RMS Mass Vapour Convergence Issue in CFX

nomi.tahir · June 28, 2016, 03:01

I am facing trouble in convergence of RMS Mass Vapour in CFX. All other RMS criteria are well met but RMS Mass Vapour value is 3.1E-3 and it decreases to 3E-3 after more than 150 iterations. I am dealing with subcooled boiling with constant heat flux in assembly. What should I do ? I mean could there be a way to ensure that it converges like there are under relaxation factors in Fluent. What should I do for convergence of RMS Mass Vapour in this case ? Your kind suggestions would be appreciated. Thanks!!

ghorrocks · June 28, 2016, 06:43

First of all: Please do not PM me with exact copies of posts on the forum. It really annoys me. If it is on the forum I will see it. I don't answer CFD questions by PM anyway.

CFX does not have the wide range of tunable Fluent has because it does not need them. The coupled solver is very robust for most applications and if it does not converge then usually it means you have a fundamental problem which cannot be fixed by simply tuning the solver constants.

As you are doing a phase change simulation you should expect problems in convergence. This FAQ will give you some general tips: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

nomi.tahir · June 28, 2016, 10:10

When I change my working fluid to water and use higher heat flux it does indeed converge . But on the other hand for any other volatile fluid at relatively lower pressure and at comparatively lower heat flux the issue remains. What do you suggest for it ?

ghorrocks · June 28, 2016, 20:53

Small time steps, double precision numerics, improve mesh quality. That is a starting point.

June 28, 2016, 03:01	RMS Mass Vapour Convergence Issue in CFX	#1
nomi.tahir New Member Join Date: Mar 2016 Posts: 11 Rep Power: 10	I am facing trouble in convergence of RMS Mass Vapour in CFX. All other RMS criteria are well met but RMS Mass Vapour value is 3.1E-3 and it decreases to 3E-3 after more than 150 iterations. I am dealing with subcooled boiling with constant heat flux in assembly. What should I do ? I mean could there be a way to ensure that it converges like there are under relaxation factors in Fluent. What should I do for convergence of RMS Mass Vapour in this case ? Your kind suggestions would be appreciated. Thanks!!

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Error message: Insufficient Catalogue Size	Paresh Jain	CFX	33	August 16, 2024 06:09
issue in mass flow rate graph	amin.z	Fluent Multiphase	0	June 2, 2015 11:13
CFX overwhelming Fluent in mass convergence of boundary layer separation case	Pierre1	FLUENT	7	March 26, 2015 22:43
Differences between CFX and OpenFOAM regarding convergence and robustness!	magjohan	OpenFOAM Running, Solving & CFD	4	February 26, 2015 11:26
CFX RMS instead of mass momentum Efficiency or Pressure ratio	Saima	CFX	2	June 15, 2011 13:01

June 28, 2016, 06:43		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	First of all: Please do not PM me with exact copies of posts on the forum. It really annoys me. If it is on the forum I will see it. I don't answer CFD questions by PM anyway. CFX does not have the wide range of tunable Fluent has because it does not need them. The coupled solver is very robust for most applications and if it does not converge then usually it means you have a fundamental problem which cannot be fixed by simply tuning the solver constants. As you are doing a phase change simulation you should expect problems in convergence. This FAQ will give you some general tips: http://www.cfd-online.com/Wiki/Ansys...gence_criteria

June 28, 2016, 10:10		#3
nomi.tahir New Member Join Date: Mar 2016 Posts: 11 Rep Power: 10	When I change my working fluid to water and use higher heat flux it does indeed converge . But on the other hand for any other volatile fluid at relatively lower pressure and at comparatively lower heat flux the issue remains. What do you suggest for it ?

June 28, 2016, 20:53		#4
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,870 Rep Power: 144	Small time steps, double precision numerics, improve mesh quality. That is a starting point.