Technicalities

Lest I be guilty of over-reliance upon the paperback writer's cheap expedient of suspense to keep you reading, I will reveal right now that they are vortex generators. Their purpose is to flush low-energy air out of the intersections. They do this by generating a tip vortex, just as any wing at a non-zero angle of attack would. The vortex, if you could see it, would look like a slender rope trailing from the vortex generator and following the contour of the intersection. In fact, when atmospheric conditions are right you can sometimes see just such vortices, made visible by condensation, trailing from vortex generators on the nacelles of some airline jets such as big-engine 737s. They appear at the same time solid and ghostly, and are capable of the most delicate, graceful and agile responses to changes in attitude or G-loading. They are, in fact, quite beautiful. In aerodynamics, aesthetic enjoyment is a frequent, albeit unintended, side effect.

Usually, we think of vortices as drag producers. Wingtip vortices, which are evidence of the effort needed to levitate thousands of pounds of stuff upon thin air, do not speed up an airplane. But under some circumstances small vortices can be used to reduce drag. To understand this hair-of-the-dog effect, it's helpful to think a little about what drag is. It is not merely a fancy term for colliding head-on with air.

When we talk about aerodynamic phenomena, we tend to treat the airplane as a stationary object with air flowing past it, as though it were mounted in a wind tunnel. While this view, which has the advantage of keeping the airplane before us for more than a second, is physically equivalent to the more realistic one of the stationary air and a moving airplane, it hampers an intuitive grasp of what's going on. Instead, for the moment let's imagine ourselves standing still as an airplane flies by. If we could see the air, it would be at rest as the airplane approached and would be pushed aside as the airplane passed. In the airplane's wake, some air would be moving in the direction the airplane moved, as if trying to follow it. This entrained air would swirl around in random ways for a few minutes before again becoming motionless. The effects would be similar to those we observe when we stand at the side of a road and a truck roars by.

The air entrained by a perfectly streamlined body consists of only a thin film adjacent to its surface. Within this film, called the boundary layer, there is a progressive change in speed: The closer to the surface it is, the closer the speed of the entrained air is to that of the moving body, until the molecules in immediate contact with the surface-which, at their scale, is of a mountainous roughness-are carried along at the speed of the airplane. The boundary layer, which is very thin at the nose but grows thicker and thicker toward the tail, takes its momentum from the airplane. After the airplane has passed, the air in the boundary layer continues to be carried forward by that momentum. It mingles with nearby stationary air, creating swirls and eddies which viscosity eventually damps out.

Nearly all of the drag of a well-streamlined object manifests itself as the momentum of the boundary-layer air that it leaves in its wake. Several factors influence boundary-layer momentum, but the dominant one, and the one most intuitively obvious, is the thickness of the boundary layer. The thicker the boundary layer, other things being equal, the more air there is in it and the greater the drag it represents.

Now air, like water, does not like to flow uphill. "Uphill," in the case of air, means from low pressure to high. Just as momentum can carry moving water up a slope for some distance, it can also propel air against rising pressure. But not for long. In the rear portion of a streamlined body, where its surfaces are converging, pressure rises. As the air closest to the surface, which is being carried along at nearly the speed of the airplane, meets this rising pressure, its momentum is insufficient to drive it up the pressure hill, and instead separation occurs.

Separation means that the boundary layer detaches from the surface and rides up over a layer of invading air that is flowing in, either from behind or from either side. This is what happens when a wing stalls. High pressure boundary-layer air from below the wing curls around the trailing edge and advances upstream, prying the boundary layer away from the surface. As the boundary layer leaves the wing surface, the curvature of its path around the wing flattens, and lift, which is the result of bending the path of passing air, is reduced.

The reverse-flowing air is being carried along by the airplane, and there has to be enough of it to occupy the space vacated by the separated upper-surface boundary layer. Accelerating that volume of air to the speed of the airplane is the source of the rapid drag rise associated with the stall.

