Ball bearing damage detection using traditional signal processing algorithms

Abstract

Fault detection and diagnosis of ball bearings has always been a challenge when monitoring rotating machinery. Specifically, bearing diagnostics have seen extensive research in the field of fault detection and diagnosis. This article reviews traditional algorithms used to detect and diagnose faulty bearings in heavy-duty milling machine tool spindle heads. Different kinds of faults have been created deliberately on the bearings of a test spindle head. The prediction effectiveness of several detection methods are tested when faults are in different stages of development.