FEMS EUROMAT 2023
Lecture
04.09.2023 (CEST)
Thermomechanical modelling of the multi-layer directed energy deposition process through flash heating with focus on remelting
AS

Alberto Santi (M.Sc.)

Technical University of Denmark

Santi, A. (Speaker)¹; Bayat, M.¹; Fabrizi, A.²; Bonollo, F.²; Hattel, J.¹
¹Technical University of Denmark, Kgs. Lyngby (Denmark); ²Universitá di Padova
Vorschau
22 Min. Untertitel (CC)

Prediction of the temperature and stress development during directed energy deposition (DED) is crucial to assess the part’s final quality before starting production. Various numerical methods have been proposed to achieve such results, taking different assumptions to minimize the computational resources needed and allowing for virtual Design of Experiments (DOE) for process parameters optimization. The Flash Heating (FH) method has been used to model the outcome of different DED geometries. When defining the geometry, the free-surface deposited by each track is usually neglected in part-scale models, and thus, as a simplifying assumption, these free surfaces are substituted by flat-surfaces. This could also complicate the numerical implementation of overlapping, which can be achieved via double-activation of the elements.

In this work, the authors propose an implementation of the flash heating technique for DED samples built with 316L. The formulation is an extension of the one developed for laser powder bed fusion modelling. The geometry is extracted from optical microscope analysis to allow the implementation of free surfaces. It has been possible to show that the remelting lines visualized after etching are due to remelting of previous tracks, and the constant volume per track should be prioritized instead of using such melting lines to reproduce the geometry to be simulated. It is shown that such a model can indeed predict the occurrence of the remelting lines.

Subsequently, an extensive analysis of the relevant simulation parameters of FH is provided, with the objective of exploiting this numerical method to minimize the computational cost. Finally, the method is tested at different process parameters to test its sensitivity towards them and ease implementation once the model is calibrated. Numerical results are coupled with experimental ones from online monitoring (temperature history during deposition) and characterization techniques (microstructural and compositional evaluation of the 3D printed parts).

Abstract

Abstract

Erwerben Sie einen Zugang, um dieses Dokument anzusehen.

Ähnliche Inhalte

© 2026