I want to do RANS with variabl
 

I want to do RANS with variable density. The governing equations for my problem known but I was wondering how to treat the favre averaged values!

- Are the variables automatically favre averaged when using a compressible RANS-model?

- When the solver computes i.e. k,epsilon ... are these the Favre averged values which I get back?

As you can see I am a little bit confused with these averaging things.

Yes, and yes. The point of
 

Yes, and yes.

The point of Favre averaging is that it is a mathematical trick designed to get us back to the same form of the equations. We _know_ how to solve the NSE in the form

ddt(rho u) + div(rho u u) = ...

When we average this equation /\ we get

ddt(/rho u \) + ....

Favre averaging allows us to split /rho u\ into separate variables, getting back to where we started again. So, yes; when using a compressible RANS model, the variables will be by definition assumed to be Favre averaged.

Gavin

Favre or RANS averaged equations are used in incompresible reacting flow ?
 

I'm simulating incompresible reacting flow case where RSM is used as turbulence model. I have doubt about what fluent is actually solving for my case. Whether Favre-averaged NSE or RANS averaged NSE.

Can anybody help ?