CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   GAMBIT meshing scheme (

rayy February 12, 2007 07:54

GAMBIT meshing scheme
Hi all...

I m a new user & I got model consists of 3 part of cylinder with different diameter. Then I union my model and become 1 volume. The middle part i only can mesh it with tet/hybrid and the other 2 parts I use Hex scheme. The question is:

1.To make meshing successful, does it mean that I shouldn't union my model but rather let it be 3 separated part with different meshing scheme?

2.If we use different scheme and different interval size on 1 model(consists of 3 parts?, does it affect the quality/accuracy of simulation?

3.Among few options on meshing scheme, which one offer better quality simulation?

Thanks in advance


Kasper Skriver February 12, 2007 08:22

Re: GAMBIT meshing scheme
1) The best option is what you've done I think. Since there is no interfaces here.

2) Maybe..

3) Hex offer the best quality solution whereas, tet is worse. (Has to do with the skewness of the cells)

The only other option is polyhedral in STAR-CCM+. This will allow you to mesh the entire volume with polyhedrals.

BR Kasper Skriver

rayy February 12, 2007 08:51

Re: GAMBIT meshing scheme
Thanks for ur answer Kasper.

I used GAMBIT and had tried mesh the entire volume with 1 scheme (tet/hybrid ). then i ckecked the histogram of meshing quality, i got bell shape histogram with the value below than 1. The worse cell is 0.9. Does this mean that my model is OK?

Kasper Skriver February 12, 2007 10:15

Re: GAMBIT meshing scheme
Depends on what kind of quality you checked for.

Was it equi angle skew(Q_EAS), equi size skew(Q_EVS) or one of the other options?

Both Q_EAS and Q_EVS shouldn't lie near 1, but near zero instead. 0,9 < Q_EAS < 1,0 is described as "very poor" quality.

Take a look at the Gambit help. It is quite usefull. User guide--> Index --> Mesh(Quality...)

rayy February 12, 2007 10:26

Re: GAMBIT meshing scheme
Thanks a lot for the advice.

All times are GMT -4. The time now is 21:30.