Abstract
This study evaluated the asphalt mixture performance of various gravel and crushed stone sources with different levels of crushed face counts as determined by ASTM D5821. Coarse aggregate angularity and texture were also evaluated with AASHTO T326 and aggregate imaging system testing. Each of the asphalt mixtures designed and tested used three asphalt binders: (a) neat PG 64-22, (b) polymer-modified PG 64-22 meeting New York State Department of Transportation (New York State DOT) elastic recovery specifications, and (c) polymer-modified PG 76-22. The permanent deformation properties of the different asphalt mixtures were measured with the asphalt pavement analyzer (AASHTO TP63) and the asphalt mixture performance tester (AASHTO TP79) with confining pressure applied to the specimens. Aggregate testing showed that ASTM D5821 correlated poorly with both AASHTO T326 and the aggregate imaging system. Situations also occurred in which aggregates had identical crushed counts but different levels of uncompacted void contents. The asphalt mixture performance tester that used confining pressure correlated well with the uncompacted voids content results of AASHTO T326. However, stresses applied in the asphalt pavement analyzer did not sufficiently mobilize the asphalt mixtures to allow differences in aggregate angularity to be clearly noted. Both asphalt tests were sensitive to asphalt binder high-temperature stiffness as determined by AASHTO TP70. Statistical analysis of the data resulted in a table that allows the New York State DOT to use aggregate angularity (as determined by AASHTO T326) and nonrecoverable creep compliance (as determined by AASHTO TP70) interchangeably to ensure that hot-mix asphalt (HMA) blends containing gravel aggregate perform as well as HMA blends containing crushed-stone aggregate.