Effect of specimen size on fatigue behavior of asphalt mixture in laboratory fatigue tests

Abstract

Laboratory fatigue testing has been extensively used to estimate the resistance to fatigue cracking of asphalt mixtures. Researchers developed a number of test methods to estimate the fatigue behavior of an asphalt mixture. However, based on the classical fatigue analysis, fatigue lives obtained from different test devices are not comparable even when they are performed under the same mode of loading and environmental conditions for the same material. The differences in specimen geometry and load configuration will result in different stress-strain distributions inside the tested specimens leading to differences in local fatigue damage. In this paper, the size effect on the fatigue life is investigated. Uniaxial tension and compression (UT/C) fatigue tests were carried out using cylindrical specimens with three different sizes. Size effect on the fatigue test results are compared and analyzed using the partial healing (PH) material model, which describes the change of the complex modulus for a unit volume due to loading. The fatigue failure points determined by the classic approach, dissipated energy ratio and the PH model were compared. The results show that fatigue behavior of the specimens in the UT/C fatigue tests can be fitted excellently by the PH model, including estimation of the endurance limit. It is concluded that fatigue behavior from the UT/C fatigue test is not dramatically influenced by specimen size. © RILEM 2012.