When Automation Goes Horribly Wrong | The Crash Of Iberia 1456

[@Top_Weeb]

I really like how you also cover accidents like this one that result in no fatalities.

[@kurtbuck99]

I'm sorry, I can't let you do that Dave. --HAL 9000

[@gomphrena-beautifulflower-8043]

Oh my gosh, what memories you stirred. I went to see 2001: A Space Odyssey at Belle Meade Theater in Nashville when it opened! I was about 10 years old; most of it was over my head at the time! But I understood it when I got a little older. Definitely before its time, this one.

[@karllegrand]

"I UNDERSTOOD THAT REFERENCE"

[@dennismayfield8846]

Spot-On, K.B.!

[@cryptoslacker-464]

LOL. See now way the Airbus was nicknamed the scarebus 🥶

[@antoniobranch]

This shows the pilots, nature and automation where all fighting each other.

[@cmonkey63]

Tail winds are no joke. In 1990 I was flying from Ottawa to Vancouver. The jet stream winds (headwinds going west) were so strong, we had to make a stop in Calgary for extra fuel. At night in a snowstorm with -14C. We were on final approach, so low you could make out the cars on the ground, when we suddenly dropped for a full 3 seconds. Quick thinking by the pilots got us back up, but my confidence in air travel never recovered after that.

[@PeterSwinkels]

While you would hope something like this never happes, at least everyone survived.

[@boyman999]

Loved MSFS footage such an easy game to look at! Keep up the awesome work!

[@RadioactiveSherbet]

Yeah, I was thinking that the computer shouldn't command a nose down when the plane is relatively level, and at too low an altitude. I can't see any situation where that's a good thing.

[@misham6547]

Aerodynamic stall because your not going to recover anyway without doing something right now

[@RubenKelevra]

Well it's kind of correct. The airspeed dropped suddenly and the angle of attack was very high. So a stall was kind of imminent. This system is in place cause pilots can do errors as all humans. In this case the wind basically fouled the planes computers long enough to cause it to take the wrong actions.

[@Milesco]

@@misham6547 : But you can't have the computer pitch the airplane down when it's so close to the ground, impending stall or not. And the pilots had commanded max engine power, so that in itself would help to avoid a stall. And if it isn't enough -- well so be it, but when you're close to the ground, the last thing you want to do is pitch the plane _down._ It's the difference between a possible crash and a certain crash.

[@ronniewall1481]

11 VIEWS A MINUTE. YOU DO GOOD YOUNG MAN.

[@MiniAirCrashInvestigation]

Thank you kind sir!

[@manojbala6870]

As an A320 trainer, things take us by surprise even now. Please don’t blame the crew Edit: Hurts to see people blame the crew. Ask the people if they can understand Aoa, more less the formula. Dear captain, If you’re reading this, you got caught in Murphy’s law

[@JamesF0790]

I mean to be fair the fact they were both flying the plane directly seems like a failure of CRM isn't it?

[@DaveChimny]

That same question came to my mind immediately after hearing it: Was there any reason to make inputs on both sides? I'm just an armchair pilot and don't know the full functionality of an Airbus cockpit but what I know is that there's no feedback of what the other pilot is doing with his sidestick. So I think it's mandatory to call out "My control!" or "I have control!" when you want to intervene. This should be the same procedure in a Boeing cockpit but the advantage is that you have the yoke in front of you clearly showing what the other pilot is doing.

[@neillp3827]

That damn plane is far too automated. Pilpts are basically trained as software trouble shooters rather than pilots when trained to fly this bird

[@DaveChimny]

​@@neillp3827 I'm not a fan of Airbus, but even I wouldn't agree here: The hardest part of being a pilot today is to understand systems. And that's the same on Airbus, Boeing or Embraer. You have to know what a plane is capable and what it can't do. That's why the pilots of Asiana 214 crashed - they didn't understand the auto throttle system of their 777.

[@patrickmollohan3082]

I'm not going to blame the crew for anything. They rode it out as it came. My problem.is with Airbus!! Making their planes think that they know more than pilots do. If the automation would have completely given ALL aspects of the controls to the pilots, even cutting off the Stall protection, they wouldve had the fine elevator control to.let the crew ride it out completely!! Those guys were basically being fed a big shit sandwich as it was!! They couldn't properly do a Go Around, nor could they land w/o the nose gear slamming the runway. I say "well done men" considering what the plane was doing. It could have easily went the other way.✈✈

[@Rhaman68]

Well, as a former Airbus Capt U must comment. The issue was not automation but crew coordination and discipline. Early in training the dangers of both pilots using the side stick and the nose wheel control simultaneously. The video explains the reasons why. Why were both pilots flying the plane when only one is supposed to be the flying pilot, the other monitoring?

[@michaelmcintyre8250]

As a retired A-320 (and other) Captain, I agree that the PNF should have kept his hand off the sidestick. Nothing good can come from having two pilots making inputs. Air France had a similar problem over the Atlantic several years ago.

