CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   MHD in Fluent (

lg88 May 25, 2012 03:42

MHD in Fluent
hello everyone
I want to use the external MHD model(electric potential method) to simulate liquld flow in a rectangular duct.But It seems that when the Hartmann number up to 100,the result didn't fit well with the analytic solution.So I wonder that the MHD model can simulate how much Hartmann number.

Thank you very much!


fferroni May 31, 2012 10:36


I've been doing my Master's thesis on incompressible MHD flows... What I've found is that FLUENT isn't all that good. The magnetic induction method works OK for Ha<100 both in cases like Hunt and Shercliff flow. The electric potential method gives erroneous results.

If you want to simulate incompressible MHD, try using openFOAM. If Hartmann numbers are low-ish, then the built-in mhdFoam solver will do just fine.

If you want to use very high hartmann numbers, i.e. fusion reactor Ha=O(10^4), that's a whole different bag of beans. You should modify icoFoam or something to use a quasi-static inductionless approximation. An explanation of how to do this is given on page 86 onwards:

Alternatively, try using CFX. I haven't tried it personally, but I was reading a paper from Korea a few weeks ago which claimed to use it successfully even for very high Hartmann numbers.

Good luck!


levka November 14, 2012 09:41

electric potential
Hello, guys!
i am wondering, is anywhere available solver for OF that have "electric potential formulation" for computing incompressible MHD with high Ha ?

And second, why did not Ms. Mistrangelo or Mr.Dousset shear their contribution to the OF society. As far as i found from their presentations and publications needed modifications to OF solver were successfully done !?

All times are GMT -4. The time now is 17:48.