time step and grid convergence

ojha.mayank485 · May 3, 2011, 02:30

Hello,

I am fairly new to this forum. Excuse me if this question is a silly one:

Am trying to attain convergence but I need to know the following questions:
1. Should one go for time-step convergence or grid convergence first.
2. Am trying to attain Time step, residual & grid independence study. But I do not know where to start from. Which one should be done first ?????

Thank you very much.

ghorrocks · May 3, 2011, 19:04

It does not really matter which you do first. Residual convergence is usually the easiest to achieve and usually can then remain constant with changing time step and convergence so that is a good one to start with. Time step convergence is often easier then mesh convergence so it is often good to do that one next.

For completeness it is best to check both the grid and time step (and convergence tolerance any anything else) once you have the final settings. Usually time step, mesh and convergence tolerance convergence is independent of each other providing you are at least close, but to be sure you should check again once you have got your final settings.

May 3, 2011, 02:30	time step and grid convergence	#1
ojha.mayank485 New Member Mayank Ojha Join Date: May 2011 Posts: 22 Rep Power: 14	Hello, I am fairly new to this forum. Excuse me if this question is a silly one: Am trying to attain convergence but I need to know the following questions: 1. Should one go for time-step convergence or grid convergence first. 2. Am trying to attain Time step, residual & grid independence study. But I do not know where to start from. Which one should be done first ????? Thank you very much.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
HELP! time step too small?	meangreen	Main CFD Forum	6	May 31, 2018 10:41
DPM: Particle Tracking	Madhukar	FLUENT	1	July 24, 2007 03:51
A Problem with setting the time step in VOF model	Le	FLUENT	2	July 20, 2006 22:00
Time Step for DES	Maged Ismail	FLUENT	1	June 27, 2005 04:43
Combustion Convergence problems	Art Stretton	Phoenics	5	April 2, 2002 05:59

May 3, 2011, 19:04		#2
ghorrocks Super Moderator Glenn Horrocks Join Date: Mar 2009 Location: Sydney, Australia Posts: 17,700 Rep Power: 143	It does not really matter which you do first. Residual convergence is usually the easiest to achieve and usually can then remain constant with changing time step and convergence so that is a good one to start with. Time step convergence is often easier then mesh convergence so it is often good to do that one next. For completeness it is best to check both the grid and time step (and convergence tolerance any anything else) once you have the final settings. Usually time step, mesh and convergence tolerance convergence is independent of each other providing you are at least close, but to be sure you should check again once you have got your final settings.