Below are the most significant sections contained in the subject interim report.  This posting does not contain analysis of Section 1.16.2 ACARS messages.  ACARS will be addressed in a separate posting.

A complete analysis between  the French and English versions are required due to the significant difference in the file size and number of pages between the two translations from BEA.

Keep in mind the French version is the official reference copy.  Double check everything BEA publishes!  Sorry, just don’t trust them…


Note: “This interim report has been translated and published by the BEA to make its reading easier for English-speaking people.  As accurate as the translation may be, the original text in French should be considered as the work of reference.”

BEA REPORT (FRENCH) pdf – 128 pages

ENGLISH VERSION pdf – 72 pages

1.1 History of Flight

On Sunday 31 May 2009, the Airbus A330-203 registered F-GZCP operated by Air France was programmed to perform scheduled flight AF447 between Rio de Janeiro Galeão and Paris Charles de Gaulle. Twelve crew members (3 flight crew, 9 cabin crew) and 216 passengers were on board. The departure was planned for 22 h 00.

Towards 22 h 10, the crew was cleared to start up engines and leave its parking position. Takeoff took place at 22 h 29.
The takeoff weight was de 232.8t (for a MTOW of 233 t), including 70.4 tonnes of fuel.

The crew contacted, successively:

• RIO DE JANEIRO approach control,
• the CURITIBA ATC centre, which cleared it to climb to FL350 at 22 h 45 min 26 s,
• the BRASILIA ATC centre at 22 h 55 min 41 s,
• the RECIFE ATC centre at 23 h 19 min 27 s, the airplane being stable at FL350,
• the ATLANTICO ATC centre on HF at 1 h 33 min 25 s.

At 1 h 35 min 15 s, the crew informed the ATLANTICO controller that they had passed the INTOL2 point then announced the following estimated times: SALPU at 1 h 48 min then ORARO at 2 h 00. They also transmitted their SELCAL code and a test was performed.

At 1 h 35 min 46 s, the controller asked them to maintain FL350 and to give their estimated time at the TASIL point.
Between 1 h 35 min 53 s and 1 h 36 min 14 s, the controller asked the crew three times for its estimated time at the TASIL point. There was no further contact with the crew.


1.5 Personnel Information

Given the length of the planned flight and in compliance with the Air France operations manual and with the regulations in force, the flight crew was reinforced.  The regulation defines reinforced crew as follows:

Flight crew where the number of members is greater than the minimum number required for the operation of the airplane and within which each member of the flight crew is able to leave his or her station and be replaced by another member of the flight crew with the appropriate qualification (4).

The airline’s procedures (5) specify that to be a back-up duty pilot, a crew member must have the same rating as the crew member that he or she is backing up and, in addition, during the captain’s rest period, a pilot with the same license as the captain must be at the controls.

From the current state of the information gathered, it is not possible to determine the composition of the flight crew on duty at the time of the event.

Note: the crew left Paris on Thursday 28 May 2009 in the morning and arrived in Rio de Janeiro in the evening of the same day.

1.6.3 Weight and balance

The aircraft left the gate with a calculated weight of 233,257 kg. The estimated takeoff weight was 232,757 kg (11), for a maximum authorised takeoff weight of 233 t. This takeoff weight broke down as follows:

• empty weight in operating condition: 126,010 kg,

• passenger weight: 17,615 kg (126 men, 82 women, 7 children and one baby (12)),

• weight in cargo compartment (freight and luggage): 18,732 kg,

• fuel weight: 70,400 kg.

(11) A quantity of 500 kg of fuel had been taken into account for taxiing between the ramp and takeoff brake-release.

(14) Air France applies a standard weight of 91 kg for a man, 72 kg for a woman and 35 kg for a child, which is compatible with the European regulations. Elaboration of the speed information

The speed is deduced from the measurement of two pressures:

• total pressure (Pt), by means of an instrument called a Pitot probe,
• static pressure (Ps), by means of a static pressure pick-off.

Probes. The Airbus A330 has three Pitot probes (see below) and six static pressure pick-offs. These probes are fitted with drains allowing the removal of water, and with an electrical heating system designed to prevent them from icing up.

Pitot_locationsPosition of the Pitot probes on the Airbus A330

pitot_pic1Pitot probe (with protection caps)

The pneumatic measurements are converted into electrical signals by eight ADMs and delivered to the calculators in that form.

