Look carefully at this photograph of a 757 during a test landing in a strong crosswind in Great Falls, Mont. The pilot "crabbed" the airplane, turning its nose into the wind during all but the final moments of the landing. Although the plane appears as if it will land to the left of the photographer, it landed on the runway to the right.

That was the urgent communique from one top federal aviation official to another last Nov. 16 after a particularly dramatic 2-hour, 17-minute flight by a Boeing 757.

Leroy Keith, who heads the Federal Aviation Administration's certification program for jetliners, was at FAA headquarters in Washington, D.C., when the phone call came.

Darrell Pederson, a Keith lieutenant, was on the line from Seattle. He was supervising Boeing's efforts to prove the 757 airworthy, and his call attracted attention because it interrupted an important meeting.

Keith recalls Pederson's message. "He said, 'Well, ... we had ... we almost lost one.' He was quite frank about it."

During a certification flight a 757 had ingested ice in its huge Rolls Royce engines, setting off cockpit alerts and creating a roar that one person on board said sounded disturbingly like a car without a muffler.

Keith said Pederson sounded shaken on the telephone. "He said it had an icing encounter and went back to Boeing Field drifting down on minimum-power setting, and got back and found the fan blades were damaged on both engines."

Later there would be differences of opinion about just how serious the problem had been, and whether the airplane really had been in any peril.

The crew intentionally had been seeking ice build-up on the airplane's wings to prove its performance under such conditions and had lingered in circumstances which any commercial pilot would have avoided It was not a situation likely to be encountered by an airplane carrying passengers, because commercial pilots don't go looking for trouble.

The objective of flight testing is to uncover potential problems, even those that are extremely unlikely. Boeing and FAA officials agree that every new airplane design has unforeseen snags that need to be discovered and corrected, and it's not fair to judge an aircraft until this process is complete.

Despite these caveats, no one took the icing incident lightly.

"It sounded really serious," Keith said. "And in fact, it was."

A Boeing flight engineer with significant responsibility for the program described the Nov. 16 flight this way: "They were descending into terrain and didn't have to power to climb. It's very dangerous. You can lose a plane that way. We don't fly with parachutes because we expect to land in the middle of a runway every time."

The engine problem, coming so late in the 757's development, posed logistical difficulties for Boeing, Rolls Royce and the 757's first customer, Eastern Airlines. The airline wanted its first plane within a month, but the FAA wasn't about to certify it until there was conclusive proof that the problem had been solved.

For their part, neither Boeing nor Rolls wanted an airplane operating with any safety question lingering.

John Winch, who directs Boeing's flight-test and certification programs, said the company went to unprecedented lengths to ensure both the 757 and 767 were thoroughly tested.

The FAA's certification procedures for the Boeing 757 and 767 are said to be the most comprehensive in history. After suffering the sting of criticism over difficulties experienced by the McDonnell Douglas DC-10 long after it was certified for flight, federal officials intensified their efforts to be certain the 757 and 767 certification programs were beyond reproach.

"I think the real reason we're being tougher is because these are a lot more complex airplanes than we had eight years ago, or 10 years ago with the 747," Keith said.

The 757 and 767 have two-man crews and equipment for low-visibility landings, both of which required additional certification efforts, said Brian Wygle, Boeing vice president of flight operations.

The 757 carries more than 100 computers, and federal inspectors sought proof that both the airplane's hardware and the intricacies of the computer software were fail-safe.

Certification is a painstaking process - and an expensive one. "You spend $1.5 million to $2 million per airplane just to put the (certification instrumentation) parts in," a Boeing engineer said. It has been estimated that flying jetliners that are equipped for flight testing and certification costs more than $50,000 an hour.

The task of certifying two airplanes simultaneously added to the challenge. At the peak of the certification effort, 17 aircraft, (including models other than 757s and 767s) were involved in flight testing, said James Lincoln, manager of the data section of Boeing flight-test engineering.

The whole program, another official said, "cost hundreds of millions of dollars."

Flight testing has a splashy reputation, a lingering image of the do-or-die pilot tempting fate to prove his machine. But a Boeing flight-test engineer said "we don't do much of that 'white knuckles and silk scarves' stuff anymore."

Still, in their more dramatic moments, flight tests aren't for the faint of heart.

Testing and certifying an airplane involves pushing it into what Phil Condit, general manager of the 757 program, calls "far corners" - performance situations one hopes an airplane will never have to encounter in actual service.

Far corners can be terrifying to the uninitiated.

A 757 cruises with maximum fuel efficiency at 80 percent of the speed of sound, or mach .8. Its maximum intended speed is mach .86. But Boeing pressed the airplane to mach .92 in flight testing. A far corner.

Such flights, said Rick Lentz, 757 flight-test aero-analysis lead engineer, "can be frightening, because the plane responds with buffeting. The tail assembly is groaning and wings are flapping - and until you've been through this a few times, you're not sure it will hold together."

