[tstorm]

I'm doing a validation study for a user-defined v2-f turbulence model in FLUENT. I used the empirical laminar flow equation for skin friction: 0.664/sqrt(Re_x) to compare results for several turbulence models. y+ is around 1.

I'm finding that Spalart-Allmaras, k-epsilon, realizable k-epsilon, k-epsilon RNG, k-omega, and k-omega SST all overpredict the skin friction coefficient by 20-50%. Interestingly enough, the v2-f models predict nearly identical skin friction coefficient distributions as the empirical equation.

I have a 2 questions:
1. Why do the standard models over predict Cf for such a simple flow?
2. Is this typical for these models?