Comparing Results between the Repeated Load Triaxial Test and Accelerated Pavement Test on Unbound Aggregate

Abstract

Unbound granular materials (UGM) are extensively used as basecourse materials around the world as they are capable of bearing relatively high traffic loads and are an economical option in comparison to bound materials. The unbound materials performance as basecourse determines the life-cycle costs of a pavement. The extent to which the repeated load triaxial test can predict the performance of unbound granular materials in the laboratory is an important parameter for road designers. Moreover, the performance of the unbound basecourse materials depends upon the moisture conditions when they are being loaded, gradation curve of the material, in situ density, permeability, and the nature of the aggregate fines (clays). There is a need to find the factors that cause the variation in the performance of the materials both in the laboratory and in-field pavement conditions to enable appropriate selection and use. This research utilizes accelerated pavement tests (APT) alongside repeated load triaxial (RLT) tests to test differently graded unbound granular materials at higher moisture contents. The objective of the research is to find the similarities and contrasts in basecourse material in both tests and to find the root causes of variation in the aggregate performance. If results of the RLT test can truly represent the results from full-scale APTs for basecourse materials, then the pavement materials industry can use the much lower costing RLT tests with confidence knowing it is representative of performance tests from APTs. The accelerated tests are performed on test pavements at the Canterbury Accelerated Pavement Testing Indoor Facility (CAPTIF) with the introduction of surface water, which replicates typical rainfall events. An increase in moisture in the basecourse layers is observed during the APTs due to entry of surface water through the surface seal, which allowed the performance of the basecourse material at higher moisture contents to be assessed. The RLT tests were conducted in parallel with the same basecourse materials at higher moisture conditions. The relative performance ranking of APTs and RLT tests is found to be the same for some materials; however, it differs in some cases. The similarities and differences in the rankings from both procedures are highlighted, and the causes of these similarities and differences are inferred and discussed.