On the application of non-destructive testing techniques on rotating machinery

The diagnosis of artificial defects in a single stage gearbox using two non-destructive techniques (vibration and acoustic emission) and advanced signal processing techniques to discriminate between different load and defect states is the scope of the present study. Wavelet based techniques were developed and utilised in order to evaluate the vibration signals and extract diagnostic information out of them. A new concept of AE transducer mounting on rotating structures, without the use of the expensive solution of the slip-ring is presented. The AE signals are analysed and their root-mean-square (RMS) values are calculated. The capability of the new approach of AE acquisition in discriminating between different loading and damage states is shown and discussed. Interesting findings on the effect of the oil temperature upon AE recordings only speculated theoretically so far are also presented. Both methods yielded interesting results and showed an ability to distinguish between healthy and defected gears.