- 30/5/2025
This special looked at disasters caused by faulty data.
Categoría
📺
TVTranscripción
00:00At the heart of every modern aircraft is a computer that's vital for safe flight.
00:06Simply put, it'd be nearly impossible to fly this airplane safely without flight computers.
00:14But those computers rely on a steady stream of accurate data.
00:19Starting to pass through the layers.
00:21When they don't get it, the results are sometimes catastrophic.
00:30What are they doing so low?
00:32My airspeed indicator's not working.
00:34Something's wrong here.
00:40What was that?
00:41We gotta get out.
00:44If they can't figure out what went wrong, this could be the end of the B-2 bomber program.
00:52How can computers designed to aid pilots hinder them instead?
00:57Investigators search for the cause of three accidents triggered by bad data.
01:03You don't just fly with a stick and a rudder and look out the window anymore.
01:07We gotta die.
01:08We gotta die.
01:09We gotta die.
01:10We gotta die.
01:13Oh, fuck.
01:16I think one zero, three other yards.
01:18I'll come back tonight.
01:19I'll wait and see you at the end.
01:20Yeah, zero.
01:21We gotta die.
01:22You're good to go, sir. Thank you.
01:40Major Ryan Link and Captain Justin Grieve
01:43are ending a four-month deployment at Anderson Air Force Base in Guam.
01:48Okay, let's head home, then.
01:50Copy that.
01:51They were beginning a 16-hour continuous flight
01:54back to Whiteman Air Force Base, which was our home base.
01:59They're returning home in a unique warplane,
02:02a B-2 bomber called the Spirit of Kansas.
02:07Tower of Death 5, start-up is complete.
02:09You can activate our flight plan.
02:11It's a 7,000-mile journey home from Guam
02:14to Whiteman Air Force Base in Missouri.
02:16Major Link is a qualified instructor on the B-2.
02:22Captain Grieve is the mission commander,
02:24with more than 2,500 hours of military flying experience.
02:30First time, CT.
02:30The very first time I flew to B-2, I felt like I was a bird.
02:41It was so smooth.
02:481.45, rotate.
02:49Next thing they know, they're going straight up.
03:04Full power.
03:07The B-2 is only 80 feet off the ground and losing speed.
03:11The plane shakes violently.
03:16It's on the verge of stalling.
03:19From my 15 years of flying the B-2,
03:22the only time I felt the B-2 shake is in the simulator.
03:31Grieve realizes they're heading for the ground.
03:33We've got to get out.
03:50I'm stunned.
03:51The B-2 has gone through 19 years and three wars without a crash.
03:55This is unheard of.
03:56With the B-2's price tag of $2 billion,
04:01this is the most expensive aviation accident of all time.
04:05We want to know what the heck just happened.
04:08Operations of the entire B-2 fleet are suspended.
04:13The U.S. Air Force appoints Brigadier General Floyd Carpenter
04:16to lead the investigation.
04:18You're never going to believe this.
04:22Video of the takeoff shows that as the plane lifted off,
04:26it pitched nose-high.
04:29Your first reaction when you see that airplane pitch up
04:32is like, what are they doing?
04:36Investigators study the mechanical and computer systems
04:39that control the jet.
04:40Perfect.
04:42What about the actuator?
04:50Okay.
04:51They quickly determined that all the flight control surfaces
04:54were functioning on takeoff.
04:56The engines were not a problem.
04:59The hydraulics weren't a problem.
05:00The flight controls were not a problem.
05:03It's a beautiful sunny morning.
05:05Why does a bomber pitch its nose up
05:06and crash in a fireball on the runway at Guam?
05:10Something happened to cause this airplane to pitch up.
05:13The heart of the B-2 bomber
05:15is its computerized flight control system.
05:23On a B-2, pilots tell the computer what they want the plane to do,
05:28and the computer determines how to achieve that.
05:32Simply put, it'd be nearly impossible
05:35to fly this airplane safely without flight computers.
05:40So did something go wrong with that system?
05:44The investigators turned to data from the bomber's flight recorder
05:48to discover what happened.
05:49The nose is lifting off the ground,
05:53but the plane registers a pitch down of minus 8 degrees.
05:58Now, obviously,
05:59this plane is not pitching down,
06:03but the computer thinks it is.
06:05That is why I tried to lift the nose higher and higher and higher.
06:09Until it stalled.
06:10So, airspeed, climb angle, and altitude are all off.
06:21How does that happen on this plane?
06:26General Carpenter wants to know
06:27why the flight computers were getting faulty data.
06:32Let's see what the pilots can tell us.
06:40We started up at 9.15.
06:42A few minutes later,
06:43Major Link saw the calibration message during start-up.
06:47Mission Commander Grieve tells investigators
06:49that the pilots received an unusual computer message.
