CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS (http://www.cfd-online.com/Forums/ansys/)
-   -   Meshing problem (http://www.cfd-online.com/Forums/ansys/82391-meshing-problem.html)

derz November 24, 2010 23:11

Meshing problem
 
http://www.flickr.com/photos/derz/5205319511/

See this picture? Notice in the middle of these regions, the mesh is coarser in the middle. How can I eliminate this? I would like the mesh to be as fine as it is at the region-boundaries, all the way through the region...

Any thoughts? I've tried playing around with many options but can't seem to eliminate it with automatic/tets meshing...

Max Efficiency November 25, 2010 06:51

On the left side in the settings look at sizing: When you decrease the Max Tet Size you force Meshing to use smaller (and more) elements to mesh your geometry.

Also try changing the growth rate. To achieve a slower growth in element size choose values close to 1 (e.g. 1.05).

akhokhay November 25, 2010 09:47

Quote:

Originally Posted by Max Efficiency (Post 284757)
On the left side in the settings look at sizing: When you decrease the Max Tet Size you force Meshing to use smaller (and more) elements to mesh your geometry.

Also try changing the growth rate. To achieve a slower growth in element size choose values close to 1 (e.g. 1.05).

this
I also think u set growing rate wrong.
u should inverse it or if boundaries are important use double method(it got different names, iirc in gambit it is bell method!) which get 2 values for each end!

akhenathon November 30, 2010 14:20

its a sizing ans infilation problem

Darren Leong December 2, 2010 22:50

Hi Derz,

1. You'll have more control by editing through CFX-mesh. Reduce the body spacing from there when meshing. There doesn't seem to be an inflation layer around your model, may have problem with the turbulence models later on.

2. You can extend the inner domain to capture the wake area and have the coarser mesh as your outer domain. This eliminates errors at the domain interfaces (lost of flow information). The outer domain should be bigger to reduce the chance of blockage effect.


All times are GMT -4. The time now is 08:18.