CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   vof grid independence and validation (http://www.cfd-online.com/Forums/fluent/44139-vof-grid-independence-validation.html)

vof_grid March 16, 2007 07:14

vof grid independence and validation
 
Can some one in the forum suggest the best grid to be used for free surface flow in 0.2 mm x 4 mm sized 2D channel. I would also wish to validate my simulation with literature.

I could not locate relavent literature discussing on grid independence for vof study in micro sized channels using fluent.

Sujith March 17, 2007 06:22

Re: vof grid independence and validation
 
I dont think a grid independence study can be done on VOF model. The vof simulations are having strong dependency on the grid. It will be better to hav a very fine mesh at the interface regions. The other regions u can hav a coarse mesh.

vof_grid March 17, 2007 16:03

Re: vof grid independence and validation
 
thanks for your post. as i go from coarser mesh to finer mesh the meniscus displacements are not same. they tend to decrease with finer mesh in the case of capillary flow in channel.

Sujith March 18, 2007 18:15

Re: vof grid independence and validation
 
a very fine grid is recommended for vof. the interface interpolation schemes(geo-reconstruction) are more accurate at finer meshes.

Bak_Flow March 19, 2007 15:40

Re: vof grid independence and validation
 
Dear VOF,

the issue here is that the physical problem has a discontinuity. The same issue arises for shock capturing in high speed compressible flows. CFD will always spread the discontinuity more than the real physical problem BUT what you have to determine is what level of smearing is acceptable for your simulation goals.

The vof discretizations will at best smear over 3 cells...what you really want to see is red and blue...but you get a range of colors at the interface. How much this smearing will affect your results si the question you need to answer.

For example if you have some integral quantity eg. drag on a hull. The smearing will lead to a higher than expected Volume of liquid above the true water line and lower than expected values for volume of liquid below the water line. However the integral over the hull may be fairly accrate since there can be a cancellation.

If on the other hand you need to know if any water leaks into a hole you just drilled....you had better get a sharp resolution...something like 1/10 of the hole height is a good start!

I hope this is of help.

Best Regards,

Bak_Flow


All times are GMT -4. The time now is 22:25.