Multi-Scale Multiphysics Simulation of the Thermo-Fluid-Metallurgical-Mechanical Conditions During Metal Additive Manufacturing

Mohamad Bayat*

*Corresponding author for this work

Research output: Book/ReportPh.D. thesisResearch

6 Downloads (Pure)

Abstract

Metal additive manufacturing (MAM) has recently attracted a lot of attention from different industrial sectors such as medical, aerospace, aviation, automotive, energy, etc., thanks to its unique capability of manufacturing customized complex geometries within a short end-to-end production time. Although MAM outweighs conventional production techniques in some certain aspects, it still needs to be improved as there are various types of defects that can originate from an improper choice of input parameters and during the course of the process.

Defects in MAM products have different varieties and cover a wide range, including surface irregularities such as dross, stair-case and balling effects, voids and porosities and finally, deflections. The voids in MAM products can either form due to insufficient energy input, leading to the so-called lack-of-fusion porosities, or they can form due to excessive energy input that causes big depression zones in the melt pool that will eventually end up in keyholeinduced porosities. The deflections in MAM products are caused by the formation and buildup of residual stresses during the process. In some extreme cases, these stresses can even lead to cracks. This signifies the fact that MAM is an inherently multiphysics process. Convection, radiation, evaporation and evaporative cooling, melt pool dynamics, recoil pressure, capillarity, the Marangoni effect, plastic deformation and yielding, creep, laser-material interaction and multiple reflections are some of the physical phenomena that take place during the course of MAM. Furthermore, the mentioned physics might occur at different lengthscales, spanning from micro-scale, to meso-scale and part-scale. Moreover, the fact that all these physics occur within a very short timespan, makes this process even more complicated.

Accordingly, any improper selection of the input process parameters in MAM can lead to the formation of defects that affect the final quality of the products or deteriorate their mechanical properties. In this scenario, numerical simulations that take the involved physical phenomena into account, can be used as a tool for investigating the impact of the input process parameters on the part quality.

The aim of this Ph.D. thesis is to model the thermo-fluid-metallurgical-mechanical conditions during MAM and at different length-scales. To this end, to fill the gap in the current literature in the field, three investigation tracks are identified in this thesis; meso-scale simulations, the Marangoni effect in MAM and finally, part-scale simulations. In one study from the first investigation track, the impact of the input process parameters in the Laser Powder Bed Fusion (L-PBF) process and on the heat and fluid flow and metallurgical conditions is studied. It is found that higher scanning speeds lead to higher cooling rates that cause finer grains sizes that eventually improve the mechanical strength of the samples. In another study from this investigation track, the formation and evolution of the lack-of-fusion porosities in L-PBF are simulated. The formation of keyhole-induced porosities is also studied in the first investigation track. Here, it is shown how the change in the morphology of the depression zone (keyhole) can change the absorption of the laser power via multiple reflections. Finally, in the last part of this investigation track, a thermo-fluid model is developed for the Directed Energy Deposition (DED) process, where the influence of the carrier flow rate on the melt pool conditions is studied. It is observed that via using higher carrier flow speeds, the height-width ratio of the DED tracks increases. In the second investigation track, a fundamental study is performed on the impact of the Marangoni effect on the heat and fluid flow conditions during the L-PBF process. Both normal and inverse Marangoni effects are simulated and analyzed. It is found that higher magnitudes of the Marangoni effect lead to more uniform temperature distribution within the melt pool, where the role of conduction becomes less significant, as the temperature gradients disappear. Finally, in the last investigation track, a part-scale model is developed for simulating the thermo-mechanical conditions during the L-PBF process, where a novel multi-scaling rule, the sequential flash heating is introduced and implemented. It shown that by refining the stripe sizes, the final predicted deformation gets closer to the one measured experimentally.
Original languageEnglish
Place of PublicationKgs. Lyngby
PublisherTechnical University of Denmark
Number of pages245
ISBN (Electronic)978-87-7475-623-1
Publication statusPublished - 2020

Fingerprint Dive into the research topics of 'Multi-Scale Multiphysics Simulation of the Thermo-Fluid-Metallurgical-Mechanical Conditions During Metal Additive Manufacturing'. Together they form a unique fingerprint.

Cite this