[eaglemckenna]

Hi there,
I am working on a case of mixed convection of an aircraft crown compartment, that is populated with a heat dissipating avionic component, the case is in 2d and has inlet and outlets.
My expertise is not in CFD per se but in the development of a Reduced Order Model using the 4 fundamental subspaces of linear algebra.
For my previous case of natural convection the dimensionless parameter that best described the case was Grashof ratio of cylinder to compartment.

For mixed convection, is the Peclet number a more appropriate parameter.
Regarding solver settings I assume implicit solver is key due to the flow dependence, would this be the case?

I am using RNG K-e enhanced wall treatment with boussineseq approximation - so for my fine grid I want a y plus of around 1? Is this correct?

Is a transient set up required, or is steady state ok - what convergence criteria should I impose, I should allow the case to run 10,000 iterations?
Max flow at inlet is 0.045m/s.

Please provide as much information as possible, help appreciated.
cheers
g