Scaling and similarity
Thermal and mass convection - General principles

After T _M , logic would suggest a direct calculation of h. In fact, an intermediate step is necessary. Any calculation of h, based even partially on purely theoretical conditions, requires experimental confirmation; in other words, we would need as many experiments as there are special cases concerning either the different geometries, or the two main classes of flow (laminar or turbulent), or the thermal boundary conditions, which we have seen can be infinitely varied.

Similarity relations...