I am using CFX-5.7.1 to estimate the pressure drop in a valve system. The computation shows some recirculation with swirling at the outlet. I switched the outlet BC to an opening (P=1atm). How can this affect the precision of my computation? I am actually trying to extrude the geometry of my outlet enough far to have a more established flow. If this is valid should I then consider the pressure of the true outlet or the nezw extruded one for calculating Delta_P ??

Tkx a lot for any advice

Hello,

from my experience, it may be difficult to get a converged solution if swirl exists at the outlet. The problem with the opening is that the conditions of the flow reentering the domain must be known (in fact thea are set within the dialog for setting up the opening bc). Your suggestion to expand the downstream geometry makes sense. You could use a quiet coarse grid and probably set a free slip boundary condition in order to not influence the pressure drop.

hth,

-Thomas

I will try that. Thank you Thomas