CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   Multiblock. Interface BC (

Kyros Yakinthos November 19, 1998 13:05

Multiblock. Interface BC
Hello to everyone. Here is a question. Suppose that you have a multiblock grid topology. You solve the N-S equations in a non-staggered CV and you must find a condition concerning let's say the velocity at the grid connectivity interface between 2 connecting blocks (or for the mass at the face..). The momentum interpolation method is adopted to avoid the pressure wingles.You have a point-to-point matching interface.

What is the best way for the boundary condition there ? I've tryed a straightforward interpolation but the result was bad.

Any suggestion ? Thank you in advance

Franck Bertagnolio November 20, 1998 08:39

Re: Multiblock. Interface BC

Concerning your boundary conditions at interface between 2 connected blocks, instead of a simple interpolation, I guess that an interface boundary condition preserving the fluxes would give much better results.

Technically speaking, it should not be too difficult in your case since the interface is conformal (point-to-point matching interface). Then, you should be able to compute the fluxes (of mass, momentum...) at the faces of one block and enforcing the latter at the faces of the neighbouring block.

Good luck.

Hansong Tang November 23, 1998 18:58

Re: Multiblock. Interface BC
Your question sounds interesting. But, could you says some details about your problem? For instance, what's your numerical schemes in subzones? Is the flow compressible or incompressible?

We have done some research about schemes for grid interfaces. Maybe we can exchange ideas.

Hansong Tang

All times are GMT -4. The time now is 02:09.