Speaker
Description
In welding processes, a large melt pool is developed when the heat source passes the top surface of the weld part. Fluid flows develop thereafter in this domain mainly induced by the Marangoni effect. This fluid flow has influences on the microstructure evolution during solidification stage. Up to now, despite its significant effect, few researches take into account the fluid flow phenomenon in the grain structure prediction. The Cellular Automaton (CA) – Finite Element (FE) tool offers the possibility to develop realistic virtual microstructure similar to the ones observed in industrial casting. In the CAFE method, two resolution scales are used. On the one hand, mass, energy and momentum conservation equations are solved at macro-scale using a FE mesh. On the other hand, a CA grid is used to model the development of the grain envelopes in the mushy zone at micro-scale. Fraction of mushy zone is then reassigned on FE mesh in a coupling approach to consider undercooled domain in thermal field evolution. Nowadays, this CAFE approach has been widely used for the simulation of solidification process and grain structure development in casting, welding or additive manufacturing processes.
Therefore, a fluid flow model allowing the simulation of liquid flow inside the melt pool is coupled with the CAFE method to provide a reliable grain structure prediction. In this simulation, the thermal field is firstly compared between simulation and experimental results when analyzing the form of the melt pool also discussing fluid flow effect on its evolution. This comparison aims at providing a reliable thermal condition for grains development. Then the simulated grain structure is compared with the EBSD measurements. Coherence is found between results, demonstrating this fluid flow effect on grain structure prediction.
| Speaker Country | France |
|---|