[@tissuepaper9962]

Forcing the plane to nose down while in landing configuration, at low altitude, and in mostly level flight is a serious bug in the automation. The dual inputs were a contributing factor, but the primary issue was with the automation. Stalling out early over the runway and landing hard on the main gear seems a lot better than hitting the runway nose first and collapsing the nose gear.

[@Milesco]

Ramon Cardona said: _"Early in training the dangers of both pilots using the side stick and the nose wheel control simultaneously."_ [sic] Then why does the Airbus design allow it?

[@Milesco]

@@michaelmcintyre8250 _"Nothing good can come from two pilots making inputs. Air France had a similar problem over the Atlantic several years ago."_ Exactly right! Indeed, that very crash prompted me to wonder at the time why Airbus would ever allow both pilots to control the airplane at the same time. That's a recipe for disaster. "No man can serve two masters." Matthew 6:24

[@tissuepaper9962]

@@Milesco I think the reason they have the automation consider both inputs is for "oh shit" moments when the PM can't keep him/herself from grabbing the stick and yanking back because the situation is so critical and scary. If both pilots give the same input at the same time that's a good indication to the plane that, whatever that input was, it needs to be acted upon *now* and not balanced against safety factors etc. by the automation. The other reason that I see as plausible is that most smaller planes have the pilot/copilot controls physically attached together, so the pilot/copilot can feel and resist each other's inputs if they're both on the controls. Airbus might have wanted to emulate that kind of dynamic when no priority is set. The situation I'm imagining is that the pilot has become incapacitated with his hand pushing the side stick full forward. If priority is set, the copilot can't really react to the pilot's erroneous inputs without flipping a switch (I think priority is set with a physical switch, I can't remember). If no priority is set, the copilot can use the sidestick to "cancel out" the inputs from the incapacitated pilot while trying to find and flip the switch.

[@matthewrowe9903]

As some one with decades in aviation accident research you do a fantastic job great vids informative and detailed keep up the good work

[@supertekkel1]

Sidestick controls alway's intrigued me; The control collum as in Boeing aircraft is much more intuitive. Each pilot can see what the other is doing, plus it gives manual feedback on what the plane's automation is doing. Think only of Airfrance flight 447.

[@melglobus]

I actually really loved the maths, thanks for taking us through the AOA calculations that the computer code does!

[@LunaticTheCat]

Me too! I got really excited when he pulled up the formula lol

[@Milesco]

@@LunaticTheCat Nerds! NERDS!!!!! _;-D_

[@LunaticTheCat]

@@Milesco Guilty as charged lol

[@franciscotrigo9656]

Very well explained!

[@MiniAirCrashInvestigation]

Glad it was helpful!

[@steves659]

Attempting to make a plane that is 'idiot proof' has been shown to be a factor in and of itself that leads to accidents. Would be interesting to see just how many times the technology saved the day versus causing the issue. Personally, I preferred the Boeing philosophy that gave us pilots the direct control of the aircraft.

[@FelixIsGood]

Pretty sure automation saved the day more than causing issues, also flights are now much more safe than 50 years ago. As human you can't keep an eye on every parameter in the cockpit a computer will always keep an eye on everything. As a computer scientist i can tell you that we even might be able to let an AI completely start, fly and land in a way which no human can do. It's more like a matter of time and a matter of the humans accepting it.

[@steves659]

@@FelixIsGood As an Airline Transport Pilot will say that humans make lousy monitors ... yes you can feed a monkey enough bananas to fly an airplane but the pilot is there for judgment.

[@FelixIsGood]

@@steves659 That is the misunderstanding most people have about AI. AI is much more capable than any human, if you are interested you can look up Deep-Mind or OpenAI, both are solving taks on a super-human level. Don't get me wrong there might be situations where 0.001% of humans are better than the AI, but they are still much better than the most. Judgment is by logic nothing only humans are capable of.

[@mathewcherrystone9479]

@@FelixIsGood AI today is only better than humans in specific tasks and only if it was trained for this exact task and often enought only for the data it was trained on. If anything unexpected happens the AI is usually screwed and doesn't know what to do.

[@MotorcycleWrites]

@@FelixIsGood AI is lightyears better than humans… at tasks which humans have to define and create the AI for. You can’t just throw a genetic algorithm at something like flying an airliner, you have to design each system deliberately and the pilots need to be aware of all of the things which the systems can and can’t do. It’s a nitpick because you’re right that we *can* design AI which is better than humans, that doesn’t mean we always actually *do*.

[@GaryNumeroUno]

Sounds like CRM broke down... the First Officer was pilot flying. The Captain should not have interfered or instead made the call... "I have control". Hopefully lessons were learnt.

[@ianjones3562]

Excellent content once again - your research is always top notch. Would love to hear details on the Korean Air 747 cargo 8509 crash that happened at Stansted UK on 22nd December 1999 . I was in the area at the time at a supermarket and heard the engine sound from having just taken off .

[@lst141]

What about the saying “I have control”

[@jjaus]

The Captain should have pushed the priority button and said that.

