PAMM · Proc. Appl. Math. Mech. 17, 611 – 612 (2017) / DOI 10.1002/pamm.201710275 Numerical simulation and validation of a solidification experiment using a continuum mechanical two-phase/-scale model Lukas Moj 1, * , Tim Ricken 1, ** , Manuel Foppe 2, *** , and Rüdiger Deike 2, † 1 University of Stuttgart, Institute of Mechanics, Structural Analysis, and Dynamics 2 University of Duisburg-Essen, Chair of Metallurgy for Iron and Steel Production A numerical simulation of a solidification experiment is presented. The experimental setup consists of an empty screwed up mould, which is filled up with molten steel and cools down due to atmospheric conditions. During the cooling process, the temperature at the centre as well as the shrinkage at the narrow side, respectively, have been recorded. The simulation has been run via the software ANSYS modified by specific provided programmer interfaces, the User Programmable Features (UPFs). All required temperature dependent material parameters have been provided by ThyssenKrupp Steel Europe. c  2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 1 Introduction Numerical simulation of casting processes are of significant importance in order to understand the material behaviour during processing. Therefore, the developed two-scale/-phase finite element model [1] has been utilized for the validation of a solidi- fication experiment. Therefore, its implementation into the software ANSYS via the User Programmable Features (UPFs) has been performed. In order to receive physically realistic results, temperature dependent material parameters for the steel alloy have been utilized. 2 Boundary value problem of the solidification experiment The solidification experiment consists of a copper mould (CM), which is filled up by a molten metallic material sample (MMS), that cools down and solidifies due to atmospheric conditions. The CM in fig. 1 a) is assembled from spare parts and fixed to Fig. 1: a) copper mould and b) numerical model a) b) a holding device via the left-bottom drilling hole. Furthermore, the smaller drilling holes at the centre are intended for the thermocouples in order to record the temperature-time distri- bution. Besides temperature, the shrinkage dur- ing cooling of the short side in its normal direc- tion has been measured by a laser device. The corresponding numerical model is illustrated in fig. 1 b). It contains two main parts, the CM in red as well as the MMS in yellow, respec- tively. Due to a symmetric geometry and bound- ary conditions, one quarter of the model has been utilized, where corresponding interface mechanical boundary conditions have been applied. Furthermore, the fixing device has been implemented by setting all its nodal displacements to zero. For the thermal boundary conditions, all external surfaces of the CM cool down via convection to the atmosphere. Furthermore, thermal-mechanical contacts have been utilized for the heat diffusion between the CM and the MMS as well as in-between the CM itself, respectively. The vertical vector of gravity holds the MMS in position. 3 Simulation results Figure 2 shows the temperature-time distribution of the entire model. The initial temperature for the CM and for the MMS has been set to room temperature as well as to the material melting point, respectively. The filling process, which takes for two ∗ Corresponding author: e-mail lukas.moj@isd.uni-stuttgart.de, phone +49 711 685 67097, fax +49 711 685 63706 ∗∗ e-mail tim.ricken@isd.uni-stuttgart.de, phone +49 711 685 63612, fax +49 711 685 63706 ∗∗∗ e-mail manuel.foppe@uni-due.de, phone +49 203 379 1326, fax +49 203 379 1426 † e-mail ruediger.deike@uni-due.de, phone +49 203 379 3455, fax +49 203 379 3464 c  2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim