Vibration signature analysis of two stage gearbox using kurtosis, skewness and crest factor

Abstract

In recent years, innovations and methodologies have been developed in mechanical equipment, mainly in rotating equipment, to increase the reliability of fault finding. World wide vibration indexes such as RMS, Kurtosis, etc. are widely accepted in industry and are also suggested by international standards. Even so, these parameters do not allow reliable diagnosis of the condition of the machinery. Their apparent simplicity of interpretation makes them more attractive. This work presents a discussion based on these traditional parameters about the diagnostic possibilities. The database used includes vibration signals of gears, taking into account different conditions of speeds and load. The results obtained show that these global parameters of vibration are limited in the exact diagnosis of fault, especially in the condition of initial faults. This method tries to acquire a baseline parameter that improves characterization of fault condition. The outcomes of velocity magnitude are compared with RMS, skewness and Kurtosis.