CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Electromagnetics (https://www.cfd-online.com/Forums/electromagnetics/)
-   -   Best way to model MHD levitation melting (https://www.cfd-online.com/Forums/electromagnetics/213478-best-way-model-mhd-levitation-melting.html)

Romain_P December 28, 2018 09:35

Best way to model MHD levitation melting
 
Hello All,

I'm currently working on a project where I want to model levitation heating and melting of a metal sample and control the shape of the liquid metal drop using the magnetic field. The magnetic field for levitation is generated by coils. So it's a liquid metal MHD problem (low magnetic reynolds number) with only a single free surface (An example is the TEMPUS experiment from NASA):


https://i.ytimg.com/vi/iC21sjlhpFE/maxresdefault.jpg

I have the theoretical knowledge for MHD but I've been trying for weeks to find a way to effectively simulate this set-up in practice, with no success up to now. So before I continue looking aimlessly for the best way to do this, I'd really like some feedback from people who have experience with such MDH problems: what is the best method to simulate such an MHD problem? Methods that I've looked into:

- Matlab: while this gives an in-depth control over the simulation, matlab tools for MHD and partial differential equations in three spatial dimensions are lacking.
- Fluent: it looks like this would be the best solution, but even after days of reading tutorials and trying to make it work, I really don't manage to. There is an MHD module built-in, but if I go with this solution, could someone explain to me in simple terms how to proceed? My only experience with fluent is the steady-state simulation of fluid flow in complex geometries, without free surfaces (and it was over 3 years ago). I have no idea how to even begin with this problem. For instance, what kind of mesh do I need to use (for some reason every mesh I import in fluent generates a null pointer error)? How do I take the time-dependent nature of the magnetic field into account? Etc.
- OpenFOAM (+ Open-source Elmer): only read about it, haven't tried it in practice yet because it seems much more complex than the other methods. If however this is the best solution I'll start learning to use it.

Any help would be greatly appreciated, and I'd love to be able to help others on here with MHD problems once I get the grasp of this.

Thank you,
Romain P

Dzhordzhs January 10, 2019 17:15

Hi Romain!

That's good that you consider OpenFOAM+Elmer. Probably you also know about EOF-Library and that this tool was build with MHD in mind.

WEB
https://eof-library.com/

Electromagnetic levitation problem
https://www.researchgate.net/publica...magnetic_field

Article explains EOF-Library
https://www.researchgate.net/publica...fluid_dynamics

I am developer of EOF-Library and I can help you with your problem, lets keep in touch.

Juris

Romain_P January 12, 2019 07:44

Quote:

Originally Posted by Dzhordzhs (Post 721584)
Hi Romain!

That's good that you consider OpenFOAM+Elmer. Probably you also know about EOF-Library and that this tool was build with MHD in mind.

WEB
https://eof-library.com/

Electromagnetic levitation problem
https://www.researchgate.net/publica...magnetic_field

Article explains EOF-Library
https://www.researchgate.net/publica...fluid_dynamics

I am developer of EOF-Library and I can help you with your problem, lets keep in touch.

Juris




Hello Juris,


Indeed, I've read quite a lot of good things about OpenFOAM + Elmer for metallurgical MHD modeling. I've responded to your message on LinkedIn, let's keep in touch. Your help is very much appreciated!


Best,
Romain


All times are GMT -4. The time now is 18:43.