[oort]

Hi!

I want simulate the water flow in a rectangular channel filled with transversal round shape fillaments.

I have build a php script that generates the vertices, blocks, edges and also the inlet and outlet patches.

When I run the blockMesh it gives a warning saying that I have 'n' undifined faces and that they will be added to default patch. I think this is normal and not problematic because I only have defined the inlet and outlet patches.

Create time


Reading block mesh description dictionary

Creating block mesh

Creating blockCorners

Creating curved edges

Creating blocks

Creating patches

Creating block mesh topology

Default patch type set to empty
--> FOAM Warning : 
    From function polyMesh::polyMesh(... construct from shapes...)
    in file meshes/polyMesh/polyMeshFromShapeMesh.C at line 576
    Found 322 undefined faces in mesh; adding to default patch.

Check block mesh topology

    Basic statistics
        Number of internal faces : 143
        Number of boundary faces : 326
        Number of defined boundary faces : 326
        Number of undefined boundary faces : 0

    Checking patch -> block consistency

Creating block offsets

Creating merge list .

Creating points

Creating cells

Creating patches

Creating mesh from block mesh

Default patch type set to empty

Writing polyMesh

end

I'm trying to use the icoFoam solver but I don't know if this is the best solution.

But what should be the inicial conditions (at /0 directory) for p and U in the three patches (inlet, outlet and defaultFaces)?

I have tried several things but none works well.

I think the problem is not in the mesh generation because blockMesh only gives that warning,
The program starts building the mesh for the modules between the fillaments. First calculates the vertices and edges for the first module and calculate for each following module. Then adds the section before the first module and then adds the section after the last module. The program adds the number of modules we wish.

The vertices/blocks/edges are defined for the first block in the next picture:

http://www.carloscompleto.com/images/grid_img_04.pdf


In the next pictures you can see that the mesh looks good:

http://www.carloscompleto.com/images/grid_img_01.jpg
http://www.carloscompleto.com/images/grid_img_02.jpg
http://www.carloscompleto.com/images/grid_img_03.jpg

When I run the blockMesh I have to edit the boundary file and change the defaulFaces from "empty" to "patch" because if I don't do it the icoFoam gives an error.

3
(
    inlet
    {
        type            patch;
        nFaces          32;
        startFace       4976;
    }
    outlet
    {
        type            patch;
        nFaces          32;
        startFace       5008;
    }
    defaultFaces
    {
        type            patch;
        nFaces          1504;
        startFace       5040;
    }
)

The time zero (/0/U) i'm using the following code but I don't know if it well defined.

boundaryField
{
    inlet      
    {
        type            fixedValue;
        value           uniform (0.002 0 0);
    }

    outlet      
    {
        type            zeroGradient;
    }

    defaultFaces    
    {
        type            fixedValue;
    value           uniform (0 0 0);
    }
}

Any way the Courant number is only smaller than 1 in the first iteration then explodes and the solver stops after some iterations and the results are horrible (with no physical credibility). I'm using deltaT = 0.00001. If I use a smaller deltaT the solver does more iterations but explodes anyway.

I'm using the default fvSchemes and fvSolution files of the cavity tutorial.

What should I do?

[hansel]

Hi oort. I'm no expert, but I've been playing with TurbFoam and it seems to be very sensitive to abrupt cell size changes. An example is where your cells go from a square cross section to rectagles with half the width. In my simulation for no reason the velocity would change there.

Maybe if you could change the cell sizes more gradually, that might help.

But like I said, I'm no expert, this is just my experience.

Steve