The Crash Detectives

by Christine Negroni

  Despite these two accidents, the Comet was flying high among the traveling public, on a steady course to three more disasters and an aviation mystery that would command the world’s attention.

  While the Comet 1’s propensity to ground-stall was unrelated to the design problem that would bring the plane to infamy, Foote’s personal experience on the wrong side of the first design error turned him into the Comet’s fiercest critic. He could never put his finger on what was wrong with the airplane, partly because the only other pilot who he knew had experienced the ground stall on takeoff was now dead. Foote was unaware of any other similar events. During a series of meetings with the British Air Line Pilots Association, in an attempt to reopen the investigation that had tarnished his reputation, Foote learned that eight other BALPA Comet pilots had been disciplined for their involvement in accidents during the brief but deadly year and a half the Comet 1 had been flying. He could also see that de Havilland was making changes associated with the ground-stall issue. The takeoff procedure had been revised a second time. Later versions of the airplane featured stall warning devices and a redesigned wing that provided more lift.

  Foote’s thoughts kept coming back to the Comet because events related to the plane were finding him like the proverbial bad penny. After his accident, BOAC had sent Foote back to flying propeller planes. He considered this a demotion, but he accepted it. A few months after his accident, Foote was visiting the training center at Hatfield; Captain Pentland was also there, for his Comet training. The two shared a car to the train station, but not a conversation. Foote told fellow BOAC pilot David Beaty that it was an uncomfortable ride.

  Imagine the scene: two captains, one sanguine about his upcoming status as jetliner captain, the other humiliated for the flaws in that same plane and now sentenced to spend the rest of his career flying yesterday’s piston-driven aircraft. It was a two-character Greek tragedy set in the backseat of a car.

  Later, when Foote heard the news about the accident that killed Pentland, he shared his worries with Beaty, suggesting that if anyone had paid attention to what he had been saying about the odd takeoff characteristics of the plane, Pentland and the ten others on the Empress might still be alive. His argument was getting traction, because shortly after that, Aeroplane magazine made a similar point. The article said that after Foote’s crash, the chance of “pilot error could be accepted” as one in a million, but with the Empress crash just four months later, the plane’s design must be considered. When pilots repeatedly make the same mistake, it “must be presumed to be too easily possible.”

  On May 2, 1953, the first anniversary of the Comet’s inaugural flight, Foote was in command of a four-engine York freighter headed to Calcutta. In the sky twenty thousand feet above him, a BOAC Comet was flying in the same direction. Foote’s plane was carrying cargo; on the jet there were happy passengers being waited on by an attentive staff. After a stop in Calcutta, many of the travelers would continue on to London; several were going to the coronation of Queen Elizabeth, on June 2.3

  Given the jet’s speed advantage, the Comet landed to refuel for its journey to Delhi well ahead of Foote’s lumbering York, but the two pilots met up at the airport. While the captain of the Comet, Maurice Haddon, waited for late-arriving passengers to board, he and Foote chatted briefly and then went their separate ways, Foote to his hotel, Haddon back to the flight deck of the airplane with the registration G-ALYV.

  When the last of the Comet’s passengers was seated, the plane departed. After takeoff, Captain Haddon confirmed clearance to climb to thirty-two thousand feet. That was the last message transmitted. Arriving at the hotel, Foote heard the news that the Comet had gone silent. BOAC tried to allay any fears. “We have not posted it as missing yet,” a spokesman for the airline told the Times of India news service. “We hope that the [air]liner has come down at one of the emergency landing grounds between Calcutta and Delhi.” Foote was not as optimistic.

  At dawn the following day, his was the lone British airplane among those from the Indian Air Force taking off in search of G-ALYV. Twenty-five miles to the northwest, Foote spied the wreckage in an area accessible only by foot. So widely strewn was the debris that it appeared the plane had disintegrated while still in the air. According to his friend Beaty, when Foote heard of the plane’s disappearance the night before, he was heard to have predicted, “It will have crashed.”

  The area where the plane went down may have been remote, but it was far from unpopulated. During an evening of sixty-mile-an-hour winds accompanied by dust storms and torrential rains, villagers reported explosions and flashes of light. One boy said he’d seen a wingless plane flying low. Where the pieces of the Comet hit the ground, a man told the Times of India news service that he heard human screams amid the fire, “but the heat was so intense that no one could approach.”

  The storm was immediately thought to have been the cause of the crash. A policeman in the town of Jangipara telegraphed the news to Calcutta, wiring, PLANE KNOCKED DOWN BY TEMPEST. When investigators began their work, they found both outer wings separated from the plane, explaining the “wingless machine” seen by the villagers. In the report the government issued a month later, the Indian inspector of accidents was circumspect, explaining that his panel had suffered from limited facilities and data and a lack of time to investigate adequately. A year or more would be needed, the report said. But they were clear about one thing: “the aircraft suffered a complete structural failure in the air” during a thunder squall. In a joint statement, BOAC and de Havilland challenged the Indians’ conclusion that this was because of either “severe gusts or over controlling or loss of control by the pilot” as little more than theorizing.

