Evaluating Laboratory Tests for Use in Specifications for Unbound Base Course Materials

Abstract

A key requirement for performance-based specification for pavement materials is the selection of laboratory tests that can predict their response under conditions similar to those encountered in the field. This paper presents the results of a study that was conducted to evaluate the capability of different laboratory tests to predict the behavior of unbound granular base course materials under cyclic traffic loading. To achieve this objective, a laboratory testing program was conducted on three types of unbound granular base course materials, namely limestone, sandstone, and granite. The laboratory testing program included conducting a static triaxial test as well as three types of repeated load triaxial (RLT) tests on the materials considered at their optimum field compaction conditions. The three types of RLT tests considered included: resilient modulus, single-stage, and multistage RLT tests. The results of the static triaxial tests showed that the considered materials had similar shear strength properties. However, the RLT test results showed a distinct behavior between those materials, such that the limestone and sandstone exhibited the highest and lowest resilient modulus values, respectively, in the resilient modulus test. In addition, the granite and sandstone accumulated the highest and lowest permanent strain, respectively, in both single-stage and multistage RLT tests. The multistage showed that the granite and limestone experienced the unstable collapse behavior at higher cyclic deviatoric stress than sandstone. In addition, the results demonstrated that the resilient strain behavior was distinct from that of the permanent strain, which indicates that the resilient modulus test cannot be solely used to evaluate the performance of base course materials under cyclic traffic loading. Finally, this study demonstrated that the multistage RLT test provides an important tool to characterize the long-term behavior of a base course material at varying stress conditions. Therefore, it is recommended to use this test in future performance-based specification.