[@RubenKelevra]

@Night Rider nope. You can override the computer if necessary

[@stormtrooper7177]

I like how it takes KZbin 13 hours to notify me about your videos...

[@patjohn775]

It’s because this talks about death and youtube is a huge mess so they can’t figure out how to separate this from stupid videos.

[@gettothepoint2707]

5:21 Traumatic High school flash backs. 😂😂😂 Thanks for that 👍

[@smcdonald9991]

Differential equations in high school?

[@Talguy21]

Thankfully, nobody had to die to teach this lesson.

[@supermaster2012]

I'm from Bilbao and this kind of wind pattern is super common when approaching Loiu Airport from either direction as the airport is essentially in the middle of a valley, parallel to its course. The airport used to be in Sondika where this wasn't that problematic but when the airport was moved to Loiu this became common. There's plenty of videos of aircradt landing at Loiu with almost 45 degrees of yaw just to couteract the wind.

[@isaaclao2380]

Yess, I am waiting for another video, keep up with the good work

[@atakorkut5110]

I haven't watched in a couple week sorry I've been busy but I'm loving the new videos man catching up now and they're great to keep it up cant wait till u hot 50k subs. i always share videos with my dad and all his mechanic buddies at the airport

[@RubenKelevra]

Kind of strange to hear that the crew flew the plane together. This kind of makes sense if the control columns are linked and you can push the input a bit further or back if you train someone. But in normal operations there shouldn't be two pilots flying.

[@crazyjhey8050]

That is one thing about airbus that really sucks the side-sticks are like game controls u can’t feel what the other stick is doing or even what the aircraft is doing. There are some private jets that have side sticks that are in sync with the other but this technology has not been used on any airbus aircraft. Boeings control columns are mirrored and even on the fly by wire aircraft they seem to do way better on control feeling

[@RubenKelevra]

@@crazyjhey8050 well if you use both sticks on an Airbus you get an audible warning: "dual input". You also got override buttons which you can use to temporarily override the other stick (are both located in the same place). Also you do have feedback on the sticks. So not sure what you're complaining about. :)

[@crazyjhey8050]

@@RubenKelevra first of all I work on the aircraft so if u don’t then u only know what what ever u read about. The artificial fell on the on sticks is nothing more than a stiffer feel as u move the stick thru the range of movement. That fell is the same if u are on the ground or if ur cruising at speed it’s artificial and it doesn’t increase with speed. As far as duel input yeah the aircraft will bitch about it but in an emergency situation I guess u can discuss with the other pilot what exactly is going on right? If it wasn’t an issue no 1 would spent the money and time on technology to synchronize both side sticks would they? Idk if u realize it but aircraft have crashed because of this exact issue. Accidents that would been prevented if one pilot knew what the other was doing so it does seem like it might be a problem.

[@PeteHartley]

Thanks very much for these great videos!

[@CuratedPile]

One thing that the average person doesn't understand (I didn't until I learned) is that constant, horizontal wind does not affect airplanes, regardless of wind speed. The plane is moving in that airmass and it has no effects other than navigational - that is it may require course (heading) adjustments to get to the intended track over the ground, and the speed over the ground would be impacted. As such, winds of 55 knots below 6000' are not an issue by themselves and cause no stresses on the airplane. Now, add gusts and vertical movement and you will affect the airplane. This case looks like it was the vertical winds that caused the trouble.

[@darenbutler7504]

Great video. Very informative and excellent graphics. Well done. 😁

[@AurioDK]

Surprised not to have seen this in the Aircrash Investigation episodes, good narrative.

[@Jet-Pack]

When the angle of attack is too high the wing won't produce any lift. Incorrect, it should be "less lift" than just before the stall. The lift during a stall is significant and non-zero. Look at an airfoil plot for the lift coefficient Cl over a 360° angle of attack range. The only times the lift coefficient reaches zero is when the angle of attack is near zero (or slightly negative) or when the airfoil is at near 180° flying backwards.

[@togathrust1047]

Excellent video!! You know what video I'm waiting for :) Keep up the great work!!

[@michaelbuckers]

All that system behaviour notwithstanding, it probably spared the plane of a much harder impact, seeing how it prevented a stall and a sharp descent rate spike right before touchdown.

[@nicholasjohnson778]

No offense, but that is not true and counters the findings from the accident investigation. The A-max protection kept the aircraft from even experiencing the buffeting from an impending stall. So there was still some performance the pilots could have squeezed out of that situation, but Airbus software blocked them. It also caused the plane to absorb the entire force of the hard landing on its nose wheel. What I would agree with is that a stall while in uncoordinated flight would have been a worse outcome. Since that would have caused one wing to stall first and impact the ground. However, the need in this scenario was to reduce the rate of descent as much as possible and for the main landing gear to absorb the impact. The automation failed at both of these.

