Jun 1, 2009

The Air France mid Atlantic mystery will be hard to unravel

Ben Sandilands — Editor of Plane Talking

Ben Sandilands

Editor of Plane Talking


The search zone for the missing Air France A330-200 is now in darkness. The airline has ‘lost hope’ for the 228 people on board AF 447, and investigators are looking for clues in a set of automated status messages that indicated a bewildering flurry of electrical systems faults that are the last information received from the flight.

The graphic above shows the cloud cover at 0950 Eastern Australian time and the zone of seasonal and often severely turbulent weather the flight was crossing is clearly visible.

The topography which includes the lights of Rio de Janeiro from which it departed and Paris where it was headed and level of moonlight over which the satellite cloud scan is projected are computer generated.

Among the first things to happen when an airliner goes missing like the Air France flight late yesterday afternoon our time and in the early morning in darkness Brazilian time is that other airlines flying the same type, an A330-200, try to pick up any clues that may affect their own operations.

That anxiety sets in long before anything can really be known. The A330 family, and the A330-200 variant, flies in its hundreds for many carriers, including more than 20 of that family in service with Qantas and Jetstar.

It is now clear from reports in France that Air France realised the flight had met with disaster about four hours after it left Rio and its operations staff had scanned with dismay the messages confirming catastrophic electrical faults on board.

The airline then worked for hours contacting relatives prior to the scheduled early morning arrival in Paris, knowing of course that there would be no arrival.

All that is known with certainty so far is that there were electrical problems of unprecedented extent followed, obviously, by a loss of control. Whether or not the jet was actually struck by lightning is not known, and jets get struck by lightning with great frequency, and often violence, and are designed to be lightning proof to very rigorous standards.

The same point can be made about the high level turbulence that is both seasonal and widespread in that area. There were many, many flights between South America and Europe in the general area that night.

Something else is thus missing from the picture, and finding physical clues as to what additionally went so terribly wrong could prove incredibly difficult given the search zone.

Speculation so soon after an accident is almost invariably wrong. Without better information it will prove to be somewhere between between hopelessly incomplete and totally wrong.

Qantas, for example, will want to know if the air data inertial reference units or ADIRUs on the Air France jet were of the same type that failed and caused some serious control incidents with its A330-300s, including the incident that hurt more than 100 passengers and crew prior to an emergency landing at Learmonth in Western Australia on 7 October last year.

That incident is still under intensive investigation by the ATSB, Airbus, and French and US authorities. But even if it was the same unit, it would be unduly speculative to read much into that coincidence at this stage.

Operators of A330s will of course not really sleep well until they can be sure of the nature of the mishap, no matter how confident they are of the airliner’s integrity. This would be true of any type of jet. And any competent airline, of which Air France is one.

So now the searching and waiting goes on. A jet with 228 people on board has undoubtedly crashed somewhere in the mid Atlantic. That is all we can say with certainty.

Note: This is a comprehensive update of the earlier post ‘When a jet goes missing’ from late last night. The best general media updates are those generated in Paris where the UK papers keep bureaus.

A time line can be found here at The Guardian online.

Breaking news resources include The Telegraph and in French the preview pages of Le Monde’s on-line subscriber site has updated summaries linked Crikey style to official statements as they are released here. If the permalink are changed, as often happens, edit the link to the home page, such as lemonde.fr or guardianunlimited.co.uk.
Just received at 3 pm unofficial but highly credible advice that the type of ADIRU being investigated in relation to the Qantas A330 accident at Learmonth last October is of different design and manufacture to that fitted to the missing Air France jet.

(Visited 3 times, 1 visits today)


Leave a comment

9 thoughts on “The Air France mid Atlantic mystery will be hard to unravel

  1. Stephen

    Ben’s comments are precisely right: the the endless media speculation right now as to cause is futile. Even finding the plane may take time, although the A330’s airplane management systems should have been communicating to Air France engineering not only the catastrophic failures as they were occurring but also the GPS coordinates of the plane. That will give the searchers a helpful basis point from which to start.

    Most aircraft crashes these days are caused not by a single factor, but by a confluence of events that may have traced its origin to a starting factor. So the real cause of this accident – presuming it is (although that is perhaps a case of holding out hope against all probability) – will likely be very difficult to determine, quite complex, and incredibly time-consuming to identify.

    All this creates real and untold anguish for the hundreds, or thousands, of family, close friends and colleagues of those on board. It may be some time before they have the certainty of knowing not only what happened but where.

    It is a terrible day for the aviation industry, but nothing in comparison to the tragedy unfolding for those whose family members, friends and colleagues were on that plane.

  2. Jackson Harding

    And can we please stop hacks using the following phrases when reporting on aviation incidents:

    “The plane disappeared from radar screens”
    – at 1100 km off the Brazilian coast there was no radar coverage. It might have appeared on a display using FANS1/ADS reports, but not radar

    “Authorities ordered the plane to return/land/et al”
    – No, the Pilot In Command makes those decisions. The “authorities” simply then help him carry out that decision.

