Macroscale modelling for efficient residual stress prediction in arc directed energy deposition

Purpose The present study is aimed at developing a computationally efficient residual stress modelling method that is suitable for arc directed energy deposition (Arc-DED) additive manufacturing applications using macroscopic continuum finite element techniques adapted from weld modelling. Design/methodology/approach The approaches were undertaken on the NeT-TG9 Arc-DED benchmark, which comprises five depositions on an SS316 L substrate. A thermo-mechanical modelling workflow was developed using guidelines from the R6 structural integrity assessment procedure and realised with FEAT-WMT and Abaqus solvers. Bead lumping (BL) and moving torch (MT) simulations were compared for computational time and accuracy. This was followed by pass lumping (PL), which simplified the BL method further with reductions in computational time. PL was implemented by adjusting the heat input to maintain the mid-length fusion zones. The thermo-mechanical simulation results were validated against experimental measurements. Findings Thermal analyses showed similar temperature distributions between the MT and BL simulations, and the experimental data. The two approaches were also comparable in their residual stress predictions, closely matching experimental measurements. BL was the more computationally efficient technique. PL also predicted the longitudinal residual stresses within the acceptable threshold for accuracy and reduced the computational time by 92% from the MT simulation. Originality/value While the methods discussed have been implemented in weld modelling, their use in Arc-DED additive manufacturing is still limited. The PL method can be applied to the TG9 benchmark and similar problems and produce accurate results. Optimisation of the technique could speed-up Arc-DED modelling to more industrially practical timeframes.