[clo]

Hi, I'm quite new so sorry if the question is stupid but after reading a lot of stuff I didn't find the aswer...
So, I' m trying to simulate a simple cube inside a wind tunnel (turbulent flow); I fixed symmetryPlanes, inlet outlet but what about the interal walls of the cube? Naturally openfoam gives me errors if I treat it as walls...
Thanx

[giampippetto]

What kind of error do you have?
If you're using turbFoam and FoamX to set the BC, you have to edit the boundaary conditions for k at the walls, since FoamX sets fixedValue but the solver requires zeroGradient.

Bye
Andrea

[giampippetto]

What kind of error do you have?
If you're using turbFoam and FoamX to set the BC, you have to edit the boundaary conditions for k at the walls, since FoamX sets fixedValue but the solver requires zeroGradient.

Bye
Andrea

[clo]

the program recognize that these walls are interal and the mesh creation is aborted...I'm working in 3D.

[giampippetto]

Looks like your mesh isn't correctly defined... you can take a look at this case for a working example, a 2d cylinder.

fatcyl.tar.gz

Bye
Andrea

[clo]

sorry I can't download the file...anyway I'm working in 3D so it doesn't look very interesting for me..

[clo]

I downloaded the file but it isn't very helpfull...thanx anyway! What I m looking for is a simple 3D case with a obstacle inside a flow to see how boundary conditions are defined on the walls of the obstacle...

[mattijs]

What are your 'internal walls'? OpenFOAM only allows boundary conditions on external faces.

[giampippetto]

A 2D case differs from a 3D only for some empty patches; the patches representing the cylinder walls have the same boundary conditions in 2D and 3D.

Bye
Andrea

[clo]

here that's the message error when I try to give empty patches at internal walls:

--> FOAM FATAL ERROR : This mesh contains patches of type empty but is not 1D or 2D
by virtue of the fact that the number of faces of this
empty patch is not divisible by the number of cells.

[clo]

Mattijs, I'm trying to specify a little volume that forms an obstacle for a straight flow. For example in the 2D case DamBreak in the tutorial there is a step:if you try to put a third dimension to that case OpenFoam crashes the calculation because it gives empty boundary conditions to the internal surfaces that are specified to generate the geometry.


I understand that my explications are quite poor but in simple words I want to create an example like DamBreak in 3D

[gschaider]

Hi clo!
(Concerning your message from 6:21. That is an output from the solver, right?)

If I understand you correctly, what you'll want to do is assign a physical boundary condition (for instance 'wall') to that cube.

'empty' is only applicable for cases with dim<3 because OF always solves in 3D and empty helps to 'stretch' the geometry in the direction of the superfluous dimensions into infinity. But that is descibed somewhere in the UserManual in much better words.

[clo]

Hi Bernhard, what you say it's right. I can't find a 3D example that shows me the difference how to treat surfaces used to create geometry and suraces that are real walls!

[andimb]

Hi!

Perhaps the nacaAirfoil tutorial for sonicTurbFoam will help you.

Andreas

[clo]

Hi Andreas! Unfortunately this is a 2D example...

[andimb]

Hi!

Yeah, but I thought, you want to know, how to set the internal walls, like the airfoil. If you set the empty patch at the to sides in this tutorial to a symmetry plane, you have a 3D case.
Ok, with 1 hexa-Element in z-direction.
But it should be the same, or?

Andreas

[clo]

but how can 1 element shows me the flow in the Z direction? (for example in the airfoil case downwash effect)? it's 2D view as 3D!

[andimb]

Hi!

It cannot. I know. But I thought the problem is to define the boundary conditions. And this is the same. It doesn't matter if there are 1 or more Elements.
Perhaps I didn't understand the problem.

Andreas

[clo]

the proble is tha t this error is shown

--> FOAM FATAL ERROR : This mesh contains patches of type empty but is not 1D or 2D
by virtue of the fact that the number of faces of this
empty patch is not divisible by the number of cells.

The only way is to find a real 3D example...

[gschaider]

Clo, tell me if I'm understand the problem you want to simulate correctly: you want to simulate a cube that is suspended (by some mystic force) in mid-air in a wind tunnel. That is a setup that rarley occurs in real live, but it's a perfectly fine example case.

Now. You and me speak of the cube as "being inside the wind tunnel" but for the unsuspecting air molecule (and therefor for the whole fluid) the wall of the cube isn't different from the wall of the wind tunnel. Therefor for the flow the cube is an external boundary and 'wall' is a good starting point for a physical boundary condition. If you want to try different boundary conditions look at table 6.3 and 6.4 in the UserGuide, but NOT table 6.2.

In my definition the wall of the cube is only an 'internal' wall if there is a grid inside the cube (is that the case?)