CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Pressure drop using Fan type BC (https://www.cfd-online.com/Forums/openfoam-solving/122199-pressure-drop-using-fan-type-bc.html)

Alexis Sack August 14, 2013 07:44

Pressure drop using Fan type BC
 
Hi everyone !

I am new to OpenFoam and I need some help… But first of all I want to thank you all for your job on CFD Online because that already helped me a lot!

I am simulating flow over a car modelled by an Ahmed body using OF 2.2.0. I want to figure out the effect of the cooling system on the drag. So I divided my calculation into 2 parts:
1. Calculation of the pressure drop due to the cooling system: I got the pressure drop over inlet velocity curve.
2. Calculation of the flow over the Ahmed body including my negative pressure drop boundary condition: I am using a Fan BC with a csv file to define the pressure drop curve.
My meshing software is HyperMesh (Altair software).

Here is my problem: when I simulate a 3 million cells mesh, I get a correct pressure drop, but if I simulate an over 4 million cells mesh, I get a pressure rise instead of a drop. The weird fact is that my files are exactly the same, only the mesh is changing.

Does anybody have an idea of what happens in my calculations?

Thanks in advance!
Alexis

Alexis Sack September 1, 2013 04:45

Pressure drop problem: solution
 
Ok I found the solution this past week!
I am posting it because I guess it may help someone one day, that is a little bit tricky.
So my problem came from parallel processing: if you are using cyclic type boundary conditions and parallel processing, you have to make sure 2 matching faces are in the same processor, otherwise you won't get proper results!
For that purpose, just change your decomposition direction and that should work good :)

coroi September 22, 2014 09:18

Quote:

Originally Posted by Alexis Sack (Post 449147)
Ok I found the solution this past week!
I am posting it because I guess it may help someone one day, that is a little bit tricky.
So my problem came from parallel processing: if you are using cyclic type boundary conditions and parallel processing, you have to make sure 2 matching faces are in the same processor, otherwise you won't get proper results!
For that purpose, just change your decomposition direction and that should work good :)

Thank you very much for this post!! It was very helpful to me!:):)


All times are GMT -4. The time now is 21:45.