[Mohsin]

hello

Could anyone please tell me how can i mesh the following figure minimum skewness (equiangle skew should be less than 0.6). It is actually a part of a cylinder with 3 inner loops i-e the cylinder is connected with 3 pipes from where air will enter. I tried Cooper but its not possible. Tetrahedral is possible but it gives me high skewness greater than 0.85.

Is there any tricky way like real splitting, virtual splitting, decomposing etc etc through which i can apporoach this case.

I will be grateful for help. Thanks
Mohsin

[-mAx-]

post a picture from your 3d domain.

[Mohsin]

thats the picture.

The middle portion of the cylinder.

I actually i created this red portion by purpose in order to apply cooper meshing scheme above and below that portion. But because of the three pipes circular regions cooper meshing cannot be applied in that red portion and tetrahedral gives high skewness.

[-mAx-]

You can mesh it wih cooper: try to split your volume by enforcing gambit to project each cylinder intake following a special direction
I did it very quick, and of course you can expect having better quality (hardly linked with angle of splits)
Maybe there is other tricks... but this one works.
Why do you want max skewness about 0.65? (it is quite low)
Sans titre.jpg

[Mohsin]

thank you Max.

before you were posting here i tried tetrahedral mesh and used size function approach. In that red portion i used "fixed size function" with 1.5 growth rate which gave me about 0.74 equiangle skewness.

Cooper meshing generally gives good quality. Your approach is very good as u splitted the circular regions and applied mesh. I will try it on my geomtery and will see the results and will post in a while.

As u asked about why i want below 0.65. Thats because lesser skewness greater mesh quality. 0.5 to 0.65 around is considered a good quality mesh design.

thank ypou again MAx

[-mAx-]

I used everyday tet-hexcore with combination of size function. This way is very powerfull since you can reach a "good" quality (max skewness 0.9) for complex geometry in few hours.
From my side (and my applications) max skewness is enough (and good results in comparison with experiments