[@pilotopolar8641]

IT'S THE FIRST TIME I WATCH YOUR VIDEOS, AND I SEE YOU'VE DONE A GOOD WORK. CONGRATULATIONS AN HERE, I AM A NEW SUBSCRIBER

[@coca-colayes1958]

I woke up to this video ,great way to start My day

[@faz_the_plane_guy1027]

Great video keep up the good work

[@MiniAirCrashInvestigation]

Wait this video isn’t live how’d you find it 😂

[@faz_the_plane_guy1027]

I don’t know 😂

[@yaysimonsays151]

9:35 oh wow perfect timing that fix will definitely be important for a few days

[@svr5423]

Computer: Radio-Altimeter says I'm close to the ground, I'm also descending and in a nose-down attitude. Let me worry about the erratic AOA first, clearly the pilots should have other priorities than not crash into the ground.

[@TrentFalkenrath]

Reducing stick angle to climb out? Sounds like releasing the brakes slightly to regain traction in a car. How interesting.

[@sivalley]

Instinctual reactions can be very hard to overcome. To use your example, that's exactly why anti-lock brakes are now standard and work exactly as you described.

[@dunbar9finger]

It's not the same. It's the opposite. In a car when using brakes manually is when you back off the brakes to find the optimal achievable deceleration and when using ABS you just intuitively jam the brake pedal down. This is the exact opposite. With the airbus software the automation *causes* the need to do the counterintuitive thing, rather than implementing a naturally occurring counterintuitive thing for you so you don't have to like ABS brakes do. Here the counterintuitive thing is that the autopilot will not let you take control unless you let go for a second, letting the controls zero out first right at the moment you need to wrest control away from the autopilot that's doing everything wrong. The autopilot software made the mistake of not realizing that unlike when you're high up, when you are just above the ground about to land keeping the nose up is more vital than stopping the stall the sudden tailwind is about to cause. A hard landing on the main wheels because you stalled out a bit higher than you should on final flare is better than a hard landing on the nose wheel. The usual logic of "it's better to dive a bit than start a stall" is not true when your altitude is just a few feet above the ground on final touchdown.

[@sarowie]

Toyota had an "unintended acceleration" "issue". When for what ever reason (that part could never be figured out) the microprocessor freaks out, the supervising microcontroller will not restart the Control task, as long as there is either gas or brake paddle active. So... brilliant. When the micropocessor task freaks out and the car accelerates full throttle, all you had to do was simple fully remove the foot from gas and breaks and then break "as normal". Note: *Fully* release the break paddle while the car accelerates full throttle. If you think that made the jury sympathize with the victims, well: Lying to NASA (yes, NASA was involved in the investigation) about a fact (Error Correcting RAM for the processor) was also on the criminal chargers and you do not need technical expertise for realizing that lying to investigators is not a good thing.

[@VexifyGaming]

Woke up just in time

[@endtimestraveller7716]

It never ceases ot amaze me at all the things that can wrong." Cripes!! Not a good advert for flying.

[@eucliduschaumeau8813]

Mountain-effect rotor winds have been known to down aircraft on more than one occasion.

[@rubybitesthedust]

When you first mentioned AOA my brain autocorrected it to AoE (Attack on Everyone) and I was so confused

[@gilbertfranklin1537]

Hey, Mini ACI, it was thoughtful not to get the math too complex for us - but I have the impression that you are personally very good at it - correct?

[@cryptoslacker-464]

It was clear as mud to me 😆

[@bernardospinelli5588]

way better now with images that correlate with what's being said, congrats

[@cherriberri8373]

Damn, automation really fucked up that flight huh

[@CyberSystemOverload]

This sidestick confusion is concerning on these Airbuses. They need to make them ACTIVE sidesticks so both of them move at the same time (like a traditional yoke). This way tactile feedback is given to both pilots as to what the other is commanding. Well done to these pilots for salvaging a confusing situation where Murphys Law hit them hard. Fantastic video and excellent explanations as always.

[@colin-nekritz]

Damn! You’re getting REALLY really good at these! You’re becoming one of my fave aviation channels. You helped me understand as well what a joke of an aircraft Airbus made, as if the lame sidesticks weren’t enough of a laughable insult to real pilots and the root cause of many disasters that the plane is just a giant video game for wannabe pilots who would be just as well playing MS Flight Simulator 2020 than ferrying actual people.

[@timburke127]

Sidesticks are lame? F16, Rafale, F22, F35, Bell 525, all insults to real pilots?

[@tookitogo]

Uh huh, a “joke” that’s now the leading manufacturer of airliners... Probably 90% of a pilot’s knowledge is aircraft-agnostic. To claim that flying an Airbus is akin to flying MS Flight Sim is profoundly ignorant. (And I say this as someone whose favorite plane to travel on is the 767, not an Airbus.)

[@rewolff2]

Small addition: In the EU meteorological reporting reports "10 km visibility" whenever it is "10 km or more". In most likelyhood it was not "between 9.5 and 10.5km" of visibility but just "clear".

[@dennismayfield8846]

'Spoiling' us, with near-flawless, visual-audio-journalism!

