Anyone got any thoughts on the relationship between y+ at the top of a boundary layer as a function of the bulk flow Reynolds number?

Fred.

hi there,

if you considered the case of the channel flow, and use the Re based on Delta (boundary layer thickness), then y+ at the top of the BL is Re(delta).

sincerely,

Frederic Felten.

OK, but would this lend itself to some kind of blending criteria for near wall/far field turbulence models.

You could compute y+ at all points in the flow (if you knew Delta, or at least had the grid to resolve it) but when does the BL end and the bulk flow begin? I was thinking that there may be a Nu+ (Mu_eff/Mu_lam) criteria that itself is Re dependent....??

Fred.

hi there,

Usually, the criteria used to determined the top of the BL is approximatively when the ratio U/Uinf=0.99

I think that some of these issues are discussed in details in: 1- "Turbulence", by Hinze J.O. (1975). 2- "Boundary layer theory", by Schlichting H. (1979).

sincerely,

Frederic Felten.

U/Uinf is all very well if one knows what Uinf is! There are a few possible pointers in Hinze for a more apriori approach. I wonder if any DNS work has shed light on this. IMO this is the main benefit of DNS work, to provide insights to enable the evolution of 'lower order' turbulence models.

Fred.

hi there,

Well if you non-dimensionalized everything appropriatly like U*=U/Uinf, then you don't need to know Uinf. so check the non-dimensional analysis and you should be able to get the criteria for the top of the BL. Now, about DNS work on this, DNS is just a way of solving the flow, it doesn't have any impact on how one set the top of the BL. At that point, i think that you need to check some articles about turbulent channel flow. For example: 1- Moser, R.D., Kim, J. and Mansour, N.N. (1999) Direct Numerical Simulation of Turbulent channel flow up to Ret= 590. Phys. Fluids, vol 11, 4, pp 943-945.

More references will be given in this paper. Sincerely,

frederic felten.

Hi,

y+ is an inner scaled coordinate. It loses its meaning after the sub-layer and the inertial layer. In these layers it is a useful scaling parameter. Using y+ to define the BLayer thickness doesn't make any sense.

chidu...