CONDITION MONITORING AND FAULT DIAGNOSIS OF SAW OF GIN MACHINE USING VIBRATION ANALYSIS

Abstract

Cotton ginning machines (saw gins) are critical in separating cotton fibers from seeds, and their saw assemblies are subject to wear, misalignment, and imbalance, which degrade performance and product quality. Early detection of these faults is essential for preventive maintenance and minimized downtime. This study investigates the application of vibration‐analysis‐based condition monitoring using accelerometer sensors, and signal processing methods including Fast Fourier Transform (FFT) and Wavelet Transform, to reliably diagnose imbalance, wear, and misalignment in saw gin saws. Experimental data were collected from a full‐scale industrial saw gin under various fault conditions. Key fault signatures are identified: imbalance produces dominant 1×RPM peaks; misalignment yields increased harmonics (particularly 2×RPM) and axial vibration; wear induces broadband noise and non‐stationary transient vibration components. Wavelet-based feature extraction improves the detection sensitivity in early‐stage faults compared to FFT alone. The results show that a combined approach can achieve fault classification accuracy above X% (to be filled with your results), enabling more effective maintenance scheduling. Implications for industrial implementation are discussed.