There was a further sooted area after the taxiway up to the broken light. Between 3,174 and 3,461 yards (2,902 and 3,165 m), there was more soot on the grass along the edge of the runway, indicating that there was extensive flame after the aircraft became airborne.
In the mile after the end of the runway, near the extended centreline, other debris was found. One piece was identified as coming from a repair on the left inner elevon, the moveable control surface at the back of the delta wing. Also found were the tail cone anti-collision light, a severely fire-damaged inspection panel from the wing’s lower surface and seven inspection panels identified as coming from the upper surface of the left wing but which showed no signs of fire damage.
Beyond that, investigators found another inspection panel from the upper surface of the left wing that also showed no signs of fire. There was also a fire-damaged piece of duct and fire-damaged structural parts that appeared to have come from the tail cone of the aircraft.
The tar had melted on the roofs of buildings in the freight zone and more debris was found there. Burn marks on the ground were also visible and a wheat field was damaged by fire 1.5 miles (2.5 km) from the end of the runway.
As the doomed flight continued, the plane shed more debris. Investigators found on the ground two hydraulic shut-off valves, one damaged by fire. Then there were two lower inspection panels from the engine casing, one melted, the other intact; debris from the wings, in particular fuel tank parts; a fire-damaged hydraulic line; and another inspection panel from the left main landing gear.
Leading up to the crash site, many small pieces of metal, honeycomb components, pieces of riveted structure and parts of the rear fuselage were found. Most of these parts showed traces of fire.
Examination of the crash site itself revealed that the aircraft had struck the ground heading 120 degrees left, practically flat with little forward speed. After the impact, it broke apart and debris spread generally to the south. The wreckage was extensively burnt. Only the front parts of the aircraft, together with a few pieces of the fuselage scattered over the site, escaped the fire on the ground. Most of the wreckage, with the exception of the cockpit, remained within a rectangle measuring 110 yards (100 m) long by 55 yards (50 m) wide. Parts of the air intake for one engine, a wheel, the front left door sill and a hinge from the aircraft’s famous droop nose were buried in the ground.
Near the hotel, impact marks were found in the asphalt. The right main landing gear was found nearby. Examination of the locking mechanism of the side strut of the left main landing gear showed that the landing gear was down and locked at the time of impact. But a 12 in. (30 cm) spacer that keeps the left main landing gear in alignment was missing. It was found, later, that it had not been replaced after recent maintenance.
The hotel itself was almost entirely flattened. Panels from fuel tank five were found inside. The outer part of the left wing, with the outer elevons still attached, was found melted on the ground. Nearby was the inner part of the wing with engines one and two still attached. The fin was resting on it with the rudder nearby. The left inner elevon was found beneath the two engines, still linked to part of the wing. It is normally located between the engines and the fuselage. The engines themselves were resting on a water tank 5 ft (1.5 m) in height. Many wing parts were found nearby, including the lower surfaces of tanks six and ten.
Parts of the cockpit had hit an electric power transformer. The pilots’ seats, the throttle levers and the autopilot were found there. It was possible to recognize the aisle between the cockpit and the cabin. One of the flight recorders – the QAR, or quick access recorder – was found there. The nose landing gear was found nearby, extended.
The QAR and the other two flight recorders were recovered within four hours of the crash. The QAR’s box was crushed and its magneto-optical disc was deformed. But the memory card, visible through the half-torn casing, seemed to be in good condition. Two of the three memory cards had been torn off at impact. The third was still in place and it contained all the information the crash investigators needed.
The cockpit voice recorder showed signs of fire and impact damage, but it was possible to recover a transcript of everything that was said in the cockpit from take-off to crash. The flight data recorder was also damaged on impact and showed signs of exposure to fire. It was taken to the Bretigny Flight Test Centre, where it was opened. The tape was in position and the winding mechanism appeared to be in good condition. However, there were black marks on the tape, various mechanisms and the inside of the casing. Once it had been cleaned with distilled ethyl alcohol and patched up where it had begun to tear, it gave a read-out of reasonable quality.
