Electrical problems could result in the loss of vital control systems
It is possible that the fury of an equatorial storm brought down Air France flight 447.
The plane’s flight path seems to have taken it through what meteorologists call the inter-tropical convergence zone.
This is where two air masses meet, sending huge storm clouds more than 40,000ft (12,000m) into the sky.
Eight years ago, former British Airways captain Roger Guiver was confronted with an enormous storm during a flight from Cape Town to London Heathrow.
“You take weather like that extremely seriously,” he says. “You don’t go anywhere near it.”
There are two potential dangers – lightning and severe turbulence.
Lightning can strike anywhere – the charge flows around the plane’s skin and can damage electrical systems.
“Static wicks” on aircraft wings are meant to dissipate electricity
But aircraft wings have what are called “static wicks” which dissipate the electricity safely.
Bored, long-haul passengers looking out of the window at the wings will spot them – thin, aerial-like structures, trailing in the slipstream.
Roger Guiver says one dramatic warning of a possible lightning strike is St Elmo’s Fire – static that flickers over the windscreen as the plane flies through a storm.
But lightning almost never causes air crashes, at least directly.
The respected Aviation Safety Network database lists just 15 incidents in more than 50 years of aviation history.
The worst was the loss of an Iranian Air Force Boeing 747 in 1976 near Madrid. Lightning ignited vapour in a fuel tank, causing an explosion.
If lightning did strike AF 447, it is more likely to have caused the electrical faults mentioned in automated maintenance messages which were sent out over a satellite network shortly before the plane disappeared.
But electrical problems may in turn have resulted in the crew losing vital control systems, or a fire.
The pilot’s main weapon against turbulence is weather radar – the receiver mounted in the nose of the aircraft can pick up signs of storm clouds ahead, which are displayed in the cockpit.
Crews aim to fly at least 10 miles (16km) around the worst storms, for reasons of safety and passenger comfort.
But it is not an exact science. Weather radars detect moisture primarily, and sometimes struggle to identify ice crystals, which can be present in the worst storm clouds.
At night, storms cannot easily be seen by eye.
Pilots try to avoid ending up in big storm clouds because of the forces they can impose on a plane.
At cruising height, the plane must be kept to precise speeds – the altitude means changes of speed can cause stalls.
But in a storm, the plane can be lifted up or thrown down in the turbulence, making it difficult for the autopilot to fly within the limits.
“It’s not frightening for us, but it’s awful for the passengers at the back,” says Roger Guiver.
Investigators will want to discover if the Air France Airbus suffered such severe turbulence that it caused catastrophic structural damage – the loss of a rudder, engine mounting, or even a broken wing.
This would be extremely rare.
The wings on newly designed aircraft are literally tested to destruction by bending them at least 50% beyond the kinds of forces produced in a storm.
As ever with air accident investigations, finding the cockpit voice recorder and flight data recorder is key.
They may be up thousands of metres below the surface.
Map of the plane’s route
But the US National Transportation Safety Board says the homing beacons they carry should be detectable down to 14,000ft (4,300m).
Retrieving them is tricky but, as military salvage experts point out, these days there is no place on the ocean floor where remotely operated vehicles cannot go.
The boxes and key sections of wreckage may be winched up, so that investigators can begin the task of explaining the fate of flight AF 447.
Full article and photos (1), (2) and (3): http://news.bbc.co.uk/2/hi/americas/8079591.stm
Photo (4): http://news.bbc.co.uk/2/hi/americas/8079122.stm
L’Airbus d’Air France volait à une vitesse “erronée” et s’est désintégré en vol, selon les messages captés au sol.
Airbus devrait publier une “recommandation” validée par le Bureau d’enquêtes et d’analyses (BEA), jeudi 4 juin, destinée à toutes les compagnies utilisant des biréacteurs A330 et rappelant que, en cas de conditions météorologiques difficiles, leurs équipages doivent conserver la poussée des réacteurs et l’assiette correctes pour garder l’avion en ligne.A ce stade de l’enquête, deux faits se dégagent : la vitesse de l’Airbus, disparu le 31 mai dans l’Atlantique sud avec 228 personnes à bord, était “erronée” et un enchaînement d’événements catastrophiques a conduit à sa désintégration en vol. Ce que confirment les messages émis automatiquement par l’appareil au cours des quatre dernières minutes de son vol, entre 23 h 10 et 23 h 14 locales (4 h 10 et 4 h 14 de Paris).
