Ice on Jets – Recurring Risk, Tolerated for Years
Federal government fiddled while planes kept crashing for the same reasons
THE PLAIN DEALER
March 1, 1993
an excerpt from DEADLY DELAYS
Part Two of a five-part series
March 10, 1989. Dryden, Ont. 12:11 p.m. : Air Ontario flight 1363, carrying 65 passengers prepares for takeoff.
In seat 7C, Nancy Ayer, a bubbly woman with a loud laugh, is excited about joining her husband in Winnipeg to celebrate his job transfer to a new sales territory.
The plane lifts slightly, the engines grinding with more power, but it fails to gain height and crashes into a dense forest.
Struggling to escape, survivors clamber over one another, Nancy Ayer runs from the plane in flames, her arms flailing. “Help me,” she cries.
Dazed, she stumbles back into the wreckage. A fellow passenger grabs and rolls her in the snow to douse the flames.
She hangs on long enough to get to a hospital where her husband, John, awaits her.
“I’m not in good shape, am I?” she asks.
“No,” he tells her. “You’re going to die.”
“You’re kidding,” she says.
“I love you,” he says.
“I love you,” she says.
Six hours later, Nancy Ayer is dead. Total fatalities: 24
Cause: “Ice and snow adhering to the upper surfaces of the wings.”
By KEITH C. EPSTEIN
WASHINGTON
It’s 1985. Ice on the wings had already caused three airline crashes.
“A hazard exists,” federal safety officials write in a confidential memo. Certain types of planes are “more sensitive” to accumulations of ice too small for pilots to spot.
But the government does nothing and, in succeeding years, other ice-sensitive planes fall from the sky, killing 81 people.
Seven years after the memo, USAir’s Flight 405 became the 10th such plane to crash, plunging into icy waters just beyond the runway at New York’s LaGuardia Airport and adding 27 people to the deadly toll of bureaucratic delay. Once again, the government’s failure to react to recurring accidents involving wing ice on Dutch-made Fokker F-28’s and two older models of McDonnell Douglas’s DC-9 led to tragedy.
Top officials in many federal agencies know such delays are common. Barry L. Harris, the Federal Aviation Administration’s acting administrator, acknowledges the government’s “lumbering lack of responsiveness.”
Charles O. Miller, a former top U.S. accident investigator, was helping the Canadians investigate the crash of the F-28 carrying Nancy Ayer and 64 other passengers, when Flight 405 went down at LaGuardia. “I felt,” he said, sighing, “like I’d come to in a time warp.”
By the time National Transportation Safety Board investigators drafted their 1985 memo, they already knew wing icing had caused crashes of an Ozark Airlines DC-9 series 15 in Sioux City, Iowa, in 1968; an F-28 near Izmir, Turkey, in 1974; a Trans World Airlines DC-9 series 10 in Newark, N.J., in 1978; and an Airborne Express DC-9 series 15 in Philadelphia in 1985.
Before each accident, pilots had looked through their windows, checking for ice. Later, the pilot of Flight 405 also would check, but there was no way he could have seen ice crystals possibly no thicker than 2 millimeters that downed his F-28.
In 1985, Airborne Express pleaded with the safety board to investigate the DC-9 accidents for a common thread.
The result was the 1985 memo, which urged that flight manuals for the earlier versions of the DC-9 emphasize the unique hazards of tiny amounts of ice on wings.
The internal memo also referred to a 1979 study by McDonnell Douglas’ chief engineer, Ralph E. Brumby, which found that the planes could become unstable and stall from even “extremely small amounts of frost.”
Government officials, manufacturers and aviation experts long have recognized that planes without slats – retractable mini-wings along the wing’s front edge – are vulnerable to small amounts of frost, snow or ice.
Pilots of most commercial jets, which have slats, can adjust them to compensate for icy conditions and give planes extra lift and stability.
While planes with slats outnumber the others 10 to 1, during the last 24 years, only three of them crashed with ice on wings. But 10 F-28s and DC-9s without slats crashed during the same period.
As early as December 1969, Fokker itself warned owners of its F-28 that ice on the wings was “particularly dangerous,” and could prevent the aircraft from lifting, sending it into an uncontrollable roll.
The safety board’s 1985 internal memo echoed these findings.
“A hazard exists,” the memo said. “Aircraft without leading edge devices are more sensitive to even light amounts of airfoil ice, which may not always be visibly detectable and which may accumulate during pre-takeoff taxi operations.”
