Condition monitoring of gearboxes using acoustic emission

Acoustic emission (AE) is one of many technologies for health monitoring and diagnosis of rotating machines such as gearboxes. Although significant research has been undertaken in understanding the potential of AE in monitoring gearboxes this has been solely applied to spur gears and slow speed roller bearings. This research presents an experimental investigation that assesses the effectiveness of both AE and vibration technologies in identifying various types of defects on in a helical gearbox; the first known attempt. Furthermore, the application of advanced signals processing techniques such as Spectral kurtosis (SK) and wavelet analysis were studied on AE and vibration signatures. It is shown that the application of advanced signal processing methods is particularly necessary for monitoring helical gears. The application of SK and wavelet analysis was found to be effective in denoising the acquired signals. The first chapter of this thesis is an introduction to this research and briefly explains motivation and theoretical background supporting this research. The second chapter summaries the relevant literature to establish the current level of the knowledge in this field. The third chapter describes methodologies and experimental arrangement utilized for this investigation. Chapter 4 discusses helical gear diagnosis for both natural and seeded surface defect. Chapter 5 reports on an experimental investigation in which several technologies such as AE, vibration and motor current signature analysis, were applied to identify the presence of a naturally fatigued pinion shaft in an operating gearbox. Chapter 6 details an investigation which compared the applicability of AE and vibration technologies in monitoring a naturally degraded roller bearing. It has been concluded that AE is a strong diagnostic tool for early diagnosis of bearings faults. However, the application of condition monitoring for helical gear diagnosis was fraught with some degree of complexity as compared to spur gears. This implies that condition monitoring of the gears using AET can be challenging. On the contrary, the applicability of AET for bearing diagnosis was promising and it offered an absolute advantage over the conventional vibrationdiagnosis. Furthermore, the application advanced signals processing methods such as Spectral Kurtosis and wavelet was found to be promising in denoise the recorded AE signals. It was also concluded that the use of different signal processing methods is often necessary to achieve meaningful diagnostic information from the signals.