• Flight Today

Ready, Set, Flap!

Birds do it, bees do it. Can two weird aircraft make aviation history doing it?

• By Graham Chandler
• Air & Space Magazine, January 01, 2002

THIS IS A STORY ABOUT TWO GUYS named Jim. Both were born in the 1940s, both grew up in the same neighborhood of Park Forest, a town on the south edge of Chicago, and both attended the same church, but neither knew about the other until one of those one-in-a-billion coincidences. They became acquainted through a seminar at the 1999 Experimental Aircraft Association fly-in at Oshkosh, Wisconsin, and discovered they were probably the two people in the world closest to achieving an aeronautical dream that has eluded humans since the dawn of time.

This morning I'm standing next to one of them, just outside a hangar at the decommissioned Royal Canadian Air Force Downsview base in Toronto. He's tall and straight, his name is James DeLaurier, and he is a professor at the University of Toronto's Institute for Aerospace Studies. Poised beside us is a little aircraft that may one day soon realize the dream of the tow Jims, a craft affectionately known as Big Flapper. It is intended to fly not with propellers, not with jets, but with flapping wings.

For centuries, attempts at building airplanes were, naturally enough, based on the way birds fly. But coming up with the perfect wing proved extraordinarily complex. Now, with the help of microcomputers and composite materials, the dream of bird-like flights is being revisited.

This morning's run is a taxi test to feel out a redesigned undercarriage. Test pilot Patricia Jones-Bowman arrives at the hangar in a beat-up Buick. She has chestnut-red hair and wears wraparound sunglasses and a blue flying suit. An ex-bush pilot, she volunteered for the project after hearing about it from her dentist, a talented woodworker who had built jigs for the Flapper's wing spars. Craning her neck, she eyes Flapper's wings, which are poised in the up position, like the wings of a hummingbird in a strobe photo.

There's not a breath of wind in the gray dawn as she takes a few complicated minutes to settle her tiny frame into the cockpit. One of the crew chiefs, Derek Bilyk, signals "All clear," and she hits the start button. The tranquil air is cracked by a dirt-bike-exhaust note, which is backed by a whining, something like that produced by the windshield wipers of an old school-bus: rr-RR rr-RR rr-RR. Big Flapper has awakened. It flaps its 41-foot wingspan like a giant rooster about to crow at the dawn. Its 24-horsepower two-stroke ultralight engine is coupled directly to the wings through a mechanical drive unit. A sprocket-and-chain mechanism drives two pylons, located just behind the pilot, up and down, raising and lowering the center sections of the three-section wings. At top engine performance, 3,800 rpm, the wings flap 1.3 times a second. But that's not achieved until the craft is on the runway. For this runup, 0.94 per second is about the maximum at which the bucking beast can be held down.

Big Flapper was built in 1996 but is the culmination of over three decades of research. DeLaurier's partner, Jerry Harris, who conceived Flapper and paid for its construction, recounts: "I first considered flapping flight in 1968, as an application for the mechanical power amplifier I was analyzing for my master's thesis at Ohio State University." He began collaborating with DeLaurier in their spare time while both were working at Battelle Memorial Institute in Columbus, Ohio, in 1973. "Natural or flapping flight was one of the few areas still open to fundamental investigation and engineering analysis," says Harris. "Fundamental" is the critical word here; as far as practical applications go, there is little at present—a fact that appears to bother none of the team members.

When DeLaurier accepted a position teaching aeronautical engineering at the University of Toronto in 1974, flapping-wing research did produce one practical offshoot of sorts: It gave some of DeLaurier's students a focus for study. Around a dozen of senior and master's theses, as well as a doctoral dissertation, have helped design Big Flapper. And one student, Theresa Robinson, is doing a master's thesis on the use of ornithopters to explore in thin atmospheres like that of Mars. But theory and models can predict only so much. Flapping-wing flight today is akin to supersonic flight in the 1940s—an unknown regime where stability and control are only educated guesses. To find out what happens, you have to do it for real.

"By far the biggest design challenge was the wing," shouts DeLaurier over the din. "I would say this wing is the most technologically complex wing in history—it's the first that has had to provide both thrust and lift." Early on, DeLaurier and Harris studied film clips of bird flight in slow motion. "There were just too many different motions happening at once," DeLaurier recalls. "The thought of modeling it was intimidating." In designing Flapper, DeLaurier figured that it would be enough to produce a careful combination of flapping and twisting to generate lift and thrust. On the upstroke, the wing twists to a positive angle relative to the fuselage, and on the downstroke, it twists to a negative one. This provides the thrust necessary for forward flight. But achieving the optimum twist was no cakewalk. Either too much twist or too little would produce inadequate lift. In fact, too much twist would take energy out of the airstream and produce negative thrust—drag.

