Mode I Fracture at Spot Welds in Dual-Phase Steel: An Application of Reverse Digital Image Correlation

Abstract

Strain fields in 600 grade dual-phase steel V-notch tensile specimens, both with and without a spot weld, were measured after mode I fracture initiation. Starting with the final image of a fully developed crack, a novel reverse digital image correlation (DIC) analysis was used to determine the path that the crack followed at the onset of fracture as well as the crack tip deformation field. This gave the pixel coordinates of grid points on both sides (i.e. fracture surfaces) of the crack path in the undeformed image. Strain fields that develop in the base material regions surrounding the two fracture surfaces were subsequently measured with forward DIC analysis. Steady state values of the crack tip opening displacement (CTOD) and crack tip opening angle (CTOA), which are important fracture parameters, were measured for the base DP600 metal. Notch tip opening displacement (NTOD) and notch tip opening angle were also measured. It was found that steady state values of the CTOD and CTOA are reached within 2 mm or so of crack growth following completion of the flat-to-slant transition of the fracture surface and stabilization of the crack tunneling effect.

Formulation of Lucas–Kanade Digital Image Correlation Algorithms for Non‐contact Deformation Measurements: A Review

“…DIC algorithms other than the forward additive type may be more advantageous in some deformation measurement applications. For example, a backward additive DIC algorithm is used by Tong [26] and Tao et al [27] to analyse the deformation field around a growing crack. The obvious advantage of such an approach is to avoid the use of a subset with a discontinuous displacement field as the subset based on the current image will have the cracked path and the free boundaries clearly defined.…”

“…It turns out that the DIC algorithms most often used in experimental mechanics as described above are more or less equivalent to the sparse optical flow version of the original so‐called forward additive Lucas–Kanade algorithm (which also accounts for affine deformation and photometric gain and bias by Lucas and Kanade ). A backward additive Lucas–Kanade DIC algorithm has only occasionally been used for analysing image sequences with growing cracks and has more recently been shown to have no noise‐induced bias when the initial (reference) image is noise‐free . Many other formulations of the DIC algorithms have rarely been evaluated at all for possible materials testing and structural monitoring applications.…”