CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (https://www.cfd-online.com/Forums/ansys-meshing/)
-   -   [GAMBIT] Problem with boundary layer - high skewness (https://www.cfd-online.com/Forums/ansys-meshing/130456-problem-boundary-layer-high-skewness.html)

Muffins February 26, 2014 08:15

Problem with boundary layer - high skewness
 
2 Attachment(s)
Hello, everyone! I'm asking for help with a GAMBIT problem. Trying to create boundary layer on volume but it results with high skewness.

I don't think it's possible to slice this volume into smaller parts. (Actually it is a part of much more complex case). Trying different surface mesh schemes didn't help as well as trying to change defaults for BLAYER.

What I cannot understand is why it doesn't preserve thickness of boundary layer, because even if I managed to reduce skewness <0,97 a layer is still too thin in some areas.

I'm out of ideas, please, suggest something...:confused:

-mAx- February 27, 2014 01:28

yoou have to be sure that your problem comes from the BL.
For that, delete the BL and mesh your volume.

As I see you your surface mesh, I suppose you want to mesh your volume with hexa. With such a topology, you won't be able.
You have to split your volume.
check this tutorial: https://www.sharcnet.ca/Software/Gam...guide/tg03.htm

Muffins February 27, 2014 07:36

Thank you very much for this tutorial. I somehow missed it.
I don't want to mesh this volume with hexa, I only want a hex BL. However now I see it's unlikely to get a good layer without splitting the volume.

-mAx- February 28, 2014 00:25

Ok if you don't want hexa volume, mesh your surface mesh with tri elements.
Your BL will be prismatic.
It should fix your problem.
Then mesh your volume with tet/hexcore (core of your volume with hexaeders)

Muffins February 28, 2014 07:34

I tried triangular meshing and it didn't help. I fiured out that increasing mesh density and thickening BL rows can partly resolve the problem. Of course, the right way is to split the volume but I don't have much time so I'll put up wth it, thank you again.


All times are GMT -4. The time now is 10:51.