  To know what really happened, the Royal Aircraft Establishment, the British government’s research-and-development agency, needed to determine the sequence of the breakup. While the official cause of the loss of the first two BOAC Comets had faulted the pilot, this time pilot action was an unlikely cause, the de Havilland/BOAC statement said. If it was an attempt to protect the reputation of the airline’s pilot training and the integrity of a jetliner that had already been sold to airlines around the world, the effect was lopsided. Airlines were not reassured; orders from Japan, Brazil, and Venezuela were canceled. Among passengers, however, the plane remained popular. BOAC had no problem selling tickets to fly on the Comet.

  Even before the Calcutta crash, de Havilland had been conducting additional research to determine if the plane was experiencing metal fatigue. The company had been prompted to do so based on what it was learning about age and fatigue on de Havilland transport planes used by the Royal Air Force. Initially, the primary focus was on the wings, but soon the U.K. Ministry of Supply, which oversaw the military planes, and the Air Registration Board, responsible for civil airplanes, asked de Havilland to broaden the study. De Havilland’s chief structural engineer, Robert H. T. Harper, agreed.

  The work began in July, with technicians repeatedly applying levels of pressure to the cabin greater than what it would experience in flight. By September, the engineers found that tiny cracks were developing in the aircraft skin at the corner of the plane’s square windows.

  This sounds like an aha moment, but hang on. Rather than causing alarm, the examiners found it reassuring. The amount of pressure applied to the cabin walls during the testing was so far beyond what the plane would experience in flight, “It was regarded as establishing the safety of the Comet’s cabin with an ample margin.”4

  What de Havilland failed to realize was that the test setup provided additional support to the structure that would not be there in flight. This was a critical oversight, one that led investigators to have a false appreciation of the cabin’s strength under pressure. This same kind of error would be made sixty years later, during the testing of the revolutionary Boeing 787 Dreamliner.

  Pressure tests were also conducted on the wings of the Comet, and in December, in much the same way, tiny cracks began to appear. In this
case, however, the engineers were alarmed. The pressure had been applied for the equivalent of six thousand hours, and that was not a lot of flight time. Many Comets had already flown more than that. BOAC immediately instituted an inspection program on its planes.

  Throughout Christmas, the manufacturer and the airline discussed whether that was enough. Did the wings need to be modified? The debate continued until it was interrupted with another Comet bursting apart in the sky.

  This was the original plane—the first jetliner to carry passengers, the plane that had starred in the newsreel films. Now it was in pieces on the floor of the sea and thirty-five people were dead.

  It happened on January 10, 1954, a clear, sunny Sunday. Capt. Alan Gibson, thirty-one, and thirty-three-year-old First Officer William John Bury had an uneventful takeoff from Rome for the last leg of a flight that originated in Karachi. Rome to London would have been a quick two hours and twenty minutes in the air.

  Captain Gibson was ascending through twenty-six thousand feet and making a radio call to a BOAC pilot in another plane, but the transmission was cut off midsentence. At thirty thousand feet, the center of the passenger cabin ripped open. The tail, the nose, and the wings outward of the embedded engines blasted off with a downward force. The tail remained largely in one piece as the plane tumbled through the air and slammed open-end first into the water.

  Before the rupture of the passenger cabin, travelers had been seated two by two on either side of the plane’s center aisle in armchair-style upholstered seats that might have made them feel as if they were enjoying a cocktail in the living room of friends. With the breakup of the airplane, some of those travelers were shot out of the ruptured fuselage. Fifteen others slammed against what remained of the fractured front bulkhead, which was fitted with a library shelf and water dispenser. The torn wings exposed the engines and released flammable kerosene that quickly ignited. The fire spread inward, burning the bodies of those who remained in the cabin.

  As in the in-flight breakup of G-ALYV outside Calcutta, G-ALYP shattered into countless pieces, descending in a cacophony of explosions and smoke. In India, the witnesses had been jungle villagers. In Italy, they were fishermen who watched for three long minutes as the plane and its contents fell into the water near the Tuscan island of Elba. Racing out in their boats, they came across the only bodies that would be recovered, the fifteen trapped on board by the bulkhead.

  Autopsies by the director of Pisa’s Institute of Forensic Medicine, Dr. Folco Domenici, would provide important clues about what happened. The victims’ organs showed the plane had experienced a split-second decompression at high altitude and blunt force injuries suggested they had died from slamming into the cabin divider.

  It took awhile for investigators to recognize the value of that information, but its most immediate effect may have been to provide a shred of comfort to the victims’ families: their loved ones could not have suffered. They died immediately.

  Floating wreckage was gathered, including the rear fuselage, the engines, and the wing center section, but the rest of the plane lay six hundred feet below the surface of the Tyrrhenian Sea. At that depth it was twice what British salvage divers could descend to at the time.5 Obtaining it would not be easy, and certainly it would not be quick.

  Investigators from BOAC, de Havilland, and the Air Accidents Investigation Branch of the U.K. Ministry of Transportation focused instead on the airplanes to which they had access. All seven BOAC Comets were grounded in what the airline called a prudent measure “to enable a minute and unhurried technical examination of every aircraft in the Comet fleet.”

