Loading…
In-situ adaptive calibration of incident signal of ultrasonic lubricant film thickness measurements
In ultrasonic reflection measurements of lubricant film thickness, an incident wave is needed as a reference signal and is typically calibrated under controlled ex-situ conditions. However, existing calibration methods have two limitations: 1) significant measurement errors are introduced due to the...
Saved in:
Published in: | Mechanical systems and signal processing 2025-02, Vol.224, Article 112236 |
---|---|
Main Authors: | , , , , , |
Format: | Article |
Language: | English |
Subjects: | |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | In ultrasonic reflection measurements of lubricant film thickness, an incident wave is needed as a reference signal and is typically calibrated under controlled ex-situ conditions. However, existing calibration methods have two limitations: 1) significant measurement errors are introduced due to the discrepancy between the calibration environment and the varying operational conditions (for example, temperature), and 2) disassembly of tribo-pairs is necessary. To address these issues, a novel method that enables in-situ adaptive calibration of incident signals is proposed. Particularly, it is found that the variation of the reflected signal against lubricant film thickness can well fit a circle in the polar coordinate, where the desirable incident signals can be extracted using the size and the location of the established circle. Importantly, the circle is iteratively updated to adapt to the variation of the working environment. This novel approach of calibrating incident signals, named the “circle-fitting method”, does not require knowledge of the acoustic properties of tribo-pairs and is thus temperature independent. For experimental validation, two test setups of a micrometer-driven calibration rig and a tilting pad thrust bearing platform have been used separately to implement ultrasound measurements of lubricant film thickness, demonstrating accurate results as evidenced by error analysis. Specifically, in the experiment with the micrometer-driven calibration rig, the measurements yielded an amplitude estimation error of 0.2% and a phase estimation error of 0.66%. |
---|---|
ISSN: | 0888-3270 |
DOI: | 10.1016/j.ymssp.2024.112236 |