ABSTRACT
ABSTRACT: During a fatigue test, damage is accumulating and the characteristic properties for the material are changing. Typically, this is expressed as a change in modulus, and when this has reached some critical value, the test is terminated. This has resulted in different definitions of failure for the two types of test, controlled stress and strain, and results in different definitions of failure and consequently different fatigue relationships. The use of dissipated energy concepts has been used as a method to bring the results of controlled stress and strain testing closer together. However, some of the analysis conducted with dissipated energy has not considered the way a specimen damages during a test. The flexural fatigue life of asphalt mixes is currently assessed by the AASHTO T321 procedure in the USA. Meanwhile, the ASTM D 7460 test method offers an alternate analysis procedure that does not rely upon a curve fit to determine the point of failure, but rather considers a maximum value in the n·S curve as the definition of cracking. The n·S curve is based upon an understanding of how energy is dissipated during the test. Recently, some mixtures with asphalt binders that contain approximately 10 to 15% polymer (defined as heavily modified in this paper) have been evaluated in these test devices. The data demonstrates that crack initiation occurs beyond the 50% stiffness reduction, which in AASHTO T321 is used as the value to terminate the test, with the assumption that all necessary data has been collected to determine the specimen’s fatigue life. The origins of both methods are discussed and data is presented that demonstrates the fatigue process that occurs in different modes of testing. In addition, analysis is presented of heavily modified mixes and these are contrasted to those made with conventional binders. Recommendations are made with regard to the interpretation of data from the various two methods.
