Understanding Air France 447
Page 2
02:10:26 The aircraft’s vertical speed reached its maximum of 6,900 feet per minute, as the airplane passed 36,000 feet with a pitch attitude of 12°.
02:10:27 The flight directors reappeared. This time, they had switched from altitude hold mode to provide guidance for maintaining the current heading and vertical speed, which at the time was a 6,000 feet per minute climb.
First Officer Robert instructed Bonin to watch his speed. Bonin replied, “okay, okay, I’m going back down.”
Some nose down inputs were made, pitch reduced slightly and thrust was reduced. But the pitch attitude remained well above normal and the airplane continued to climb.
Robert said, “Go back down. According to that we’re going up. According to all three you're going up, so go back down.”
Bonin replied, “Okay,” and again Robert said, “You’re at...go back down.”
02:10:32 The pitch attitude was reduced to about 10° and the vertical speed decreased to 4,000 feet per minute as the airplane climbed through 37,000 feet. The angle of attack, typically at 2.5° for cruise, had so far remained below about 4°.
02:10:36 The flight director bars disappeared again. The airspeed on the left side displayed correctly, indicating 223 knots. Fifty knots of airspeed had been lost in the climb from the time the autopilot disconnected 31 seconds earlier. The standby instrument’s airspeed remained abnormally low (the right side display is unknown).
02:10:42 The flight directors reappeared for one second. They provided guidance to maintain a climb rate of 1,400 feet per minute, the current vertical speed.
02:10:47 The flight directors reappeared for 53 seconds (until 2:11:40), again the guidance was to maintain the climb rate of 1,400 feet per minute. The thrust levers were moved back to 2/3 between the idle and climb settings, and the N1 decreased to 85%.
02:10:50 Attempts were made to call the captain back by ringing a call chime that sounds in the crew rest area.
02:10:51 Bonin continued to make nose up inputs and the pitch attitude increased to 16°. The airplane no longer had sufficient energy to maintain the flight director target vertical speed. A series of small roll movements began. Bonin countered each with lateral stick input. The stall warning triggered again and remained on for the next 45 seconds.
02:10:57 The stall angle of attack was reached and the airplane began to shake due to the stall buffet. Thrust was advanced to TOGA (Take Off Go Around - full power). Bonin continued to make nose up inputs as the airplane continued a shallow climb. The engine speed increased to 103% N1.
02:11:00 Robert said, “Above all, try to touch the lateral controls as little as possible, eh.”
02:11:04 The roll oscillations continued - 9° to the right, 16° to the left, and Bonin’s inputs increased, using up to full left and right stick input to counteract them.
02:11:07 The airspeed indication on the standby instrument increased from 129 to 183 knots. The last of the pitot icing had cleared and all three airspeed indications were then displaying correctly. (The normal indicated airspeed for Mach .81 at FL 380 is about 260 knots.)
02:11:10 A maximum altitude of 37,924 feet was reached. The airplane had gained 3,000 feet and lost 90 knots of indicated airspeed in the one minute and seven seconds since the autopilot disconnected.
02:11:20 The stall warning still sounding, Robert said, “But we’ve got the engines, what’s happening?”
02:11:22 Nose up inputs were maintained but the airplane began to descend. The airplane started to bank consistently to the right. Left stick inputs were made which slowed the rolling motion for a few seconds.
02:11:24 Robert again asked, “Do you understand what’s happening or not?”
02:11:30 The bank angle increased to the right. Bonin held full left stick with virtually no effect on the bank angle. The airplane began a right turn that would bring it around 225° before it hit the water three minutes later.
02:11:32 Bonin said, “I don’t have control of the airplane any more now. I don’t have control of the airplane at all.”
02:11:38 Robert said, “Controls to the left,” and made two brief full-left control inputs. Bonin continued to hold his sidestick full left and moved it full aft where it was held for almost 40 seconds. The descent rate increased to 10,000 feet per minute.
The pitch attitude fluctuated between about 10° and 16° nose up, while the angle of attack steadily climbed to 40°.