The fear is personal, not corporate, Lentz added. Anxiety is normal for a person who hasn't been through it before, although Boeing is confident the airplane will perform as intended, he said.

The vibrations of some maneuvers, Lentz said, "will literally rattle your teeth."

Not every dramatic test takes place in the air. Boeing routinely destroys one airplane of each model to see what it takes, to see if it's really as tough as the engineers say it is.

The 757 test happened last July 16. Enormous pressures were applied to a 757 airframe inside a hangar. The wings were bent upwards ... first two feet ... then five ... eight ... 10 ... At 11 feet, 6 inches of deflection, both wings snapped.

"It's like loading a bridge," Condit said. "You just keep loading it until the thing finally goes 'kaboom!' That's exactly the sound it makes. I felt it in my knees. I don't know if that was the excitement or the boom"

The results were pleasing, because the airplane proved 12 percent stronger than engineering estimates, and because, in a tribute to Boeing engineering and quality control, both wings failed at the same place and at almost the same moment-just 14 thousandths of a second apart.

IMAGINE WHAT must go through the mind of a test pilot about to take off in a jetliner which never has flown before.

Wind-tunnel tests say the aircraft will fly. Engineers say the design will soar like a dream. Mechanics and inspectors have looked over the huge machine.

But will it really fly?

John Armstrong, chief test pilot for the Boeing 757 program, said he felt little anxiety about taking off for the first time every in a 757 on Feb. 19 of last year.

"Excitement" is what he remembers of the moments before hurtling into the air from Renton Municipal Airport.

Armstrong and Boeing called it a "perfect first flight" upon landing, although later they admitted that a design problem, later corrected, prompted them to temporarily shut down one of the757's two engines during the flight.

Armstrong, piloting the "No. 1" 757, had general good fortune throughout the 11-month flight-test program-a program which involved five different airplanes, each conducting separate testes specified years in advance.

Like the rest of the 77 project, the requirements of a 1,254-hour flight-test program were scheduled with precision back in late 1978 and 1979, before Boeing irrevocably committed itself to the financial and other risks of a new-airplane program.

A chart created at the end of 1978 shows Armstrong's 757 was to fly 375 hours of tests, between February and December 1982 when certification between February and December 1982 when certification was to be complete. The timetable was met.

The inside of Armstrong's 757 would be almost unrecognizable to a frequent flyer on regular commercial-airline flights.

There are few seats. Water barrels, like metal beer kegs in appearance, are positioned where seats might otherwise be. By filling the barrels in differing configurations, Boeing engineers can simulate the effects of different passenger loads.

There are racks of computerized equipment attached to thousands of wires which snake through the airplane. The wires carry signals from sensors all over the plane, and computers record the data and help make sense out it.

Four thousand channels of information were stored simultaneously by the on-board flight-test computers on the 757. A decade ago, the 747 flight-test computer equipment filled a jumbo jet from stem to stern and succeeded in recording only 800 channels, while th e757 accomplished a much larger task with room to spare inside a shorter, much narrower airplane. That's a result of a decade of computer evolution.

Several engineers and technicians are on board during test flights, monitoring the gear and the airplane's performance. The atmosphere is business-like, competent, yet markedly relaxed. A flight attendant from a commercial airline might wince at the freedom of movement inside a test-flight airplane, even during takeoffs or landings.

WHEN LOCAL TELEVISION viewers see news stories on the 757 or 767, they are sometimes treated to rather unusual footage of a Boeing airplane touching its tail to the runway during takeoff. It's an interesting sight, and not always explained by the newscaster, who may be reporting Boeing sales figures or some other issue unrelated to flight tests.

The tail dragging is known as "Vmu" testing, and it was one of the missions of 757 No.1. It took place last June in southern California, at Edwards Air Force Base, the same place the Space Shuttles have landed.

Twenty-eight times the 757 dragged its tail down the Edwards runway. Each e as a test of the characteristics of the 757 at minimum-speed takeoffs (Vmu stands for Velocity-minimum unstick, with "unstick" signifying the wheels departing the runway).

The object was to determine the lowest speed of safe takeoff under various conditions so that a schedule could be established to guide pilots in selecting appropriate take-off speeds.

Avoiding damage to the airplane during the tail scraping involved careful work by the pilots and the temporary addition of an oak skid to the bottom side of the rear of the airplane.

Pilots lifted the nose of the airplane off the ground rapidly, lowering the tail in what is called "rotation." When they sensed the tail was about to touch the ground, they would slow the rate of rotation.

"It's a hard test to fly, to be precise about it," said Les Berven, the FAA pilot who flew the Vmu tests with Armstrong. "You have to rotate just fast enough to make sure you've gotten the tail on the ground before life-off, but not so fast that you hit it."

The judgment, Berven said, is "set of the pants."

Though Armstrong's 757 was put through its paces relatively uneventfully, the opposite was true of 757 No. 3, flown by a fellow test pilot, Kenny Higgins.

Problems began the first time the airplane was flown: the landing gear would not retract fully.

