphalt binder. This asphalt was graded
as a PG 70-22 using the Superpave performance graded classification system. It should be
noted that this asphalt binder barely missed being graded out as a PG 70-28 because its m-
value is 0.001 lower than the required 0.300.

25
Table 4.5: Aggregate Gradation of CMHB-C Mix on IH 35 in Austin
Percent Passing
Delta Matls. Cap. Aggr. Cap. Aggr. Cap Aggr.
Size, mm
C-Rock D-Rock F-Rock Scalpings
25 100.0 100.0 100.0 100.0
22.4 100.0 100.0 100.0 100.0
16.0 94.0 100.0 100.0 100.0
9.5 5.0 76.0 100.0 100.0
4.75 2.0 7.0 58.0 100.0
2.00 1.0 2.0 7.0 85.0
0.425 0.0 0.0 3.0 55.0
0.180 0.0 0.0 2.0 46.0
0.0075 0.0 0.0 0.7 39.0

Table 4.6: TF&A AC-45P Asphalt Binder Properties

Property Test Result Specification
Unaged Properties:
G*/sin # oC 2.30 kPa 1.00 kPa min.
RTFO-Aged Properties:
G*/sin # oC 3.39 kPa 2.20 kPa min.
PAV-Aged Properties:
G*sin # oC 1855 kPa 5000 kPa max.
o
S @ -12 C 141 MPa 300 MPa max.
m @ -12 oC 0.361 0.300 min.

4.2.3 CMHB-F Mix in New Braunfels, Texas

The CMHB-F mix was laid down on FM 1103 in New Braunfels, Texas. The field
gradation and the laboratory gradation replicated in this research effort for this mix are
shown in Figure 4.4; the asphalt content is 6.5 percent of the total mass of the mix. The
black circles represent boundary limits that the gradations must meet to satisfy CMHB-F
requirements.

26
100
90
80

Percent Passing
70
60
Lab
50
Field
40
30
20
10
0
0.075 0.3 0.6 2.36 4.75 9.5 12.5 19 25
Sieve Size (mm) to 0.45 Power

Figure 4.4: Aggregate Gradation for CMHB-F Mix on FM 1103 in New Braunfels, Texas

Colorado Materials supplied the aggregates for this mix. The gradations for the
aggregate stockpiles are shown in Table 4.7. TF&A supplied the AC-20 asphalt binder. This
is the same asphalt binder as the one used in the BIH-35 mix in New Braunfels. Table 4.2
shows the properties of this asphalt binder.

Table 4.7: Aggregate Gradation of CMHB-F Mix on IH 35 in Austin

Percent Passing
Colorado Materials Colorado Materials Colorado Materials
Size, mm
D-Rock F-Rock Screenings
22.4 100.0 100.0 100.0
16.0 100.0 100.0 100.0
12.5 99.1 100.0 100.0
9.5 73.9 100.0 100.0
4.75 7.3 68.8 100.0
2.00 7.3 7.5 88.8
0.425 2.8 3.8 46.9
0.180 2.6 3.1 34.4
0.075 2.3 2.9 26.1

4.3 LIMITATIONS AND ASSUMPTIONS

As with any effort to replicate mixes that are already laid in the field, there are
limitations to the extent to which the mixes can be reproduced. It should be noted that the
goal of this research effort was to replicate what was in the field in terms of aggregate type,
aggregate gradation, asphalt type, and asphalt content. However, designs are based on
specimens compacted in the laboratory. Therefore, the difference between compaction

27
method in the laboratory and compaction method in the field is taken into account.
Compaction devices used in the design stage (i.e., the Texas gyratory compactor or the SGC)
adequately prepared the mixes for this research effort. It should be noted that all mixes were
compacted to an air void content of 4, plus or minus 0.5 percent, in order to ensure
consistency within the different mixes as well as between different mixes.
It is impossible to use exactly the same aggregates and asphalt used in the field
projects unless some of the materials are stored during construction for future use (as was
done on the Westrack project). Therefore, the original aggregate and asphalt producers were
contracted to acquire materials that were as close to the materials used in the actual projects.
This use of the original producers is the best that could be done.
However, the gradations and asphalt contents used in this research effort are based on
the data collected and recorded in the QC/QA summary sheets. For this reason, the
replicated mixes actually represent the field mixes and not those originally designed in the
laboratory. This approach was chosen because it is common for the contractor to adjust the
mix design during production, and the good-performing mixes were selected based on their
performance in the field and not because of the performance of the laboratory-compacted
specimens used in the design phase.
Volumetric data for the samples tested from each of the mix designs are shown in
Appendix A.

28
CHAPTER 5. EXPERIMENT DESIGN

The overall research effort is divided into six phases. The first phase involves
modifications to TxDOT’s static creep tester and the SST. The next four phases involve the
evaluation of different test parameters. The experiment designs for these four phases will be
presented separately in the subsequent sections. The last phase, Phase VI, will discuss
recommendations for developing acceptance criteria, if possible. It should be noted that the
numbers shown in the subsequent tables represent the number of replicates used for that
combination of factors.
5.1 PHASE II: EVALUATING THE EFFECT OF SAMPLE DIAMETERS AND
TESTING MACHINES

In Phase II, the effect of sample diameters and testing machines was evaluated. The
three base mixes (A, B, and C) discussed in Chapter 4 were used in this phase. These mixes
were molded to a diameter of 100 mm (3.9 in.) using the Texas gyratory compactor (TGC)
and molded to a diameter of 150 mm (5.9 in.) using the Superpave gyratory compactor
(SGC). The specimen heights were kept constant at approximately 51 mm (2 in.). In Phase
II, all the tests were conducted at 40 oC (104 oF) on molded specimens. Table 5.1 shows this
experiment design. Three replicates were used for each combination of levels.

Table 5.1: Experiment Design for Phase II

Mix
Static Creep
Diameter (mm) A B C
Machine
100 3 3 3
TxDOT (TGC)
150 3 3 3
100 3 3 3
SST (SGC)
150 3 3 3

5.2 PHASE III: EVALUATING THE EFFECT OF SAMPLE PREPARATION

In Phase III, the effect of sample preparation is evaluated. The same three base mixes
used in Phase II (A, B, and C) were used in this phase. The cut specimens were prepared by
saw-cutting a 51 mm (2 in.) section out of the middle of a regular SGC-compacted specimen.
The molded specimens were molded in the SGC directly to the required height of 51 mm (2
in.). Phase III, all tests were conducted at 40 oC (104 oF). Table 5.2 shows this experiment
design. Three replicates were used for each combination of levels.

29
Table 5.2: Experiment Design for Phase III
Mix
Static
Creep Preparation A B C
Machine
Molded 3 3 3
TxDOT
Cut 3 3 3
Molded 3 3 3
SST
Cut 3 3 3

5.3 PHASE IV: EVALUATING THE EFFECT OF ASPHALT CONTENT AND

TEMPERATURE

In Phase IV, the effect of sample preparation was evaluated. The same three base
mixes used in Phases I and II (A, B, and C) were used in this phase. In Phase IV, all the tests
were conducted on the TxDOT static creep tester at 40 oC (104 oF) and 64 oC (147.2 oF) on
molded specimens. Table 5.3 shows this experiment design. Three replicates were used for
each combination of levels.

