CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > OpenFOAM

Why OpenFOAM much slower than fluent in NACA0012 simulation (transsonic situation)

Register Blogs Members List Search Today's Posts Mark Forums Read

Like Tree2Likes
  • 1 Post By akidess
  • 1 Post By olivierG

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   October 22, 2016, 05:14
Default Why OpenFOAM much slower than fluent in NACA0012 simulation (transsonic situation)
  #1
New Member
 
shadow
Join Date: Aug 2013
Location: China
Posts: 4
Rep Power: 11
xucloud77 is on a distinguished road
NACA0012 is a basic benchmark in aerodynamics. We run a simple case with M=0.8 and attack angle = 1.25, which is a compressible steady state problem.

In OpenFOAM, we use the rhoCentralFoam solver. It takes 58000 seconds CPU time to reach the final steady state. With the same grid and coeffs, Fluent only takes 600 seconds CPU time to get convergence. In Fluent, we selected the steady problem and density based solver.

Does any body know the reason why Fluent is 100 times faster than OpenFOAM? We are really shocked by the results.

I guess this may be caused by several reasons:
  1. Under the restrict of CFL condition, the delta time is nearly 1.0e-7. The boundary layer grids have large length-width ratio, will it affect the delta T?
  2. I doubt about the time stepping in steady problems. Fluent does not have setting about time stepping but iteration for convergence.
These are about the models of solver. But other optimizations such as boundary condition, precondition of flow may be added to Fluent.
xucloud77 is offline   Reply With Quote

Old   October 24, 2016, 03:47
Default
  #2
Senior Member
 
akidess's Avatar
 
Anton Kidess
Join Date: May 2009
Location: Germany
Posts: 1,377
Rep Power: 28
akidess will become famous soon enough
What about using rhoSimpleFoam?
newbee09 likes this.
__________________
*On twitter @akidTwit
*Spend as much time formulating your questions as you expect people to spend on their answer.
akidess is offline   Reply With Quote

Old   October 24, 2016, 04:23
Default
  #3
New Member
 
shadow
Join Date: Aug 2013
Location: China
Posts: 4
Rep Power: 11
xucloud77 is on a distinguished road
Akidess, thank you for your reply.

We have used rhoSimpleFoam to run the case under OpenFOAM-4.0. Without changing the configuration, the case did not convergence got float point exception error.

I think there must be some problem in the settings of flux. The default flux scheme is linear.

I put all the case files and log here.
https://dl.dropboxusercontent.com/u/...SimpleFoam.zip

Last edited by xucloud77; October 24, 2016 at 09:02. Reason: correct
xucloud77 is offline   Reply With Quote

Old   October 24, 2016, 05:13
Default
  #4
Senior Member
 
Olivier
Join Date: Jun 2009
Location: France, grenoble
Posts: 272
Rep Power: 16
olivierG is on a distinguished road
Hello,

Your case look really strange:
- you set the turbulence model to laminar
- but you set in 0 dir k/epsilon ... and use wall function on empty patch....
- T and p seem also strange (p: fixedValue !)
- use "slip" on airfol for T and p
- do not use GAMG for p solver
- mesh seem to have high aspect ratio too.
...

Is this really the case you are comparing to Fluent ?
If yes, your case is really misconfigured.


regards,
olivier
newbee09 likes this.
olivierG is offline   Reply With Quote

Old   October 24, 2016, 09:08
Default
  #5
New Member
 
shadow
Join Date: Aug 2013
Location: China
Posts: 4
Rep Power: 11
xucloud77 is on a distinguished road
Hello olivierG, sorry for my careless. I correct the case, and rerun it. But the problem happened again. Any other option should be corrected or add to make rhoSimpleFoam work and get the correct results?

I think this kind of steady state pressure-based solver could be more efficient than rhoCentralFoam. But now the most urgent is to make rhoSimpleFoam work.

https://dl.dropboxusercontent.com/u/...SimpleFoam.zip
xucloud77 is offline   Reply With Quote

Old   October 24, 2016, 10:23
Default
  #6
New Member
 
Lee Howe
Join Date: Dec 2014
Posts: 2
Rep Power: 0
leehowe is on a distinguished road
Quote:
Originally Posted by xucloud77 View Post
Hello olivierG, sorry for my careless. I correct the case, and rerun it. But the problem happened again. Any other option should be corrected or add to make rhoSimpleFoam work and get the correct results?

I think this kind of steady state pressure-based solver could be more efficient than rhoCentralFoam. But now the most urgent is to make rhoSimpleFoam work.

https://dl.dropboxusercontent.com/u/...SimpleFoam.zip
Maybe you can try foam-ext version, I know there are several compressible steady state solvers such as steadyCompressibleFoam.
leehowe is offline   Reply With Quote

Reply

Tags
efficience, fluent, openfoam, steady state

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On


Similar Threads
Thread Thread Starter Forum Replies Last Post
Running UDF with Supercomputer roi247 FLUENT 4 October 15, 2015 14:41
should I convert from FLUENT to OpenFOAM ? mrenergy OpenFOAM 7 December 12, 2013 13:40
New OpenFOAM Forum Structure jola OpenFOAM 2 October 19, 2011 07:55
Fluent elbow in Openfoam chemeng OpenFOAM 1 January 21, 2010 04:52
OpenFOAM vs Fluent for cylinder at Re%3d150 lr103476 OpenFOAM Running, Solving & CFD 40 December 18, 2008 10:09


All times are GMT -4. The time now is 09:38.