Rotating blade vibration analysis using photogrammetry and tracking laser Doppler vibrometry

Online structural dynamic analysis of turbomachinery blades is conventionally done using contact techniques such as strain gauges for the collection of data. To transfer the captured data from the sensor to the data logging system, installation of telemetry systems is required. This is usually complicated, time consuming and may introduce electrical noise into the data. In addition, contact techniques are intrusive by definition and can introduce significant local mass loading. This affects the integrity of the captured measurements. Advances in technology now allow for the use of optical non-contact methods to analyse the dynamics of rotating structures. These include photogrammetry and tracking laser Doppler vibrometry (TLDV). Various investigations to establish the integrity of photogrammetry measurements for rotating structures involved a comparison to data captured using accelerometers. Discrepancies that were noticed were attributed to the intrusive nature of the contact measurement technique. As an extended investigation, the presented work focuses on the validation of photogrammetry applied to online turbo machinery blade measurements, using TLDV measurements. Through a frequency based characterisation approach of the dynamics of the two scanning mirrors inside the scanning head of a scanning laser Doppler vibrometer (SLDV), TLDV is employed in developing a system that can be used to achieve a perfect circular scan with a Polytec SLDV, (PSV 300). Photogrammetry out-of-plane displacements of a laser dot focused on a specific point on a rotating blade are compared to displacements captured by the laser scanning system. It is shown that there is good correlation between the two measurement techniques when applied to rotating structures, both in the time and frequency domains. The presence of slight discrepancies between the two techniques after elimination of accelerometer based errors illustrated that the optical system noise floor of photogrammetry does contribute to inconsistencies between photogrammetry and other measurement techniques. (C) 2016 Elsevier Ltd. All rights reserved.