|Your account||Today's news index||Weather||Traffic||Movies||Restaurants||Today's events|
Designing the 757
June 21, 1983
It came to rest at an Eastern Airlines gate where top Boeing officials were waiting to give the airline's president, Frank Borman, his first ride in the airplane he had helped launch with a $900 million order four years earlier.
A gate agent stepped up to the 757's door, popped out a butterfly-shaped handle and turned it clock-wise. Grasping the door firmly, while the top brass of Boeing and Eastern looked on, she pushed and pulled.
The 323-pound door moved only a few inches. It wouldn't budge beyond that. Try as she might, she couldn't get it open.
"See?" said Paul Johnstone, then Eastern's senior vice president for operations.
Johnstone chuckled later and explained that he had purposely chosen a small woman to open the heavy door or try to so Boeing executives could see first-hand that something was wrong. "I had mousetrapped'em," he said.
Although the 757 passenger door met elaborate engineering criteria and reliability tests, in Eastern's view the airplane at Dorval International was a flawed product.
The airline was to take delivery of the first airplane in December, just three months away, and it wanted a door every gate agent, regardless of his or her weight or strength, could open.
The 757 door took about 70 pounds of strength to open, twice as much as a 727 door. There were several reasons, including that it weighed more because the 757 sits higher off the ground and the door must contain a longer escape slide for emergency evacuation.
Boeing hadn't ignored the question of how much strength was necessary to open the door, said Jim Johnson, 757 director of engineering. On the contrary, engineers had calculated everything from the door's weight to the viscosity of its oil to the effects of friction on the door's bearings and rollers.
D.P. Tingwall, chief project engineer for engineering computing, said door loads were examined by computer and the design of cam parts was modified on the basis of the computer's findings.
"We didn't give proper consideration to a small framed woman with light weight," Boeing's Johnson said. A small person didn't have enough leverage to move the door, he said.
Boeing first encountered the problem last July 8, when Nancy Ballard, a 115-pound Eastern gate agent at Seattle-Tacoma Airport, had enormous difficulty opening a 757 door in a test at Boeing field.
"This young girl damn near got herself a hernia trying to open the damn door," said Johnstone, now retired from Eastern. In fact, Ballard came away from the test with a muscle bruise the size of a baseball on one arm, where she had repeatedly leaned for leverage while trying to pus the door open.
"I was able to open the door, but believe me, it was a strain," she said.
It took a dozen engineers eight 56-hour weeks to solve the problem by designing a dual-spring mechanism, but the solution hadn't yet been installed when Johnstone sprung his mousetrap in Montreal.
THE DOOR PROBLEM wasn't the only thing discovered relatively late in the development of the 757 that required re-engineering.
In the fall of 1981, an anesthetized 4-pound chicken was loaded in a pneumatic gun and fired at 360 knots head-on into a stationary 757 cab.
The expectation was that the chicken would deflect off the cabin's sloping metal roof. Instead, it pierced the airplane's skin.
"It looked like you had thrown a shot put through it," said Ed Pottenger, a Boeing engineer.
This shocking result, and the realization that it might be repeated if the 757 hit a bird in flight, led to some urgent redesigning of the cabin roof of both the 757 and 767. The challenge was great because several 767s already were flying and had to be cut apart.
The changes were particularly painful because beefing up the cab added 70 pounds to the weight of each airplane, and saving weight is an aeronautical engineering objective pursued with almost religious intensity.
"There are people in the Boeing Co. who kill their grandmothers for five pounds. I'm dead serious," said Leroy Keith, the Federal Aviation Administration official who oversees certification of jetliners and other transport aircraft.
Overcoming those obstacles was only a small part of the engineering that went into the door and cabin of the 757 and the door and cabin represented but a fraction of the 65,000 "engineering events" involved in creating the airplane's detailed design.
At its peak, 1,500 engineers and a like number of assistants were involved in the 757 project. Although it was building two new airplanes simultaneously, Boeing didn't want to swell its engineering ranks temporarily, so it uprooted engineers from subcontractors around the country.
Often using computers (which helped design 47 percent of the airplane's parts), the engineers tackled large questions such as what shape the tail should be and how far back the wings should be swept, and such seemingly small questions as whether a piece of hardware should be hollow and how many inches wide a restroom should be.
Along the way, the engineers helped contribute to a dramatic change in the nature of the airplane: The 757 Boeing started to build is far different from the 757 it actually built, largely because of the improvements in the airplane's cockpit and electronics.
