Mayday Air Crash Investigations - S10 E06 - Whos in Control
Category
📚
LearningTranscript
00:00Ladies and gentlemen, we're beginning our descent towards Amsterdam-Skipel Airport.
00:06We'll have to turn that off for now, sir.
00:08Turkish Airlines Flight 1951 is preparing to land in Amsterdam.
00:14Amsterdam, Turkish 1951, descending to 7000, speed 250.
00:23The crew is flying a state-of-the-art Boeing 737.
00:28Flaps 15.
00:30Localizer alive, localizer capture.
00:38In the final moments of the flight, the landing turns into a catastrophe.
00:43The plane falls like a rock.
00:46The crash of Turkish Airlines Flight 1951 involves the most popular plane on Earth.
00:51With nearly one and a half million passengers boarding 737s every day,
00:55investigators need to figure out if the problem was with the plane or with the pilots flying it.
01:00Made it! Made it!
01:01Made it!
01:02Made it!
01:03On the morning of February the 25th, 2009,
01:17Turkish Airlines Flight 1951 becomes the first plane to crash at Amsterdam's Skipel Airport in more than ten years.
01:34It smashed into the ground really hard and made a tremendous noise.
01:41The plane hits the ground in a muddy field just north of runway 18 right.
01:51Since the crash was so close to the airport, rescue workers arrive quickly.
01:58Most of the passengers have survived, but many are badly injured.
02:05Survivors are taken straight to local hospitals.
02:09Images of the Amsterdam accident quickly spread around the world.
02:17This is the third crash of a passenger jet in the past six weeks.
02:22The sudden nature of this accident adds to the mystery.
02:29It doesn't take long for the Dutch Safety Board to arrive at the scene.
02:33They will be investigating this accident, but they won't be alone.
02:40The crash involved an American-made plane,
02:43so the U.S. National Transportation Safety Board sends Joe Sidor,
02:47one of its most experienced investigators, to Amsterdam.
02:50When it's a non-U.S. registered aircraft that crashes overseas such as this Turkish Airlines,
02:56we are the state of manufacture and design of the airframe.
03:01And also, in this case, the engines.
03:04Fuselage in three large pieces, engines forward at the main wreck site.
03:10Flight 1951 was one of the most advanced aircraft in the skies,
03:15the Boeing 737-800 series.
03:18It's designed to travel longer routes at higher altitudes.
03:22The new generation 737 is still the best airplane Boeing ever built.
03:27We developed an airplane that had an improved wing, improved avionics, simpler systems that required less maintenance.
03:37Investigators know this isn't just any plane.
03:40The 737 is the world's best-selling commercial jet.
03:44Finding out why this one crashed is imperative.
03:48There are more than 5,000 of them in the skies.
03:52They carry about one and a half million passengers a day.
03:55Investigators must quickly determine if there's a flaw with the plane that could cause another accident.
04:01What they know so far is that flight 1951 had been traveling from Istanbul, Turkey, to Amsterdam.
04:09There were 128 passengers on board the early morning flight, including four engineers from Boeing.
04:16There was no mention on board of any kind of trouble in the cockpit.
04:31The crash has killed nine people, including three members of the Boeing team and the pilots in the cockpit.
04:43See if you can get me some aerials of the crash.
04:45There are eerie similarities to another recent accident involving a Boeing aircraft, British Airways Flight 38.
04:52Approximately a year before this, there had been a 777 short landing at Heathrow, which had a dual-engine flame out.
05:00In that accident, a Boeing 777 fell to the ground short of the runway.
05:06The British Airways pilots reported that both their engines stopped delivering power just before landing.
05:13At the time of the Turkish Airways crash, the cause of that accident hasn't been found.
05:21Like in that case, investigators have plenty of clues to work with.
05:25The plane and its engines are largely intact.
05:28The flight data recorder and cockpit voice recorder are found in good condition.
05:33There are also plenty of survivors to describe what happened.
05:38According to the passengers, the landing had been routine,
05:41but then suddenly the plane simply dropped out of the sky and hit the ground.
05:46But perhaps the biggest clue comes from the crash site itself.
05:56The wreckage is not spread out.
05:58It tells investigators that the plane could not have been traveling forward at high speed when it hit the ground.
06:04The way the aircraft had crashed, it did appear to be some sort of a landing accident in which there was moderate control of some sort.
06:13The pattern of debris and the passenger reports point investigators to an immediate suspect, the engines.
06:21The engines issue was a very big issue in my thought process at the time initially.
06:27There's no evidence of fire on the fuselage.
06:30In many crashes, fuel in the plane's tanks ignites on impact.
06:34The lack of fire raises an obvious question.
06:37Did the engines stop running because flight 1951 had simply run out of gas?
06:42That was one of the first thoughts that I had was, did this airplane have fuel aboard?
06:46Because otherwise, how does a 737 literally fall out of the sky on approach to an airport?
06:52But the location and condition of the plane's engines suggest that perhaps they didn't quit in flight.
06:57Sure looks like it was running.
