Purpose of relaxation factors

Mohsin · April 30, 2010, 10:08

hello

I think that if we would reduce the under relaxation factors then the result will take a lot of time to converge and if we dont change the under relaxation factors then the result will take short time to converge and it doesn't have any affect on the final solution.

So reducing under relaxation factors just only reduces the time if we disregard time thing what other benefit we can get from under relaxation factors. Please comment. Thanks
Mohsin

leonardo.morita · April 30, 2010, 11:02

Hello,
reducing under-relaxation factors can help a lot to achieve convergence in some cases, damping oscillations which may lead to divergence. Indeed, their use is almost always compulsory and you won't get your case converged if you don't set appropriate values.

Mohsin · April 30, 2010, 11:46

Can u tell me how to get the optimum relaxation factors or using the default would be good option rather than reducing or increasing.

I have a cylinder and air is coming from upside and goes into the 3 outlets.

My residual for conitniuty is 10^-5 it touches 2*10^-5 and then goes up and doesn't go until; 10^-5. I dont know what might be the problem.

leonardo.morita · April 30, 2010, 11:59

Well, optimum values depend on the case

but Fluent User's Guide recommends us to start with default values and eventually decrease them if residuals keep increasing.
And 10-5 isn't a bad value for continuity at all...once again it depends on the case, numerical schemes, initialization values etc. Search "continuity residuals convergence" in the forum for dicussions about judging convergence using residuals. You'll see that they're not that important...

Mohsin · April 30, 2010, 12:03

ya you are rite but sometimes increasing relaxation factors decreases residuals.

Chris D · April 30, 2010, 12:57

Residuals are a measure of how much the solution is changing in "time". Under relaxation factors limit how much the solution is allowed to change in "time". If you decrease the under relaxation factors, you decrease the rate of change of the solution, which you see as a decrease in the residuals. It doesn't mean that your solution is converged.

April 30, 2010, 10:08	Purpose of relaxation factors	#1
Mohsin Senior Member Mohsin Mukhtar Join Date: Mar 2010 Location: South Korea Posts: 249 Rep Power: 17	hello I think that if we would reduce the under relaxation factors then the result will take a lot of time to converge and if we dont change the under relaxation factors then the result will take short time to converge and it doesn't have any affect on the final solution. So reducing under relaxation factors just only reduces the time if we disregard time thing what other benefit we can get from under relaxation factors. Please comment. Thanks Mohsin

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Relaxation Factors for Transient solvers	philippose	OpenFOAM Running, Solving & CFD	19	March 20, 2014 05:39
relaxation factors and time accuracy	Mike	Main CFD Forum	7	May 21, 2005 13:41
Relaxation Factors	Tim	Phoenics	3	June 30, 2004 03:03
relaxation factors	zhujianguo	Phoenics	1	August 5, 2003 07:17
relaxation factors adjust	zhujianguo	Phoenics	1	July 15, 2003 12:11

April 30, 2010, 11:02		#2
leonardo.morita Member Leonardo Giampani Morita Join Date: Apr 2009 Location: Paris, France Posts: 58 Rep Power: 17	Hello, reducing under-relaxation factors can help a lot to achieve convergence in some cases, damping oscillations which may lead to divergence. Indeed, their use is almost always compulsory and you won't get your case converged if you don't set appropriate values.

April 30, 2010, 11:46		#3
Mohsin Senior Member Mohsin Mukhtar Join Date: Mar 2010 Location: South Korea Posts: 249 Rep Power: 17	Can u tell me how to get the optimum relaxation factors or using the default would be good option rather than reducing or increasing. I have a cylinder and air is coming from upside and goes into the 3 outlets. My residual for conitniuty is 10^-5 it touches 2*10^-5 and then goes up and doesn't go until; 10^-5. I dont know what might be the problem.

April 30, 2010, 11:59		#4
leonardo.morita Member Leonardo Giampani Morita Join Date: Apr 2009 Location: Paris, France Posts: 58 Rep Power: 17	Well, optimum values depend on the case but Fluent User's Guide recommends us to start with default values and eventually decrease them if residuals keep increasing. And 10-5 isn't a bad value for continuity at all...once again it depends on the case, numerical schemes, initialization values etc. Search "continuity residuals convergence" in the forum for dicussions about judging convergence using residuals. You'll see that they're not that important...

April 30, 2010, 12:03		#5
Mohsin Senior Member Mohsin Mukhtar Join Date: Mar 2010 Location: South Korea Posts: 249 Rep Power: 17	ya you are rite but sometimes increasing relaxation factors decreases residuals.

April 30, 2010, 12:57		#6
Chris D Senior Member Chris Join Date: Jul 2009 Location: Ohio, USA Posts: 169 Rep Power: 17	Residuals are a measure of how much the solution is changing in "time". Under relaxation factors limit how much the solution is allowed to change in "time". If you decrease the under relaxation factors, you decrease the rate of change of the solution, which you see as a decrease in the residuals. It doesn't mean that your solution is converged.