Analysis of acoustic signals from rotating machines for wear detection

Abstract

Automated Condition Monitoring (ACM) of rotating machines is necessary to avoid sudden machine breakdown due to wear and tear of Rolling Element Bearings (REBs). Hence, detecting faults in REBs well before machine outage is important. The aim of the proposed system is to develop a prototype to detect mechanical wear, by capturing acoustic signals with frequencies between 20 Hz to 20 kHz, from the machine under test, using non-contact electret microphone sensor. The frequency spectra of the acoustic signal from a healthy bearing is compared with the acquired acoustic signal. This system aids in scheduled replacement of faulty parts, thereby reducing the machine downtime, increasing reliability and customer satisfaction, but the results do not infer on the severity of the fault present in the REBs. The system also helps in predicting Remaining Useful Life (RUL) of the machine under consideration.