06:59When we first looked at where the engines ended up, the initial impression was that they probably were producing thrust at impact,
07:07given that they were so far forward of the main wreckage.
07:11That was just a very general conclusion.
07:16Only the flight data recorder can tell investigators how much power the engines were generating in the seconds before the crash.
07:25Lots of fuel. Rules that out.
07:28It doesn't take long to discover that there was plenty of fuel in the plane's tanks.
07:32Flight 1951 definitely did not run out of gas.
07:36Passengers report that in the final seconds before the crash, the plane hit what felt like turbulence.
07:43It points investigators to a well-known culprit, a microburst.
07:48A microburst is a powerful column of air that shoots out of storm clouds.
07:53It can literally slam a low-flying plane to the ground.
07:56If an airplane flies into that at approach speeds, you're not going to be flying anymore.
08:00You're going to come out of the sky.
08:02Certainly, it was one of the things that all of us took a look at at the beginning.
08:05Was there a microburst?
08:06In 1985, a Delta Airlines flight was caught in a microburst while landing at Dallas-Fort Worth Airport.
08:20The plane hit the ground short of the runway.
08:23One hundred and thirty-seven people were killed.
08:29Investigators learned that there were heavy clouds above the airport at the time of the accident.
08:34A powerful gust of wind may well have accompanied them.
08:38The flight data recorder will have recorded wind speeds outside the plane.
08:45Investigators will need to analyze the flight data to prove the microburst theory.
08:50In the meantime, the rescue has led to an unusual discovery.
08:54There was three pilots in the cockpit, which is unusual.
08:58This is a two-crew cockpit.
09:00So why was that third pilot there?
09:03Since none of the three pilots survived the accident,
09:06it's all the more urgent for investigators to retrieve the data from the cockpit voice recorder.
09:12It records conversations in the cockpit.
09:15They're in luck.
09:16Because of the way that the aircraft crashed,
09:19access to the recorders and the condition of the recorders was excellent.
09:23The reason for the third pilot is soon uncovered.
09:26Flying standard arrival route.
09:28Visibility 3,500 meters.
09:31Expected to decrease to 2,500 meters.
09:34For First Officer Murat Cesaire, this had been a training flight of sorts.
09:39He was new to the airline and was being shown the intricacies of landing at Amsterdam's airport.
09:45In the Jepsen charts, which is what all pilots use to navigate to and from Schiphol,
09:50there's 102 pages of information on Schiphol alone.
09:53So there's dozens of approaches.
09:55Runway 18 to the right has three high-speed exits to the left.
10:00The captain, Hassan Arasan, was doing double duty.
10:04He was training his first officer.
10:07Make small corrections as we get close to the runway.
10:10And was in command of Flight 1951.
10:13It's because Captain Arasan was teaching that there was the third pilot in the cockpit.
10:18Olgai Ozhgur was a safety pilot.
10:21He was there to keep an eye on the flight's progress during this training mission.
10:25The purpose of that second set of eyes is to make sure that the captain and the first officer,
10:33if they are in a situation where it's a little bit of a training portion of the flight,
10:37that they don't miss something.
10:39We got a clean recording.
10:41The voice recorder reveals that the three crew members began preparations for landing when still above 8,500 feet.
10:50Amsterdam, Turkish 1951, descending to 7,000, speed 250.
11:00But the voice recorder has picked up an unusual sound.
11:03A warning that makes no sense at this stage of the flight.
11:07One of the investigators from Boeing was an engineering pilot that came and helped with the CVR.
11:15And he's listening to see when or are there any unusual sounds that can be heard that would not be normal.
11:21Speed okay for ILS-18, right.
11:26Descent to 4,000, ILS-18, right.
11:30The alarm keeps sounding.
11:32It's the landing gear configuration warning horn.
11:35Captain Arasan continually dismisses it.
11:39Turkish 1951, descend to 4,000, ILS-18, right.
11:45Landing gear.
11:50Is that the landing gear warning?
11:54They're at 8,300 feet here.
11:56The warning is a solid clue.
11:59But investigators can't yet see how it could possibly have caused a crash.
12:05The crew of Turkish Airlines Flight 1951 got warnings to extend their landing gear while still thousands of feet in the air.
12:13On the initial listen, we heard a gear warning horn occur as the aircraft was approaching when it was still up and it was coming in about 10,000 feet and below.
12:24Investigators now turn to the flight data recorder to help solve some of the mysteries surrounding this flight.
12:30The analysis of wind speeds outside the aircraft is completed.
12:34It's clear none are drastic enough to have brought down the plane.
12:38Evidence of a microburst.
12:43But the flight data recorder does provide some valuable insight into the cause of the landing gear warning.
12:51One of the instruments that measures altitude had the plane already on the ground.
12:57That's when we looked and saw the radio altimeter data.
13:00On the recorder it said about 8,000 feet and then immediately it went down to about minus eight.
13:05Minus eight feet is an indication that the aircraft's on the ground, but of course it's still at 2,000 feet.
13:12The Boeing 737 is equipped with two separate altimeters.
