Turbulence Modelling
 

I want to find out more about turbulence models for analysing flow over a backward facing step. The best model to use or which model is suited to certain conditions. Also what assumptions do these models require or use for accurate results. I'm trying to analyse reattachment length for different turbulence models. Where can I find information? Richard

Re: Turbulence Modelling
 

Hi;

The k-e turbulence model is a popular and computationaly tractable and robust model. The old version of this model(Std. k-e) is not good to model a flow field with recirculation or with separating zone. This version can not predict real length of separation zone. But the recent version of this model(RNG k-e) is suitable for modeling of flow field with separation zone or recirculation. We can consider the low reynolds number effects in this version.

However the best model but very computationaly expensive to predict the separation zone is Reynols stress model.

The software documentation is very good to understand turbulence modelling in fluent and others.

Hope this help you

ROOZBEH

Re: Turbulence Modelling
 

I have been working on supersonic gas flow thru a convergent-divergent (Laval) Nozzle, using Fluent. The inlet conditions are 50 bar, 3000 K, Back pressure is 5480 Pa with 216 K. I have used k-E RNG model, with inlet and outlet pressure as boundary condition. (Outlet pressure defined as the same as back pressure)

The results show an abrupt reduction in Mach number just before the exit line and only near the symmetry axis of the nozzle. No abnormality found near walls. At the point where Mach number decreases, static temp has increased, total pressure decreased, but surprisingly the static pressure and density remains constant. Just before this abnormality, the radial velocity becomes negative, that is, the gas flows towards axis of symmetry; which in my opinion is the cause of reduction/ disturbance of axial velocity.

I need help in understanding the cause of this phenomenon. Also, please advise what is the right turbulence model for my case. The analytical, one dimensional solution suggests an exit mach number of 3.6, whereas I am getting 3.4 to 3.5 at the exit line except the region of abnormality where it has decreased upto 3.0

Thanks Sohail Ahmed