dolfin team mailing list archive

["Johan Hoffman" <jhoffman@xxxxxxxxxx>]

> Anders Logg wrote:
>> On Tue, Jan 15, 2008 at 08:33:07AM +0100, Murtazo Nazarov wrote:
>>
>>>> Related question for both incompressible flow and elasticity
>>>> problems.  Suppose I have a plane of symmetry that will allow me to
>>>> reduce my computational domain.  If
>>>>
>>>> \sigma_ij
>>>>
>>>> is the relevant stress tensor, then I will have that
>>>>
>>>> t^k_i \sigma_ij n_j = 0
>>>>
>>> Ones you have tangential and normal vectors, you can include this as a
>>> boundary term to the ffc form?
>>>
>>
>> FFC/DOLFIN only have built-in facet normals (FacetNormal). There is no
>> FacetTangent but it would be easy to add (if we just define what we
>> mean by the tangents on a face of a tetrahedron).
>>
>>
> The normals and tangents I mean should be on the nodes not on sides of
> element. This will also be available in UNICORN.

Yes, this (only facet normals) is one of the reasons the implementation for
Dirichlet type bc's (such as slip bc) is not straight forward using
FFC/DOLFIN today.

Although, possibly facet normals may be ok for Neumann type bc's.

/Johan

>
> /murtazo
>
> _______________________________________________
> DOLFIN-dev mailing list
> DOLFIN-dev@xxxxxxxxxx
> http://www.fenics.org/mailman/listinfo/dolfin-dev
>