AF 447: Air France briefs pilots on pitots and investigation status

A document that appears to be a detailed Air France briefing to its pilot workforce from four days ago (ie just after the BEA press conference last week) has turned up. Seems pretty clear that it’s genuine.

You can see the original in French here, or the same thing here where it was originally posted. And read on for a translation into English. Pprune contributor Squawk Ident has come up with this mostly automated translation below with some tweaks. Unfortunately I’ve got to dash right now and can’t check it though until later, but looks OK. It’s very educational on the pitot tube history.





Roissy, On June 18, 2009


On June 15, a point on the inquiry was made by the Management of thesafety of the company in front of executives PN of the Air Operations.This “N°5″ information” recapitulates all the points approached at thetime of this presentation.


The accident of AF447 took place in the international water; this isthus the State where the aircraft is registered that leads thetechnical inquiry. Four working groups were made to this end by theBureau d’Enquêtes et d’Analyses. Air France contributes to thisinvestigation in bringing its expertise to work with these groups ofwhich each members is engaged to respect the confidentiality ofinformation.

Only the BEA is authorized to communicate on the progress of theinquiry. It will publish a preliminary report before the the 30th ofJune.

In parallel to the technical inquiry and in accordance with the FrenchRight, a judicial enquiry is carried out by the GTA (Gendarmerie duTransport Aérien) under control of an instruction judge of the Court ofParis.

The rules specific to the company envisages in addition theinstallation of an internal inquiry. For this reason the Head officedesignated 5 permanent members including two flight crew staffrepresentatives.

Finally 2 other investigations will be led at the initiative of the CHSCT PNT and PNC.


At the time of the accident the plane was leaving a zone of convectionrelated to intertropical convergence. The satellite photos diffused todate show clouds spreading out at high-altitude, but thesephotographies do not indicate directly the position of the stormy cellsthat the crews of different companies flying on this road met. Acomplementary work of analyses is in progress.

The main informations that we have about the plane result from automatic maintenance messages.

These messages are embodied in the information transmitted by the planewithout intervention of the crew to alow maintenance teams to preparethe interventions upon the arrival of the plane. These messages are noteasily exploitable for an investigation and cannot replace data fromCVR (Cockpit Voice Recorder) and DFDR (Digital Flight Dated Recorder).

More particularly, the format of the messages and logics oftransmission do not allow to reconstitute with certainty the chronologyof appearance of the anomalies in the cockpit.

The majority of these messages are the consequence of the anomalies andthe inconsistencies of aerodynamic speeds; they were detected by theflight control calculators and the automatic pilot. Among these one canbe quoted:

The loss of characteristic speeds

The passage in secondary law of flight (Alternate Law)

These associated anomalies and their consequences are tested during thecertification of the plane; this allows to guarantee that the planeremains controllable in this situation.

All of these messages and their significance will be very certaily published in the preliminary report of the BEA.

At this stage, nothing in these messages allows to establish:

- A loss of power supply,

- A loss of the screens of piloting (PFD and Stand-by horizon),

- A faulty operation of the ADIRU which could have involved an incident of the type of which Qantas company knew recently.

Only one certainty: the sequence of the messages does not allow to explain the accident by itself.


Many erroneous assertions were advanced in connection with theanemometric probes equipping the Airbus fleet. We present to you inchronological order a summary of the reports and especially of theactions carried out by the company since 2001.

August 2001:

Following fluctuations and/or losses of the indications of theaerodynamic speed on A330 and A340 reported by certain companies, theDGAC publishes an “Airworthiness Directive” (AD) by plane type, toimpose the replacement of the probes of Pitot ROSEMOUNT P/N 0851GR,either by probes GOODRICH P/N 0851HL, or by SEXTANT (THALÈS) P/NC16195-AA; this operation having to be finished before the 31 December2003.

Indeed, the official services allot these events to the presence ofcrystals of ice and/or of quantities of water which exceeds thespecifications of the probes Pitot ROSEMOUNT P/N 0851GR. In accordancewith the “Airworthiness Directive”, the model SEXTANT (THALÈS) P/NC16195-AA is installed on the Air France A340 fleet ; as from December2001, Air France receives its first A330 which is origin-equipped ofthe probes SEXTANT (THALÈS) P/N C16195-AA. No event of this type willbe reported on Air France A330 and A340 until May 2008.

