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

how would under relaxation factor effect convergence or results?

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

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   May 5, 2020, 07:25
Default how would under relaxation factor effect convergence or results?
  #1
Senior Member
 
Weiqiang Liu
Join Date: Feb 2018
Posts: 277
Rep Power: 6
Weiqiang Liu is on a distinguished road
hi all,

I am using fluent to calculate propane combustion in a micro-channel. I set the under relaxation factor of energy to 0.8 and iteration can be started normally. after a huge number of iterations, the results seems to be converged. however, when I increase the under relaxation factor of energy, variables in monitor position started to vary again. when I further increase the energy under relaxation factor, divergence happened immediately.

I am wondering how the under relaxation factor would effect convergence or results.

Best regards

weiqiang
Weiqiang Liu is offline   Reply With Quote

Old   May 5, 2020, 07:45
Default URFs
  #2
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,948
Blog Entries: 1
Rep Power: 32
vinerm will become famous soon enough
Are you running a transient simulation or steady-state? In any case, 0.8 is a rather small URF for energy since energy time-scales are large. In other words, energy URF needs to be high, usually higher than 0.9 because energy takes longer to diffuse. With low URF values, all the errors are not reduced, particularly low-frequency errors. As you increase the URF, these errors change and the solution updates to a new value. Work with URF between 0.92 and 0.96. Else, you will have to play with Multigrid controls but do not touch those if you do not know what those mean.
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   May 5, 2020, 07:54
Default
  #3
Senior Member
 
Weiqiang Liu
Join Date: Feb 2018
Posts: 277
Rep Power: 6
Weiqiang Liu is on a distinguished road
Quote:
Originally Posted by vinerm View Post
Are you running a transient simulation or steady-state? In any case, 0.8 is a rather small URF for energy since energy time-scales are large. In other words, energy URF needs to be high, usually higher than 0.9 because energy takes longer to diffuse. With low URF values, all the errors are not reduced, particularly low-frequency errors. As you increase the URF, these errors change and the solution updates to a new value. Work with URF between 0.92 and 0.96. Else, you will have to play with Multigrid controls but do not touch those if you do not know what those mean.
hi vinerm,

thanks very much for your reply! I am running a steady state simulation. actually, I am trying to reproduce results in a published paper. I spend almost 2 million iterations to get seemingly converged results. the general trend of my results are very similar to results in the literature. However, the temperature magnitude of mine is 300k lower.

I don't know why. I tried to increase URF of energy. Then temperature would increase very slowly. when I increase URF to 0.99, flame, namely high temperature region disappears immediately. when URF is set to 1, divergence happened.

Best regards

Weiqiang
Weiqiang Liu is offline   Reply With Quote

Old   May 5, 2020, 08:43
Default Temperature Discrepancy
  #4
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,948
Blog Entries: 1
Rep Power: 32
vinerm will become famous soon enough
300 K is a rather large discrepancy and cannot be due to URF. Either this is related to units, but I hope you have already checked those or related to physical setup. Fluent always works in SI and reports data in K; papers usually report data in C.
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   May 5, 2020, 09:14
Default
  #5
Senior Member
 
Weiqiang Liu
Join Date: Feb 2018
Posts: 277
Rep Power: 6
Weiqiang Liu is on a distinguished road
Quote:
Originally Posted by vinerm View Post
300 K is a rather large discrepancy and cannot be due to URF. Either this is related to units, but I hope you have already checked those or related to physical setup. Fluent always works in SI and reports data in K; papers usually report data in C.

IMG_0529.jpgIMG_0525.jpg
Hi vinerm,

You can check my results and the screen shot of literature results. It shows the centerline and wall temperature profiles. The general trends are very similar. However, Maximum temperature in my case is 1760k and maximum temperature in literature is 2050k.

I tried to check the input fuel mass or heat dissipation condition. However, these conditions are very simple which can not be different with those in literature.

I am wondering is something wrong for my UDF to define gas phase and surface reaction rate.

Best regards

Weiqiang
Weiqiang Liu is offline   Reply With Quote

Old   May 5, 2020, 09:18
Default Reactions
  #6
Senior Member
 
vinerm's Avatar
 
Vinerm
Join Date: Jun 2009
Location: Nederland
Posts: 2,948
Blog Entries: 1
Rep Power: 32
vinerm will become famous soon enough
UDF may or may not have a bug. Look for the material properties because those define the heat of reaction.
__________________
Regards,
Vinerm

PM to be used if and only if you do not want something to be shared publicly. PM is considered to be of the least priority.
vinerm is offline   Reply With Quote

Old   May 5, 2020, 09:27
Default
  #7
Senior Member
 
Weiqiang Liu
Join Date: Feb 2018
Posts: 277
Rep Power: 6
Weiqiang Liu is on a distinguished road
Quote:
Originally Posted by vinerm View Post
UDF may or may not have a bug. Look for the material properties because those define the heat of reaction.


I used CHEMKIN format thermal and transport data just like the author did. Thanks for your suggestions. I will check further.

Best
Weiqiang Liu is offline   Reply With Quote

Reply

Tags
convergence, results, under relaxation

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
problem with Min/max rho tH3f0rC3 OpenFOAM 8 July 31, 2019 09:48
Relaxation - Automatic Convergence Control poonkm Phoenics 2 August 15, 2014 00:36
Convergence with reduced under relaxation and time scale factors saisanthoshm88 CFX 3 March 20, 2013 06:04
Convergence with reduced under relaxation and time scale factors saisanthoshm88 FLUENT 0 March 19, 2013 10:08
Force can not converge colopolo CFX 13 October 4, 2011 22:03


All times are GMT -4. The time now is 02:59.