  De Havilland threw itself behind newer, safer versions of the airplane. The engineers examined the structures and the systems. They discussed all the possible scenarios. They designed modifications to guard against anything that could have happened, even if no proof existed that the potential problem had anything to do with the disaster. If the investigators found something that didn’t happen but could have, it became a high-priority fix. Note that, because you don’t see it happening every day.

  Lord Brabazon, the chairman of the Air Registration Board and a member of the Air Safety Board, summed up the work by saying, “Modifications are being embodied to cover every possibility that imagination has suggested as a likely cause of the disaster. When these modifications are completed and have been satisfactorily flight tested, the Board sees no reason why passenger services should not be resumed.” So even though the cause of the crashes remained a mystery, on March 23 the Comets returned to the sky.

  Of the seven airplanes subjected to the alterations described by Lord Brabazon, one was a two-and-a-half-year-old Comet, registration G-ALYY, with twenty-seven hundred flight hours. In February, the plane had undergone an eleven-pounds-per-square-inch pressurization test to check its structural soundness. G-ALYY was leased to and operated by South African Airways to service the London-to-Johannesburg route. On April 2 and again on April 7, the plane was subjected to more inspections, perhaps one too many.

  After a panel was removed from the plane to give access to a test inspector, the panel was not reinstalled correctly. The plane departed on its first leg to Johannesburg on April 7. On arrival in Rome, the mechanics were horrified to discover loose bolts inside the right wing and “an equal number of missing bolts” from the panel at the wheel well. Opening the plane up for inspection had created a hazard. That nothing had gone wrong as a result seemed a blessing, but the blessing was short-lived.

  Other maintenance issues kept the plane in Rome for a day. At 6:32 the following evening, former South African military pilots Capt. Willem Karel Mostert, thirty-eight, and thirty-two-year-old First Officer Barent Jacobus Grove, took the plane and its fourteen passengers into the air. Next stop: Cairo, three hours away.

  Radio calls were normal. The crew checked in at seven thousand feet, and then at eleven thousand feet. Then, on the way to thirty-five thousand, they made a transmission that also included telling controllers their anticipated arrival time in Egypt. Then nothing more was heard. Calls from Rome and Cairo went unanswered.

  The scenario was all too familiar.

  It took a day just to find the oil slick indicating that the flight had likely ended in the Tyrrhenian Sea off the coast of the volcanic island of Stromboli. Six bodies were found along with some airline seats, but in an area where the sea depth was thirty-three hundred feet, the downed plane was as inaccessible as if it had flown into outer space. Concluding that the wreckage would never be recovered, authorities realized that whatever they determined about this accident would be an extrapolation from what they found in the Elba crash.

  Both planes had come apart on ascent at roughly the same altitude and with tremendous force. The plane that crashed at Elba had flown 1,290 pressurized flights; the South African Airways plane had completed 900. Again, the autopsies were performed by Dr. Domenici, who confirmed that a rapid decompression had taken place. The question was why.

  The British government pulled the airworthiness certificate of the Comet 1. This time the planes were on the ground for good.

  Into this puzzle came one of the era’s most provocative thinkers, Alan Turing, who broke Germany’s Enigma code and was the subject of the 2014 movie The Imitation Game. Turing developed the Automatic Computing Engine, a machine that automated complex equations so that they could be completed faster than humans could solve them. Parts of G-ALYP retrieved from the sea were subjected to exhaustive testing and comparison to an undamaged Comet, and for that, Turing’s Pilot ACE computer was used to run the many calculations required.

  The investigators didn’t rely entirely on that newfangled thing called a computer. At the Royal Aircraft Establishment in Farnborough, workers also constructed an enormous tank into which would fit the fuselage of an entire Comet 1 jetliner, with the wings protruding out on either side like arms out of an undershirt. Beginning in early June 1956, water was alternately pumped into and out of the cabin to create 8.25 pounds of pressur
e per square inch against the walls. This was to simulate the effect of flight at forty thousand feet. Each infusion and release of water lasted about three hours, a typical flight leg for a Comet. Every day, the test airplane accumulated virtual flights and experienced actual stress.

  While that was happening, the wings were flexed up and down as if the plane were in flight. After one thousand applications, a “proving flight” was conducted and the pressure to the inside of the plane was increased to eleven pounds per square inch.

  On June 24, as the water was being pumped into the plane for the proving flight, the needles of the pressure gauge passed 8. They climbed to 9 and then 10. But it got no farther than 10.4 inches of pressure, because the fuselage ruptured at the cabin ceiling. An eight-foot section gaped open as much as three feet at the widest part. The slice ran through an area that included the cutout for the emergency escape hatch.

  That might sound like the end of the story, but was far from it. It was as if a light had been turned on a previously darkened path. Investigators could see where they needed to go, but not what they would find on arrival. The effort to retrieve wreckage from G-ALYP off the coast of Elba received new energy. Two months later, nearly 70 percent of it had been recovered, enough to show that there were several places where a fatigue crack could have originated.

  It was not possible “to establish with certainty the point at which the disruption of the skin first began,” the report of the Comet inquiry read, but the long, erratic, and oftentimes faint line charting the course of this mystery was soon to end.