Hi guys,

I am currently modelling a backwards facing step at relatively low Re (<<1e-6).

I have tried the Realizable k-e model with non-equilibrium wall functions, but I am having trouble with convergence.

My Y+ values are of the order 1, since the step region is relatively small compared to the rest of the domain, and I need sufficient nodes to capture the flow features - so if I use a y+>30, I just don't have enough nodes in the region of interest.

Anyone have suggestions? I am thinking k-w SST model, since it should have a low-Re correction - but would y+ = 1 be good for this model?

Thanks! Riaan

Hi Riaan

Ever thought about a laminar flow regime at your Re ?

br, Max

I doubt the flow will be laminar in the re-attachment region, or in the circulation region aft of the step.

I assume your Re of 1e-6 is based on inlet velocity and inlet cross section. I am quite sure that there is no transition to turbulent flow even in the areas you have mentioned. By the way, I think there is a FLUENT tutorial about your case, which highlights laminar and turbulent flow over a backward facing step.

Cheers