The wreckage of the passenger cabin was found mixed with the debris from the hotel. The seats and most of the victims were found in the same area. None were alive. Engines three and four were found to the right of the passenger cabin.
The central instrument panel showing the engine speeds, fuel flow and exhaust gas temperatures were also found. The lever that operated the nose and visor and the landing gear selector were in the down position. All the landing gear was down.
The engines showed no signs of fire damage before the crash. Engine one showed signs that it had hit a hard object before it hit the ground. Engine two showed signs of damage from a soft object. It was determined that both were turning much slower than engines three and four.
Wheel one was burnt, but the wheel and tyre showed no signs of damage before the crash. Wheel two was also burnt, but part of the tyre was missing. It also had not been burnt before the impact. Nor had wheel five, which was found intact. Wheel six was intact and not burnt.
A partial reconstruction of the aircraft took place between 1 October 2000 and 31 January 2001. However, due to the condition of the wreckage, not much was learnt this way. It was not possible to reconstruct the surfaces located near the landing gear well, nor the majority of the right wing. But a melted piece appeared to have some small punctures.
The seats from the cockpit showed that the crew was in position for take-off with the flight engineer between the captain and the flight officer. Passengers were in the seats assigned to them on boarding. Their seat belts were fastened.
The lower surface of the port wing showed signs that it had been exposed to flame and the dry bay between the fuel tanks showed that it had been exposed to very high pressure after takeoff. This was thought to have been caused by the combustion of kerosene inside the wing. It was noted that the alloy used to make the majority of the structure of the aircraft weakened rapidly when exposed to high temperatures. At around 572°F (300°C), it is six times weaker than at normal working temperatures.
The investigators concluded that the fire under the port wing had started between V
1
and V
r
, rotation speed – that is, the speed where the plane actually takes off. On impact, the plane was engulfed in fire so intense that exposed plastic parts of the nearby hotel were melted. Firemen from Le Bourget were the first on the scene, but due to the scale of the fire they were only able to limit its spread and attend to the injured. Paris Charles de Gaulle Rescue and Fire Fighting Service then turned up with twelve vehicles, including six with foam firefighting systems. They poured more than 180,000 litres of water and 3,800 litres of emulsifier on to the flames. With the help of reinforcements from neighbouring fire stations, they managed to get the fire under control after three hours. It was clear that no one on board could have survived.
The examination of the wreckage told the same story as had been captured on film and seen by eyewitnesses. The first phase of the take-off was completely normal. But during the final acceleration, several people heard explosions. The first was followed by the appearance of a flame under the wing, between the left engine casing and the fuselage a few seconds before the plane took to the air.
Some people reported seeing pieces fall on the runway immediately after the first noise of explosion. Several people said that the conflagration began with a small flame like that from a blowtorch. It then grew much wider, enveloping the left engines, and longer until it was the length of the fuselage. The flame was accompanied by thick black smoke. According to some observers, the plane swerved slightly to the left, but then stabilized slightly off the centreline of the runway.
After take-off, numerous small pieces were seen to fall from the aircraft. After passing the airfield’s freight zone, the aircraft was no longer climbing. The angle of attack seemed to be constant and the landing gear remained extended. It flew over the RN 17, the road outside the perimeter of the airfield at around 200 ft (60 m), then made a sharply banked left turn. After turning nearly 180 degrees, it went nose up and struck the ground with its left wing. There was a conflagration followed by one or more explosions.
Other Concorde cabin crew said that, given their experience, the crew of Flight 4590 could not have failed to notice the significant changes in the aircraft during take-off. They would have noticed the engine surges, the lateral and longitudinal accelerations and the strange smell.
Investigators noted that Concorde’s tyres had burst on fifty-seven occasions – twenty-seven on British Airways’ planes and thirty on those of Air France. Nineteen of these tyre bursts were caused by foreign objects on the runway. In twelve cases, the wing had been damaged and in six the fuel tanks had been penetrated. But none had resulted in the rupture of a tank, a fire or a significant loss of power in two engines.
