Evaluation of Caltrans Test Method for Measuring Resistance of Compacted Bituminous Mixture to Moisture-Induced Damage (05-1976)**
Haiping Zhou, California Department of Transportation

This paper presents an evaluation of test results from two round robin tests conducted by the California Department of Transportation (Caltrans) in 2003 using the newly developed test for measuring resistance of compacted bituminous mixture to moisture induced damage (California Test 371) and the development of preliminary precision statements resulting from the round robin tests. The test is a modification of the AASHTO T-283-89 procedure using the recommendations of NCHRP Project 9-13. Twenty laboratories statewide participated in the tests. Thirteen of the laboratories participated twice. The material used in the round robin tests was dense-graded hot-mix asphalt. The loose mixtures were prepared by one laboratory and the same Rice value was provided to all laboratories. The evaluation includes examination of test data outliers, evaluation of means, precision, and test repeatability. The evaluation indicates the following: • The standard deviation of the tensile strength ratio (TSR) values from the two round robin tests falls within the ASTM D 4867 suggested between-laboratory precision criterion. However, the TSR range exceeds the ASTM D2s criterion. • More than half of the participating laboratories met the ASTM D 4867 within-laboratory precision criteria. • For evaluation of test repeatability, data from 11 participating laboratories indicate that nine laboratories achieved the same mean air void content and saturation for both round robin test series. Also, at least 4 of 11 laboratories achieved the same mean tensile strength for both round robin test series. Preliminary precision statements for inclusion in the newly developed test method were developed as below: • Within-Laboratory Precision o Single-operator standard deviation (1s) of tensile strength (unconditioned or conditioned samples) is 74 kPa o D2s range is 210 kPa • Between-Laboratory Precision o Multi-laboratory standard deviation of the TSR is 8% o D2s range is 23% The recommended precision statements developed as a part of this study may be refined as more test data are generated.