Basic Facts:
Tuesday afternoon Paris time a Concorde supersonic passenger aircraft owned and operated by Air France crashed during takeoff from Charles de Gaulle Airport near Paris, France. In an eyewitness account aired by ABC shortly after the crash, a professional pilot reported watching the Concorde trailing a long plume of flame from one of its left engines just before it lifted off the runway. He specifically commented that the flame he witnessed was far in excess of anything the aircraft's afterburners could produce in normal operation. Less than 40 seconds later the Concorde's nose rose sharply and the aircraft rolled to the left and went nose down and inverted into a small hotel, killing all one hundred passengers and nine crew aboard, plus four people on the ground.

Later in the day, a picture (see above) was circulated showing the aircraft skimming the trees at a very low altitude trailing a plume of flame from one or more engines which was as long as the length of the fuselage (204 feet). The photo confirmed the already highly reliable analysis of the pilot-witness.

The Concorde was bound for New York's Kennedy Airport on a charter flight, carrying mostly German tourists, none of whom survived.

The aircraft involved was placed in service with Air France in 1980, and was one of the younger Concordes. It had logged over twelve thousand hours of flying time in the intervening twenty years, and five thousand takeoff and landing sequences. This is in comparison to some subsonic passenger jets that are approaching sixty thousand hours of flight time and almost as many cycles (or takeoff-to-landing sequences).

Discussion:
First, all of us should take a moment to think of the men and women who lost their lives in this accident, and the pain and grief of those who were left behind. I say that not to engage in emotional manipulation, but to remind you that even the most clinical analysis needs to hang tight to the reality that we're discussing members of our human family and not just facts and figures.

What caused this crash is unknowable until a full probe has been completed (which will be done by French authorities, who are quite competent at air accident investigation). Keep in mind, however, that there is never just one cause to an air accident.

In this case, the eyewitness reports and the photograph of the crippled aircraft conclusively show that something catastrophic occurred sometime during the takeoff sequence. The proof of that is the extremely long and robust plume of flames under the left wing. Initially it appeared that the plume of flame might have been coming from one or more of the engines, but that now does not appear to be the case. In fact, a tire - probably the right front tire on the left main landing gear - exploded and/or disintegrated sometime during the take off roll after reaching commit speed. The debris from that tire was undoubtedly thrown up and forward, part of it being ingested by number two engine (killing the engine quickly) and some of it being ingested by number one engine (which slowed down and then came back up to full power). In addition, the debris from the tire either directly or indirectly punctured the underside of the wing creating a massive fuel leak which immediately ignited creating the plume of flame. And finally, the landing gear retraction system was apparently damaged by the same explosion - all of this occurring while the aircraft was in the vulnerable position of being too fast to stop on the runway and yet below rotate speed at which it could lift off the runway. One of the remaining questions that has yet to be answered is whether or not engine two actually exploded (in other words, an uncontained engine failure) peppering the underside of the wing with debris. Because the investigators have found no engine parts on the runway and only tire debris it appears unlikely that number two exploded, but because it rolled back to an RPM of zero, it was clearly, massively damaged.

So what could this plume of fire mean?
Now, if just one engine had failed, even with a gigantic plume of fire trailing behind, the remaining three engines at full power should be able to thrust the aircraft to a safe altitude and keep it in the air long enough to turn it around and land it. Concorde pilots practice engine-failure-on-takeoff scenarios in each simulator training session, and the performance capabilities of the aircraft even at heavy gross weight are well known. This accident, in other words, can not be explained by a single engine loss by itself.

Concorde's engines are grouped two to a side, with engines 1 and 2 side-by-side in a metal box under the left wing, and engines 3 and 4 in a similar metal box under the right wing. There is a titanium "firewall" separating engine 1 from engine 2, and another separating engine 3 from engine 4. These "firewalls" are designed to be tough enough to prevent parts from an exploding engine from coming through to the other side and causing a second engine to fail, but it's not a foolproof system, and there is a possibility that a single engine disintegration could have destroyed power on the adjacent engine - although it does not appear it is the case in this accident.

From what investigators have said so far, we know that the aircrew was presented with a single engine loss that left them with a control problem from the very beginning. With an engine out on one wing and all the remaining engines at maximum power, the aircraft has a powerful tendency to roll over toward the side with the inoperative engine.

