[Ahmed Khattab]

Hi All,

i am using a steady state solver "simpleFoam", i know that solver stops when it either achieve tolerance limit or end time set in controlDict.

in my case the solver achieves convergence after 13,000 time step, while it is set to 10^-9, i tried with smaller tolerance 10^-6 but it about to hit 7,000 time step and still running.

is it any rule of thumb for suitable iterations number for a steady state solver, or it differs based on mesh size or any thing else?

thanks,

[Ahmed Khattab]

Hi All,

i have got a clue for my problem, although my mesh has no problems with orthogonality or skewness, there is a high aspect ratio between cell dimensions. that length of cell boundary in x and y direction was much greater that is in z direction.

running same solver on a finer mesh but with cells that has same length in all direction showed much faster convergence rate.

thanks,

[Heremus]

If your residuals don't change anymore. You can always watch into your case in between time steps. If your flow properties don’t change every 1000 steps, or what you deem necessary, you have reached your solution. Residuals alone don't give you unfortunately the information that the simulation is finished.

[eric]

You need to provide more information about the problem you are trying to simulate, and give the residual plots. The "acceptable" residual depends on what you are simulating and what level of accuracy you need in the results. That being said, 1E-09 sounds excessively low.