CFD Online Discussion Forums

CFD Online Discussion Forums (
-   STAR-CCM+ (
-   -   Free surface modelisation (

mintsa November 10, 2011 12:18

Free surface modelisation
I want to simulate a free surface, the geometry is a parallelepiped, I apply gaussain pressure on the top face, I want to see how the free surface moves.
I used the VOF model but I can not visualize the deformation, I can use morphing for this?

Thanks You

abdul099 November 11, 2011 15:02

What do you want to do???

To see the free surface, create an iso surface and apply the volume fraction field function with an iso value of 0.5.
Don't take it the wrong way, I'm happy to help with any question. But I recommend to read in your preferred literature about the principles of the VOF model. You don't seem to be aware what VOF means and how it works. Morphing means deforming the mesh, which is only neccessary to change the geometry of your domain.

mintsa_icsd November 14, 2011 04:12

Thank You for your reply abdul099
But I used VOF model, I saw the interface between the two phases, but the result was wrong. I had apply gaussian pressure on the top surface (because it impossible to apply it on the free surface), but the second phase disapear at the end of simulation. Or normally there must have just a deformation on the free surface.

abdul099 November 14, 2011 18:28

Okay. My car makes a strange noise. Can you tell my what's broken? I suppose, you can't since you don't have any information about the car and the noise. That's the same with your problem. Nobody can help you when you don't describe your problem a little bit better! There's no visionary here!

You can't apply anything to the free surface since it is only an iso-surface of a specific quantity with a specific value. It is no boundary. And you don't want to apply anything on the free surface since it is a RESULT of your simulation!

When a phase disappears, check your setup. Especially boundary conditions, mesh and time step. The same to do when the result is "wrong", check your settings. You have to make sure, your mesh is fine enough, your time step is small enough, you use enough inner iterations etc... When your setup is right, you will get an appropriate result.

By the way, I'm just a retard who doesn't know about Gaussian pressure. What is it and what do you want to see?

mintsa November 15, 2011 04:29

I really like your metaphor of the car.
To do simple, my system is a glass of water, blowing over the surface of the water then the surface is deformed.
I want to see this deformation.
I just modeled the contents of the glass, starting fraction volume is 0.5.
The sides of the above is an pressure outlet ( where I fixed Gaussian pressure P= A*exp(Rē/rē)).
all others are walls (no-slip).
I arrive visualized deformation (on the iso-surface) but in the end there is no longer of water in the glass (which is not logical given the pressure)
The chosen models are Eulerian multiphase, gravity, Implicit Unsteady, laminar, Multiphase Interacrion, Multiphase Mixture, segregated flow, VOF, three dimensional.
Thanks for help

abdul099 November 15, 2011 18:15

Okay, much better now ;) I've got an idea what could be wrong, but still need some information. To make it easier for you, I will ask for specific information...

What do you mean with "I just modeled the contents of the glass, starting fraction volume is 0.5."? Did you model only the liquid phase or is there a gas phase above the liquid level? Is it a "deep" or a shallow glass? How big is the diameter, how long is it?
Does your pressure boundary drive a significant flow? Which velocities can be seen?
What about the starting volume fraction, did you initialize the whole domain uniformly with 0.5?

mintsa_icsd November 16, 2011 05:29

I modeled gas phase above liquid phase (but the geometry is just a cylinder). The cylinder represents a shallow glass ( the height is 0.05 m, the diameter is 0.06m).
I think that pressure boundary drives flow, because I can see in the same surface a reversed flow.
I sow just the air velocity (imposible for me to see the water velocity).
And yes the starting volume fraction is intialized uniformly with 0.5 (I used this field function $$Position[2]<=0.025)?1:0.

All times are GMT -4. The time now is 13:30.