y+ is the dimensionless distance showing where the cell nearest the wall falls within the boundary layer. Reports usually quote a single average; what should be examined is the distribution.
Two legitimate strategies
Resolving to the wall. Target y+ < 1; the viscous sublayer is resolved by the mesh. Required for quantities that depend on the near-wall region — separation, heat transfer, drag.
Wall functions. Stay in the y+ ≈ 30–300 range; the near-wall region is not resolved but bridged with semi-empirical profiles. The mesh becomes cheaper. Acceptable when the flow is attached and the near-wall region follows the classical logarithmic profile.
The dangerous band: 1 to 30
This range is the buffer layer, where neither approach is fully valid. Too coarse to resolve to the wall, too fine for a wall function. If a mesh's y+ distribution falls largely in this band, the mesh is caught between two strategies and the near-wall behaviour of the result is unreliable.
Where it is high and where it is low matters
y+ depends on the local velocity, so on the same mesh it comes out very low at a stagnation point and very high in an accelerating region. The acceptance criterion is this: be in the target range in the region where the decision is made.
If a blade averages y+ = 1 but reaches y+ = 40 in the separation region of the suction side, that mesh cannot answer the question. Conversely, y+ = 200 in a volume where heat transfer is irrelevant is not a problem.
It must match the model
Running a model designed for near-wall resolution, such as SST, at y+ = 50 throws away the model's main advantage [1]. Model choice and mesh decision are made together. See the turbulence model note and the boundary layer mesh note.
References
- Menter, F. R. (1994). Two-equation eddy-viscosity turbulence models for engineering applications. AIAA Journal, 32(8), 1598–1605. DOI: 10.2514/3.12149
- Launder, B. E. & Spalding, D. B. (1974). The numerical computation of turbulent flows. Computer Methods in Applied Mechanics and Engineering, 3(2), 269–289. DOI: 10.1016/0045-7825(74)90029-2