Learning Forum

[MYSELF]

I am working on fluid mixing using fluent 19.2. in the convergence plot my results do not converge. I observed the straight line after some iterations. I am taking the rms value 10^-6 for continuity, velocities and concentration values for convergence. why it is not converging??

[MYSELF]

 

convergence plot that i am getting is like this

[ringelf1]

It does not converge because all the lines have to be under your convergence criteria. And when you set all of them up to 10^(-6) the white and the light blue line are below that criteria. and that's why the calculation is still running.

[DrAmine]

Please check judging convergence chapter. Better to rely on monitoring key quantities to judge convergence .

[Karthik Remella]

Hello,

I agree with ringlef1. Just to quickly add...

Generally, it is a good rule of thumb to set all your convergence criteria to 1e-6. But sometimes, depending on the problem you are trying to solve, it might be difficult to achieve. You might have to relax your convergence criteria in such cases. A good check would be to see what your monitors are doing. If you are looking for a steady solution and your monitors show this behavior (and your residuals change by at least 3-4 orders of magnitude - i.e., 1 to 10^-3 or 10^-4), then it would seem like you have a decently converged solution.

Thanks.

Best,

Karthik

[MYSELF]

thank you all for replying me on this problem. it means it will be good if i change my residuals to 1e-4 or 1e-5, then i can get converged solution. Is there any deviation in results? 

[DrAmine]

Convergence is not only by decreasing residuals. Have a look into monitor plots and mass and flux imbalances. Default residual criteria are fair but you can tighten them a but this does not however answer the question why they are leveling out so early. Afain: the Manual is your friend.