PSI - Issue 17

Pedro J. Sousa et al. / Procedia Structural Integrity 17 (2019) 835–842 Author name / Structural Integrity Procedia 00 (2019) 000 – 000

839

5

4. Results

The analysis was performed for two regions of interest, Fig. 5. The inner region comprises the PCB mounted camera sensor and the outer region comprises the mechanical front-facing support. Due to the shadows in the interface between the two materials, there is no visible continuity between the inner and outer regions. As such, the relative absolute displacement cannot be accurately measured without developing a fringe-counting system. Nonetheless, it is possible to measure the relative rotation between these regions and obtain their separate behavior.

Fig. 5. Regions of interest of the specimen, in red

The deformed readings were acquired with the heat source turned off, in order to ensure that the turbulent hot air currents would not affect the measurements, when hitting the specimen. As the acquisition in all three axes is not simultaneous, there was a minor temperature difference between the first and last axis, starting at 39.4°C and ending at 38.3°C, which gives an average of 38.85°C and a maximum difference of 1.1°C. The specimen was also tested multiple times to ensure repeatable and reproducible results. From the images acquired in the reference and deformed stages, it is possible to obtain a phase map of the difference between them using the Carré method for each axis. As expected in ESPI, there is significant speckle noise in the data, which must be filtered before any further processing. Furthermore, as these results are always contained in the [0, 2  ] range, it is necessary to perform an unwrapping operation in order to obtain a continuous phase map. These procedures were performed following a Windowed Fourier Filter (WFF) approach and a quality guided (QG) phase unwrapping method developed by Kemao [13] – [16]. The results from these two stages are shown in Fig. 6 and Fig. 7, respectively.