Residual stresses are inherent in parts manufactured using the wire + arc additive manufacturing (WAAM) technique, resulting in unpredictable mechanical response and structural integrity (Colegrove et al.: J Mater Process Technol 213:1782-1791, 2013). An effective post-processing technique, which enhances the mechanical properties of WAAM parts, is rolling. This study investigates the vertical and pinch rolling effects on residual stress distribution in WAAM components. Initially, a WAAM model was created using a thermo-mechanical finite element modelling approach and validated against the experimental results. Subsequent to the validation of the model, the effect of the main parameters involved in vertical and pinch rolling processes, namely the rolling depth, the curvature depth of the roller, the roller shape, transversal displacement, rolling direction and roller thickness, was investigated. The results from this study show that the residual stress profile in the vertical rolling process applied on a WAAM wall can be enhanced by increasing the rolling depth and curvature depth of the roller. Moreover, it is shown that in the pinch rolling process, the residual stress profile is sensitive to the rolling direction and more compressive residual stresses can be induced into the wall by applying fewer passes of rolling using thicker rollers.

No abstract

No abstract

This is the first of two manuscripts that presents a computationally efficient full field deterministic model for laser powder bed fusion (LPBF). A new Hybrid Line (HL) heat input model integrates an exponentially decaying (ED) heat input over a portion of a laser path to significantly reduce the computational time. Experimentally measured properties of the high gamma prime nickel-based superalloy RENÉ 65 are implemented in the model to predict the in-process temperature distribution, stresses, and distortions. The model accounts for specific properties of the material as different phases. The first manuscript presents the HL heat transfer model, which is compared with the beam-scale exponentially decaying model, along with the melt pool geometry obtained experimentally by varying the laser parameters. The predicted melt pool geometry of the beam-scale ED model is shown to have good agreement with experimental measurements. While the proposed HL model exhibits lesser accuracy in predicting the melt pool geometries, it can predict the cooling rates and nodal temperatures as accurately as to the ED model. Moreover, under large time integration steps, the HL model becomes more than 1,500 times faster than the ED model.

No abstract