I'm modelling a 2D swirling jetflow, where a swirling pipe flow (simulated as an inlet velocity distribution) enters a larger duct/pipe (Roback & Johnson, 1983). My simulations produce absolutely terrible results, the circumferential velocity is zero almost from the start and the recirculation area is not being modelled correctly. Does anybody have any experience with modelling 2-D turbulent/swirling pipe flow with STAR-CD and encountered problems with that? I'm running out of ideas and don't know what else to try, other than switching to FLUENT...any idea's are welcome!

I use standard k-e, cyclic boundary conditions,and a (r,theta,z) grid of (200,1,200) for a geometry (61mm,1deg,1016mm)

Nijso

I don't think anything is wrong with your computations. It is well known that the standard k-e model is not able to produce "good result" when the flow is swirling. You can eventually improve the result, if you use a nonlinear version of the k-e model.

Find someone who will look at you model. Have you asked the support.

Have you checked y+? If you use 200 Cells for 61mm then y+ should be very low. Move the inlet bc upstream into the pipe (away from the step).