Effect of Specimen Fabrication Variables on the Measured Mechanical Properties of Asphalt Mixtures Tested in the Asphalt Mixture Performance Tester (AMPT)

The scope of this research involves the evaluation of the effects of specimen preparation variables-mixing temperature, binder time at mixing temperature, mixer type, mixing time, loose mix conditioning temperature, loose mix conditioning depth, loose mix stirring, mold loading, placement in mold, additional time at compaction temperature before loading in mold, gyratory specimen height, and test specimen air voids-on tests such as the Dynamic Modulus and Flow Number conducted using the Asphalt Mixture Performance Tester (AMPT). The initial testing program to evaluate the precision of the Dynamic Modulus and Flow Number tests (described in National Cooperative Highway Research Program Report 702) indicated that "specimen fabrication is a major component of the between-laboratory variability" for both tests. The expected outcome of this study was that AMPT users-Federal Highway Administration (FHWA), state user agencies, and industry-understand the importance of proper specimen preparation and the effect of specimen preparation variables on the results obtained from the AMPT Dynamic Modulus and Flow Number tests. The use of AMPT Dynamic Modulus at 20 degrees C is the most sensitive portion of the modulus-time curve and can provide a general idea of sample preparation impact on other mixture tests. This information comes at critical timing as the United States begins to implement performance-engineered mixture designs, balanced mixture designs, and performance-based pay factors. Based on the findings of the study, recommendations may be made regarding changes to specimen preparation procedures, which are expected to produce more reproducible test results.