Table 5.3: Experiment Design for Phase IV

Mix
Asphalt Content Temperature (oC) A B C
40 3 3 3
Optimum
64 3 3 3
40 3 3 3
Optimum + 0.5%
64 3 3 3
40 3 3 3
Optimum + 1.0%
64 3 3 3

In addition to the test matrix shown in Table 5.3, the BIH-35 mix (A) was also
molded at an asphalt content of 0.5 percent less than optimum.
5.4 PHASE V: EVALUATING GOOD AND POOR MIXES

In Phase V, good-performing and poor-performing mixes in the Central Texas region

were selected for this evaluation. Details of the reasons why these mixes were selected are
discussed in Chapter 4. In Phase V, all tests were conducted at 40 oC and 64 oC on molded

30
specimens. Table 5.4 shows this experiment design. Three replicates were used for each
combination of levels.
Table 5.4: Experiment Design for Phase V
Type of Mix Temperature (oC) Good Mix Poor Mix
40 3 3
Superpave 12.5 mm
64 3 3
40 3 3
Superpave 19.0 mm
64 3 3
40 3 3
CMHB-C
64 3 3
40 3 3
CMHB-F
64 3 3

5.5 PHASE VI: DEVELOPMENT OF ACCEPTANCE CRITERIA

The first phase of this experiment focused on evaluating the effect of specimen
diameter (compounded with compaction method) and testing device. The second phase of
this experiment focused on evaluating two specimen-preparation methods for SGC-
compacted specimens and testing device. Although part of the overall picture, these two
phases do not contribute directly to the development of acceptance criteria.
The third phase of this experiment focused on evaluating the effect of asphalt content
and testing temperature. The fourth phase focused on differentiating the mix properties of
good and poor mixes. Results from these two phases, coupled with engineering judgment
and with experience obtained from previous research efforts, will be used to develop some
kind of acceptance criteria.
Figure 5.1 shows an example of how acceptance criteria may be developed. In this
example, mix property value (P) is plotted for good-performing and poor-performing mixes.
Accepting the fact that good-performing mixes have higher P values than poor-performing
mixes, a particular P value can be selected to delineate between good and poor mixes. In this
example, a value A is selected to delineate between good-performing and poor-performing
mixes.

31
Property, P
Y

Good Mixes Poor Mixes

Figure 5.1: Example of a Method to Develop Acceptance Criteria

If a clear distinction cannot be made between good-performing and poor-performing
mixes, then engineering judgment will be used to determine what appropriate guidelines can
be recommended for developing acceptance criteria.

32
CHAPTER 6. DISCUSSION OF RESULTS

The modifications to TxDOT’s static creep test and the SST (Phase I) are discussed in
Chapter 3. The results for Phases II through V are discussed in this chapter. For these four
phases, three properties were evaluated — stiffness, permanent strain, and slope. Tests
results from all the tests can be found in Appendix A.
Based on the variability within replicates in this study, practical significance was
determined to be differences greater than 20,000 kPa (2,900 psi) for stiffness, differences
greater than 0.1 x 10-3 mm/mm for permanent strain, and differences greater than 1.0 x 10-8
mm/mm/sec for slope.
6.1 PHASE II RESULTS: SAMPLE DIAMETER AND TESTING MACHINE
EVALUATION

In Phase II of this research effort, the effect of specimen diameter and testing
machine was evaluated. The two specimen diameters used were 100 mm (3.9 in.) and 150
mm (5.9 in.). The two testing machines used were TxDOT’s static creep tester and the SST.
Three replicates for each combination of levels were tested. It should be noted that the
specimens were all molded specimens and that they were all tested at 40 oC (104 oF).
Table 6.1 shows the levels at which the main effects or interactions are significant for
stiffness, permanent deformation, and slope.

Table 6.1: Statistical Analysis for Phase II

p-value
Factor/Interaction
Stiffness Perm. Strain Slope
Mix 0.0032 0.0087 0.0327
Diameter 0.0061 < 0.0001 < 0.0001
Machine 0.0185 0.5819 0.2241
Mix*Diameter 0.0004 0.0067 0.0261
Mix*Machine 0.0322 0.5867 0.4908
Machine*Diameter 0.0021 0.8879 0.6462
Three-Factor Interaction 0.0073 0.3496 0.4354

For stiffness, all main effects and interactions are statistically significant at an α level
of 5 percent. Even at an α level of 1 percent, only machine (main effect) and the interaction
between machine and mix are not statistically significant. Figure 6.1 shows the column chart
and line graphs for stiffness. Figure 6.2 shows the interaction between machine and mix.
The significant interaction between the three factors does not allow for any observable trends
in stiffness. From a practical standpoint, the observed stiffness for all the combinations,
except Mix B and Mix C at 150 mm (5.9 in.) run on the SST, is essentially the same because
the range of stiffness within three replicates is about 10,000 (1,450) to 20,000 kPa (2,900
psi).

33
160000
140000
120000
Stiffness, kPa 100000
80000
60000
40000
20000 C
0 B Mix
100 A
150
ETS 100
ETS 150
SST
SST
Diameter, Machine

Figure 6.1: Phase II Column Chart for Stiffness (ETS denotes TxDOT’s EnduraTEC Systems
[ETS] device)

160000
Mix A, ETS
Stiffness, kPa

140000 Mix B, ETS

120000 Mix C, ETS
Mix A, SST
100000
Mix B, SST
80000
Mix C, SST
10 150
Diameter, mm

Figure 6.2: Phase II Line Graph for Stiffness (ETS denotes TxDOT’s EnduraTEC Systems
[ETS] device)

34
Figures 6.3 and 6.4 show the column chart for permanent strain and slope,
respectively. In both cases, the diameter, mix, and interaction between diameter and mix are
significant at an αOHYHORISHUFHQW7KHPP LQ VSHFLPHQVKDYHKLJKHUSHUPDQHQW
strain and higher slope than the corresponding 150 mm (5.9 in.) specimens. This disparity
can be due to boundary effects having a greater effect on the 150 mm (5.9 in.) specimens
(aspect ratio of 0.33) than on the 100 mm (3.9 in.) specimens (aspect ratio of 0.50). Figures
6.5 and 6.6 show the interaction between diameter and mix for permanent strain and slope,
respectively. An interesting trend was observed in permanent strain and slope: For 100 mm
(3.9 in.) specimens, Mix B had the highest permanent strain, while for 150 mm specimens,
Mix B had the lowest permanent strain. For 100 mm (3.9 in.) specimens, Mix A had the
lowest slope, while for 150 mm (5.9 in.) specimens, Mix B had the lowest slope. These
trends are due to the interaction between mix and diameter. From a practical standpoint, in
half of the combinations, there are practical differences between specimens of different
diameters for both the permanent strain and the diameter.

0.40
0.35
Permanent Strain, 10-3

0.30
0.25
mm/mm

0.20
0.15
0.10
0.05 C
0.00 B Mix
100 A
150
ETS 100
ETS 150
SST
SST

Diameter, Machine

Figure 6.3: Phase II Column Plot for Permanent Strain (ETS denotes TxDOT’s EnduraTEC
Systems [ETS] device)

35
5.0

Slope, 10–8 mm/mm/sec

4.0

3.0

2.0

1.0
C
0.0 B Mix
100 A
150
ETS 100
ETS 150
SST
SST
Diameter, Machine

Figure 6.4: Phase II Column Plot for Slope (ETS denotes TxDOT’s EnduraTEC Systems
[ETS] device)
Permanent Strain, 10 –3

0.40
0.35 Mix A, ETS
0.30 Mix B, ETS
mm/mm

0.25 Mix C, ETS

0.20 Mix A, SST
0.15 Mix B, SST
0.10 Mix C, SST
100 150
Diameter, mm

Figure 6.5: Phase II Interaction between Mix and Diameter for Permanent Strain (ETS
denotes TxDOT’s EnduraTEC Systems [ETS] device)

36
5.0

10 mm/mm/sec
Mix A, ETS
4.0
Mix B, ETS

Slope,
3.0
Mix C, ETS
2.0
Mix A, SST
1.0

–8
Mix B, SST
0.0 Mix C, SST
100 150
Diameter, mm

Figure 6.6: Phase II Interaction between Mix and Diameter for Slope (ETS denotes TxDOT’s
EnduraTEC Systems [ETS] device)

6.2 PHASE III RESULTS: SAMPLE PREPARATION AND TESTING MACHINE

EVALUATION

In Phase III of this research effort, the effect of specimen preparation and testing
machine was evaluated. The two preparation methods are molded and cut, as discussed
earlier in Chapter 1. The two testing machines used were TxDOT’s static creep tester and
the SST. Three replicates for each combination of levels were tested. It should be noted that
all the specimens were 100 mm (3.9 in.) in diameter, and they were all tested at 40 oC (104
o
F).
Table 6.2 shows the level at which the main effects or interactions are significant for
stiffness, permanent deformation, and slope.

