JEE_V5_N4_RP2 Ashwani Kumar Chandel Raj Kumar Patel Journal on Electrical Engineering 2230 – 7176 5 4 11 17 Bearing, Condition, Faults, Induction Machine, Monitoring, Wavelet Transform Condition monitoring and fault diagnosis of electrical machinery are of great concern in industries. Pre fault detection in machinery can save millions of rupee required as emergency maintenance cost. In the present paper a methodology for the diagnosis of bearing faults in an induction machine has been given. The proposed discrete wavelet transform methodology extracts principal features from vibration signals caused by the faulty bearings and subsequently analyse these for determining the type of fault. The discrete wavelet transform is fed with the accelerometer signals and it is found that the impulses appear periodically with time period corresponding to defect frequencies. The results thus obtained reveal that the proposed pre-faults detection method is very efficient and effective. April - June 2012 Copyright © 2012 i-manager publications. All rights reserved. i-manager Publications http://www.imanagerpublications.com/Article.aspx?ArticleId=1858