On 14 June 1979, Air France’s F-BVFC was taking off from Washington Dulles Airport, when tyre six deflated and lost its tread. This caused tyre five to burst, leading to the destruction of the wheel. Debris thrown up caused small punctures in tanks two, five and six. After some unsuccessful attempts to retract the landing gear due to the failure of the hydraulic system, the crew landed the aircraft back at Washington, DC, twenty-four minutes later.
On 9 August 1981, British Airways’ G-BOAG was taking off from New York’s Kennedy Airport when tyres one and two burst, leading to minor penetration of tank five. Tyre five burst when G-BOAB was taking off from London Heathrow on 5 November 1985, causing damage to the landing-gear door. Tank five was penetrated, probably by a piece of the door mechanism.
On 29 January 1988, G-BOAF was taking off from London Heathrow when it lost ten nuts from wheel three. A bolt punctured tank seven. Tyre four burst on G-BOAF when it was landing at London Heathrow on 15 July 1993, damaging the gear door mechanism. Tank eight was damaged, probably by a piece of this mechanism. The tyre burst when the braking system jammed. Then tyre two burst when G-BOAB was taxiing at London Heathrow on 25 October 1993, again due to the braking system jamming. There was damage to the water deflector and tank one suffered minor penetration, probably from a piece of the deflector.
Four of these events occurred during take-off. One was caused by an object on the runway. In two cases the tyre burst occurred for reasons that were not determined. In the other case, it was due to a tyre deflating while the aircraft was rolling at high speed. One of these events resulted in an aborted take-off. In the three others, the aircraft took off, then returned to the airport to land instead of proceeding to its destination.
By far the most serious was the accident that had taken place at Washington, DC, on 14 June 1979. Most of the structural damage resulted from the impact of pieces of wheel rim on the wing. Three tanks, whose skin is just
1
/
20
in. (1.2 mm) thick, were penetrated, one by a piece of rubber from a tyre. The resulting leaks lost fuel at a rate of four gallons a second. It was considered that the risk of a fire was low and it was not necessary to install protection for the underside of the wing. In the later accidents, the tanks were not penetrated by pieces of the tyre itself but by debris dislodged by the tyre, and modifications were made.
After the crash in 2000, tests were made to see what would happen to tyres if they were damaged by a curved metal strip similar to the one found on the runway afterwards. At a Goodyear technical centre in the United States, the tyres were installed on the side of a trolley. A load of about 25 tons was applied to each tyre. The sample strips were stood on edge on a concrete surface and a truck towed the trolley over them at 6 mph (10 km/h). With the strip in that position, it was flattened by the tyre. But in other positions it cut into the tyre. Eventually, a position was found where it cut the tyre in a similar way to the piece found on the runway at Charles de Gaulle airport.
Further experimentation showed that if the rupture took place at 60 metres per second (216 km/h; 134 mph), a piece of tyre weighing 2.5 kg would be flung off. At higher speeds, heavier pieces were thrown off. These would hit the underside of the wing at up to 65 metres per second (234 km/h; 145 mph).
The metal strip found on the runway did not come from Concorde. It was clearly part of a plane so investigators set about trying to find the aircraft it had come from. The piece was a wear strip from an engine fan reverser cowl. It was suspected that this came from the DC-10 operated by Continental Airlines that had taken off from Charles de Gaulle Airport five minutes before Concorde, bound for Newark, New Jersey. Only one plane, an Air France Boeing 747, had taken off in the interim. Investigators headed for Houston, Texas, home of Continental.
They examined the plane and the aft support of its fan reverser. The lower left wear strip, which was about 17 in. (44 cm) long, was missing. This would not be noticed when the fan-reverser cowl was closed. The support was painted with green epoxy primer like that found on the metal strip on the runway after Concorde had crashed. The position where the missing part should have been was covered in red mastic, also like the runway strip. Some rivets were still in place. The right wear strip appeared to be an original part, but another of the left wear strips had been replaced and the level of wear on the strip adjacent to the missing strip exceeded the tolerances specified by the manufacturer.