Question about convergence

Anna Tian · October 29, 2012, 11:15

Hi

May I ask why very fine mesh always need more iterations to converge?

flotus1 · October 29, 2012, 11:38

For time-explicit solvers, this can be explained quite eaisly:

The timestep is limited by the CFL-criterion, thus the CFL-number may not be greater than 1 (or 2 for some solvers).
A small cell size means small timestep to satisfy the CFL-criterion.

In a smaller time, the solution "travels" a shorter way in the computational domain. Thus the change of the solution is smaller with each timestep.
To achieve the same change from the initial condition compared to a coarse-mesh-computation, more time steps are needed.

October 29, 2012, 11:15	Question about convergence	#1
Anna Tian Senior Member Meimei Wang Join Date: Jul 2012 Posts: 494 Rep Power: 16	Hi May I ask why very fine mesh always need more iterations to converge? __________________ Best regards, Meimei

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Time step dependence of convergence behavior of steady state simulations in CFX	Chander	Main CFD Forum	5	December 23, 2013 06:31
Convergence Question	racerdude777	Main CFD Forum	2	March 14, 2011 23:07
Convergence Question	James	FLUENT	2	October 29, 2005 15:59
Question about convergence criteria?	Thomas	FLUENT	7	April 8, 2005 20:38
help! cyclone convergence question	Lcw	FLUENT	1	January 26, 2005 16:29

October 29, 2012, 11:38		#2
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,427 Rep Power: 49	For time-explicit solvers, this can be explained quite eaisly: The timestep is limited by the CFL-criterion, thus the CFL-number may not be greater than 1 (or 2 for some solvers). A small cell size means small timestep to satisfy the CFL-criterion. In a smaller time, the solution "travels" a shorter way in the computational domain. Thus the change of the solution is smaller with each timestep. To achieve the same change from the initial condition compared to a coarse-mesh-computation, more time steps are needed.