Table 6.2: Statistical Analysis for Phase III

p-value
Factor/Interaction
Stiffness Perm. Strain Slope
Mix 0.0007 0.0250 0.0210
Sample Preparation 0.1227 0.6350 0.4305
Machine < 0.0001 0.0200 < 0.0001
Mix*Sample Preparation 0.0196 0.7666 0.5736
Mix*Machine 0.0006 0.1073 0.2669
Machine*Sample Prep. 0.9410 0.5514 0.2344
Three-Factor Interaction 0.2165 0.2668 0.7402

For stiffness, permanent strain, and slope, testing machine and mix (main effects) are
statistically significant at an α level of 5 percent. For stiffness, the interaction between
testing machine and mix is also statistically significant at an α level of 5 percent. Figures
6.7, 6.8, and 6.9 show the column charts for stiffness, permanent strain, and slope,
respectively.

37
160000
140000

Stiffness, kPa
120000
100000
80000
60000
40000
20000 C
0 B Mix
ETS A
SST ETS
mold mold SST
cut
cut
Machine,
Sample Preparation

Figure 6.7: Phase III Column Chart for Stiffness (ETS denotes TxDOT’s EnduraTEC
Systems [ETS] device)

0.35
0.30
Permanent Strain, 10-3

0.25
mm/mm

0.20
0.15
0.10
0.05 C
0.00 B Mix
ETS SST A
mold mold ETS
SST
cut
cut
Machine,
Sample Preparation

Figure 6.8: Phase III Column Chart for Permanent Strain (ETS denotes TxDOT’s
EnduraTEC Systems [ETS] device)

38
4.0
3.5
3.0

m m /m m /se c
S lope , 10 -8
2.5
2.0
1.5
1.0
0.5 C
0.0 B M ix
E TS A
SST E TS
m old m old SST
c ut
c ut
M a chine ,
S a m ple P re pa ra tion

Figure 6.9: Phase III Column Chart for Slope (ETS denotes TxDOT’s EnduraTEC Systems
[ETS] device)

With the SST, Mix B has the greatest stiffness, followed by Mix C, and then Mix A.
With the EnduraTEC Systems (ETS) device, there does not appear to be a clear trend. From
a practical standpoint, there is little difference in stiffness among the mix, machine, or
sample preparation method. The only practical difference is between the ETS and the SST
for Mix B. Mix A and Mix C exhibit this difference, with such difference explained by the
interaction between mix and sample preparation.
For permanent strain, there are no clear trends, and there does not appear to be a
practical significance for any of the main effects.
Slope results with the SST are consistently higher than those with the ETS, and these
differences are of practical significance. There is little practical significance among mixes,
though Mix B consistently exhibits lower slope values for all combinations of machine and
sample preparation.
6.3 PHASE IV RESULTS: ASPHALT CONTENT AND TEMPERATURE
EVALUATION

In Phase IV of this research effort, the effect of asphalt content and temperature was
evaluated. The three asphalt contents used were optimum, optimum plus 0.5 percent, and
optimum plus 1 percent. In addition, Mix A (the primary mix) was also tested at optimum
minus 0.5 percent. The two temperatures used were 40 oC (104 oF) and 64 oC (147 oF).

39
Three replicates for each combination of levels were tested. It should be noted that all the
specimens were molded and that they were all 150 mm (5.9 in.) in diameter.
Table 6.3 shows the level at which the main effects or interactions are significant for
stiffness, permanent deformation, and slope.

Table 6.3: Statistical Analysis for Phase IV

p-value
Factor/Interaction
Stiffness Perm. Strain Slope
Mix 0.9626 0.7665 0.4580
Asphalt Content 0.0222 0.0482 0.0001
Temperature 0.0803 0.7628 < 0.0001
Mix*Asphalt Content 0.5711 0.8800 0.4153
Mix*Temperature 0.8539 0.6970 0.3487
Temperature*Asphalt Content 0.1054 0.0591 0.0001
Three-Factor Interaction 0.6948 0.8982 0.2660

In this Phase IV, asphalt content (main effect) is statistically significant at an α level
of 5 percent for stiffness, permanent strain and slope. Asphalt content and temperature are
also statistically significant at an α level of 0.01 percent for slope. Figures 6.10, 6.11, and
6.12 show the column charts for stiffness, permanent strain, and slope, respectively.

40
110000

100000

Stiffness, kPa
90000

80000

70000

60000

50000 C
-0.5 0.0 B Mix
0.5 1.0 A
40 40 -0.5 0.0
40 40 0.5
64 64 1.0
Diff. from Optimum
64 64
AC Content, Temp.

Figure 6.10: Phase IV Column Chart for Stiffness

0.40
0.35
-3
Permanent Strain, 10

0.30
0.25
mm/mm

0.20
0.15
0.10
0.05
0.00 C
-0.5 0.0 B
0.5 Mix
1.0 A
40 40 -0.5 0.0
40 40 0.5
64 64 1.0
64 64
Diff. from Optimum
AC Content, Temp.

Figure 6.11: Phase IV Column Chart for Permanent Strain

41
3 .0

S lop e, 10 m m /m m /sec
2 .5

2 .0

1 .5
-8

1 .0

0 .5

0 .0 C
-0 .5 B
0 .0 0 .5 Mix
1 .0 A
40 40 -0 .5
40 0 .0 0 .5
40 64 1 .0
64 64
D iff. from Op tim u m 64
AC C o nten t, T em p .

Figure 6.12: Phase IV Column Chart for Slope

There is a clear trend of decreasing stiffness, increasing permanent strain, and

increasing slope with increasing asphalt content at 40 oC (104 oF). This finding is to be
expected. However, the magnitude with which the responses change with asphalt content
varies for Mix A, Mix B, and Mix C. The trends for stiffness, permanent strain, and slope
can be seen more clearly in Figures 6.13, 6.14, and 6.15, respectively. Some of the “kinks”
in the plots for 64 oC (147 oF) do not conform with the expected results, but are not entirely
uncommon in a sensitivity analysis of this nature.
Mixes A, B, and C showed a 15, 12, and 33 percent decrease in stiffness,
respectively, at 40 oC (104 oF). This finding shows that Mix C is more sensitive to changes
in asphalt content. At 64 oC (147 oF), the decreases were -1, 11, and 12 percent for Mixes A,
B, and C, respectively. It should be noted that only the decrease in stiffness for Mix C at 40
o
C (104 oF) is of practical significance.
At 40 oC (104 oF), Mixes A, B, and C showed a 38, 19, and 41 percent increase in
permanent strain, respectively. This shows that Mixes A and C are more sensitive to changes
in asphalt content. At 64 oC (147 oF), decreases in permanent strain of 44, 13, and 16 percent
were observed for Mixes A, B, and C, respectively. It is interesting to observe increases in
permanent strain with a 1 percent increase in asphalt content at 40 oC (104 oF) and decreases
in permanent strain with a 1 percent increase in asphalt content at 64 oC (147 oF). It should
be noted that the changes in permanent strain are not of practical significance.

42
At 40 oC (104 oF), Mixes A, B, and C showed a 30, 41, and 23 percent increase in
slope, respectively. This finding shows that the slopes of all the mixes are fairly sensitive to
changes in asphalt content. At 64 oC (147 oF), decreases in slope of 3, 32, and 13 percent
were observed for Mixes A, B, and C, respectively. It is interesting to observe increases in
slope with a 1 percent increase in asphalt content at 40 oC (104 oF) and decreases in slope
with a 1 percent increase in asphalt content at 64 oC (147 oF). It should be noted that the
changes in slope are not of practical significance.

105000
100000
95000 Mix A, 40C
Stiffness, kPa

90000 Mix B, 40C

85000 Mix C, 40C
80000 Mix A, 64C
75000 Mix B, 64C
70000 Mix C, 64C
65000
60000
-1 -0.5 0 0.5 1 1.5
Difference from Optimum AC Content, %

Figure 6.13: Phase IV Interaction between Mix and Temperature for Stiffness

0.40
0.35
Permanent Strain

0.30 Mix A, 40C

0.25 Mix B, 40C
Mix C, 40C
0.20
Mix A, 64C
0.15
Mix B, 64C
0.10 Mix C, 64C
0.05
0.00
-1 -0.5 0 0.5 1 1.5
Difference from Optimum AC Content, %

Figure 6.14: Phase IV Interaction between Mix and Temperature for Permanent Strain

43
3.0

2.5 Mix A, 40C

Mix B, 40C

Slope
Mix C, 40C
2.0
Mix A, 64C
Mix B, 64C
1.5 Mix C, 64C

1.0
-1 -0.5 0 0.5 1 1.5

Difference from Optimum AC Content, %

Figure 6.15: Phase IV Interaction between Mix and Temperature for Slope

6.4 PHASE V RESULTS: GOOD VERSUS POOR MIX EVALUATION

In Phase V of this research effort, the effect of good and poor mix was evaluated.
Three replicates for each combination of levels were tested. It should be noted that all the
specimens were 150 mm (5.9 in.) in diameter and that they were all tested at 40 oC (104 oF)
and 64 oC (147 oF). Table 6.4 shows the level at which the main effects or interactions are
significant for stiffness, permanent deformation, and slope.

