by Palmer, Bill
At this same time (02:11:45) the captain entered the cockpit. He was not informed of the nearly 3,000 foot climb and he arrived to find the airplane descending through its original cruise altitude of 35,000 feet at 10,000 ft/min. The pitch attitude was 15° nose up, banked 32° to the right and increasing. The PF’s sidestick had been full left and full back, and the stall warning had silenced.
The thrust levers were retarded to idle and as a result the nose pitched down from about 15° nose up to about 12° nose down. We can be sure that this was the result of the thrust reduction as the sidestick was held full back this entire time, and the elevators were at full nose up as well. The decreased pitch reduced the angle of attack slightly to around 38°, which was enough to push the indicated airspeed up to about 80 knots, revalidating the angle of attack information and activating the stall warning again. Meanwhile, the airplane was descending at 15,000 feet per minute.
Even though the stall warning sounded again, the sidestick remained full back, the elevators full up, and thrust at idle. The nose then pitched up to about level with the horizon and the angle of attack increased, causing the indicated airspeed to fall again and the stall warning to silence.
Almost 20 seconds later at 02:12:04 Bonin said, “I have the impression we have some crazy speed, what do you think?” and he started to deploy the speed brakes. This validates the theory that he thought they were going too fast. Robert commanded, “No, above all don’t extend the…”
“No? Okay,” Bonin replied.
“Don’t extend,” Robert said, and then they were retracted.
At 02:12:10, descending through 29,000 feet, the thrust levers, having been at idle for the last 20 seconds, were then advanced to the climb detent, and the engines spooled up. The aft deflection of the sidestick was relaxed, but the elevator remained full nose up. With the additional thrust from below the wings the nose pitched up from 10° nose down and reached 8° nose up over the next few seconds.
02:12:20 The bank angle, which had been to the right for the past minute, momentarily reached wings level, but continued to oscillate between 20° right and 18° left. Each rolling motion was counteracted with full lateral stick input.
02:12:33 The thrust levers were then pushed forward to the TOGA position again, but little additional thrust was available. The stick was moved halfway forward. The elevator moved from its full-up position, and the pitch reduced from 8° nose up to 2° nose down. The stab trim also started to move from its full nose-up position. The indicated airspeed rose above 60 knots yet remained below 80 knots, and the stall warning triggered again for about 2 seconds. This indicates that some pitch control, or at least pitch influence, was still available.
02:12:40 Another pitch oscillation brought the nose from 2° down to about 8° nose up and then about 8° nose down. The indicated airspeed rose briefly to over 150 knots and the stall warning sounded for about 5 seconds.
02:12:42 A brief exchange illustrated First Officer Bonin’s uncertainty about the altitude. With the stall warning sounding, he said, “What are we here? On alti what do we have here?”
The captain said, “it’s impossible,” and Bonin repeated the question, “In alti what do we have?”
First Officer Robert replied, “What do you mean, on altitude?”
Bonin said, “Yeah, yeah, I’m going down, no?”
Robert replied, “You’re going down, yes.”
At 02:12:45 the airplane passed through 20,000 feet. The bank angle increased to 40° to the right and Bonin held full left stick again to try to correct the excessive bank angle. The sidestick averaged approximately neutral in pitch but the nose pitched down to about 8° below the horizon, as it rolled right. The airspeed rose, angle of attack data was declared valid again, and the stall warning sounded for about 10 seconds. With the nose pitched down and the stall warning on, the sidestick was moved further aft.
The captain and First Officer Robert instructed Bonin to ‘get the wings horizontal’ three times, and Bonin replied, “That’s what I’m trying to do. I’m at the limit with the roll.” He had been holding left stick for most of the previous 90 seconds.
02:13:00 The captain suggested, “The rudder bar. Wings horizontal, go gently, gently.”
02:13:02 Bonin initially and briefly input left rudder to counteract the right bank, and as the wings returned towards horizontal, a right sidestick input was made and rudder input also changed to the right. The yaw damper’s rudder commands intensified and Bonin’s rudder input had little effect.
02:13:10 As a result of constant aft stick input, the elevator returned to full nose up. The airplane pitched up toward 15° and the angle of attack returned to greater than 45°, reaching even as high as 60°.27
02:13:18 Bonin remarked, “We’re there, we’re there, we’re passing level one hundred” (10,000 feet). “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 “won’t do anything,” but he apparently reset PRIM 1 and SEC 1 anyway.
02:13:36 Bonin called out, “Nine thousand feet.” The sidestick was still about half way back, the elevator and stab were at full nose up (trying to comply with Bonin’s pitch-up order), and the thrust levers were in TOGA.
Robert said, “Climb climb climb climb!”
To which Bonin replied, “But I’ve been at max nose up for a while.”
Realizing that Bonin had been doing the wrong thing for some time, the captain said, “No no no, don’t climb!”
Robert said, “So, go down,” and pushed his own stick forward for five seconds while the thrust levers were reduced to climb. Bonin continued to pull back.
