solver for multi region?
 

Hi,

I would like to solve for magnetic field (Maxwell equation) induced by permanent magnets. Each magnet has different material properties. Hence, I am considering a new solver with multi regions. chtMultiRegionFoam is similar, but, much more complex because it solves both fluid and solid.

Has anyone tried anything similar, ie, one governing equation (vector, not scalar) but, with multiple regions?

Thanks!

Pei

I don't think it should be too hard. I modified chtMultiRegionFoam for a simple diffusion equation (for a scalar) between two regions where the diffusivity changes between the two regions.

Here's my model: multiRegion-chtMultiRegion.tgz

And you can see the basic result here: http://www.cfd-online.com/Forums/ope...tml#post251624

Thanks a lot benk!

I will study your implementation. Maybe we can write a tutorial on this later.

Pei

I can help with a tutorial on this, although I'm still trying to understand chtMultiRegionFoam myself.

implicit approach for region coupling
 

Dear foamers

I am simulating a multie-region problem in OF.

As far as I now, cht multiregion solver implements explicit approach for coupling the regions. (It modifies the boundary value at the coupled interface by iterations).
I want to know if there is an implicit approach for coupling.
Like making fvMatrix for different regions and assembling them and finally solving one general fvMatrix?

best,

Hossein

Hello Hossein,

In the OpenCFD branch there is an implicit mode for chtMultiRegionFoam since version v2112: https://www.openfoam.com/news/main-n...mplicit-cyclic

If I remember correctly, there was a thread about it somewhere on the forum where people noticed different results when using the implicit mode compared to the original explicit one.
Tobias opened a bug report about it: https://develop.openfoam.com/Develop.../-/issues/2536
It's still open so I don't know what is the current behavior in the latest version.

Hope this helps,
Yann

Hi

Than you very much for your response.

As I checked the solver in the latest version, it solves fvMatrix of each region ,seperately. it seems that, it couples the regions explicitly.



best regards,

Hoss

Hi

Than you very much for your response.

As I checked the solver in the latest version, it solves fvMatrix of each region ,seperately. it seems that, it couples the regions explicitly.



best regards,

Hoss

Hello Hoss,

When the implicit mode is used, one single energy equation is solved at once for all regions.

Regards,
Yann