06:54Hey, Chief, we're seeing an Air Data Cal message.
06:58Air Data Cal stands for Air Data Calibration.
07:02In the simplest terms,
07:04the Air Data Calibration gives the aircraft
07:07its orientation to the universe.
07:10There are 24 sensors near the nose of the B-2
07:15that constantly measure air pressure.
07:18The plane's computers use those readings
07:21to calculate altitude, airspeed, and angle of attack.
07:25If they get out of balance,
07:27it calls for an Air Data Calibration,
07:29and the pilots call out maintenance.
07:32Can you put it in maintenance mode for me?
07:34If any of those 24 sensors provides an air pressure reading
07:42that differs significantly from the others,
07:45a recalibration is needed
07:47to bring them all back into agreement.
07:50You're good to go, sir.
07:54Pitot heat is on.
07:57Less than an hour after the recalibration...
07:59Clear right.
08:01Good to line up.
08:02The pilots maneuver the massive bomber
08:04to the start of the runway.
08:08Okay.
08:09Go on.
08:11Everything was 100% routine
08:13until we hit 100 knots.
08:14That's when we got the FCS master caution.
08:18Before I could even push the button,
08:20it rescinded itself.
08:21Stop or go?
08:26Go.
08:27Warning or send it.
08:30In that scenario,
08:31if I was the pilot in command,
08:32I would continue.
08:33Why?
08:34It's not safety of flight.
08:37When we hit 145,
08:38I called for Major League to rotate.
08:41Then it all very quickly went to hell.
08:51Investigators delve deeper
08:54into the calibration performed that day.
08:59You're good to go, sir.
09:02These three weren't just off by a little.
09:05They are way off.
09:08They discover that three sensors on the B-2
09:11were providing incorrect air pressure readings.
09:13100 knots.
09:15And that a master caution alarm
09:17flickered on for a few seconds
09:19just before takeoff.
09:20At that point,
09:24we really didn't know
09:25how they're related,
09:26but we figured they might be.
09:28What was that?
09:30FCS caution.
09:34The recalibrated sensor
09:35started providing faulty air data again.
09:38But six seconds later...
09:40Stop or go?
09:41Go.
09:42The flight computer resolved
09:44the sensor discrepancy
09:45and canceled the warning.
09:49What is going on
09:50with this plane's sensors?
09:54Okay.
09:55So,
09:56let's start at the top.
09:59Why the need to recalibrate
10:00in the first place?
10:01Huh.
10:10So,
10:11they got delayed by a day.
10:15Investigators learned
10:16that while the pilots
10:16waited to resume their mission,
10:19the spirit of Kansas
10:20sat on the tarmac.
10:24Okay.
10:24so the jet was left outside
10:28for 24 hours
10:29because of the delay.
10:31So what?
10:33Weather records show
10:35that the night before the accident,
10:37a tropical rainstorm
10:38settled over the base.
10:42Tell me,
10:44a rainstorm
10:45didn't take down
10:45a $2 billion airplane.
10:47tests confirmed
10:54that the sensors
10:54exposed to the heaviest rainfall
10:56became saturated
10:58and needed recalibrating.
10:59They did the recalibration.
11:14You're good to go, sir.
11:15Why did the recalibrated sensors
11:17then produce bad data?
11:20But they still ended up
11:21with faulty data.
11:23They nearly killed it.
11:32Full power.
11:34A state-of-the-art
11:35B-2 bomber
11:36stalled on takeoff
11:37because of faulty data
11:39about its climb angle.
11:41U.S. Air Force investigators
11:43still can't explain
11:44why that happened.
11:45And the pressure is on
11:47to find answers.
11:48If they can't figure out
11:50what went wrong,
11:52this could be the end
11:53of the B-2 bomber program.
11:57OK.
11:58So,
11:59they recalibrate
12:01at
12:0109.34.
12:07Nearly an hour later,
12:10the master caution
12:11lights up
12:12because of an air data issue.
12:14So,
12:15what happens
12:16in those 56 minutes?
12:19General Carpenter
12:21reviews what the pilots
12:22told them about the flight.
12:24Routine taxi.
12:26They hit the pitot heat,
12:28wait for the timer,
12:30and they're off.
12:33Pitot heat
12:34is a system of small heaters
12:36connected to each
12:37of the plane's sensors.
12:38Just before takeoff,
12:39the pilots activate
12:41the heaters
12:41so the sensors
12:42don't freeze
12:43when the plane
12:44reaches colder temperatures
12:45at higher altitudes.
12:49Pitot heat is on.
12:52General Carpenter
12:53asks an engineer
12:54who built the system
12:55if the pitot heat
12:57could have affected
12:58the recalibrated sensors.
13:00Clear right,
13:01good to line up.
13:01And then
13:03they turn on pitot heat
13:05when they get
13:05to the runway.
