AAPT
Abstracts for the March 7-9, 2005 Annual Meeting

Government Engineers Forum- Sunday 1:00 pm, Frank Fee Presiding

Workshop Session-Monday 10:15 am, Larry Michael Presiding

FIELD VALIDATION OF MACROTEXTURE-BASED HOT-MIX ASPHALT SEGREGATION DETECTION METHODS

by Gerardo W. Flintsch, Kevin K. McGhee, and Edgar de Leon

Abstract

This paper discusses the efforts carried out in Virginia to validate the use of macrotexture measurements for identifying and measuring hot-mix asphalt (HMA) segregation as detected by more conventional laboratory procedures. The foundation of this project was a series of field experiments designed to sample the uniformity of typical VDOT paving mixtures and to test the validity of available equations to predict target macrotexture values. Eight HMA mixes with variable maximum aggregate size and volumetric properties were considered. At each location, the research team conducted dynamic macrotexture measurements and selected the areas to conduct static tests based on those measurements. The static tests included the following: static macrotexture using a Circular Track Meter, density using nuclear and electromagnetic devices, and core extraction and testing. The laboratory tests performed on the cores included: core height (mm), Specific Gravity, void content, laboratory permeability, asphalt content, and gradation. The researchers demonstrated that macrotexture measurement holds great promise as a tool to detect and quantify segregation for quality-assurance purposes. Correlations of texture with void level/density and asphalt content (traditional measures of segregation) are strong, especially for mixes with moderately-sized aggregates.

DETERMINING MINIMUM LIFT THICKNESS FOR HOT MIX ASPHALT (HMA) MIXTURES

by E. Ray Brown, M. Rosli Hainin, and L. Allen Cooley, Jr.

Abstract

One of the main issues that has been continuously discussed among asphalt technologists is the minimum lift thickness necessary to provide a high probability of adequate compaction under a reasonable compactive effort. In many cases, pavements constructed with thin lifts have been difficult to compact to the desired density level. Thinner lifts tend to cool faster and restrain the movement of aggregate particles during compaction. To date, little research has been performed to determine the minimum lift thickness needed to achieve desirable air voids during construction. This research project was undertaken to determine the minimum lift thickness for HMA mixtures.
A total of 36 mixes were designed comprising various gradations, aggregate types and nominal maximum aggregate sizes (NMAS). Laboratory samples were prepared using both Superpave gyratory compactor (SGC) and automated vibratory compactor (AVC), and seven field sections were built to evaluate the relationship between air voids and thickness to nominal maximum aggregate size (t/NMAS). The results of the study indicated that t/NMAS has a significant impact on the compactibility of HMA mixtures. For improved compactibility, it is recommended that the t/NMAS be at least 3.0 for fine-graded mixes and at least 4.0 for coarse-graded mixes. Layers thinner than that recommended will likely require more compactive effort to achieve adequate density. The cooling rate for mixes less than 38 mm is very high making compaction difficult. Hence for best results, the thickness should be at least 38 mm even during warm weather. Paving in cold weather may require an even greater layer thickness.

EVALUATING WISCONSIN WARRANTY PROJECTS: BEFORE, DURING AND AFTER THE WARRANTY

by Ervin L. Dukatz, Jr. and Scot W. Schwandt

Abstract

This paper is the third in a series on the Wisconsin HMA warranty pavement experience. This paper is an update on the 24 projects previously covered in presentations in 1996 and 2001. The performance histories of all 55 warranty projects constructed through the end of 2003 are tracked from their construction date to the present. This data included all the projects that have passed their 5-year warranty periods. The background on development of Quality Control specifications in Wisconsin is presented to aid in explaining how the Warranty specification in Wisconsin has extended pavement life, improved quality and reduced costs. Future directions of the warranty specification are discussed.