[@antoniofdez620]

very good explanation.

[@neiss2]

Love your accident/incident reports, yet here I'd like to make an important correction. I've flown myself to Bilbao airport in similar conditions and the main issue with Bilbao is not the strong winds but the fact that the runway runs parallel to two mountain ridges, one south of the airport and the other north of it. When there are strong winds from the south, they create a particular turbulence throughout the whole ILS path down to the same threshold of the runway that sometimes forces the airport to remain closed even under CAVOK conditions, due to the strong turbulence. Landing in the (in)famous south-wind conditions at Bilbao is known to every passenger that lives in the city and forces the pilots to adjust as landing under windshear conditions, hence flying a highly automated aeroplane like the A320 wasn't the optimal thing until Airbus hadn't corrected its programming.

[@swtcaroline93]

Why wasn't the plane in landing configuration, or was it? Wouldn't that cancel the AOA protections? I really enjoy your videos!

[@Soordhin]

It was, and no, landing configuration in itself does not cancel AOA protections, it just adjust the values used for the present configuration. After all, AOA protection is there to prevent stalling the plane, which would have helped Turkish Airlines with their 737 in Amsterdam for example.

[@swtcaroline93]

@@Soordhin True, that was pretty close to the ground. Didn't think of it that way.

[@tomstravels520]

Yes it was in landing configuration but at the time the AOA protections were the same no matter what the altitude. Now the AOA protections are only disabled at about 200ft I think

[@mikehenderson631]

Wow man you went to a flight simulator that is so cool that you did that instead of using the videos of people send you for their up the airplane

[@cameronbooker445]

Thats a neat sentence. I'm hanging out drinking my own urine and smoking Clorox wipes trying to figure it all out.

[@jimenusky]

Yeah the Bilbao airport is infamous for its windy landings... I've only flown into it (as passenger) once and I was lucky enough to have a smooth touchdown lol 😣

[@campbelldutch75]

And hundred years from now and thousands more LIFES we will still be learning from mistakes lol

[@jessasto947]

A nice one to wake up to, thank you again! Do you know other videos you recommend with flight computer problems? this is fascinating!!

[@sharoncassell9358]

There are some on Mentour pilot Petter the Swede. Often the automation confuses the pilots or the autopilot cuts off unbeknownst to the pilots and didaster ensues.

[@mann2520]

First notification this hour😁😁😁since I unlocked my phone

[@AviationNut]

What sucks is that this was almost a brand new plane, Iberia just got the plane from Airbus just one year earlier. The aircraft had little over 1400 hours on the airframe and a little over 800 flight cycles. But i guess that's what insurance is for, it still sucks seeing a new aircraft crash, luckily everyone made it out alive.

[@sharoncassell9358]

Many times the plane is new and as pilots try to learn its habits horrific things happen. Its very unfortunate and I'm always happy when they save the plane & pax of course. They cost 20 million and up. God bless simulators prior to flight.

[@Oops-IMeantToDoThat]

What I don't understand, and I've heard this on a number of crash debriefs, is why BOTH pilots are flying. There us supposed to be 1 flying and 1 monitoring. And, for some reason, NTSB never comments on it as a contributing or main cause

[@robgoffroad]

This is why I hate computers that try to do your job. I have yet to find a traction control or stability control system in a car that I can't out perform... in a plane, like we saw here, it was the cause of the problem. I'm glad they had them attempt to fix it.

[@kilroy1964]

And the complicated part is that AOA is relative to the direction of the air flow? Not. Good vid though!

[@janipt]

I dont understand whats the problem here. Automation absolutely cannot allow pitch up if the AOA is too high because then the plane would stall and crash. Maybe im missing something

[@gilbertfranklin1537]

Altitude - when you are virtually on the ground all bets are off.

[@haschid]

The pilots should have went around after the first updraft. It was an unstable approach. And why in the hell were they fighting for the control of the aircraft? Terrible CRM. As far as I understand, automation saved everyone's life in this case.

[@sharoncassell9358]

No way Jose'. A Tesla csr just crashed into parked cars as it driver was asleep. People have to monitor the vehicle automation & correct it if it goes awry. We will not be replaced by robots.

[@aerohub6694]

Amazing! Even if it takes more time i still want more flight simulation!!

[@MiniAirCrashInvestigation]

Thanks I loved the sim as well haha

[@DelfinaKS]

Great video. I am just an aviation enthusiast. However, if I understand it correctly form your videos as well as those by other aviation educations, this particular issue would not have happened in a Boeing aircraft as the side stick and rudder inputs are direct and not automatically adjusted by the flight computer (except in the infamous MCAS used in the new MAX). I am not sure why manufacturers are going for more and more automation that is difficult to override.

