Why Reynolds Number Dominates Small UAV Design

The parameter that changes everything

Reynolds number is the ratio of inertial to viscous forces in a flow:

Re = ρ · V · c / μ

For a transport aircraft, chord-based Reynolds number sits in the tens of millions. For a small fixed-wing UAV with a 0.2 m chord flying at 15 m/s, it is roughly 200,000. A hand-launched micro-UAV may operate below 50,000.

This is not a modest quantitative shift. It is a change of physical regime, and aerofoil data measured at high Reynolds number does not transfer.

What goes wrong below Re ≈ 500,000

At high Reynolds number the boundary layer transitions to turbulence early and naturally. A turbulent boundary layer resists separation well, because turbulent mixing continuously carries high-momentum fluid from the outer flow toward the surface.

At low Reynolds number the boundary layer remains laminar over much of the chord. Laminar layers carry far less momentum near the wall, so when they encounter the adverse pressure gradient past the point of maximum thickness, they separate.

Frequently the separated shear layer then transitions to turbulence and reattaches further downstream, enclosing a region of recirculating flow known as a laminar separation bubble. The bubble thickens the effective body, raises pressure drag, and — most dangerously — can burst without warning as angle of attack increases. The result is an abrupt stall with little of the buffet that normally warns a pilot.

The practical consequences

Aerofoil selection changes entirely. Sections optimised for high Reynolds number, including many NACA 6-series laminar-flow profiles, perform poorly here. Aerofoils developed specifically for low Reynolds number — the Selig, Eppler and Wortmann families — manage the bubble deliberately rather than assuming it away.

Thickness and camber shift. Very thick sections separate early at low Reynolds number. Moderate thickness combined with generous camber usually outperforms a scaled-down high-Reynolds-number section.

Lift-to-drag ratio falls hard. A section reaching an L/D of 150 at Re = 3,000,000 may achieve 60 at Re = 200,000, and 30 below 100,000. Range and endurance estimates built on full-scale polars will be substantially optimistic.

Surface finish stops being cosmetic. At high Reynolds number, minor roughness is lost within an already-turbulent layer. At low Reynolds number, roughness — or a trip strip placed deliberately — can improve performance by forcing transition ahead of the separation point.

Designing for it rather than around it

Turbulators are standard practice on sailplanes and small UAVs: a trip strip, a zigzag tape, or a small step positioned just ahead of where separation would otherwise occur. They cost a little skin friction and can save a great deal of pressure drag by preventing the bubble from forming or from bursting.

The wider lesson is that scaling an aircraft is not a geometric operation. Reynolds number scales with size and speed, and the physics changes as it falls. Validate at the Reynolds number you will actually fly at, not the one your reference data was measured at.

Rough regime guide

Before you commit

Obtain or generate polars at your design Reynolds number. Tools such as XFOIL, used within their known limitations, are far better than extrapolating from handbook data measured an order of magnitude away. Where endurance or range is a contractual figure, validate with a flight test before quoting it.