A METHOD FOR CONDITION MONITORING AND FAULT DIAGNOSIS OF SAW GIN MACHINES USING VIBRATION SIGNALS

Abstract

In this study, we propose a new diagnostic method for monitoring the technical condition of saw gin machines using vibration signal analysis. Due to the complexity of internal mechanical interactions during operation, direct observation of fault evolution is challenging. Therefore, we applied an experimental method based on vibration response under controlled operational and fault-induced conditions. Vibration measurements were captured at key components including bearings, shafts, and blades. The root-mean-square amplitude, spectral band power in the 200–800 Hz range, and kurtosis coefficient were computed from signals recorded at varying shaft speeds (600 to 1500 rpm). The observed increase in RMS amplitude from 0.12 g (healthy) to 0.29 g (faulty) indicated sensitivity to early-stage faults. Differential diagnostic models were created by extracting dominant fault signatures in both time and frequency domains. Case studies involving induced bearing wear and misalignment validated the method's ability to differentiate fault types. Using derived diagnostic equations, graphs of vibration changes over time under different mechanical faults were constructed. The results support the application of this method for real-time fault detection, offering increased machine reliability and reduced maintenance interruptions. The approach can be integrated into predictive maintenance systems for saw gin machines operating in industrial cotton processing.