Session I-Monday 1:30 pm, Frank Fee Presiding

DEVELOPMENT OF SUPERPAVE HIGH-TEMPERATURE PERFORMANCE GRADE BASED ON RUTTING DAMAGE

by Alaeddin Mohseni, Dr. Samuel Carpenter, P.E., and John D'Angelo

Abstract

SUPERPAVE asphalt binders are selected using the Performance Grade (PG) methodology introduced by SHRP. The PG grading system has two components; low-temperature and high- temperature. The work presented in this paper provides an improved methodology to select the high temperature PG grade for asphalt binders. An enhanced high-temperature transfer function for pavement temperatures was developed through a comprehensive application of the Integrated Climatic Model (ICM). This model was developed using hourly climatic data, providing a more accurate representation of the high-temperature conditions in the pavement. Using the hourly high-temperature calculations, a performance based procedure utilizing current empirical models for HMA stiffness and rut predictions was developed.
This procedure is a damage based comparison of rut development and the impact of the use of different PG graded binders to compare the accumulation of damage. It accounts for extended hot periods in certain climates, which could not be considered in the original selection procedure. Models were developed to extend the PG damage concept from the limited number of stations with hourly temperature data to the entire database using daily temperature data. A Degree-Days concept was developed to relate the number of days the pavement temperature is over 10 C to the PG damage calculation. The performance based nature of the damage concept allows for a mechanistic inclusion of traffic amount, and the traffic speed. Recommendations for grade bumping for traffic and speed were presented. Variability was included to provide protection against the one-time extreme temperature event. An existing software program was modified to include the new algorithm for implementation.

RATIONAL FATIGUE LIMITS FOR ASPHALT BINDERS DERIVED FROM PAVEMENT ANALYSIS

by Rodrigo Delgadillo and Hussain Bahia

Abstract

Superpave specification has some critical gaps, most of them related to the performance characterization of modified bitumens. This issue is generally recognized by the State Highway Agencies, industry and researchers [1]. In 2001 NCHRP 459 [1] report was published, offering a revised system for testing and evaluating asphalt binders based on damage behavior. For fatigue characterization of binders, a new specification parameter was proposed, called Np20. One of the main advantages of the new parameter was the inclusion of the pavement structure in the specification. This paper starts with a review of why G*sind has to be replaced by the new parameter Np20. The NCHRP 459 proposal for using Np20 was, however, only conceptual and lacked the details required for implementation, specific criteria and limits. The main purpose of this paper is to show how pavement structural analysis can be use to implement the binder fatigue specification parameter Np20. Pavement analysis using multilayer elastic theory program was used to determine the strain levels in the pavement for 6 different pavement structures typically used in practice. From those strains, the fatigue life of 19 different binders was calculated. Known relationships between mixture and binder strain, and between mixture and binder fatigue life, were used to perform the calculations. The results allowed determining the suitability of the 19 binders for different traffic volumes, speeds and pavement structures. Testing conditions (temperature and input energy) for the conditions of Wisconsin and specification limits for Np20 are proposed.

SIMPLE METHOD TO OBTAIN ASPHALT BINDERS LOW TEMPERATURE PROPERTIES FROM ASPHALT MIXTURES PROPERTIES

by Mihai Marasteanu, Adam Zofka, Timothy Clyne, Xinjun Li, and Jim McGraw

Abstract

The purpose of this research was to investigate the possibility of developing a simple method that could be used to obtain the low temperature properties of the asphalt binders contained in an asphalt mixture without using the extraction and recovery process. This information is critical in developing blending charts for mixtures that use RAP. Limited tests performed with the Bending Beam Rheometer (BBR) on thin beams of mixtures showed a lot of promise in obtaining reasonable values for the mixture stiffness. The recently proposed Hirsch model was used to backcalculate the binder stiffness from the BBR mixture stiffness and the results were compared to the extracted binder stiffness. The results were used to recalibrate one of the parameters in the Hirsch model to closely match the predicted and the experimental binder stiffness at 60 seconds.

POLYPHOSPHORIC ACID MODIFIED ASPHALT : PROPOSED MECHANISMS

by Gaylon L. Baumgardner, J-F. Masson, John R.Hardee, Andy M. Menapace, and Austin G. Williams

Abstract

Asphalt binders have been chemically modified with polyphosphoric acid (PPA) to improve high temperature rheological properties without adversely affecting low temperature rheological properties since the early 1970's. More recently, PPA has been used in Superpave performance- grade (PG) binders that need an extended range between the high and low temperature performance requirements. The mechanism of chemical modification of asphalt with PPA remains in great part unknown. This paper presents results that will help to better understand the mechanisms of chemical modification with PPA. PPA modified and unmodified asphalts from different crude sources were analyzed by chemical methods (n-heptane insolubles), thin-layer chromatography (TLC), gel-permeation chromatography (GPC) and atomic force microscopy (AFM). The results indicate that the mechanism of PPA action depends on the base asphalt. In one case PPA affected a phase dispersed in asphalt, in the other case it affected the asphalt matrix. In both cases, PPA caused stiffening of the modified phase. Three stiffening mechanisms are proposed.

