[leejc]

I have a 3-D cylindrical domain (mesh is cartesian) and would like to add a constant azimuthal (theta direction) component to the inlet boundary for U. Now, there doesn't seem to have a primitive patch field type for this. Does anyone know how this can be done without having to calculate the list of U for each cell surface on the patch?

Many thanks in advance.

[schmidt_d]

Hi Jerry,

Are all three vector components of the boundary invariant in time? If so, I would recommend funkySetFields. It allows you to set fields with expressions. Normally, it is used for setting initial conditions, but if it is writing volVectorFields, I suspect that it will write the values for the boundaries too.

See:
http://openfoamwiki.net/index.php/Co...funkySetFields

-David

[schmidt_d]

Jerry,
Ignore funkySetFields. I forgot about groovyBC, another of Bernhard's whimsical masterpieces. That would cut right to the chase. It lives in the OF1.5 dev breeder too.

see http://openfoamwiki.net/index.php/Contrib_groovyBC for documentation.

If it would help, my student Kshitij wrote a brief PDF that explains the nomenclature used in OF's mixed BC's. Email me and I can send it to you.

-David

[schmidt_d]

All,
It makes more sense if I publish this here. Wish me luck with the attachment. It is a gzipped PDF (to get it small enough to be posted as an attachment).
-DPS

Originally Posted by schmidt_d

Jerry,
Ignore funkySetFields. I forgot about groovyBC, another of Bernhard's whimsical masterpieces. That would cut right to the chase. It lives in the OF1.5 dev breeder too.

see http://openfoamwiki.net/index.php/Contrib_groovyBC for documentation.

If it would help, my student Kshitij wrote a brief PDF that explains the nomenclature used in OF's mixed BC's. Email me and I can send it to you.

-David