The 757 has 95,000 different part types and a total of 3 million separate parts. Engineers selected or designed each one. Other engineers decided when material would be needed to make the parts and when the parts themselves would be needed.
Phil Condit, who held the top two engineering posts on the 757 project before being named its vice president and general manager in January, said he didn't make important engineering choices so much as designate times when they must be made by rank-and-file engineers.
"I couldn't absorb enough data to possibly make these kinds of decisions," he said.
Although the engineering details of the 757 far exceed the grasp of any single human mind, the problems involved in the design of the passenger door and cockpit offer some insight into the complexities of conceiving and creating a new airplane.
A BOEING 757 passenger door doesn't attract attention, which is just fine with its creators.
"A door looks simple," Condit said. "That's the way you want it to look. You don't want the passenger worrying about whether it's going to work."
In fact, the simple-looking passenger door contains about 500 parts, held together by 5,900 rivets. Its mechanical systems were designed by a battalion of engineers and fashioned by custom-made tools that cost millions of dollars.
The curving 4-inch-thick door must contain highly reliable mechanisms, including a system to control the speed at which it rotates on elaborate hinges, and a system that enables the door to power itself open and deploy an escape slide in an emergency.
The slide is stored in the door, but whenever the door is closed and "armed" the slide automatically attaches to the sill of the doorway. In an emergency, the slide is pulled out of the door as the door opens. The slide inflates automatically.
The slide must reach the ground in an emergency even if the airplane is resting nose-up and tail-down, or is listing to one side with a broken landing gear. And the slide must inflate rapidly and reliably even in 25-knot winds.
The complexities of the 757 passenger door are all the more remarkable because of the utter simplicity of the door's basic concept. The door is, in essence, a plug not unlike a bathtub stopper or a bottle cork.
But unlike a cork in a bottle, which is wedged in from the outside, the 757 door is wedged from the inside. The pressurized cabin air helps hold the door in place.
With the first turn of the door handle, internal mechanisms unlatch the door and reduce its height and wedge shape by folding in "gates" at the top and bottom. The door swings into the cabin briefly, unplugging the doorway, then slides back through the opening at a 25-degree angle and swings wide to fold against the outside of the airplane.
But to think of the door only as a piece of hardware is to overlook what is perhaps its most telling characteristic: compromise. The door, like the whole airplane, is as much a collection of engineering trade-offs as it is a collection of parts.
"To the last detail, everything we do it a compromise," Johnson said. Boeing engineers refer to these compromises as "trades," and there are lots of them in a door.
For example, the door must be wide enough to provide passengers comfortable entry, yet not so wide it robs seating space. Its window must give adequate vision of what is outside, and yet not use up too much of the space needed for its mechanical innards.
The door must be strong enough to hold out an alien environment of sub-zero temperatures, low air pressures and speeds approaching the sound barrier. And yet, like the rest of the airplane, it must be as light as possible to maximize fuel efficiency.
"I can't overstress the complexity of this set of trades that are continually going on," Condit said. "How much off am I? What happens if I put a little more wing area on? How does that balance?"
Engineers have different concepts of what is important. "You get a hydraulics guy, and he thinks the airplane ought to have 80 million miles of hydraulic lines in it," said John Armstrong, chief test pilot of the 757 program. "And the propulsion guy thinks that the airplane's just a vehicle to carry his engines around."
"Weight is the all-important driving force behind almost everything," said Keith, the FAA official. "you can build something that is totally safe, fire-resistant and fail-safe, but it would be made out of titanium and it would be heavy and it would be prohibitively costly. So you've got a series of trades.
"It's just one series of compromises. In performance, handling qualities, systems, reliability, comfort, economy, structure, fabrication."
That all the trades are made and an airplane is created and tested in just four or five years is remarkable, Keith said. "You've got a magnificent flying machine out of the deal in a relatively short time period. It just never ceases to amaze me."
EARLY in 2978, a cockpit designer named Tom White made a pie-in-the-sky suggestion a suggestion that, if accepted, would fundamentally change the future of Boeing's new family of jetliners.
The 757 program was a year away from its eventual launch and the proposed configuration of the airplane kept changing. "Semi-fluid," one designer called it.
But Boeing was certain of one thing: although much of the 757 would be new, including its engines, wings and interior design, to save development costs it would use updated versions of the 727 cockpit, tail and body cross-section. In short, it would be a 727 derivative.