13:07He was really concerned
13:09when he saw
13:10and understood
13:11like he only could
13:13that moisture
13:14in the system
13:15and the data
13:15that we were putting
13:16into it
13:16to fix it
13:17could cause
13:18such a problem.
13:20Recalibrating
13:21the sensors
13:21brought the three wet ones
13:23in line with the others.
13:25But turning on the pitot heat
13:26evaporated the moisture,
13:29putting the sensors
13:29back out of alignment.
13:32So the data
13:33that was put in
13:34on the calibration
13:35now is invalid again
13:36and the flight computers
13:37are now trying to resolve
13:38the issue
13:39with these sensors.
13:50What was that?
13:52FCS caution.
13:55So then
13:56why did the faulty
13:57air data warning
13:58disappear so quickly
13:59after it came on?
14:01It's seconds
14:02from takeoff.
14:05So it has to make a choice.
14:10The B-2's
14:11flight computer
14:12is constantly receiving
14:13four separate
14:14data streams
14:15from all 24 sensors.
14:17If there's a discrepancy
14:19in the data
14:20from those streams,
14:21the computer
14:22is programmed
14:22to select
14:23any two
14:24of the channels
14:25to proceed.
14:28It just chose wrong.
14:29The flight computer
14:30selected a channel
14:31that contained
14:32incorrect information
14:33about the pitch angle
14:34from the recalibrated sensors.
14:36145, rotate.
14:3712 seconds later,
14:42the bad data
14:43caused the B-2
14:44to pitch up
14:45abnormally.
14:48Once they
14:49rotated the aircraft
14:50on speed
14:51as they thought
14:52and left the ground,
14:54they were along
14:55for the ride
14:55at that point.
14:57Full power!
14:57That steep pitch up
14:59caught the spirit
15:00of Kansas crew
15:00completely off guard,
15:02leaving them vulnerable
15:03to the effects
15:04of the faulty data.
15:07There's no way
15:08a pilot
15:09or a maintainer
15:10would understand
15:11the system
15:11well enough
15:12to realize
15:13what could have happened
15:14when they did
15:15that data calibration.
15:22Within two months,
15:23the B-2's
15:24flight computer
15:25was reprogrammed
15:26to address the fault.
15:28There have been
15:28no issues
15:29with bad data since.
15:32In the end,
15:33safety is everything.
15:35It doesn't matter
15:36if it's an airliner,
15:38a fighter jet,
15:39a space plane,
15:40or a stealth bomber.
15:43As sophisticated
15:45as aircraft have become,
15:47there are still situations
15:48where crucial information
15:50is communicated
15:51by radio.
15:52In those scenarios,
15:54a failure
15:54to check
15:55and recheck
15:56that data
15:56can result in disaster.
16:02Independent Air
16:03Flight 1851
16:05cruises above
16:06the Atlantic Ocean.
16:08There are
16:09137 passengers
16:10on board
16:11the Boeing 707.
16:14Santa Maria,
16:15Independent Air 1851,
16:17requesting that report.
16:19Flight engineer
16:20Jorge Gonzalez
16:21checks the weather ahead.
16:24He's a U.S.
16:25Air Force veteran,
16:27as is Captain
16:30Leon James Doherty.
16:33Let's request
16:34descent.
16:37Santa Maria control,
16:39Independent Air 1851,
16:41would like to descend.
16:42First officer
16:44Sammy Adcock
16:45started with Independent Air
16:46two weeks ago.
16:47The crew was a balanced crew
16:51that had flown together
16:53for about a week
16:54or ten days
16:55and they were doing fine.
16:56The flight from
16:58Bergamo, Italy
16:59will refuel
17:00at Santa Maria,
17:01an island
17:01in the remote Azores,
17:04before continuing
17:04to Punta Cana
17:06in the Dominican Republic.
17:08The airport
17:08at Santa Maria
17:09is not a busy one.
17:11It has no radar.
17:12There were only
17:13three scheduled
17:14airline flights
17:15per day.
17:15It doesn't justify
17:16the cost
17:17of air traffic
17:18control radar.
17:22You're cleared
17:233,000 feet
17:25on Q&H
17:261027.
17:28Runway will be
17:29one-niner.
17:31Expect
17:31Isla's approach.
17:33Runway one-niner
17:34report reaching
17:353,000.
17:36The controller
17:39is relying solely
17:40on position reports
17:42from the pilot.
17:43So they have
17:44to be meticulous
17:45about documenting
17:46the altitude
17:47that the aircraft
17:48is at.
17:51The crew
17:53needs to fly
17:53over the 2,000-foot
17:55Pico Alto mountain
17:56to reach the airport.
17:59On that day,
18:00the weather
18:00in Santa Maria
18:01was fine.
