Prediction of Billet Fracture at Two-High Screw Rolling Piercing

Two-high screw piercing of steel 50 billets was performed at feed angles of 15 and 18 degrees. Piercing was stopped at 10–12 mm distance from the rear end of the billet. Axial fracture was registered at 15 degrees feed angle in front of the mandrel nose, while no fracture was observed at 18 degrees feed angle. Piercing for similar parameters was simulated using DEFORM finite element analysis software, and several criteria were used for fracture prediction. Ayada criterion was ineffective, as evidenced by the comparison with experimental data. The normalized Cockcroft–Latham criterion was effective for fracture prediction only in the axial zone. Results of this study showed that the most effective technique for fracture prediction is the construction of a dependence of “accumulated strain–stiffness coefficient of tension” along the trajectory of the point movement.