Numerical and experimental analysis of melt layer formation in serial hot gas welding

Abstract Serial hot gas welding is an increasingly important contactless joining process for thermoplastic components, particularly for applications requiring complex weld seam geometries and low-particle processing conditions. However, process design is still largely based on empirical knowledge since the interactions between hot gas flow, heat transfer and melt layer formation are not yet sufficiently understood. In this work, a simulation-based approach for the process design of serial hot gas welding is presented. A computational fluid dynamics (CFD) model created in ANSYS Fluent 2025 R2 was used to simulate hot gas flow, convective heat transfer and transient temperature evolution during the heating process. Polypropylene (PP) was selected as the representative thermoplastic material. Based on the simulated temperature fields, the formation and thickness of the melt layer were determined. Experimental validation was performed using infrared temperature measurements, high-pressure joining tests and transmitted-light microscopy combined with image-based melt layer quantification. The simulation results show good agreement with the experimental investigations. Surface temperature deviations between simulation and experiment remained below 10%, while the simulated melt layer profiles and thicknesses correspond well with the experimentally observed melt layers. Furthermore, the simulations enabled the identification of unfavourable flow conditions, local overheating, insufficient heating and non-uniform melt layer formation. The presented approach demonstrates the potential of simulation-assisted process design to reduce experimental effort and support the optimisation of hot gas tools and joining geometries in serial hot gas welding processes.