Importance of Characterizing Heat Transfer in Both the Tool and Workpiece for Friction Stir Welding Thermal Model Validation

Accurate thermal modeling of friction stir welding (FSW) requires correct representation of heat generation and heat partitioning at the workpiece/tool interface. However, most published models are validated against temperature measurements from only one side of this interface, so agreement with experiment does not guarantee that interfacial heat transfer is correctly represented. This study evaluates whether one-sided temperature validation is sufficient, using steady-state FSW of AA 6061-T6 aluminum with an H13 steel tool. Temperatures were measured with thermocouples embedded in both the workpiece and the tool, and a Eulerian thermomechanical model was developed in ForgeNxt. The viscoplastic friction coefficient and the workpiece/tool heat transfer coefficient were calibrated against tool temperatures only, workpiece temperatures only, and both simultaneously. Tool-only calibration reproduced tool temperatures within 2.5% but overpredicted workpiece temperatures by 18% on average; workpiece-only calibration achieved 5% average workpiece error but underpredicted tool temperatures by 26%. Sensitivity analysis showed that workpiece temperatures were governed primarily by the friction coefficient, while tool temperatures were sensitive to both friction coefficient and heat transfer coefficient. No single parameter pair reproduced both temperature sets, indicating that one-sided validation can produce misleading agreement and that two-sided validation is necessary to achieve accurate interfacial heat partitioning.