|
[Sponsors] |
|
April 1, 2003, 06:42 |
Gambit probs please help!!!
|
#1 |
Guest
Posts: n/a
|
I've been having great difficulties meshing a model of a formula one cars front geometry I have meshed my faces using a quad/tri pave I get the following error message when I mesh the volume :
Initialization failed to mesh 32 nodes Error: TG_Mesh_Domain failed with error code 1 Im using a tetrahedral/hybrid mesh The largest skewed element I have is .69 Please help, I think im going to go mad Thanks Steve |
|
April 1, 2003, 08:30 |
Re: Gambit probs please help!!!
|
#2 |
Guest
Posts: n/a
|
Sometimes gambit fails to create pyramides into the right direction (due to misspointing surface normals), hence they try to penetrate existing cells or the side triangles of the pyramides are missing for the tet-meshing tool.
Try to export your surface grid and have a look to the facette normals with tgrid. If necessary flip them and try to generate the mesh within tgrid (with tgrid 3.4 even pyramides are created automatically). If you want to be sure that the pyramides are created in the right way, mesh them first manually. Volker |
|
April 1, 2003, 11:00 |
Re: Gambit probs please help!!!
|
#3 |
Guest
Posts: n/a
|
Thanks for the advice but i dont have easy access to tgrid at the moment is there anything i can do in gambit?
Thanks |
|
April 2, 2003, 03:37 |
Re: Gambit probs please help!!!
|
#4 |
Guest
Posts: n/a
|
If the reason is really what I described and you don't want to use triangles only: no. Maybe with gambit 2.1 there will be a workaround.
The only remedy at the momemt is to avoid quadrilateral facettes. Let me hear whether it worked |
|
April 3, 2003, 06:10 |
Re: Gambit probs please help!!!
|
#5 |
Guest
Posts: n/a
|
Hi thanks for the reply again. Im still fairly new to meshing so this may be a bit of a dumb question but hpw would i generate it just using triangles i have meshed all the surfaces with a triangle pave but still came up with the same problem before it comes up with the final message it displays this:
initialization failed; perturb boundary nodes and try again what does this mean and is there a was of doing this? Any help would be great Thanks Steve |
|
April 3, 2003, 07:02 |
Re: Gambit probs please help!!!
|
#6 |
Guest
Posts: n/a
|
This means that the tgrid process which runs in the background of gambit cannot create its initial unrefined mesh which connects all existing nodes.
After initialisation is done tgrid/gambit refines the coarse grid until it looks useable. Initialisation fails to perform due to different reasons:e.g.: -surface-mesh is too coarse -there are highly skewed surface-cells -there are zones with high size ratios (small faces aside of big ones) -there are surface-cells with high aspect ratios Try to refine your surface grid. And try to get tgrid, since it is much better for analyzing and locating surface grid defects. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Converting Fluent mesh files to Gambit neutral files | samuelkc | ANSYS Meshing & Geometry | 9 | April 21, 2021 06:28 |
Gambit help: Cube inside cube | Jack Martinez | FLUENT | 13 | August 11, 2010 06:29 |
Gambit 2.3.16, Xming or Exceed...HELP! | bambam3417 | FLUENT | 10 | May 7, 2010 12:39 |
Gambit -dev X11 doesn't work | Ervin Amet | FLUENT | 0 | October 28, 2007 08:33 |
Athlon 64 Gambit Compatibility? | Derick | FLUENT | 3 | January 17, 2006 17:28 |