[ns_chrys]

I think this question has been asked quite a few times on CFD-online, but the answers I have found have not been satisfactory (either they will agree with a solver manual and not with a well established text on turbulence such as Wilcox's Turbulence Modelling or vice versa)

So getting to the point =>

I understand that for the purposes of modelling, there are technically two zones in a turbulent boundary layer (and solving for each layer has its own y+ requirements):
1) inner layer = viscous sub-layer (y+ <=5) & Buffer layer (5 < y+ < 30)
2) Outer layer = overlap log-law layer (y+ >= 30) + Fully turbulent

The log law layer connects the inner layer and the outer layer and can be considered to be a layer on its own - which is fine! After going through the chapter on turbulence modelling in Frank White's viscous fluid flow and comparing my notes from Wilcox's text and the StarCCM+ user manual, I think I have a pretty good understanding on the reason why the y+ values are what they are for every zone except the buffer layer:

What is so "special" about the buffer layer that all solver manuals (and none of the textbooks) tell one to choose a y+ value that does not fall in this range? White's text has a formula {pg 419} (due to Spalding in 1961) that merges the viscous sub layer + buffer layer and the log law layer. And apparently its well accepted.

If this is the case, then I'm sure the solvers of today have that formula incorporated. Which leads me to ask why do their manuals say don't use a y+ between 5 and 30 {which btw is also the conclusion of this paper = Wall y+ Strategy for Dealing with Wall-bounded Turbulent Flows; due to Salim M. Salim and S.C. Cheah}