[DBFIU]

Hello,

I wanted to start a general 'rules of thumb' thread for this topic. I am a heat transfer analyst and I am being tasked to perform some CFD design sidework, I think it would be a powerful combination if I could generate my own HTCs using CFD.

Can someone please guide me in the right direction when it comes to things you aren't or are supposed to do?

For instance, I use SST model in CFX, ICEM meshed. What is a good target y+ value if I want to see legitimate heat transfer coefficients?

Let's assume this is not conjugate and that the body and gas temperatures are close enough not to warrant changes in gas properties at the surface. Ruling out all these variables, I want to use CFX to produce HTC's, sorry if this was answered before; I searched and therei s little info on this subject.



Thanks,

- D

[julien.decharentenay]

Hi D,

I have got 2 rules for you:

Rule 1: Benchmark
Use test cases with experimental data/analytic results/engineering design guides to validate the approach (mesh, model, convergence criteria) you take. Obviously choose cases that represents some of the physics you are trying to model.

Rule 2: Assess your results
Check the effect of the mesh, convergence and other parameters on the solution.

Regards,