Air France 447 wreckage found in Atlantic

Crash: Air France A332 over Atlantic on Jun 1st 2009, aircraft entered high altitude stall and impacted ocean
By Simon Hradecky, created Friday, Jul 29th 2011 16:35Z, last updated Saturday, Jul 30th 2011 20:27Z

The BEA have released their third preliminary report in French. While it doesn't appear there have been additions to the factual part already released in report #2, report number #3 contains first analysis of the crash.

In there the BEA reported, that the aircraft departed Rio de Janeiro with the captain in the left hand seat, the first first officer in the right hand seat and the second first officer in the observer seat.

At 01:55Z the captain left his seat in order to take his rest, the second first officer took the captain's seat. The first first officer (in the right hand seat) became pilot flying (PF), the second first officer pilot monitoring (PNF).

The captain was observing the briefing the two first officers conducted. The PF pointed out there was a little bit of turbulence ahead according to weather radar's image but they couldn't climb, pointing out the FMS indication of maximum possible cruise level of 375. Subsequently the captain left without leaving any instructions with the first officers, especially not assigning tasks. This absence of formal framework later led to a breakdown of optimal crew cooperation between the two first officers.

The turbulence increased, at 02:08Z the PF decided to turn left by 12 degrees selecting heading mode into the autopilot. The PF remarked he would prefer to climb to FL360. Intensive weather radar observations followed however were interrupted by a sensation of quick temperature increase and the appearance of an odour that distracted the crew for more than a minute. The PNF finally identified the smell as ozone.

Discussions in the cockpit did not suggest any malfunctions.

At 02:09:46Z the background noise on the cockpit voice recorder significantly changed, the BEA analysed the sound was consistent with the characteristics of presence of ice crystals, the change in sounds getting the attention of the first officers who had not experienced this before. The PNF takes action and suggests to slow down to Mach 0.8 and engages engine anti-ice on.

At 02:10:05Z a sharp drop of airspeed indications occurred likely due to obstruction of the pitot probes by ice crystals. The autopilot and autothrust systems disengaged as result. The presence of turbulence led to the aircraft rolling right to a bank angle of 8 degrees, the nose pitches up to 11 degrees nose up within 10 seconds. The PF announces "I have control" and makes a quick left nose up input almost to the mechanical stops of the side stick. The nose pitches further up and at 02:10:10Z the stall warning activates.

Between 02:10:18Z and 02:10:25Z the PNF reads various ECAM messages, then attracts the attention of the PF to the loss of speed. At that time both left hand and ISIS airspeeds were below 100 knots, the aircraft was climbing by 7000 feet/min and the aircraft rolls left and right, with the bank angle remaining between -10 and +10 degrees of bank, as result of PF inputs. Both first officer recognized the loss of airspeed indications, however none called for the unreliable airspeed procedures. Comparing the three airspeed indications the PNF urges the PF multiple times to lower the nose. The aircraft was climbing through FL370 and continued to climb.

At 02:10:34Z the left airspeed sharply rose to 215 KIAS and became valid again, the speed indication of the stand by instrument (ISIS) remained wrong. The BEA analysed that the speed was 60 knots from the point of autopilot disengaging until that time, the 60 knots loss of speed being consistent with the aircraft climbing 2000 feet in that configuration.

At 02:10:47Z the thrust levers were moved to about 2/3 of travel resulting in 85% N1. The climb rate reduced to 100 feet/min, the angle of attack reduced to about 3 degrees and the roll is controlled. The first officers now attempt to call the captain back to the cockpit. The BEA analyses that after a quick trim and levelling off as result of PF inputs the trajectory of the aircraft appeared controlled again, however the initial increase in altitude was excessive and should have been pointed out by the PNF. However, none of the pilots called out speeds, vertical speeds, pitch angles or altitude.

At 02:10:51Z the stall warning activated again, the pitch angle had increased to about 6 degrees nose up. As result of PF inputs the pitch angle increased from 6 to 13 degrees and the angle of attack increased to 10 degrees. Rapid onset of buffeting occurred. 5 seconds later the thrust levers are placed into the TOGA detent. The BEA analysed that was about the point where the aircraft went out of control.

The PF maintains inputs to keep the nose up between 11 and 23 degrees nose up with the angle of attack being betwen +11 and +18 degrees.

At 02:11:07Z the ISIS airspeed becomes valid at 183 KIAS, and all three indicated airspeeds agree. The PF selected ADR3 as input for his primary flight display. The vertical speed drops through 0 and becomes highly negative (around 4000 feet/minute sink rate), the airspeed reduces further to 160 KIAS.

At 02:11:20Z the PF states twice he has lost control of the aircraft.

