The behaviour of a longitudinal V-shaped crack, on the surface of a continuously cast steel slab, is studied during hot rolling. The analysis is carried out by means of the commercial FE-code LS-DYNA3D. Process parameters obtained from industry are used as a reference. The slab of initial width 1000 mm and thickness 220 mm is rolled down to 30 mm. It is assumed that the material can be treated as rigid-perfectly plastic and that the cracks do not propagate. The latter assumption is in agreement with industrial observations for a steel grade similar to that analysed here. The aim of the study is to investigate the possibility of controlling the plastic deformation so that the cracks disappear or so that their deteriorating effects are minimised. The analysis is focused upon the influence of friction, roll radius and rolling schedule on the change in the shape of a crack of initial depth 20 mm and a crack angle of 6°. The reliability of the simulations is checked by pilot-plant experiments using aluminium as the model material for steel. The results indicate that it is not possible to prevent the bottom side surfaces of the crack from coming into contact, especially not for small reductions/pass and small roll radii. The influence of friction was found to be marginal. Contact between the crack surfaces is found already at the beginning of the rolling, as the V-shaped crack is being changed to Y-shape. Considering the upper part of the crack, this remained open during the majority of the schedules studied. However for heavy reduction/pass and a large roll radius, this part of the crack is closed also, but not before the final passes. If the bottom side surfaces of the crack are in complete contact, they are prevented from further oxidisation. It is assumed that for such conditions the deteriorating influence of the bottom part of the defect decreases during subsequent rolling. During the elongation of the workpiece, the oxide flake of the Y-crack bottom is broken into splinters with oxide free material in between, making the creation of a high performance weld possible. Provided that this supposition is correct, the best results should be obtained for light reductions/pass at the beginning of the rolling, resulting in an early closure of the crack bottom, followed by heavy reductions/pass enabling also the closure and oxide splintering of the upper part of the folded crack.
Read full abstract