A SYNTHESIS OF THEORIES AND MECHANISMS OF BITUMEN-AGGREGATE ADHESION INCLUDING RECENT ADVANCES IN QUANTIFYING THE EFFECTS OF WATER

by Arno W. Hefer, Dallas N. Little, and Robert L. Lytton

Abstract

A synthesis of research on bitumen-aggregate adhesion is presented within the framework of theories and mechanisms established in the general field of adhesion science. An effort is made to convey the material with perspective and to introduce terms and concepts that would aid in understanding the theories and associated mechanisms. The ability of these theories to place bitumen-aggregate adhesion and the effect of moisture on a quantitative level is considered and practical implications discussed. Experimental support is included where available. The theory of (weak) boundary layers, mechanical theory, electrostatic theory, chemical bonding theory, and thermodynamic theory are presented. Thermodynamic theory is the most widely applied by adhesion scientists due to its universal energy basis. Compatibility of different bitumen-aggregate pairs can be assessed if their surface free energies are known. In addition, surface energy in combination with other material properties serves as an important ingredient in performance prediction, including stripping potential. Electrostatic theory reveals that important additional mechanisms evolve in the presence of water. It is shown how the thermodynamic model, previously used to assess bitumen-aggregate adhesion, can be extended to include the effects of electric double layer forces as well as pH of water at the bitumen-aggregate interface.

Session II-Tuesday 9:00 am, Erv Dukatz Presiding

COMBINED LABORATORY AGEING / MOISTURE SENSITIVITY ASSESSMENT OF HIGH MODULUS BASE ASPHALT MIXTURES

by Gordon D. Airey, Young K. Choi, Andrew C. Collop, Abigail J.V. Moore, and Richard C. Elliott

Abstract

Over the last ten year there has been a general trend in the UK to use progressively stiffer base materials, due to their expected "long life" performance. However, recent concerns over the durability of these materials has led the UK Highways Agency (HA) to temporarily suspend the use of 15 and 25 penetration grade bitumen and establish a research programme to investigate the long- term durability of High Modulus Base (HMB) material. This paper describes the relative performance of a new combined ageing / moisture sensitivity laboratory test at evaluating the resistance of HMB mixtures to moisture induced damage compared to the standard procedure used in AASHTO T283. The test is known as the Saturation Ageing Tensile Stiffness (SATS) test and consists of initial saturation prior to placing compacted asphalt cylindrical specimens in a moist, high temperature and pressure environment for an extended period of time. The stiffness modulus measured after the test divided by the stiffness modulus measured before the test (retained stiffness modulus), and the specimen saturation after the test (retained saturation), are used as an indication of the sensitivity of the compacted mixture to the effects of combined ageing and moisture. The test has been successfully used to reproduce in the laboratory the loss of stiffness modulus observed on a trial site where HMB material had been placed. The two water sensitivity conditioning methods have been used to assess the resistance to moisture damage of six HMB asphalt mixtures. The mixtures have incorporated different aggregate types (basic and acidic), two 15 penetration grade bitumens, two aggregate gradations (UK and French designs), varying binder and air void contents and different fillers. The results from the SATS test have been compared directly with those obtained on the same mixtures using the AASHTO T283 procedure. Both the SATS and AASHTO method have been able to rank the moisture conditioning performance of the different mixtures with the order of ranking being identical. The HMB mixtures containing the acidic aggregate (including low air void content and high binder content combinations) produced lower ratios of retained stiffness (and strength for AASHTO T283) compared to the basic aggregate mixtures as well as an acidic aggregate mixture including hydrated lime filler. However, the relative performance of the two tests (in terms of retained stiffness ratio) differed considerable for the moisture susceptible acidic aggregate mixtures with the retained stiffness values from the AASHTO T283 procedure being approximately double that of the SATS test at nominally the same saturation levels. The combined ageing / moisture conditioning SATS test can therefore be considered to be a more severe procedure compared to the accelerated water conditioning with freeze-thaw cycle AASHTO T283 method. In addition, the SATS test was the only procedure that was able to reproduce the 60% reduction in stiffness modulus found for the acidic aggregate HMB mixture in the field.

