CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   modeller/mesher (http://www.cfd-online.com/Forums/fluent/27919-modeller-mesher.html)

Shane Gillis September 30, 2000 17:29

modeller/mesher
 
Are there any other good Modellers & meshers for use with Fluent other than Gambit?

John C. Chien September 30, 2000 18:38

Re: modeller/mesher
 
(1). Yes, I think so. (2). Before they invent the Gambit with the help of another company, Fluent uses another company's product called Geomesh by ICEM CFD company. (3). Geomesh is a simpler version of ICEM's other product such as ICEM/ddn, ICEM/HEXA, ICEM/TETRA , ...(4). So, Fluent used to depend on another company's product like Geomesh. And today, you can still use geomesh, or ICEM's Hexa or Tetra. I have been using ICEM/HEXA. It's based on a separate concept of blocking topology and projection on the geometry. (5). Before Gambit, I also have used Patran. So, basically, Fluent is not a company which is good at geometry and mesh generation. It depends on another company's products all the time. (6). So, to answer your questions, you can use Geomesh from ICEM cfd, ICEM/Hexa or ICEM/Tetra, or Patran. (7). The integration of modules for commercial cfd codes has been one main issue. As a result, the user has to spend a lot of time and effort to go through different modules, sometimes using the same command for different tasks, to get the results. It is not a very productive approach. (8). So, using these CAD like geometry and mesh generation is one way to get the job done. But it is extremely time consuming, including Gambit, if you are dealing with complex and routine problem. (most are these type anyway) (9). I am tired of these commercial approaches. So, I will probably stop answering the questions soon. It is not the right approach for cfd in the long run. In the short term, they can be used to keep the subject of cfd alive. So, from that point view, it is quite useful. I mean you can check out these codes. They are still quite useful. (But not effective).

Greg Perkins October 1, 2000 23:24

Re: modeller/mesher
 
John,

you comment that 'It is not the right approach for cfd in the long run' - could you expand on this comment. What is your vision for cfd in 'the long run'??

Regards

Greg Perkins

John C. Chien October 2, 2000 09:51

Re: modeller/mesher
 
(1). send e-mail, we can exchange ideas, if I have the time.

Kai Kang October 2, 2000 10:40

Re: modeller/mesher
 
Gambit is made very general CAD-oriented program, if you are doing something specific, like mixing studies, you might want to get MixSim, or for HVAC: Airpak. They have specially designed mesher/modellers that makes the modelling easier. Many other companies also have such products and you just need to shop arround.

Depends on the application.

David Minns October 2, 2000 13:58

Re: modeller/mesher
 
Hi,

The best thing I can say to you is:

www.icemcfd.com

ICEM CFD is one of the best tools for CFD mesh generation, over the years we have provided meshing for:

Fluent STAR CD CFX

Regards

David

Joern Beilke October 2, 2000 14:58

Re: modeller/mesher
 
And if you are not happy with the mesh-quality of ICEM/Hexa you can use Gridpro (www.gridpro.com) :)

Chetan Kadakia October 3, 2000 06:23

Re: modeller/mesher
 
You can also use ANSYS, NASTRANS, or IDEAS. These packages are more often used for FEA, but I import grids from ANSYS all the time. I believe Fluent also puts out a program called T-Grid which is supposed to be for more complicated meshes, if I am correct. What problems do you have with Gambit? Let us know which package you find to work out the best.

Shane Gillis October 3, 2000 10:11

Re: modeller/mesher
 
Gambit wasn't loading & at the time, I did not know why, so I was looking for alternative meshers & modellers in case I found it necessary to switch.

John Chawner November 16, 2000 11:58

Re: modeller/mesher
 
You might consider Gridgen. See http://www.Pointwise.com


All times are GMT -4. The time now is 11:50.