Identification of defects in thrust ball bearings in the presence of external vibrations

Slow speed bearing is the most critical component for the rotary machinery running at lower speeds. So there is need for the vibration monitoring of bearing to detect fault at early stage. Examples of bearings used in slow speed machinery are paper mills, wind turbine power plant, air preheaters and conveyor belts used in mining industries. Defect detection in slow speed bearing is difficult since high noise to signal ratio and also weak bearing defect signal. At such low speeds the energy released from bearing defects is small and coupled with surrounding noise. Identification of the signals of defective slow speed thrust ball bearing of mechanical components in the presence of external vibration is a difficult task. Therefore, this research work has been carried out to improve the defect detection of slow speed thrust ball bearing in presence of external vibration using self-adaptive noise cancellation (SANC). Circular defect (diameter = 400µm) is present on the inner race of the bearing and external vibrations have been imparted to the defective bearing with the help of running compressor. It has been noticed that the defective bearing signal to noise ratio has been significantly improved after the implementation of SANC. The noise has been suppressed and the defect frequencies are clearly visible in the vibration spectrum. In this report in SANC technique the least mean square (LMS) algorithm has been used.