Interpretation of convergence plots in CFX

Jmac · August 24, 2011, 21:39

Hi all
I am currently developing a cfd simulation of a shower mixer in CFX, and am having some problems with attaining a decent convergence. I understand that possible faults in the model can be found by analysing features within the convergence plots (eg large oscillations), is there a reference that anyone can point me to that explains these features, and what they say about the model?
Thanks,
James

TrazPT · September 2, 2011, 14:17

If your steady-state residuals exhibit large oscillations, it's quite likely your physical flow is inherently time-varying. I suggest running the application as a transient solution, saving some .trn files along the way, and then check to see if the convergence behavior improves within each time step. If so, create a transient animation to find the location of the unsteadiness.

Josh · September 5, 2011, 15:03

Hey James,

Think about it logically. Look at the locations of maximum residuals - is something physically important happening there, or is the mesh quality/resolution poor? You may also find inspiration in an introductory CFD book or this page: http://www.cfd-online.com/Wiki/Ansys_FAQ#Postprocessing

August 24, 2011, 21:39	Interpretation of convergence plots in CFX	#1
Jmac New Member James McLachlan Join Date: May 2011 Posts: 6 Rep Power: 15	Hi all I am currently developing a cfd simulation of a shower mixer in CFX, and am having some problems with attaining a decent convergence. I understand that possible faults in the model can be found by analysing features within the convergence plots (eg large oscillations), is there a reference that anyone can point me to that explains these features, and what they say about the model? Thanks, James

September 2, 2011, 14:17	Research Specialist	#2
TrazPT New Member Mark Troscinski Join Date: Jan 2010 Location: McMurray, PA Posts: 1 Rep Power: 0	If your steady-state residuals exhibit large oscillations, it's quite likely your physical flow is inherently time-varying. I suggest running the application as a transient solution, saving some .trn files along the way, and then check to see if the convergence behavior improves within each time step. If so, create a transient animation to find the location of the unsteadiness.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Flow through porous medium using CFX: lack of convergence	Chander	CFX	3	July 23, 2014 07:32
Creating plots using Ansys CFX	Marios	CFX	4	April 18, 2008 00:24
Relief Plots in CFX	Michael	CFX	15	June 30, 2006 09:01
Convergence problems	Chetan	FLUENT	3	April 15, 2004 20:13
CFX 4.4 installation problem	Pandu Sattvika	CFX	1	December 1, 2001 05:07

September 5, 2011, 15:03		#3
Josh Senior Member Joshua Counsil Join Date: Jul 2009 Location: Halifax, Nova Scotia, Canada Posts: 366 Rep Power: 18	Hey James, Think about it logically. Look at the locations of maximum residuals - is something physically important happening there, or is the mesh quality/resolution poor? You may also find inspiration in an introductory CFD book or this page: http://www.cfd-online.com/Wiki/Ansys_FAQ#Postprocessing