Identification of Heat Transfer Parameters for Gravity Sand Casting Simulations

Abstract

Gravity sand casting simulations require accurate modelling of heat transfer phenomena to reliably evaluate the expected quality of the produced parts. Average model parameters can be easily retrieved from a validated database. However, these parameters are highly dependent on the specific sand used and the actual forming process in the foundry. Furthermore, the heat transfer from the solidifying alloy to the mould surfaces is not precisely known, so simulation models usually use typical values for overall heat transfer coefficients. Most research works investigate individual parameters, whereas heat transfer phenomena largely arise from their interaction together. Therefore, the present work describes a combined experimental and computational method based on genetic algorithm techniques for determining the most important parameters for heat transfer in a sand mould. The experiments examine both virgin and reused sand, as these are alternatively used in the foundry for mould forming. The density, thermal conductivity, and specific heat capacity of the different sands are identified, along with heat transfer coefficients. The counterproof simulations demonstrate that the standard parameters are quite reliable for virgin sand. However, in the case of reused sand, the identified parameters lead to more reliable results.