Degree Name

Doctor of Philosophy


School of Mechanical, Materials and Mechatronic Engineering, Faculty of Engineering


The condition monitoring of slow speed roller bearings has been investigated. A test-rig was designed and constructed to enable detailed measurement of horizontal rotating bearing acceleration forces in both the axial and radial plane in the speed range of 0.5 to 10 revolutions per minute. These accelerations were carried out at both sonic and ultrasonic sampling rates to establish which technique is the most appropriate. Overall bearing displacement and surface temperatures were measured. Strains generated in the stress frame by the loading of the bearing were monitored along with the power used to drive the test-rig. Measurements were obtained from two full-size slew bearings operating in Bluescope Steel Limited. One bearing operated at 4.3 rpm continuously in the vertical plane. The other slew bearing operated intermittently and with partial rotation at approximately 1 rpm in the horizontal plane. During this project, the concepts of Symmetry and Stability have been developed as a fundamental approach to information analysis. A considerable number of novel signal processing methods including; Kurtosis/Correlation dimension plots, Symmetry State Space (SSS), Symmetric Wave Decomposition (SWD), Compressed Eigenvector Deconvolution Spectral Analysis (CEDSA), Ring Matrix Fault Values (RMFV) have been developed. These methods all utilize symmetry, antisymmetry, symmetry ’breaking’, stability and enable the assessment of which sensor methodology combination is best for the situation considered. It will be shown, among other things, that ultrasonic measurements using sensors designed for Acoustic Emission (AE) permit an implementation of an early warning system for slow speed bearings. This will enable the operator to carry the minimum inventory in bearings and to plan shut downs without incurring additional costs from unplanned outages resulting from failed bearings.

02Whole.pdf (7896 kB)