A wing can be partially stalled and go on flying. In particular, air close to protrusions on the wing surface tends to lose energy and separate prematurely, while the flow over other portions of the wing remains attached to the surface. Wing root intersections on low-wing airplanes often tend to stall well before the rest of the wing. They may be partially stalled even at climbing speed.

The purpose of wing root fillets is to fill in the channel formed by wing and fuselage, reducing the steepness of the pressure hill that it creates. But a similar effect can be achieved, with less weight, by a vortex generator. Rather than change the physical contour of the surface, the vortex generator directs a stream of high-energy air into the separation-prone area. "High-energy" simply means "fast-moving." The natural question is, where does it get this air?

That is the genius of the vortex generator. It is really a short wing flying along at a high lift coefficient with its tip outside the boundary layer. The rapidly spinning vortex that it creates drags air from outside the boundary layer toward the surface. The rope of the vortex resembles a stream of water from a hose, flushing the "dead" air in the bottom of the boundary layer back along the wing.

So much for the Cessna 303 and its leading-edge vortex generators. Vortex generators also appear in many other contexts. They can be used to help air round a sharp corner, as at the back end of a Cessna 337; to delay stall and reduce landing speed, as in various aftermarket systems that are applied close to the leading edges of wings; to bring high-energy air into contact with ailerons, improving roll response at high angles of attack; and to remedy boundary-layer thickening from transonic shock. There are also vortex generators that don't look like vortex generators. Sawtooth and notched leading-edges, for instance, create a vortex that grows stronger with increasing angle of attack, staving off spanwise movement of the boundary layer in the same way that a wing fence would. Some airplanes with thick, strongly swept leading edges have small oval surfaces protruding vertically below the leading edge; called vortilons, they too are intended to delay spanwise migration of the boundary layer.

There is another aerodynamic device, the strake, that functions very similarly to a vortex generator. In fact, a strake is a vortex generator, but of a different shape and location from the miniature wings we usually associate with that name.

The most common application of a strake is the dorsal fin, which many airplanes have, whether they need it or not. A dorsal fin-the narrow triangle projecting forward from the base of the vertical fin-has a particular purpose: to prevent hard-over rudder lock, an unpleasant phenomenon that can occur if an airplane yaws sufficiently to stall the fin. When air flows diagonally across the dorsal fin, it curls over the leading edge-which should be sharp, to encourage clean separation-and forms a vortex that pulls high-energy air into contact with the fin. Not all airplanes need dorsal fins; they are mostly associated with spin recovery, and most airplanes aren't intended to spin. But they have become an expected element, to the point that, as with nosewheel shimmy dampers, designers sometimes toss them in rather than risk finding out later, from somebody's bitter experience, that they were, in fact, needed.

Strakes are sometimes found ahead of the horizontal tail for the same reason-to prevent its stalling at very high angles of attack, as would occur, again, in a spin.

The ability of a strake to keep airflow attached to wings at very high angles of attack has made them de rigueur in modern fighter design. The F-16 and F-18 have them, as do practically all fighters designed since the 1960s. The F-22 appears to have no strakes, but close examination of the engine air intakes reveals a sharp-edged upper corner intended to produce a vortex. This system works, incidentally, only for wings of low aspect ratio, because the effect of the strake is felt over the entire span.

Vortex generators seem like a discordant element in aerodynamic design, violating the rule that smoothness is desirable above all things. I am tempted to assert that they do not occur in nature, but if I do, a scholarly reader is bound to write pointing out that some benthic invertebrate or nubbly cetacean has them. The small finger-like feathers that protrude from the midspan joint of some bird's wings may be vortex generators, but they may also just be loose feathers. Not everything in nature has a purpose, and birds have brains, which are better at controlling airflow than any vortex generator. I am doing my best to think of a fish with vortex generators, and I can't. Bugs, especially small ones, have turbulators-some really tiny ones have wings made entirely of hair-but they're a different story, having more in common with the dimples in golf balls than with miniature wings. The vortex generator is, perhaps, like the wheel, a distinctly human contribution to the universe of useful things.