Establishing correction solutions for Scanning Laser Doppler Vibrometer measurements affected by sensor head vibration

TL;DR

This paper develops and tests correction schemes for Laser Doppler Vibrometer measurements affected by sensor head vibrations, significantly improving measurement accuracy through sensor placement and theoretical analysis.

Contribution

It introduces practical correction methods for SLDV vibrations, including sensor placement strategies and a theoretical framework for vibration correction.

Findings

Correction schemes reduced error by up to 27 dB.

Single accelerometer option lowered RMS phase error by 30%.

Theoretical analysis supports sensor placement and correction effectiveness.

Abstract

Scanning Laser Doppler Vibrometer (SLDV) measurements are affected by sensor head vibrations as if they are vibrations of the target surface itself. This paper presents practical correction schemes to solve this important problem. The study begins with a theoretical analysis, for arbitrary vibration and any scanning configuration, which shows that the only measurement required is of the vibration velocity at the incident point on the final steering mirror in the direction of the outgoing laser beam and this underpins the two correction options investigated. Correction sensor location is critical; the first scheme uses an accelerometer pair located on the SLDV front panel, either side of the emitted laser beam, while the second uses a single accelerometer located along the optical axis behind the final steering mirror. Initial experiments with a vibrating sensor head and stationary target confirmed the sensitivity to sensor head vibration together with the effectiveness of the correction schemes which reduced overall error by 17 dB (accelerometer pair) and 27 dB (single accelerometer). In extensive further tests with both sensor head and target vibration, conducted across a range of scan angles, the correction schemes reduced error by typically 14 dB (accelerometer pair) and 20 dB (single accelerometer). RMS phase error was also up to 30% lower for the single accelerometer option, confirming it as the preferred option. The theory suggests a geometrical weighting of the correction measurements and this provides a small additional improvement. Since the direction of the outgoing laser beam and its incident point on the final steering mirror both change as the mirrors scan the laser beam, the use of fixed axis correction transducers mounted in fixed locations makes the correction imperfect. The associated errors are estimated and expected to be generally small, and the theoretical basis...