Wired July 2005
Saving the Pentagon's Killer Chopper-Plane
22 years. $16 billion. 30 deaths. The V-22 Osprey has been an R&D nightmare. But now the dream of a tilt-rotor troop transport could finally come true
By Ron Berler
Eight RH-53D Sea Stallion transport helicopters lift off in twilight from the nuclear-powered aircraft carrier USS Nimitz in the Gulf of Oman, bound for a makeshift airstrip 600 miles away in the middle of Iran's Dasht-e-Kavir desert. Their clandestine mission, Operation Eagle Claw, is to rescue 53 Americans held hostage in the US embassy in Tehran.
At the airstrip, codenamed Desert One, six bumblebee-shaped C-130 Hercules transport planes wait to refuel the Sea Stallions. The helicopters are there to carry Delta Force commandos 270 miles to a staging area in the mountains outside Tehran, then raid the embassy the next night. The mission can't proceed any faster. The Vietnam-era Sea Stallions have limited range and cannot refuel in-air.
The helicopters enter Iranian airspace below 200 feet to avoid radar detection. Somewhere over the desert, they get trapped in a large haboob, a storm of dust as fine as talc. Visibility drops to near zero. One Sea Stallion drops out after a warning light flashes. A second reports gyro failure and turns back. A third loses its hydraulic pump. Only five fully functioning helicopters reach the airstrip.
Concerned that the mission is too hobbled to succeed, President Carter orders the team to abort. As the aircraft prepare to evacuate, one Sea Stallion shifts position on the airstrip to allow a C-130 to take off. The pilot lifts off, banks left, and loses his bearings in a welter of dust and downwash. He banks back to the right and collides with the C-130, his rotors slicing into the transport plane's fuselage. Both aircraft burst into flames; eight servicemen die.
The fiasco at Desert One in 1980 highlighted the Pentagon's need to replace its antiquated fleet of transport helicopters. The Sea Stallion and its 1960s-era cousin, the CH-46 Sea Knight, were too slow and, in their old age, had become maintenance nightmares and safety hazards. Incoming Navy secretary John Lehman, a pilot during the Vietnam War, thought he had an answer. When he was working for Henry Kissinger on the National Security Council in the 1970s, he saw photographs of a curious-looking experimental aircraft called the XV-15. It was a tilt-rotor hybrid - equal parts helicopter and plane, able to take off vertically and hover like a helo and then swivel its tilt-rotor pods (called nacelles) forward to fly like a traditional fixed wing.
After President Reagan put Lehman in charge of the Navy in 1981, Lehman traveled to the Paris Air Show to see the XV-15 in action. It wasn't as cool as, say, the British short-takeoff/vertical-landing Harrier jump jet. In fact, it didn't look remotely aerodynamic - more like a moving van stuck between two 38-foot-wide windmills. But Lehman was smitten. "It was very easy to fly," he says, "far more stable than a traditional helicopter, and simpler and safer than a Harrier. I was convinced it was what we needed." Lehman pushed the plane through the Navy's acquisition process.
In 1983, the Navy awarded Bell Helicopters and Boeing Aircraft a $68.7 million joint contract to design an aircraft based on the XV-15. This was the V-22, nicknamed the Osprey. It would carry two dozen geared-up marines or 10,000 pounds of weaponry, fly 2,100 nautical miles at 25,000 feet with just a single, in-air refueling, and land anywhere, no runway required.
At least, that was the theory. It's been 22 years, and the skies aren't exactly crowded with Ospreys. After more than two decades and $16.4 billion, the history of the V-22 is a sorry tale of cost overruns, shoddy construction, and managerial incompetence. Thirty people have died in four Osprey crashes, making the V-22 one of the killingest experimental planes ever. The program has teetered on the brink of elimination since almost the beginning.
