Determination of fundamental tensile failure limits of mixtures
Conference Paper
Overview
Additional Document Info
View All
Overview
abstract
Cracking is one of the most influential distresses that governs the service life of asphalt concrete pavements. Acquiring more insight into the fracture behavior of asphalt mixtures is essential to improve the cracking resistance of asphalt pavements. Recent work at University of Florida (1,2) has focused on the development of a new visco-elastic fracture mechanics-based crack growth law which identified the fracture energy density of mixtures as a fundamental crack growth threshold. The work that led to the development of the HMA fracture mechanics framework was based solely on Superpave IDT test results, therefore, it is of considerable interest to evaluate the existence of a fundamental fracture threshold for other laboratory test configurations. The work presented in this paper focuses on the identification and evaluation of fundamental tensile failure limits of hot mix asphalt mixtures (HMA) from multiple laboratory test configurations. Fracture energy densities and tensile strengths were evaluated for three different mixtures, including unmodified and heavily polymer modified mixtures, from the Superpave IDT, the Semi-Circular Bending Test (SCB) and the Three Point Bending Beam Test (3PB). Experimental analyses were complemented by a Digital Image Correlation System capable of providing dense and accurate full field strain estimations and thus suitable for describing the cracking behavior of the material at crack initiation. Tensile failure limits and the resulting fracture behavior in the three test configurations were predicted using a displacement discontinuity boundary element method with tessellations. Both experimental and numerical results indicate that fundamental HMA tensile failure properties can be determined from different test methods when appropriate analysis and interpretation methods are used. It was found that significant damage and first fracture occur locally in specimens associated with all test methods prior to peak loading, while analysis based on peak load and approximate analysis may lead to erroneous results.