In a one-page memo dated March 23, 1978, White asked why the company shouldn't forget about making the 757 derivative. Instead, he proposed putting the all-new Boeing 767 wide-body nose and state-of-the-art digital electronics on the narrow-body 757. It would be a challenge, since there was almost four feet of difference in the diameters of the two airplanes, but he thought it was possible.
White was suggesting more than just a nose job for the 757. It would be a complete personality change. It would make the 757 a sister of 767, which was a 1980s airplane, rather than a half-sister to the 727, a 1960s airplane. But it also would cost Boeing a fortune in additional development costs and add substantial weight to the airplane.
White argued the updating would make the 757 attractive to airlines for years and might result in the FAA eventually approving a common pilot rating for both airplanes. This would cut operating costs for airlines flying bot the 757 and 767.
The idea of creating a common cockpit for the 757 and 767 wasn't new to White's boss, Del Fadden. It had come up every few months. But never before had a designer developed the idea to imaginatively or made it seem within the realm of possibility. Fadden encouraged White.
But the suggestion went nowhere. It just wasn't what Boeing had in mind.
Over the summer of 1978, the 767 program was launched with United Airlines as its first customer. The 757 program was announced in late August, although it wouldn't get the official go-ahead from Boeing until the following March.
During that summer it dawned on Boeing management that the 757 was looking less and less like a 727 derivative. There was talk about using a non-727 tail for improved aerodynamics, and the latest cockpit design called for an advanced safety-and-maintenance monitoring system that could reduce the flight crew from three to two. Without a formal policy decision having been made, the 757 was evolving away from the 727.
In October, Ken Holtby, a Boeing vice president who had run the 747 division for four years, took charge of coordinating development of the757 and 767. Management felt the two programs sometimes were plowing the same ground.
"One of the things that became immediately obvious was that many of the decisions that had been made on one program or another really should have been applied across the board," Holtby said. "We found a lot of differences between the airplanes that really couldn't be justified."
Holtby told the product-development organization to study ways to increase the commonality of the two airplanes. Much of the job fell to Doug Miller, a chief designer.
One of the first things Miller did was call Tom White. Together they went to Boeing's Everett plant to look at a mockup of the 767 cab, which used a new design Boeing had been developing for years. They set out to prove the cab could be used on the narrower 757.
Soon there were a lot of people working on the common-cockpit idea, and excitement grew. Pete Morton, senior project engineer on the 757, became a powerful advocate. H.G. Stoll, Morton's counterpart on the 767, remembers a phone call in which Morton spelled out the common cockpit idea.
"At first I thought it looked like it was way out as far as an idea," Stoll said. "After all, (if) somebody says you're going to take the front end off a Cadillac and put it on your Datsun, your first reaction is that it's not a good idea."
Morton and Miller approached Condit, who, as chief engineer, was impressed with the idea of common parts between the airplanes, but thought a common FAA pilot rating was "pretty elusive, a pretty high-risk think to be going after."
Eventually, the idea worked its way up the corporate ladder to Holtby, the vice president who had urged greater commonality between the airplanes.
"I had to be persuaded," Holtby recalls. "There are a lot of factors that go into that kind of a decision, including our capability. Quite a few of the guys were recommending against it because they didn't feel we had the resources to do it. So there was quite a bit of debate."
Boeing decided to take the gamble.
Today, both airplanes have identical cockpits, developed jointly by 757 and 767 engineers...Boeing perceives the evolution of the cockpit as a triumph...White still seeks new design ideas...Condit, who is only 41, has risen from engineering to become vice president and general manager of the 757 division.
And hanging on Condit's office wall, framed and signed, is one of White's original renderings of how he thought the 767 cab could be fitted to the 757.
INSIDE and out, the cab it the most expensive part of a jetliner's fuselage to engineer and manufacture.
The cab's exterior is of irregular shape. None of the metal is flat; even the contours are not uniform. Inside the cab, an impressive number of electronic and mechanical devices must be sandwiched into a small area.
Morton estimates 120 to 150 engineers were involved in creating the cab and cockpit on the 757 and 767, including those who designed the exterior, crafted the interior and figured out how to fit in all the instrumentation.
Cab design is based on the position of a pilot's eyes, hands and feet. The 757 and 767 cockpits are designed for people from 5-foot-2 to 6-foot-3. The shorter height was included in the design in the expectation that women pilots will become numerous during the useful life of the airplanes.
Boeing selected earth tones of brown and beige for the interiors because a NASA study found those colors reduce anxiety in high-stress environments. Sheepskin-covered seats were installed.
