CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Pre-Processing (https://www.cfd-online.com/Forums/openfoam-pre-processing/)
-   -   Fluent Outflow boundary condition implemented in OF? (https://www.cfd-online.com/Forums/openfoam-pre-processing/73921-fluent-outflow-boundary-condition-implemented.html)

Fluent Outflow boundary condition implemented in OF?

Hello FOAMers,
is there someone that knows if a BC similar to Fluent's outflow BC has already been implemented in OpenFOAM?
Fluent's guide says that:
Quote:
 Outflow boundary conditions in FLUENT are used to model flow exits where the details of the flow velocity and pressure are not known prior to solution of the flow problem. You do not define any conditions at outflow boundaries (unless you are modeling radiative heat transfer, a discrete phase of particles, or split mass flow): FLUENT extrapolates the required information from the interior.
Can be ok setting U, k, epsilon as inletOutlet, inletValue \$innerField and p as zeroGradient both at the inlet and at the outlet? Since U is directed to the external of the domain at the outlet, the inletOutlet BC will use U as zeroGradient, in such a way that both U and p are not fixed in that boundary.

 dinonettis March 19, 2010 13:29

if we consider the case of a flow over a wing within a square domain (boundaries: inlet, outlet, top, bottom) what I usually do, using the solver simpleFOAM, is to adopt the following settings:

inlet: U (k,epsilon) inletOutlet, p zeroGradient
outlet: U (k,epsilon) zeroGradient, p outletInlet
top, bottom: U (k,epsilon) freestream, p freestreamPressure.

if you want to avoid setting even the pressure at the outlet you can try simply changing the BC at the outlet from outletInlet to zeroGradient or freestreamPressure (actually the value is not needed in both). However I'm not sure the problem is well-posed in this way. Is the flow subsonic or supersonic??
let me know,

dino

Hi Dino,
The flow is subsonic, actually it is an incompressible flow. If I am right, freestream bc can be used on compressible flow only, thus that is not a possibility for me. That's why I chose inletOutlet everywhere.
As for my case, I cannot set the pressure at the outlet, since I do not know what its value is. I am trying to reproduce results from a Fluent simulation, and they used an outflow bc at the outlet, thus that is why I asked such a question.
Thanks for any suggestions you will give.
Cheers,

 gwierink March 23, 2010 04:09

Quote:
 I cannot set the pressure at the outlet, since I do not know what its value is.
Is it an idea to use zeroGradient also for p?

Quote:
 Is it an idea to use zeroGradient also for p?
Well... this is what I did in the beginning:
Quote:
 setting U, k, epsilon as inletOutlet, inletValue \$innerField and p as zeroGradient both at the inlet and at the outlet. Since U is directed to the external of the domain at the outlet, the inletOutlet BC will use U as zeroGradient, in such a way that both U and p are not fixed in that boundary
In that case, if the U inletValue is low (between 0.5 and 3 m/s), the solution is stable, but when raising U to 11 m/s, the solution explodes. However, I am worried the divergence is only a matter of iterations, even with low U value. As Dino noticed, I guess the problem is not well posed in this way...

 alib022 March 30, 2017 10:34

Quote:
 Originally Posted by maddalena (Post 251225) Well... this is what I did in the beginning: In that case, if the U inletValue is low (between 0.5 and 3 m/s), the solution is stable, but when raising U to 11 m/s, the solution explodes. However, I am worried the divergence is only a matter of iterations, even with low U value. As Dino noticed, I guess the problem is not well posed in this way...