[Tobi]

Hi all,

i am simulating a twisted flow field in a diffuser geometry.
In ANSYS I get the vertex-breakdown effect but with OF not.

I used the inletOutlet BC for U and T and set a fixedValue pressure at the outlet.

Could it possible that I have to extend the outlet with a normal pipe ? In ANSYS we do so but I thought with the inletOutlet BC its possible to do it without that pipe.

Tobi

[hannes]

Hello Tobi,

I guess you are doing URANS? Or LES?

I think you need to be more specific about the solver settings (fvSchemes and fvSolution).
I remember that the onset of vortex breakdown is predicted too late if the solver is too dissipative (due to schemes or turbulence modeling).

Regards, Hannes

[Tobi]

hi,

i am modelling with rans and k-epsilon, k-omega, and sst on a modified rhoSimpleFoam solver ...

The schemes are gauss upwind and gauss linear couse other schemes are not converge.

Its more a BC question what BC I set on the diffuser outlet couse there is the vortex breakdown too - in few regions. Or should I made an new outlet being away from the original outlet with the length of three x diffuser diameter ?

Tobi