This is known as an "asymmetric trust" condition, and the only thing that can keep the aircraft flying straight (and not rolling inverted) would be the pilots pushing the rudder pedals to almost full right deflection in the direction of the good engines.

Now, here's the problem: There is a minimum speed (called a VMCA, or Velocity of Minimum Control/Air) below which even the rudder isn't enough to keep the aircraft from rolling inverted and crashing. If the pilots have enough power to keep the craft flying faster than that minimum speed, they can stay airborne. But if reduced engine power, and perhaps metal damage to the exterior of the aircraft's wing from the engine failure/explosion is producing too much drag, the aircraft may slow down until the speed drops below that critical VMCA. In fact, there is a two-engine VMCA and a three-engine VMCA. This aircraft lifted off the runway above the three-engine minimum control airspeed and thus the pilots were able to continue to fly even though they could not accelerate or climb. However, the airspeed that they were able to maintain was 30 - 40 knots below the minimum controllable airspeed for two engines out on one side, therefore, when number one engine on the left side - the remaining engine on that side - died, the aircraft instantly became uncontrollable.

What could the pilots have done?
That is a legitimate question, but first keep in mind the fact that the flight lasted for less than 40 seconds, leaving the pilots little if any time to react.

There is a specific "commit" speed on takeoff above which you no longer have enough runway left within which to stop the aircraft if you decided to abort. Above V1 (the commit speed) you are all but guaranteed to run the aircraft off the far end of the runway, and an aircraft full of fuel can turn into a bomb as it plows off the end. This, in other words, is not a choice a careful pilot would make once he's accelerated above V1. Instead, he'd continue the takeoff, supposedly secure in the knowledge that he had enough power to handle a single engine loss.

Also, the fact that they're trailing a 200 foot plume of fire cannot be known by the pilots because they can't see behind the wing. Even if the pilots knew from the engine instruments that they'd suffered an engine loss, there would be no way to know that it had also become a massive blowtorch. Therefore, it would be unlikely that a sudden engine or multiple engine failure at high speed late in the takeoff roll would have prompted the pilots to abort.

In addition, because the Concorde's wing does not have "leading edge devices" and the type of massive flaps on the trailing edge used by other jets to slow their takeoff and landing speeds, Concorde must achieve nearly 220 knots for a normal heavy takeoff, as compared to 140-160 knots for a fully loaded 747. That's a considerable difference, requiring a lot more thrust, and reducing the margins for error.

Once airborne with a diminished power situation at heavy takeoff weights, the aircraft would be in a very marginal position aerodynamically. While fighting to just sustain flight until they could accelerate and gain some altitude, every pound of thrust, and every control movement, would be critical. There is no question they would be fighting hard to keep the wings level (for maximum lift), and under no circumstances would they have tried to turn back to the airport while skimming treetops at no more than a hundred feet. Reports, in other words, that seemed to indicate that the flight crew purposefully avoided a larger hotel or other structures are simply wrong. This aircrew was doing their best to just maintain control. They weren't searching for a place to crash, because they had no intention or desire to put the aircraft on the ground anywhere but an airport. The fact that they did is conclusive evidence of progressive loss of control.

Does this mean the airplane is unsafe now?
Not necessarily. However, there is obviously a significant design vulnerability in the Concorde that has shown up many times in the past with exploded tires, damage hydraulics, punctured fuel tanks, and engines ingesting tire debris after such an explosion. While we are far from final determination of all the events that led to this accident, one thing is painfully clear: the design vulnerability and the proximity of the engine to the tires and the ability of the engines at high power settings on take off to suck up the debris from an exploding tire (leading to one if not two engine failures on one side) means that something substantive must be changed in the equation to permit Concorde to operate with significant assurance that this accident can never happen again. Whether this involves using more "beefed up" tires, greater inspection of existing tires, reducing the service life of the tire, or some combination of the above, it is difficult to see how either Air France or British Air can continue to operate the Concorde without making some substantive change or identifying a significant difference in the existing fleet from what was on the main landing gear of the accident aircraft.

And there's one other key point that needs to be made: This accident should in no way diminish the incredible accomplishments of the Concorde in operating for 27 years without scratching a passenger. Even though its era may be coming to an end in the near future, the Concorde's ability to routinely travel with passengers at twice the speed of sound and deliver them safely has validated the highest expectations of its designers.

More later.
John J. Nance

Back to top or previous page