    “The terrified passengers….”
    – Nuff said.

  3. Ken Borough


    In May 1979, an American Airlines DC10 (AA191) mysteriously crashed near O’Hare (Chicago) Airport with the loss of all 271 souls on board. Shortly after the accident, the US FAA ordered the grounding of all DC10s until the cause of the accident was known and aircraft rectified so as to ensure that a similar event did not occur. The cause was determined and all affected DC10s were made good. The type continued to operate successfully and safely for many years.

    As the loss of Air France’s A330 is not known and as there have been several instances relating to near catastrophic loss of control, I wonder if it is now time for the authorities to order the grounding of the type until we know why these events are occurring. Admittedly, the A330 has had, till now, operated without hull losses, but can we afford to lose any more lives without knowing what is behind these unfortunate events? On another perspective, is the Airbus family so technically advanced and reliant on automated systems that the pilots who fly them are ‘losing their feel’ (call it situational awareness if you will) for flying?

  4. Ben Sandilands


    The resolution to this may depend on the much tighter search zone Le Monde reports as in effect as daylight returns. It is clear now that the technology and determination to retrieve the flight data recorders from the ocean floor exists, should they be located.

    The AF jet used Honeywell for its ADIRU units, where Qantas for example uses Northrop Grumman units.

    I doubt that the A330s would be grounded on suspicion rather than sudden, urgent evidence of a hitherto unknown issue that can’t be addressed with new SOPs.

    Some similar examples. The Bombardier Q400 is now being described by the FAA as having some significant ice vulnerabilities but the emphasis is on deficient pilot selection and checking at Colgan Aviation which operated the Continental service that crashed at Buffalo. If it was grounded until these icing issues were fixed the consequences for Qantas would be dire. And the type certainly has form when it comes to main gear and nosewheel issues as well.

    The 737 was permitted to continue in service for years after doubts emerged over the loss of rudder control. It took a very long time for Boeing to acknowledge the real reasons for the fatal accidents caused and the hundreds of other loss of control incidents, one of them each occurring on Ansett and Australian 737s too. The rectification program took many years.

    SAAB 340s have a well documented icing sensitivity too. As did the early ATRs, but again they were both allowed to remain in service as the gamble was made that changing the operational rules would overcome the deficiences of the former while redesign of the de icing system for the latter was mandated within a fixed time.

    There were of course two nearly simultaenous issues with the DC-10. One which caused the disastrous depressurisation and floor collapse on the Turkish DC-10 near Paris did not cause a fleet wide grounding while the floor was modified and the cargo hold locks improved, yet the terrible Chicago accident, traced to improper engine handling leading to engine/wing strut failures did cause a grounding. I think the former issue was really the more serious, which forshadowed the section 41 defects on earlier 747s but it was the latter, driven in part by grandstanding that caused the grounding.

    I think the administration of air safety is full of such examples of too much or too little intervention. But I agree whole heartedly for the need for a major effort to understand this A330 crash. If it proves to be something that encompasses all the engine/wing combinations and ADIRU and other systems architectures then drastic action would not only be merited by compelled. Or we might find a lump of space debris in it (something I’m sure will bring down a jet one day) and have gone through a traumatic dislocation of the air transport industry for nothing.

    It is a tough call.

  5. Jackson Harding

    I was living near Chicago at the time and there was no mystery about the crash of AA191. The port engine detached from the wing at or shortly after rotation, slammed into the upper surface of the wing and severed all the hydraulics, not just to that wing but to all controls. A plane spotter in the O’Hare terminal filmed the whole thing and sold the pictures to the newspapers for a kings ransom.

    The engine detached due to cost cutting at AA maintenance (sound familiar?) and the hydraulic failure was attributed to a design flaw. Under competitive pressure from Boeing with the B747 and Lockheed with the L-1011 the Douglas team rushed the DC-10 to the market. They only had three hydraulic systems, and they were all run along the same route, making it possible to breach all of them simultaneously.

  6. George Sydney

    should there be a GPS on the plane?

    Why it is so hard to locate it?

  7. Rodrigo

    Hi Ken,

    Another (very recent) incident involving severe turbulence that was not mentioned anywhere, also affecting an A330, can be found here: http://en.wikipedia.org/wiki/TAM_Flight_8095

    I know it’s all speculation, but it seems to me that severe turbulence is becoming a quite common problem on the A330s…

  8. Ben Sandilands


    I remain puzzled by the wide zone originally defined too because at least one of the automated messages from the jet included GPS coordinates. Even allowing for uncertainty in the trajectory taken by the jet after that message was sent, or swift currents, the search zone seemed very large, until much earlier today, when it was drastically reduced and significant surface debris was found.

  9. mystery of the missing crew

    […] riddle of a missing crew from the Mary Celeste with a trunk dating back to the conquistadors? …The Air France mid Atlantic mystery will be hard to unravel …The Air France mid Atlantic mystery will be hard to unravel … highly credible advice that the type […]

Share this article with a friend

Just fill out the fields below and we'll send your friend a link to this article along with a message from you.

Your details

Your friend's details