Un journal de Sao Paulo, le Jornal da Tarde, a eu connaissance de la chronologie et de la teneur de ces messages, selon lui grâce à une source proche de la compagnie française. Ces informations ont été publiées, mercredi 3 juin, par le très sérieux Estado de Sao Paulo (les deux quotidiens appartiennent au même groupe de presse).A 23 heures, soit vingt minutes avant l’entrée prévue de l’Airbus dans l’espace aérien du Sénégal, le pilote envoie un message signalant qu’il traverse une zone de fortes turbulences, dans des cumulo-nimbus chargés d’électricité et des vents violents. Les satellites météo attestent que les vents soufflaient à 160 km/h.
Les problèmes commencent à 23 h 10 lorsqu’un message signale la déconnexion du pilote automatique. On ignore si cette opération a été effectuée par l’équipage ou si elle a été provoquée par les systèmes de sécurité. Cette déconnexion intervient, en effet, automatiquement lorsque les ordinateurs détectent une panne grave.
Des pannes multiple
L’équipage a-t-il voulu dévier de sa trajectoire pour éviter les zones les plus dangereuses ? Si oui, l’opération s’annonçait délicate. Selon un pilote cité par l’Estado, le contrôle manuel d’un avion à haute altitude est “extrêmement difficile”. Une chose est sûre : à partir de cet instant, l’Airbus est piloté manuellement.
A la même minute, un autre message informe que le “fly-by-wire”, autrement dit les commandes électriques de vol qui activent les volets et les ailerons, passe sur le régime “alternative law”. Cette alimentation de secours est actionnée automatiquement en cas de pannes électriques multiples. L’avion conserve alors suffisamment d’électricité pour voler, mais plusieurs systèmes de contrôle de la stabilité de l’appareil sont détériorés. Dans une telle situation, une alarme sonne pour alerter le personnel de cabine.
A 23 h 12, deux autres messages auraient signalé des pannes dans deux équipements fondamentaux, Adiru (Air Data Inertial Reference Unit) et ISIS (Integrated Standby Instruments System). Ces ordinateurs fournissent des informations capitales sur l’altitude, la vitesse et la direction du vol.
A 23 h 13, de nouveaux messages indiquent des pannes électriques dans l’ordinateur principal (Prim1) et l’auxiliaire (Sec1) de vol. L’ultime message est envoyé à 23 h 14. Il signale : “cabine en vitesse verticale”. C’est l’indice d’une dépressurisation, cause ou conséquence d’une désintégration en vol.
Dans les milieux proches de l’enquête, on conteste formellement la chronologie des informations – et notamment la panne de l’Adiru et de l’ISIS au moment indiqué – tout autant que les conclusions “hâtives” qu’en tire l’Estado.
Les informations de la presse brésilienne tranchent avec la prudence affichée par le BEA français en charge de l’enquête. Lors d’un point presse, mercredi 3 juin, Paul-Louis Arslanian, directeur général du Bureau, a insisté sur les difficultés de l’enquête qui “venait seulement de commencer et ne sera pas facile”. Il a ajouté : “Nous avons déjà rassemblé beaucoup d’éléments.” Il a, par exemple, confirmé l’émission de messages automatiques durant environ trois minutes avec “une grande concentration de messages sur une minute”.
Ces messages, a ajouté M. Arslanian, “donnent des indications. Reste à les comprendre et à les mettre en perspective”. Quant aux deux “boîtes noires” qui contiennent les conversations de l’équipage et les paramètres du vol, M. Arslanian n’est pas “d’un optimisme total sur une éventuelle récupération”, mais le BEA a “déjà travaillé sans enregistreurs”, précise-t-il. Un premier rapport devrait être rendu à la fin du mois de juin.
