Dear All

I have been using turbulence models in CFD programs for quit some time now. So I thought I better start to get a better insight in how it affects the Navier Stokes Equation. What I have thought until now is that a turbulence model gives an additional contridution to the viscosity used in the Navier Stukoes equation!! Then I started reading and discovered that what the turbulence models actually contribute with is the Reynolds stress tensor. Is that it?? Does a turbulence model like the k-e only calculate a Reynolds stress tensor that is used in the Navier Stokes equation?? And how is this then included in the linerised equation?? Is it just a part of the convective coefficient??

Regards

Bo

A statistical turbulence model provides a closure for the unknown correlations that represent the turbulent fluxes of momentum (Reynolds stress tensor). The Reynolds stresses may be determined directly by solving a set of modelled transport equations for the Reynolds stresses themselves. The length scale must almost be determined, usually by solving for the dissipation rate e. However, conventional 2-equation models, like the k-e model, make use of the Boussinesq stress-strain relationship that relates the Reynolds stresses to the mean rate of strain through a scalar eddy viscosity. You can see how this enters the momentum equations if you substitute the Boussinesq stress-strain relationship for the Reynolds stresses in the statistically-averaged momentum equations.