RELATIONSHIP BETWEEN TRIAXIAL SHEAR STRENGTH AND INDIRECT TENSILE STRENGTH OF HOT MIX ASPHALT

by Terhi K. Pellinen and Shangzhi Xiao

Abstract

Testing conditions for the indirect tensile (IDT) strength test were instigated in order to determine the Mohr-Coulomb failure equivalency between the triaxial shear strength and the IDT strength. The time-temperature superposition principle was applied to determine equivalent testing conditions. Testing verified that the time-temperature superposition principle continues to hold at large failure strains. A predictive technique to obtain the cohesion (c) of an asphalt mixture based on the IDT strength was studied and the overall ratio of (c) to IDT strength turned out to be 1.73, which agrees very well with the findings by other researches. The error associated with the prediction of (c) varied between 0.9 to 38% for individual mixtures, averaging less than 21%.

FIELD EVALUATION OF WITCZAK AND HIRSCH MODELS FOR PREDICTING DYNAMIC MODULUS OF HOT-MIX ASPHALT

by Raj Dongre, Leslie Myers, John D'Angelo, Chuck Paugh, and Jagan Gudimettla

Abstract

Dynamic modulus values were predicted using the Witczak and Hirsch Models for asphalt mixtures from five pavement construction sites across the United States where the FHWA Mobile Asphalt Laboratory conducted localized materials testing. Superpave G* and phase angle data was measured for original (unaged), RTFO, and PAV aged asphalt binders used at these sites, at various temperatures and frequencies. The binder data and hot-mix volumetric properties were used as input in the Witczak and Hirsch models to predict hot-mix dynamic modulus at various temperatures and frequencies. Dynamic modulus was also measured for mix from each site using the new Superpave Simple Performance Tester (SPT) in unconfined compression. Dynamic modulus values predicted from equations were then compared to the values measured in the laboratory to evaluate the predictive capability of the Witczak and Hirsch models. Master Curves were also generated. In addition, the Global Aging System proposed by Mirza and Witczak was investigated using the dynamic modulus measurements and predictions.
The results indicate that both models provide reasonable predictions of dynamic modulus within the scope of this study. The Hirsch model was found to be easier to use, more accurate, and was capable of slightly wider range of applicability at higher temperatures. Both models require further refinement to better predict dynamic modulus of production samples that have the same gradation, but deviate in asphalt binder content and volumetric properties. It was also found that the Global Aging System overpredicted the severity of aging that occurs during mixing/placement of hot-mix asphalt. Shift factors calculated using binder data alone were found to give reasonably accurate master curves for hot-mix data. It was also found that the binder's Useful Temperature Range (UTR) calculated by algebraically summing the high and the low continuous PG grade temperatures may be used to accurately predict the activation energy EA. The hot-mix shift factors may then be determined using the predicted EA and the Arrhenius shift factor equation.
NCHRP 1-37A design guide software was used to predict the performance of a selected conventional asphalt concrete pavement. Both measured and predicted dynamic modulus data from each site was used in the design guide software for a Level 1, Level 2, and Level 3 analysis. In general, the AC rutting results predicted by the design guide were found to be inconsistent at any Level analysis with the magnitude of differences in the Superpave PG grades of asphalt binders used (from PG58-28 to PG70-22). The resulting IRI values were also not found to change significantly. In the Level 3 analysis, some trend was observed between the AC rutting and the high temperature PG grades, but the differences were not significant. The default A and VTS values used in Level 3 analysis were found to be significantly different than those determined in this study.
Based on the above results, recommendations are made for improvements in the Witczak and Hirsch predictive models. Recommendations are also given for refinement of the NCHRP 1-37A design guide models for characterizing asphalt materials.

OVERLAY TESTER: A SIMPLE PERFORMANCE TEST FOR THERMAL REFLECTIVE CRACKING

by Fujie Zhou and Tom Scullion

Abstract

Reflective cracking is a major concern when placing an asphalt overlay on a cracked pavement. The opening and closing of joints and/or cracks and associated thermal stresses produced by daily temperature cycles is a major contributor to reflective cracking. This mechanism has been simulated in the laboratory using the Texas Transportation Institute's Overlay Tester. The test specimens used are 150 mm long and can be conveniently cut from either field cores or from laboratory molded specimens. The test is simple to conduct and the results are obtained generally in less than 1 hour. Over the past decade the focus of asphalt mixture design has been largely to address the rutting concerns. The Overlay Tester device described in this paper appears to be a tool to let the designer balance the competing requirements of both rut and crack resistance.
Preliminary studies on replicate specimens found the test to be highly repeatable. Sensitivity studies indicated that the Overlay Tester provides reasonable results, in that raising the asphalt performance grade and decreasing the testing temperature will lead to shorter reflective cracking life. However, increasing asphalt content will significantly improve the reflective cracking resistance. Furthermore, in a series of controlled tests on typical Texas mixtures it was determined that aggregate absorption appears to have a major impact on overlay test performance. In this paper the Overlay Tester concept is validated on five case studies in Texas. In each case cores from the pavement were tested and their laboratory results correlated well with actual field performance.
A preliminary framework for using the overlay tester as part of a mixture design process is proposed. The Overlay Tester is recommended to characterize the reflective cracking resistance, and the Hamburg wheel tracking test and/or APA is proposed for evaluating rutting resistance. These devices provide the designer with tools to balance the conflicting rutting and cracking requirements. This framework is used to select the optimum asphalt content for two projects in Texas.

