Evaluation of the staircase and the accelerated test methods for fatigue limit distributions

There are many existing test methods and data reduction techniques for stat istical properties of fatigue strength at a specific fatigue life, but none of them have been validated by experiment and simulation. Validation by si mulation is preferable because high cycle fatigue tests are usually very su bjective, not repeatable, and time consuming. This paper compares two quali tatively different approaches for evaluation of fatigue limit distributions . The analysis is based on computer simulations. The first approach contain s two methods of data reduction (by Dixon and Mood as well as that by Zhang and Kececioglu) based on simulated staircase test data. Another approach i ncludes two methods of extrapolation from shorter fatigue lives (the ray-pr ojected and the parallel-projected methods). The analysis in the first appr oach is based on the assumed target distribution of the fatigue limit, whil e the analysis in the second approach is dependent on the method of initial extrapolation of the target distribution to the shorter lives. Several sim ulation-based algorithms for the generation of fatigue limit data and low c ycle fatigue data were developed based on a given statistical distribution of the fatigue limit. The simulated test results were statistically evaluat ed against the given distributions with a target mean and two different coe fficients of variation of fatigue Limits (COV (S-FL)). It was concluded tha t for a regular COV (S-FL) (say 3.3%), the parallel-projected method is the best, followed by the Dixon and Mood approach. In an extremely large COV ( S-FL) (say 30%), the Dixon and Mood method is recommended. (C) 2001 Elsevie r Science Ltd. All rights reserved.