Colleagues, I have a 3-D, compressible flow CFD solution with a nicely refined wall mesh (Y+<1, multiple mesh in near-wall regions, etc.). I would like to post-process the solution across the wall boundary layer at varisou locations and cross plot the results to compare to the universal velocity distributions laws. My dilema is that the freestream flow has a continually varying velocity characteristic and I am not sure how to accurately determine the "edge of boundary layer". (Simple rules like Vedge=0.99*Vfreestream are not applicable, since Vfreestream varies continually). How can I determine the "edge of boundary layer" location accurately from a general 3-D result, which tells me flow variables as a function of normal distance from a wall point? Any insights into this is appreciated.

You might want to try the 'displacement' boundary layer, which is a lot less arbitrary than the usual '% of free stream.'

I just put 'displacement boundary layer' into Google and got 1,880,000 'hits' including some current text books.

Good luck!