Why Hold the Aircraft Still and Move the Air? Relative Motion Is All That Counts

The forces on a body moving through air depend on the relative motion between the two and not on which of them is moving, so a stationary model in a moving stream experiences the same lift and drag as a moving model in still air. That equivalence turns an experiment that requires launching something and chasing it into one where the object sits still and instruments are attached to it directly. Forces can be measured on a balance beneath the model, pressures at hundreds of points on its surface, and the flow itself visualised with smoke, tufts or laser sheets. Nothing has to be recovered afterwards, and the conditions can be repeated exactly.

The layout follows from the need for smooth uniform flow at the test point:

• •A fan or compressor driving the air, usually placed downstream of the model
• •A settling chamber with screens and honeycomb that straightens the flow
• •A contraction that accelerates the air and reduces its turbulence
• •The test section, where the model sits and the measurements are made
• •A diffuser that slows the air again to recover energy
• •A closed circuit returning the air, in most large installations

Testing a model a tenth the size of the real aircraft does not automatically tell you about the real aircraft, and resolving that is the central difficulty. What governs the flow is a combination of size, speed, density and the stickiness of the air, and the same combination must be matched between model and full size for the behaviour to correspond. A model a tenth the size therefore needs the air to be ten times faster, or ten times denser, or considerably colder, to produce matching conditions. Facilities achieve this by pressurising the air to many atmospheres or by cooling it with liquid nitrogen to extremely low temperatures, both of which are expensive and both of which exist for exactly this reason.

Measuring forces gives numbers and seeing the flow gives understanding, which is why visualisation techniques have developed alongside the measurement. Smoke released from a fine tube upstream traces a line through the flow and shows where it separates from a surface, which is the single most informative thing to know about a wing. Short tufts of wool taped to the surface lie flat in attached flow and thrash about where it separates, which costs nothing and works. Oil mixed with pigment painted onto the model dries into a pattern recording the surface flow. Laser sheets illuminating tiny seeded particles, photographed twice in quick succession, give the velocity at thousands of points at once and have become the standard research method.

The technique spread well beyond aviation and several applications now account for more testing than aircraft do. Cars are developed in them continuously, since drag determines fuel consumption and downforce determines cornering, and motorsport teams have historically spent enormous sums on facilities until the governing bodies restricted the hours permitted. Buildings are tested at model scale for wind loading and for the pedestrian-level gusts that tall towers generate at their base, which has forced redesigns of several completed buildings. Bridges are tested for the oscillation that destroyed one famously in 1940. Athletes, cyclists and ski jumpers are tested directly. Computational simulation has replaced a large share of the routine work and has not replaced the final verification.

Forces depend on relative motion, so a stationary model in moving air experiences what a moving model in still air would, which allows instruments to be attached directly and conditions repeated exactly. Screens, a contraction and a diffuser produce smooth uniform flow. Matching a small model to a full-size aircraft requires matching a combination of size, speed and air properties, which is why facilities pressurise or deeply cool the air.