Labeling Algorithm for Outer-Race Faults in Bearings Based on Load Signal

Abstract

Rolling element bearings are essential components for the proper functioning of many types of rotating equipment. Diagnosing faults in bearings has traditionally been done using signal processing techniques inspired by physics, wherein acceleration signals are analyzed using time-frequency analysis methods. To study the effect of bearing damage on acceleration signals, experiments are typically performed aiming for a natural propagation of a spall. However, the extent of spall severity during the test remains uncertain. It is possible to disassemble and reassemble the bearing for visual inspection. Nevertheless, previous studies observed that the vibration signal would drastically change if this operation was conducted repeatedly, impacting the identification of trends in the acceleration signal. The objective of this study is to provide a method which can assist with labeling the spall size in endurance tests without the necessity of disassembling and reassembling the test rig. To address this issue, a new algorithm, based on the load cell signal was developed to assess the spall size using low-speed measurements. This algorithm enables the identification of the circumferential angle at which the rolling element interacts with the spall and is only carrying a partial load. The algorithm has been validated through visual inspections conducted during the experiment. This algorithm makes it possible to estimate the spall size without the need for visual inspection in subsequent experiments. A labeled endurance test contributes to a better understanding of spall propagation, such as the effect of speed, load, and material properties on the propagation speed. This study demonstrates how the load signal can be used for fault labeling with relatively simple and common techniques. This approach will enable the tackling of advanced and more complex problems in future endeavors, such as fault severity estimation and even prognosis.