13:16One measures air pressure to determine the plane's height above sea level.
13:21That reading is displayed prominently in the cockpit on both pilots' flight display.
13:28The plane is also equipped with a radio altimeter.
13:31It's made up of four antennas.
13:34Two transmit signals to the ground and two others read the signal that bounces back to determine the plane's height.
13:41It's precise. It's very, very precise.
13:45Pressure altimeters can sometimes be not as accurate, and radar altimeters are 100% accurate if they're working properly.
13:58One antenna feeds the reading to the first officer's display.
14:03The other feeds the captain's instruments.
14:07In the case of flight 1951, the captain's side was wrong most of the flight.
14:14Investigators go back over the CVR and make a puzzling discovery.
14:19Amsterdam, Turkish 1951. Descending to 7000. Speed 250.
14:31Radio altimeter.
14:32Captain Arasan seems to have known that the landing gear warning was being caused by a faulty radio altimeter.
14:39The airplane thought that it was low to the ground and the gear was not down.
14:42And the captain recognized that the problem was really in the radio altimeter, showing him that they should be on the ground.
14:49And he goes, it's just the radio altimeter.
14:52Throughout much of the approach, the captain's radio altimeter had been displaying a reading of minus eight feet, triggering the warning to lower the gear.
15:00They treated it like it was a nuisance.
15:03Turkish 1951. Descend to 2000.
15:072000 1951.
15:12Investigators dig for any other abnormalities.
15:18They learned that with flight 1951 still about 17 kilometers from the airport, controllers directed the pilots to begin their final turn to line up with the runway.
15:29Turkish 1951.
15:30Turkish 1951.
15:31Left heading.
15:322-1-0.
15:33Clear approach.
15:341-8.
15:35Right.
15:36Left 2-1-0.
15:37Clear ILS.
15:38Turkish 1951.
15:392-1-0.
15:40Set, sir.
15:41This turn puts flight 1951 in line with runway 18 right.
15:54It's equipped with an instrument landing system, which sends out a signal outlining the ideal descent path to the foot of the runway.
16:03The autopilot follows that glide path until the plane is a few hundred feet from the ground.
16:08Then the pilot takes over.
16:10It makes landing almost effortless.
16:13The ILS is pretty easy to follow.
16:15It's a video game.
16:16My daughter has flown in a simulator and can land a 737 using the ILS.
16:27The crew begins configuring their plane for landing, unfazed by the warning horn that's repeatedly triggered by the malfunctioning radio altimeter.
16:36Flaps 15.
16:39Ten kilometers out, flight 1951 picks up the ILS signal that will guide the plane to the runway.
16:46Localizer alive.
16:49Localizer capture.
16:52The safety pilot, Olgay Özgür, now reminds Captain Arasan about the failed altimeter.
16:58We have radio altimeter failure, sir.
17:01Okay.
17:02Turkish 1951.
17:07Runway 18, right.
17:12Clear to land.
17:13Clear to land.
17:16Investigators are stumped.
17:18The crew knew about the malfunction and continued their approach.
17:22How had it then caused them to crash?
17:25Clearly there was more to this accident than a faulty altimeter.
17:28The whole premise of airline safety, the way we build the airplanes, the way we fly them, is based on the idea that we can have any number of failures and we should still be able to arrive safely.
17:39The radio altimeter is just one instrument.
17:41There's no way in the world that that one instrument, if it fails, should be a major cause of worry that we're going to have a crash.
17:49Investigators wonder if the crew had been given proper guidance for their approach.
17:59They turn to exchanges between the pilots and the controller who guided them in.
18:06They carefully review every instruction.
18:08Turkish 1951.
18:09Ascend to 4000.
18:10Speed okay for ILS 18.
18:11Right.
18:12Turkish 1951.
18:13Ascend to 2000.
18:14Turkish 1951.
18:15Left heading 210.
18:16Clear approach.
18:1718.
18:18Right.
18:19By following the controller's instructions, the crew made their final turn much too close to the runway.
18:40If they intercept properly, they should be here.
18:44International guidelines call for approaching planes to intercept the signal that guides them to the runway from below.
18:51It's so pilots don't have to make any drastic last minute course corrections to get to it.
18:57The intercept here, they had to descend.
19:00But flight 1951 was given instructions that brought it to the threshold of the glide slope while still way above it.
19:07It's a common practice at Schiphol because it gets planes to the runway faster.
19:12Because they were so close, they had to capture the glide slope from above.
19:16Although it is an unusual situation, it is one that can be handled by a flight crew if it is managed properly.
19:23Approaching a glide slope from above is more difficult, mostly because the crew has to suddenly slow the plane and descend rapidly to intercept the signal.
19:32We also call this a slam dunk approach.
19:35And some pilots like it, some pilots don't.
19:38It's a little bit harder and things happen quicker when you're above the glide path trying to intercept from above.
19:44And it's just a challenge for a lot of pilots.
19:47The approach from above increased the crew's workload, but it's standard practice at Schiphol Airport.
19:53I've flown into Schiphol dozens of times and I expect it.