18:03There was just
18:03some cloud cover
18:05in the vicinity
18:06of Pico Alto.
18:0913 minutes
18:10from the airport,
18:11the pilots
18:11ready the plane
18:12for landing.
18:14Okay,
18:15I'm going to go ahead
18:16and put the ILS
18:16in on my side.
18:17Okay.
18:19ILS,
18:20or Instrument Landing System,
18:22is a navigation system
18:23that provides
18:24vertical and horizontal
18:25guidance to the runway.
18:28That means
18:29that he can fly
18:30in clouds
18:31until he gets
18:32on to final approach.
18:36As they close
18:37in on the island,
18:39an alert
18:39in the cockpit
18:40tells the pilots
18:41that they are
18:42500 feet above
18:43their selected altitude.
18:45Don't know
18:45if we're going
18:46to get visual
18:46or not here.
18:50Starting to pass
18:50through the layers.
18:53But as they enter
18:54the cloud bank
18:55over the island,
18:57disaster strikes.
18:58I can't keep
18:59this damn thing
19:00level.
19:02Pull up.
19:05Pull up.
19:08Pull up.
19:18Four and a half miles
19:20from Santa Maria Airport,
19:22Portuguese firefighters
19:23and civilians
19:24discover what remains
19:25of Independent Air
19:27Flight 1851.
19:29The jet hit
19:30Pico Alto
19:30very near
19:31the top
19:32of the mountain.
19:35The accident site
19:37was a chaotic scene.
19:39The wreckage
19:40was thrown
19:40all over the mountain.
19:42All 144 passengers
19:45and crew on board
19:46are dead.
19:53By daybreak,
19:54the Portuguese
19:55Civil Aviation Authority
19:56arrives at the crash site.
19:58They are soon joined
19:59in the Azores
20:00by investigators
20:01from the U.S.
20:02National Transportation
20:03Safety Board.
20:07So, this is the flight path.
20:10The plane impacts
20:12the east side
20:13of Pico Alto.
20:15Correct.
20:16Together,
20:17they examine
20:18the wreckage pattern.
20:20They're all concentrated
20:21right here.
20:22sending debris
20:26over the mountain ridge.
20:30Scars on the ridge
20:31offer the first clues.
20:34Looks like low-angle,
20:35high-velocity.
20:37They must have been
20:38flying level.
20:39Sea fit?
20:40Looks like it.
20:43They quickly recognize
20:45this incident
20:45as a controlled flight
20:47into terrain
20:48or sea fit.
20:50What was the altitude
20:51where they hit?
20:531,795 feet.
20:58And Pico Alto?
21:00What does the
21:01Jeppesen shark show?
21:02It's listed as
21:031,936 feet.
21:06They struck the mountain
21:08140 feet below
21:09what was listed
21:10on their chart.
21:11What were they doing
21:13so low?
21:15The possibilities
21:16were, one,
21:17the crew was off course.
21:20Two,
21:21there was an altitude error.
21:22It was our job
21:23to find out
21:24which of those
21:24would explain the accident.
21:28What's the minimum
21:29sector altitude?
21:303,000 feet.
21:33The minimum
21:34sector altitude
21:35is at least
21:361,000 feet above
21:37all terrain
21:38within 25 miles
21:40of the airport.
21:41Plenty of room.
21:43Why did these guys
21:44descend so low
21:45when they knew
21:45by the charts
21:46that the minimum
21:47altitude was
21:483,000 feet
21:49and there was
21:50a mountain there?
21:53What was the
21:54minimum sector altitude
21:55you gave them?
21:59The team checks
22:00to see if the
22:01air traffic controller
22:02made an error.
22:053,000 feet.
22:07You're sure?
22:09Yeah, I even
22:09made a note of it
22:10on my flight strip.
22:14He was perfectly calm.
22:16He had a normal behavior.
22:18He thought he had
22:18done a good job.
22:20Were they advised
22:21to use the ILS?
22:23Yes.
22:26Expect ILS approach.
22:29Runway one niner
22:30report reaching 3,000.
22:31I couldn't find anything
22:35that explains
22:36what happened
22:37from his part.
22:38Maybe the altitude alert
22:39survived.
22:41Let's take a look.
22:43The altitude alert
22:44sounds a warning
22:45when the plane
22:46approaches
22:46a pre-set altitude.
22:55Investigators discover
22:56that the altitude
22:57was set
22:57to 2,000 feet,
22:591,000 below
23:01the required altitude
23:02of 3,000 feet.
23:04This shocked
23:05all of us.
23:06We just couldn't believe it,
23:08how that could happen.
23:11They hope
23:12the cockpit voice recorder
23:14will reveal
23:14why the first officer
23:16input the wrong data
23:17into the altitude alert.
23:22Independent 1851,
23:24roger.
