High-cycle fatigue life prediction issues in helicopter components

In the determination of fatigue life using the safe-life method, S-N curves are a key component. However, the process of generating S-N data usually involves significant testing time and cost. This leads to a dilemma in terms of assigning a sensible number of cycles to define a ¿runout¿ test duration for unfailed specimens. On the one hand, a longer duration is preferred, to obtain a more accurate measurement of the fatigue life. On the other hand, there are temptations to stop the test at a lower number of cycles for economy, and also for reliability, since short experiments are less likely to be disturbed by unexpected events, such as electrical fluctuations. This paper discusses the effect of the definition of run outs on the S-N curve and ultimately the effect on fatigue life predictions. These effects will be emphasized in helicopter context, in terms of economical fight and safety.