The main objective of the study described in this paper was to evaluate different conventional and modified overlay materials for use in California. Twelve Heavy Vehicle Simulator (HVS) experiments were carried out and all were simulated using the computer program CalME. The purpose of the simulations was to enable “virtual” experiments to be conducted under exactly uniform conditions, and to enable extrapolation to other loading and climatic conditions. Six HVS experiments were done on sections where the overlays were placed on a two year old pavement which had had only light trafficking. This rutting experiment had uni-directional loading at elevated temperatures. For six other sections the overlays were placed on sections that had already been loaded to cracking in a previous HVS experiment. Reflection of cracks, in an existing pavement, through a new overlay is difficult to predict using Mechanistic-Empirical models. A simple model for calculating the strain at the tip of a crack was developed based on a large number of 2D and 3D finite element calculations. This response model was used in CalME, to determine damage to the overlay and to predict the appearance and propagation of visible cracks. Other models in CalME were from calibration studies on new pavements.

The overlay materials were dense graded asphalt concrete, gap graded asphalt rubber hot-mix, and several asphalt concrete materials with modified binders. The overlays were placed in thicknesses of 38 mm to 95 mm. The master curves of the asphalt materials were determined from frequency sweep tests in the laboratory and layer moduli were also backcalculated from Falling Weight Deflectometer (FWD) tests. The parameters of the fatigue damage models (for modulus reduction) were determined from beam fatigue tests at constant strain in the laboratory and the permanent deformation parameters from Repeated Simple Shear Tests at Constant Height (RSST-CH). The model parameters from the laboratory tests were used in CalME for simulation of the HVS experiments and calibration factors were determined from these simulations.

During the HVS tests temperatures were measured at different depths, the applied loads were recorded, and the resilient and permanent deformations were measured at several depths in the pavement, using Multi Depth Deflectometers (MDDs). A Road Surface Deflectometer (RSD) and a laser profilometer were used to record the resilient surface deflections and the permanent deformation profiles, respectively. Any surface cracking was also recorded. The results were imported into the CalME database and the experiments were simulated, hour by hour, using the incremental-recursive procedure. Care was taking to ensure that the calculated resilient deflections matched the measured deflections reasonably well during the full experiment

For the empirical model of permanent deformation a calibration factor was determined to relate the laboratory deformation to the in situ down rut. A relationship for prediction of reflection cracking was also developed. Values measured periodically during the full experiment were used in the calibration, not only the start and end points of the tests.