Table 6.4: Statistical Analysis for Phase V

p-value
Factor/Interaction
Stiffness Perm. Strain Slope
Mix 0.1675 0.9644 0.0077
Asphalt Content 0.2009 0.0591 0.0009
Temperature 0.6269 0.0009 0.1292
Mix*Asphalt Content 0.3032 0.5729 0.1646
Mix*Temperature 0.2502 0.9750 0.0185
Temperature*Asphalt Content 0.4012 0.0622 0.0013
Three-Factor Interaction 0.5766 0.4883 0.1388

For Phase V, temperature is significant at an α level of 0.09 percent for permanent

strain, and mix and asphalt content are significant (main effects) at an α level of 1 percent.

44
Figures 6.16, 6.17, and 6.18 show the column charts for stiffness, permanent strain, and
slope, respectively.

100000

95000

90000
S tiffne ss, kP a

85000

80000
75000
70000
65000 40 good
40 poor Te m p,
60000
64 good
A Good/
B C 64 poor
V P oor
W X
M ix

Figure 6.16: Phase V Column Chart for Stiffness

0.40

0.35
-3
Permanent Strain, 10

0.30

0.25
mm/mm

0.20
0.15

0.10
0.05 40 poor
0.00 40 good Temp,
64 poor
A Good/
B C 64 good
V W Poor
X
Mix

Figure 6.17: Phase V Column Chart for Permanent Strain

45
4.0

3.5
Slope, 10 mm/mm/sec 3.0

2.5

2.0
-8

1.5
1.0

0.5 64 poor
0.0 64 good Temp,
40 poor
A Good/
B C 40 good
V Poor
W X
Mix

Figure 6.18: Phase V Column Chart for Slope

At 40 oC (104 oF), Mix A and Mix V (Superpave mixes) have a lower stiffness,
higher permanent strain, and higher slope for the poor mixes compared to the corresponding
good mix. Mixes B and C (intermediate and fine mixes from the sensitivity analysis in Phase
IV) also show the same trends as those for mixes A and V. These results are expected.
However, Mix W and Mix X (CMHB mixes) have a higher stiffness and practically similar
permanent strain and slope for the poor mixes as compared to the corresponding good mix.
This result is unexpected. For the good mixes, the Superpave mixes have higher stiffnesses
than the CMHB-C mixes, though there are no clear trends in the permanent strain and slope.
It should be noted, however, that mix stiffnesses are highly dependent on aggregate and
asphalt type and quality, which differ between mixes in this study. No conclusions can or
should be made regarding a comparison between Superpave and CMHB mixes based on this
study. It should also be noted that the differences between good and poor mixes are not of
practical significance, except for Mix C.
At 64 oC (147 oF), the trends are less obvious and the differences observed are less
significant than those observed at 40 oC (104 oF). This finding is surprising. At higher
temperatures, the asphalt binder is less stiff and is more likely to deform. One possible
explanation for this result is the low stress level used in this test — 69 kPa (10 psi). AAMAS
and other studies have used a stress level of 207 kPa (30 psi) for unconfined uniaxial static
creep tests. It is possible that a higher stress level will increase the difference in the
responses for this static creep test.

46
6.5 PHASE VI RESULTS: DEVELOPMENT OF ACCEPTANCE CRITERIA

As mentioned in Chapter 5, an effort was made to develop acceptance criteria for

Superpave mixes, if possible. However, the results from this research effort have not shown
practical differences in the measured responses. Consequently, clearly defined acceptance
criteria cannot be developed based on the findings presented herein. However, the lessons
learned from this research effort are discussed in Chapter 7, and recommendations are made
for future efforts to develop acceptance criteria.

47
48
CHAPTER 7. CONCLUSIONS AND RECOMMENDATIONS

This research effort sought to answer a number of questions regarding the static creep
behavior of Superpave mixes. Phase I involved modifications to the current TxDOT static
creep tester and to the SST. Conclusions from Phases II through V are listed below and are
followed by a discussion of the overall conclusions and recommendations for future research.
7.1 CONCLUSIONS FROM PHASES II THROUGH V

Conclusions from Phase II

All main effects and interactions were statistically significant for stiffness at an α
level of 5 percent. However, there were no practical differences in stiffness.
Mix and diameter (main effects) and the interaction between mix and diameter for
permanent strain were statistically significant at an α level of 5 percent. The differences
owing to diameter — differences resulting from the interaction between diameter and mix —
were of practical significance in some cases but not in other cases. The differences resulting
from machine type were not of practical significance.
Mix and diameter (main effects) and the interaction between mix and diameter for
slope were statistically significant at an α level of 5 percent. The differences resulting from
diameter were significant in some cases and not significant in other cases. This finding was a
result of the interaction between diameter and mix. The differences resulting from machine
type were of practical significance in most cases.
Conclusions from Phase III
Mix and machine (main effects) were statistically significant for stiffness, permanent
strain, and slope at an α level of 5percent. However, there were no practical differences for
either of the two statistically significant main effects for both stiffness and permanent strain.
For slope, the SST gave slope values that were consistently higher and of practical
significance. There was no practical difference between slope values between sample
preparation methods.
The interaction between mix and sample preparation and mix and machine was also
statistically significant at an α level of 5 percent.
Conclusions from Phase IV
Asphalt content was statistically significant for stiffness, permanent strain, and slope
at an α level of 5 percent. At 40 oC (104 oF), there was a clear trend of decreasing stiffness,
increasing permanent strain, and increasing slope with increasing asphalt content. However,
only one mix showed differences of practical significance. At 64 oC (147 oF) there was a
decrease in stiffness, permanent strain, and slope from optimum to optimum plus 1 percent
asphalt. However, several “kinks” in the data were observed at the optimum plus 0.5 percent
asphalt content.

49
Temperature and the interaction between asphalt content and temperature were also
significant at an α level of 5 percent for slope.
Conclusions from Phase V
Temperature was statistically significant for permanent strain, and asphalt content
was statistically significant for slope at an α level of 5 percent. At 40 oC (104 oF), the good
Superpave mixes and the two other base mixes have higher stiffnesses, lower permanent
strains, and lower slopes than the corresponding poor mixes. The two coarse matrix high
binder (CMHB) mixes exhibited the opposite behavior. Overall, these differences were not
of practical significance. At 64 oC (147 oF), the good mixes exhibited similar (from a
practical standpoint) if not higher stiffnesses, lower permanent strains, and lower slopes than
the poor mixes.
7.2 CONCLUSIONS AND RECOMMENDATIONS

In the study of diameters and testing machines, only the effect of machine on slope
was of practical significance in almost all cases. Based on the limited scope of this study, it
is concluded that the SGC specimens molded to 51 mm (2 in.) in height exhibit creep
properties similar to those of a Texas gyratory compactor specimen compacted to the same
height. However, there were practical differences in slope values between specimens tested
on the ETS and on the SST.
In the study on sample preparation and testing machine, only the effect of machine on
slope was consistently of practical significance. Based on the limited scope of this study, it
could be concluded that sample preparation methods (molded or cut) do not affect creep
properties of SGC-compacted mixes from a practical standpoint. However, there were
practical differences in slope values between specimens tested on the ETS and on the SST.
In the study on asphalt content and temperature, there was a consistent decrease in
stiffness, increase in permanent strain, and increase in slope with increasing asphalt content
for tests run at 40 oC (104 oF). However, most of the differences were of no practical
significance. The differences observed at 64 oC (147 oF) were even less significant and
much more inconsistent in terms of whether higher values were observed at 40 oC (104 oF) or
64 oC (147 oF).
In the study of good versus poor mixes, it was observed that poor Superpave mixes
exhibited lower stiffnesses, higher permanent strains, and higher slopes as compared with the
corresponding good mixes. The intermediate and fine mixes from the sensitivity analysis in
Phase IV also show the same trends. However, the poor CMHB mixes exhibited higher
stiffnesses and practically similar permanent strains and slopes for the poor mixes as
compared to those in the corresponding good mixes. This finding could be interpreted to
mean that the asphalt content used in the field may not be the optimum for resisting
permanent deformation. However, in actual construction, many other factors play into the
choice of asphalt content.