19:57If the controller's instructions had somehow overtaxed this crew, their conversations would indicate it.
20:04They're just five kilometers from the runway.
20:07One thousand.
20:08Check.
20:09Flaps forty.
20:10Speedbrake.
20:11Speedbrake arm.
20:12Green light.
20:13One thing at a time.
20:14Landing gear.
20:15Gear down.
20:16Three green.
20:17Flaps.
20:18Flaps forty.
20:19Green light.
20:20Five hundred.
20:21All lights on.
20:22Please warn the cabin crew.
20:23Okay.
20:24Cabin crew, take your seats.
20:25Then real trouble.
20:38A stall warning.
20:39Speed, sir.
20:40I have control.
20:41100 miles of speed.
20:42I have level of speed.
20:44There's no no need to stop.
20:47Super there until a warning.
20:50100 miles of speed.
20:54Arisan fought to save his plane.
20:57But just 400 feet above the ground,
20:59at only one and a half kilometers from the runway,
21:01the Boeing 737 suddenly fell straight down.
21:05It only took a few seconds for it to hit the ground.
21:15Turkish 1951.
21:16The recording sheds light on the final minutes of the flight.
21:20The crew was configuring their plane for landing
21:23well after it should have been done.
21:25Flaps.
21:26Flaps 40. Greenlight.
21:28Most airlines have regulations that call for a flight to be stabilized,
21:32to have all checklists completed
21:34by the time the plane hits 1,000 feet.
21:37In instrument conditions,
21:39you're required 1,000 feet to have basically everything done.
21:42The airplane is configured, you have slowed,
21:44you have run your before landing check.
21:46And you've received your landing clearance.
21:49And from 1,000 feet on in,
21:50you just monitor the instruments
21:52and we're looking for the runway.
21:55Please warn the cabin crew.
21:59In fact, this crew was still running their checklist
22:02up to the moment the crisis hit,
22:04460 feet above the ground.
22:06This approach was not stabilized.
22:12Because the aircraft was unstable,
22:14the flight crew was in a very high workload environment
22:16in the last 1,000 feet of flight.
22:19The radio altimeter was malfunctioning.
22:22The aircraft was giving off warnings.
22:24The crew was assigned a challenging approach.
22:27And they were executing a checklist late.
22:30But none of this explains why Flight 1951 crashed.
22:34In these type of accidents,
22:35you can never get inside the head of the pilots.
22:39And that's a very frustrating type of accident.
22:42But the flight data recorder does provide another intriguing clue.
22:46Moments before Flight 1951 hit the ground,
22:49the plane's engines were at idle,
22:51hardly providing any power.
22:53Perhaps this accident is a repeat of the Heathrow incident.
22:56The engines, it was interesting to note,
22:59were at idle approximately the last two minutes of the flight
23:04until the very end when the thrust was increased again.
23:09That was a big red flag right there.
23:12The question is, is why was that the case?
23:15But then they spot something that's very different
23:17from the accident at Heathrow.
23:19It's hard flare mode.
23:21For some reason,
23:23while still more than 1,000 feet above the ground,
23:26the plane's computer began preparing to touch down.
23:31In retard flare mode,
23:33engine power is reduced to idle by the flight computer.
23:37And the plane's nose automatically pitches up
23:40to the flare position.
23:44Planes should only be in this configuration
23:46just before they touch the ground.
23:49The autopilot raises the nose to break the descent.
23:52The autothrottles brings the power back to flight idle.
23:54And you touch down with the power
23:56either all the way in idle
23:57or just about to be in idle.
23:59But flight 1951 went into a slow,
24:03nose-up position well before touchdown,
24:05causing the plane to fly slower and slower
24:08throughout its descent.
24:10Speed!
24:11So why was flight 1951 in landing mode?
24:14Speed!
24:16And why hadn't any of the three crew members
24:19noticed how slowly they were flying?
24:31So, what else was going on when the engines went to idle?
24:35The trouble seems to start with the malfunctioning altimeter.
24:39We had to look at the system as a whole
24:41and see how that minus eight affected
24:43the other systems on the aircraft.
24:46And that was a very big portion of this investigation.
24:49We had to say,
24:50how did the autopilot use that data?
24:53More importantly,
24:54how did the autothrottle use that data?
24:56The computer that flies the plane
24:58consists of two main systems,
25:00the autopilot and the autothrottle.
25:03The autothrottle determines
25:04how much power to ask the engines for,
25:07while the autopilot controls
25:08the plane's altitude and direction.
25:14The two systems work independently of each other,
25:17and only one of the radio altimeters
25:19provides information to the autothrottle.
25:21In this case,
25:23I had to learn everything there was
25:24about radio altimeters and autothrottle systems,
25:26which I didn't know before.
25:28The pieces of the puzzle begin coming together
25:30when they find the connection
25:32between the faulty radio altimeter
25:34and engine power.
25:36The radio altimeter provides information
25:39to the autothrottle from the captain's side.
25:41The only altimeter feeding information
25:52to the autothrottle was the captain's,
25:55and it was wrong.
