Problems with simpleFoam with flow in pipe
 

Hi everyone!! I need a little help. I'm performing a 3D simulation of laminar, incompressible flow in a straight pipe. When analizing my results, i found them a little bit confusing. The thing is, that, instead of finding that the flow is parabolic, it develops into a parabolic-like profile but "translated" from the origin, that is, a parabolic manner but that in the walls is not zero. The detalis of the results can be seen in the attached images.

As boundary conditions for U, i set a parabolic profile with max_vel = 0.46 m/s at inlet, zeroGradient at outlet, and zero for the wall.
For p, i set it to zero at the outlet. (see the o files for U and p attached)

I also attach the fvSolution and fvSchemes files.

In addition, i check the residuals, and they seem to converge nicely (see Ux residual plot attached)

The tubes dimensions are diamater = 0.003 m and lenght = 0.01 m, the cinematic viscosity (nu) is 3e-6, and turbulence is off.

Anyone may have an idea of what is happening?

Extra-comment: I also performed the simulations with a surfaceNomalFixedValue for inlet velocity, and happens the same, the flows develope into this "shifted" parabolic profile.

Helpppp!!! jajajaja

Hi Fernando,
here is a case with surfaceNormalFixedValue as boundary condition and your specifications.
My velocity profile looks ok...
Run "blockMesh" first...
Martin

Thanks a lot!!! Apparently, although i was setting turbulence off, in RASmodel i was setting kepsilonmodel, instead of laminar.... :D. I think that would correct my results!

Graciassssssssss

Hi Martin :),

I am entry level Open Foam User,

When I saw in the system/fvSchemes and system/fvSolution files..I found k and epsilon equation are set to solve..
==================================================
divSchemes
{
default none;
div(phi,U) Gauss upwind;
div(phi,k) Gauss upwind;
div(phi,epsilon) Gauss upwind;
div(phi,R) Gauss upwind;
div(R) Gauss linear;
div(phi,nuTilda) Gauss upwind;
div((nuEff*dev(grad(U).T()))) Gauss linear;
}

laplacianSchemes
{
default none;
laplacian(nuEff,U) Gauss linear corrected;
laplacian((1|A(U)),p) Gauss linear corrected;
laplacian(DkEff,k) Gauss linear corrected;
laplacian(DepsilonEff,epsilon) Gauss linear corrected;
laplacian(DREff,R) Gauss linear corrected;
laplacian(DnuTildaEff,nuTilda) Gauss linear corrected;
}
===========================================


Since the problem is laminar, why are you solving for k and e equation..?

can you help me ???

Hi,

these settings are ignored if simulation is set to be laminar in constant/turbulenceProperties or constant/RASProperties.

Martin

Dear Martin,

Thank You for quick reply...:):)