Vibration analysis of nylon gear box utilizing statistical method

Gears are important element in a variety of industrial applications such as medical and automotive sectors. In the medical field it is used in surgical instruments, Biopsy instrumentation and Dental implant products. In automotive sectors like Electrical Throttle Controls, Steering Systems, Instrument Clusters, Window and Door Latch systems. Though the plastic gears have less heat dissipation, high wear resistance and self-lubrication properties, there may be defects in nylon gears in runtime applications. The defects include surface fatigue like pitting, scuffing, scratching and some common defects like flash, warping, bubbles etc. The gear defects are often identified under switch off condition and gear are checked. There are defect testing methods like image processing. In this work the defects were identified when the system is in running condition, by analysing the vibration signals obtained from the system under varying load and speed conditions. Gears are designed based on the parameters like module, pitch circle diameter, no of teeth and bending load capacity. In this work, the experimental setup is fabricated for testing the gears at three different sets of speeds. By varying the speed of transmission between two shafts both in loading and unloading conditions, the sound signals are acquired using sound sensor and sound signals are processed using MATLAB. The timedomain signal and frequency domain signals are used to detect the failure in nylon gear. Using the parameters like kurtosis, RMS, variance and standard deviation value the good gears and defective gears are compared. By using this method the defects in nylon gears can be identified easily. (C) 2019 Elsevier Ltd. All rights reserved.