Statistical characteristics and logarithmic EWMA control method of bearing raceway's surface roughness by superfinishing

Abstract

The bearing raceway is the main working surface of the bearing and good surface integrity is required. Since superfinishing is an important method for achieving high surface quality, it is usually used as the last processing to play a very essential part in improving the surface integrity of bearing raceway. In this paper, we proposed a superfinishing surface quality control method, by which we could determine whether there were systematic factors that caused surface quality anomalies by measuring and analysing the roughness data without directly measuring process parameters and environmental indicators. Firstly, we verified through experiment that the surface roughness by superfinishing did not obey the normal distribution but obeyed the logarithmic normal distribution. We interpreted this statistic characteristics by the mechanism of the relationship between roughness and random factors in the superfinishing. Then, according to above findings, and considering the possible trend of surface roughness during continuous processing, we designed a logarithmic exponentially weighted moving average (EWMA) method to test the surface roughness and judge whether it was under control. Finally, the effectiveness of this surface quality control method in superfinishing was verified by applying it to the actual bearing raceway outer ring machining.