At 02:11:37Z the PNF pushes the priority button and provides a large left input in reaction to a high right bank, calling "controls on the left". He almost instantly released the priority button and ceased control inputs again without required calls.

At 02:11:42Z the captain entered the cockpit, just before the stall warning stops, the airplane was at 35800 feet MSL, 9100 feet/min sink rate, airspeeds below 100 KIAS, pitch 12 degrees nose up and engines at 102% N1. Neither first officer provides an account to the captain what is happening. The BEA analysed that when the stall warning stopped all three AoA computations had become invalid due to too low an airspeed. Airspeeds are no longer displayed on both PFDs.

At 02:12:04Z the PF remarked he believed they were in overspeed, the BEA assumes because of the excessive background noise in the cockpit. None of the two other pilots supported that hypothesis which was in contradiction to instrument indications, high pitch angle and high rate of descent.

Until impact now the stall warning activates when the nose is lowered and silences when the pitch angle increases. FDR data suggest that as soon as the stall warning activates the pilots react with nose up inputs which cause the stall warning to silence again.

The BEA analysed that none of the three pilots ever worked out which indications to trust and which not.

The BEA analysed that both first officers had been trained unreliable airspeed emergency maneouvers at low altitude which required them to disengage flight directors and autoflight systems, then adopt an attitude between 10 and 15 degrees nose up. At altitude however the stall warning would activate at an angle of attack just above +4 degrees.

The BEA listed following findings:

- The Captain’s departure occurred without clear operational instructions

- The crew composition was in accordance with the operator’s procedures

- There was no explicit task-sharing between the two copilots

- The weight and balance of the airplane were within operational limits

- The crew had noticed returns on the weather radar

- The crew made a heading change of 12° to the left of its route

- The AP disconnected while the airplane was flying at upper limit of a slightly turbulent cloud layer

- There was an inconsistency between the measured speeds, likely as a result of the obstruction of the Pitot probes in an ice crystal environment

- At the time of the autopilot disconnection, the Captain was resting

- Even though they identified and announced the loss of the speed indications, neither of the two copilots called the procedure "Unreliable IAS"

- The copilots had received no high altitude training for the "Unreliable IAS" procedure and manual aircraft handling

- No standard callouts regarding the differences in pitch attitude and vertical speed were made

- There is no CRM training for a crew made up of two copilots in a situation with a relief Captain

- The speed displayed on the left PFD remained invalid for 29 seconds

- The approach to stall was characterised by the triggering of the warning, then the appearance of buffet

- A short time after the triggering of the stall warning, the PF applied TO/GA thrust and made a nose-up input

- In less than one minute after the disconnection of the autopilot, the airplane was outside its flight envelope following the manual inputs that were mainly nose-up

- Until the airplane was outside its flight envelope, the airplane’s longitudinal movements were consistent with the position of the flight control surfaces

- Neither of the pilots made any reference to the stall warning

- Neither of the pilots formally identified the stall situation

- The invalidity of the speed displayed on the ISIS lasted 54 seconds

- The Captain came back into the cockpit about 1 min 30 after the autopilot disconnection

- The angle of attack is the parameter that enables the stall warning to be triggered; if the angle of attack values become invalid, the stall warning stops

- By design, when the speed measurements were lower than 60 kts, the 3 angle of attack values became invalid

- Each time the stall warning was triggered, the angle of attack exceeded its theoretical trigger value

- The stall warning was triggered continuously for 54 seconds

- The airplane’s angle of attack was not directly displayed to the pilots

- Throughout the flight, the movements of the elevator and the THS were consistent with the pilot’s inputs

- The engines were working and always responded to the crew’s inputs

- No announcement was made to the passengers

Air France commented: "It should be noted that the misleading stopping and starting of the stall warning alarm, contradicting the actual state of the aircraft, greatly contributed to the crew’s difficulty in analyzing the situation. During this time, the crew, comprising both First Officers and the Captain, showed an unfailing professional attitude, remaining committed to their task to the very end. Air France pays tribute to the courage and determination they showed in such extreme conditions. At this stage, there is no reason to question the crew’s technical skills."

The French BEA released new safety recommendations mainly recommending to introduce an AoA indication on the cockpit instruments, introducing additional training for manual aircraft control at altitude, additional definitions for the role of a relief captain to ensure proper task sharing in the cockpit as well as recommendations regarding flight data recorders.