02:13:40 A synthetic voice announced, “DUAL INPUT” five times, indicating that both sidesticks were displaced from neutral. Robert had failed to announce that he had taken control of the airplane, or to use the takeover push-button to cut out Bonin’s sidestick commands. Bonin’s nose-up and Robert’s nose-down pitch commands cancelled each other out. Robert commanded, “So give me the controls, the controls, the controls to me.”
Bonin released the stick and said, “go ahead, you have the controls, we are still in TOGA, eh?”
At 02:13:49, with the sidestick held forward by Robert, the pitch attitude decreased to about 5° nose down. The indicated airspeed rose to about 90 knots and the stall warning sounded again. Robert eased up on the forward stick input and, despite having said he had the controls, Bonin pulled back on his sidestick for a few seconds.
02:14:00 The nose began to rise again
02:14:05 The captain said, “Watch out, you’re pitching up there, you’re pitching up.”
02:14:07 Robert began pulling full back on his sidestick, saying, “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°.
Bonin added, “Well, we need to, we are at four thousand feet!”
02:14:10 The captain warned again, “You’re pitching up.”
02:14:16 At 2,500 feet above the water, and within the operating range of the radar altimeters, the ground proximity warning system (GPWS) activated, warning of the high vertical speed close to the surface. The synthetic voice said, “SINK RATE, PULL UP!, PULL UP!”
The captain gave permission, “Go on, pull.”
“Let’s go, pull up, pull up, pull up!” said Bonin.
The thrust levers were moved to TOGA both pilots pulled their sidesticks aft as the airplane pitched up toward 16°.
Bonin pushed the takeover push-button on his sidestick cutting out Robert’s commands. “PRIORITY RIGHT” was announced by the synthetic voice. A red arrow in front of Robert illuminated and pointed to the right where Bonin was pulling full back on the stick.
Bonin, still not having understood what happened said, “[expletive] we’re going to crash”
The GPWS commanded, “PULL UP!”
“This can’t be true” Bonin
said.
“PULL UP!” repeated the GPWS.
Bonin pled, “But what’s happening?”
The captain commanded, “Ten degrees pitch.” The GPWS continued to alert “PULL UP!”
Robert pushed his sidestick forward but Bonin still held the takeover push-button so Robert’s inputs were blocked. Both engines were operating at near maximum thrust and the angle of attack was in excess of 45°. The airplane impacted the water 16° nose up, in a 5° left bank with a forward ground speed and vertical speed both at 107 knots (10,900 ft/min).
Were there more indications lost than just airspeed?
It is clear that for about 90 seconds, the crew of AF447 had a hard time determining if they were climbing or descending, as the airplane fell from 32,000 to 10,000 feet. Vertical speed indications were erratic. Most probably due to the effect of the high angle of attack on the static ports.
The following exchanges beg the question if the crew had lost more instrument indications than just airspeed.
PF: Pilot Flying
PNF: Pilot Not Flying
CA: Captain
02:10:39
The air data and attitude/heading sources for the right side were switched to ADR 3 and IR 3 (FO on 3).
PF
02:11:58
“I don’t have vertical speed indication. I have no more displays”
At this time the airplane was descending through approximately 32,000 feet at 15,000 ft/min.
PNF
02:12:02
“We have no more valid displays.”
PF
02:12:04
“I have the impression that we have some crazy speed, no what do you think?”
PNF
02:12:13
“We’re pulling. What do you think about it, what do you think, what do we need to do?”
CA
02:12:15
“There I don’t know, there it’s going down.”
02:12:15-19
The air data and attitude/heading sources for the captain were switched to ADR 3 and IR 3 (CA on 3).
CA
02:12:23
“The wings to flat horizon the standby horizon.”
PNF
02:12:27
“You’re climbing.”
PNF
02:12:28
“You’re going down, down, down.”
CA
02:12:28
“Going down?”
PF
02:12:30
“Am I going down now?”
PNF
02:12:31
“Go down.”
CA
02:12:32
“No, you climb there.”
PF
02:12:32
“I’m climbing okay, so we’re going down now.”
CA
02:12:34
“You’re climbing.”
PF
02:12:42
“On alti what do we have here? In alti what do we have?”
CA
02:12 43
“It’s impossible.”
PNF
02:12:45
“What do you mean, on altitude?”
PF
02:12:47
“Yeah yeah yeah I’m going down, no?”
PNF
02:12:50
“You’re going down, yes.”
PF
03:13:18
“We’re there, we’re passing level one hundred.”
02:13:32
The air data source was selected back to NORM.
The crew, or at least the pilot flying, had an obvious difficult time determining if the airplane was ‘going up’ or ‘going down.’ It is unlikely that he suddenly forgot how to read an altimeter. There is no mention in the report indicating the loss of indications other than airspeed. During this entire exchange, while there was confusion on whether they were going up or down, they were in fact going down in the range of 15,000 feet per minute. That is over 170 miles per hour - in the vertical axis!