Speed calculation by the ADR

The CAS and Mach number are the main items of speed information used by the pilots and the systems to control the aircraft. These parameters are elaborated by three calculators, called ADIRU, each consisting of:

• an ADR module which calculates the aerodynamic parameters, specifically the CAS
and the Mach,
• an IR module that provides the parameters delivered by the inertial units, such as
ground speed and attitudes.

The ADRs use the measured pressure values to calculate the CAS and the Mach. The diagram below shows in simplified form the overall architecture of the airspeed measuring system:


There are therefore three speed information elaboration systems that function independently of each other. The “Captain” probes feed ADR 1, the “First Officer” probes feed ADR 2 and the “Standby” probes feed ADR 3.

Only the standby instruments such as the ISIS elaborate their speed and altitude information directly from the pneumatic inputs (“standby” probes), without this being processed by an ADM or ADR. Systems that use the speed information

The speeds calculated by the ADRs are used, in particular, by the following systems:

• fly-by-wire controls system,
• engine management system,
• flight management and guidance system,
• ground proximity warning system,
• transponder,
• slats and flaps control system.

1.6.7 Checks and maintenance of the Pitot probes

The Pitot probes and maintenance actions are described in the operator’s maintenance manual.

The Pitot probes are subject to a daily visual inspection by a mechanic, who checks their
general condition. The crew performs the same type of check before each flight. During Type C checks, the following operations are performed on the Pitot probes:

• cleaning of the complete probe using compressed air (“blowing” operation),
• cleaning of the drains with a specific tool,
• test and check of probe heating by the standby electrical power supply system,
• check of the sealing of the circuits.

In the case of speed inconsistencies being reported by the crew, corrective actions are the same as those in the Type C checks.

1.6.8 ACARS communication system

The ACARS system, integrated in the ATSU on Air France’s Airbus A330, is used to transmit non-vocal messages between an aircraft and the ground by VHF or satellite communication. It can be used in particular by operators to transmit information in real time (meteorological data, flight progress information, etc.).

There are three major categories of message that can be transmitted:

• non-vocal (ATC) communication messages with an air traffic control centre (CPDLC in particular),
• operational communication messages (AOC) with the operator’s operations centre,
• maintenance messages, exclusively from the aircraft to the maintenance centre.

ACARS messages are transmitted as a priority by VHF or by satellite if VHF is unavailable. They pass through an ACARS service provider’s server (ARINC or SITA) before arriving at the operator’s centre.

ACARS NetworkInformation relative to the network (processing by the ground station and/or service provider’s server) and information relative to the satellite (type of message, channel used, etc.) is added to the useful message.

The ATC and operational messages are generated by the ATSU. The maintenance messages are generated by the CMC and transferred to the ATSU before being transmitted. Of these three types of message, it is the ATC messages that have the highest priority.

Note: the operator can configure part of the ATSU (the AOC part in particular) so as to filter the maintenance messages transmitted or to send specific types of information relative to the flight. F-GZCP was programmed to automatically transmit its position approximately every ten minutes.

1.6.9 Centralised Maintenance System

The aircraft has a Centralised Maintenance System (CMS) whose role is to facilitate maintenance operations. It acquires and saves certain messages transmitted by the Flight Warning System (FWS) or the test functions integrated in the systems (BITE). It generates maintenance reports, including the CFRs (when the aircraft is in flight) and PFRs (once the aircraft has landed). The CMS groups together two Central Maintenance Calculators (CMC) and the various systems’ integrated test functions. Flight Reports (CFR and PFR)

The CFR is made up of all the maintenance messages generated on-board an aircraft in flight. Once on the ground, the system generates a more elaborate report, called the PFR. A maintenance-related message may be:

• a fault message reflecting the triggering of a monitoring process which may inform on the status or functioning of the system concerned,
• a cockpit effect message reflecting an indication presented in the cockpit (for example an ECAM message or a flag).

Note: the term ‘fault’ means the triggering of a monitoring process that may, in certain cases, refer to a failure.

There are three classes of fault messages:

• class 1: these have operational consequences and are accompanied by at least one cockpit effect (not necessarily recorded in the CFR),
• class 2: these do not have any operational consequences; they are accompanied by one or more “MAINTENANCE STATUS” messages that are only brought to the attention of the crew via the ECAM’s STATUS page once on the ground,
• class 3: these messages can only be consulted on the ground, by using each calculator’s BITE systems; these messages are therefore not included in the CFR or PFR.

