CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   FLUENT (https://www.cfd-online.com/Forums/fluent/)
-   -   Fluctuating velocities during iterations in Fluent (https://www.cfd-online.com/Forums/fluent/206078-fluctuating-velocities-during-iterations-fluent.html)

ujjwalmohanty August 29, 2018 09:43

Fluctuating velocities during iterations in Fluent
 
1 Attachment(s)
Hi,

I am trying to simulate a flow over an equilateral triangle in 2D. I generated the mesh using ICEM with a minimum quality of 0.98.



To define a Reynolds Number = 40, I used the following data:


Density =1; Viscosity = 1; Height = 2 (Under reference values tab : Area=2, Length = 2); Velocity = 20m/s.
Convergence criteria for momentum, x velocity and y velocity : 1e-11.


Even after 30000+ iterations, the momentum, x- velocity, y-velocity is fluctuating.
Could anyone please tell me whether this means that there is something wrong with the problem setup? If yes, what could it be?

LuckyTran August 29, 2018 12:20

I hope you realize you are looking at residuals and not the actual velocities.

Of course there will be fluctuations, nothing in the code prevents them from fluctuating. But hopefully it is a bounded asymptotic convergence, meaning they fluctuate less than some threshold. It seems your threshold of 1e-11 was too ambitious and you obtained instead 1e-06 and 1e-07 for x and y momentum equations.


Is it a problem? It depends how picky you are. The result is the result. If the result is not good enough then you need to do something. In you case I see you have outputted lift and drag coefficients and you notice these also fluctuate at some level. If this fluctuation is too much for you then you can worry about it.

ujjwalmohanty August 29, 2018 13:50

Quote:

Originally Posted by LuckyTran (Post 704513)
I hope you realize you are looking at residuals and not the actual velocities.

Of course there will be fluctuations, nothing in the code prevents them from fluctuating. But hopefully it is a bounded asymptotic convergence, meaning they fluctuate less than some threshold. It seems your threshold of 1e-11 was too ambitious and you obtained instead 1e-06 and 1e-07 for x and y momentum equations.


Is it a problem? It depends how picky you are. The result is the result. If the result is not good enough then you need to do something. In you case I see you have outputted lift and drag coefficients and you notice these also fluctuate at some level. If this fluctuation is too much for you then you can worry about it.




Thanks LuckyTran for your inputs. Yes I know those are the residuals. I didn't put it the right way. Sorry for that.

For the residuals I am getting a straight line at 1e-07.
And yes, the Cd does fluctuate but only the fourth decimal place.



Thank you.
:)


All times are GMT -4. The time now is 07:24.