Effects of loading and sample geometry on acoustic emission generation during fatigue crack growth: Implications for structural health monitoring

Abstract The reliability of traditional non-destructive methods for crack detection is well understood and characterised using Probability of Detection (POD) curves. Structural Health Monitoring (SHM) techniques in contrast remain largely unquantified. The performance of the Acoustic Emission (AE) technique for damage detection and location in potential SHM applications is underpinned by the intensity of AE signal generation from the damage site. In this paper, factors influencing the rates of emission of Acoustic Emission (AE) signals from propagating fatigue cracks were investigated. Fatigue cracks were grown in specimens made from 2014 T6 aluminium sheet while observing the effects of changes in crack length, loading spectrum and sample geometry on rates of acoustic emission. Significant variation was found in the rates of AE signal generation during crack progression from initiation to final failure with a number of distinct phases identified in that progression implying different failure mechanisms operating at particular stages in the failure process. A new ‘probability of hit’ method for quantifying crack detecting capability using AE is also presented.