But it never went away, propped up by genuine need, pork barrel politics, and the hope that the money already spent wasn't money wasted. Now the weird hybrid plane has entered a critical test phase called operational evaluation - the last hurdle before full production. The Osprey made it to op-eval once before, five years ago, and failed spectacularly. After an intense few years of engineering and test flights, years of tearing the plane apart and putting it back together under a fix-it-or-kill-it threat from the Pentagon, the Osprey is back. At military bases across the country, from New River Marine Corps Air Base in North Carolina to Edwards Air Force Base in California, pilots and engineers are testing the plane under combat conditions: extreme heat and cold, desert sand, high-altitude flying, aircraft carrier takeoffs. If all goes well, the evaluation will end in July and construction of the fleet will begin in 2006, and the first Osprey squadron will fly in fall 2007.
The Pentagon is confident it has a winner. The engineers and pilots believe they have solved the problems, both technological and organizational, that made the Osprey seem like little more than a deadly boondoggle. The Marine Corps has already ordered 360 Ospreys. The Air Force Special Forces is in for 50, and the Navy for 48. Sticker price: roughly $73 million apiece (GlobalSecurity.org, a defense consulting group, estimates the figure is really more like $105 million). "The Sea Knight can fly marines 50 miles from ship to beach, where the enemy is generally waiting," says Marine lieutenant colonel Kevin Gross, a manager on the Osprey team and the program's former flight test director. "With the Osprey, we'll be able to carry them past the beach, around the threat, around the weather, across any terrain, to where the enemy is weakest, where we can dictate the battle."
The V-22 barely survived the 1990s. The cold war was over, terrorism was a distant threat, and military spending was under scrutiny. When the Osprey budget ballooned from a projected $2.5 billion in 1986 to $30 billion in 1988 without a single test flight, defense secretary Dick Cheney tried to zero out the funding. Congress, not the Pentagon, kept the budget at the minimum level. The plane finally flew in 1989, and two years later it had its first crash. On June 11, 1991, an Osprey prototype hovering in helicopter mode - about 15 feet off the ground - wobbled. The left nacelle hit the runway and the plane dropped, bounced a few hundred feet, and burst into flames. The two pilots aboard sustained minor injuries in the accident, which investigators traced to incorrect wiring in a flight control system.
A second crash a year later generated more attention. At Quantico Marine Air Base, in a special ceremony designed to win congressional support, an Osprey was supposed to roar over the airfield like a military jet, then dramatically drop to the tarmac in helicopter mode. But as members of Congress and government officials watched, the aircraft's right engine caught fire. The plane plunged 500 feet into the Potomac River, killing all seven marines aboard. The program was grounded for 11 months. The Navy's investigation blamed a leak of gearbox fluid from the right nacelle into the engine.
For almost a decade, the program limped along, overbudget and behind schedule. Then, in April 2000, a third aircraft crashed, killing 19 marines - the worst military flight accident since a Marine jet sliced through a ski gondola cable in the Italian Alps in 1998, killing 20. The plane, which had taken off from Yuma Marine Corps Air Station, was bound for Marana Northwest Regional Airport near Tucson, Arizona. Flying in tandem with a second Osprey, its mission was to load passengers in Yuma - a simulation of a rescue from, say, an overseas embassy - and fly them to safety.
The lead Osprey approached the Marana airport about 2,000 feet too high, but rather than circle to shed altitude, the pilot decided to land. He descended dangerously fast, hitting the runway hard. The second aircraft, with a crew of 4 plus 15 fully outfitted marines on board, followed the lead plane and came down even faster - more than 2,000 feet per minute going just under 45 miles per hour. At 245 feet above the ground, the Osprey lost lift in its right rotor, stalled, and rolled over before it could issue a mayday. It crashed and exploded, killing everyone aboard.
The accident was attributed to a little-understood flight phenomenon called vortex ring state, or VRS, in which a helicopter descending rapidly at low forward speed drops into its own turbulence. Its rotors lose their grip on the air, and the bird drops out of the sky. That news especially shook the Osprey community - it suggested that the plane might be fundamentally flawed. One squadron member reportedly turned in his wings.