September 2007:

Airbus emits technical notes (Service bulletins or SB) which recommend,without being mandatory and outside all context affecting thenavigability of the planes, the replacement of the probes THALÈS P/NC16195-AA installed on fleets A320/A330/A340 by new model probes THALÈSP/N C16195-BA.

It is indicated that this model improves the behaviour of the probe bylimiting the consequences of water ingestion by strong rains and byreducing the risk of icing. After analysis, the Air France technicalteams decide to launch this modification on the A320 fleet which hasincidents with losses of speed indications at low altitude in case ofstrong rains. They decide to replace probes on A330/A340 by the newmodels only in the event of failure, the A330/A340 fleet having then noincidents with loss of speed informations.

May – August 2008:

A first incident of icing of the probes occurs on a Air France A340with temporary loss of speed indications; followed by a second incidentin July 2008 and then of 3 incidents in August 2008, all on A340.Airbus is immediately questioned on the origin of these incidents andto the measures allowing to cure it.

September and October 2008:

Many exchanges with the Airbus technical teams take place. Anexploitation incident is recorded in September and one in October 2008.In six months, 7 incidents were thus recorded, when no incident of thiskind had been reported before.

Airbus answers that:

o The supposed origin of these incidents is a icing by crystal formation in the probes of velocity measurement,

o The new model THALÈS P/N C16195-BA was not conceived to answer theproblem of icing and thus should not bring significant improvement withthis problem,

o The probes installed are in conformity and even exceed the lawfulrequirements in term of airworthiness and of safety of the flights.

November 2008:

Following various follow-up from Air France technical services, Airbuscorrects its September 2007 technical notes in a dated November 12,2008 edition that withdraws any mention of a contribution of the probeTHALÈS P/N C16195-BA improved resistance to icing.

On 24 November2008 a meeting between the technical directions of AirFrance and Airbus discusses lengthly of the incidents with loss ofspeed informations. Air France requires that a technical solution bequickly brought to solve these incidents. Airbus confirms again thatthe origin of these incidents is an icing of the probes, that the lastmodel THALÈS P/N C16195-BA does not treat of the subject and that theprobes installed are in conformity with the airworthiness requirementsand safety of the flights.

February 2009:

Facing our insistence to find a solution, wind-tunnel tests areundertaken by Thalès and Airbus on the behavior of the probe THALÈS P/NC16195-BA.

March 2009:

At the end of March 2009, two new incidents of exploitation arerecorded of which a first on A330. That brings the total number ofincident to 9, including 8 on A340 and one on A330.

Airbus, again requested on several occasions, answers by confirming thepresumption of icing of probes and refers to a procedure of maintenanceand checking of the probes.

April 2009:

In a letter of April 15, 2009, Airbus informs of a new element: theprobe THALÈS P/N C16195-BA does not have vocation to answer the problemof icing of the probes, but the tests carried out by Thalès show abehavior definitely better than that of the former model. Taking intoaccount the limitations of the wind tunnel tests, Airbus suggests anexperimentation on Air France planes to check if an improvement isconfirmed in real situation. Without awaiting this experimentation, AirFrance decides to immediately extend this measure to its entirelong-distance Airbus A330/A340 fleet and to replace the totality of theprobes speed. One internal technical document launching themodification is established dated April 27, 2009. The beginning of theplanes modification is planned as of reception of the parts, at a rateof several planes per week, from June 1.

May 2009:

Air France requires THALÈS to accelerate the delivery schedule of theprobes. Those are delivered starting May 26, 2009 at a rate of 12 tubesof Pitot per week. The program thus could be accelerated.

Since the accident:

Without prejudging of a link between the anemometric probes and theinconsistencies in the indications of speeds presented to the pilots,Air France has decided to accelerate its plan of replacement of theprobes THALÈS P/N C16195-AA on the fleet Airbus. Since last June 12,all A320,A330,A340 Airbus, in exploitation within Air France areequipped with the probes of last generation THALÈS P/N C16195-BA.