Full article: http://www.lemonde.fr/societe/article/2009/06/04/l-airbus-d-air-france-volait-a-une-vitesse-erronee_1202293_3224.html#ens_id=1200707
Airbus Warns of Speed Problems After Jet Crash
Brazilian aircraft crew on search for the missing airliner
In the first hint that malfunctioning airspeed indicators played an important role in the crash of an Air France jet bound for Paris earlier this week, Airbus issued a warning to all its customers to follow established procedures when pilots suspect the devises are not properly functioning. The message, approved by French investigators, said that the reminder was sent “without prejudging the final outcome of the investigation,” but clearly pointed to the possibility that mismanaging the plane’s speed had been one step in a cascade of on-board failures, leading to the crash northeast of Brazil and the death of all 228 people on board.The message noted that “there was inconsistency between the different measured airspeeds” in the Airbus 330 that crashed. The inconsistency was among the error messages that were sent by the plane’s automatic systems by radio back to an Air France maintenance base.
The warning was issued as the Brazilian air force on Thursday recovered the first piece of floating debris in the Atlantic Ocean from the Air France jet that crashed, and as more questions than answers were emerging over how and when the plane may have broken apart.
The Brazilian military Web site reported that a helicopter crew extracted a structural support piece about eight feet long — which might have come from the jet’s cargo hold — about 340 miles northeast of Brazil’s Fernando de Noronha islands. It also recovered two buoys, which airplanes carry as standard emergency equipment.
Airspeed on jets is measured by the combination of a tube that faces forward, called a Pitot tube, and an opening on the side of the plane, known as a static port. By comparing the pressure in the Pitot tube that is created by the oncoming wind with the pressure from the static port, an airspeed instrument can determine out how fast the plane is moving through the air. The model that crashed, an A330, has three pairs of tubes and static ports. But other instruments can also be involved in calculating air speed, and the notice to airlines, called an Accident Information Telex, did not specify the nature of the inconsistency.
The message went to airlines that operate all of the Airbus products, from narrow-body A318 models to the double-decker jumbo A380.
Failure to properly manage an inconsistency has been cited in several crashes of big jets of various manufacturers.
The Airbus notice refers to the Quick Reference Handbook and the Flight Crew Operating Manual, which is a more detailed volume that is also kept in the cockpit.
For all the models, however, the advice is the same: keep the plane level and keep the throttle setting in place, while troubleshooting the problem.
So far, the scant physical evidence from the crash has not helped investigators determine a cause.
A senior Brazilian military official said that a 12-mile-long fuel slick found on the surface of the ocean seemed to rule out a mid-air fire or explosion as the cause of the disaster.
But a pilot for Air Comet who was flying in the vicinity on Sunday night told the Spanish newspaper El Mundo in Thursday’s edition that he saw a bright flash of white light at the same time the Air France Flight 447 disappeared about 700 miles off the coast of Brazil on its way from Rio de Janeiro to Paris.
The chief executive of Air France, Pierre-Henri Gourgeon, said at a private meeting with families that the plane disintegrated either in the air or when it slammed into the ocean, and that and there were no survivors.
In acknowledging that it had found more debris on Thursday, news reports quoted a Brazilian military official as saying the color suggested it was internal parts of the plane.
“We don’t know if it’s the fuselage, (but it’s) probably internal because we found brown, white and two yellow bits,” Ramon Borges Cardoso, the director of the Air Space Control Department, said on Brazil’s Terra web site. “That doesn’t correspond to the external part of the airplane but rather the internal part, where you have the baggage hold, seats and covers.”
Mr. Cardoso said ships in the area were collecting debris to take to the operations base in Fernando de Noronha, a 12-hour round trip.
Experts are questioning whether extreme turbulence, a direct lightning strike, the speed of the aircraft during the stormy weather, or a pre-existing problem with the plane might have caused it to break up only four hours into the 11-hour flight.
Without the black boxes containing the plane’s voice and data recorders — which officials say may never be recovered the from the ocean — the only clues besides the scant debris exist in the series of 10 satellite signals.
The pilot sent a manual signal at 11 p.m. local time indicating that the plane was passing through an area of black, electrically charged cumulonimbus clouds, usually containing lightning and violent winds, according to The Associated Press. The news service, quoting an unidentified aviation official, presented a chilling timeline of events in the cockpit that reveal the plane’s ensuing problems within a 14-minute span, but not what caused them.
