[vkannama]

Dear all,

I am new to openfoam and trying to solve a simple 3D cylinder laminar flow problem. I obtained the geometry from Cylinder mesh using blockMesh with m4 macro and, using paraview, everything is good with geometry. The BCs are: inlet velocity is 1m/s and outlet pressure is 0. I earlier solved this problem in Nek5000 and both the velocity and pressure-drop matches with the theoretical values. However, in openfoam, the velocities and pressure are both strange. For example, the axial pressure drop is supposed to be linear, while my result is shown attached.

Please point any mistakes from the relevant input files pasted below, or anything else I might be doing. I will be grateful.

Thanks,
Ajay

******** controlDict file ******
application icoFoam;

deltaT 0.005;

******** fvSolution ******
solvers
{
p
{
solver PCG;
preconditioner DIC;
tolerance 1e-07;
relTol 0;
}

pFinal
{
$p;
relTol 0;
}

U
{
solver smoothSolver;
smoother symGaussSeidel;
tolerance 1e-07;
relTol 0;
}
}

PISO
{
nCorrectors 2;
nNonOrthogonalCorrectors 0;
pRefCell 0;
pRefValue 0;
}

******** velocity BC/IC (U) ******
dimensions [0 1 -1 0 0 0 0];

internalField uniform (1 0 0);

boundaryField
{
inlet
{
type fixedValue;
value uniform (1 0 0);
}

walls
{
type noSlip;
}

outlet
{
type zeroGradient;
}
}
******** pressure BC/IC (p) ******
dimensions [0 2 -2 0 0 0 0];

internalField uniform 0;

boundaryField
{

inlet
{
type zeroGradient;
}

outlet
{
type fixedValue;
value uniform 0;
}

walls
{
type zeroGradient;
}
}

[Gret]

I have the same problem

[alexeym]

Hi,

a simple 3D cylinder laminar flow problem

Could you elaborate? Basic question is "Why icoFoam?"

[Saideep]

Hi;

I did not use blockmesh but generated an .stl and ran my simulations using icoFoam and I get reasonable results.

1. I am not sure if 'noSlip' is a keyword added to latest versions of OpenFOAM but you can replace it with
wall
{
type fixedValue;
value uniform (0 0 0);
}

2. I guess the time step size is quite high. Try with 5e-7.

3. If possible try to refine your mesh. The finer it is, I see the results converge better to reality.

Let me know if you want my case file. I will drop it.

[Saideep]

Found my case easier. Attached the case.

Just run:
1. surfaceFeatureExtract
2. blockMesh
3. snappyHexMesh -overwrite
4. icoFoam

[vkannama]

Originally Posted by Saideep

Found my case easier. Attached the case.

Just run:
1. surfaceFeatureExtract
2. blockMesh
3. snappyHexMesh -overwrite
4. icoFoam

Hi Saideep, this worked very nicely! I see a parabolic velocity profile and linear pressure drop. Don't really understand the steps you've written. I will get back to you if I have any trouble. Thank you.

Originally Posted by alexeym

Hi,
Could you elaborate? Basic question is "Why icoFoam?"

Hi Alex, I thought icoFoam is the simplest solver for my purpose (laminar, not steady state). Are there any alternatives? Please let me know. Would like to try others.