23:25You're clear
23:25to 3,000 feet
23:27on QNH 1027.
23:30Clear to 2,000.
23:33Did the first officer
23:34just say 2,000?
23:36Maybe he heard
23:372,000,
23:383,000
23:38and mistook it
23:40for 2,000.
23:42The controller
23:42should have corrected
23:43him too,
23:44but doesn't.
23:46Why?
23:47Hang on a second.
23:53The first officer's
23:54readback of 2,000 feet
23:55isn't in the controller's
23:56transcript.
23:57So he never heard it?
23:59How is that possible?
24:06Investigators compare
24:07the transcripts
24:08from the plane
24:09and from the tower.
24:14Hang on.
24:17They were talking
24:24at the same time.
24:25Yeah.
24:26They canceled
24:26each other out.
24:29You're clear
24:30to 3,000 feet
24:32on QNH 1027,
24:35runway 19er.
24:36It was stunning.
24:50We were stunned
24:50to finally understand
24:53how the two transmissions
24:55overlapped
24:56perfectly
24:57to cancel
24:58each other out.
24:58I was an investigator
25:00for over 30 years
25:01at the NTSB.
25:03I've never encountered
25:04anything like this before
25:05and I never encountered
25:06anything like this afterwards.
25:08But this discovery alone
25:10isn't enough
25:11to explain
25:12why the plane
25:13was so low
25:14when it crashed
25:15into the mountain.
25:19So the controller
25:20doesn't hear it.
25:20Investigators have discovered
25:22that critical communications
25:23about Flight 1851's
25:25required altitude
25:26were missed
25:27because of overlapping
25:29radio calls.
25:31But there's still
25:32another question
25:33that worries them.
25:35Something doesn't line up.
25:40The plane's altimeter alert
25:42was incorrectly set
25:44to 2,000 feet.
25:47Even at that altitude,
25:48they would have cleared
25:49the ridge by some 200 feet.
25:52What else affects altitude?
25:55Q and H.
26:00Q and H
26:01is an air pressure value
26:02pilots input
26:03so their altimeter
26:04can accurately measure
26:06the plane's height.
26:07That value can change
26:09with weather conditions.
26:10If the altimeter
26:12is set to a higher Q and H
26:13than is actual,
26:15the aircraft
26:15will actually be lower
26:17than what the altimeter
26:18is telling the pilot.
26:19It says here
26:21they were given
26:22a Q and H
26:23of 1019
26:24at 144 p.m.
26:28Then they were given
26:29a reading of
26:291027
26:31at 156 p.m.
26:35Investigators calculate
26:36the change in altitude
26:37between the two
26:38different settings.
26:40That Q and H
26:41brought them
26:42at least 200 people
26:43at 2,000.
26:44Right into the mountain.
26:45They asked the controller
26:55about the two different
26:57settings he gave
26:58to the pilots.
27:011019
27:02was on the Met report.
27:0512 minutes later
27:06when you gave them
27:07the 3,000 foot clearance
27:08you said
27:10Q and H
27:121027.
27:17Sorry,
27:17I don't know
27:18where I got
27:18that second number.
27:23Investigators
27:24turn to the CVR
27:25to see if the pilots
27:26caught the controller's mistake.
27:28After the first officer
27:30gets the minimum
27:31descent altitude wrong
27:33he questions
27:35the Q and H.
27:37Is that what they said?
27:381027 on the millibars?
27:39Yep.
27:42He was right
27:43to question it.
27:44But the captain
27:44just dismisses it.
27:47And the first officer
27:48just accepts it.
27:52Because of the different
27:53experience levels
27:54because of the prominence
27:56and assertiveness
27:57with which the captain
27:58said it.
27:59Shut off any discussion.
28:00Makes me wonder
28:01what else these guys missed.
28:04Investigators
28:05return to the cockpit
28:06voice recording
28:07to hear how the pilots
28:09briefed each other
28:10before approaching
28:11the landing.
28:12Altimeters.
28:14Set and cross-checked.
28:15Set and cross-checked.
28:17Seatbelts and shoulder harness.
28:23Secured.
28:25We're having fun now.
28:26Hey, hey.
28:27That's it?
28:30Stop the tape.
28:33They don't even mention
28:34the minimum descent altitude.
28:37Or pico alto.
28:38Here.
28:43The lack of detail
28:45is shocking.
28:49Had they done
28:50the approach briefing
28:51and reviewed the information
28:53and articulated
28:54the information
28:54as required.
28:56It's highly unlikely
28:58in my opinion
28:59that they would have
29:00struck the mountain.
29:02Starting to pass
29:03through the layers.
29:08Pull up.
29:10Terrain.
29:10Terrain.
29:11Pull up.
29:12The captain's ultimately
29:13responsible
29:14was the pilot
29:15in command.
