CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Any weakness or limits of CFX for secondary flows (http://www.cfd-online.com/Forums/cfx/19513-any-weakness-limits-cfx-secondary-flows.html)

J.T. March 7, 2003 14:41

Any weakness or limits of CFX for secondary flows
 
Any one has the epxerience of running the turbulent secondary flows by using CFX? Thanks

Glenn Horrocks March 9, 2003 17:11

Re: Any weakness or limits of CFX for secondary fl
 
Hi JT,

I think you'll find that CFX5 is actually one of the best commercial codes around for solving things like secondary flows. It is simulations like that where the the coupled solver in CFX5 really shows its strength over the usual uncoupled technique.

If you are talking about CFX4, well it can solve these types of flows but the simulation time will probably be considerably longer due to its use of the uncoupled solver.

Depending on exactly what turbulent secondary flow you are talking about, you may require the Reynolds stress model to simulate it. CFX 4 and 5 both have a number of Reynolds stress models available. The new CFX 5.6 beta version also contains a stress-omega model which I am yet to try out, but looks interesting for low Reynolds number secondary flows.

Glenn Horrocks

J.T. March 12, 2003 19:39

Re: Any weakness or limits of CFX for secondary fl
 
Thank you very much for the infomation! I looked into CFDesign, their results for backward facing step for 3-D turbulent flows are not very good. I am trying to use CFX. Is that easy to get started if you got a pretty complicate geometries? In other word, is their mesh generation automatic or user-friendly? How do you like their post-processor?

Thanks!

Peter March 12, 2003 22:21

Re: Any weakness or limits of CFX for secondary fl
 
The automatic mesh generation in CFX5 still needs a lot of improvement. If your geometry consists of the curved surface or fillet, then you may find that the mesh elements will tend to squish together at this region when you use CFX5 mesh generator. The louisy mesh may destroy the whole solution depending on what problems you have. What I suggest to you is to try importing your mesh from other mesh generation software and solve your problem in CFX5. It does have some excellent solvers anyway.

J.T. March 13, 2003 14:11

Re: Any weakness or limits of CFX for secondary fl
 
Hi, Peter Did you check their benchmark test with turbulent flow over a 3-D lid-driven cavity? I will ask them to provide the tested results. Since ANSYS bought the CFX, do you think ANSYS's mesh generation will provide best results? If I got the answers I will discuss with you. Thanks!

Robin March 16, 2003 10:52

Re: Any weakness or limits of CFX for secondary fl
 
Hi Peter,

Bunching such as this can easily be fixed by setting proper mesh controls.

Robin


All times are GMT -4. The time now is 18:03.