CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)
-   -   OpenFOAM 1.7.1: Oscillating CFL number (https://www.cfd-online.com/Forums/openfoam-solving/80111-openfoam-1-7-1-oscillating-cfl-number.html)

LarsPT September 15, 2010 19:24

OpenFOAM 1.7.1: Oscillating CFL number
 
Hi,

I want to simulate the capillary rise of a fluid in a closed rectangular container in a low-gravity environement using interFoam in OpenFOAM 1.7.1. So, I have only walls and symmetryPlanes as BCs. The container has a square cross section with length 1 cm and a height of 10 cm. The contact angle of the fluid is 0.

Here are the BCs (symmetryPlane is of course of type symmetryPlane):
Code:

alpha1:
walls
{
  type    constantAlphaContactAngle;
  theta0 0;
  limit    gradient;
  value  uniform 0;
}

p_rgh:
walls
{
  type    fixedFluxPressure;
  adjoint no;
}

U:
walls
{
  type    fixedValue;
  value  uniform (0 0 0);
}

My problem is, that the CFL number is kind of oscillating. So it increases up to 0.2 then makes a jump up to 8 or 10 and then it jumps back to 0.1 or even lower. After a while it crashes with sigfpe error. I already changed deltaT and maxDeltaT but it didn't help. The new capillaryRise tutorial doesn't work either. It shows the same phenomena and crashes with sigfpe error.

So is there someone else, who has had the same problems or who can help me?

Thanks in advance!

Lars

ata September 17, 2010 02:07

OpenFOAM 1.7.1: Oscillating CFL number
 
Hi Lars
Did you examine your case with bigger contact angles?
Best regards
Good luck

Ata

LarsPT September 17, 2010 09:40

Thanks for your answer.

No, I didn't, yet. But maybe it's a good idea. For now I tried to use a more viscous fluid to reduce the dynamics of the rise.

I read some papers about CFD (VOF) and contact angles. Is there any documentation how interFoam deals with the singularity of the no-slip BC?

Lars

ata September 18, 2010 01:50

Hi Lars
How are you?
I hope you are doing very well.
What you mean "singularity of the no-slip BC"?
You can see the PhD thesis of Onno Ubbink entitled by:
"Numerical prediction of two fluid systems with sharp interfaces"
I hope that help you.
Best regards

Ata

LarsPT September 18, 2010 03:55

So, I tried OpenFOAM-1.6 and it's performs stable but the results are not really good. 40 to 50 % error to the analytical comparison. I think, I have to work on the mesh and on the time step to get some higher accuracy. OpenFOAM-1.7.1 still crashes after a while. I don't think it's a bug, it is only because of the special case.

Now to the singulartiy and the no-slip condition. Usually the velocity at the wall should be zero but the fluid has to move anlong the wall in order to wet the surface. I read about an aproach of disabling the no-slip BC and replace it by a no-friction BC. The pressure loss is then compensated by an artificial body force. The intensity of the body force is unfortenatly a fit parameter and depends on the case.

By the way, I have the PhD thesis of Ubing and the one of Rusche, but I didn't get too much into it, yet. I think it will be worth doing so. Thank you for your help! But I still wonder why OF-1.7.1 has so many problems with this case while it is working with the former version!?

Lars


All times are GMT -4. The time now is 18:37.