Unlike the CFR, the PFR presents correlations between the fault and cockpit effect messages. The relative positions of the messages in a CFR and in the corresponding PFR may therefore be different. Maintenance message acquisition by the CMC

The CMC acquires certain ECAM messages from the FWC, in the order that the latter transmits them. This is not necessarily the order in which those messages were displayed on the Engine Warning Display (E/WD). Up to one hundred messages can be acquired in one second. The messages indicating a flag or an advisory are received from the DMCs and must be confirmed for between 2.4 and 3 seconds in order to be acquired. They are timed once this confirmation has been made.

DisplaysThe fault messages are received from the BITE of the various systems. When a system detects a fault, it transmits a fault message to the CMC containing:

• the ATA code (six digits) of the equipment concerned by the fault,
• the name of the system that detected the problem, called the source,
• the message’s class (1 or 2),
• a message describing the fault,
• information on whether the fault is of a lasting (“HARD”) nature or not (“INTERMITTENT”).

When the CMC receives this type of message, it opens a one-minute correlation window corresponding to the first three or four digits of the ATA code. During this period, all the fault messages that may have been received including those same three or four first ATA code digits are grouped together. Once the minute has elapsed, the CMC closes the correlation window and applies the priority rules between the correlated messages in order to generate an overall message:

• class 3 messages are not taken into account,
• a class 1 message takes priority over a class 2 message,
• for two messages of the same class, a message reporting an internal fault (the system detects a fault in its own operation) takes priority over one reporting an external fault (the system observes a fault in another system),
• and as a last resort, the oldest message takes priority.

The overall message generated then contains the priority message’s information (ATA code, source, etc.), to which is added the list of names of the other systems called identifiers that have generated correlated messages. It is this overall message that then appears in the CFR or PFR. So, no information on the descriptions of the messages transmitted by the identifiers is given; only the description of the priority message is saved. Furthermore, if the source or one of the identifiers has transmitted a class 2 message, its name is preceded by an asterisk (*).

The following theoretical sequence is given as an example:


All the messages are timed to the nearest minute. The timing of an ECAM message consists of the time of its acquisition by the CMC, and that of a fault message is the time at which the correlation window opened. It is therefore possible in a CFR to find an ECAM message preceding a fault message that is nevertheless timed one minute before it. So, for example:

**SEE PDF FILE** Transmission of maintenance messages by the CMC

In order to transmit the messages by ACARS, the CMC sends them to the ATSU. ECAM messages are transmitted in real time as soon as they are acquired. Flag or advisory messages are transmitted as soon as they have been confirmed.

Fault messages are transmitted as soon as the corresponding correlation window is closed.

1.6.10 Radio communications system

The Airbus A330’s radio communications system consists of the following equipment:

• VHF and HF transmitters-receivers
• RMPs,
• audio integration systems: ACP and AMU.

Each VHF / HF transmitter-receiver can be controlled by one of the three RMPs. VHF equipment

There are three identical VHF communication systems installed. Each system includes:

• a transmitter-receiver in the avionics compartment,
• an antenna on the upper part of the fuselage for VHF 1 and VHF 3, and on the lower part of the fuselage for VHF 2. HF equipment

The aircraft has two HF communication systems. Each system includes:

• a transmitter-receiver in the avionics compartment,
• an antenna coupler situated at the root of the stabiliser,
• a shared antenna integrated in the leading edge of the fin.

Since the HF system has a range of several thousand kilometres, a large number of communications are received.

Furthermore, the quality of the transmissions may sometimes be poor. Communications may also be interrupted due to natural phenomena.

A SELCAL call system, transmitting a light and sound signal, informs the crew when a ground station is attempting to contact them.

1.18.4 Procedures to be applied in case an unreliable speed indication is detected On the date of the accident, the operator’s procedures mention that the following actions must be carried out from memory by the crew when they have any doubt concerning the reliability of a speed indication and when control of the flight is “affected dangerously”:


If conduct of the flight does not seem to be affected dangerously, the crew must apply the UNRELIABLE SPEED INDICATION / ADR CHECK procedure (see appendix 9).

For information, the “Memory Item” in the Airbus QRH relative to the same fault is shown below in the version in force on the date of the accident.


1.12 Localisation of the bodies and aircraft parts

The French and Brazilian navies found debris belonging to the aircraft from 6 June onwards. All the debris known to the BEA was referenced in a database. By 26 June, this database included 640 items.

Whenever the information is available, the position, the date and the time of their recovery are indicated. The chart below shows the position of all of the bodies and debris thus georeferenced.

The bodies are represented by red circles and the debris by white circles. The tail fin (vertical stabiliser), found on 7 June is represented by a yellow diamond.


