Search

A super-layer deposition method is developed for 3D macroscopic finite element modeling of heat transfer at part scale during the powder bed fusion (PBF) process. The proposed super-layer strategy consists of the deposition of batches of several layers. The main consideration is to deal with the effective heating times and with the inter-layer dwell time in a reasonable way. The material is deposited at once for each super-layer thanks to level set and mesh adaptation methods, while the energy input is prescribed, either by respecting the layer-by-layer thermal cycle or in a single thermal load. The level set method is used twice: first to track the interface between gas and the successive super-layers of powder bed and second to track the interface between the part in construction and the nonexposed powder. To preserve simulation accuracy, adaptive remeshing is used to maintain a fine mesh near the evolving construction front during the process. Simulation results obtained by means of this super-layer method are presented and discussed by comparison with those obtained by layer-by-layer strategy, considered here as a reference. It is shown that, when respecting certain conditions, temperature evolutions and distributions approaching the reference ones can be obtained with significant savings on computation time. Assessment is first performed on simple part, then on a more complex configuration. [ABSTRACT FROM AUTHOR]

A three dimensional model for material deposition in Selective Laser Melting (SLM) with application to Al2O3-ZrO2 eutectic ceramic is presented. As the material is transparent to laser, dopants are added to increase the heat absorption efficiency. Based on Beer-Lambert law, a volumetric heat source model taking into account the material absorption is derived. The Level Set method with multiphase homogenization is used to track the shape of deposed bead and the thermodynamic is coupled to calculate the melting-solidification path. The shrinkage during consolidation from powder to compact medium is modeled by a compressible Newtonian constitutive law. A semi-implicit formulation of surface tension is used, which permits a stable resolution to capture the gas-liquid interface. The formation of droplets is obtained and slight waves of melt pool are observed. The influence of different process parameters on temperature distribution, melt pool profiles and bead shapes is discussed.

A modelling approach is developed for the description of microstructure formation in the industrial AM1 Ni-base superalloy. Solidification and homogenization simulations are first carried out using a microsegregation model, before using the local compositions as an input for precipitation calculations, in order to characterize the influence of segregation on precipitation. First, the precipitation model was validated by comparing simulated and measured evolutions of the average precipitate radius during isothermal heat treatments at 1100 ∘C and 1210 ∘C. The chained microsegregation and precipitation simulations indicate that the global sequences of precipitation events remains are qualitatively the same at the different locations in the microstructure, but the growth and dissolution kinetics are strongly influenced by the local compositions. Local supersaturations have a larger effect on the average radius of the precipitates than certain stages of the precipitation heat treatment.

Dans le domaine du soudage, les propriétés finales du cordon sont fortement liées à la structure de grains développée au cours des procédés de fusion / resolidification. La maîtrise des propriétés de l'assemblage final passe ainsi par une amélioration de la connaissance de sa structure de ce domaine. Dans cet objectif, un modèle couplé Automates Cellulaires – Eléments Finis est proposé pour simuler le développement, en volume, de cette structure, dans le cadre du soudage TIG. Ce modèle est appliqué au soudage d'acier Duplex 2202 et l'évolution de la structure de grains selon les paramètres procédés est discutée. In the welding area, the final properties of the weld bead are mainly induced by the grain structure developed during the melting and solidification steps. The mastery of the properties of the joining will be achieved with a better knowledge of the developed grain structure. A 3D coupled Cellular Automaton – Finite Element model is proposed in order to simulate the grains development in TIG process. This model is applied to the welding of a duplex stainless steel grade. The grain structure evolution is discussed for the various process parameters.