my residual does not reduced

yaseen wsu · April 27, 2016, 10:02

hello
my free surface model in steady state, it is residual is greater than 10^-4 for most equations, even I used monitor point for mass flow rate at outlet there are fluctuation. note I used smaller and larger time scale in both cases same thing. can anyone suggest me how to residuals become lower than 10^-4 which is critical
thanks
http://www.cfd-online.com/Forums/att...1&d=1461765708

liucheng8602 · April 27, 2016, 18:32

There are many reasons for convergence problem.
Please refer to CFX help files about convergence and here is a link http://www.cfd-online.com/Wiki/Ansys...gence_criteria I hope it could help.

I think first you should check if your convergence problem global or local. If it's a local problem caused by the mesh, improve mesh quality. Then please be careful about the turbulence model you selected.

yaseen wsu · April 27, 2016, 22:44

Quote:

Originally Posted by liucheng8602

There are many reasons for convergence problem.
Please refer to CFX help files about convergence and here is a link http://www.cfd-online.com/Wiki/Ansys...gence_criteria I hope it could help.

I think first you should check if your convergence problem global or local. If it's a local problem caused by the mesh, improve mesh quality. Then please be careful about the turbulence model you selected.

thanks for your reply
I read this thread before, I think my mesh quality is ok in terms of orthognal and skewness only i have problem in expansion rate due to inflation. i used RNGk-e model, because flow swril and highly curvature

-Maxim- · April 28, 2016, 02:39

In case you have several domains, display the monitor with residuals for each domain. Also have a look at the MAX residuals and not just the RMS.

You might see that for example "V-Mom in Domain XYZ" has the highest residuals. The go to CFX POST, create an isovolume with V-Mom in Domain XYZ and select "Above Value" and chose 0.001 in case your criteria is 10^-3 for MAX residuals. Choose a bright color. Then evaluate the isovolume (location, size, etc). Maybe take some screenshots for us

Residuals in a steady-state simulations can be high in case there are unsteady flow phenomena like turbulent vortices - the only solution here is to switch to a transient simulation.

yaseen wsu · April 28, 2016, 06:27

thanks a lot Maxim
no, I have only one domain only I used local parallel divided into 8 partition, thanks for your suggestion, Yes I want to simulate in transient to show how difference between steady or transient. but unfortunately I have residual problem.

-Maxim- · April 28, 2016, 07:06

you should still display the MAX residuals and create the isovolume.

yaseen wsu · April 28, 2016, 15:31

http://www.cfd-online.com/Forums/att...1&d=1461871657
http://www.cfd-online.com/Forums/att...1&d=1461871671
http://www.cfd-online.com/Forums/att...1&d=1461871682

-Maxim- · April 29, 2016, 02:21

Quote:

Originally Posted by yaseen wsu

http://www.cfd-online.com/Forums/att...1&d=1461871657
http://www.cfd-online.com/Forums/att...1&d=1461871671
http://www.cfd-online.com/Forums/att...1&d=1461871682

Invalid Attachment specified. If you followed a valid link, please notify the administrator

yaseen wsu · April 30, 2016, 06:18

the following attachment may be worked, which is isovolume of V-mom for value 10^-3 residual
https://drive.google.com/open?id=0B3...jRKaVF0VU1NZ00
https://drive.google.com/open?id=0B3...DJoeHBhSWlBeXM

ghorrocks · April 30, 2016, 09:46

When you attach a file to the forum you then have to do something with it - either make it an image or a downloadable link. It appears you are not doing anything with the attachments you upload and that is why all your attachments are invalid.

yaseen wsu · May 1, 2016, 01:45

isovolume of v-mom for 10^-3 max. residual
https://drive.google.com/open?id=0B3...DJoeHBhSWlBeXM
https://drive.google.com/open?id=0B3...EdmZGNlYnJVS3c
https://drive.google.com/open?id=0B3...jRKaVF0VU1NZ00

ghorrocks · May 1, 2016, 08:48

Your images are not of the residual. It is just of the velocity.

yaseen wsu · May 1, 2016, 09:47

yes of v-mom velocity which has max. residual, the follow is my residuals

http://www.cfd-online.com/Forums/att...-rersidual.jpg

ghorrocks · May 1, 2016, 09:58

I think you have misunderstood Maxim's suggestion. The suggestion was to plot a isosurface of the u/v/w velocity residual (not the u/v/w velocity). This will show you where in the simulation are the highest residuals and this is likely to be the location where your convergence problem occurs.

The equation residuals are not in the results file by default. You have to go to the output tab in CFX-Pre and select the option which adds them to the results file. Then you will have the residuals in the results file as additional variables for you to display.

yaseen wsu · May 1, 2016, 11:04

ohhh, according to your suggestion before running the model, from CFX PRE in output tab add residuals and then after running in CFX POST plot them, otherwise there is no residuals in cfd post. that is right???

ghorrocks · May 1, 2016, 20:24

Yes, that is right. You will need to rerun the simulation. Or you could restart the old one, adding the residuals option, and just run it for 1 iteration.