50
7.3 FUTURE RESEARCH

As suggested in the conclusions and recommendations, agencies are cautioned against

blindly accepting the conclusions of this research effort. Adequate care must be taken to
identify the context of the recommendations and adapt them to each unique set of
circumstances. In order to extend the applicability of these conclusions, additional research
is required to verify the findings for a broader range of materials (asphalt and aggregates).
Only then can the findings be applied in a broader sense.
One underlying comment throughout the discussion of results and conclusions is the
absence of practical differences in the results, even in the presence of statistical differences.
A concern in the current test setup is the fairly large variability mentioned in Chapter 6.
With a rather large variation between the three replicates, it is hard to say that one mean is
very different from another. One possible solution would be to increase the stress level at
which the test is run. Some other methods, including the Asphalt Aggregate Mixture
Analysis System (AAMAS), have used a stress level of 207 kPa (30 psi), and this stress level
should be investigated in the future. Such a stress level will make the test more of a “torture
test” and possibly help to better differentiate between mixes.
Another factor of interest is the aspect ratio. Owing to limitations in the testing
equipment, the height of the test specimen was maintained at approximately 51 mm (2 in.),
even though the diameter was increased from 100 mm (3.9 in.) to 150 mm (5.9 in.). This
height reduced the aspect ratio of the test specimen from 0.5 to 0.33. Further research can be
conducted where the aspect ratio is kept constant while the diameter of the specimen is
increased. For a 150 mm (5.9 in.) diameter specimen, a specimen height of 75 mm (3.0 in.)
is recommended to maintain a constant aspect ratio.
Owing to the lack of practical differences between the mixes, no attempt was made to
develop acceptance criteria. However, it is hoped that the work performed in this project will
contribute to the growing knowledge regarding the use of the static creep test as a means of
evaluating the rut resistance of asphalt concrete mixes.

51
52
BIBLIOGRAPHY

American Association of State Highway and Transportation Officials, AASHTO Provisional

Specifications, American Association of State Highway and Transportation Officials,
June 1998.

Brown, R., Dukatz, E., Huber, G., Michael, L., Scherocman, J., and Sines, S., Performance of
Coarse-Graded Mixes at Westrack — Premature Rutting, Final Report, U.S.
Department of Transportation, Federal Highway Administration, June 1998.

Cominsky, R. J., Huber, G. A., Kennedy, T. W., and Anderson, M. A., The Superpave Mix
Design Manual for New Construction and Overlays, Strategic Highway Research
Program, National Research Council, Washington, D.C., 1994.

Dunn, R., Survey Finds Superpave Use Expected to Double This Year, FOCUS, U.S.
Department of Transportation, Federal Highway Administration, September 1999.

Finnie, I., and Heller, W., Creep of Engineering Materials, McGraw-Hill, New York, 1959.

Hill, J. F., The Creep of Asphalt Mixes, Journal of the Institute of Petroleum, Vol. 59, 1973.

Kennedy, T. W., Huber, G. A., Harrigan, E. T., Cominsky, R. J., Hughes, C. S., Von Quintus,
H., and Moulthrop, J. S., Superior Performing Asphalt Pavements (Superpave): The
Product of the SHRP Asphalt Research Program, Strategic Highway Research
Program, National Research Council, Washington, D.C., 1994.

Khedr, S. A., “Deformation Mechanism in Asphalt Concrete,” Journal of Transportation,

ASCE, Vol. 112, 1986.

Kinder, D. F., A Study of Both the Viscoelastic and Permanent Deformation Properties of a
New South Wales Asphalt, 13th Australian Road Research Board, New South Wales,
Australia, 1986.

Kronfuss, R., Krzemian, R., Nievelt, G., and Putz, P., Verformungsfestigkjeit von Asphalten
Ermittlung in Kriechtest, Bundesministerium fur Bauten and Technik,
Strassenforschung, Heft 240, Wien, Austria, 1984 (in German).

Lai, J. S., and Anderson, D. A., “Irrecoverable and Recoverable Non-Linear Viscoelastic
Properties of Asphalt Concrete,” Transportation Research Record 468, 1973.

Little, D. N., Button, J. W., and Youssef, H., Development of Criteria to Evaluate Uniaxial
Creep Data and Asphalt Concrete Permanent Deformation Potential, TRR 1417,
Transportation Research Board, National Research Council, Washington, D.C., 1991.

53
Little, D. N., and Youssef, H., “Improved ACP Mixture Design: Development and
Verification,” Research Report 1170-1F, Texas Transportation Institute, Texas A&M
University System, March 1992.

Mahboub, K., and Little, D. N., “Improved Asphalt Concrete Mixture Design,” Research
Report 2474-1F, Texas Transportation Institute, Texas A&M University System,
1988.

McGennis, R. B., Anderson, R. M., Kennedy, T. W., and Solaimanian, M., Background of
Superpave Asphalt Mixture Design and Analysis, U.S. Department of Transportation,
Federal Highway Administration, Washington, D.C., February 1995.

Mitchell, J. K., Fundamentals of Soil Behavior, John Wiley and Sons, Inc., 1976.

National Highway Institute, NHI Course No. 13152: Superpave for the Generalist Engineer
and Project Staff, Participant Manual, FHWA HI 97-031, U.S. Department of
Transportation, Federal Highway Administration, July 1997.

Roberts, F. L., Kandhal, P. S., Brown, E. R., Lee, D. H., and Kennedy, T. W., Hot Mix
Asphalt Materials, Mixture Design, and Construction, NAPA Research and
Education Foundation, Lanham, Maryland, 1994.

Sousa, J. B., Harvey, J., Painter, L., Deacon, J. A., and Monismith, C. L., Evaluation of
Laboratory Procedures for Compacting Asphalt Aggregate Mixtures, Institute of
Transportation Studies, University of California, 1991.

Tahmoressi, M., “New Hot Mix Design Procedure for Coarse Matrix High Binder Mixtures,”
Technical Quarterly, Texas Department of Transportation, July 1993.

Texas Department of Transportation, Creep Test Analysis, 5-lb Recovery Hold, Materials and
Tests Division, Texas Department of Transportation, June 1995.

Texas Department of Transportation, Creep Test Analysis, 5-lb Recovery Hold Follow Up,
Materials and Tests Division, Texas Department of Transportation, June 1995.

Texas Department of Transportation, Method of Compacting Test Specimens of Bituminous

Mixtures, Materials and Tests Division, Texas Department of Transportation, March
1991.

Texas Department of Transportation, Special Specification Item 3116, Quality

Control/Quality Assurance of Hot Mix Asphalt, Materials and Tests Division, Texas
Department of Transportation, 1993.

54
Texas Department of Transportation, Static Creep Test, Test Method Tex-231-F, Manual of
Testing Procedures, Volume 1, Texas Department of Transportation, September
1995.

Van der Loo, P. J., “Creep Testing: A Simple Tool to Judge Asphalt Mix Stability,”
Proceedings of the Association of Asphalt Paving Technologists, Vol. 43, 1974.

Van der Loo, P. J., “The Creep Test: A Key Tool in Asphalt Mix Evaluation and Prediction
of Rutting,” Proceedings of the Association of Asphalt Paving Technologists, Vol. 47,
1978.

Viljoen, A. W., and Meadows, K., The Creep Test — A Mix Design Tool to Rank Asphalt
Mixes in Terms of Their Resistance to Permanent Deformation Under Heavy Traffic,
Tech Note: TP/36/81, National Institute of Road Research, Pretoria, South Africa,
1981.

Von Quintus, H. L., Scherocman, J. A., Hughes, C. S., and Kennedy, T. W., NCHRP Report
338: Asphalt Aggregate Mixture Analysis System (AAMAS), Transportation Research
Board, National Research Council, Washington, D.C., 1991.

Yandell, W. O., The Prediction of the Behavior of Elastoplastic Roads During Repeated
Rolling Using the Mechano-Lattice Analog, HRR 374, 1971.