RUT RESISTANCE AND VOLUMETRIC COMPOSITION OF ASPHALT CONCRETE MIXTURES

by Donald W. Christensen, Jr. and Ramon Bonaquist

Abstract

The paper presents an accurate model for estimating mixture rut resistance using the concept of mixture resistivity, which is the inverse of the permeability of the mixture aggregate to the specific binder used in the mixture, at the critical temperature for rutting in a given application. Laboratory testing using the repeated shear at constant height test showed very good correlations between resistivity, compaction effort, and maximum permanent shear strain, a good indicator of rut resistance. To verify the usefulness of the proposed model, field rutting data from MN/Road, WesTrack and the NCAT Test Track were modeled, using resistivity, Ndesign, and relative field density as predictor variables. A single model resulted which accurately predicted field rutting rates for all three projects (r2 = 89%). This model was used to analyze the effects of changes in mixture composition, binder grade, and compaction on rut resistance, and also in a preliminary analysis of the current Superpave mix design system. The most important factor affecting rut resistance in HMAC mixtures is binder grade, followed by aggregate fineness relative to VMA. Large variations in rut resistance are possible in mixtures of a given type; experience and judgment are needed to ensure proper performance levels are obtained.

Symposium: Tuesday 1:00 pm, "Quiet Asphalt Surfacing Technology," Joe Button Presiding

Session III-Wednesday 9:00 am, Rita Leahy Presiding

A FULL-SCALE PAVEMENT STRUCTURAL STUDY FOR MECHANISTIC-EMPIRICAL PAVEMENT DESIGN

by Angela L. Priest and David H. Timm

Abstract

As pavement structural design continues to progress toward mechanistic-empirical (M-E) methodologies, local agencies will be called upon to develop some of the key inputs and models for M-E design. Local calibrations for environment, materials and traffic along with equations to link pavement response to field performance are critical to the successful introduction and adoption of M-E design. The structural study currently underway at the National Center for Asphalt Technology (NCAT) Test Track is an example of one such approach to develop the local calibrations and required equations for flexible pavements and can serve as a guide to others in the development of full-scale pavement testing facilities. There are many critical components that must be addressed in order to have a successful structural study including the design of the test sections, instrumentation selection, sensor installation, data collection and analysis. These issues are discussed in light of the NCAT Test Track to aid in the planning and execution of future structural experiments as local agencies continue to address the needs of M-E pavement design.

A PRACTICAL LOOK AT THE SIMPLE PERFORMANCE TESTS: LOUISIANA'S EXPERIENCE

by Louay N. Mohammad, Ph.D., Zhong Wu, Ph.D., Leslie Myers, Ph.D., P.E., Sam Cooper, P.E., and Chris Abadie, P.E.

Abstract

The Superpave volumetric mix design procedure developed during the Asphalt Research Program of the Strategic Highway Research Program did not include a mechanical "proof" test similar to the ones commonly used in the Marshall mix design or Hveem mix design such as the Marshall stability and flow tests or the Hveem stabilometer method, respectively. The Superpave mix design method, however, did use strict requirement to material specifications and volumetric mix criteria to ensure satisfactory performance of mix designs that was intended for low volume traffic. In addition, the original Superpave mix design protocol required mix verification for intermediate and high volume traffic through advanced materials characterizations tests utilizing the Superpave Shear Tester test protocols. It was quickly recognized the complexity of those test protocols for routine mix design application and that a simple performance test is needed to compliment the Superpave volumetric mix design procedure. In response to this need, NCHRP Project 9-19, Superpave Support and Performance Models Management, recently recommended three candidate Simple Performance Tests (SPTs) to complement the Superpave volumetric mixture design method. These are flow time (FT), flow number (FN), and dynamic modulus |E*| tests. In addition, the dynamic modulus test was selected for the HMA materials characterization input utilized in the 2002 Empirical and Mechanistic Guide for Design of New and Rehabilitated Pavement Structures, developed under NCHRP Project 1-37A. This paper presents a practical look at the results of a cooperative evaluation of similar Superpave mixtures utilizing two tests of the SPTs: the dynamic modulus and Flow Number. Two different dynamic modulus predictive models, Witczak and Hirsch, were also evaluated.

