Right - maybe I should think about these things before I say them ;-)

So now the question is... if you were in my place and wanted to set a
u-bc_val type residual for boundary elements... what would you do for
tri's and tets?  It seems like a post-processing step is the best way
(so that something from an element with only a node on the boundary
doesn't have a chance to contribute anything).  What would that
post-processing step be?  Are there any facilities already in libMesh
that could help out?

I mean, I can certainly loop over the local_active_elements and add
all the nodes on the boundaries to a std::vector<Node *> right after I
read the mesh... but there is a bit of a problem with adaptivity in
this case... and it feels hackish.  But I suppose I was hoping that
this is what a BoundaryMesh was... so I guess it's not all that much
different.

Thanks,
Derek

On Thu, Jun 12, 2008 at 3:25 PM, Benjamin Kirk <[EMAIL PROTECTED]> wrote:
>> So how are other people doing boundary conditions with tri's and
>> tets?  With Dirichlet you can just use a penalty to swamp everything
>> out.  But with Nuemann?
>
> Neumann is not an issue because the BC is in terms of the boundary integral,
> so you only consider it with elements whose sides intersect the boundary.
> The elements who only touch the boundary through a node do not enter into
> the Neumann BC.
>
>

-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
Libmesh-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-users

Reply via email to