The McDonnell Douglas DC10 lacks a safety system included on other U.S.-built jumbo jets that could prevent the kind of control loss that was a major factor in the May 25 crash in Chicago that killed 273 persons, The Washington Post has learned.
The Federal Aviation Administration, which certified the DC10 as air-worthy in 1971, apparently was persuaded by McDonnell Douglas that the DC10 could be controlled without the system that gives added lift at lower speeds.
At the time the aviation industry was developing its wide-bodied jets in the 1960s, the FAA proposed regulations specifically requiring such a safety system. The agency modified its rule to make it more general at the suggestion of the industry. Lockheed, Boeing and McDonnell Douglas each arrived at a different solution to the question of control loss, but Lockheed and Boeing went one step further than Douglas.
The system in question concerns control surfaces called slats located on the wings of jetliners. Slats are huge metal plates on the front edge of the wing that are extended out and down from the wing during takeoff and landing. That extension gives the wing added lift at low speeds that makes it possible for jet airplanes to land and take off on relatively short runways.
McDonnell Douglas sought to demonstrate during test flights for the DC10 that its jumbo could be safely flown under normal conditions with "asymmetrical slats," or when the slats on one wing are extended farther than on the other wing.
The result of such a condition is that one wing has a much greater ability to life an airplance than the other wing. When such an unbalanced situation becomes extreme, a plane will roll out of control. That apparently is what happened when the DC10 crashed in Chicago.
Just as American Airlines Flight 191 lifted its nose off the runway at O'Hare International Airport, the engine and its support pylon under the left wing fell off. In the process, investigators believe, the hydraulic lines that control the slats on the left wing were severed and wind pressure pushed the slats back into the left wing.
On the right wing, however, the slats remained extended. The right wing, therefore, had a high degree of life. The left wing did not. The plane rolled 107 degrees to the left and fell to the ground.
Under the McDonnell Douglas design, there was nothing to keep the slats extended on the left wing once both of the hydraulic systems that served that area were ripped loose. Both Lockheed's £1011 and Boeing's 747 have brakes on the slats to hold them out in the event of a total hydraulic or (in the case of Boeing) pneumatic failure.
Pilots interviewed agreed that although the engine fell off the CD10 at a critical moment, the American Airlines pilot would have had a much better chance of saving the plane if the left wing slats had remained extended. He would have had somewhat less of a chance if the right-wing slats had retracted - and thus placed the control surfaces on both sides in balance. The worst possible situation was the one that occurred - fully extended slats on the right wing and no slats extended on the left wing.
John Brizendine, president of Douglas, Aircraft Co., a division of McDonnell Douglas, testified in Congress last week that the company had flown the DC10 with asymmetrical slats. "The test results established that even in these conditions the aircraft remained controllable," he said.
Brizendine was asked by a reporter if the flight-test program included a condition of asymmetrical slats at takeoff. "I would doubt it," Brizendine said. "I don't remember exactly. That was eight or nine years ago."
Don Armstrong, a test pilot for the FAA who flew the DC10 during its certification trials, said in an interview, "We did two flight tests [with asymmetrical slats]. But we did not assume in the tests that the slats became asymmetric at liftoff or immediately thereafter," as happened in Chicago.
As a prerequisite to certification, the DC10 had to meet "special conditions" imposed on it by the FAA. One of those special conditions says:
"It must be shown by analysis or test, or both, that the airplane is capable of continued safe flight and landing after any...failures or jamming in the flight control system and surfaces (including...life...systems) within the normal flight envelope, without requiring exceptional piloting skill or strength...."
In the case of the DC10, that special condition covers the slat situation, according to Chris Christie, of FAA's engineering office.
The lack of a system other than hydraulics on the DC10 to prevent asymmetrical slats relfects a basic difference in design philosophy between McDonnell Douglas on the one hand and Boeing and Lockheed on the other, according to aeronautical engineers, senior airline maintenance officials and federal investigators.
McDonnell Douglas sought to show that the DC10 could be flown with asymmetrical slats. Lockheed and Boeing sought to prevent asymmetrical slats from ever happening. Either approach is acceptable under FAA rules.
The DC10 engineers built two rebundant hydraulic systems along the left wing so that if one failed, the other would power the slats. If both failed, the plane would still fly.
Boeing, which uses pneumatic (compressed air) power to drive the sltas on the 747, placed a totally independent disc brake on each slat, tied to a sensor.If the sensor finds that the slats are slipping into asymmetry, the brakes lock the slats in place. This would give the pilot a balanced, controllable situation.
Lockheed, on its £1011, has a mechanical interconnection among all its slats, plus an electrical sensor, plus a brake outboard of the engine and pylon.
Warren Stauffer, who was in charge of the design loads determination for the £1011 program and who is not Lockheed's director of engineering was interviewed in Burbank. He said that "early in the design of the £1011 we recognized the need for addressing the problems of the possible loss of a wing-mounted engine and pylon...."
Lockheed assumed, Stauffer said, that if the engine and pylon came off the plane, the hydraulic systems would be lost, and would the electrical sensors that test for asymmetry. In that case, a totally independent sensor, well beyond the eingine and pylon area on the bar that drives the slats, would sense retraction, and lock the slats in the extended position.
John L. Cooke, a spokesman for McDonnell Douglas, said that is was the opinion of the McDonnell Douglas engineers that "one thing you can't predict...is precisely what damage might be done to the leading edge of the wing and the way it might come apart" when an engine and pylon separate.
When the FAA was first writing regulations for the wide-bodied airplanes in 1968, it drafted one that said "the motion of high life...control surfaces must be synchronized by a mechanical interconnection" unless the plane would be safe otherwise.
FAA files show that an objection was filed by Aerospace Industries Association of America, the trade organization representing the airplane builders. The proposed standard dictated a specific design as a means of compliance, AIA complained, and the intention of the regulation could be met by other means.
It is not known from the records so far made available by the FAA just how much discussion there was between McDonnell Douglas and the FAA on the slat question during the 43 months it took the DC10 to win certification.
It is clear from interviews, however, that the matter was of substantial concern to the FAA because of the emphasis placed on lift devices in the special conditions for the DC10. However, the word slat somehow disappeared from the FAA regulations bebween the time they were proposed and the time they were adopted and does not appear in the special condition entitled that applies to the DC10.
Certification is essentially a series of secret meetings between the FAA and the manufacturer. To open the process, FAA Administrator Langhorne M. Bond told Congress last week, could mean a manufacturer would have to place a proprietary data in the public record. To keep it closed, congressmen suggested, means that interested parties such as pilots and passengers lack the opportunity to hear the technical debate and to comment on it.
Of the three types of U.S. jumbo jets, only the Lockheed £1011 has a mechanical interconnection among all of its slats on both wings.
The two slats on each wing closest to the fuselage of the DC10 are interconnected mechanically, but the five other slats on each wing are not. The slats near the fuselage remained extended in Chicago.
On the 747, there are brakes to prevent the slats from retracting "even if we've lost all other systems," a Boeing spokesman said.
Richard Sliff is now retired from the FAA and lives in Maryland. Sliff was the acting director of flight standards for the FAA's western region when he signed the special conditions the FAA imposed on the DC10 in 1968.
Interviewed last week, Sliff said there is no way planes with "high lift" could be flown safely in such an anymmetrical condition. "I've never tested in [in flight]," says Sliff. "It's evident what happens." CAPTION: Illustration, DC10, By Robin Jareaux - The Washington Post