25:57It showed minus eight feet
25:58throughout most of flight 1951's approach.
26:03It's beginning to look like
26:04the faulty radio altimeter
26:06triggered the events that led to the crash.
26:09Investigators need to know
26:10what went wrong with it.
26:13On a 737,
26:15the transmitting and receiving antennas
26:16for both radio altimeters
26:18are lined up underneath the cockpit.
26:21Three of the antennas
26:23were all but destroyed in the crash.
26:25They can't be tested.
26:27But one antenna from the captain's side
26:29is undamaged.
26:32Investigators consider two possibilities.
26:34A failure of one of the components,
26:37or some sort of interference
26:38that caused the faulty reading.
26:40The only component that survived the crash
26:43checks out.
26:49The computers that control the system
26:51also work.
26:53But investigators do make
26:55a curious discovery about them.
26:57They aren't the same ones
26:58that were installed on the plane
27:00when it was delivered to Turkish Airlines
27:02seven years ago.
27:02This find changes the focus
27:05of the investigation.
27:07The maintenance aspect
27:08of this accident aircraft
27:11was one that we looked at
27:13as deeply as we could.
27:14When the plane's maintenance log is studied,
27:27investigators find that the radio altimeter
27:29on this plane had a problematic history.
27:32We got additional data from Turkish Airlines.
27:36And that data showed that
27:38on this one aircraft
27:39of the past, I believe,
27:42over a thousand flights,
27:43there was about 150 flights
27:45that had faulty radio altimeter systems.
27:48The documents show
27:49that a little more than a year
27:51before the crash,
27:52both computers were replaced.
27:54because of complaints,
27:55they were causing faulty readings.
27:58One of the incidents involved
27:59a radio altimeter reading
28:00of minus eight feet.
28:04So that was telling us
28:06that there was an issue
28:07that had been there.
28:08The issue did not just occur
28:09on this flight.
28:11The faulty readings persisted.
28:15Mechanics repeatedly swapped
28:16the computers and replaced the antennas
28:18to try to solve the problem.
28:22It's determined
28:23that Turkish Airlines
28:24tried several ways
28:25to fix the altimeter,
28:27but they couldn't find
28:28a repair that worked.
28:33At the time of the accident,
28:35Turkish Airlines
28:36had a fleet of 52
28:37Boeing 737-800 series airplanes.
28:46It's on page 93.
28:48When we reviewed
28:49the maintenance data,
28:51we found that radio altimeter problems
28:52had been written up
28:53several times
28:54on both the accident airplane
28:56and the fleet.
28:59Investigators discover
29:00that in the year
29:01before the crash,
29:02Turkish Airlines
29:03dealt with 235 system faults
29:06with the radio altimeters
29:07on their 737s.
29:11Fixes ranged from replacing
29:13and exchanging antennas,
29:15cleaning of the systems,
29:17exchanging and replacing
29:18the computers,
29:18and installing gaskets
29:20and installing gaskets
29:21to shield the system
29:22from possible water damage.
29:25It's not like they weren't
29:26doing anything about it.
29:28The Turkish Airlines
29:29maintenance personnel
29:30knew that the radio altimeter
29:33problem was one
29:35of their highest issues
29:37with regard to maintenance.
29:3916 of those altimeter repairs
29:41were made to the plane
29:42that crashed in February 2009.
29:45If the problem
29:47was so widespread,
29:48investigators wonder
29:49why it hadn't caused
29:50serious problems
29:51before this accident.
29:55They don't have to dig
29:57too far back
29:57to find out
29:58that in fact it had
29:59on this very same plane.
30:03On two recent flights,
30:04they had the exact same problem.
30:07Twice in the 48 hours
30:08leading up to the accident,
30:10the radio altimeter
30:11showed a negative reading,
30:13putting the plane
30:15into retard flare mode.
30:17Both times,
30:18the crew noticed the problem,
30:21disengaged the autothrottles,
30:23and brought the plane in
30:24for a safe landing.
30:31You just disconnect it
30:32and fly the airplane.
30:34In the months after the crash,
30:36other operators come forward
30:37with similar stories.
30:39In Australia,
30:41in the Netherlands,
30:42in Canada,
30:43in Austria,
30:44pilots report
30:45their 737s
30:46going into retard flare mode
30:48when the left radio altimeter
30:50showed a faulty reading.
30:52Each of those crews
30:53reacted the same way.
30:55They disengaged the autothrottle
30:57and pushed the power
30:58back up manually.
31:01They all landed safely.
31:05Things are going to break
31:06on an airplane,
31:07and usually you're able
31:08to identify that
31:09and take that out of,
31:12make it so that it's not a threat
31:13for the landing.
31:15In 2008,
31:18Boeing received a whopping
31:192,569 reports
31:22of faulty radio altimeters
31:23on their latest 737s.
31:26But very few of those cases
31:27involved the plane
31:28going into retard flare mode.
31:31Hardly any reports at all.
31:32Boeing also tried,
31:35but couldn't find
31:36the cause of the failures.
