PSI - Issue 28

Mohamed Ali Bouaziz et al. / Procedia Structural Integrity 28 (2020) 393–402 M.A. BOUAZIZ et al/ Structural Integrity Procedia 00 (2019) 000–000

395

3

(a)

(b)

Fig. 1. Sample geometry and raster orientations. (a) µ-SENT sample dimensions (mm) ; (b) +/- 45 ° deposition.

The experimental setup consisted of a numerical Keyence VHX-1000 microscope for the surface observation, a tensile micromachine and a triggering system. The later allowed images to be recorded at specific rates when the specimen was continuously loaded and each image to be related to the corresponding applied load. Since the notch was welded due to the temperature of filament deposition and the dimension of the notch (i.e., 150 µm in width), the magnification power was set to the lower lens range value (100×) in order to monitor a wide enough region to contain the notch opening. When this opening was detected, a zoom (2×) in the vicinity of the notch root was operated with no interruption of the test. This second part of the experiment is analyzed hereafter. DIC provides displacement fields (Sutton, M. A.; Orteu, J. J.; Schreier 2009). From an image of the surface in reference and deformed states, displacement vectors were retrieved by registering subsets in the original image within the deformed image. DIC resolutions are now sufficient to analyze experiments performed at various scales (Forquin et al. 2004; Sutton et al. 1999). In the present study, Ncorr (Blaber, Adair, and Antoniou 2015) was used to analyze the microscopic images obtained during the test (Marae-Djouda et al.2020). 3. Crack growth measurement To measure crack growth, a precise estimation of the crack tip position was needed. An adaptation of the method proposed by (Mathieu et al. 2013) was carried out to locate the crack tip position at the surface of the printed material. This method is based on an FE model to compute displacement fields over the considered surface. DIC results are taken as boundary conditions. The main idea of the method is that each node on the crack path can be chosen as crack tip but only one node gives the minimum value of the least root mean squared displacement gap between measured and computed values, and it is the best approximation for crack the tip location. The method is schematically shown in Figure 2, where the outer contour of DIC measurements, represented in green, are prescribed to the FE model, and internal nodes are used for comparison with measurements. Nodes on the crack path (red dashed line) are tested to be crack tip positions. The nodal displacement difference between DIC and FE analyses was computed, and the displacement gap consisted in the root-mean-square difference           m n m comp m c meas m m c n 1 2 2 , 1 u x u x x x  (1) where x c is the considered crack tip node, n m the number of measurements located at x m , u meas the measured displacement, and u comp computed displacement.