Failure behaviour of strain and acceleration signals using various fatigue life models in time and frequency domains

This study aims to determine the appropriate frequency domain methods for durability predictions of lower suspension arm based on real-life random vibration data. In frequency domain, the power spectral densities (PSD) of the vibration signals were obtained and then analysed using three PSD cycle counting models, including the Lalanne, Dirlik and Narrowband methods. Correlation between fatigue lives predicted by time domain and frequency domain methods were evaluated to determine the appropriate frequency domain methods for durability prediction suspension lower control arm. Results show good correlation between fatigue life predicted with vibration fatigue and strain-life approaches. Therefore, it was deduced that road surfaces affect the fatigue life of the lower control arm, whereby the rougher the surface, the lower the fatigue life of the lower control arm. Lalanne and Dirlik PSD cycle counting algorithm were found to be the most suitable method for vibration fatigue analysis of lower suspension arm.