The flight continued, with the wheels hanging out of the airplane at a strange angle, but it was cut short at 39 minutes. The basic airworthiness of the airplane was established, however.

Five days later, a more difficult problem developed.

Higgins was bringing in No.3 for a landing at Boeing field after a flight of more than three hours when he had trouble with the airplane's flap system. A transmission part froze, forcing the airplane to land about 200 knots too fast.

To Boeing's total surprise, all four tires on the right-hand side caught fire when the brakes locked up. The wheels were ground flat. On the left-hand side, two tires and wheels were flattened.

Fire trucks moved in an put out the flames. Boeing workers jacked the airplane up and changed wheels and tires, then taxied the airplane off the runway and into a hangar.

Flight-test problems are instructive, and Boeing set out to learn from the bad landing. In this instance, finding the cause of the difficulties was challenging - "a real witch hunt," in the words of Jim Johnson, director of engineering on the 757 project.

The snag was in computer programming.

In the end, it was proved that the flap problem was unrelated to the locking of the wheels, which was a failure of the plane's anti-skid system. A freakish set of electrical impulses confused four or five computers controlling the anti-skid system, Condit said.

It took several weeks before the bugs were all workout, although temporary repairs were made within two days.

"It took us about a day and a half to isolate the problem," Johnson said. "Very quickly, through what we call cuts and jumpers, where we go into a circuit board and make a cut and put in a new wire ... we were flying."

The same fixes were made in all the 757s - and in all the 767s, where the same potential problem existed.

But Higgin's troubles with No.3 weren't over. He had yet to fly the Nov. 16 "ice flight."

Ice on a jetliner can be deadly, as was demonstrated by the crash of an Air Florida 737 into the Potomac River near Washington, D.C.'s, National Airport on Jan. 13, 1982. The problem is most threatening when the ice is irregularly shaped, because it can dramatically change the shape of the airplane's airfoil, destroying much of its lift.

For most of its 2 hours and 17 minutes, the Nov. 6 flight was uneventful. The pilots were intentionally building up two inches of ice on the 757's wings, then shedding it with anti-ice systems.

But the unexpected struck rudely toward the end of the flight when chunks of ice broke off the center hubs of the two Rolls Royce engines and damaged the fan blades.

Rather than heating the engine's "spinner core" - the hub in the center of the outer fan blades-Rolls had elected to use a flexible tip it believed would flex to keep ice from building up.

The Nov. 16 flight proved dramatically that the flexible tip wouldn't always work.

Immediately upon ingestion of the ice, both engines began to rumble and cockpit instruments showed high levels of engine vibrations. The vibration could be felt throughout the airplane, including the cockpit.

Higgins and Dick Paul, the FAA pilot on board, cut back the engine power, alternately idling one engine, then the other. They aborted the tests, retracted the flaps and raised the landing gear.

The left-side engine was vibrating particularly badly on the return to Boeing Field, and the pilots agreed to land with the engine idling rather than possibly push it too far by running it hard.

Finding ice hadn't been easy for Boeing. There had been a weeks-long search for the appropriate test conditions. Then, suddenly, there was an abundance of ice-and an unexpected problem.

Engineers in the back of the airplane, monitoring banks of instruments and watching the ice on the wings, were pleased with the amount of ice they finally had found.

But the cockpit, where the engine performance was alarming the pilots, there was no elation. Upon landing, Higgins told the FAA: "I was not happy."

ROLLS AND BOEING solved the problem by substituting a heated spinner cone for the flexible tip. It was a rush job, with round-the-clock shifts in the Rolls Royce's plant at Derby, England, producing the spinner on a few day's notice.

The airplane was certified and rushed into service by Eastern Airlines, which took delivery of its first two 757s at the end of December.

But the FAA made the certificate valid for only six months. The airplane was perfectly safe, the FAA said, but it wanted a seat in the cockpit repositioned so that FAA personnel who occasionally ride along on commercial flights could have a better view of pilot activities.

Boeing and the FAA dug in over the issue, and it looked for a time as if the matter would end up in court. But in late May , the FAA granted a permanent certificate after Boeing agreed to move the 757 seat just 7 inches.

Hundreds of Boeing flight-test employees gathered at Longacres on Jan. 22 to celebrate the certification of the 757. The FAA came in form more than a little ribbing, including a skit in which an outlandish chair with a chicken attached to it was displayed and proclaimed to be, by a supposed FAA representative, a "damn good seat."

But the flight-test crews gave themselves a bad time, too. The test pilots and their airplanes were roasted in good humor.

When it was Higgin's turn, a top Boeing engineer named Pete Morton explained the astronomical odds against all the problems Higgins had encountered with 757 No. 3. He gave Higgins a T-shirt with the slogan "Extremely Improbable."

Higgins, having weathered a stormy certification filled with unlikely events, replied: "Extremely improbably to me means that it happened yesterday."

Part  1 |  2 |  3 |  4 |  5 |  6 |  7 |  8