Understanding near-wall refinement for backward-facing ramp
Hello CFD fellow Warriors,
I would like to ask a few fundamental questions today regarding the choice of appropriate mesh refinement strategies and required non-dimensional wall scale y+, for the SST .
Firstly, the entire domain is representing an external flow over a hump at 4.5 m/s and the length is ~3.5 m. The incline ramp is 5.5 degrees and the decline ramp is 16 degrees. The Reynolds Number is 5.5x10^5.
I am currently having a dilemma with some of the options in the Fluent control panel for the SST model and I have tried to understand the curvature correction and the low-Re correction switched on for all simulations. The main questions are listed below -
1. Based on my limited knowledge the low-Re model tries to account for regions of the flow field where the Reynolds numbers are not turbulent and are experiencing transitions or laminar behaviour. It tends to dampen the turbulence viscosity. Does this negatively impact my boundary layer studies?
2. I am currently using a y+ < 1 for most of the ramp (Image) and I think this should be OK with the Low-Re correction model. Is there an actual lower limit for the y+ that is recommended for the SST model? I thought that near-wall resolution is good if I am not considering using the wall function approach.
Please share your thoughts, experiences and comments with turbulence modelling and boundary layer growth.
Hi Ovi, if you won't get any replies here, try to ask the admin to move the thread to the "Main CFD" forum. They have excellent experts on that RANS stuff there.
|All times are GMT -4. The time now is 17:10.|