Content area
Abstract
Heat-flow models based on the finite difference and finite element methods have been developed to calculate the steady-state thermal field in billet and slab moulds. A transient heat-flow analysis which employs an implicit alternating-direction finite-difference method has been formulated expressly to investigate the effect of surface boiling on billet mould temperatures. The models were validated by comparing predictions with experimental measurements reported in the literature.
The thermal distortion of billet moulds have been computed using a three-dimensional elasto-plastic finite-element model. It has been shown that thermal expansion is the dominant component of the total distortion while the contribution due to bending associated with the through thickness gradient is relatively small. The distorted mould exhibits an outward bulge in the meniscus region, which has a maximum value in the 0.1('(TURN))0.3 mm range, approximately 6 cms below the meniscus. The peak distortion is bounded above and below by regions of negative and positive taper respectively. Measurements of wall movement in an operating billet caster compare favourably with the predicted distortions. The periodic movement of the mould wall observed in the meniscus region has been attributed to nucleate boiling in the cooling channel.
A systematic analysis of the effect of operating variables on wall temperatures has revealed that owing to the thicker walls, peak slab mould temperatures are higher than in billet moulds. It is therefore recommended that materials with enhanced softening resistance be employed in slab moulds. In contrast to this softening is unlikely in billet moulds but however owing to higher cold-face temperatures boiling may occur in standard flow-rates; water quality must therefore be adequate to avert deposition of scale and the attendant high-wall temperatures and excessive distortion. A comparison of the thermal distortion profiles calculated for six industrial billet moulds has shown that mould constraints and wall thickness have a profound influence on permanent distortion. Lowering the meniscus level or modifying the constraints so as to minimize the restraints to thermal expansion tend to reduce the plastic strain and could be beneficial from the standpoint of mould life.
A mechanism has been proposed relating longitudinal corner cracking and rhomboidity in billets to an intermittent boiling cycle in the mould. It has been demonstrated that if boiling on two adjacent faces of the mould is out-of-phase with boiling on the other two, uneven cooling and rhombiodity would result. This can lead to corner cracks by hot-tearing in interdendritic regions near the solidification front.
