NDE Full-Field Vibration Analysis with High-Speed Video

High-speed cameras provide a remote contactless NDE technology which requires line-of-sight. It can be implemented for monitoring vibrations in machinery, piping, structural components. The analysis of the recorded vibrations can be used to detect mechanical problems and defects, as well as monitor structural health. This technology is also suitable in difficult to access locations or locations which require safety distances. It provides a full-field vibrational map within the field-of-view of the camera lens. Three NDE tests were carried out with the camera-based vibration measurement technique: (i) on concrete samples with embedded corroded and non-corroded rebar and concrete panels with pre-stressed tendons; (ii) on an industrial mixer; (iii) on steam generator tubes. The vibration data were extracted from the videos and analysed. For monitoring or investigating structural vibration of components such as the tested ones, frame rates in the order of several hundred per second are adequate to capture the frequency range of interest. The camera-based vibration measurements were carried out using an iPhone 13. The videos in this study were acquired at 240 frames per second using the "slo mo" mode at a resolution of 1080 pixels. In the concrete experiments, a rebound hammer was used to generate vibrations in the samples. For the industrial mixer, analysis was conducted on the vibration response of the mixer and its tubing connections to evaluate its dynamic behavior at various rotational speeds. Lastly, vibration analysis was carried out on a straight tube array built in-house to investigate the dynamic behavior of steam-generator tubes subjected to various flow conditions. The videos were processed with software MeScope VT440 for vibrational analysis. The amplitude measurements and the amplified motion and displacements are based on a created mesh on the sample surface with one distance measurement. The video processing creates an amplified apparent 2D motion of the individual pixels frame by frame. An in- house code developed on LabVIEW and previously verified was also used to process the recorded videos and extract the time-domain vibration amplitudes. The processed vibration data were found to be in very good agreement, demonstrating the potential of the contactless vibration measurement technique for structural monitoring or diagnostics.