[7R0Y]

Hi there,

I am running several geometries which resemble closed channel flow (L>>W>>H), implementing periodic boundary conditions (to solve for fully developed flow along the length of the channel), with some form of obstruction on the bottom surface (ex. pyramidal "bumps", height of the bumps are ~1/2H).

I am performing a parametric study to vary flow rates and I am expecting laminar flow for the lower flow rates (Re<200) and am unsure when transition will occur with each geometry.

I am able to achieve convergence with a laminar model for low Reynolds (based on channel hydraulic diameter) cases but when flow rate is increased (Re>~350)the solution diverges or monitors show oscillating behaviour. Monitors are: (1) integrated pressure along centerline plane and (2) vertex averaged velocity magnitude at a point where vortices/separation would likely occur.

I have a few questions, if anyone can shed some light it would be greatly appreciated.

1. For low Re (Re<~200) cases that converge with a laminar model is it safe to assume the results are accurate?

2. For Re > 350 would I require a fully turbulent model or may a transition model be needed? Any advice on turbulence models would be a bonus, I have yet to delve into the differences.

3. Is it accurate to say that the cases that do not converge with the laminar flow are in transition or are turbulent?

Cheers,
Troy