VERIFICATION FOR THE CALIBRATED PERMANENT DEFORMATION MODELS FOR THE 2002 DESIGN GUIDE

Mohamed M. El-Basyouny, Matthew W. Witczak, and Sherif El-Badawy

Abstract

The new design approach adopted in the 2002 Design Guide utilized a mechanistic-empirical pavement design procedure. The 2002 Design Guide provides time series predictions of several major distress types. One such distress mechanism is the permanent deformation (rutting). The Design guide solution predicts the total pavement rut depth, as a function of time, by predicting and summing the individual rutting contributed by all rut susceptible layers (asphalt, granular base, subbase and subgrade). This necessitated that a set of rutting models be implemented in the new design method to reflect the new design procedure. The pavement rutting prediction models included in the Design Guide nationally were calibrated by using actual sections from the LTPP database, to reflect the field performance of the flexible pavements.
To ensure that the calibrated model is as accurate as possible and that the predicted model trends are as close as possible to what experience, practical knowledge and reasonable engineering judgment of the performance of the asphalt concrete pavements allows; an extensive sensitivity analysis was conducted using a wide variety of salient variables that were felt to have an impact on pavement rutting considered in the Design Guide. This paper contains an in-depth detailed analysis for the entire sensitivity study.
In this study, a typical section was run using the 2002 Design Guide software. Different levels for each variable considered were used while keeping the other variables constant. The results of the sensitivity runs and the performance of the section with respect to the variables was found o be very logical and rational. In fact, the mechanistic-empirical design approach appears to provide the user with a much more powerful tool to assess the complex interaction of design parameters to performance than is currently available in any other design methodology in the world.

PREDICTION AND CALIBRATION OF PAVEMENT FATIGUE PERFORMANCE USING TWO-STAGE WEIBULL APPROACH

by Bor-Wen Tsai, M.O. Bejarano, J. T. Harvey, and C.L. Monismith

Abstract

An integrated two-stage Weibull model was established to take into account both crack initiation and crack propagation of laboratory controlled-deformation flexural fatigue beam tests at various testing conditions and material properties. Correction factors which calibrate between the laboratory testing results and in-situ pavement performance were obtained by comparing the simulated stiffness deterioration curve with the in-situ stiffness deterioration utilizing accelerated pavement testing facilities-Heavy Vehicle Simulator (HVS) sponsored by the California Department of Transportation Accelerated Pavement Testing Program (CAL/APT). Deflection data from multi-depth deflectometer (MDD) was used to back-calculate the stiffness deterioration of asphalt concrete based on the Odemark-Boussinesq method. In the two-stage Weibull simulation, an experimental design including two fatigue damage accumulation mechanisms ("repetition-hardening" and "stiffness-ratio-hardening"), two air-void content patterns ("exponential decreased" and "constant"), and three wander patterns ("uniform", "normal", and "no-wander") was evaluated and compared. Results indicate that the two-stage Weibull approach is quite promising in fatigue pavement performance prediction and can be utilized toward more complicated in-situ applications.

VERIFICATION OF THE CALIBRATED FATIGUE CRACKING MODELS FOR THE 2002 DESIGN GUIDE

by Mohamed M. El-Basyouny and Matthew W. Witczak

Abstract

The new design approach adopted in the 2002 Design Guide utilized a mechanistic- empirical pavement design procedure. This necessitated that a new set of fatigue cracking models to be used in the new design method to reflect the new design procedure. The pavement fatigue cracking prediction models included in the Design Guide were calibrated by using actual sections from the LTPP database, to reflect the field performance of the flexible pavements.
To ensure that the calibrated model is as accurate as possible and that the predicted model trends are as close as possible to what experience, practical knowledge and reasonable engineering judgment of the performance of the asphalt concrete pavements allows; an extensive sensitivity analysis was conducted using a wide variety of salient variables that were felt to have an impact on both types of fatigue cracking (top-down and bottom-up) considered in the Design Guide. This paper contains an in-depth detailed analysis for the entire sensitivity study.
A typical section was run using the 2002 Design Guide software. Different levels for each variable considered were used while keeping the other variables constant. The results of the sensitivity runs and the performance of the section with respect to the variables was very logic and relevant to the performance expected.

