[Nek5000-users] Anisotropic conductivity

[nek5000-users at lists.mcs.anl.gov]

Hi Alireza,

There is a paper on my website describing the approach:

http://www.mcs.anl.gov/~fischer/pubhtml/fischer_sci05.pdf

Basically, you need to change the geometric factors after
nek has set them.  I normally do this in userchk, because
the first call to userchk implies that the code is ready
to start solving the equations and that (most) all of the
requisite variables are set.    Therefore, if you change
them, they won't be reset against your wishes.

(This is trickier if you have ifuservp=.true.,
because in that case nek will keep resetting the diffusivity...,
but maybe it will be ok---we can look into that.)

I can send an example where I've set up anisotropic 
diffusion problem... (attached here, in fact, but I've
not tested recently).  It's a 2D case.

Let me know if you need assistance with the 3D case or 
have questions with this one -- the only tricky part is 
figuring out where the components of the symmetric 3x3 
tensor are stored.

Best regards,

Paul


On Tue, 26 Jun 2012, nek5000-users at lists.mcs.anl.gov wrote:

> Hi Neks,
>
> I want to solve for a convective heat transfer problem with a diagonal 
> conductivity tensor which has different constant values in x and y 
> directions. I assume I should use udiff, but I'm not sure how to implement 
> it. I wonder if you have any idea.
>
> Regards,
> Alireza
>
> _______________________________________________
> Nek5000-users mailing list
> Nek5000-users at lists.mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
-------------- next part --------------
A non-text attachment was scrubbed...
Name: aniso.tgz
Type: application/x-gtar
Size: 4514 bytes
Desc: 
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20120626/f2c5927c/attachment.tgz>