[@shi01]

Well, statistically most aviation accidents happen because of human error. We always tend to think of the cases where the pilot(s) heroically safed the plane in one way or another and tend to blank out the cases where simple pilot mistakes where the main reason for a disaster. We don't often hear about the cases where automation actually safed the plane because there's nothing interesting to write about. And even in cases that did get covered by media, we tend to praise the pilots. The best example for this is the miracle of the hudson. Because purely from a flying standpoint Captain Sullenberger would have stalled the plane i think on three occasions purely from the inputs he has given on the stick, and only the envelope protections of the A320 keept the aircraft flying/gliding. Sullenburger did a great job no question, but it wasn't all his skill alone that safed the plane in the end.

[@DelfinaKS]

@@shi01 Automation is helpful but the problem starts when you the automation takes over and refuses to allow the person flying the plan to control it. If automation is given as a tool to help the pilot, that is good. If it is used as a replacement for investing in pilot training (like they did with the MCAS), that is a horrible strategy.

[@shi01]

@@DelfinaKS Well, that always depends on the circumstances. There are situations when not doing what the pilot wants is the right solution. MCAS is actually a bad example for this, because it didn't depend on any pilot input at all. It was mainly designed to conform with FAA regulations and to do that as cheaply as possible. You can't really compare that to a fully fledged fly by wire system.

[@DelfinaKS]

@@shi01 My point is that it is just plain stupid to design a system that ignores pilot input and not even inform pilots about it in their training. That was the central issue with MCAS. If there are systems that ignore pilot inputs, there should always be a manual override to that and pilots must also be informed how the system works. Otherwise, you should just get rid of pilots and use software to fly planes, but it is inhumane to have a pilot be a mute witness to a disaster planned by a programmer who wrote the automation.

[@lucas29476]

So this is the exact same issue as the (very minor) one affecting the last moments of the Hudson ditching?

[@Musikur]

No, in that case, Sully intentionally used the α-max protections to keep the nose as high as possible so they would touch the water at the lowest possible speed. Alpha protection was a hero on that day.

[@JamesF0790]

The tail already dug in on the Hudsun landing, if there hadn't been for the alpha protection when Sully ordered the plane nose up it would have hit at a steeper angle, dug in more and likely would have torn the plane apart.

[@jackwebb5917]

I still had high school flash backs!!!

[@gd8205]

MSFS2020 with replay. Nice. Videos are going to be so much better. Now you just need aaaall accident aircraft to be developed for the sim 🙃

[@moiraatkinson]

Sounds as if, rather than having a Flight Engineer onboard, planes nowadays should travel with a Systems Analyst..

[@tzadiko]

I don't understand why AOA protection doesn't seem to take into account extremely low altitude

[@maxsido2149]

Man you need more Subs ps did my Sugesstons arrive on yours

[@MiniAirCrashInvestigation]

Did you suggest a video? I’m sorry I don’t know

[@maxsido2149]

@@MiniAirCrashInvestigation yes 2 of them

[@MiniAirCrashInvestigation]

I’m sorry I didn’t get them can you post them here?

[@maxsido2149]

@@MiniAirCrashInvestigation the Panavia Tornado Droptank Incident and the Umea Gippsland 8 Crash

[@MiniAirCrashInvestigation]

Thanks! I’ll check them out!

[@virajtemghare3267]

it may be a good idea to intall a small camera to get a view like on 1:02 and relay it to the crew. It may help the crew to visually assess the situation in times of trouble

[@tissuepaper9962]

The last thing the cockpit needs is another data stream and another computer system to go with it, IMO.

[@mitalibose1527]

Surprised that he didn't make any videos on Ryanair

[@jlw22356]

That just flew right over my head!

[@LunaticTheCat]

Automation overall saves lives, but unfortunately when it goes wrong, it can also take lives. Luckily no one died in this crash.

[@donwald3436]

Airbus is HAL 9000.

[@funnyfarm5555]

Was the plane repaired or scrapped. Pilots disciplined, or retrained?

[@tronas8515]

I know the max has bad press. But to be fair any issue with the trim was a memory item well before the max issues for every 737 in service, and wasn't dealt with properly... But this kind of scenario, this is why Boeing is better. The pilots tried something and the plane said no.

[@JimDean002]

I'm not sure I go with the last sentence where we blame the wind. The wind has been blowing over there for a billion years. This is basically someone building an extremely complicated system which encountered landing conditions they had not programmed for. The end of the day the computer wasn't smart enough to know that it was pushing an aircraft nose down hard feet from the ground. That's a programming issue not a wind issue

[@isaaclao2380]

Also this made me remember that air france A320 airshow crash

[@MiniAirCrashInvestigation]

Oh yeah that was tragic :/

[@colin-nekritz]

Real planes have yokes between the pilots legs where god intended, none of this video game joystick crap.

[@Keestral]

@@colin-nekritz this has nothing to do with yokes vs side sticks but everything to do with flight control systems. Check you C.W Lemoine video on the report of a F35 crash that happened recently. The same thing basically happened, the pilot asked for something that should have saved the plane and the computer decided to ignore them.

