Welding is one of the most popular joining technologies in industry. Depending on the materials to be joined, the geometry of the parts and the number of parts to be joined, there is a wide variety of methods that can be used. These joining techniques share a common feature: the material in the weld zone experiences different thermo-mechanical history, resulting in significant variations in material microstructure and spatial heterogeneity in mechanical properties. To optimize the joining process, or to refine the design of welded structures, it is necessary to identify the local mechanical properties within the different regions of the weld. The development of full-field kinematic measurements (digital image correlation, speckle interferometry, etc.) helps to shed a new light on this problem. The large amount of experimental information attained with these methods makes it possible to visualize the spatial distribution of strain on the specimen surface. Full-field kinematic measurements provide more information regarding the spatial variations in material behaviour. As a consequence, it is now possible to quantify the spatial variations in mechanical properties within the weld region through a properly constructed inverse analysis procedure. High speed tensile tests have been performed on FSW aluminium welds. The test was performed on an MTS machine at a cross-head speed of up to 76 mm/s. Displacement fields were measured across the specimen by coupling digital image correlation with a high-speed camera (Phantom V7.1) taking 1000 frames per second. Then, through the use of the virtual fields method it is possible to retrieve the mechanical parameters of the different areas of the weld from the strain field and the loading. The elastic parameters (Young’s modulus and Poisson’s ratio) are supposed to be constant through the weld. Their identification was carried out using the virtual fields method in elasticity using the data of the early stage of the experiment. Assuming that the mechanical properties (elastic and plastic) of the weld are constant through the thickness, the plastic parameters were identified on small sections through the specimen, using a simple linear hardening model. This method leads to a discrete identification of the evolution of the mechanical properties through the weld. It allows the understanding of the slight variations of yield stress and hardening due to the complexity of the welding process.
Full-field modal analysis using high-speed 3D digital image correlation