CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Transition SST - Intermittency/Tu Decay/Y+ (

Kloz July 28, 2011 08:29

Transition SST - Intermittency/Tu Decay/Y+
Hi, I'm using the Transition SST model in Fluent V13 to model a transitional flow on a flat plate in a wind tunnel. I'm unsure about how to set the intermittency on the velocity inlet. Zero? 1? Any ideas there?
Regarding turbulence intensity, the manual mentions estimating the Tu decay. I have very little decay between my inlet and plate, but it exists nonetheless. How do I go about estimating that decay? Or is just trial and error until I reach the desired turbulence intensity values at the plate?
The manual also states that for the transition models y+<1 is desired. Seems pretty steep a requirement, I'm wondering if the 1-4 range might be sufficient. More importantly, do I also need to have the y+ on the tunnel walls less than 5?
Finally, sorry for the basic question, but what's the best way to identify the laminar and turbulent regions on the plate and the point of transition onset?


Kloz July 31, 2011 12:16

Any advice regarding how to set the intermittency on the velocity inlet? It's the biggest question mark at the moment.

sail July 31, 2011 22:20


i reccomand some investigation on the papers of menter and the phd thesis of Langtry. the gamma-retheta model is not as easy as it looks and requires some tuning to get meaningful answers.

y+<1 is a must, anything bigger will move the transition point upstream. it is well documented in the fluent theory manual.

Kloz August 29, 2011 11:04

So I've taken a look at Menter and Langtry's paper, 'A Correlation-Based Transition Model Using Local Variables'. I'm trying to replicate their results for one of the test cases they used to validate the model, the Schubauer flat plate experiment from the NACA TN 3489. I'm using the same inlet conditions they specify in the paper as well as the second order upwind scheme but for some reason transition just isn't being triggered. My Cf plot is exactly that of the laminar computation and consistent with the Blasius solution. Not really sure what I'm missing. Do I need to play around with wall roughness or wall material specifications for example? I've also tried upping the freestream turbulence intensity (to 0.5%, 1% & 3% compared to Menter's 0.18%) and playing with the viscosity ratio (10 & 120 compared to Menter's 5) but no luck, still laminar flow. Y+ is <1 all over the plate.

All times are GMT -4. The time now is 18:39.