55
56
Appendix A

Summary of Volumetric Properties and Static Creep Test

Results

57
Sensitivity Analysis
4 inch samples of SC (L) Blend
ETS

Common Volumetric Properties C1 C2 C3 C4 C5 Ave. Std. Dev. COV, %

Percent Asphalt, Pa 4.8 4.8 4.8 4.8 4.8
Specific Gravity of Binder,Gb 1.031 1.031 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.534 2.534 2.534 2.534 2.534
Stone Effective Specific Gravity, Gse 2.634 2.634 2.634 2.634 2.634
Voids in Mineral Aggregate, VMA 11.5 11.3 12.1 11.7 11.9 11.7 0.32 2.7
Voids Filled with Asphalt, VFA 58.4 57.5 60.2 59.0 59.5 58.9 1.03 1.8
Percent Absorbed Asphalt, Pba 1.5 1.5 1.5 1.5 1.5
Percent Effective Asphalt, Pbe 3.3 3.3 3.3 3.3 3.3
Percent Passing 0.075 mm sieve 6.9 6.9 6.9 6.9 6.9
Dust Proportion 2.07 2.07 2.07 2.07 2.07

Test Data
C1 C2 C3 C4 C5 Ave. Std. Dev. COV, %
Mix Bulk Specific Gravity, Gmb 2.355 2.361 2.341 2.350 2.346 2.351 0.008 0.3
Ave. Max. Theo. Density, Gmm 2.451 2.451 2.451 2.451 2.451 2.451 N/A N/A
Percent Density 96.1 96.3 95.5 95.9 95.7 95.9 0.32 0.3
Percent Air 3.9 3.7 4.5 4.1 4.3 4.1 0.32 7.7
Ave. Specimen Ht., mm 51.143 50.595 51.505 50.821 51.455 51.104 0.395 0.8
Ave. Specimen Ht., in. 2.014 1.992 2.028 2.001 2.026 2.012 0.016 0.8
Total Strain, (mm/mm)x1000 0.680 0.629 0.721 0.688 0.743 0.692 0.044 6.3
*Creep Stiffness, kPa 101279 108331 95699 100474 92048 99566 6166 6.2
*Permanent Strain, (mm/mm)x1000 0.34 0.29 0.31 0.27 0.39 0.32 0.05 14.7
Recovery Efficiency, % 50.5 53.9 57.6 60.2 48.0 54.0 4.99 9.2
*Slope of SS curve,(mm/mm/sec)xE8 3.7 3.5 3.6 2.7 3.9 3.5 0.46 13.2
Ave. Strain Difference, (mm/mm)xE6 133 134 135 95 264 152 65 42.5
Max. Strain Difference, (mm/mm)xE6 208 192 208 167 366 228 79 34.5
Sensitivity Analysis
4 inch samples of SC (L) Blend
MTS

Common Volumetric Properties C6 C7 C8 C9 C10 Ave. Std. Dev. COV, %

Percent Asphalt, Pa 4.8 4.8 4.8 4.8 4.8
Specific Gravity of Binder,Gb 1.031 1.031 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.534 2.534 2.534 2.534 2.534
Stone Effective Specific Gravity, Gse 2.634 2.634 2.634 2.634 2.634
Voids in Mineral Aggregate, VMA 12.1 12.0 11.3 11.8 11.6 11.8 0.32 2.7
Voids Filled with Asphalt, VFA 60.2 59.9 57.5 59.4 58.5 59.1 1.10 1.9
Percent Absorbed Asphalt, Pba 1.5 1.5 1.5 1.5 1.5
Percent Effective Asphalt, Pbe 3.3 3.3 3.3 3.3 3.3
Percent Passing 0.075 mm sieve 6.9 6.9 6.9 6.9 6.9
Dust Proportion 2.07 2.07 2.07 2.07 2.07

Test Data
C6 C7 C8 C9 C10 Ave. Std. Dev. COV, %
Mix Bulk Specific Gravity, Gmb 2.341 2.343 2.361 2.347 2.354 2.349 0.008 0.4
Ave. Max. Theo. Density, Gmm 2.451 2.451 2.451 2.451 2.451 2.451 N/A N/A
Percent Density 95.5 95.6 96.3 95.8 96.0 95.8 0.34 0.4
Percent Air 4.5 4.4 3.7 4.2 4.0 4.2 0.34 8.1
Ave. Specimen Ht., mm 51.723 51.036 51.728 50.709 51.064 51.252 0.454 0.9
Ave. Specimen Ht., in. 2.036 2.009 2.037 1.996 2.010 2.018 0.018 0.9
Total Strain, (mm/mm)x1000 0.854 0.919 0.956 1.207 0.932 0.974 0.136 14.0
*Creep Stiffness, kPa 78539 73459 70223 56035 72624 70176 8466 12.1
*Permanent Strain, (mm/mm)x1000 0.32 0.35 0.42 0.49 0.41 0.40 0.07 16.6
Recovery Efficiency, % 63.1 61.4 56.3 59.4 56.5 59.3 2.99 5.0
*Slope of SS curve,(mm/mm/sec)xE8 3.1 3.1 3.1 3.0 3.7 3.2 0.28 8.8
Ave. Strain Difference, (mm/mm)xE6 255 379 172 571 254 326 156 47.7
Max. Strain Difference, (mm/mm)xE6 383 540 301 816 298 468 218 46.6
Sensitivity Analysis
4 inch samples of SC (L) Blend
SST

Common Volumetric Properties C11 C12 C13 C14 C15 Ave. Std. Dev. COV, %
Percent Asphalt, Pa 4.8 4.8 4.8 4.8 4.8
Specific Gravity of Binder,Gb 1.031 1.031 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.534 2.534 2.534 2.534 2.534
Stone Effective Specific Gravity, Gse 2.634 2.634 2.634 2.634 2.634
Voids in Mineral Aggregate, VMA 11.4 12.0 12.0 11.7 11.7 11.8 0.25 2.1
Voids Filled with Asphalt, VFA 58.1 60.0 59.9 59.0 59.0 59.2 0.78 1.3
Percent Absorbed Asphalt, Pba 1.5 1.5 1.5 1.5 1.5
Percent Effective Asphalt, Pbe 3.3 3.3 3.3 3.3 3.3
Percent Passing 0.075 mm sieve 6.9 6.9 6.9 6.9 6.9
Dust Proportion 2.07 2.07 2.07 2.07 2.07

Test Data
C11 C12 C13 C14 C15 Ave. Std. Dev. COV, %
Mix Bulk Specific Gravity, Gmb 2.357 2.342 2.343 2.350 2.350 2.348 0.006 0.3
Ave. Max. Theo. Density, Gmm 2.451 2.451 2.451 2.451 2.451 2.451 N/A N/A
Percent Density 96.2 95.6 95.6 95.9 95.9 95.8 0.25 0.3
Percent Air 3.8 4.4 4.4 4.1 4.1 4.2 0.25 6.0
Ave. Specimen Ht., mm 51.239 51.398 51.317 51.055 51.493 51.300 0.166 0.3
Ave. Specimen Ht., in. 2.017 2.024 2.020 2.010 2.027 2.020 0.007 0.3
Total Strain, (mm/mm)x1000 0.715 0.677 0.749 0.699 0.710 0.030 4.3
*Creep Stiffness, kPa 98094 101427 94805 98134 98115 2703 2.8
*Permanent Strain, (mm/mm)x1000 0.34 0.37 0.40 0.34 0.36 0.03 7.9
Recovery Efficiency, % 52.4 45.3 46.6 51.4 48.9 3.49 7.1
*Slope of SS curve,(mm/mm/sec)xE8 5.3 4.2 4.9 4.7 4.8 0.46 9.6
Ave. Strain Difference, (mm/mm)xE6 6 14 18 14 13 5 38.7
Max. Strain Difference, (mm/mm)xE6 48 48 71 95 66 22 34.3

*No data could be read from the output file for C14.
Sensitivity Analysis
6 inch samples of SA (J) Blend
ETS (saw-cut)

