The use of high frequency stress waves for monitoring gears
The aim of this research is to investigate the feasibility of using stress waves for condition monitoring of gears. The project involved setting up an experimental rig, carrying out experimental work, acquiring stress waves signatures, and processing the signals. It has been shown that stress waves can successfully be employed for early detection of incipient gear failure. A experimental gearbox was employed during the experiments. Miniature ultrasound transducers, both sensitive and sufficiently small, were manufactured and installed on the stationary outer race of the rolling element bearing of the gearbox to detect stress waves from the meshing gears. The stress waves signals from the transducers were digitised and digitally processed to extract relevant information. The signatures were high-pass filtered at a cut-off frequency of 200 kHz, thus representing exclusive ultrasonic frequencies. A new statistical parameter, Energy Index, was developed and performed on the stress wave signatures which were segmented to represent individual gear teeth. Along with this new parameter, the classical statistical parameters, (Peaks, RMS, Standard Deviation, Kurtosis, etc.) were also performed. Conclusive results are presented in graphical form in terms of Cumulative Energy Indices' and Energy Indices in polar form for individual gear teeth. A new algorithrn was also developed and presented for the envelope detection of signal by iterative peak detection. Although no direct comparison was made between condition monitoring of gears using stress waves and methods such as low frequency vibration analysis and wear debris analysis, it is apparent that stress waves monitoring offers a much earlier warning of incipient gear failure because the technique can detect material defonnations which are precursors to changes in the dynamic properties of gears and the occurrence of wear debris. The technique, therefore, can predict incipient failure much earlier, extending the lead-time before failure, and as a result, minimising sudden failures which may have catastrophic consequences.