Boeing ran tests on how computer screens should offer information, seeking answers to such questions as whether pilots might be distracted if screen automatically changed displays. (Answer: yes. So Boeing designed systems so that nonessential screen would update displays only when requested to do so, or in emergencies.)
Boeing engineers also used computers in wind-tunnel studies to simulate the airplanes' handling characteristics long before they ever flew. Test pilots expressed their preferences, and engineers changed computer software and the airplanes' control surfaces until they found configurations that gave both airplanes desirable flying qualities.
The goal, which Boeing says it achieved, was to build two airplanes that seemed similar to pilots, even though they are quite different.
"Cockpits are the place in the airplane that have the most compromises that I know of," Morton said. "Everything comes together there.
"For example, if I want a good view of a panel, I don't want a big control column there," he said. "But the guy that's responsible for the control column would like a good meaty one that a guy can wrap his hands around and really pull. And he doesn't are if it blocks my instruments.
"And then I want a nice compact cockpit. I don't care if there's a quarter of an inch behind a panel. But the guy who has to go buy the equipment wants 18 inches behind there. And there's another guy with the responsibility to cool it. He puts ducts in the back of that thing..."
The 757's front window, or "No.1 windshield" as it is called, demonstrates the compromises that can be involved in a single jetliner component.
A big window gives a pilot good outside vision, but leaves less room for instrumentation. Glass contributes nothing to the strength of the cabin, so big windows mean big posts, which block vision. Structural engineers would just as soon have portholes as picture windows.
Glass must be strong enough to withstand the impact of a large bird at high speed. But thick glass doesn't transmit light as well as thing glass.
Glass is a poor insulator, so large windows can chill a cab at night and let in excess sunlight by day. Big windows can mean glare problems, too.
The aerodynamics of the cabin are crucial, both for fuel efficiency and to keep noise levels low. Curved windshields help aerodynamically, but can create optical distortions, including double-light reflections.
The No.1 windows, which are made in England, are the same for both the 757 and 767. There are two of them, one on the right side and one on the left. They are flat. The side windows, known as Nos.2. and 3, are curved and differ between the two airplanes because their fuselages have different shapes.
The narrower 757 has smaller No. 2 and 3 windows and they are closer to the pilots' shoulders. But Boeing shaped and positioned them so that they give essentially the same field of view as the larger and more distant side windows on the 767.
The reason? Again, to make the airplanes feel alike to pilots.
In some places, engineers could not resolve differences between the two airplanes, but the variances aren't major. For example, the No. 2 windshield opens in both planes to allow an escape route for pilots, but the mechanisms differ due to fuselage shapes.
AFTER it stops in Montreal last September, the 757 flew on to England with a load of Eastern and Boeing officials.
On the way, a duck hit one of the cockpit's No. 2 windows, not an unusual incident.
"It's usually not a big deal," said Les Berven, and FAA pilot who was co-piloting the flight. "All it did was just to make him into jelly and he slid down the side of the window."
The window didn't break but then Boeing knew it wouldn't because the window had gone through a series of "chicken tests."
Boeing is a little touchy about the subject of chicken tests, and points out they are required by the FAA. Here's what happens:
A live 4-pound chicken is anesthetized and placed in a flimsy plastic bag to reduce aerodynamic drag. The bagged bird is put in a compressed-air gun.
The bird is fired at the jetliner window at 360 knots and the window must withstand the impact. It is said to be a very messy test.
The inch-thick glass, which includes two layers of plastic, needn't come out unscathed. But it must not puncture. The test is repeated under various circumstances the window is cooled by liquid nitrogen, or the chicken is fired into the center of the window or at its edge.
"We give Boeing an option," Berven joked. "They can either use a 4-pound chicken at 200 miles an hour or a 200-pound chicken at 4 miles an hour."
The British government requires that the metal above the windows also must pass the chicken test. This was the test the 757 failed. It had not been conducted on the 767, which has not British customers.
The 757 failure meant both airplanes had to be modified, since the metal overheads are structurally identical. Sixteen 767 cabs already had been completed, and had to be cut apart so reinforcing metal could be installed.
Mort Ehrlich, and Eastern Airlines senior vice president, said he watched Airbus Industrie conduct chicken tests in Toulouse, France.
"A few of us who were there uttered the classic remark about how hard it is to be a chicken in Toulouse," he said. "I guess the same is true in Seattle."
Home delivery | Contact us | Search archive | Site map | Low-graphic
NWclassifieds | NWsource | Advertising info | The Seattle Times Company
Back to top