CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Unsteady Case (http://www.cfd-online.com/Forums/fluent/39165-unsteady-case.html)

Q January 8, 2006 10:07

Unsteady Case
 
I want to know when I perform unsteady cacluation using fluent. Let us suppose I use fixed time stepping method.

And I use certain delta_t=x1 and maximum number of iteration 20 and I also use maximum number of iteration 10 and 4. What is the difference I will get in the results.

How can I optimze delta_t and maximum number of iteration?

Thanks

Q

galary January 8, 2006 21:00

Re: Unsteady Case
 
In my opinion, maximum number of iteration is number of equation iteration solutionin in one time step.Usually you should choose number of iteration to satisfy convergence and get a stable solution in the time step.In face unsteady case in one time step is regarded as steady case.


Muhammad Shakaib January 9, 2006 03:36

Re: Unsteady Case
 
Generally if your time step is correct than there is convergence before 30 iterations this means the residuals drop to atleast an order of magnitude and if you have placed a monitoring point for any variable it becomes stable for last few iterations. The no. of iterations '10' appears to be enough for small time step (than needed) but 4 iterations does not seem suitable.

Q January 9, 2006 04:48

Re: Unsteady Case
 
Hello I am running a unsteady case I would like to know what kind of message I would be getting during this iterations whether I will get message that the solution is converged in 10 iterations if suppose I am using the maximum number of ietartions to be 30.

Q

Muhammad Shakaib January 9, 2006 08:50

Re: Unsteady Case
 
While iterating if the residuals drop below the values (of residuals) that you have set message will be ' solution is converged' and next time step will be started.

Q January 9, 2006 12:24

Re: Unsteady Case
 
what is the typical value of residuals you set?

Muhammad Shakaib January 14, 2006 00:08

Re: Unsteady Case *NM*
 

Muhammad Shakaib January 14, 2006 00:11

Re: Unsteady Case
 
Sorry for prevoius empty reply. I just clicked the'post message' due to mistake. I keep my residuals for continiuty 1e-5 for others 1e-3. But these values depends on the case so this may not help you.

Thanks


All times are GMT -4. The time now is 06:12.