The timeline of the recovery of the bodies and debris from the aircraft found between 6 June and 18 June, 2009 and known to the BEA on 26 June, 2009, can be found in appendix 4.

1.12.2 Identification of the items recovered

The identification of the debris shows that it consists mainly of light items belonging to the cabin fittings and holds (bulkheads, galley, ceiling or floor panels, seats, overhead baggage bins, cabin and hold lining).

Approximately thirty pieces are external parts of the plane (vertical stabiliser, pieces of the radome, the engine cowl, the under belly fairing, the flap actuator fairing, the trimmable horizontal stabiliser and the secondary control surfaces).

The identified debris thus comes from all the areas of the plane.

An ELT distress beacon with manual tripping was also recovered. This had not been actuated. Its switch was found in the “OFF” position.

Visual inspection

A first visual inspection brought to light the following.

The tail fin was damaged during its recovery and transport but the photographs available made it possible to identify the damage that was not the result of the accident. The middle and rear fasteners with the related fragments of the fuselage hoop frames were present in the fin base. The distortions of the frames showed that they broke during a forward motion with a slight twisting component towards the left.

Part of the radome was found, representing approximately a fifth of its circumference along its upper part.


The galley, identified as G2, located at the level of door 2 on the right-hand side, was not very distorted. Baskets and racks were compressed in the lower part of both galley carts.



The distortions observed in the metal vertical reinforcements of a toilet door showed evidence of great compressive forces.


Fragments of the walls of the flight crew rest module were crumpled and those of the ceiling were deformed downwards. The floor was curved under the effect of a strong upward pressure from below. The connecting brackets between the floor and the walls were bent backwards.


1.12.4 Summary of visual examination

Observations of the tail fin and on the parts from the passenger (galley, toilet door, crew rest module) showed that the airplane had likely struck the surface of the water in a straight line, with a high rate vertical acceleration.


1.13 Medical and Pathological Information

Sailors from the Frigate Ventôse recovered about thirty bodies. A visual examination of the bodies showed that they were clothed and relatively well preserved. All of them were handed over to the Brazilian Navy to be transferred to the Recife morgue.

At this stage of the investigation, the BEA has not yet had access to the autopsy data.


On the basis of the first factual elements gathered in the course of the investigation, the following facts have been established:

  • • The crew possessed the licenses and ratings required to undertake the flight,
  • • The airplane possessed a valid Certificate of Airworthiness, and had been maintained in accordance with the regulations,
  • • the airplane had taken off from Rio de Janeiro without any known technical problems, except on one of the three radio handling panels,
  • • no problems were indicated by the crew to Air France or during contacts with the Brazilian controllers,
  • • no distress messages were received by the control centres or by other airplanes,
  • • there were no satellite telephone communications between the airplane and the ground,
  • • the last radio exchange between the crew and Brazilian ATC occurred at 1 h 35 min

15 s. The airplane arrived at the edge of radar range of the Brazilian control centres,

  • • at 2 h 01, the crew tried, without success for the third time, to connect to the Dakar ATC ADS-C system,
  • • up to the last automatic position point, received at 2 h 10 min 35 s, the flight had followed the route indicated in the flight plan,
  • • the meteorological situation was typical of that encountered in the month of June in the inter-tropical convergence zone,
  • • there were powerful cumulonimbus clusters on the route of AF447. Some of them could have been the centre of some notable turbulence,
  • • several airplanes that were flying before and after AF 447, at about the same altitude,

altered their routes in order to avoid cloud masses,

  • • twenty-four automatic maintenance messages were received between 2 h 10 and 2 h 15 via the ACARS system. These messages show inconsistency between the measured speeds as well as the associated consequences,
  • before 2 h 10, no maintenance messages had been received from AF 447, with the exception of two messages relating to the configuration of the toilets,
  • • the operator’s and the manufacturer’s procedures mention actions to be undertaken by the crew when they have doubts as to the speed indications,
  • • the last ACARS message was received towards 2 h 14 min 28 s,
  • • the flight was not transferred between the Brazilian and Senegalese control centres, 69
  • • between 8 h and 8 h 30, the first emergency alert messages were sent by the Madrid and Brest control centres,
  • • the first bodies and airplane parts were found on 6 June,
  • • the elements identified came from all areas of the airplane,
  • • visual examination showed that the airplane was not destroyed in flight ; it appears to have struck the surface of the sea in a straight line with high vertical acceleration.