The program took another hit on the ground. Odin Leberman, then a lieutenant colonel and the Osprey squadron commander at New River, ordered marines in his command to falsify Osprey maintenance records. He did it to make the plane appear more reliable than it was, to increase its chances of winning new funding. "We need to lie or manipulate the data, or however you wanna call it," he said in a meeting. A maintenance crew member was secretly running a tape. Leberman was later relieved of duty.
Then, on December 11, 2000, another Osprey went down. Its crew was practicing instrument approach landings at night, and the plane fell 1,600 feet into a boggy forest near the New River base; all four marines aboard died. One of the nacelles had a catastrophic leak in the hydraulic system, and to compensate, the pilot hit a systems-reset button. Nothing happened. He hit it again. And again, at least eight times. Later, investigators found a glitch in the aircraft's software. Each press of the button had, for some reason, caused the plane to decelerate, making an accident even more inevitable.
On December 12, 2000, the V-22 was grounded indefinitely.
The Osprey had reached a crisis. Several engineers transferred to other aircraft projects. The program manager left. The chief engineer was promoted off the project. A new regime at the Pentagon was demanding that the plane get fixed in two years or get canceled. The entire culture of Osprey-making had to be fixed, and a series of significant technological problems had to be solved - fast.
Around Christmas of 2000, Ken Baile was offered the assistant chief engineer position. He hesitated. "I talked to my wife, I talked to my dad. I asked myself: Is this plane dependent on some technology that can never work? Do I want to associate my career with this? I talked to a lot of the engineers who'd remained. I read the reports. I took the job because I was satisfied the Osprey's technology was sound."
Under the weight of a stack of accident reports and assessments from the Pentagon, other government agencies, and various independent commissions, Air Force colonel Craig Olson, the new program manager for the military, set out to change Osprey Country (the flyboys' nickname for the test center at Patuxent River Naval Air Base). Olson realized that engineers had been in too much of a hurry, pressured by Defense higher-ups and Congress. "Before 2001, we were schedule-driven," he says. "Meeting a funding deadline was more important than making sure we'd done all the testing we could."
Evidence of that permeated the program. A Navy investigation following the 1992 crash found that a warning light had flashed in the Osprey's cockpit shortly after takeoff from Florida's Elgin Air Force Base. Yet the pilot kept going. He skipped a stop in Charlotte, North Carolina, under pressure from command not to be late to the fundraising ceremony at Quantico. After the Marana crash, a GAO report revealed that in 1997 and 1998, with pressure mounting to get the Osprey on budget and on schedule, program officials eliminated 70 of 103 planned tests, including rapid descent while carrying a full load. At a Senate Armed Services Committee hearing after the New River crash, program managers admitted they'd known about the hydraulic leak problem for six months. They hadn't corrected it, they testified, because an important funding deadline was approaching and the plane couldn't afford to fall any further behind schedule.
That pattern of eliminating flight tests to stay on schedule worried no one more than the test pilots, who already felt their concerns were being ignored. "I remember one time our onboard primary mission computers were failing, and we'd have a black cockpit for 8 or 12 seconds before the software recognized the problem and the backup kicked in," says Gross, the former flight test director. "That's too long to go with no cockpit displays, not knowing our air state. When we wrote it up in our deficiency reports, the program office got extremely angry and frustrated with us. They didn't want us slowing down the program because of unfavorable reports."
But the increased scrutiny from Congress and the media after the last crash scared Pax River straight. "I think another crash would shut us down," Olson says.
"People now actually read my flight test reports," says Steve Grohsmeyer, an Osprey test pilot. "Shortcuts don't happen anymore."
As a result of the culture changes - and with the Pentagon's 2003 deadline looming - the Osprey engineers dived back into their technology. Armed with the conclusions from a decade of reports and analyses, they set out to correct the failures behind each of the four crashes.