Senior officials at the safety board refused to make the memo public and failed to seek the changes suggested by their investigators. The safety board can only suggest remedies, only the FAA can require them. Yet the safety board failed to give the FAA the suggestions in the internal memo.
“We didn’t feel we were on real solid ground,” said William G. Laynor, deputy director of the safety board’s office of aviation safety. “More research was needed. We were busy with other things. Plus, industry was already taking action.”
He was right about industry. Independently, McDonnell Douglas had published new training material about wing ice hazards that might have helped if only the government had required airlines and pilots to read it.
“Clearly, pilots weren’t getting the message,” said Brumby, the McDonnell Douglas chief engineer. “It’s not like McDonnell Douglas and pilots talk directly to each other. But it was so frustrating to see this happening to Learjets and F-28s and DC-9s,” he said. “And then the Continental crash happened.”
In November 1987, after de-icing, a Boise-bound Continental Airlines flight at Denver waited 27 minutes for takeoff. When the DC-9 series 10 crashed, killing 28 people, the board no longer doubted the evidence.
Yet today the FAA still hasn’t acted on the board’s urgent pleas after the Continental crash to force airlines to use a longer-lasting antifreeze and end-of-runway de-icing pads common in Europe. Crashes due to wing ice have all but disappeared in Europe in the past two decades.
The jelly-like, glycol-based antifreeze lasts for 45 minutes, three times as long as the weaker North American version. Many U.S. airlines object to the added cost. Spraying a Boeing 727 with the stronger substance adds about $200 to the cost, but the need to redo the job is less likely.
Nor did the FAA include information in flight manuals to help pilots of planes without slats check for ice before takeoff.
More than a year after the Denver crash, the FAA said it “does not believe there is anything unique” about planes without slats.
NTSB did nothing to counter that claim. The safety board’s Laynor acknowledged the issue “fell through the cracks.”
Board member Susan M. Coughlin agreed. The board “should have communicated its strong disagreement … swiftly and clearly,” she said.
Even so, if the FAA had at least complied with a 1981 request from the safety board, the risk would have been less. The safety board had asked the aviation agency to include specific warnings about ice and limitations of de-icing fluids for each kind of aircraft.
The FAA refused, arguing it needed to improve weather forecasts first. The FAA still had not distributed wing icing warnings when a Ryan International Airlines DC-9 series 15 crashed in Cleveland in February 1991, another victim of ice on the wings.
Such a warning, which would have included specific information for the F-28, could have warned Flight 405 pilots John Rachuba and Wallace Majure that the de-icing fluid applied to their wings was good for only 12 to 15 minutes. The jet sat on the runway for 35 minutes.
The Air Line Pilots Association, meanwhile, continued to hound the safety board. “The system has failed,” the pilots union complained in a formal filing following the Continental crash in Denver. “Information documented in the three previous DC-9-10 accidents foretold of the unique hazard.”
The similarities of “almost identical” crashes cannot be ignored, the union said. “The DC-9 series 10 aircraft … is more adversely affected … than other DC-9s.”
The pilots association demanded that the airlines modify the planes, making them safer.
Neither the safety board nor the FAA was willing to go that far. NTSB thought its other recommendations would be sufficient.
But the 1989 crash of Canadian F-28 with Nancy Ayer aboard should have erased any doubt about the susceptibility of F-28s to wing icing. By the end of the year, three other F-28’s had crashed in icy conditions, in Argentina, Turkey and Korea. Fokker itself warned U.S. owners to take off within 15 minutes of being de-iced.
In Washington, nothing happened.
Then on Feb. 17, 1991, just after midnight, a light snow was falling at Cleveland Hopkins International Airport. The two-man crew of Ryan International Air Flight 590, a DC-9 series 10 cargo jet on a mail run to Indianapolis, failed to have the plane de-iced, though it had been on the ground 26 minutes. On takeoff, the plane lifted and stalled.
Both crewmen died.
Cause: “Ice contamination on the airplane’s wings,” the safety board concludes.
“We now find ourselves making the case once more for a safety issue that we were confident existed as early as 1985,” complained the safety board’s Coughlin.
