Effect of oxide scale failure in hot steel rolling on subsequent hydraulic descaling: numerical simulation

A thermal-solid model has been applied to simulate the behaviour of secondary oxide on the surface of the strip during hydraulic de scaling. This model uses the finite element method with detailed modelling of the roll, stock and an oxide layer on the stock. The model allows for the analysis of deformation, viscous sliding along the oxide/metal interface, cracking and spallation of the oxide scale from the metal surface. The oxide scale failure is predicted taking into account temperature dependence of the main physical phenomena, namely thelmal expansion, stress-directed diffusion, fracture and adhesion. A water jet impinging on a hot stock surface with oxide scales, cracked and partly spalled after the rolling pass, is simulated as transient boundary conditions for pressure and heat transfer on the basis of available results of experimental investigations and theoretical assumptions of hydraulic descaling. The boundary conditions are dependent on descaling conditions, such as spraying pressure, discharge rate, spray angle and spraying distance. The mechanism of descaling at hot rolling temperature is investigated from the analysis of operational factors in hot strip rolling and the study of oxide scale and oxide/metal interface properties. An investigation is made of the temperature fields during hydraulic descaling of heated steel strip and the associated thermal stresses contributing to separation the scale from the metal surface. The descalability of steel is strongly influenced by the state of failure in the scale after rolling, characterised by the presence of through-thickness cracks and partly spalled scale fragments. The measurement of separation loads, critical for the modelling of through-thickness cracking or failure due to crack propagation along the oxide/metal interface, is discussed. The results of the computer-based modelling are discussed with regard to the availability of measured data and model assumptions and simplifications.