[Fauster]

Dear "Star-ccm+er" !

I would like to calcul drag and lift coefficient for a body. It is appropriate to define a "Flow split outlet" BC on the outlet face (with a ratio of 1) instead of "pressure outlet" BC ? For exemple for wind loading on buildings.

Thanks for your help,

[me3840]

No, flow split outlets are for internal systems. You need to use a pressure condition and take care of atmospheric pressure gradients due to gravity.

[Fauster]

Thanks,

At the begining I wanted to set a field function for the static pressure (at the outlet face) :

Ps = -rho.g.z

But the calculation behaved very badly. Maybe I need to use a free stream condition... I am kind of confused because I read many differents things on the internet about appropriate BC for external flows.

Intuitively I would have set a zero gradient BC on lateral and top faces. And a pressure outlet condition on the outlet face. But many authors provides to use a free stream BC on the top face or even a wall BC on lateral faces.

What are your thought ?

Thanks a lot

[marmot]

if you make the boundaries far enough away from your buildings you can use a symmetry, slip wall on top, sides and pressure for the outlet. Like you are simulating a building inside of a wind tunnel

Perhaps though you want to change wind directions, thus instead of a box you have a half sphere, dome for your domain and thus free stream bc would be better.

[Fauster]

Thanks,

I didn't thought about using half sphere for fluid domain. I should investigate that.