Airfoil Flow Lab

Binns Flight Services

A 2D "smoke tunnel" showing airflow around a wing section. Smoke lines enter from the left (the relative wind), and the airfoil pivots about its quarter-chord as you change the angle of attack.

Things to try

• Smooth flow: at low AoA the smoke follows both surfaces and leaves the trailing edge smoothly.
• Creeping separation: as you approach the critical AoA, the flow starts separating at the trailing edge and the separation point (orange marker) moves forward.
• Stall: past the critical AoA the flow breaks away from most of the upper surface, lift drops, and drag rises sharply. L/D collapses.
• Recovery: lower the AoA again. The flow does not reattach instantly. You usually have to go a few degrees below the stall angle, and it then takes time for the separation point to move back. Watch the dot on the C_L graph trace a hysteresis loop.
• Compare airfoils: the thin airfoil stalls abruptly from the leading edge. Thicker and cambered sections stall more gently from the trailing edge. Camber gives lift at zero AoA.
• Flaps: lowering the flap adds camber. You get more lift at the same AoA and a higher C_Lmax, but the wing stalls at a lower angle of attack and drag goes up a lot. Compare the curve with the dashed flaps-up curve on the graph. At 30° you can see some flow separating over the flap itself. AoA is still measured from the wing's chord line (flaps up).
• Starting vortex: make a sudden AoA change and look behind the trailing edge (turn on "Vortices").

Readouts

C_L, C_D and L/D are 2D section values (an infinitely long wing), so there is no induced drag. That's why L/D is much higher than a whole airplane's. Stall angles and C_Lmax are typical of these NACA sections at full-scale Reynolds numbers. The stall warning comes on about 2° before the critical angle, like a stall horn.

About the model

This is a teaching model, not a research CFD code. The flow around the airfoil comes from a panel method (potential flow). Separation is driven by a dynamic stall model (Beddoes–Leishman style), and the turbulent separated wake is made by shedding discrete vortices from the separation point and the trailing edge. The trends are physically realistic, but the exact numbers are not wind-tunnel data.

Keys: ↑/↓ or ←/→ change AoA · Space pause · R reset