Torsional vibration signal analysis as a diagnostic tool for planetary gear fault detection. (1st February 2018)

• Record Type: Journal Article
• Authors: Xue, Song; Howard, Ian
• Is Part Of: Mechanical systems and signal processing. Volume 100(2018)
• Page Start: 706
• Page End: 728
• Publication Date: 2018-02-01
• Subjects: Torsional vibration -- Planetary gear fault diagnosis -- Lumped-parameter model -- Finite element model -- Signal processing techniques; Structural dynamics -- Periodicals; Vibration -- Periodicals; Constructions -- Dynamique -- Périodiques; Vibration -- Périodiques; Structural dynamics; Vibration; Periodicals; 621
• DOI: 10.1016/j.ymssp.2017.07.038
• Languages: English
• ISSNs: 0888-3270

Abstract

Highlights: A 20 degree of freedom lumped mass coupled torsional and transverse planetary dynamic model simulated gear tooth damage. Four simulation studies are presented including healthy and damaged sun, planet and ring gear cases. Torsional vibration using phase domain averaging of planetary gear components successfully detected gear tooth faults. Carrier arm rotational averaging of torsional vibration shows sun, planet and ring gear teeth faults. Advanced signal processing methods and diagnostic indicators showed clear evidence of localized gear tooth damage. Abstract: This paper aims to investigate the effectiveness of using the torsional vibration signal as a diagnostic tool for planetary gearbox faults detection. The traditional approach for condition monitoring of the planetary gear uses a stationary transducer mounted on the ring gear casing to measure all the vibration data when the planet gears pass by with the rotation of the carrier arm. However, the time variant vibration transfer paths between the stationary transducer and the rotating planet gear modulate the resultant vibration spectra and make it complex. Torsional vibration signals are theoretically free from this modulation effect and therefore, it is expected to be much easier and more effective to diagnose planetary gear faults using the fault diagnostic information extracted from the torsional vibration. In this paper, a 20 degree of freedom planetary gear lumped-parameter model was developed to obtain the gear dynamic response. In the model, the gear mesh stiffness variations are the main internal vibration generation mechanism and the finite element models were developed for calculation of the sun-planet and ring-planet gear mesh stiffnesses. Gear faults on different components were created in the finite element models to calculate the resultant gear mesh stiffnesses, which were incorporated into the planetary gear model later on to obtain the faulted vibration signal. Some advanced signal processing techniques were utilized to analyses the fault diagnostic results from the torsional vibration. It was found that the planetary gear torsional vibration not only successfully detected the gear fault, but also had the potential to indicate the location of the gear fault. As a result, the planetary gear torsional vibration can be considered an effective alternative approach for planetary gear condition monitoring.