An important engine-mounting support that broke on the DC10 that crashed in Chicago was such a low-priority maintenance item that it probably was never inspected on many other DC10s before the May 25 accident, investigators have told The Washington Post.
Furthermore, the "possible design problem" Federal Aviation Administrator Langhorne Bond cited when he grounded all U.S.-operated DC10 Wednesday centers on the failure of that support, known technically as a flange on the aft attach bulkhead of the pylon.
The support was scheduled for a routine check under FAA-approved maintenance program only once every 20,000 flying hours - about 5 1/2 years of normal service. Because of the random-sampling method the FAA permits on many airplane parts, the supports on only 17 percent of the DC10 fleet had to be inspected at that 20,000-hour interval, according to investigators and the official FAA maintenance manual.
The flange and the aft bulkhead on all DC10 pylons were ordered inspected by the FAA as a result of preliminary findings in the investigation of the Chicago crash, which killed 275 people. Cracks found in that flange ultimately led to the grounding of the planes.
Before those inspections, the aft bulkhead on many of the 138 DC10s in the U.S. fleet had not had their first check on the part because they had not been flying long enough or because they were not included in the 17 percent. There were only 59 DC10s flying in the United States in 1972, for example.
Although other areas of the plyon assembly - the hugh support system that is supposed to hold the engine to the wing - were inspected regularly, the aft bulkhead and the flange were concealed from view by a series of cover plates.
Because of that, "It is quite possible the aft bulkhead was never seen" on many DC10s before the FAA order, said Frank Taylor, director of accident investigation for the National Transportation Safety Board.
These facts call into question the adequacy of the FAA's certification and inspection oversight, FAA officials concede.
"The first thing we have to do is get a safe airplane and get it back in the air," said M. Craig Beard, chief of the FAA's aircraft engineering division in Los Angeles. "Then we have to take a look at the total program. It won't take as long to get the DC10 back in the air."
FAA procedures will be questioned in at least three congressional hearings, the first of which begins Monday. Already, it is known, for example, that:
Continental Airlines first discovered about two years ago that it was possible to crack the aft bulkhead flange through a maintenance procedure the FAA outlawed last week. The FAA did not find out about that incident until after the Chicago accident, although McDonnell Douglas, manufacturer of the DC10, did know about it, Beard said.
McDonnell Douglas, seeking certification for the DC10 in the late 1960s, submitted reams of engineering data of the FAA's western regional office in Los Angeles. The FAA discovered last week, eight years after it certified the DC10, that the data provided did not substantiate the "failsafe intergrity" of the pylon, Beard said.
"Fail-safe intergrity" means that, if the aft bulkhead fails, the pylon should remain on the plane and function long enough for the failure to be discovered during scheduled maintenance.
"This doesn't mean that the aft bulkhead is not fail-safe," Beard said. "It only means that the data does not substantiate it . . . Doulgas is very busy right now trying to show us you can have a substantial cut in that flange without a [pylon] failure."
Despite all this, the key issue in the DC10 crash may not be the engine and pylon assembly, but the hydraulic system.There is growing evidence, safety board investigators said, that two of the three supposedly independent hydraulic systems on the crashed plane failed and "thus turned the pilot into a passenger," in the words of one expert.
The hydraulic system is the power-steering of the aviation world. Without it, the pilot cannot direct the enormous control surfaces on the wings and tail.
The crashed plane's incident history, on file with the FAA records center in Oklahoma City, shows it had suffered at least two partial hydraulic system losses before the crash. Investigators say, however, that there is nothing unusual in that plane's history, including the partial hydraulic failures. "Any airline would be happy to have a plane that did that well," one expert said.
Nonetheless, the DC10's hydraulic system may be the subject of some design questions before the inquiry is over. The loss of an engine on American Airlines Flight 191, while undoubtedly creating control problems for the pilot, should not under normal circumstances have made the plane unflyable, pilots and American Airlines officials have said.
Safety board officials are working on the theory that, somehow, the hydraulic system failed on the left wing. Therefore, the leading edge devices that extend in front of the wing to give it extra lift at relatively slow takeoff speeds retracted.
That would deprive the wing of needed lift and throw the plane out of balance - particularly if the leading edge devices stayed out on the right wing, as they apparently did.
"We have double-checked all the witnesses to see if they can tell us about the slats [leading-edge devices] retracting just as the plane took off," said safety board spokesman Ed Slattery. "They can't say that; it happened too fast. The consensus is that they saw a little bit of fog spewing from where the engine cut loose; that could have been fuel, or it could have been hydraulic fluid."
The engine and most of the pylon came off the wing as the plane's nose lifted off the runway. "We don't know what plumbing came off with that pylon," Slattery said. "There are fuel lines, hydraulic lines, electric systems - all kinds of things in there. We're studying it."
While FAA teams and the safety board are concentrating on design questions and the pathology of the accident, the AA has also set up a number of teams around the country to look at DC10 maintenance practices.
Formal orders of investigation, giving the FAA subpoena powers, were issued Thursday against McDonnell Douglas and the eight airlines that operate DC10s in the United States.
Maintenance programs for new airliners are worked out by the FAA, with consultation from the manufacturer. That group, called a maintenance review board, establises maximum time periods between examinations of everything on an airplane.
An individual airline is permitted to modify the board's maintenance program with approval from the FAA. As time passes and experience is gained with the aircraft, some intervals are extended, others are shortened.
But an individual airline has great flexibility within that program to do as it wihes, as long as it meets the broad outlines of the approved maintenance programs. Specific tools and procedures are rarely specified.
Since 1975, the number of FAA inspectors assigned to airlines has fluctuated between 242 (in 1977) and 270 (in 1975). The current number is 266, according to FAA figures.
In the same period, the size of the U.S. airliner fleet has decreased slightly.
There is another way to look at it. American Airlines employs about 6,000 mechanics at its maintenance base at Tulsa, and the FAA has seven inspectors there to watch them and the more than 240 airplanes they work on.
"There is no way we're going to hire enough inspectors to watch every one of those 6,000 mechanics," said George House, who heads the FAA program in Tulsa. "All we can do is spot-check."
Does the FAA really know if there is another kind of airplane in the fleet flying around with an undiscovered, uninspected problem similar to the aft pylon flange?
"Let's be honest with ourselves. We don't," said Beard. "I'm not going to BS you. How are you absolutely sure of anything in life?" CAPTION: Illustration, DC10 Bulkhead, The cracked flange suspected in the DC10 crash was in the rear of the pylon and concealed by cover plates. By Richard Furno - The Washington Post