Detection of faulty accelerometer mounting from response measurements

Data integrity has become more critical with the growth in remote semi-automated machine health monitoring. The mounted resonance frequency of an accelerometer, if correctly mounted on a flat surface, is normally quite repeatable, for example at different locations on the same object. A method has r...

Full description

Saved in:
Bibliographic Details
Published in:Journal of sound and vibration 2020-07, Vol.477, p.115318, Article 115318
Main Authors: Randall, R.B., Smith, W.A.
Format: Article
Language:English
Subjects:
Accelerometers
Cepstral analysis
Data integrity
Fault diagnosis
Faulty accelerometer mounting
Faulty sensor readings
Flat surfaces
Frequencies
Frequency ranges
Industrial applications
Modal analysis
Mounted resonance frequency
Mounting
Remote monitoring
Resonance
Resonance changes in structure and sensor system
Resonant frequencies
Response spectra
Separation of forcing and transfer functions
Signal processing
Transducers
Transfer functions
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Data integrity has become more critical with the growth in remote semi-automated machine health monitoring. The mounted resonance frequency of an accelerometer, if correctly mounted on a flat surface, is normally quite repeatable, for example at different locations on the same object. A method has recently been developed which largely separates forcing and transfer functions in response signals, by applying an exponential window in the cepstrum. This technique can show the resonance peaks in response spectra at all measurement points, and has been applied successfully in operational modal analysis. This paper proposes how this technique can be used to detect a faultily mounted accelerometer in industrial applications. Changes in the natural frequencies of the structure should show up equally in all measurements, but deterioration in the mounting condition of an individual accelerometer would only affect the mounted resonance frequency for that transducer. This paper demonstrates these principles using measurements on a gear test rig, operating with a range of different speed profiles, with two similar accelerometers, one having correct and then faulty mounting of two different severities. Measurements are also compared for accelerometers with a wider frequency range.
ISSN:0022-460X
1095-8568
DOI:10.1016/j.jsv.2020.115318