The application of acoustic emission method for ultrasonic fatigue testing monitoring

The ultrasonic fatigue testing (USFT) is an effective method for rapid determination of the fatigue properties of structural materials under high cycle (≥106 cycles) loading. However, the occurrence and accumulation of fatigue damage with this test method remain uncertain due to the limitations of the existing measurement methods. Currently used monitoring methods allow detecting the fatigue cracks, but only in the late stages of failure. Despite the superior sensitivity to localized processes in materials, the use of the acoustic emission (AE) method in ultrasonic testing is extremely difficult due to the presence of resonant noise. This work aimed to suppress resonant noise and extract the signal for early detection of fatigue damage. The authors tested the samples of the AlSi9Cu3 aluminum alloy under the asymmetric cyclic loading (R=0.1) at a resonant frequency of 19.5 kHz with a non-threshold AE registration. The fracture surfaces were analyzed by electron and optical microscopy. The authors processed AE by two different methods: (1) the digital filtering method consisted of detecting resonant noise and removing it from the spectrum; (2) the φ-function method consisted of differentiating the spectrogram by time. The processed spectrograms were integrated by the frequency with further extraction of the AE events using the threshold method. The digital filtering method revealed a correlation between AE signals and fatigue damage, whereas the undamaged control sample showed no signals. The φ-function technique demonstrated ambiguous results, showing high AE activity on the control sample.