31:38They concluded
31:39that the radio altimeter problem
31:41was not a threat to safety
31:42because the 737
31:44gives off enough warnings
31:45so that crews
31:46can intervene
31:47and land safely.
31:51In fact,
31:51in every instance
31:52where the radio altimeter failed,
31:55crews were able to recover.
31:56Turkish Airlines flight 1951
32:02seems to be the one exception.
32:08Investigators still don't know why.
32:11It really got us wondering
32:12of what happened.
32:14And that's when we started
32:15to look really closely
32:16at the actions
32:18of the flight crews
32:19while it was on that final approach
32:21in less than 1,000 feet.
32:25Finally,
32:26as investigators again
32:28revisit the last minutes
32:29of flight 1951,
32:31the circumstances
32:32of the tragedy
32:33become clear.
32:35They see a remarkable
32:37sequence of events
32:38that transpire
32:39to bring down this plane.
32:41So,
32:42what was happening
32:43when the plane
32:44went into retard flare mode?
32:45They discover
32:46that the plane
32:47went into landing mode
32:48and pulled back power
32:50at the worst possible moment,
32:52exactly as the crew
32:53was descending
32:53to meet the glide slope.
32:55It masked
32:56what was actually happening.
32:59As the crew
33:00configured their plane
33:01to drop down
33:02to meet the glide slope,
33:04they expected the plane
33:05to slow down
33:06as part of that maneuver.
33:07But the plane
33:08was actually slowing down
33:10because the computer
33:10was in landing mode.
33:13That's why
33:13none of the three pilots
33:15said anything about
33:15the throttles
33:16moving to idle.
33:17It was insidious.
33:19Where it first captured
33:20in the retard mode,
33:21it didn't hurt them at all
33:22because they were actually high
33:23and they were a little bit fast.
33:25And the pilots
33:26actually wanted
33:27the power back anyway.
33:28In fact,
33:28the throttles may have
33:29already been
33:29in the flight idle mode
33:31as they were trying
33:32to get down
33:33and slow down.
33:35The trouble starts here
33:36at 8,300 feet,
33:3713 miles out
33:38from the airport,
33:39minutes before the crash.
33:41Amsterdam.
33:49Turkish 1951
33:50descending 7,000
33:52speed 250.
33:56Turkish 1951
33:58descend to 4,000.
33:59Speed OK
34:00for ILS 1B.
34:03Right.
34:05Grading ultimate.
34:11Would the crew
34:12have known
34:12that because of
34:13that radio altimeter
34:14they would have gone
34:15to a retard flare mode
34:16in the throttles?
34:17No.
34:19It was a common problem
34:20at the airline
34:21that the crew
34:21couldn't see the risk
34:22it posed this flight.
34:24We have an airplane
34:25that was malfunctioning
34:26in a very minor way
34:27but in a way
34:28that if not caught
34:29could and did
34:30metastasize
34:31into something
34:32much more virulent.
34:34Turkish 1951
34:36descend to 2,000.
34:382,000.
34:391951.
34:41Turkish 1951
34:45left heading
34:46210
34:47cleared approach
34:4818 right.
34:51Left 210
34:52cleared ILS
34:54Turkish 1951.
34:56Now left
34:56at 210 degrees
34:58maintaining 2,000 feet
35:00brings the flight
35:01in right here.
35:035.5 miles out.
35:05They now have to
35:06intercept
35:06the glide slope
35:08from above.
35:08At 2,000 feet
35:12with the glide slope
35:13below them
35:13the pilots
35:14have to reduce
35:15their speed
35:16while descending
35:16steeply.
35:18Speed 140.
35:20They believe
35:20the throttles
35:21are moving back
35:22for the descent
35:22to the glide slope.
35:24In fact
35:25the auto throttle
35:26is slowing
35:26the plane down
35:27because it's gone
35:28into landing mode.
35:30It will continue
35:31to slow the plane
35:31until it stalls.
35:33What we found
35:34is that
35:34when the flight crew
35:36was doing
35:36their before landing
35:37checklist
35:38each one of them
35:39was doing something
35:40while they should
35:41have been
35:42monitoring their airspeed.
35:44For the next
35:44100 seconds
35:45no one notices
35:47what's happening
35:48until it's too late.
35:51Established altitude set.
35:531,000.
35:54Check.
35:56Flaps 14.
35:58Speed set.
35:59The experienced pilot
36:00recognized that
36:01the first officer
36:02was probably a little
36:02bit behind
36:03on the approach
36:03so he calls
36:04for flaps 40
36:05and moves the lever
36:06informing the first officer
36:08that he has done so.
36:10He's trying to help
36:10the first officer
36:12catch up
36:12to the position
36:13of the aircraft.
36:14Speed break.
36:14Speed break armed.
36:16Green light.
36:17One thing at a time.
36:19Landing gear.
36:20The plane is now
36:21700 feet from the ground.
36:23Gear down.
36:24Three green.
36:24Flaps.
36:25Flaps 40.
36:27Green light.
