Frequency Spectrum Filtering For Machinery Fault Diagnostics

Abstract This paper is a research effort to provide a stone in the building of machinery fault detection and diagnostics. It relies on using the spectrum analysis of the vibration signal using FFT. Depending on the expected faults of a specific machine the technique of bandpass filtering is applied on the frequency spectrum of the vibration signal generating bandpass spectra helping in easier fault diagnostics. Central frequencies and proper bandwidth are assigned to generate the new spectra of the defected machine. A case study of an industrial fan is used an application for the proposed technique.