The man with the fan

"It was definitely a paradigm shift," says Design News 2004 Engineer of the Year Paul Bevilaqua, recalling when DARPA awarded contracts to develop a common, affordable, strike fighter to Lockheed Martin, McDonnell Douglas, and Boeing in the early-1990s. "Typically, the military issues some pretty tight performance specs. "But under the first-ever application of a new, cost-based design initiative, they said, 'Build us an F-16 that can land vertically, is stealthy, and costs $24 million,'" says Bevilaqua. "We engineers, who typically complain that, 'These damn requirements are impossible,' were now saying ' How can we design an airplane without any damn requirements?"Ironically, though, it was the lack of specs that forced these engineers to really understand at a fundamental level the trade-offs among stealth, maneuver, payload, and range. That knowledge helped them come up with the only propulsion system that achieves both supersonic flight and vertical lift.

Paradigm shift just might be the perfect way to characterize how a team of engineers at Lockheed Martin Skunk Works led by Bevilaqua developed the innovative shaft-driven, lift-fan propulsion system. This patented technology—which for the first time allows an airplane to fly at supersonic speeds and land vertically—played a key role in Lockheed Martin's much-publicized, $200B contract win in October 2001 to build the new F-35 Joint Strike Fighter. The first flight test of a prototype airplane is targeted for 2005. The first operational flight is scheduled for 2008, with an estimated 5,000 aircraft to be built over the life of the program.

Vertical Lift a Holy Grail

Up until now, the concept of a supersonic, short takeoff/vertical landing (STOVL) aircraft had been a kind of Holy Grail for aerospace engineers. The challenge always came down to how to produce the additional thrust necessary to achieve vertical lift in a plane without incurring a huge penalty elsewhere. Since the 1950s, design teams have experimented with various, often elaborate schemes to create vertical lift—ranging from afterburning of the exhaust (probably one of most dangerous things you can do, given the high temperatures and velocities of the lift jets) to various ways of using engines to drive a big lift fan (severely limiting top speed or requiring a heavy rotor with a complicated folding mechanism).

It's the classic engineering trade-off story: Short takeoffs and vertical landings require a higher thrust-to-weight-ratio (>1) than is available in a typical fighter jet (;.75). But increasing the engine size to get the required vertical thrust is a no-win situation: The larger the engine, the higher the weight and drag—which limit speed. Likewise, the large-diameter rotors that give helicopters the thrust required for vertical lift in hover increase drag and reduce speed. "Unfortunately, the vast majority of STOVL designs are compromises, at best," says Bevilaqua, pointing to the AV-8B Harrier as an example.

The only successful STOVL airplane in operation, the1960s-era, sausage-like Harrier, uses a large, turbofan engine for both lift and cruise propulsion. One of the reasons that the design is successful is because it employs continuous thrust vectoring, which involves using nozzles to divert the exhaust flow either horizontally for cruise or vertically downward for takeoff and landing. This scheme is more efficient than stopping and starting a rotor, like the x-wing concept. But a large-diameter fan means a large-diameter aircraft. "When you're thick and tubby like the Harrier, you're not going to get to Mach 1," says Bevilaqua. "Supersonic planes need to be long and slender, like a missile."

Three Decades Studying STOVL

Bevilaqua knows all about the trade-offs, because he himself has been focusing on STOVL for his entire career. While working on an MS in Engineering Science and a PhD in Aeronautics and Astronautics at Purdue, his research focused on turbulent wakes—perfect training for someone who would go on to work on turbulent lift jets.

In 1971, Bevilaqua left campus and became an Air Force Lieutenant. He was assigned to Wright Patterson AFB, where he eventually became Deputy Director of the Energy Conversion Lab headed up by Hans Von Ohain, the inventor of the German jet engine. Fortuitously, Von Ohain happened to be studying STOVL aircraft. "It was Hans who showed me what those T-S diagrams actually meant," says Bevilaqua.

Bevilaqua received his PhD from Purdue in 1973, and left the Air Force in 1975 to become Manager of Advanced Programs at Rockwell International's Navy Aircraft Plant. In 1980, he signed on to do single-engine STOVL studies for NASA that would ultimately lead to the Joint Strike Fighter. He left Rockwell in 1985 and joined Lockheed as Chief Aeronautical Scientist to help set up a research lab in Los Angeles. "My charter was to invent something that would get a new line of business started for the company," recalls Bevilaqua.

He identified two possibilities: A hypersonic airplane or a supersonic STOVL airplane. NASA was still interested in the latter, sponsoring additional study contracts at four companies, including Lockheed Martin. "They evaluated four different concepts, including various thrust augmentation schemes and a Lockheed study of the Rolls Royce tandem-fan engine. "None looked clearly superior," says Bevilaqua, but the tandem-fan concept, which involved separating the engine fan into fore and aft fans connected by a common drive shaft, influenced his later design work on the F-35.