I have spoken with some of the world’s leading experts on this airplane, both test pilots and experts on its recording systems. They are not relying on the English translation of the transcript. There was no reason for the altitude display to have been lost. They stated that if it had been, there would have been other faults registered with the air data and display management computers, but there were none.
As they were descending through 30,000 feet, First Officer Bonin declared, “I have a problem, it’s that I don’t have vertical speed indication.” In the prior seconds, the angle of attack exceeded 45°, and the recorded vertical speed had become erratic. In simulator recreations I have seen the vertical speed indication go blank when the vertical speed reached 20,000 feet per minute (shown below). I take him at his word that he did not have a vertical speed indication. However, all indications show that the altimeter continued to operate.
The crew saw an airplane with operating engines, pitched up, erratic airspeed, an altimeter moving many times faster than they have ever seen one move, and a vertical speed indicator that was blank, erratic, or pegged beyond the limit of its normal display range. It most likely presented a picture that did not make any sense. The captain remarked, “It’s impossible.” This was a situation where they could not imagine that all of these indications were true, but what to believe? Alternate sources for the data were selected, but it made no difference.
There are also hints that the attitude displays on the two primary flight displays may also have been lost for an unknown period of time. However, Airbus and the accident investigators state that they do not believe that happened.
Both first officers declared that they had “no more valid displays.” This is not a statement that would be expected with the loss of just airspeed and vertical speed. Then the source for the left side instruments, both air data and heading/attitude were switched to the ADR 3 backup, having been previously selected to the right side.
The captain said, “The wings to flat horizon, the standby horizon.” Why he would refer to the standby horizon if the two primary attitude displays were operating is a mystery. The most logical explanation is that they were not working. No other similar incident had reported a loss of attitude display. That is until a Mihin Lanka A321 (Mihin Lanka is the budget airline of the Democratic Socialist Republic of Sri Lanka) in March, 2013. The airplane was en route from Jakarta, Indonesia to Colombo, Sri Lanka.
My sources revealed that the flight was cruising at FL 340, 500 miles southeast of Colombo Sri Lanka (a position approximately 4° North latitude), and encountered turbulence at which time the autopilot disconnected with multiple ECAM alerts. The crew lost airspeed indications as the airspeed values fell below 30 knots. Altitude indications jumped 580 feet due to the loss of a reliable Mach value to correct the static pressures. The flight control law downgraded to Alternate Law and the autothrust was lost. The crew also reported the loss of both attitude displays on their Primary Flight Displays (PFDs) for up to 20 seconds. The loss of an attitude display in this type of scenario is certainly unexpected, and seems inconsistent with the condition.
A preliminary evaluation was said to have revealed that at least two of the three AOA probes, the total air temperature probe, and pitot tubes, and engine ram air pressure sensors had suffered an icing incident. But there was no evidence of any loss of attitude display as reported by the pilots since all the inertial reference data remained valid.
Perhaps the crew of AF447’s apparent confusion was due to inoperative attitude displays on their PFDs, an erratic or inoperative vertical speed indication, and a rapidly declining altitude indication. We will never know for sure.
The BEA recommended that aircraft undertaking public transport flights with passengers be equipped with an image recorder that makes it possible to observe the entire instrument panel. This equipment would have clarified the “no more usable displays” statement and resolve questions such as the pilots reporting loss of att
itude displays when the feeding data remained valid.
Could it Have Happened in a Boeing?
There were many voices that were quick to say that had this been a Boeing, with a conventional yoke control column and not the Airbus sidesticks, that the accident would not have happened. The theory being that the other pilot would see that the pilot flying was pulling back for long periods of time and corrected it.
We can never know for sure, but support of that theory comes from a segment of the cockpit voice recording when the airplane was passing through about 9,000 feet:
02:13:36 Bonin called out “Nine thousand feet”. At this time the sidestick was about half way back, as it had been for the previous 40 seconds, though no one else could see this.
Robert remarked, “Climb climb climb climb!”
To which Bonin replied “But I’ve been at max nose up for a while.”
The captain, realized that Bonin has been doing the exact wrong thing - despite prolonged stall warnings for some time, stated, “No no no, don’t climb!”
Robert said, “So, go down,” and pushed his own stick forward for 5 seconds while the thrust levers are reduced to climb. But Bonin continued to pull back. A synthetic voice announced, “DUAL INPUT” five times, indicating that both sidesticks were displaced from neutral. Robert failed to announce that he had control of the airplane, or to use the takeover push-button to cut out Bonin’s sidestick commands. The result was that the two pitch commands cancelled each other out. Robert commanded, “So give me the controls, the controls, the controls to me.”
The excerpt from the flight data recorder tracing below, is from the time that the above exchange took place. The top set of lines graph the sidestick position of each pilot. Above the 0 line is nose down, below is nose-up command. The purple line defines the elevator position, which is at or near the full nose up limit at this point. The bottom line shows the pitch attitude. Keep in mind that about 3° is the normal cruise level attitude.