The first one, in 1991, was relatively easy. A flight control system had been miswired, and the engineers and mechanics fixed it soon after the accident. Crash number two, a year later, was tragic - seven marines killed in front of an audience - but the pilot knew he had an engine problem. Under pressure to get to his destination, he'd simply ignored the warning light. A more safety-minded, less time-crunched culture would make sure that kind of thing didn't happen anymore. More important, the nacelle was redesigned to prevent fluids from pooling - eliminating a potential fire hazard.
The third and fourth accidents, though, were trickier. Even two years after the third Osprey went down, pilots and designers worried about the mysterious aerodynamic problem of vortex ring state. The problem was that nobody knew much about VRS. When airplane wings or helicopter rotor blades cut through the air, they create a region of low pressure above them and high pressure below. That differential creates lift, but maintaining it depends on the smooth flow of air over both surfaces. Spinning helicopter blades turn the air beneath that high-pressure zone into chop - drop into that turbulence and the air stops sticking to the blades. The prop stops pushing, and the bird stops flying.
Lead test pilot Tom MacDonald of Boeing was assigned the VRS problem. "It was this mystery area," he says. "So little research had been done on it. People wondered: Would it swallow planes alive?"
MacDonald and the engineers worked out a system. He'd take the plane to 10,000 feet, putting enough air between him and the ground so he'd be able to recover if he got into trouble. Then he'd pull the nacelles back until they were almost vertical, in helicopter conformation, slow his forward airspeed, and try to induce VRS.
"We'd fly all day long," says Gross, copilot on a few of the test runs. "We'd fall 2,000 or 3,000 feet and recover. We'd fly back up to 10,000 feet, repeat the exercise at 1,000 feet per minute, then 1,500, then 2,000, all the way up to 5,000 feet per minute. Then we'd do it again, this time changing our airspeed." (A typical rate of descent for a 747 passenger jet on runway approach is 700 to 800 feet per minute.) In the process MacDonald, a former Marine pilot, quadrupled the published knowledge base on VRS.
What he found was that vortex ring state is surprisingly hard to induce. He had to fly slower than 40 knots while keeping the plane in a steady position for at least five seconds, and then descend at a hot 2,200 feet per minute. He also found that in an Osprey, he could recover from the condition relatively easily, provided he had 2,000 feet of altitude to play with. In the end, the team didn't alter the aircraft. Solution: Install a simple warning system. When a pilot pushes an Osprey toward VRS, a light flashes in the cockpit and a voice cautions, "Sink rate." And Osprey pilots now know to pay attention to those warnings.
The hydraulic leaks that had plagued the Osprey - and caused the last crash - required an entirely different kind of problem-solving. Engineers on the ground would have to rip the engines apart and start over. The investigation into the 2001 accident showed that a tangle of tubes in a nacelle had chafed against a main hydraulics line. Chafing had been a problem for years; the titanium hydraulic tubes are ultralightweight but brittle and relatively fragile.
The solution seemed obvious: Rejigger the plane's hundreds of feet of hydraulic lines so none of them touch. But that meant remodeling the guts of the nacelles, finding new space for fuel and electrical lines as well. "The technology wasn't in question," says Don Courson, the lead hydraulic engineer. "It was more of a design issue."
So Courson's team stripped a nacelle down to its frame and panels. Then, system by system, they started replacing parts - the prop rotor gearbox, the tilt axis mechanism, the engine. Finally, they redesigned and reinstalled the hydraulics lines.
Line clearance isn't a common mechanical problem. Pop the hood of a car and you'll find bundles of wires everywhere. The Osprey is unique because the nacelles shift and rotate, pulling lines taut and causing them to rub against adjacent surfaces. To be safe, the engineers decided that all hydraulic lines would have a half-inch of clearance. That's not easy to design or build. "Imagine a million-piece jigsaw puzzle," Courson says.
To make sure they'd solved that puzzle, they put the test plane back together and flew it for five hours. Then they took it apart again and checked the lines. No problem. They reassembled it and flew it for 10 hours. Took it apart. Checked the lines. Once they got up to 35 flight hours with no chafing and no leaks, they knew they had it right.
