CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   VOF Geo-Reconstruct (

Amir December 6, 2004 14:55

VOF Geo-Reconstruct
Hello, I was wondering if anybody knows whether CFX 5.7 has VOF geo-reconstruct model Thanks Amir

Neale December 7, 2004 22:53

Re: VOF Geo-Reconstruct
No, CFX-5 doesn't use reconstructing volume tracking ala. Hirt, Nichols, Youngs, Kothe, Rider, etc...

It uses a TVD compressive advection scheme for the volume fraction equations instead.


Amir December 8, 2004 16:42

Re: VOF Geo-Reconstruct
Thanks for the hint, I will go through the manual and read about it. Do you have any comments on the accuracy of that comparing to VOF? Thanks

Neale December 15, 2004 20:27

Re: VOF Geo-Reconstruct
Reconstructing volume tracking will probably be a bit better in terms of the fidelity of the approximation because you explicitly track the interface. However significant numerical issues do arise with respect to estimating the interface normal (gradient of the volume fraction). See Rider, Kothe, etc... for more info on that. Reconstruction is very sensitve to how that is done.

For many real world applications a compressive, bounded, advection scheme does an excellent job and is much easier to maintain, extend and allow more feature matrix interactions on unstructured grids than with a reconstruction algorithm.

If you run free surface calculations in CFX-5 you should start with the homogenous MPF model. If that is too diffusive for your application then try inhomogenous, which will allow the two (or more) phases to separate and keeps a sharper interface.


All times are GMT -4. The time now is 19:21.