[aerohead]

Originally Posted by freebeard

I'm having trouble sorting out whether the viscosity of the fluid needs to be greater or less at smaller scales.

Gnats feel air like we do water, so maybe a less viscous fluid like ethanol?

Here's a CFD analysis of fifth-wheel teardrops: www.sciencedirect.com: Shape optimisation of teardrop trailers to minimise aerodynamic drag in articulated lorries

1) It all come down to Reynolds number, which must always be 'supercritical.' ( one-million or above ).
2) 'Kinematic viscosity' is part of the equation used to calculate Rn. It must be known, just as with air density ( rho ) in the drag force calculation.
3) 'Scale' is the other consideration, since 'Length' of the body under consideration is literally one of the factors used in the equation used to calculate Rn .
4) 'Velocity' is the limiting factor in scale testing, as there will be a point where flow is so great that air becomes 'compressible' and supersonic shockwave drag can be present around the body, due to it's super-velocity effects.
5) We must limit ourselves to 'low-speed' aerodynamics, of 'subsonic' flow.