02:11:42 The captain entered the cockpit. The indicated airspeeds fell below 60 knots which rendered the angle of attack sensors invalid. The stall warning silenced. As the speeds fell below 30 knots the airspeed indication was replaced by a red SPD flag. The airplane was right wing low with varying bank angles up to 45°, even though Bonin’s sidestick was full left in an attempt to control it.
The captain asked, “What are you doing?”
Robert replied, “What’s happening? I don’t know, I don’t know what’s happening.”
Bonin said, “We’re losing control of the airplane there.”
Robert added, “We lost all control of the airplane, we don’t understand anything, we’ve tried everything.”
02:11:47 The pitch attitude porpoised from 8° to 15° nose up. The thrust levers were moved to idle and the nose pitched down to about 11° below the horizon. The vertical speed increased to 15,000 feet per minute. The engine speed decreased to 58% over the next 20 seconds.
02:11:55 The stall warning triggered briefly.
02:12:04 Bonin said, “I have the impression that we have some crazy speed, no, what do you think?” He then deployed the speedbrakes and the nose pitched up level with the horizon. Robert told him not to extend the speedbrakes and then they were retracted. The nose returned to 8° below the horizon, then pitched up again.
The airplane continued to descended with vertical speeds between 10,000 and 15,000 feet per minute as the pitch attitude oscillated between -8° and +15°.
Each time the nose pitched down the angle of attack reduced slightly (though remaining above 30°), the airspeed indication reappeared with speeds between 60 and 80 knots, and the stall warning reactivated.
The bank angle rocked between 20° and 40° to the right, accompanied by corresponding yawing motions and back and forth lateral accelerations. Both pilots and the automatic yaw damper function fought to keep the airplane upright.
02:12:10 The thrust levers were positioned to the CLB (climb power) detent and the N1 increased from 58% to 105% in 10 seconds.
02:12:30 For the first time in a minute, the bank angle passed through wings level as it oscillated left and right, the PF made large sidestick deflections to counter every roll action.
02:12:33 The thrust levers were moved from the climb detent to TOGA.
02:12:44 The airplane descended through 20,000 feet.
02:12:45 The airplane began to roll right again, and the bank angle exceeded 40° within a few seconds. Full left sidestick was again applied and held steady for 20 seconds.
02:13:18 The airplane descended through 10,000 feet. Bonin said, “We’re there, we’re there, we’re passing level one hundred” (10,000 feet). They had already descended 28,000 feet.
Robert said, “Wait, me, I have, I have the controls, eh.” He made a left input for about seven seconds, but Bonin never released his sidestick, and a synthetic voice announced “DUAL INPUT.”
Bonin expressed his shared confusion again, “What is… how come we’re continuing to go down now?”
Robert instructed the captain to see if a reset of the flight control computers could help. The captain remarked that it would not do anything, but he reset primary and secondary flight control computers, PRIM 1 and SEC 1, anyway.
02:13:36 Bonin called out “Nine thousand feet.” His sidestick was still about half way back, the elevator and stabilizer were full nose up (trying to comply with Bonin’s pitch-up order) and the thrust levers remained at TOGA.
Robert said, “Climb climb climb climb!”
Bonin replied, “But I’ve been
at max nose up for a while.”
Robert pushed his sidestick all the way forward for a five seconds, “DUAL INPUT” sounded, and the captain said, “No no no, don’t climb!”
Robert said, “So, go down” and pushed his own stick forward again while the thrust levers were pulled back to climb power. Bonin, however, continued to pull back, and “DUAL INPUT” sounded again.
02:13:45 Robert said, “So give me the controls, the controls to me, controls to me.”
Bonin acknowledged, “Go ahead, you have the controls, we are still in TOGA, eh.”
Robert lowered the nose to 7° below the horizon, the airspeeds displayed once more, along with the stall warning.
02:13:52 Seven seconds later, despite Robert saying he had the controls, Bonin began to pull back on the sidestick again. “DUAL INPUT” was announced, and the airplane started to pitch up.
