[rietis]

Dear OpenFOAM users.

Here is my code for Velocity:

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.3.0                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       volVectorField;
    object      U;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 1 -1 0 0 0 0];

internalField   uniform (0 0 0);

boundaryField
{
    fixedWalls
    {
        type            fixedValue;
        value           uniform (0 0 0);
    }

    roof
    {
        type            slip;
    }

    inlet
    {
       type            fixedValue;
        value           uniform (0.5 0 0);
    }
    outlet
    {
        type            outletInlet;
    outletValue    uniform (0.5 0 0);
        value           (0 0 0);
    }
}

// ************************************************************************* //

I have tried several BC. These were the best. The problem is that after OpenFOAM solves the problem in paraview viewing the velocity it increases. With this I mean after scaling it finds a place where the velocity is higher like 0.625 or 1 m/s which is not physical If my inlet flow is at 0.5 m/s and the outflow the same. It gets worse when I take a smaller mesh.

Can somebody explain what I am doing wrong?

Cheers

Raitis.

[thiagopl]

Hi rietis,

Give us more details about your problem. Which solver are you using? How is your geometry?

Regards

[rietis]

Sorry.

I am using PISO solver. Pressure for all walls and patches are zero gradient.

The geometry is two boxes 3 by 3 and 3 by 3, which are conected by a path like 0.5 m from the top, and in the botton there are inlet and outlet faces.

the inlet and outlet are located 0.05 m or 5 cm from the wall.

The geometry is made from hex blocks. I tried to do snappyHex mesh, but were not able to do that sadly as well, to refine the mesh at the walls.

I hope I explained more precious this time.

Cheers,
Raitis.

[betakv]

Hi rietis,

the outlet BC for velocity should be a inletOutlet with inletValue uniform (0 0 0)

[rietis]

Thank you for your replie.

The problem is that I tried with the inletOutlet with inletValue uniform (0.5 0 0), but still there is that at certain time it jumps up the inlet value.

I tried these as well:
inletValue uniform (0 0 0);
value (0.5 0 0);


inletValue uniform (0.5 0 0);
value (0 0 0);

And pretty much every combination, even though some of them where not physical and did not bring to a good flux.

Do you know what inlet profile does the BC inletOutle give?

cheers Raitis.

[betakv]

Have you checked boundary conditions for pressure? There should be a fixedValue on outlet and zeroGradient on other patches

[rietis]

Fixed value pressure on outlet? No it is zero gradient.
Do I need to calculate the pressure or just guss?

[rietis]

As I understand I can use to determine the pressuere on the outlet like this:

P=F/A wher F=rho*v^2/2*A.

So it just p=rho*v^2/2?

[betakv]

It depend on your case. If it is an incompressible case I suggest to set fixedValue to 0. If it is a compressible case you can set e.g. 1e5 but it depend on your case

[rietis]

it is a incompreesible case so I will try later this day to set on 0 and will report if it fixed the problem.

Thank You.

cheers

Raitis

[thiagopl]

Hi,

You can use fixedFluxPressure as BC for pressure on the walls.
Regardless your BC's, according to your description your case has a restriction to the passage of the flow, right?

Regards.

[rietis]

Yes it really did help when I put 0 for pressure in the inlet.

[rietis]

It did not work for a finner mesh. After getting the results it wee as terrible as before.
fixedFluxPressure bc, how do I implement this?

[betakv]

Can you post your case?

[rietis]

From blockMesh there are these bondaries (it containts like 60 blocks and it is preatty long)

boundary
(
    fixedWalls
    {
        type wall;
        faces
        (     
 .....

    );
    }
    roof
    {
        type wall;
        faces
       (            
.....
        );
    }
    
        inlet
    {
        type patch;
        faces
        (            
      ....
        );
    }

    outlet
    {
        type patch;
        faces
        (            
       ....
        );
    }
    
    problemFaces
    {
    type wall;
    faces
    (            
....
    );
    }


    zeroFaces
    {
    type empty;
    faces
    (            
....
    );
    }
);

For velocity I did

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.3.0                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       volVectorField;
    object      U;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 1 -1 0 0 0 0];

internalField   uniform (0 0 0);

boundaryField
{
    fixedWalls
    {
        type            fixedValue;
        value           uniform (0 0 0);
    }

    roof
    {
        type            slip;
    }

    inlet
    {
        type            fixedValue;
        value           uniform (0.74 0 0);
    }
    outlet
    {
        type            outletInlet;
        outletValue     uniform (0.74 0 0);
    value        uniform (0 0 0);
    }

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

    zeroFaces
    {
        type            empty;
    }
}

// ************************************************************************* //

Actually I wanted to implement parabolicVelocity for inlet, but I was not able to download the liberies for it.

Then for the pressure I have this:

/*--------------------------------*- C++ -*----------------------------------*\
| =========                 |                                                 |
| \\      /  F ield         | OpenFOAM: The Open Source CFD Toolbox           |
|  \\    /   O peration     | Version:  2.3.0                                 |
|   \\  /    A nd           | Web:      www.OpenFOAM.org                      |
|    \\/     M anipulation  |                                                 |
\*---------------------------------------------------------------------------*/
FoamFile
{
    version     2.0;
    format      ascii;
    class       volScalarField;
    object      p;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 2 -2 0 0 0 0];

internalField   uniform 0;

boundaryField
{
    fixedWalls
    {
        type            zeroGradient;
    }

        roof
    {
        type            zeroGradient;
    }

    inlet
    {
        type            zeroGradient;
    }
    outlet
    {
           type            fixedValue;
    value        uniform 0;
    }

    problemFaces
    {
        type            zeroGradient;
    }
    zeroFaces
    {
        type            empty;
    }


}

// ************************************************************************* //

The geometry is two boxes conecting with one path, and there is a different inlet and outlet, at the wall, meaning the inlet is not moved any mm from the wall.

Cheers Raitis.

[betakv]

Boundary seems to be okay. Have you checked the mesh with checkMesh?

[rietis]

Yes I did check the mesh with check mesh. IT was ok. There was a problem, but I fixed it before. The speed of the velocity increases more at the outlet it looks like. And Preatty much every simulation I try to make it is like that. (I tried a different geometry with differenr BC, but still the same). Do I need to specify k or nuS ?

I am really confused, I am trying to solve this for an almost 3-4 months. I thought I got it after put the pressure but it was just for a not very fine mesh.

Do you have any other suggestions?

Cheers,

Raitis.

[rietis]

Here I got a little bit father. I put zerogradient for outlet and it reduce significant the anomoly, but it is still there, I think it should be like that.

Here is the next think I want to make a profile for inlet.
In the attachment there are two files. One is showing the inler place and the other the profile.

In the picture which shows the inlet area is positioned in yz plane, and in the other it is zx plane.

I want that the profile is with these condition:
1) It is normal to yz plane so it is in the n (1 0 0) direction.
2) First I would like to make it z^4. (shown in the picture)
3) How would I be able to see the prfile after in the mesh?

How do I make it to the normal in x-direction? Here is my code (which is not finished), but is this how can I do it?

    inlet 
    { 
           type                           groovyBC;         
   variables                 "fz=pos().z^4;"; 
   valueExpression "vector(fz,0,0)";     
}

It looks to simple, and not right as well.