Author(s)

B. R. Rabb

Abstract

The aim of the fatigue analysis is to obtain a reliable design using the mean value of the fatigue limit reduced to the appropriate probability of failure as the allowed stress amplitude. This reduction requires a thorough knowledge of both the population mean and the population standard deviation. Furthermore, the correct probability distribution of the fatigue limit must be anticipated. Fatigue testing is required to obtain reliable estimates of the fatigue limit and standard deviation. The staircase or up-and-down test method is usually used. The obtained sample values must be transferred to population values by applying the rules of statistics, i.e. by applying confidence levels. However, due to the nature of the staircase test, it is very difficult to obtain a correct value of the sample standard deviation. This can be demonstrated by deriving the cor

Keywords