[@isaaclao2380]

@@colin-nekritz UM Airbus use side stick, not a yoke... Except the A300 and A310

[@youtubeuser1052]

I'm not a pilot, but it's baffling to me that Airbus designed planes to accept input from both pilots at the same time. There should be a big obvious switch in the middle of the plane that determines which side has control. What good can ever come from both pilots providing inputs? If they are both providing correct inputs then it would be just as good to have the plane listen to only one of them. And if one or both of the pilots is providing bad inputs, combining that input with the other pilot isn't likely to have as good a result as just having the other pilot take over. The pilot monitoring should be doing exactly that, monitoring. If the pilot monitoring thinks the pilot flying is doing so badly, then the pilot monitoring should declare a go around or in the worst case, flip the big switch and take over full control.

[@tomstravels520]

You mean like the priority takeover button ON THE SIDESTICK

[@youtubeuser1052]

@@tomstravels520 That depends. Does that button make the situation described in the video impossible? Was the button the result of a redesign because Airbus realized that no good can ever come of both pilots providing input simultaneously and hoping that the sum of their inputs would be better than just one pilot flying? Because I'm still baffled by the fact that Airbus ever thought that "blending" the separate inputs of two pilots would ever be better than ensuring that only one person can ever be in control. I'm imagining what it would be like to drive a car if the person in the passenger seat also had a steering wheel and was able to increase or decrease the amount of wheel turn from what the driver intends.

[@tomstravels520]

@@youtubeuser1052 most likely as then the aircraft would have pitched up slowly as it would only be taking commands from the sidestick with the button pressed. The reason for the design is because if one pilot move their sidestick down and the other up....how is the computer supposed to know which one it should listen to so it takes the average of the 2. That’s why the button exists.

[@youtubeuser1052]

@@tomstravels520 Well, good to know that Airbus realized their design blunder and added that button to fix it. From the description in this video it sounds like it used to be optional. It sounds like they used to allow both pilots to make control inputs at the same time and left it up to the pilots to know that doing so was a bad idea. So my comment is really about "why did they EVER think it was a good idea to allow simultaneous dual input?" In a non-fly-by-wire plane it makes sense that in an emergency you might need both pilots to physically supply enough muscle power to move control surfaces, but when they switched to computer control they should have immediately discarded the idea of adding up or averaging the pilots and enforced a rule of "the computer will only ever obey one person at a time".

[@tomstravels520]

@@youtubeuser1052 no that button has ALWAYS been fitted and should have been pressed here as airbus knew originally it could be a possibility that both pilots would try to get control. You clearly don’t fully understand the control logic of the airbus. If one pilot moves their sidestick left and the other pilot moves their sidestick the exact same the. How is the computer supposed to know which one to obey? Then if one moves their sidestick to the right but other stays left...again who should it be listening to. Combining the inputs actually makes slightly more sense because then it is summing up both inputs and responding to both (but no further deflection that 1 sidestick). Only one pilot should be moving the sidestick at any time. If the other has to take over then say “I HAVE CONTROL” and press the priority takeover button so only your sidestick is obeyed

[@Sky10811]

When u use the term "heavy turbulence" r u referring to the same term in the sense which pilots use? Would be great to be consistent: did you mean heavy as moderate or as severe? thank u

[@sivalley]

Turbulence is a fluid dynamics term meaning "not smooth" so pilots saying there is turbulence is no different than saying rapidly changing winds. Your perception that there is a difference is the real question here. Not that it's your fault given how some people misuse terms all the time. 😊

[@MothaLuva]

There are three types of turbulence: light, moderate and severe. The term „heavy“ turbulence is not necessarily used. Check out the differences.

[@sivalley]

@@MothaLuva Sorry, but turbulence is turbulence. Aeronautical operator degrees of severity do not reflect on how turbulent flow behaves. The terms of severity were developed so those less adept at fluid dynamics could better categorize risk. But risk is risk, so that makes the categorizations moot. Aircraft have historically crashed due to even "mild" turbulence from the wake of another craft.

[@MothaLuva]

@@sivalley Get some flight hours in different kinds of turbulence, then we talk about this „turbulence is turbulence“ thing. And, if the wake of one aircraft is enough to crash another, there is nothing „mild“ about it. You can believe me, I have been in one wake (with an MD 80 during initial climbout, ending up banking almost 90 degrees right within a second, maybe second and a half). That’s the reason, why there is specific separation during approach and take off with specific „wake turbulence category“ airplanes (light, medium, heavy, super, depending basically on their MTOW). For aircraft operators it is not important how turbulent flow behaves, it’s important how violent the effects of it are.

[@imaPangolin]

Oddities like this sudden tailwind, human pilots can recognize and counter. I don’t think we are anywhere near close to pilotless airplanes.

[@MIchaelArlowe]

The wind hit it and the front fell off?

[@julianfierro7034]

Oh my god, I'm back in A320 Gen Fam

[@marvin4827]

No matter what you plan for, life don't go by the book...