Common Volumetric Properties A1 A2 A3 Ave. Std. Dev. COV, %

Percent Asphalt, Pa 5.1 5.1 5.1
Specific Gravity of Binder,Gb 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.545 2.545 2.545
Stone Effective Specific Gravity, Gse 2.605 2.605 2.605
Voids in Mineral Aggregate, VMA 13.0 13.5 13.8 13.4 0.40 3.0
Voids Filled with Asphalt, VFA 60.7 62.1 62.9 61.9 1.11 1.8
Percent Absorbed Asphalt, Pba 0.9 0.9 0.9
Percent Effective Asphalt, Pbe 4.2 4.2 4.2
Percent Passing 0.075 mm sieve 5.3 5.3 5.3
Dust Proportion 1.26 1.26 1.26

Test Data
A1 A2 A3 Ave. Std. Dev. COV, %
Mix Bulk Specific Gravity, Gmb 2.334 2.321 2.313 2.323 0.011 0.5
Ave. Max. Theo. Density, Gmm 2.417 2.417 2.417 2.417 N/A N/A
Percent Density 96.6 96.0 95.7 96.1 0.44 0.5
Percent Air 3.4 4.0 4.3 3.9 0.44 11.2
Ave. Specimen Ht., mm 52.365 50.826 49.775 50.989 1.303 2.6
Ave. Specimen Ht., in. 2.062 2.001 1.960 2.007 0.051 2.6
Total Strain, (mm/mm)x1000 0.592 0.559 0.598 0.583 0.021 3.6
*Creep Stiffness, kPa 115356 122415 114706 117492 4276 3.6
*Permanent Strain, (mm/mm)x1000 0.24 0.21 0.18 0.21 0.03 14.3
Recovery Efficiency, % 60.2 62.4 70.7 64.4 5.54 8.6
*Slope of SS curve,(mm/mm/sec)xE8 2.4 2.3 1.6 2.1 0.44 20.8
Ave. Strain Difference, (mm/mm)xE6 192 45 176 138 81 58.6
Max. Strain Difference, (mm/mm)xE6 249 84 229 187 90 48.1

Sensitivity Analysis
6 inch samples of SA (J) Blend
Sensitivity Analysis
6 inch samples of SA (J) Blend
SST (saw-cut)

Common Volumetric Properties A4 A5 A6 Ave. Std. Dev. COV, %

Percent Asphalt, Pa 5.1 5.1 5.1
Specific Gravity of Binder,Gb 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.545 2.545 2.545
Stone Effective Specific Gravity, Gse 2.605 2.605 2.605
Voids in Mineral Aggregate, VMA 13.3 13.7 13.3 13.4 0.23 1.7
Voids Filled with Asphalt, VFA 61.8 62.8 61.7 62.1 0.61 1.0
Percent Absorbed Asphalt, Pba 0.9 0.9 0.9
Percent Effective Asphalt, Pbe 4.2 4.2 4.2
Percent Passing 0.075 mm sieve 5.3 5.3 5.3
Dust Proportion 1.26 1.26 1.26

Test Data
A4 A5 A6 Ave. Std. Dev. COV, %
Mix Bulk Specific Gravity, Gmb 2.324 2.314 2.325 2.321 0.006 0.3
Ave. Max. Theo. Density, Gmm 2.417 2.417 2.417 2.417 N/A N/A
Percent Density 96.2 95.7 96.2 96.0 0.25 0.3
Percent Air 3.8 4.3 3.8 4.0 0.25 6.3
Ave. Specimen Ht., mm 51.682 49.217 50.570 50.490 1.234 2.4
Ave. Specimen Ht., in. 2.035 1.938 1.991 1.988 0.049 2.4
Total Strain, (mm/mm)x1000 0.413 0.670 0.543 0.542 0.129 23.7
*Creep Stiffness, kPa 167980 103283 126056 114670 16103 14.0
*Permanent Strain, (mm/mm)x1000 0.10 0.29 0.29 0.29 0.00 0.0
Recovery Efficiency, % 75.8 56.7 46.6 51.7 7.2 13.9
*Slope of SS curve,(mm/mm/sec)xE8 1.8 3.2 3.8 3.5 0.4 12.1
Ave. Strain Difference, (mm/mm)xE6 59 47 20 34 19 57.0
Max. Strain Difference, (mm/mm)xE6 118 99 72 86 19 22.3

*Specimen A4 was not included in the statistical analysis because the results are for a re-test. The original test
yielded no data in the datafile.

Sensitivity Analysis
6 inch samples of SA (J) Blend
ETS (molded to 50.8 mm)

Common Volumetric Properties S3 S4 S6 Ave. Std. Dev. COV, %

Percent Asphalt, Pa 5.1 5.1 5.1
Specific Gravity of Binder,Gb 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.545 2.545 2.545
Stone Effective Specific Gravity, Gse 2.605 2.605 2.605
Voids in Mineral Aggregate, VMA 13.2 13.0 13.1 13.1 0.10 0.8
Voids Filled with Asphalt, VFA 61.4 60.9 61.0 61.1 0.26 0.4
Percent Absorbed Asphalt, Pba 0.9 0.9 0.9
Percent Effective Asphalt, Pbe 4.2 4.2 4.2
Percent Passing 0.075 mm sieve 5.3 5.3 5.3
Dust Proportion 1.26 1.26 1.26

Test Data
S3 S4 S6 Ave. Std. Dev. COV, %
Mix Bulk Specific Gravity, Gmb 2.327 2.332 2.331 2.330 0.003 0.1
Ave. Max. Theo. Density, Gmm 2.417 2.417 2.417 2.417 N/A N/A
Percent Density 96.3 96.5 96.4 96.4 0.11 0.1
Percent Air 3.7 3.5 3.6 3.6 0.11 3.0
Ave. Specimen Ht., mm 50.414 50.547 50.703 50.555 0.145 0.3
Ave. Specimen Ht., in. 1.985 1.990 1.996 1.990 0.006 0.3
Total Strain, (mm/mm)x1000 0.748 0.756 0.729 0.744 0.014 1.9
*Creep Stiffness, kPa 91716 90642 93963 92107 1695 1.8
*Permanent Strain, (mm/mm)x1000 0.20 0.23 0.21 0.21 0.02 7.2
Recovery Efficiency, % 72.9 69.8 71.7 71.5 1.56 2.2
*Slope of SS curve,(mm/mm/sec)xE8 1.2 3.0 1.2 1.8 1.04 57.7
Ave. Strain Difference, (mm/mm)xE6 280 251 117 216 87 40.3
Max. Strain Difference, (mm/mm)xE6 401 364 191 319 112 35.2
SST (molded to 50.8 mm)

Common Volumetric Properties S1 S2 S5 Ave. Std. Dev. COV, %

Percent Asphalt, Pa 5.1 5.1 5.1
Specific Gravity of Binder,Gb 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.545 2.545 2.545
Stone Effective Specific Gravity, Gse 2.605 2.605 2.605
Voids in Mineral Aggregate, VMA 13.3 13.0 13.0 13.1 0.17 1.3
Voids Filled with Asphalt, VFA 61.7 60.8 60.7 61.1 0.55 0.9
Percent Absorbed Asphalt, Pba 0.9 0.9 0.9
Percent Effective Asphalt, Pbe 4.2 4.2 4.2
Percent Passing 0.075 mm sieve 5.3 5.3 5.3
Dust Proportion 1.26 1.26 1.26

Test Data
S1 S2 S5 Ave. Std. Dev. COV, %
Mix Bulk Specific Gravity, Gmb 2.325 2.333 2.334 2.331 0.005 0.2
Ave. Max. Theo. Density, Gmm 2.417 2.417 2.417 2.417 N/A N/A
Percent Density 96.2 96.5 96.6 96.4 0.20 0.2
Percent Air 3.8 3.5 3.4 3.6 0.20 5.7
Ave. Specimen Ht., mm 51.058 50.578 51.142 50.926 0.304 0.6
Ave. Specimen Ht., in. 2.010 1.991 2.013 2.005 0.012 0.6
Total Strain, (mm/mm)x1000 0.753 0.785 0.716 0.751 0.035 4.6
*Creep Stiffness, kPa 91657 88837 96387 92294 3815 4.1
*Permanent Strain, (mm/mm)x1000 0.30 0.36 0.25 0.30 0.06 18.2
Recovery Efficiency, % 60.2 54.1 65.1 59.8 5.48 9.2
*Slope of SS curve,(mm/mm/sec)xE8 3.2 4.1 3.6 3.6 0.45 12.4
Ave. Strain Difference, (mm/mm)xE6 63 10 47 40 27 68.0
Max. Strain Difference, (mm/mm)xE6 120 48 96 88 37 41.7
Sensitivity Analysis
6 inch samples of SM Blend (SA + 0.5% AC)
ETS (molded to 50.8 mm)

