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

CFX vs Fluent convercenge residuals

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

Like Tree3Likes
  • 1 Post By ghorrocks
  • 1 Post By thomasmoura
  • 1 Post By ghorrocks

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   June 3, 2017, 16:57
Default CFX vs Fluent convercenge residuals
  #1
New Member
 
Join Date: May 2017
Posts: 3
Rep Power: 9
thomasmoura is on a distinguished road
Hello all,

I'm doing some comparisons between CFX and Fluent in terms of speed and accuracy.

I've done CFX runs where I set the target residuals at 1e-6 (as company methodology). When doing the Fluent runs I've came across that the recommended values were 1e-3...

Anyone could explain me why this 3 order magnitude difference, as this has direct impact in the run duration?
thomasmoura is offline   Reply With Quote

Old   June 4, 2017, 01:55
Default
  #2
Member
 
Quang Le Dang
Join Date: Jun 2012
Posts: 61
Rep Power: 14
SPH_CFD is on a distinguished road
Quote:
Originally Posted by thomasmoura View Post
Hello all,

I'm doing some comparisons between CFX and Fluent in terms of speed and accuracy.

I've done CFX runs where I set the target residuals at 1e-6 (as company methodology). When doing the Fluent runs I've came across that the recommended values were 1e-3...

Anyone could explain me why this 3 order magnitude difference, as this has direct impact in the run duration?
1e-3 is just default in numerical theory (lower this value incraeses the accuracy of solution). But we always set to 1e-6 for convergence criteria.

Sent from my SM-A520F using CFD Online Forum mobile app
SPH_CFD is offline   Reply With Quote

Old   June 4, 2017, 07:28
Default
  #3
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,863
Rep Power: 144
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
I suspect CFX and Fluent calculate the residuals by different methods, so the numbers need to be compared with care.

To compare CFX vs Fluent objectively you need to compare a well resolved model in both cases. This means mesh independent, adequate convergence and any other tuneable parameters correctly set. It also means that the same settings are not necessarily the best for both - for instance the mesh which gives mesh independence could well be different in both software. Do not assume the same mesh is suitable for both codes.
elia_l likes this.
ghorrocks is offline   Reply With Quote

Old   June 4, 2017, 09:14
Default
  #4
New Member
 
Join Date: May 2017
Posts: 3
Rep Power: 9
thomasmoura is on a distinguished road
Quote:
Originally Posted by SPH_CFD View Post
1e-3 is just default in numerical theory (lower this value incraeses the accuracy of solution). But we always set to 1e-6 for convergence criteria.

Sent from my SM-A520F using CFD Online Forum mobile app
Yes I understand that, but the "default" of CFX is 1e-4... Do you currently set 1e-6 in Fluent?
thomasmoura is offline   Reply With Quote

Old   June 4, 2017, 09:23
Default
  #5
New Member
 
Join Date: May 2017
Posts: 3
Rep Power: 9
thomasmoura is on a distinguished road
Quote:
Originally Posted by ghorrocks View Post
I suspect CFX and Fluent calculate the residuals by different methods, so the numbers need to be compared with care.
Yes I suspect the same, so I am asking if anyone can confirm this and have a guideline like "fluent residuals are 10x greater" or something like that.
elia_l likes this.
thomasmoura is offline   Reply With Quote

Old   June 4, 2017, 11:14
Default
  #6
Member
 
Quang Le Dang
Join Date: Jun 2012
Posts: 61
Rep Power: 14
SPH_CFD is on a distinguished road
Quote:
Originally Posted by thomasmoura View Post
Yes I understand that, but the "default" of CFX is 1e-4... Do you currently set 1e-6 in Fluent?
Yes, always 1e-6 (at least) for both FLUENT và CFX or the orther commercial softwares.
SPH_CFD is offline   Reply With Quote

Old   November 7, 2018, 16:25
Default
  #7
New Member
 
Join Date: Oct 2018
Posts: 22
Rep Power: 8
elia_l is on a distinguished road
Quote:
Originally Posted by ghorrocks View Post
I suspect CFX and Fluent calculate the residuals by different methods, so the numbers need to be compared with care.

To compare CFX vs Fluent objectively you need to compare a well resolved model in both cases. This means mesh independent, adequate convergence and any other tuneable parameters correctly set. It also means that the same settings are not necessarily the best for both - for instance the mesh which gives mesh independence could well be different in both software. Do not assume the same mesh is suitable for both codes.
I think the same.. Residuals for fluent "deal more" with imbalances (conservation equations); for CFX should be more or less the difference between values that the same variable (p, u, v, w) assumes for two consecutive iterations.


Please correct me if I'm wrong.
elia_l is offline   Reply With Quote

Old   November 7, 2018, 18:36
Default
  #8
Super Moderator
 
Glenn Horrocks
Join Date: Mar 2009
Location: Sydney, Australia
Posts: 17,863
Rep Power: 144
ghorrocks is just really niceghorrocks is just really niceghorrocks is just really niceghorrocks is just really nice
I don't about know the residuals in Fluent, but for CFX the residuals are calculated by a propriety algorithm which has only a general outline described in the documentation. So it is not possible to compare the exact approach as it is not public.
elia_l likes this.
__________________
Note: I do not answer CFD questions by PM. CFD questions should be posted on the forum.
ghorrocks is offline   Reply With Quote

Reply

Tags
convergence criteria

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
[ANSYS Meshing] Solver preference - CFX vs FLUENT gcoopermax ANSYS Meshing & Geometry 1 May 31, 2017 20:17
Mesh and Solve Times for CFX, Fluent, CD-adapco Jade M Main CFD Forum 4 August 28, 2012 03:54
Fluent and CFX Ale Main CFD Forum 7 July 30, 2008 22:14
Jobs in cfd - fluent or cfx? jobman Main CFD Forum 6 July 5, 2006 16:02
CFX vs STAR-CD and Fluent William CFX 20 October 29, 2001 14:55


All times are GMT -4. The time now is 01:46.