And while the wing is twisting, it also needs to maintain the right degree of bending rigidity. DeLaurier explains the concept by asking me to imagine holding the ends of a cardboard tube. "Try and twist it," he says, "and it remains rigid. But slit it down the length of one side and overlap the edges and you can twist it while it remains stiff." DeLaurier and Harris pulled off a similar innovation by conceiving a shear-flexing design. Each wing is composed of two flexing sections of polyester; during flapping the sections slide over the wing ribs. The sections are attached to what is essentially a double trailing edge: One section slides over the other, like two edges of the cardboard tube.

Harris and DeLaurier first tested the concept with a 10-foot-span radio-controlled model the named "Mr. Bill" (after the misfortune-prone "Saturday Night Live" creation). It took years of research, they say before Mr. Bill was able to fly with their shear-flexing design and demonstrate their method of three-axis control, which they later used in Big Flapper.

In the full-size ornithopter, the pilot controls pitch by manipulating the horizontal stabilizer. But the third function of a flapping wing, lateral control, hasn't been included on Flapper's wing. The shear-flexing wing design precludes the use of standard ailerons for direct roll control. So Flapper is designed to turn solely by rudder. The pilot will bank by a technique known as yaw-roll coupling. Lateral stick input deflects the rudder, yawing the aircraft. The windward wing experiences an increase in angle of attack and airspeed and therefore enhanced life, which rolls the aircraft in the direction of the turn.

Flapper has yet to demonstrate that it can actually do any of this. During a test in November 1998, as Flapper exceeded 50 mph, it started to lift off the runway, then smacked down on the ground so hard that the nose gear failed. To keep the craft on the runway while the speed increased, the team shortened the nose gear so Flapper had a nose-down angle. That suppressed lift buildup. "What we hadn't accounted for," says DeLaurier, "was that the new down force from the wings was imposing compressive loads in our vertical struts beyond the design limit." During another test, this one in October 1999, Jones-Bowman reached 56 mph—just short of Flapper's predicted 57-mpg takeoff speed—when one of the vertical struts buckled. The team reinforced the struts.

Once we get clearance from the tower, Jones-Bowman comes up on the throttle and Flapper beats its wings down the runway. Taking their cues from the rhythm of the wings, the fuselage pulsates up and down, the tail takes little dips, the main gear does mini-squats, and the wing supports flex. The cadence picks up as Jones-Bowman accelerates to 35 mph, turns around, and does it again. We're in formation at her two o'clock, sitting in the back of a rusty VW pickup, when Flapper's engine sputters and dies.

The crew pushes Flapper off the active runway. After 20 minutes, the problem is declared to be a blockage in the fuel filter. An engine failure in flight has been thought of, and DeLaurier calculates that wing loading—essentially, the difference between the pressure of the air above the wing and that below it—would overcome the engine compression and drive the wings to their full-up position. "Lateral control would be pretty sensitive with all that dihedral," he says. "But with small and tender inputs, it would be flyable."

By now the wind is picking up. Since sideslipping into a crosswind without both aileron and rudder is dicey, testing the aileron-less Flapper is best done in calm, limp-windsock mornings, so testing is called off for the day.

I ask Jones-Bowman how it feels to be inside the Big Flapper. "I'm steering, controlling, bouncing, watching instruments and everything," she says. "Once I rev up, it's up and forward, down and backward, and the stick goes with me. I tell myself not to try and stop it but it's difficult. If I do, PIOs are gonna be a problem." ("PIOs" are pilot-induced oscillations—the pilot's attempts to correct pitching motions actually increase their amplitude, rather than diminish them.)

A few months following my interview, Jones-Bowman resigned as project test pilot, citing safety concerns related to the liftoff incident. Jack Sanderson, a longtime ultralight and homebuilt enthusiast, jumped at the chance to replace her. "I had to go on a crash diet," he says. "But once I got in the cockpit it felt natural."

