Robert Lundström

Asphalt Fatigue Modelling using Viscoelastic Continuum Damage Theory

ABSTRACT Three conventional asphalt concrete mixtures containing binders of different penetration, obtained from one and the same source, are characterised with regard to fatigue behaviour at three different temperatures using uniaxial testing. The results are analysed using a viscoelastic continuum damage model. The model eliminates strain and stress level dependency as well as mode of loading dependency. Fatigue paths at different temperatures can be predicted using the time-temperature superposition principle, if the material is not too soft. Difficulties in predicting the fatigue damage evolution at soft conditions might be explained by the occurrence of additional softening mechanisms such as hysteretic heating during testing.

Characterization of Asphalt Concrete Deterioration Using Monotonic and Cyclic Tests

A viscoelastic continuum damage mechanics (CDM) model is used to assess whether or not it would be possible to predict fatigue characteristics of asphalt concrete mixtures by the aid of experimental results obtained using monotonic tests. The CDM-model is shown to effectively eliminate strain rate dependence at monotonic testing and strain and stress amplitude dependence as well as mode of loading dependence of fatigue testing. The model is also shown to be able to model effects of different temperatures by the use of the time-temperature superposition principle, at least under conditions of not too low a stiffness. However, based on the results presented, it is difficult to accurately predict fatigue results based on characteristic curves obtained from monotonic tests under all testing conditions. The reason may be difference in damage mechanisms of the two types of loading (monotonic vs. cyclic).

Influence of Hysteretic Heating on Asphalt Fatigue Characterization

This paper presents a study focusing on the influence of hysteretic heating on asphalt samples during laboratory fatigue testing. The experimental test setup for material characterization and temperature measurements, including its effect on fatigue test results, as well as theoretical aspects on hysteretic heating, are described. The experimental part of the investigation concerns linear viscoelastic and cyclic fatigue characterization of six asphalt concrete mixtures using uniaxial testing. All the mixtures show nominally identical volumetric properties (aggregate size distribution, binder and air void content) but different binder properties. Three base bitumens and three polymer modified binders were used. The cyclic fatigue tests were carried out at 0, 10, and 20°C using controlled strain and stress modes and different excitation amplitudes. In order to acquire knowledge regarding temperature changes during fatigue testing, several experimental techniques were used. The main thermal study was performed using thermocouples attached to the midheight envelope surface of each sample. The sample surface temperature distribution and its evolution during fatigue testing were investigated using an infrared thermal camera. Furthermore, a limited study of the magnitude of difference between surface and maximum temperature inside the sample was carried out using thermocouples embedded during gyratory compaction. When compared, each method shows advantages and disadvantages regarding simplicity and reliability. In principle, the three methods provide similar results, but the type of information obtained differs among the methods. The use of thermocouples attached to the envelope surface during fatigue testing provides accurate and consistent results of global temperature that can be used to investigate the influence of heating on asphalt fatigue characteristics. By use of thermal measurements and a continuum damage model, it was possible to show a pronounced effect of heating on fatigue behavior. The influence of heating was especially obvious at high excitation amplitudes and elevated temperatures, i.e., conditions where the material produces high amounts of viscoelastic dissipated energy as well as temperature sensitive material behavior.

Fatigue Modeling as Related to Flexible Pavement Design

ABSTRACT A literature study of rheological and fatigue modeling of asphalt mixtures is presented. Theoretical aspects on structural modeling, rheological behavior and the fatigue integration in design procedures are reviewed. In principle, pavement design methods can be categorized in three broad groups: empirical, semi-mechanistic and fully mechanistic methods. Pavement design is generally performed using semi-mechanistic methods comprising analytical or numerical structural response models and deterioration modeling based on transfer functions and shift factors. In the case of fatigue deterioration, several approaches have been elaborated e.g. classical models, fracture mechanics and damage mechanics. The approaches differ regarding theoretical foundation and evaluation methods used. Recognizing significant limitations concerning theoretical basis as well as lack of empirical support for current design methods, a shift in paradigm from semi-empirical methods towards more advanced fully mechanistic methods have been initiated. According to this approach, improved pavements are achieved by appropriate design methods which are capable of predicting fatigue resistance in the actual pavement environment, and thus taking into account complex stress conditions, influence of temperature and material characteristics, such as aging and healing.

Influence of Pavement Materials on Field Performance: Evaluation of Rutting on Flexible, Semi-rigid and Rigid Test Sections after 7 Years of Service

ABSTRACT This paper presents results obtained so far from a Swedish full-scale test, comprising in total 19 different rigid, semi-rigid and flexible test stretches. The purpose is to compare rutting performance of the different pavement categories and types using extensive laboratory tests and field measurements after seven years of trafficking. The rigid pavements withstand rutting better than the two other pavement categories due to comparably small abrasion and negligible permanent deformation. The semi-rigid pavements show less rutting than the flexible pavements, which primarily is the result of the negligible permanent deformation in the cement bound layers. Even though the rigid pavements show very small rutting, it is probably not the main distress type and the primary trigger for maintenance activities for this pavement type. In the case of semi-flexible and flexible pavements, both laboratory tests and field measurements indicate that significantly improved rutting performance can be obtained using appropriate asphalt mixtures manufactured using wear-resistant aggregates and stiff binder.

Causes of rutting in flexible and semi-rigid test sections after 14 years of service

Rutting is a major distress and is commonly targeted in design-build contracts as a key requirement, but at the same time, contemporary design methods usually provide scarce information on evolution in absolute terms. The objective of this paper is to investigate and analyse rutting results from a large full-scale road test. The analyses concerned magnitudes and the causes of rutting with a main focus on flexible and semi-rigid structures: one Reference, one high-performance asphalt (HPA) and one asphalt on a lean concrete (LC) base. Field measurements and sampling for the current study comprised acquiring transversal profiles and coring pavement samples. The results suggest that for the HPA and the LC base pavements, rutting is mainly caused by studded tyre wear and densification of the asphalt layers. For the conventional reference pavement additional rutting, most likely in the lower layers, was noted.