CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Solution limits (http://www.cfd-online.com/Forums/fluent/31419-solution-limits.html)

 ap May 6, 2003 16:04

Solution limits

FLUENT limits solution values according to values specified in the Solve->Controls->Limits dialog box.

Is there a way to disable limitations?

Hi and thanks

 Alex Munoz May 6, 2003 21:31

Re: Solution limits

Hi ap

Risky business what you want to do!! I really suguest that you do not inactivate limit solutions.

Some of them are used to redefine variables in case that your case solution reach the limits, and thereby provide convergence.

I think that you should modify those limits based in a complete understanding of their impact on the solution. They could found as follow

/solve/set> limits Minimum allowable absolute pressure (pascal) [1] Maximum allowable absolute pressure (pascal) [5000000] Minimum allowable k (m2/s2) [1e-14] Minimum allowable epsilon (m2/s3) [1e-20] Maximum allowable turbulent/laminar viscosity ratio [100000]

Best regards

Alex Munoz

 heydari May 7, 2003 03:54

Re: Solution limits

Hi, I think you want to continue, your running to get to a fixed value, so you can use your limitaion at very low values,10e-10.

 ap May 7, 2003 08:14

Re: Solution limits

My limitation is on the turbulent viscosity ratio and it happens during the first second (I use 0.001s as time step) of the solution.

I refined the grid and changed boundary conditions, but nothing changed. I'm sure my data are correct, so I don't know how to avoid limitation.

If I try to set the turbulent viscosity ratio to 10^20 (the max value accepted by FLUENT solver), I also have the limitation during the first 1000-1500 time steps (time step = 0.001s).

I always obtain a converged solution, but how should I consider it if the limitation is present? I'd like to see what would be the real solution predicted by the model and not the "limited" solution.

I know why limitation takes place. I just want to see what happens if I remove it. But to do this I need to completely disable the limiting function of FLUENT. Is this possible?

Thanks

ap

 Alex Munoz May 7, 2003 12:00

Re: Solution limits

Hi ap

Keep in mind that the turbulent viscosity ratio fix the dissipation rate of turbulent kinetic energy to a minumum value, according to the proportion mut~1/epsilon, where mut is the turbulent viscosity and epsilon the dissipation rate of turbulent kinetic energy. As a result, when you increase the turbulent viscosity ratio, you indirectly decrese the dissipation rate of the turbulent kinetic energy. Therefore, I think that your approach do not lead you to a truly solution of the flow field.

It seems that your domain is very complex and thus the spatial ot the temporal discretization of the domain is not enough to provide resonable results.

I really do not how to solve this problem, since you already has refined the grid. Perhaps you should try LES, this turbulent model does not have solution limits.

Regards

Alex Munoz

 ap May 7, 2003 12:54

Re: Solution limits

Thanks :)

 All times are GMT -4. The time now is 17:52.