02:14:05 The captain warned, “Watch out, you’re pitching up there.”
02:14:07 Robert pulled back on his sidestick, and added, “I’m pitching up, I’m pitching up.” The thrust levers were retarded to idle for two seconds and the nose pitched up reaching 16°.
The captain warned, “You’re pitching up.”
Bonin said, “Well, we need to, we are at four thousand feet!”
02:14:16 At 2,500 feet from the surface, the Ground Proximity Warning System (GPWS), announced, “SINK RATE, PULL UP.”
The captain gave permission, “Go on, pull,” and Bonin enthusiastically remarked “Let’s go, pull up, pull up, pull up!”
The thrust levers were moved to TOGA and a nose-up command was applied by both pilots as the airplane pitched up toward 16° again.
02:14:22 Bonin pushed the takeover push-button on his sidestick cutting out Robert’s commands. The synthetic voice announced, “PRIORITY RIGHT,” and a red arrow illuminated on the glare-shield in front of Robert pointed to Bonin, who had taken control.
Bonin, still not having understood what happened said, “[expletive] we’re going to crash.”
The stall warning announced, “STALL, STALL.”
The GPWS warned, “PULL UP!”
“This can’t be true,” said Bonin.
“PULL UP!” shouted the GPWS.
Bonin: “But what’s happening?”
The captain commanded, “Ten degrees pitch attitude,” while the GPWS continued to alert “PULL UP.”
Robert pushed his sidestick forward, but Bonin was holding his takeover push-button down and his sidestick full back. Robert’s locked out inputs had no effect.
02:14:28, quarter past midnight local time. The A330 impacted the water at a 45° angle, 16° nose up in a 5° left bank with a forward ground speed and vertical speed both at 107 knots (123 mph, 10,900 feet per minute). The airplane was crushed from below and shattered into thousands of pieces. Survival was impossible. All 228 people, who came from 32 nations; 126 men, 82 women, 7 children, and one infant were killed.
02:47 - 05:30 Unaware that AF447 had crashed, the four control centers working the flight communicated with each other several times and questioned the estimated times of the flight’s progress, and noted that none had had radar or radio contact since 01:35. Attempts had been made to contact the flight by radio call, relay from other aircraft, ACARS messages (text messages) via the Air France Operations Control Center (which were rejected), and numerous direct SATCOM (Satellite telephone communications) call attempts to the aircraft.
04:59 Air France Operations Control Center contacted the Dakar controller; both noted their inability to contact the airplane.
05:23 The disappearance of the flight was registered and the search and rescue process was triggered.
08:22 - 09:09 The first emergency messages were sent by the Madrid and Senegalese control centers.
11:04 A Brazilian airplane took off to begin search and rescue operations.
Jun 6, 2009: Five days after the accident, the first floating wreckage was found. Over the next 12 days the remains of fifty victims (forty-five passengers, four flight attendants, and the captain), and about 1,000 parts and pieces of floating debris were recovered by French and Brazilian naval forces.
June 10 - July 10, 2009: Undersea searches were made to detect signals transmitted by the two flight recorders' emergency locater beacons, without success. The undersea terrain in the search area was extremely rugged with variations in depth from 2,300-14,000 feet over short distances. The search for the flight recorders was compared to searching for a shoebox in New York City.
July 2009: The first interim report on the accident was published
July - August 2009: Searches for the wreckage were conducted, including attempts to detect the underwater locator beacons (which were damaged in the breakup) with side-scan sonar and a Remotely Operated Vehicle (ROV), without success.
December 2009: The second Interim report on the accident was published.
April - May 2010: The third campaign of underwater searches for the wreckage continued without success, including deep water sonar and remote vehicles.
March 25, 2011: The fourth campaign of undersea searches began.
April 2, 2011: The wreckage was found 12,800 feet deep on a flat area surrounded by terrain. At that depth it is permanently dark with temperatures in the mid 30’s°F (2 °C to 3 °C). It was 6.5 miles from its last position transmitted by ACARS.