Common Volumetric Properties M5 M7 M10 Ave. Std. Dev.COV, %

Percent Asphalt, Pa 5.6 5.6 5.6
Specific Gravity of Binder,Gb 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.545 2.545 2.545
Stone Effective Specific Gravity, Gse 2.596 2.596 2.596
Voids in Mineral Aggregate, VMA 14.5 14.7 14.7 14.6 0.12 0.8
Voids Filled with Asphalt, VFA 61.5 61.4 61.9 61.6 0.26 0.4
Percent Absorbed Asphalt, Pba 0.8 0.8 0.8
Percent Effective Asphalt, Pbe 4.8 4.8 4.8
Percent Passing 0.075 mm sieve 5.3 5.3 5.3
Dust Proportion 1.09 1.09 1.09

Test Data
M5 M7 M10 Ave. Std. Dev.COV, %
Mix Bulk Specific Gravity, Gmb 2.304 2.305 2.300 2.303 0.003 0.1
Ave. Max. Theo. Density, Gmm 2.393 2.393 2.393 2.393 N/A N/A
Percent Density 96.3 96.3 96.1 96.2 0.11 0.1
Percent Air 3.7 3.7 3.9 3.8 0.11 2.9
Ave. Specimen Ht., mm 51.065 50.762 50.589 50.805 0.241 0.5
Ave. Specimen Ht., in. 2.010 1.999 1.992 2.000 0.009 0.5
Total Strain, (mm/mm)x1000 0.659 0.785 0.893 0.779 0.117 15.0
*Creep Stiffness, kPa 103983 87226 76619 89276 13797 15.5
*Permanent Strain, (mm/mm)x1000 0.14 0.26 0.27 0.22 0.07 32.4
Recovery Efficiency, % 78.7 67.4 70.2 72.1 5.88 8.2
*Slope of SS curve,(mm/mm/sec)xE8 1.9 2.0 1.8 1.9 0.10 5.3
Sensitivity Analysis
6 inch samples of SN Blend (SA - 0.5% AC)
ETS (molded to 50.8 mm)

Common Volumetric Properties N5 N6 N9 Ave. Std. Dev.COV, %

Percent Asphalt, Pa 4.6 4.6 4.6
Specific Gravity of Binder,Gb 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.545 2.545 2.545
Stone Effective Specific Gravity, Gse 2.601 2.601 2.601
Voids in Mineral Aggregate, VMA 12.5 12.7 12.9 12.7 0.20 1.6
Voids Filled with Asphalt, VFA 63.1 63.7 64.3 63.7 0.60 0.9
Percent Absorbed Asphalt, Pba 0.9 0.9 0.9
Percent Effective Asphalt, Pbe 3.8 3.8 3.8
Percent Passing 0.075 mm sieve 5.3 5.3 5.3
Dust Proportion 1.41 1.41 1.41

Test Data
N5 N6 N9 Ave. Std. Dev.COV, %
Mix Bulk Specific Gravity, Gmb 2.335 2.330 2.324 2.330 0.006 0.2
Ave. Max. Theo. Density, Gmm 2.431 2.431 2.431 2.431 N/A N/A
Percent Density 96.1 95.8 95.6 95.8 0.23 0.2
Percent Air 3.9 4.2 4.4 4.2 0.23 5.4
Ave. Specimen Ht., mm 50.831 51.215 50.844 50.963 0.218 0.4
Ave. Specimen Ht., in. 2.001 2.016 2.002 2.006 0.009 0.4
Total Strain, (mm/mm)x1000 0.577 0.690 0.772 0.680 0.098 14.4
*Creep Stiffness, kPa 118181 99099 88638 101973 14980 14.7
*Permanent Strain, (mm/mm)x1000 0.15 0.17 0.19 0.17 0.02 11.8
Recovery Efficiency, % 74.8 75.0 75.9 75.2 0.59 0.8
*Slope of SS curve,(mm/mm/sec)xE8 1.3 1.6 1.3 1.4 0.17 12.4
Sensitivity Analysis
6 inch samples of SP Blend (SB + 0.5% AC)
ETS (molded to 50.8 mm)

Common Volumetric Properties P5 P6 P10 Ave. Std. Dev.COV, %

Percent Asphalt, Pa 5.3 5.3 5.3
Specific Gravity of Binder,Gb 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.542 2.542 2.542
Stone Effective Specific Gravity, Gse 2.609 2.609 2.609
Voids in Mineral Aggregate, VMA 13.2 13.3 13.3 13.3 0.06 0.4
Voids Filled with Asphalt, VFA 59.7 60.1 60.3 60.0 0.31 0.5
Percent Absorbed Asphalt, Pba 1.0 1.0 1.0
Percent Effective Asphalt, Pbe 4.3 4.3 4.3
Percent Passing 0.075 mm sieve 5.7 5.7 5.7
Dust Proportion 1.32 1.32 1.32

Test Data
P5 P6 P10 Ave. Std. Dev.COV, %
Mix Bulk Specific Gravity, Gmb 2.331 2.328 2.326 2.328 0.003 0.1
Ave. Max. Theo. Density, Gmm 2.413 2.413 2.413 2.413 N/A N/A
Percent Density 96.6 96.5 96.4 96.5 0.10 0.1
Percent Air 3.4 3.5 3.6 3.5 0.10 3.0
Ave. Specimen Ht., mm 50.841 50.621 50.723 50.728 0.110 0.2
Ave. Specimen Ht., in. 2.002 1.993 1.997 1.997 0.004 0.2
Total Strain, (mm/mm)x1000 0.773 0.738 0.820 0.777 0.041 5.3
*Creep Stiffness, kPa 88387 92455 83504 88115 4482 5.1
*Permanent Strain, (mm/mm)x1000 0.28 0.22 0.28 0.26 0.03 13.3
Recovery Efficiency, % 64.4 70.0 66.3 66.9 2.85 4.3
*Slope of SS curve,(mm/mm/sec)xE8 2.3 1.9 2.2 2.1 0.21 9.8
Sensitivity Analysis
6 inch samples of SR Blend (SC + 0.5% AC)
ETS (molded to 50.8 mm)

Common Volumetric Properties R5 R7 R10 Ave. Std. Dev.COV, %

Percent Asphalt, Pa 5.3 5.3 5.3
Specific Gravity of Binder,Gb 1.031 1.031 1.031
Stone Bulk Specific Gravity, Gsb 2.534 2.534 2.534
Stone Effective Specific Gravity, Gse 2.631 2.631 2.631
Voids in Mineral Aggregate, VMA 12.7 12.3 12.7 12.6 0.23 1.8
Voids Filled with Asphalt, VFA 58.3 56.7 58.3 57.8 0.92 1.6
Percent Absorbed Asphalt, Pba 1.5 1.5 1.5
Percent Effective Asphalt, Pbe 3.9 3.9 3.9
Percent Passing 0.075 mm sieve 6.9 6.9 6.9
Dust Proportion 1.78 1.78 1.78

Test Data
R5 R7 R10 Ave. Std. Dev.COV, %
Mix Bulk Specific Gravity, Gmb 2.336 2.348 2.336 2.340 0.007 0.3
Ave. Max. Theo. Density, Gmm 2.431 2.431 2.431 2.431 N/A N/A
Percent Density 96.1 96.6 96.1 96.3 0.28 0.3
Percent Air 3.9 3.4 3.9 3.7 0.28 7.6
Ave. Specimen Ht., mm 50.529 50.718 50.626 50.624 0.095 0.2
Ave. Specimen Ht., in. 1.989 1.997 1.993 1.993 0.004 0.2
Total Strain, (mm/mm)x1000 0.746 0.805 0.693 0.748 0.056 7.5
*Creep Stiffness, kPa 91726 85130 98873 91910 6873 7.5
*Permanent Strain, (mm/mm)x1000 0.20 0.24 0.19 0.21 0.03 12.6
Recovery Efficiency, % 72.7 70.7 72.6 72.0 1.13 1.6
*Slope of SS curve,(mm/mm/sec)xE8 2.1 2.1 1.8 2.0 0.17 8.7