Discrete Wavelet Transform Based Measurement of Inner Race Defect Width in Taper Roller Bearing

Abstract

Local bearing defect monitoring and measurement has been a challenging area of research for profitable use of motion. A technique based on decomposition using Symlet wavelet is implemented for measuring inner race defect width in a taper roller bearing. It is difficult to estimate the defect width in inner race because defect moves continuously with respect to accelerometer. The bursts are selected from the signal for measuring the defect width on the basis of their amplitude. The degree of ambiguity in detecting the entry point in the groove gets reduced by using Symlet5 wavelet based decomposition which is almost linear phase nature and hence, sharpness in the signal gets maintained even in case of sudden change induced in raw signal. The technique has been successfully implemented for measuring defect width ranging from 0.4714 to 1.8145 mm. The measurement of defect width has also been verified by image analysis. Maximum difference in measurements has been found to be 6.68 % for the defect width of 0.4714 mm at no load condition and this gets reduced to 1.21 % with increase in load. This method has also potential in dimension measurement of surfaces having dissimilar and uneven surface characteristics.