CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Strange behaviour because of contact angle (interfoam) (https://www.cfd-online.com/Forums/openfoam-solving/83825-strange-behaviour-because-contact-angle-interfoam.html)

 Kim123 January 12, 2011 11:16

Strange behaviour because of contact angle (interfoam)

As this is my first post in this forum I would like to say "Hello" to everybody.

I just started with OpenFOAM and mesh-based CFD in general. I wanted to simulate the process shown on the following picture.

http://img98.imageshack.us/img98/2114/overviewc.png

A viscous fluid is pulled under a blade (by a conveyor belt) to form a thin tape. The conveyor belt speed is given through the lower wall boundary velocity in the desired direction (here x direction). My first approach was using the damn break tutorial as a starting point. Using the settings from the tutorial was quite successfull.The only thing i changed was the the viscosity of the former "water". It was set to 4 Pa s.

I played a little around and used different boundary conditions so that the surface tension effects between the wall and interface are no longer ignored.

After some reading I came up with with following setting for the walls in the alpha1 file:

Wall
{
type constantAlphaContactAngle;
theta0 30;
value uniform 0;
limit none;
}

and the p_rgh:

Wall
{
type fixedFluxPressure;
}

Now strange things happen as seen in the following screenshot:

http://img209.imageshack.us/img209/9...actangle30.png

These effects are directly related to the contact angle and seem to be more extreme at higher viscosity difference between the two fluids. If an angle of 90° is chosen everything works fine.

I was wondering if somebody could tell me the cause of the extreme curving of the surface. In my opinion the contact angle should not have such a big effect on the fluid behaviour in the reservoir. I suspect it might have something to do that simulation is not "relaxed" at the first timestep in the area near the walls (contact angle of 90° because of initial geometry) and the resolution of the mesh.