Open Access

This thesis aims to develop a FE-based model of a dieless wire drawing process for wires made from magnesium alloys. To this end a general material model of pure magnesium and a model of the dieless wire drawing process are developed. Based on the general pure magnesium model an alloy specific model for AZ31 wire is developed. The performance of both models is assessed using experimental data generated on a dieless wire drawing prototype. The process model is conceptionally split into the thermal and mechanical response of the wire. The thermal model is validated by axial temperature profiles and the mechanical model is vali-dated by CSA-reduction and wire force. Both behaviours are validated separately before combin-ing the thus created models into a thermomechanical model of the dieless wire drawing process. The thermal material model is developed for pure magnesium. An initial assumption of limited correlation between content of alloying elements and thermal behaviour, was disproven. As a results in addition to alloy-specific mechanical data, thermo-electric data is recorded to achieve thermal validity of the model. This is done by identifying the experimental maximum temperature of the drawn wire for a given heating power and calculating the necessary input power of the in-duction heating device to achieve this temperature in simulation. The mechanic material model is based on experimental stress-strain curves recorded for each investigated wire materials in addi-tion to pure magnesium data, based on literature. Results show the thermomechanical magnesium models to be mostly valid, provided process parameters stay within the range of available data on the mechanic material performance. Where the model is forced to extrapolate material behaviour, simulation quality drops. This ap-plies for wire temperature and CSA-reduction. Estimations of wire force are shown to be invalid. For AZ31 wire the thermal model generated valid temperature profiles of the wire. The thermo-mechanical model for AZ31 is shown invalid as both CSA-reduction and wire force deviate from experimental results.