Nevertheless in an information bulletin published on June 8, Airbusconfirms that the Airbus world fleet can be exploited with one of the 3types of anemometric probes which equips world fleet, namely THALÈS P/NC16195-AA, THALÈS P/N C16195-BA and GOODRICH P/N 0851HL.


We intend to recall that any step of prevention imposes three requirements;

– a requirement for transparency,

– a requirement for reactivity,

– and a requirement of pro-activity.

Whatever the circumstances, it is advisable to proceed in this way.

This is why, we have decided to keep you regularly informed of thestate of advance of the inquiry and this, in an agreement with the BEAthat remains the only one in charge of the communication of the factualelements.

In addition, we named two staff representatives within the Internal Commission

– Mrs L G for Commercial flight crew

– Mr O R for the Technical Flight crew.

This commission reserves the right to raise recommendations constantly if it feels it need.


2 names deleted.

PNT : Personnel Navigant Technique (Cockpit Crew)

PNC : Personnel Navigant de Cabine (Cabin Crew)

DGAC: Direction de l’Aviation Civile

CHSCT Comité d’Hygiène, de Sécurié et des Conditions de Travail

The members of the CHSCT are elected Unions members. CHSCT is competentfor all matters related to the security, hygiene and work conditions ofthe employees in a given society.

, , , , , , , , , , ,

5 Responses to AF 447: Air France briefs pilots on pitots and investigation status

  1. Michael June 23, 2009 at 1:44 am #

    Why are they asking for confidentiality and later … for transparency? Why not transparency all the way? Would confidentiality results in transparency? This is the typical French hypocrisy: they may be have already recovered the black boxes and hide them …

  2. facts not fiction June 23, 2009 at 5:46 am #

    It was nice to have a up to date report of the investigation! Without any theories or assumptions! Thank you Kieren Daly for this post! Very informative!

    One of the interesting facts is that Air France Had tried very hard to keep their aircraft as safe as possible. with the continued assurance by Airbus until earlier this year of the safety of the old type pitot probes! So far very little has been mentioned about Airbus’ slow response to the 330/340 pitot icing problems! Another issue/question that is kept silent is that the thales pitot old and new could be inferior to the bf goodrich pitot (of which there are no known difficulties). The pitot probes would not be the only thing that could have brung down flight 447 but nevertheless an important part of the investigation!

    thanks again for this informative post

  3. William June 23, 2009 at 12:49 pm #

    Am curious about the value of finding the black boxes at this point.

    I for one support the ongoing effort to locate the CVR/CDR.

    That said…
    Considering the financial liabilities implicit with the revelation of a major aircraft design defect, I am inclined to think that (from a cost perspective at least) there are those individuals who would just as soon hope the recorders remain in the deep.

  4. Dennis Horne June 25, 2009 at 7:38 am #

    Is it not possible to transmit some data from the flight data recorder (FDR) and cockpit voice recorder (CVR) the same way as ´error/trouble´ data is sent to base by ACARS (aircraft communication addressing and reporting system), say, when a pilot pushes a MAYDAY/PAN PAN button?

    In this crash the pilots did not make a m´aidez (= help me) or panne (= breakdown) call, possibly because they were unaware of problems until the aircraft was (almost) out of control.

    My guess is — for what it´s worth, as a private pilot/airline passenger — the aircraft entered a CB (cumulonimbus = thundercloud), had an upset and quickly became uncontrollable; maybe unavoidably maybe not. Would an automatically-triggered system have had time to transmit enough data to be useful?

  5. John Raby July 15, 2009 at 12:35 am #

    An earlier report in the news media suggested the finding of the Captains body indicated he may not have been on the flight deck at the time of the accident.

    Is it possible that the remaining pilots who presumably had less experience may have misjudged the nature of the convective cloud/radar returns and allowed the aircraft to penetrate severe weather.

    This situation is one of the many concerns I personnaly have expresssed with regard to the push for the MPL. Demonstrated skills in a flight simulator and academic knowledge are no substitute for experience.

Leave a Reply