Though one of the limiting factors was the availability of a turbofan engine with enough thrust, DARPA provided Bevilaqua with funding to develop something new that would work. The engineers struggled to define the design space, having only the most general specs to work with."Typically what the government did in the past was give us the performance specs, and we'd evaluate our alternative design concepts and see how affordably they met the requirements," says Bavilaqua. "We would choose the lightest, most affordable concept as our best design."

Now, engineers were confronted by the challenge of transforming one requirement—a weight not to exceed 24,000 lbs—and general performance goals into a successful new aircraft design. To do this, they performed a parametric analysis on all of the key design variables—from thrust-to-weight ratio to wing loading, generating hundreds of curves across whole ranges of performance characteristics.

Since the engineers felt that it would ultimately come down to what engine size was required, they plotted the curves that they felt represented minimum performance requirements for transonic acceleration, instantaneous turn, and sustained turn all on the same graph—in essence defining the design space for STOVL operation. What the analysis revealed was disheartening: The space defined by the intersection points of the curves indicated that the size of the engine that would be required for hover using direct lift techniques would probably be too heavy to actually hover or too large to fly at supersonic speeds.

"After I concluded that direct lift was too risky, I invited some of the best Skunk Works engineers to come in and brainstorm ways that we could increase the thrust in a jet engine," says Bevilaqua.

The team came up with plenty of innovative ideas, ranging from using a laser beam to set off pulsed explosions to charging the exhaust with ions and using a magnet to generate an electromotive force. They even tried some unusual brainstorming techniques, like asking: "If someone famous invented a STOVL aircraft, what would it be like?"

"With one month to go on the DARPA contract, all I had to report was that Hercules would have invented the Harrier and Houdini would have invented the Stealth. Then one day I was struggling with the conclusion that there seems to be no better way to get power out of a jet engine than with a turbine. And no better way to get power forward than with a shaft. And no better way to generate thrust than with a fan," says Bevilaqua. "A lift fan and cruise engine would give me two lift jets, which could provide pitch control. So I decided to bleed off the engine bypass air for roll control. Then it just hit me that bleeding off the bypass air would increase the engine's effective nozzle, which would increase the power produced by the turbine engine, which I could then use to drive the lift fan."

So was born the concept for the innovative, dual-cycle, lift-fan propulsion system patented in 1993 by Bevilaqua and fellow Skunk Works engineer Paul Shumpert, who worked with Bevilaqua to prove out the concept. It would take eight more years to demonstrate it in flight.

Dual-Cycle Designs Delivers the Thrust

The whole trick in STOVL design is to figure how to get additional thrust from the propulsion system, without incurring a huge penalty in weight, efficiency, or aircraft range. Lockheed Martin's lift-fan propulsion system hinges on its ability to harness power from both the front and the back of a conventional turbofan engine, thereby providing thrust for both lift and cruise.

The system consists of a turbofan engine, with a compressor, combustor, two turbines (low pressure for driving the fan and high pressure for driving the compressor), and exhaust nozzle. The 50-inch-diameter, vertically mounted lift fan that provides cool air for the downward thrust is powered with a drive shaft connected to the front of the engine fan by a clutch. A unique feature of the engine is its ability to cycle between turbofan (cruise thrust but no shaft power) and turboshaft (shaft power and reduced thrust) modes by changing the nozzle exit area. The rapid transfer of thrust between the engine and fan provide pitch control, while two auxiliary nozzles in the wings provide roll control.

The additional power to drive the fan comes from the exhaust jet of the turbofan engine. Like a turbocharger on a high-performance car, which uses the energy in the exhaust gases to spin a turbine that pushes more air through the engine, Bevilaqua's system is able to extract more thrust from the engine by transferring the energy in the exhaust jet to a larger mass of air. This is done by adding another turbine stage to the engine.

With this extra boost in thrust, engineers were able to avoid having to oversize the engine, which in turn avoided a weight penalty. The design will also allow Lockheed Martin to easily produce variants of the design for the Navy and the Air Force, which do not need STOVL capabilities, simply by removing the fan.

A Risky Design Flies High

Although the engine, drive shaft, and clutch were all within the state-of-the-art, Bevilaqua says that Lockheed Martin's design was viewed as more risky than Boeing's approach, which was based on a direct-lift cruise engine with a single fan. So it was particularly rewarding for Lockheed engineers when during flight testing in 2000 and 2001, the JSF X-35 demonstrator proved to be the only design to meet all performance requirements. In one particularly breath-taking moment, British test pilot Simon Hargraves conducted the first-ever vertical takeoff and landing as Bevilaqua stood watching from the sidelines with his wife and daughter.

"It wasn't supposed to be the vertical takeoff and hover test. Simon was just going to check control response. But he gave the plane just a little bit of throttle, and it jumped up 30 feet straight in the air. It was so smooth, so perfect, and so under control," says Bevilaqua

As an engineer who has spent his entire career developing STOVL aircraft—he's now focusing on ways to apply the lift-fan system to transport aircraft—Bevilaqua's thinking is more of the Blue-sky variety, not surprisingly. But clearly his ideas are grounded by sound engineering, and a high degree of optimism. "Sure, our design was risky. But I always believed that fundamentally it was going to work."