[pepe.aero]

Hi all,

i'm trying to assign unsteady inlet boundary condition (ramp-shaped UX profile) working on a multiphase problem.
In OF 2.1.1 I can do it using the uniformFixedValue type by this way in the 0/U file:

inlet
{
type uniformFixedValue;
uniformValue csvFile;
csvFileCoeffs
{
fileName "$FOAM_CASE/u_ramp.txt"
outOfBounds clamp;
hasHeaderLine true;
refColumn 0;
componentColumns (1 2 3);
}
}


where the u_ramp.txt is something like the following:

Time,Ux,Uy,Uz
0.00, 0.0, 0.0, 0.0
0.01, 0.0, 0.0, 0.0
0.02, -0.066363, 0.0, 0.0
0.03, -0.132726, 0.0, 0.0
0.04, -0.199089, 0.0, 0.0
0.05, -0.265452, 0.0, 0.0
0.06, -0.331815, 0.0, 0.0
0.07, -0.398178, 0.0, 0.0
0.08, -0.464541, 0.0, 0.0
0.09, -0.530904, 0.0, 0.0
0.10, -0.597267, 0.0, 0.0
0.11, -0.663630, 0.0, 0.0
0.12, -0.730000, 0.0, 0.0
0.13, -0.730000, 0.0, 0.0
0.14, -0.730000, 0.0, 0.0
0.15, -0.730000, 0.0, 0.0


The problem is: Paraview reports this error trying to visualize results:

ERROR: In /home/utkarsh/Kitware/superbuild/paraview/src/paraview/VTK/IO/vtkOpenFOAMReader.cxx, line 6535
vtkOpenFOAMReaderPrivate (0x298a3d0): Wrong list type for uniform field

Is not all. In the variable list available to be visualized I can't find U for inlet boundary, and I find U (of cell) just for internalMesh, while the interpolated one does not appear.

In my idea this is not a post-processing problem, but my doubt is if I setup the unsteady boundary condition properly.
Has anyone experiences with a ramp inlet in OpenFOAM?

Thanks for your support,

Kind regards,

Andrea

[michielm]

I have noticed that ParaView doesn't always understand the uniformFixedValue boundary condition if I load the files from vtk of with a .foam file.

However, if I use paraFoam it works well, because apparently there is some conversion that fixes the problem. Have you used paraFoam to open your dataset?!

[pepe.aero]

Hi Michiel,

thanks a lot for your reply. I was used to use the paraview command instead of paraFoam. You're right, using paraFoam the problem is fixed. Good.

A second issue is: OpenFOAM shows me some convergence problems using the unsteady boundary condition. I'm using a ramp at inlet to grow up gradually the velocity in order to assign a smooth velocity profile and reach convergence in less iterations.
But the result at the moment is that I reach no convergence by this way. So, do I have to change something in the numerical parameters (fvSchemes, fvSolutions) in order to suit unsteady BCs?

Thanks again for your help!

Regards,

Andrea

[michielm]

Hi Andrea,
To be able to answer that I need to know more about the case you are trying to model.
What are you modelling and what are the boundary conditions you are applying?!

[pepe.aero]

Hi,

sure I give you all details.
I'm working on naval application of a hull in a multiphase flow. The problem can be referred to the wigley-hull tutorial, but I'm trying to improve different features, such as an unsteady ramp inlet boundary condition.
The hull is in a 3D rectangular domain and I've attached files regarding numerical parameters and all boundary conditions.
Notice that the initial position of free surface (alpha1 is mass fraction) is set with command setFields just before the start of the computation.

I really appreciate your support!

Kind regards,

Andrea

[michielm]

Hi Andrea,
I can't really find a clear reason for the case not to converge.

Do you have convergences issues in the inner or the outer loop of your iterations?