“The facts are disturbingly similar to previous accidents,” McDonnell Douglas said in a private March 21 letter to some airlines, referring to all DC-9 accidents to that point. The DC-9 series 10 and 15 wings are “sensitive to small amounts of ice, snow, freezing precipitation and frost.” Crews should “taxi back for a second de-icing if a delayed takeoff … raises any question of wing condition.”
McDonnell Douglas quietly decided it was time for a fix. Company insiders described the remedy as an unimpressive set of five or six electrical relays, two pressure switches, a couple of brackets, some tubes and wires – no more than a few thousand dollars in parts.
Hot air would be conveyed from the engines to the wings using technology available a quarter-century ago, when the company made the planes. Retrofitting those older models would be easy.
Today, nearly two years later, it still isn’t required.
The evidence continued to pile up on the desks of aviation experts in government and industry; the FAA, still unwilling to acknowledge the problem, refused to act.
In June, a story in New York’s Newsday raised questions about the DC-9 series 10’s susceptibility to icing. The story was widely distributed at the aviation agency and the safety board – but the FAA again took no action.
Concluding its investigation of the Cleveland crash, the safety board blamed the FAA for “a lack of appropriate response … to the known critical effect” of ice on DC-9s.
Early last year, the FAA finally agreed with one of the safety board’s suggestions: to include a warning in flight manuals that “wings without leading edge devices are particularly susceptible to loss of lift due to wing icing,” and to recommend a hands-on check of the wing.
Still, nothing required crews to do what the manual suggested, or to use the longer lasting antifreeze, and there was no mention of other, ice-sensitive DC-8s and F-28s.
The FAA decided in February 1992 to conduct a survey to identify the other planes – a task agency spokesman Fred Farrar said “probably” could have been accomplished in a single afternoon.
The FAA was still working on its “survey” when USAir’s Flight 405 crashed. The ice on its wings, too small for the pilot to see, “resulted in an aerodynamic stall and loss of control,” the safety board officially determined Feb. 17.
But the safety board also faulted the airline industry and the FAA for their `failure’ to establish de-icing safeguards for the crew.
This time, the FAA was convinced. Within a month of the crash, it admitted the “difference between hard wings and slatted wings.”
Even so, Anthony J. Broderick, an FAA associate administrator, insisted: “Until the tragedy of Flight 405, I know of no one who clearly recognized this factor.”
“I can’t understand what made him say that,” counters NTSB’s Laynor.”It’s been hard to miss.”
Within a month, the FAA even agreed to consider the easy remedy contemplated by McDonnell Douglas. Price tag: $26,500 per airplane. The airline industry opposed it, saying the $800,000 fix was too costly and the matter died.
“That’s ridiculous,” counters Charles Eastlake, a professor of aerospace engineering at Embry-Riddle Aeronautical University in Daytona Beach, Fla. The fix, he said, is both easy and inexpensive.
Other stories from this series:
Deadly Delays: Bureaucracy is Killing Us – Long after wreckage is examined for clues, causes determined, and solutions urged – thousands of Americans still risk death or injury in similar accidents. (The Plain Dealer)
Ice on Jets – Recurring Risk, Tolerated for Years – Despite pinpointing causes of multiple commercial airliner crashes, Washington does nothing to require simple, proven remedies advocated for years
Flight 405: The Story of Four Passengers- Strangers on a plane, going about the routine business of flying, trusted that commercial aviation had become as safe as airlines and the government could make it. They were wrong.
Killer Trucks – Why the Slaughter Won’t Stop – Trucks with self-adjusting brakes would have fewer accidents, causing less damage and saving hundreds of lives. Yet the government dawdled in requiring them.
While Grownups Squabbled, Children Died – Battles between automakers and regulators for more than a decade stalled development of safer child seats.
Feds Shrug Off a Life-Saver for Commuter Planes – Commuter airline flights crashed repeatedly into the ground for lack of a simple device urged for years by federal safety officials. The FAA declined. Planes kept crashing.
Cessnas Crash, but Agencies Do Nothing - For decades, Cessnas chocked from a carburetor flaw known to the manufacturer and the government. Yet pilots had never heard of the problem, and the government required no fix.
Yellow Coffins – Modern school buses are among the safest means of transport. Yet when accidents occur, children are often trapped. Still, the government for years allowed preventable tragedies to recur.
Safety Board Has No Teeth – The National Transportation Safety Board is widely known for investigating accidents. What many people don’t realize is that it’s powerless – a toothless tiger.