Near the start of his project, DeLaurier had heard that other attempts were under way to make the first flapping-wing flight. But he never met the other Jim until one of his students attended a Flapping Forum at the 1999 Oshkosh fly-in and heard Jim Theis talk about his project. The following May, the two Jims got together at Zumbrota, Minnesota, where Theis' ornithopter project, Nighthawk, lay in parts. After that meeting, Theis and project manager Brian Said, perhaps inspired by Big Flapper (or by the thought of Flapper beating Nighthawk), accelerated their efforts to make the first ornithopter flight. BY late October they had finalized a plan to fly within a year. Things were well under way when Theis died last January. "Jim Theis was one of the most innovative researchers I've ever met," DeLaurier says today. "If our respective design approaches hadn't been so far along, I'm sure that we would have collaborated." The Nighthawk project is now in the hands of Brian Said.

For hundreds of miles along all Ocean Boulevards of Florida's Atlantic coast, pelicans enjoy catching ocean breezes, using the occasional flap of a wing to correct for gusts or to gain a little altitude. Brian Said lives here with his wife Winnie, in a palm-shaded house near the Jupiter Inlet, and he envisions human beings delighting in the same sort of flying. The deep-thinking engineer is surrounded by waist-high stacks of logbooks and the da Vinci-esque sketches of Jim Theis, whom he met in 1976 at Florida Atlantic University.

Said's ornithopter design philosophy differs from DeLaurier's. He wants to fly more like birds: control flapping angles and rates, no vertical tail, soar when the currents are right, turn with wing warping rather than yaw-roll coupling to a vertical rudder.

Without Theis, Said has more to do than DeLaurier and Harris. The original Nighthawk, built more than 20 years ago, was not an ornithopter—it had no flapping-wing drive—but rather a smaller-than-average ultralight with a pusher drop. It was designed to prove the team's system for lateral control. Said explains that in the wing warping that the Wright brothers used for lateral control, the warping was coupled—when on wing warped, the other warped in the opposite direction. That produces adverse yaw, he says, and that's why the Wrights had to add a vertical stabilizer. Then Said pulls out a case of video tapes to demonstrate his and Theis' alternative—"independent reverse wing warp."

We watch scenes reminiscent of early black-and-white movies of flight. In them, Theis lays prone in Nighthawk I's cocoon-like cockpit, gets airborne to roughly 10 or 12 feet at about the speed of a man running, and flies back and forth across a field, turning at each end like a bird. There's no vertical stabilizer or rudder, just a separately controlled pigeon-like tail that moves up-down, left-right, and twists around its long axis, plus wings that "reverse warp" independently. "With independent reverse wing warp," Said explains, "when you want to make a right turn, the right wing is twisted to a more positive angle." That increases the angle of attack, and therefore the lift and drag, on the right wing, which, when coordinated with tail inputs, turns the aircraft to the right. The movement is similar to the way a bird turns while soaring, deflecting airflow to one side of the other by adopting asymmetrical wing positions and twisting its tail. Said says there's no adverse yaw.

Testing progressed nicely until 1979, when, on a final test run over a polo field late in the day with low visibility, Theis pulled up over a scoreboard. On the other side was a speaker post; on wing struck it, and Nighthawk rolled over, slamming Theis into the ground. His back broken, he spent the rest of his life in a wheelchair. "Jim was undaunted, but [the accident] set back our program," says Said.

For the next 20 years, Theis and Said dabbled with designs for a flapping-wing version of Nighthawk, but earning a living took priority (Theis had his own engineering design firm, and Said is an engineer at Lockheed) and the project languished. That is, until the visit by DeLaurier. "After that we sat down and got our heads together seriously for a solid month—the longest sustained effort yet," says Said. Using Nighthawk's wing, Theis worked into the night completing all stress and performance analyses and computer-simulated lift and thrust curves. He and Said perfected a design for a flapping drive mechanism powered by hydraulic actuators. "It was a real light bulb for me," says Said, who like hydraulics because they enable the pilot to exert control over flapping angles and rates, with micro-computers used to regulate response to control inputs. "The actuator is designed for a high repetition rate and high fatigue tolerance," he says.

But research needs to be funded. "We hadn't actively sought external funding," says Said. For now, funding is split between Said and the Theis family, supplemented by donations of parts and materials from corporations.

Said is in the process of transporting the Nighthawk project from Theis' Minnesota home to Florida. Then he will install the flapping wing drive unit and press on toward a first flapping flight. He is looking for funding to accelerate the effort. The Theis family name may stay with it: Jim's son Charlie is a Delta Air Lines pilot, and he says he'd love to fly it.

Will humans ever produce a workable technology for flapping-wing flight, or is it really just for the birds? "It may or may not make sense to extend it to machines," says DeLaurier partner Jerry Harris, "but people are compelled to try."