INFLUENCE OF ASPHALT BINDER PROPERTIES ON THE FATIGUE PERFORMANCE OF ASPHALT CONCRETE PAVEMENTS

by Bor-Wen Tsai and C.L. Monismith

Abstract

A number of studies have suggested that limits for the loss stiffness of the binder G*sinė in the AASHTO binder specification is, by itself, not a sufficient indicator of the fatigue performance of asphalt concrete (AC) in asphalt pavement structures. To assist the Pacific Coast Conference on Asphalt Specifications (PCCAS) in deliberations pertaining to the AASHTO binder specification, this study was undertaken. The purpose of the study was to evaluate the influence of nine binders, selected by the PCCAS participants, on pavement fatigue performance for a representative range in pavement sections, environmental conditions, and traffic loading.
The program has included: 1) tests on the binders to define their rheologic response over a range in temperatures and times of loading; 2) fatigue tests on mixes containing the binders; and 3) analysis of the simulated fatigue performance of the mixes in pavement structures in four environments of the Pacific region. With the exception of the thin AC pavement structure, mixes containing the nine binders provided adequate simulated fatigue performance (estimated ESALs equal to or exceeding design ESALs) at a 90 percent reliability level.
Two alternatives for the binder fatigue requirement are briefly discussed, i.e., retain the current specification requirement or eliminate it. If the current fatigue requirement is eliminated, the question to be asked by the pavement community is what might happen without it? For example, could materials which are different than those used in this investigation (all of which met the current criterion and performed adequately according to the pavement simulations) result from changes in production? Until this question can be resolved, it is recommended that the prudent course is to maintain the current requirement.

Session IV-Wednesday 1:00 pm, Steve Brown Presiding

THE "DBN" MODEL: A THERMO-VISCO-ELASTO-PLASTIC APPROACH FOR PAVEMENT BEHAVIOR MODELING. APPLICATION TO DIRECT TENSION TEST AND THERMAL STRESS RESTRAINED SPECIMEN TEST

by Francois Olard and Herve Di Benedetto

Abstract

Mix behavior is complex. Following applied loading, very different types of properties can be observed, such as linear viscoelasticity (LVE) for very small strain amplitudes, non linearity for higher strains, fatigue or rutting for a great number of cycles, brittle or ductile failure, thixotropy and healing. These different "typical" kinds of behavior are introduced. Then, the proposed "DBN" law is presented. This law is very versatile and provides an effective means of describing the different types of mix behavior by using the same formalism. An emphasis is especially placed on explaining how the law can cope with different "typical" mix behaviors. The complexity of the description and the number of constants can be very different following desired sophistication for the modeling. In particular, some comparisons between experimental DTT and TSRST results and numerical simulations based on the DBN law are presented. Some of the presented aspects still need further research work.

PRACTICAL APPLICATION OF CONTINUUM DAMAGE THEORY TO FATIGUE PHENOMENA IN ASPHALT CONCRETE MIXTURES

by Donald W. Christensen, Jr. and Ramon Bonaquist

Abstract

A useful modification of continuum damage equations originally presented by Kim et al. is described in this paper. Several practical applications of these relationships are presented, including reduction of laboratory fatigue data and analysis of fatigue damage in asphalt concrete pavements. A rapid procedure for performing and analyzing laboratory fatigue tests is presented. Data gathered using this procedure was combined with flexural fatigue data produced during the Strategic Highway Research Program to develop a semi-empirical equation for estimating fatigue life from volumetric composition, degree of compaction and asphalt binder rheological type (R-value). Based upon these analyses, the authors recommend that for applications requiring fatigue resistant asphalt concrete, design VFA values below 70% should be avoided, and good field compaction should be emphasized. Further research is needed to evaluate the relationships among fatigue resistance, healing rate and asphalt binder rheology.