36:29In their haste
36:30to complete
36:31their checklist
36:32none of the three
36:33crew members
36:33noticed the warnings
36:34that their speed
36:35is dropping dangerously.
36:37First a red bar
36:38appeared on their
36:39flight display.
36:41Then when the
36:42airspeed continued
36:42to drop
36:43a flashing box
36:44appeared around
36:45their actual airspeed
36:46to draw the pilot's
36:47attention to it.
36:49At this point
36:50no one sees either.
36:51Cabin report confirmed.
36:53The aircraft
36:53is now 600 feet
36:55from landing.
36:56When things start
36:56changing colors
36:57it's a warning to you,
36:59it's a caution to you
36:59that you're approaching
37:01the limits of the aircraft.
37:02Mist approach.
37:04Altitude set.
37:05500.
37:06So all these indications
37:07the crew has
37:08in front of them
37:09saying that
37:10the aircraft
37:11is slowing down.
37:11during that time
37:13they were still
37:14completing their checklist.
37:15Of course the aircraft
37:16is getting closer
37:17to the ground.
37:18In fact it's less
37:19than 500 feet
37:20from touchdown.
37:21And right before
37:22the stick shaker
37:24started
37:24the captain
37:26told the safety pilot
37:27please warn the cabin crew.
37:29Cabin crew
37:30take your seats.
37:33Speed set.
37:34I have control.
37:35By the time
37:36they advanced
37:36the tunnels
37:37to full power
37:37it was unrecoverable.
37:40They were too low
37:40for the engines
37:41to catch up.
37:42And that's it.
37:44It's now too late
37:45to save this plane.
37:51They all knew
37:52about the altimeter
37:52problem
37:53but knowing
37:54didn't help.
37:55Boeing also
38:03didn't foresee
38:04the potentially
38:05deadly effect
38:06of a faulty altimeter
38:07but on February
38:09the 25th 2009
38:10it triggered
38:12an unusual
38:12sequence of events
38:13that brought down
38:14a jetliner
38:15and killed
38:16nine people.
38:19The official report
38:20into the accident
38:21blames it
38:22on a convergence
38:22of circumstances.
38:23There is never
38:25ever just one
38:26cause
38:26to an airline
38:27accident.
38:28It simply
38:28doesn't exist.
38:29Maybe someday
38:29God will swap
38:30one out of the sky
38:31but until then
38:33there's never
38:33one cause.
38:34The Dutch report
38:35also points out
38:36that Boeing
38:37could have realized
38:38the problem
38:38with the radio
38:39altimeter system
38:40could have had
38:40an impact
38:41on safety
38:42given that no one
38:43foresaw
38:44how that failure
38:45might cause a crash.
38:47The Turkish
38:47Airlines accident
38:48raises a big
38:49question.
38:51Are airplanes
38:52becoming too
38:53complex?
38:53Investigators
39:04have determined
39:04that Turkish
39:05Airlines flight
39:061951
39:07crashed mainly
39:08because the
39:09pilots didn't
39:09recognize the
39:10consequences of
39:11the warnings
39:11they were getting.
39:15Turkish
39:161951
39:17descent to
39:184000.
39:18This is not
39:19the first plane
39:20to crash
39:20because the
39:21crew didn't
39:21understand
39:22what their
39:22plane was
39:22telling them
39:23our problem
39:24is not the
39:25automation
39:25our problem
39:26is the
39:26depth of the
39:27training
39:27and the
39:28ability of
39:28the human
39:29beings
39:29to recover
39:30from mistakes
39:31made in
39:31interfacing with
39:32the automation.
39:33Mika Ensley
39:35studies the
39:36relationship
39:36between pilots
39:37and technology.
39:39We haven't
39:40really designed
39:41the automation
39:41to take best
39:43advantage of
39:44what people do
39:44well and take
39:46away the parts
39:47that people don't
39:48do well.
39:49In 1996,
39:51the crew of an
39:51Aero Peru 757
39:53crashed when the
39:54pilots couldn't
39:55decipher
39:55contradictory warnings
39:57about their
39:57altitude and air
39:58speed.
40:00The plane flew
40:01into the
40:01Pacific Ocean.
40:0261 passengers
40:03and nine crew
40:04members were
40:05killed.
40:07In 1995,
40:08the flight
40:08management system
40:09on a Boeing
40:10757 could and
40:11should have
40:12steered the
40:12plane to a safe
40:13landing in
40:14Cali, Colombia.
40:16But a last-minute
40:17change to the
40:18flight plan meant
40:18the crew had to
40:19reprogram their
40:20computer.
40:21They mistakenly
40:22entered a course
40:22that took them
40:23headlong into a
40:249,000-foot mountain.
40:27159 people died
40:28in the crash.
40:29The cautionary
40:30tale here is that
40:31we can get this
40:31equipment, we can
40:32get these silicon-based
40:33units, if you will,
40:34so sophisticated that
40:36we can't talk to
40:37them effectively.
