CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Surface Tension Force (http://www.cfd-online.com/Forums/openfoam-solving/60581-surface-tension-force.html)

gopala June 14, 2005 06:40

Hello, I am trying to analy
 
Hello,

I am trying to analyze individual forces acting on the interface for every time step. I was not able to understand the implementation of surface tension force in the interFoam solver. Is it exactly the "Brackbill's CSF model" ? It would be very helpful if I can get the expression for surface tension force (volume force) so that I can print it out every time step.

Thanks in advance

/Vinay

henry June 14, 2005 06:55

The surface tension is impleme
 
The surface tension is implemented as

sigma*kappa*grad(gamma)

where kappa is the surface curvature. sigma*kappa is returned by interface.sigmaK() and you should look in interfaceProperties.C to see how the curvature is evaluated. There are a couple of implementations with the less-good alternatives commented out. The force is included into the flux prediction step of the pressure equation (pEqn.H) rather than in the momentum equation:

fvc::interpolate(interface.sigmaK())*fvc::snGrad(g amma)

so that it is a force on the face balanced by pressure gradients and bouyancy. This is equivalent to a staggered-mesh formulation which is more robust and accurate than if it were included as a momentum source and then interpolated. The problem for you now is that surface-tension is not a cell "volume force" but a cell-face force.

gopala June 14, 2005 07:32

Thanks for the quick reply.
 
Thanks for the quick reply.

> The problem for you now is that surface-tension is not a cell "volume force" but a cell-face force:

This means that if I want to print out surface-tension force as volume force then I should use the following expression:

volVectorField Fcsf = interface.sigmaK()*fvc::grad(gamma);

volScalarField magFcf = mag(Fcsf);

is this right ?

/vinay

henry June 14, 2005 08:04

That will give you a surface-t
 
That will give you a surface-tension force as a volume force but it will not be consistent with that actually used in the code, i.e. the integrated force will not balance with the other forces as it should but it will be a rough estimate and a good indicator.


All times are GMT -4. The time now is 02:48.