CFD simulation of metal transfer and melt pool in additive manufacturing with electric arc
||CFD simulation of metal transfer and melt pool in additive manufacturing with electric arc|
||SNIC Medium Storage|
||Isabelle Choquet <email@example.com>|
||2021-01-01 – 2022-01-01|
The present project application is in the continuation of a former SNIC project in CFD simulation of additive manufacturing (AM) with Gas Metal Arc (GMA) heat source to deposit metal layer by layer. In this process a thermal plasma arc is formed, and its energy is used to fuse metal in order to join the additive metal with the metal layer previously deposited. CFD modeling of this multi-physics process is still in its infancy. The existing models need improvement to become self-consistent and predictive, as well as assessment at each step of the developments.
We did develop, test and apply in the open source CFD tool OpenFOAM a solver for computing a thermal plasma arc and its interaction with a refractory electrode. Arc and electrode are coupled through space charge layer and an ionization layer (https://doi.org/10.1016/j.cpc.2017.09.010). We did also develop and test a second solver for melt pool and metal transfer. This second solver, applied in the former project SNIC-2020/5-432, is the first object of the present study. It accounts for thermal flow, temperature dependent thermodynamic and transport properties, melting and solidification, free surface deformation with thermocapillary force, and electromagnetism. We will apply this solver to finalize the 1-year projects SNIC-2020/5-432, 6-167 that were shortened down to 4 months due to the retirement of Hebbe. We will also test new model developments intended to improve the model self-consistency. Our present focus is thus on performing computations for:
1) completing mesh convergence study to finalize (see project report SNIC 2020-5/432)
a) the validation along the free space-directions of the melt pool and metal transfer simulation model for different process conditions,
b) the study of the three different assumptions currently used for modelling the electromagnetic force , and thus
c) the writing of the related two manuscripts currently under preparation to be submitted to international scientific journals.
2) applying the simulation model to single and multi-layer deposition to gain AM-process knowledge for different GMA process conditions met in production, and prepare a publication.
3) testing a new version of the melt pool and metal transfer model extended including the simulation of the arc thermal plasma (also developed in OpenFOAM at University West) to improve the model self-consistency.
The test cases are designed for time-dependent problems requiring small time steps (of the order of 1µs), and mesh cells small enough to capture the thermocapillary flow. Furthermore, the calculations need to be long enough to reach full development of the melt pool and then cool down the deposited metal in order to confront calculation results to experimental measurements for both single and multi-layer deposition.