CFD Online Discussion Forums

CFD Online Discussion Forums (
-   OpenFOAM Running, Solving & CFD (
-   -   Mesh motion applied to an internal interface (

lr103476 September 13, 2007 13:33

Dear all, Right now, I am s
Dear all,

Right now, I am struggling with the following. I would like to define an interface within the mesh and the flow needs to go through the interface. Secondly, that interface needs to move according to a moving body which is located inside the region surrounded by the interface.

Two questions:
1) What is the easiest way to define such an interface. Using gambit, the interface becomes a wall :-(. Will setFace do the job?

2) Which available boundary condition do I have to use on that interface, such that flow is flowing through. I managed to move the interface (insideSlider / outsideSlider) of the mixer2D, but the boundary condition on the interace remains unclear to me....

Btw, I am using icoDyMFoam with some dynamicFvMesh based libraries to account for the motion of a body, and preferably some interfaces.

Thanks and regards,

lr103476 September 14, 2007 02:32

Anyone? Is there a boundary co
Anyone? Is there a boundary condition for interface patches, where the flow goes through?

Thanks, Frank

msrinath80 September 14, 2007 03:19

Frank, If I understand your qu
Frank, If I understand your question correctly, then by interface you actually mean some kind of porous media or membrane?

lr103476 September 14, 2007 03:25

Well, it is just a patch, cons
Well, it is just a patch, consisting of an array of cell faces. But indeed, you can see it a a infinite small membrane, which has to have no effect on the flow! In Fluent, you can do anything with such an internal interface.

I like to specify the motion of certain internal interfaces in order to have some more control on the overall moving mesh quality.

Regards, Frank

hjasak September 14, 2007 03:29

In the tet FEM solver I wrote
In the tet FEM solver I wrote this is quite easy: you can group the faces and apply a constraint defining the point motion there directly on the matrix. The rest is trivial.



lr103476 September 14, 2007 03:35

Hello Hrvoje, Uhmm, I don'
Hello Hrvoje,

Uhmm, I don't really understand how to do that. Do I have to program a new library or can I just apply a BC?

Besides, I really like to use the new finite volume based mesh motion, for my problems it is a lot faster....


zhan001 March 12, 2008 22:07

Hi Frank, Have you get any
Hi Frank,

Have you get any progress on this issue? I am at the beginning stage of the problem which is much similar with yours.
1. How do you deal with the membrane? Are you using nonlinear elasticity model? (I mean when the membrane got stretched)
2. How do you implement the FSI task with OpenFOAM? I did not see an adequate solver for this purpose.
3. Does your problem involve extremely large membrane deformation?

Sorry for so many asking, thanks.

All times are GMT -4. The time now is 13:00.