[@GuillermoSTD]

Hehe, I had to smile when I was watching this video and it mentioned Bilbao, as that's where I live. It's not that they're very high at all, about 2,200 ft at their highest point, but the city is more or less surrounded by small mountains, so I can see how this peculiar orography can create pretty difficult conditions for takeoffs and landings at our local airport. Glad to see that this scary incident prompted a response from Airbus and served to improve air safety.

[@MiniAirCrashInvestigation]

Thanks for the context!

[@sarahalbers5555]

Guillermo., isn't that in the Basque region? I have always be fascinated by that area, rarely hear it mentioned. Off topic, but I named one of my cats Basque name. Thanks Mini, this was interesting. I flew the A320 alot, I know it had some eccentricities. Keep up the great work.!

[@markodom3841]

The theoretical benefit of most technology is that the plane won’t forget things or be asleep at the switch as humans can be - the common thread of most pilot error crashes which were most of the total crashes before FCCs. Thus, technology in both Airbus and Boeing has generally made air travel safer overall. But in this case where the pilots knew what needed to be done but the technology wouldn’t let them, then technology has overshot our overall objective. The programming of computers is based on a number of assumptions which is fine if all the assumptions are correct. The assumptions failed to incorporate this scenario.

[@Ghazibinhafiz]

In summary, the "protection" system of AOA has priority over pilot command of sidestick, prevent the pilots to take the correct action, which is to FLARE, last few seconds before touch down. Crew is not to blame in the case. However, Airbus should be blamed for this accident. Anyway, Software has been updated since then to convert aircraft to "Direct" law, which a term means the aircraft will be converted to "Normal" aircraft where the pilots have full authority to control the aircraft

[@theMoerster]

Ah yes...Bilbao Baggins Airport.

[@spddiesel]

Seems like the easiest fix to this is to have the automatic stuff turn off below a certain elevation, say 200'.

[@wilberburroughs3133]

I was behind them that day flying a Bae ATP

[@RubenKelevra]

Isn't there a warning "dual input" if two pilots use the sidestick to warn about that behavior?

[@busteraycan]

I think that alarm only triggers if the stick inputs are opposite.

[@planespotting8727]

I really like your videos. ❤️👍🏻 I am a planespotter and have tons of pictures taken mainly at Arlanda Stockholm. Let me know if you wanna use my pictures in your videos. If you’re looking for a specific aircraft, airline, type or general images. Just let me know 👍🏻

[@maureen-paulbarnes-vonkulm480]

Why didn’t ground detection cancel that?

[@Bobrogers99]

In times of conflict, does the computer know best? Does the pilot know best? Do either of them have all relevant data? Can they work together? It may be a coin flip.

[@sparkyobrian6417]

there is the crux of design, Airbus gives the final limits to the aircraft computers ( bank angle etc) ; boeing gives the pilots the final decision

[@SuperHeatherMorris]

@@sparkyobrian6417 In your example of the Airbus limiting bank angle, the aircraft would (in Normal Control Law) limit the bank angle to 67 degrees which is the angle at which in a steady turn the G loading is at the maximum structural limit. Why would even a Boeing pilot want more than 67 degrees?

[@aeomaster32]

@@SuperHeatherMorris G load doesn't depend on bank angle alone. You can have zero G at any bank angle.

[@SuperHeatherMorris]

@@aeomaster32 Of course you can. However, the Airbus control system was designed for an airliner, not to do aerobatics. For that reason absolute bank angle was limited to 67 degrees as any greater angle of bank in a level turn would exceed the aircraft's G limit. In other words there would never be a need for a greater angle. The normal maximum angle of bank was is much less than this (I can't remember the exact number, but about 30 degrees). There is also a maximum pitch angle of 25 degrees up and 15 degrees down as to go outside those would very quickly get you to some other limit. To get a greater angle of bank than 30 degrees, you have to hold the controls into the bank, as soon as you relax pressure on the side stick the aircraft rolls back to 30 degrees.

[@xmlthegreat]

The problem is that the software is usually unaware of the plane's current intended action, or the phase of flight. All routines seem to be on, all the time. Context awareness is lacking, and I'm surprised that software testing did not find this issue. Usually software in safety critical scenarios like this is tested to hell and back, but I guess in 2001 that may not have yet been the norm.

[@aeomaster32]

No one can anticipate every scenario reality can throw at you. That is the flaw in putting automation in control. Automation should be advisory, not in charge.

[@wazza7575]

Just on thing.. planes don't 'stall out'. They 'stall' or aerodynamically stall.

[@johnpritchard5410]

The video is relevant to the story, which is not always the case here.

[@casparcoaster1936]

Jesus H. I went postal (in my mind) when I couldn't find disable button for my Lexus key fob,; kept locking me out! If no cockpit overide for Boeing by wire, I'd go after the CEO (in my mind).

[@tissuepaper9962]

I lose my shit when my car yells at me for opening the door with the engine on, I would love to be able to get it to STFU. I know about CO, car! I'm not stupid enough to leave the engine running in a confined space! Just stop pestering me!

[@terp2726]

The chart was more informative than the instruction on math.