COMPREHENSIVE EVALUATION OF DAMAGE IN ASPHALT MASTICS USING X-RAY CT, CONTINUUM MECHANICS, AND MICROMECHANICS

Injun Song, Dallas Little, Eyad Masad, and Robert Lytton

Abstract

Asphalt pavements experience damage due to traffic loading under various environmental conditions. Damage is caused by viscoplastic flow and microcracks, fracture due to fatigue, or fracture due to thermal effects. Asphalt pavements have the capability to recover from some of this damage depending on binder surface and rheological properties, filler surface properties, and length of rest periods.
Asphalt mastic (asphalt and mineral filler) properties play a major role in controlling damage and healing. This paper addresses the development of a comprehensive methodology used to characterize damage and healing in asphalt mastics. The methodology relies on nondestructive imaging techniques (X-ray CT), principles of continuum damage mechanics, and principles of micromechanics. The X-ray CT yields a damage parameter that quantifies the percentage of voids (cracks and air voids) in a specimen. The continuum damage model parameters are derived from the relationship between applied stress and pseudo strain. The micromechanics model relates the damaged mastic modulus to a reference undamaged modulus. This relationship is a function of internal structure properties (void size, film thickness, and percentage of voids), binder modulus, aggregate modulus, and bond energies within the mix. The internal structure parameters are all obtained using X-ray CT.
This methodology is used to characterize damage in asphalt mastic using the Dynamic Mechanical Analyzer (DMA). Damage is monitored as a function of loading cycles, rest period, and moisture state (dry and wet conditions). The damage parameter measured by X-ray CT correlated very well with the predictions of the continuum and micromechanics models. All damage parameters were able to reflect the accumulation of damage under cyclic loading and were also able to capture the influence of moisture conditioning on damage. The healing due to rest periods was captured using X-ray CT, and it was reflected in the continuum and micromechanics damage models. Although this paper focused on fatigue cracking and healing at room temperatures, the methodology developed can be used to assess damage due to different mechanisms such as permanent deformation and low temperature cracking.

A MICROMECHANICS-BASED APPROACH FOR DETERMINING PRESENCE AND AMOUNT OF RECYCLED ASPHALT PAVEMENT MATERIAL IN ASPHALT CONCRETE

by William G. Buttlar and Eshan V. Dave

Abstract

The economical use of pavement milling in recent years has resulted in the availability of significant amounts of recycled asphalt pavement (RAP). The use of RAP in new asphalt concrete mixtures can result in cost savings for both aggregate and asphalt binder. When properly utilized in mix design and production, lower costs can be realized without a sacrifice in the serviceability of the pavement. However, cost savings may not be realized by the owner and there may be a loss in serviceability of the pavement when unauthorized use occurs. This study was conducted to modify and/or develop a test to determine the presence and amount of RAP in post-production mixtures as a practical quality assurance tool. A testing and analysis procedure was developed to identify the presence and amount of RAP in asphalt mixtures. The development of the procedure was carried out by testing of laboratory as well as plant mixtures with known RAP amounts. To validate the most promising method two sets of blind samples were evaluated. An extensive testing of RAP materials from various sources was also carried out to determine the effect of RAP variability on the newly developed procedures. From a literature review it was identified that asphalt binder properties change most dramatically with aging and therefore a procedure based on testing of asphalt binders was developed. An extensive procedure coupled with an analysis technique based on micromechanics was developed for determining amount of RAP in the asphalt concrete.

CHARACTERIZATION OF PAVEMENT DAMAGE DUE TO DIFFERENT TIRE CONFIGURATIONS

by Imad L. Al-Qadi, Pyeong J. Yoo, and Mostafa A. Elseifi

Abstract

In response to the introduction of new generations of wide-base tires, pavement responses to different tire configurations were measured at the Virginia Smart Road. The investigated new generations of wide-base tire have a wider tread and a greater load-carrying capacity than conventional wide-base tires. Results of the experimental program indicated that the steering axle was the most damaging of all the tested axles. In addition, measured stresses and strains were used to evaluate potential fatigue and rutting damage resulting from different tire configurations. A finite element (FE) parametric study was then conducted to evaluate different failure mechanisms that were not quantified in the field. The developed three-dimensional FE models accurately simulated actual tire tread sizes and applicable contact pressure on each rib and incorporated laboratory-measured pavement material properties into the analysis. The developed FE models were also calibrated and properly validated against stress and strain measurements obtained from the experimental program. Four failure mechanisms were considered in the pavement damage evaluation: fatigue cracking, primary rutting, secondary rutting, and top-down cracking. Results indicated that the first new generation of single wide-base tires would cause relatively greater pavement damage than conventional dual tires. On the other hand, a recently-developed second size new generation of wide-base tire would induce similar pavement damage as that of conventional dual tires.