April 28, 2016, 02:39		#4
-Maxim- Senior Member Maxim Join Date: Aug 2015 Location: Germany Posts: 415 Rep Power: 12	In case you have several domains, display the monitor with residuals for each domain. Also have a look at the MAX residuals and not just the RMS. You might see that for example "V-Mom in Domain XYZ" has the highest residuals. The go to CFX POST, create an isovolume with V-Mom in Domain XYZ and select "Above Value" and chose 0.001 in case your criteria is 10^-3 for MAX residuals. Choose a bright color. Then evaluate the isovolume (location, size, etc). Maybe take some screenshots for us Residuals in a steady-state simulations can be high in case there are unsteady flow phenomena like turbulent vortices - the only solution here is to switch to a transient simulation. gmaschio likes this.

April 30, 2016, 06:18		#9
yaseen wsu Senior Member yaseen Join Date: Oct 2015 Location: Hawler Posts: 174 Rep Power: 10	the following attachment may be worked, which is isovolume of V-mom for value 10^-3 residual https://drive.google.com/open?id=0B3...jRKaVF0VU1NZ00 https://drive.google.com/open?id=0B3...DJoeHBhSWlBeXM Last edited by yaseen wsu; May 1, 2016 at 01:43.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
simpleFoam error - "Floating point exception"	mbcx4jc2	OpenFOAM Running, Solving & CFD	12	August 4, 2015 02:20
Unstabil Simulation with chtMultiRegionFoam	mbay101	OpenFOAM Running, Solving & CFD	13	December 28, 2013 13:12
pimpleFoam: turbulence->correct(); is not executed when using residualControl	hfs	OpenFOAM Running, Solving & CFD	3	October 29, 2013 08:35
calculation stops after few time steps	sivakumar	OpenFOAM Running, Solving & CFD	7	March 17, 2013 06:37
Orifice Plate with a fully developed flow - Problems with convergence	jonmec	OpenFOAM Running, Solving & CFD	3	July 28, 2011 05:24

April 27, 2016, 18:32		#2
liucheng8602 New Member Chen Liu Join Date: Jun 2014 Location: Beijing, China Posts: 3 Rep Power: 11	There are many reasons for convergence problem. Please refer to CFX help files about convergence and here is a link http://www.cfd-online.com/Wiki/Ansys...gence_criteria I hope it could help. I think first you should check if your convergence problem global or local. If it's a local problem caused by the mesh, improve mesh quality. Then please be careful about the turbulence model you selected.

April 28, 2016, 06:27		#5
yaseen wsu Senior Member yaseen Join Date: Oct 2015 Location: Hawler Posts: 174 Rep Power: 10	thanks a lot Maxim no, I have only one domain only I used local parallel divided into 8 partition, thanks for your suggestion, Yes I want to simulate in transient to show how difference between steady or transient. but unfortunately I have residual problem.

April 28, 2016, 07:06		#6
-Maxim- Senior Member Maxim Join Date: Aug 2015 Location: Germany Posts: 415 Rep Power: 12	you should still display the MAX residuals and create the isovolume.

April 28, 2016, 15:31		#7
yaseen wsu Senior Member yaseen Join Date: Oct 2015 Location: Hawler Posts: 174 Rep Power: 10	http://www.cfd-online.com/Forums/att...1&d=1461871657 http://www.cfd-online.com/Forums/att...1&d=1461871671 http://www.cfd-online.com/Forums/att...1&d=1461871682

April 30, 2016, 09:46		#10
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	When you attach a file to the forum you then have to do something with it - either make it an image or a downloadable link. It appears you are not doing anything with the attachments you upload and that is why all your attachments are invalid.

May 1, 2016, 01:45		#11
yaseen wsu Senior Member yaseen Join Date: Oct 2015 Location: Hawler Posts: 174 Rep Power: 10	isovolume of v-mom for 10^-3 max. residual https://drive.google.com/open?id=0B3...DJoeHBhSWlBeXM https://drive.google.com/open?id=0B3...EdmZGNlYnJVS3c https://drive.google.com/open?id=0B3...jRKaVF0VU1NZ00

May 1, 2016, 08:48		#12
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Your images are not of the residual. It is just of the velocity.

May 1, 2016, 09:47		#13
yaseen wsu Senior Member yaseen Join Date: Oct 2015 Location: Hawler Posts: 174 Rep Power: 10	yes of v-mom velocity which has max. residual, the follow is my residuals http://www.cfd-online.com/Forums/att...-rersidual.jpg

May 1, 2016, 09:58		#14
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	I think you have misunderstood Maxim's suggestion. The suggestion was to plot a isosurface of the u/v/w velocity residual (not the u/v/w velocity). This will show you where in the simulation are the highest residuals and this is likely to be the location where your convergence problem occurs. The equation residuals are not in the results file by default. You have to go to the output tab in CFX-Pre and select the option which adds them to the results file. Then you will have the residuals in the results file as additional variables for you to display.

May 1, 2016, 11:04		#15
yaseen wsu Senior Member yaseen Join Date: Oct 2015 Location: Hawler Posts: 174 Rep Power: 10	ohhh, according to your suggestion before running the model, from CFX PRE in output tab add residuals and then after running in CFX POST plot them, otherwise there is no residuals in cfd post. that is right???

May 1, 2016, 20:24		#16
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,703 Rep Power: 143	Yes, that is right. You will need to rerun the simulation. Or you could restart the old one, adding the residuals option, and just run it for 1 iteration.