Investigation of parameters for fault detection of worm gear box using denoise vibration signature

In industrial applications, worm gearboxes are a key element. In a worm gearbox, as the material of a worm wheel is softer than that of a worm screw, the worm wheel gear is vulnerable to failure through various modes like pitting, wearing out, or tooth breakage during the sliding process. Due to this, it is essential to monitor the failure of the worm wheel gear of the worm gearbox, and it has gained importance for the diagnosis of faults in gearboxes. The present work focuses on the investigation of the effect of worm wheel tooth breakage, worm wheel bearing outer race, and varying load on vibration signature amplitude and frequency domain statistical features such as root mean square (RMS), crest factor, kurtosis, mean, peak to peak, skewness, sample variance, and standard deviation. The experimental setup is fabricated to conduct the experimental trials. An OR34 FFT analyzer with NVGate software is used to acquire the frequency domain vibration signature. Experimental results show that captured vibration signature amplitude for healthy worm wheel and bearing increased as fault occurred on the worm wheel, and bearing and frequency domain statistical features value changed with the change in fault location in the worm gearbox.