But the hydraulic leak wasn't the only problem in the fourth crash. Control software had warned the pilot of a problem, but it hadn't told him what to do about it - in fact, the pilot's attempts to get the aircraft under control had made things worse. "That was a little unnerving," Grohsmeyer says. "The last thing a pilot wants is to feel he's lost control of his plane." There were thousands of lines of code, and weak spots could surface anywhere.
Baile's team created what they called a Triple Lab. It consisted of a flight simulator, every bit of software installed in an Osprey, and a full hydraulics system to power it all. They spent a year in the Triple Lab, with a test pilot, letting their minds run wild. "We imagined every conceivable systems failure," Baile says.
The first glitch they tried to induce was the one that caused crash number four. The accident report said that a warning light had confused the pilot about the nature of the problem. And the flight manual hadn't explained what the light meant either. In the end, the engineers rewrote the software that set off the warning, and they fixed the bug that caused the Osprey's catastrophic deceleration before the crash. They changed a majority of the existing code and improved cockpit displays. "We wanted to remove all ambiguity, to make sure the pilot would know what and where the trouble was," Baile says.
By mid-2003, the program was back on track. "Nobody's questioning the Osprey's technology anymore," Olson says. That's not entirely true - some critics still think the plane simply can't fly safely. But they've grown quiet as the Osprey comes to the end of op-eval, almost five years since its last crash.
On an April morning 25 years after the disastrous mission in Iran, a V-22 taxis to a grass strip between two runways at Pax River. It's one of the base's seven Ospreys, gearing up for a test run.
Covered in primer paint, its Marine Corps insignia barely visible, the craft looks characteristically awkward. From 75 yards away, I see an airman give a hand signal. The plane's twin Rolls-Royce engines rev, generating a backwash so powerful that my sunglasses fly off. The two dozen flight engineers in the blockhouse-like building behind me are spared the windstorm; they're all staring at computers, reading telemetry.
The Osprey's rotors, pointing upward, dig into the air with an earsplitting thrum. The 16.5-ton aircraft rises to 15 feet. It floats for half a minute, then tilts its rotors and slides 300 feet to the right. Then to the left. The three-man crew is checking tolerances at low hover. "See how smooth it flies?" says Scott Trail, shouting over the prop noise. Trail, a major in the Marines, is one of a dozen test pilots on the Osprey, and, like MacDonald, he's a big fan. Three years ago he flew troops into Afghanistan in Sea Knights, the Vietnam-era transport helicopter. Twice as old as most of the guys on board, the Sea Knights were so creaky that Trail could carry only eight passengers at a time (even though the chopper has seats for 24).
"Watch this," he says. The Osprey starts to move forward, gathering speed like a dragster tearing down the track. The nacelles start to rotate forward - 10 degrees, 30, 70, 90 - until the helicopter has morphed into an airplane and is roaring at 290 miles per hour. In 15 seconds the conversion is complete. The plane shoots skyward and disappears over the horizon.
That was one of many tests the Osprey will have to pass. Right now, at air bases around the US, V-22s are performing similar - and more complicated - maneuvers, all part of the op-eval cycle.
Even if everything works, the plane faces one last hurdle: Congress can still say no. Depending on the economy, the military's needs, political pressures, and concerns about deficit spending, Congress may balk at the program's final cost. "There's a limit to our resources. You have to decide on priorities," explains Representative Mark Udall (D-Colorado), a member of the House Armed Services Committee, who says he plans to vote to continue funding the planes.
Some of the biggest believers are those who started the program. "The fact is, we have an airplane that will save a lot of lives and give us a lot of capabilities we don't currently have," says Lehman, the former Navy secretary who funded the project back in 1983. "When I look at the aircraft, I think of how useful it would have been in Afghanistan and Iraq. That's why the military won't give it up."
Mike Lieberman, a military affairs aide on the House Armed Services Committee, has a more pragmatic view: "My God, we've thrown so much money at it, we have to get something out of it."
© Copyright 2005, The Condé Nast Publications Inc.