29:16But each crew member
29:16failed in that respect.
29:18Pull up.
29:22Pull up.
29:23The pilot's
29:30investigators conclude
29:31that the pilots
29:32failed to follow procedures
29:33that would have caught
29:35the bad data
29:36provided by the air traffic controller.
29:38One doesn't necessarily
29:41completely trust
29:43any system
29:44or any data resource.
29:46The information
29:47that one receives
29:48either from one's senses
29:50or from the instruments
29:51of the aircraft
29:51or from outside
29:53navigational aids
29:53has to be balanced
29:55with one's own experience
29:57and judgment.
29:57When bad data slips
30:02into the second-by-second
30:03calculations
30:04that a pilot makes
30:06the consequences
30:07can be tragic.
30:14Gregorio Luperon
30:16International Airport
30:17in the Dominican Republic.
30:22Passengers on board
30:23Bergen Air Flight 301
30:25have a nine-hour flight
30:27to Frankfurt
30:27ahead of them.
30:31Captain,
30:32take your seats
30:32ready for takeoff.
30:35Exterior lights.
30:36The pilots prepare
30:37their Boeing 757
30:38for takeoff.
30:40Check.
30:40First officer
30:41Aikut Gergen
30:42has fewer than 75 hours
30:44in the plane.
30:46Ready for takeoff.
30:47Captain Ahmed Erdem
30:49is one of Bergen Air's
30:50most senior pilots.
30:52Alpha Luma Whiskey 301
30:54ready for takeoff.
30:55301
30:56ready for takeoff.
30:57Runway 08.
30:58Clear for takeoff.
30:59Runway 0830.
31:00Moulis Evrenesolo
31:02is the relief pilot.
31:04A good flight.
31:05Good flight.
31:06Good flight.
31:06Good flight.
31:07Power set.
31:09Okay, check.
31:10My airspeed indicator's not working.
31:18The captain's airspeed
31:23should read the same
31:24as his first officer's.
31:26120.
31:26But it doesn't.
31:29Is yours working?
31:31Yes, sir.
31:35You tell me.
31:37Erdem instructs
31:38his first officer
31:39to call out
31:39when the plane
31:40reaches takeoff speed.
31:42V1.
31:48Rotate.
31:48Shortly after takeoff,
31:59the captain's airspeed indicator
32:01appears to be functioning normally.
32:02Climb thrust.
32:04Climb thrust.
32:05Climb thrust.
32:09One minute and 30 seconds
32:11into the flight,
32:12the crew engages the autopilot.
32:14Center autopilot on, please.
32:17Center autopilot
32:18as command.
32:25Then the computer
32:26reports a problem.
32:28Rudder ratio
32:29mock airspeed trem.
32:31The alert warns
32:32that the plane
32:33is traveling too fast.
32:35Yes, trem.
32:37Something's wrong here.
32:41Captain Erdem's
32:42airspeed indicator
32:43shows the plane
32:44traveling at 325 knots.
32:47But the first officers
32:48indicates the plane
32:50going much slower.
32:54Mine shows only 200 now
32:55and decreasing, sir.
32:58Both of them are wrong.
33:01Now,
33:02a far more serious
33:03warning sounds.
33:08This alarm indicates
33:09the plane
33:10is approaching
33:10the maximum speed
33:12at which it was designed
33:13to fly.
33:15Now it's 350, yes?
33:20Captain Erdem
33:21decides to do
33:22what the alerts
33:22tell him to do,
33:24slow down.
33:26And let's take that
33:27like this.
33:32The result
33:33is terrifying.
33:34shortly after takeoff,
33:42Bergen Air Flight 301
33:43is in trouble.
33:47Cop!
33:48Cop!
33:48Cop!
33:49The cockpit
33:50fills with the ominous
33:52sound of the stick shaker.
33:56It means that the airplane
33:58is about to attain
34:00a speed so low
34:02so that it cannot
34:02sustain flight.
34:10The plane's
34:11attitude direction indicator
34:13shows that its nose
34:14is pointing dangerously high.
34:16A-V-I.
34:24Suddenly,
34:25the plane rolls hard
34:27to the right
34:28and begins to dive.
34:29Fast levers!
34:30Fast!
34:39Come on!
34:40Oh, what's happening?
34:49Bergen Air Flight 301
34:50vanishes from radar.
34:57No survivors are found.
34:59The Dominican Republic's
35:03Aircraft Accident
35:04Investigation Bureau
35:05asks the U.S.
35:07National Transportation
35:08Safety Board
35:09to assist in their
35:10investigation
35:11of the downed
35:12Boeing 757.
35:15Any signal
35:16from the recorders?
35:23It takes three weeks
35:25to recover
35:25Flight 301's
35:26black boxes.
35:27That flight data
35:32recorder
35:33was our key.
