URF vs Time Step Size

Ahmed Alkaisi · February 20, 2018, 00:58

Hi,
What is better to achieve the convergence?
Lowering URF or reducing Time step size?
Thanks

BlnPhoenix · March 6, 2018, 05:40

Quote:

Originally Posted by Ahmed Alkaisi

Hi,
What is better to achieve the convergence?
Lowering URF or reducing Time step size?
Thanks

In general, i would only touch URF if the solution is diverging.

Ahmed Alkaisi · March 10, 2018, 00:36

Quote:

Originally Posted by BlnPhoenix

In general, i would only touch URF if the solution is diverging.

Thanks for replying
But don't you think reducing time step has a bigger effect on solution than URF?
and what is the lowest value of URF you can use?
Regards

BlnPhoenix · March 10, 2018, 06:11

Quote:

Originally Posted by Ahmed Alkaisi

Thanks for replying
But don't you think reducing time step has a bigger effect on solution than URF?
and what is the lowest value of URF you can use?
Regards

To be honest i can not give you a good answer to when to prefer a lower URF instead of a lower time step, since i never faced such a situation. But my guess is, if lowering the time step again and again doesn't not help at all, i would look into lowering the URF for a given time step size. There is then probably a optimum (calculation time wise) of time step and URF when you have a difficult problem that is not easy to stabilize.
Theoretically speaking the lowest URF would be something very small but > 0.

Ahmed Alkaisi · March 19, 2018, 01:00

Thank you for your reply

hamed.majeed · March 19, 2018, 12:00

Isn't URF somehow related to the Courant number as well. If that is the case then URF will effect courant number, likewise, time-step also effects courant number. Hence, stability of the problem would depend on URF and time-step size!!

BlnPhoenix · March 20, 2018, 08:18

Quote:

Originally Posted by hamed.majeed

Isn't URF somehow related to the Courant number as well. If that is the case then URF will effect courant number, likewise, time-step also effects courant number. Hence, stability of the problem would depend on URF and time-step size!!

I'm not aware of any such relation. Can you give a source for this?

Regards

hamed.majeed · March 20, 2018, 09:30

In Fluent theory guide (V19)>21.4.4 Steady-state iterative algorithm>21.4.4.2 Under-relaxation of equation.

The CFL is a solution parameter in the pressure-based coupled algorithm and can be written in terms of $\alpha$ :

(1- $\alpha$ )/ $\alpha$ =1/CFL

Under-relaxation factor= $\alpha$

BlnPhoenix · March 20, 2018, 09:43

Quote:

Originally Posted by hamed.majeed

In Fluent theory guide (V19)>21.4.4 Steady-state iterative algorithm>21.4.4.2 Under-relaxation of equation.

The CFL is a solution parameter in the pressure-based coupled algorithm and can be written in terms of $\alpha$ :

(1- $\alpha$ )/ $\alpha$ =1/CFL

Under-relaxation factor= $\alpha$

I don't understand this. I know the CFL definition only for time dependent simulations, here it is used for steady state applications. Also URF = 1 is commonly used in transient simulations for certain equations. The given formula here however yields CFL = 0 for $\alpha$ = 1.

So maybe this is an alternative usage/definition of CFL, which in turn i don't see a use for in multiphase flows since these are pretty much always transient..

But thanks for making aware of this!

hamed.majeed · March 23, 2018, 10:39

Quote:

Originally Posted by BlnPhoenix

I don't understand this. I know the CFL definition only for time dependent simulations, here it is used for steady state applications. Also URF = 1 is commonly used in transient simulations for certain equations. The given formula here however yields CFL = 0 for $\alpha$ = 1.

So maybe this is an alternative usage/definition of CFL, which in turn i don't see a use for in multiphase flows since these are pretty much always transient..

But thanks for making aware of this!

The definition I told you is for implicit relaxation of equations.

February 20, 2018, 00:58	URF vs Time Step Size	#1
Ahmed Alkaisi Member Ahmed Alkaisi Join Date: Aug 2015 Location: Australia Posts: 80 Rep Power: 11	Hi, What is better to achieve the convergence? Lowering URF or reducing Time step size? Thanks

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
[Other] Contribution a new utility: refine wall layer mesh based on yPlus field	lakeat	OpenFOAM Community Contributions	58	December 23, 2021 03:36
Transient simulation not converging	skabilan	OpenFOAM Running, Solving & CFD	14	December 17, 2019 00:12
simpleFoam error - "Floating point exception"	mbcx4jc2	OpenFOAM Running, Solving & CFD	12	August 4, 2015 03:20
Micro Scale Pore, icoFoam	gooya_kabir	OpenFOAM Running, Solving & CFD	2	November 2, 2013 14:58
Problems in compiling paraview in Suse 10.3 platform	chiven	OpenFOAM Installation	3	December 1, 2009 08:21

March 19, 2018, 01:00		#5
Ahmed Alkaisi Member Ahmed Alkaisi Join Date: Aug 2015 Location: Australia Posts: 80 Rep Power: 11	Thank you for your reply

March 19, 2018, 12:00		#6
hamed.majeed Senior Member Hamed Abdul Majeed Join Date: May 2012 Location: New Orleans, LA, US Posts: 147 Rep Power: 14	Isn't URF somehow related to the Courant number as well. If that is the case then URF will effect courant number, likewise, time-step also effects courant number. Hence, stability of the problem would depend on URF and time-step size!!

March 20, 2018, 09:30		#8
hamed.majeed Senior Member Hamed Abdul Majeed Join Date: May 2012 Location: New Orleans, LA, US Posts: 147 Rep Power: 14	In Fluent theory guide (V19)>21.4.4 Steady-state iterative algorithm>21.4.4.2 Under-relaxation of equation. The CFL is a solution parameter in the pressure-based coupled algorithm and can be written in terms of $\alpha$ : (1- $\alpha$ )/ $\alpha$ =1/CFL Under-relaxation factor= $\alpha$