No abstract

No abstract

The fatigue of asphalt mixes constitutes one of the main types of distress of pavement structures. In laboratory, different procedures and testing equipments were implemented to describe the fatigue properties of asphalt mixes. Among them, the repeated indirect tensile test is used to determine the fatigue properties of asphalt mixes and also, their resilient modulus. Generally, fatigue tests are conducted using one of the two basic types of loading: controlled-strain or controlled-stress. Both loading modes produce fatigue prediction models relating the initial response (tensile strain or stress) of asphalt mixture to the fatigue life. In recent years, the application of more fundamental approaches has been suggested. One of these approaches is based on the principles of the Fracture Mechanics. The most popular relationship used in fracture mechanics is the Paris´ law. This paper presents an application of the Repeated Indirect Tensile Test to determine the fundamental parameters governing the crack propagation process in asphalt mixtures. The basic adopted hypotheses, the used data reduction techniques and the results obtained for an asphalt mixture considered as an example are presented, discussed and compared with other cited in the analysed literature.

Accelerated pavement testing (APT) is an effective testing procedure to evaluate asphalt pavements. With APT it is possible to determine and measure the structural response and pavement performance under a controlled, accelerated damage accumulation in a compressed period of time. However, different types of APT technologies can lead to different results. Full-size loading devices simulate road traffic accurately, but are expensive, while down-scaled size simulators are cost effective, nevertheless further away from reality. In this work, two types of APT mobile load simulators with different loading characteristics are compared with respect to pavement response in the field and in the laboratory. The MLS10 is a full-size simulator, whereas the MMLS3 is a one-third scale device. The relationship between the devices was studied in terms of the measured strains induced by both machines in the same pavement. Therefore, a testing field was instrumented with strain gauges and first trafficked with MLS10. Later, a slab of the instrumented pavement was cut off the road and tested in the laboratory with the smaller MMLS3. Furthermore, the structure of the pavement was modelled with a viscoelastic finite element method model and the moving loads of both machines were simulated considering size, speed and approximate footprints of their tires. As for the pavement materials, the properties of the different asphalt layers were determined in the laboratory. Experimentally acquired strain data were used to validate the models. Stress fields under different loading and environmental conditions were analysed and compared. The evaluation shows that the models can predict the pavement response under different loading conditions. However, they still need to be improved to increase the accuracy under different conditions. Further, the analysis of the strains show that both load simulators induce a different stress-strain situation and scaling of the pavement should be considered.

No abstract