Laminar-turbulent transitions models?


#1

I was wondering if SimScale offers any of the models for laminar to turbulent flow transitions, eg gamma-ReTheta model, or k-omega-v2 model. There seems to be some recent developments in this area in OpenFOAM (see discussion here), but I couldn’t see any of these as options in SimScale’s Analysis Types.


#2

@gsokoll,

Thanks for your question. You can read more about the flow models supported by SimScale in this post

With regards to the laminar to turbulent flow transition model, these are not yet offered on SimScale as the demand for them hasn’t been there.

Do you have a specific application you are working on right now that requires this, would be interested to hear more?


#3

Hi Anna. The application is drag optimisation of wing-fuselage intersections and tailplanes, for gliders and large UAV’s. For the expected operating conditions, a significant portion of the flow will be laminar but turbulent transition will occur and impact the results. SST has been used to date, but it would be very interesting to be able to benchmark some of the newer laminar-turbulent models also.


#4

@gsokoll - very interesting application! So the gamma-ReTheta model would be your preferred model? Or which models would be the most relevant / helpful?


#5

hello,

a big disadvantage of (all) turbulent models are that they are only valid if the whole boundary layer is turbulent.
The predicted drag is too high and if a airfoil is simulated the stall-angles are also higher than predicted.

With a (Gamma-Re) transition model the simulation is a bit more expensive in compution (it needs two more equation to be solved) but it is more accurate, if a transition occurs

Since openfoam 2.1.0 a transition model is implemented (kklOmega), maybe simscale will offer this boundary-model soon.

source: http://openfoam.org/release/2-1-0/physical-turbulence-combustion/


#6

@dheiny - To be honest, I am not sure which would be more appropriate - gamma-ReTheta or k-omega-v2 (I need to do some more reading !). But I believe either model would be a significant improvement over assuming fully-turbulent or fully-laminar flow models for this sort of application.

@mjosic - my understanding from the discussion below, the existing kklOmega model has some issues, in particular, growth of Laminar Kinetic energy when separation occurs, and for this reason the new model k-omega-v2 is recommended instead of klOmega.

http://www.cfd-online.com/Forums/openfoam-programming-development/85382-gamma-retheta-turbulence-model-predicting-transitional-flows-6.html#post607577