Modelling and diagnosis of faults in simple bevel gear train

Abstract

Gears are a fundamental component of spinning and rotating machinery that transmits power. Different loads and speeds can be applied to gear systems. Faults in gears can lead to breakdown of machines. Therefore, it is essential to keep an eye on the condition of the gears in order to avoid machine failure. Vibration, current and acoustic signature analysis are some of the fundamental methods for monitoring the state of different machineries. Modelling of a system (a virtual prototype) can be used to create enormous amounts of data with different system variables. It can replace the cost and labor involved in experimental analysis. In the present work, two models of a simple bevel gear train are taken into consideration to analyze the response of the system under wear and tear in time as well as frequency domain. The first model is a multi-body dynamics model, developed in CAD based multi-body simulation tool MBS ADAMS and the second model is a six degrees of freedom mathematical model developed in block diagram based Matlab-Simulink environment. In modelling, six different test cases of a simple gear train with healthy pinion, pinion with 20% broken teeth, 40% broken teeth, 60% broken teeth, 80% broken teeth on one tooth and one completely missing tooth are taken for consideration to study the system responses. Analysing the response of the systems, it can be inferred that, the faulty simple gear train (gears with greater amount of wear and tear) is having higher amplitude of vibration compared to that of healthy simple bevel gear train. The amplitude of vibration increases with the wear rate of tooth of the gear. In envelope spectrum, the amplitude of peaks at gear mesh frequencies is more for faulty gear train compared to the healthy bevel gear train which shows the manifestation of faults.