Translated Transcript (PIC: Captain, PF: Pilot Flying, PNF: Pilot monitoring):
Time Source
02:00:33 PF Well a bt of turbulence which you just saw… we will … we should find it again before, in fact we are in the cloud layer unfortunately as we can not climb too much for the moment because of the temperatue which is sinking less than expected. This is reducing the REC MAX for us a little lower to go for 37 (FL370)
02:06:44 PF The ITC, there it is between SALPU and TASIL
02:06:54 PF Minus 42, we are not going to use the anti ice it is still there
02:07 PF You see we are really on the limit of the cloud layer.
02:08:07 PNF Can you maybe turn a bit to the left? I agree that we are in manual, right?
02:08:19 PNF What I call manual means we are not in managed mode (Nav)
02:09:54 PNF Here, I’ll reduce the speed a bit for you.
02:10:03 PF Do you want to switch to Ignition Start?
02:10:06 PF I have control
02:10:09 PF Ignition Start
02:10:11 PNF What is this?
02:10:14 PF We don’t have a good… We don’t have a good indication of ….
02:10:17 PNF We have lost the speeds, engine thrust A T H R engine lever thrust
02:10:18 PF … speed
02:10:22 PNF Alternate Law Protections
02:10:24 PNF Wait, we are about to loose
02:10:25 PNF Wing Anti-Ice
02:10:27 PNF Watch your speed, Watch your speed
PF Ok, ok, I will descend back
PNF You are stabilizing
PF Yeah
PNF You are descending back
02:10:33 PNF According to the three you are climbing, now you are descending.
02:10:35 PF Agreed
02:10:36 PNF You are at… descend back
PF It is going, we are descending back
02:10:39 PNF I’ll put you on A T T (*) (Selecteur ATT /HDG is put in position F/O on 3)
02:10:42 PF We are, yes we are in climb
02:10:49 PNF Where is he, eh?
02:10:56 PF TOGA
02:11 PNF Try to use the lateral controls as few as possible hey!
02:11:03 PF I am in TOGA
02:11:06 PNF … is he coming or not
02:11:21 PNF We have certainly the engines, what is happening?
02:11:32 PF I don’t have control of the aircraft, I don’t have control of the aircraft at all
02:11:38 PNF Command to the left (taking control)
02:11:41 PF I have the impression that we have speed…. (we are in speed)
02:11:43 PIC Hey, what are you doing?
PNF What is happening, I don’t know, I don’t know what is happening
02:11:53 PIC Ok, take, take this
02:11:58 PF have a problem, I have no more vertical speed here
PIC Agreed
PF I have no indication at all.
02:12:04 PF I have the impression that we have a crazy speed, no, what do you think?
[Speedbrakes are deployed]
02:12:07 PNF NO, don’t extend them! Certainly not.
02:12:13 PNF What do you think, what doyou think, what do we have to do?
02:12:15 PIC I don’t know, it descends.
02:12:10 PF Here, that is good, we have wings level, no it doesn’t want.
PIC Wings to level, the horizon, the backup horizon.
PNF The horizon (secondary)
02:12:26 PNF The speed?
02:12:27 PNF You are climbing,
VS Stall Stall
PNF You are descending, descending descending
02:12:30 PF I am descending?
PNF Descend!
02:12:32 PIC No, you are climbing
02:12:33 PF Here, I am climbing, okay, right so lets descend (or okay we are descending) (unclear)
02:12:42 PF OK, we are in TOGA
02:12:42 PF On the altitude where are we?
02:12:44 PIC this is not possible
02:12:45 PF On alti(tude) we are where?
02:12:45 PNF What do you mean on altitude?
PF Yes, yes, yes, I am descending there, no?
PNF Yes, you are descending.
PIC Hey, you are in…. put the wings level,
PNF Put the wings level!
PF That is what I am trying to do
PIC Put the wings level
02:12:59 PF I am at the limit of, with the warping
PIC The rudder
02:13:25 PF What, how is it that we are continuing to descend at the limit there?
02:13:28 PNF Try to find what you can do with the controls up there, The primaries e.t.c.
02:13:32 PF At level 100
02:13:36 PF 9000 ft
02:13:38 PIC Carefull with the rudder!
02:13:39 PNF Climb, climb. Climb, climb
02:13:40 PF But I am at the limit of the nose since a while
PIC No, no, no, don’t climb
PNF So descend
02:13:45 PNF So, give me the controls, to me the controls.
PF Go ahead, you have the controls, we are still on TOGA
02:14:05 PIC Careful, you are nose high (cabres? )
PNF I am nose high?
PF Well, we need to, we are at 4000 ft
02:14:18 PIC Go, Pull
PF Go, Pull pullpull
02:14:26 PIC Ten degrees pitch

Editorial note: at 02:12:04Z the transcript mentions the speedbrakes have been deployed, the FDR graphics as well as the remainder of the report do not mention at all whether the speedbrakes have been extended or not.

Andrew33 wrote: »

^^^^
A harrowing read indeed.