Roughness Transfer Model Based On The Monte Carlo Method: Application To Steel Sheet Temper Rolling

The work is motivated by the automotive customer requirements for a roughness with high peak density (RPc > 75 peaks/cm) on steel sheet surfaces to maintain a good strip surface finish after painting. For that purpose, a new approach is utilized to compute roughness transfer and predict final roughness on uncoated steel sheets (average roughness, skewness, peak density) as a function of temper rolling conditions. The approach is based on the Monte Carlo method combined with the Abott curves and an auto-correlation function to numerically generate roughness profiles for roll and strip respectively. To compute the roughness transfer due to the interaction between roll and strip surfaces during rolling, a composite Abott curve that combines roll and strip Abott curves is used with an asperity crushing model integrated with a rolling model. This enables simulating two roughness transfer mechanisms in the roll bite: the strip asperity crushing mechanism by roll roughness and the strip asperity ploughing mechanism due to roll-strip interfacial sliding speed. The model shows the importance of considering the two mechanisms to predict roughness transfer correctly when strip elongation is high, typically a few percent, and roll bite friction is low. In these conditions, the forward slip at the exit of the roll bite tends to erode strip roughness significantly and decrease considerably the peak density on the strip. For lower elongation levels (~1%) and high roll bite friction, the ploughing mechanism becomes negligible and roughness transfer is controlled by the crushing mechanism only. With these two mechanisms, the model reasonably predicts the evolution of average roughness and peak density over a wide range of temper rolling conditions (new/worn work rolls, hard/soft grades, various elongations), but skewness is more difficult to predict for some trials.