CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   K-omega SST transition fully turbulent (

PvS July 16, 2011 08:34

K-omega SST transition fully turbulent

I have read almost every thread about Yplus and K-W SST model. I have set my wall function to Automatic with enabled Transitional Turbulence to Fully Turbulent.

Now could I just ask one simple question

Does my Yplus has to be less than 1 (y+<1) when the automatic wall function is on! and how many Inflations layers must I use near the wall for my mesh?

Im dealing with a Reynolds Number of approximatley 5000
my hydraulic diameter is about 0.003m

If I have to give 25-30 inflation layers to make the simulation more accurate what is the recommended growth rate and the inflation layer option

1) first layer thickness?
2) LAst Aspect ratio?
3)Total Thickness?

I know The SST model is based on the k-omega model and has the same automatic wall treatment
Improves on standard k-omega model, which proves too sensitive to freestreamturbulence.
Accounts for transport of turbulent shear stress and gives highly accurate predictions of the onset and the amount of flow separation
This is a good default choice

Could somebody please clear this for me because I've been hearing that if you re using automatic wall function you do not need to be worrying about Yplus But however when check my Yplus in CFX post or when I monitor my Wall plus on my geomerty I get y+=22 and when I dont put any inflation layers the solution oscillates and does not reach the residual criteria.

So please help!

ghorrocks July 17, 2011 06:29

Your mesh does not have to be y+<1 but you probably have not achieved an accurate simulation if it is not the case. The transition model requires a finer mesh than the fully turbulent model.

The turbulence transition model is tuned for external flows, primarily airfoils. You seem to be using it for internal flows. While it is the best model around for transition, be aware that it might not be so accurate for internal flows.

I grow my inflation layers at 1.05 for rough results and 1.02 for more demanding results when turbulence modelling is critical.

All times are GMT -4. The time now is 00:28.