I have a pipe with a sudden expansion from 0.25 mm to 1 mm. Fully developed flow is at the inlet and a reynolds number of 500.
I use k-omega SST turbulence model because laminar simulation hasn't converged for this scenario (for lower Reynolds number, laminar and turbulent simulation both show the same result!).
The only problem is, that the k-omega SST predicts false turbulence in the pipe before the expansion. I have a small deviation from fully developed laminar flow there, which should not happen.
I think this might be a common problem with RANS turbulence models, because one part of my geometry is fully laminar.
Is there any paper what deals with this effect?
How could I prevent this?