Beari̇ng fault detecti̇on under unbalance effect usi̇ng vi̇brati̇on analysi̇s

Bearing failures can be caused by a variety of reasons including inappropriate operating conditions, installation problems, and inadequate lubrication. The work presented in this thesis addresses the use of vibration analysis in the detection of failure occurring in a cylindrical roller bearing in the presence of unbalance. First of all, theoretical vibration simulations are carried out for a cylindrical roller bearing which has a local fault on its inner or outer race. In order to verify the correctness of the established numerical model, a bearing test rig is designed and a cylindrical roller bearing is used for the tests. A local fault is seeded artificially on the other race and the bearing is then tested under a constant working load, but the effect of unbalance is introduced. The resulting vibration signals are processed and analysed by means of Fourier transform and short time Fourier transform (STFT) in order to detect the presence of both fault and unbalance. It has been found that the presence of fault and the effect of unbalance are clearly revealed by the STFT when the results from the healty and faulty bearings are compared.