| Summary: | Rolling element bearings, an essential part of virtually all rotating machinery, fail unpredictably. Systems to measure bearing condition are therefore essential to avoid the financial and human consequences of bearing failure. An impulsive vibration condition monitoring technique known as shock pulse measurement has been used commercially for over thirty years. Its success is well documented yet it is based purely on empirical evidence. No attempt to understand the theory behind it was academically published until the work of Ho (1998). This research extends Ho's work to better understand the determination of rolling element bearing condition from the measurement of impulsive vibration. A rolling element bearing continually produces small impulsive shocks due to the interaction of roughness asperities on the rolling surfaces. A theoretical, stochastic model has been developed that accurately predicts the pattern of such pulses. It predicts patterns produced by various lubricant film thickness in good condition bearings, and those that will be produced by bearings with damage. It can consequently predict the deterioration of a bearing as film thickness decreases through to the onset of damage. A major contribution to the understanding of shock pulse measurement has been made in recognising that the asperity shock pulse patterns are different to the pattern of enveloped pulses produced by the resonant transducers and instrumentation typically used to measure them. Theory has been developed to successfully explain this difference in terms of asperity pulse shape. Appropriate compensation has been included in the model to adjust its output for the various shapes of pulse produced by different asperity interactions. Experiments covering a practical range of bearing sizes and operating conditions, for both bearings in good condition and bearings with damage, have been conducted to validate the model. Low cost, portable instrumentation and resonant transducers have been designed and manufactured to obtain the shock measurements. These devices are capable of being developed into cost effective commercial products with a minimum of additional effort. A better understanding has been gained of the characteristics of piezoelectric transducers when used for shock pulse measurement. This understanding allows transducers to be optimally designed for a given application. A rig and procedure have been developed to obtain the calibration values for resonant transducers used for impulsive vibration measurement. This calibration technique takes into account the pulse shape theory and can be developed into a commercial calibration process with a minimum of effort. Collaboration with a world leading manufacturer of tapered roller bearings, including full access to their existing resources, was a major contribution to the success of this research.
|
|---|