About 10 minutes after the manual signal, the plane sent out a series of automatic messages to indicate the autopilot had disengaged, a computer system had switched to alternative power and that controls to keep the plane stable had been damaged. About three minutes later, more automatic messages reported the failure of systems to monitor air speed, altitude and direction.
The last automatic message, which was received at 11:14 p.m. local time, signaled a loss of cabin pressure and complete electrical failure.
The system controlling these automatic signals, installed on all newer Airbus models, does not indicate the airplane’s location. Rather, it is designed to speed maintenance efforts by alerting maintenance technicians on the ground to problems before a plane lands, so they can be cleared up without delaying the next flight.
William R. Voss, president of the Flight Safety Foundation in Alexandria, Va, said in a telephone interview Thursday that the pattern of debris could indicate an in-flight break-up. Big debris that floats, Mr. Voss said, is generally not a flat sheet of aluminum, like part of the fuselage, but a more box-like structure, often the tail, and the tail often comes off first.
In this case, he said, searchers have come upon a big floating piece that was on the airplane’s likely track, indicating that it separated from the plane at high altitude, early in the event.
He noted that what was first reportedly radioed back from the automatic system was the auto pilot disengaging. This occurred, Mr. Voss said, either because a pilot saw something going wrong and took manual control or because the computerized auto pilot had a problem “and gave it back to him.”
“The sequence is pretty consistent with things going to hell from there,” he said.
Eleven planes and helicopters are involved in the search operation, the Brazilian military said, including a French ship with two remotely controlled submersible crafts that can be used to explore as deep as 19,600 feet and a United States Air Force P-3 Orion. A spokesman for the National Transportation Safety Board, Ted Lopatkiewicz, said his agency had worked with “civilian and military U.S. government agencies” to look for radar or satellite data for the area at the time of the crash but had not found any.
The federal safety board said late Wednesday that it would be part of the investigation; under international treaty, it will be involved because the engines were manufactured in the United States by General Electric. The role of the engines in the crash, if any, has not been determined.
But unlike other crash investigations that the agency joins, it will not be sending representatives, at least not yet. Peter Knudson, an agency spokesman, said that at the moment “there is no place to go.”
In many international crashes, the safety board makes its laboratory available for deciphering the cockpit voice recorder and helps in interpreting the flight data recorder. But in this case, involving a French airline and a European-built plane, that work will probably be done in France, experts said.
The main value of the debris found so far may be as a clue to the location of more important parts of the plane that are certain to have sunk, notably the black boxes. Each is equipped with a device that sends out an audio beep that in favorable conditions can be heard at 5,000 meters — about 3.1 miles. That signal begins to fade after 30 days.
The ocean is more than four miles deep in some parts of the area, and, while water is an excellent transmitter of sound, the sound waves are reflected at boundary layers where the water changes temperature, according to Duncan W. Schofield, a principal engineer at Honeywell, which built the boxes. Searchers can lower microphones to listen, but those are towed through the water at a few miles per hour, increasing the need to narrow the search area.
Aviation experts said the flight data recorder was designed to track more than 400 categories of data, including the strength of turbulence and the functioning of various cockpit systems. And the cockpit voice recorder, if recovered, would indicate who was in the cockpit.
More than four hours into the flight, the captain, who is required to be at the controls for takeoff and landing, would typically have gone off for dinner and a nap, according to airline experts, leaving a relief pilot and the first officer at the controls, both typically less experienced but well qualified for what is ordinarily a very quiet phase of flight.
The plane was flying through an area of powerful thunderstorms, but there is no clear indication that the weather was unusual for the region.
“I don’t see anything unusual about these storms,” said Timothy A. Vasquez, a former Air Force meteorologist whose company, Weathergraphics, published a Web page showing the plane’s path and satellite data of storm intensity at the time of the crash. “Planes have flown through a lot worse; I’ve seen worse squall lines in Kansas and Missouri,” he said.
Full article: http://www.nytimes.com/2009/06/05/world/europe/05plane.html?hp