35:36Investigators
35:37immediately
35:37notice something
35:38unusual
35:39about the flight.
35:4015 degrees pitch
35:42nose up
35:42seems high.
35:46It's almost
35:46the maximum.
35:48Shortly after
35:49the autopilot
35:50is switched on
35:51the plane's
35:52nose
35:52pitches upward.
35:54They also see
35:56that the indicated
35:57airspeed
35:57is much higher
35:58than it should be.
36:00350 knots.
36:02It can't be right.
36:05They focus
36:06their attention
36:07on the sensor
36:08that feeds
36:09airspeed information
36:10to the gauges
36:11the pitot tube.
36:15When the plane
36:17travels forward
36:17an increase
36:18in air pressure
36:19inside the pitot tube
36:21causes the
36:22airspeed indicator's
36:23needle to move.
36:24But if a pitot tube
36:25becomes blocked
36:26it can send
36:28faulty readings
36:29to the plane's gauges.
36:30We started
36:32carefully looking
36:33at what could
36:35cause that
36:35kind of thing
36:36to happen.
36:38Regulations
36:39state that
36:39any time a plane
36:40will be on the
36:41ground for an
36:42extended period
36:42of time
36:43pitot tube
36:44covers must be
36:45placed on the
36:46tubes to prevent
36:47blockage
36:48usually by dirt
36:49or ice.
36:51Records
36:52indicate that
36:53the plane
36:53sat on the
36:54tarmac in
36:55Puerta Plata
36:55for 25 days
36:57before taking
36:58off for
36:58Frankfurt.
36:58I'd like to
37:02ask you
37:02some question
37:03about the
37:04breaking airplane.
37:05Investigators
37:06wonder if
37:07mechanics
37:07neglected to
37:08cover the
37:09pitot tubes
37:09after maintenance.
37:11Did you put
37:12the pitot
37:13covers back
37:14on them
37:14when the
37:15maintenance
37:15was finished?
37:17It didn't
37:17have any
37:17covers with it.
37:19We didn't
37:19take any
37:19off and
37:20we didn't
37:20put any
37:21on.
37:21and that's
37:29when we
37:29discovered that
37:30the pitot
37:31had not
37:31been covered
37:32for the
37:3225 days
37:33that the
37:33aircraft
37:34remained
37:34parked at
37:35the
37:35international
37:36airport
37:36in
37:36Puerto
37:37Plata.
37:38The
37:39investigation
37:39team
37:40concludes
37:40that in
37:41that time
37:42some sort
37:43of blockage
37:43developed in
37:44the pitot
37:45tube on the
37:45captain's side.
37:46Nobody knows
37:47for sure.
37:48The evidence
37:48is 7,000
37:49feet down
37:50in the
37:51Atlantic Ocean.
37:52A blocked
37:53pitot tube
37:53could have
37:54given the
37:54pilots bad
37:55data in
37:56the form
37:56of conflicting
37:57airspeed
37:57readings.
37:59But how
38:00could that
38:01lead to
38:01such a
38:02disaster?
38:04And what
38:05we've got
38:05for you
38:05is a
38:06computer
38:06signal.
38:06Using
38:07information
38:07from the
38:08black boxes
38:08the NTSB
38:09pieces together
38:10the flight's
38:11last moments.
38:13The captain
38:13actually realized
38:14that his
38:15speed
38:16indicator
38:16was not
38:17working.
38:17He was
38:18working?
38:18Yes,
38:18sir.
38:20The data
38:21shows that
38:22the problems
38:23on Bergen
38:23Air Flight
38:24301 began
38:25when the
38:26autopilot
38:26took over.
38:27Central
38:28autopilot
38:28on,
38:28please.
38:29Autopilot.
38:30Is it
38:31possible that
38:31the flight
38:32computer was
38:33just as
38:33confused about
38:34the airspeed
38:35as the
38:35crew?
38:36If so,
38:37how did
38:38that contribute
38:38to the
38:39crash?
38:42Right
38:42after the
38:43autopilot
38:43is engaged,
38:44the plane's
38:46nose begins
38:47to rise.
38:48Investigators
38:49have discovered
38:49that Flight
38:50301's autopilot,
38:52sensing that
38:52the plane was
38:53traveling too
38:54fast,
38:55raised its
38:55nose to
38:56reduce speed.
38:57It soon
38:58rises to
38:5915 degrees,
39:00then it
39:01stays there.
39:03The autopilot's
39:04a pretty smart
39:05guy.
39:05He already
39:06knows he's
39:06got all the
39:07power that
39:08he's going
39:08to get for
39:09the climb.
39:10The only
39:11thing for
39:12the autopilot
39:12to do is
39:13raise the
39:13nose.
39:15But the
39:16autopilot is
39:17reacting to
39:18faulty airspeed
39:19information from
39:20the captain's
39:21pitot tube.