40:38And when they go
40:38berserk or they
40:39have a problem or
40:40we misprogram them,
40:42we end up putting
40:43ourselves and our
40:44passengers in danger
40:45while we're trying
40:46to figure it out.
40:47NASA is working
40:48on something called
40:49the Integrated
40:50Intelligent Flight
40:51Deck.
40:52It's a project
40:53aimed at helping
40:53humans better use
40:55the technology that
40:55surrounds them in
40:56the cockpit.
40:57They're looking at
40:58automated flight
40:59manual systems,
41:00voice recognition
41:00systems, they've
41:02developed new kinds
41:03of microphones that
41:04are bone-conducting
41:06chips.
41:06They're developing
41:07new displays for
41:08understanding the
41:09ground environment,
41:11for being able to
41:12detect where other
41:13aircraft are in the
41:14environment.
41:15What some
41:15researchers are
41:16finding is that
41:17the best technology
41:18shouldn't replace
41:18pilots but work
41:19with them.
41:21Really integrating
41:22people with
41:23technology more
41:23effectively has to
41:24do with designing
41:26the displays so
41:27that you can really
41:28understand what it's
41:29doing and you can
41:30make it simpler to
41:31understand how to
41:32make it do what it is
41:33you want it to do.
41:34You shouldn't have to
41:34push 16 buttons
41:35through 8 levels of
41:37menus to figure out
41:38what's going on with
41:38the system.
41:39It should be integrated
41:40and presented
41:40effectively.
41:41It should be as
41:42easy to communicate
41:43with as the person
41:43next to you.
41:45Boeing and Airbus,
41:46the two largest
41:47manufacturers of
41:48passenger planes,
41:49take very different
41:50approaches to the
41:51relationship between
41:51humans and technology.
41:54Airbus gives the
41:55flight computer much
41:56of the decision-making
41:57power in the cockpit.
41:59In their view,
41:59this is a way to
42:00prevent a lot of
42:01human errors by making
42:02sure the airplane
42:03doesn't do something
42:04that's going to cause a
42:05crash, even if the
42:06humans want them to.
42:07But Boeing has a
42:08different view.
42:09Its philosophy is to
42:10provide information to
42:12pilots and have them
42:13make decisions.
42:14Having more information
42:16is better for the
42:17pilot than having the
42:18pilot in the loop, in
42:21the equation, so to
42:22speak.
42:23I kind of like that.
42:25Airbus will argue
42:26vociferously and in
42:28continuous fashion that
42:29that view is archaic and
42:31incorrect.
42:32I think the jury is still
42:33out.
42:34The final report into
42:35the Turkish Airlines
42:36tragedy blames the
42:37crash of flight 1951
42:39partly on a failure of
42:40technology.
42:41The erroneous radio
42:42altimeter data caused
42:44the autothrottle to go
42:45to an improper mode
42:46that is, of course, not
42:48a good situation.
42:49The cause of the
42:50radio altimeter failure
42:51was never uncovered.
42:53The Dutch investigation
42:54asks Boeing to improve
42:55the reliability of the
42:57system.
42:57We learned a lot about
42:59the radio altimeter
43:00system and how it
43:02affects the autothrottle
43:03system.
43:04Boeing is currently
43:05working on system
43:07improvements to prevent
43:11this type of autothrottle
43:14event from occurring again.
43:18But the report also
43:20faults the crew for not
43:21noticing that their airspeed
43:23was dropping dangerously low
43:24in spite of the fact that
43:26there were three pilots on
43:27board.
43:28Forget that you've got an
43:29autothrottle.
43:30You look at the airspeed and
43:32you look at the altitude.
43:33You look out the damn
43:34window if you want to, but
43:36airspeed is a crucial
43:38component of staying alive in
43:40an airplane.
43:41And you always need to know
43:42what your airspeed is.
43:44But to Mika Ensley, the
43:45crew's failure to monitor
43:46instruments is entirely
43:48understandable and may be
43:49more the fault of the
43:50instruments than the crew.
43:52It's actually very difficult
43:54for people to be monitors
43:56of automation.
43:57One of the things that
43:57people don't do a good job
43:59of actually is monitoring.
44:00We're very good on-the-spot
44:02decision-makers.
44:03We're very good at coming up
44:04with creative solutions to
44:06problems.
44:07But repetitive monitoring is
44:09the kind of thing that
44:09actually people aren't very
44:10good at at all.
44:11So what we have to do is
44:12design automation to work
44:14with people in a way that
44:15keeps them more actively,
44:16cognitively involved and in
44:17the loop and not just
44:19monitoring a piece of
44:19automation to say, is it
44:21doing what it's supposed to
44:22be doing?
44:23Who's the ultimate computer?
44:25The pilot.
44:26The individual who should be
44:27able to say, I don't know
44:28what this thing is doing to
44:29me, but I'm punching it
44:30off and flying the airplane.
44:31Fly the jet.
44:33Probably the smartest thing
44:34we ever learned to say in
44:35our training in the
44:36airlines, fly the jet.
44:37Do that first or nothing
44:38else counts.
44:40That's what they forgot to
44:41do.