April 26 – May 13, 2011: Beginning of the fifth campaign of undersea searches. The BEA team, made up of twelve investigators and experts, devoted itself to the localization and recovery of the flight recorders, mapping the accident site, then recovering airplane parts that were useful to the safety investigation.
May 1-3, 2011: 23 months after the accident, discovery and recovery of the Flight Data Recorder (FDR) and Cockpit Voice Recorder (CVR).
May 13, 2011: Readout and analysis of the flight recorders at the BEA headquarters began.
May 21 – June 3, 2011: Continuation of undersea operations. The ship and equipment were made available by the BEA to representatives of the judicial authorities, which made it possible for them to recover the remains of one hundred and three victims.
July 2011: The third interim report on the accident was published.
July 2012: The final accident report was issued by the BEA.
How could one of the world’s most modern airliners, with an excellent safety record and qualified crew, fall out of the sky and disappear?
Like any aircraft accident, Air France 447’s crash was a result of a chain of events and circumstances. Remove any one of these, and the accident would likely not have happened.
The crew was unable to maintain control, stalled the airplane, and crashed within four and a half minutes of autopilot disconnection. Communication issues delayed the discovery that the airplane was lost by several hours, and added two years to the search for the wreckage.
Amazingly, no component on the airplane actually failed. But, the pitot tubes were overcome by conditions that were not anticipated, causing a loss of reliable airspeed data. A cascade of system disconnections, downgrades, and human errors followed, and the loss of the flight in this manner shocked the airline industry.
Weather, design factors, pilot competence, training, and human factors all played a part.
To understand what really happened requires an understanding of the human element, the weather, and the machine. This is what I will provide.
Chapter 2: The Flight Crew
There were three pilots: Captain Mark Dubois, First Officer David Robert, and First Officer Pierre-Cedric Bonin.
58-year-old Captain Marc Dubois had joined Air France in 1988 and at the time of the accident had approximately 11,000 flight hours, including 1,700 hours on the Airbus A330, all as captain.
He earned his Private license in 1974. While working as a flight attendant for Air France between 1976 and 1982, he earned his commercial certificate (1977), instrument rating (1978), instructor rating (1979), mountain airport rating (1980), and took the written tests for fir
st class professional and airline transport ratings.
He flew a range of Cessna, Piper, and Beech aircraft and several models of light twin turboprops.
In 1982 he earned his 1st class professional pilot’s license2, and worked as a demonstration pilot for Intra Avia Service Company in the first months of 1983. He worked for various companies until August of 1984, and was an independent pilot until February of 1988 when he was hired by Air Inter at age 37.
While at Air Inter he received a Caravelle XII and A300 type ratings, his airline transport pilot certificate, and 1st class pilot instructor rating.
In March/April 1997 he received his A320 type rating as Air Inter and Air France merged. In June 1998, he received a 737-200 type rating and at the age of 47 was first appointed as captain. A new A320 type rating (within Air France) was issued in May 2001.
In October 2006, at age 55, he received his A330 type rating and had an unsatisfactory line test flight in January 2007. His training was extended, and a month later he passed.
An A340 rating was added in August 2007 with a captain checkout complete in September.
Captain Mark Dubois had flown 16 trips to South America since he arrived in the A330/A340 division nearly two years earlier.
37-year-old David Robert had over 6,500 total hours, almost 4,500 hours of which was in the A330.
He earned his basic license and passed his airline pilot theory test in 1992 at 20 years old. In 1993 he received his professional pilot’s license and multi-engine instrument rating. At this time Air France stopped pilot hiring and drew up a waiting list due to economic issues.
In 1997 he trained as an air traffic control engineer at the French civil aviation university, ENAC. In August 1997, Air France called and he delayed joining in order to finish his training at ENAC. In July 1998, age 26, he started training at Air France and earned an A320 type rating in November of that year. His airline transport license was issued in April 2001.
He received his A340 type rating in February 2002, followed by an A330 type rating in October.