39:22The autopilot
39:23gets its data
39:24only from the
39:25captain's pitot
39:26tube, the
39:28one that was
39:28blocked.
39:29Moments
39:29later, it
39:30sends out two
39:31different warnings.
39:32Rudder ratio
39:33mock airspeed
39:34travel.
39:35And the
39:36airplane begins
39:36picking up
39:37warnings.
39:38Rudder ratio,
39:39a variety of
39:40things that
39:40the airplane
39:41is sensing
39:42problems.
39:45Investigators
39:46realized that
39:47Captain
39:47Erdem wrongly
39:48concluded that
39:49both airspeed
39:50indicators were
39:51malfunctioning.
39:53In fact,
39:54his first
39:55officer's gauge
39:56was always
39:57correct.
39:57The plane
39:58was actually
39:59traveling much
39:59too slowly.
40:01What can
40:02we do?
40:04But that
40:04needle
40:05continued to
40:06climb around
40:07the clock
40:08face.
40:08until it
40:09activated the
40:11overspeed
40:12warning.
40:14The autopilot
40:15system sends
40:16a warning that
40:17the plane is
40:17traveling too
40:18fast.
40:19But in
40:20reality, it's
40:21slowing down.
40:25They're going
40:26slow.
40:27They think
40:27they're going
40:27too fast.
40:28And confusion
40:29has set in.
40:30Let's pull the
40:31airspeed.
40:32Let's see.
40:33That's when
40:34Captain Erdem
40:35makes the
40:35gravest error
40:36of all.
40:37You can see
40:38that he now
40:38pulls back
40:39on his
40:39throttles.
40:40At the
40:41plane's
40:41already slow
40:42speed.
40:50Pulling back
40:51on the
40:51throttles
40:51puts the
40:52plane on
40:52the verge
40:53of stalling.
40:54And that's
40:55when he gets
40:55the stick
40:55shaker.
40:57God!
40:58God!
40:59God!
40:59The autopilot
41:00is programmed
41:01to always
41:01disconnect
41:02when the
41:03stick shaker
41:03activates.
41:04With one
41:05warning telling
41:06him the plane
41:06is flying too
41:07fast, and
41:08another indicating
41:09it's flying too
41:10slow, Captain
41:11Erdem is
41:12overwhelmed.
41:13That
41:16Mako airspeed
41:17warning horn
41:18combined with
41:19the stick
41:19shaker was
41:20a tremendously
41:21mind-boggling
41:22experience.
41:24They're in
41:24direct opposition
41:25of each other,
41:26and two warnings
41:27that you would
41:27never expect to
41:28get, one right
41:29behind the other.
41:30In response to
41:31the investigation's
41:32findings, the
41:33FAA asks Boeing
41:35to change some
41:36of those warnings.
41:37Those changes
41:39include the addition
41:40of a new warning,
41:41which alerts both
41:42pilots when their
41:43instruments disagree.
41:46In addition,
41:47Boeing modifies
41:48its planes so
41:48that pilots can
41:49easily choose
41:50which pitot tube
41:51the autopilot is
41:52using for airspeed
41:53readings.
41:55You can't fly a
41:57modern aircraft
41:58without the proper
41:59data, because you
42:00don't just fly with
42:01a stick and a rudder
42:02and look out the
42:02window anymore.
42:04Your data is
42:05your flight.
42:08More than
42:091,400 Boeing
42:11planes worldwide
42:12are affected by the
42:13new directives.
42:18One final question
42:20remains.
42:21What blocked the
42:22plane's pitot tubes
42:23and caused the
42:24airspeed indicator to
42:25malfunction?
42:26What's happening?
42:28With Bergen Air Flight 301's pitot tubes
42:33lost on the ocean floor,
42:35investigators look at the
42:37area surrounding the
42:38airport in Puerta Plata
42:39and think they've found
42:41a likely suspect,
42:43the mud-dauber wasp.
42:45when a mud-dauber is looking for an area to build its nest,
42:57it's looking for a site, a place that's more or less tubular.
43:01They didn't put covers on the pitot tubes, so there was an opportunity to get something
43:13like a mud-dauber in that pitot tube.
43:16more than ever, aircraft depend on complex flight computers for safe flight.
43:24Even the smallest scrap of bad data can prove to be deadly.
43:29Aircraft and aircraft systems have become much more sophisticated over the years.
43:39But even with the most sophisticated aircraft,
43:42it's essential and necessary for pilots to know
43:46how to deal with those systems,
43:48how to understand how they work,
43:50how to not use them if they're giving them bad data,
43:53and how to manage whatever they have to continue to fly the airplane.
43:59we're not sure then.
Recomendada
44:03
|
Próximamente
47:00
1:11:01
47:22
52:36