Evaluation of the prediction ability of ductile fracture criteria over a wide range of drawing conditions

In this study, we selected six ductile fracture (DF) criteria to assess their capability to predict material fracture when high-alloy steel was cold drawn at a wide range of reduction ratios and die semi-angles in a single-pass drawing process. A user-defined subroutine VUSDFLD for each DF criterion was coded in Fortran and integrated into ABAQUS. Twenty-eight drawing tests were performed using a combination of reduction ratio (r) range (10–49 %) and die semi-angle (α) range (4°–25°). Test results were compared with the results of finite element simulations. The results showed that, if three macroscopic variables (namely, stress triaxiality, maximum principal stress, and equivalent stress) and a cut-off value are combined in a DF criterion, the fracture of the material being drawn can be predicted at a sufficient level. This DF criterion showed a much higher prediction accuracy than the other five DF criteria, especially in the operating region (α < 10° and r > 40 %) where the producer of cold-drawn steel bar (or rod) products prefers to increase productivity and improve quality.