A novel experimental research on vibration characteristics of the running high-speed motorized spindles

Research output: Contribution to journalArticle

Abstract

Because of the complicated running condition and many unpredicted factors such as unmodelled dynamics and external disturbances, the fault analysis of the high-speed motorized spindle is proved difficult. In this paper, a novel experimental method is proposed to research the vibration characteristics of the running high-speed motorized spindle. The method consists of four steps. Firstly, the vibration signal measurement and processing system are built according to data collector and signal analysis software, and then, the vibration signal of the spindle extracted from sophisticated experimental environment is studied by using harmonic wavelet transform for its advantages, such as ultra-narrow band, high resolution detection and ability of extracting weak signal. After that, local frequency domain zooming technique of harmonic wavelet packet is employed to study the vibration spectrum of the spindle at eight different rotating speeds. Furthermore, the axis orbit of the rotor is purified successfully, which can serve as verification basis for subsequent on-site fault diagnosis. Finally, a comparative analysis of eight different vibration signals under the same load but different speeds condition is carried out, the spectral components are classified into two types: the repeated ones and the regular ones. The accordance between experimental results and theoretical analysis suggests that the proposed method in this paper is effective.

Details

Authors
Organisations
Research areas and keywords

Subject classification (UKÄ) – MANDATORY

  • Materials Engineering

Keywords

  • High-speed motorized spindles, Vibration, Harmonic wavelet analysis, Fault diagnosis
Original languageEnglish
Pages (from-to)2245-2252
JournalJournal of Mechanical Science and Technology
Volume27
Issue number8
Publication statusPublished - 2013
Publication categoryResearch
Peer-reviewedYes