TOO SASSY!!

are u saying the airbus 320 is sturdier than the boeing 737? i don’t buy it 😮

It will eventually be found that the Fly By Wire system downed the 3 aircraft on 9/11 as well. Flight MH370 and China Flight 5735 are Both Caused by Fly By Wire failures.

As a private pilot student, we learn on day 1 about positive transfer of controls. I say my controls he says your controls and I repeat my controls. But apparently these pilots each wanted to assert their ability to save the plane by both taking control of a fly by wire system at the same time. I really dislike airbus' fly by wire / stick system. Just put a yoke, with a stick it's more difficult imo to position the plane properly. It's not a game

no ty

That was in strong contention for the title of this video haha

@@GreenDotAviation I see what you did there. Nice touch.

"This is highly irregular"

HAL, pretty please, may I....

That's krass; but, perfectly on spot accurate .

So... As I said in my comment earlier, this wasn't autopilot-related. What we're talking about here is the EFCS (Electronic Flight Control System), which is a completely different kettle of fish. In the simplest terms, EFCS (or fly-by-wire) describes the computer-driven electronic control systems of the aircraft, whereas FMS/FMC/autopilot describes the computer systems which determine and automate the flightpath of the aircraft. The most important thing to understand is that - contrary to significant misunderstandings as to how the EFCS systems work - the systems are designed to comply with anything the flight crew request to the best of their ability, whilst at the same time checking those inputs against the safe flight envelope. It's inaccurate to state that the EFCS ignores pilot inputs, rather it tries to comply within the limits it knows to be safe. These limits are not static values, they're constantly computed on the basis of the data coming in from the aircraft's sensors. What happened in this case is that the EFCS software was designed to protect against a change in AoA which would cause the aircraft to stall. Furthermore, the system was designed such that dual inputs from both sidesticks would be "summed". This is why the breakdown in CRM over which of the flight crew was supposed to be in control was an issue. The reason the EFCS initially failed to command nose-up was because the summed inputs would have almost instantly risked a stall condition. This was then exacerbated by the tailwind further compromising the "safe AoA" calculation, which caused the nose-down input. In engineering terms, this is what we would consider an extreme "edge case" - namely an existing issue caused by intentional operation outside of what would be expected made worse by an external factor. It's understandable that if one considers the "safe" margins to be fixed, what you're suggesting would work for this one particular scenario - however the margins are very much dynamic. This is how Airbus took the scenario as presented in this case and applied it to their existing design, hopefully preventing a recurrence.

@@turricanedtc3764 this answer was long and detailed, so thank you. But it's not quite satisfactory in terms of answering the original question. Whether the critical AoA is fixed or dynamically calculated is irrelevant. At the end of the day, the EFCS had that information, since it had determined that the pilot's inputs would have have put the plane beyond that. Once it calculated that critical angle, why not just set the pilots input to that max allowed value, alpha max I believe. The question is, if the supplied inputs from the pilots puts the plane beyond the safe flight envelope calculated by the ECFS, why not set it to the limit, or near it? Why simply "ignore" it all together? And I know you were talking about the semantics of that word, but it's rather apt here as the plane literally did not do what the pilots commanded. In fact, it did the opposite, it pitched down. Ergo, it ignored their commands.

@@turricanedtc3764 Yeah, I meant the fly-by-wire system, not the autopilot. (It doesn't help that the title of the video puts the word "autopilot" in everyone's head...) In any case, thanks for the detailed explanation. I'm still not sure why the summing of both inputs would be an important factor, as that implies that the fly-by-wire system would have deemed a _single_ nose-up input to still be safe under these conditions. Given the nose-down that happened immediately after, it sounds like the safe limit for nose-up was already close to zero anyway.

Because... engineers didn't foresee such a case to happen, in other words it was a software design flaw.

@@hakanevin8545 That much is obvious :P. My point was that the system was designed to deal with unsafe input, which raises the question why they chose to handle it by disregarding the summed input altogether instead of limiting it to the maximum safe value.

I was thinking the same thing lol.

I can’t let you do that, Dave.

It was the fly-by-wire syst.

"Never buy software updates from the gas station"

Truly. The Best!

No it's not

I'm binge watching for three days now 😂

This was the one game I could zone out on and play alone as a kid. Learning how to fly commercial jets with my joystick and mapping out flights through Las Vegas or NYC was magical and worth my first upgraded graphics card. One of these days I’m gonna do a computer build that can handle this version. Looks a bit more interesting than MSFS 98 ha.

@@jeffrey.a.hanson it’s MSFS 2020, a completely new version. I would totally recommend getting into it all! Hit me up if you need any help as moving to this sim was great for me

@@K1OIK yeah what’s wrong?

@@K1OIK you removed your reply to me, which you said “try to make it sound like you made it out of 4th grade”. First off, the country which I live in doesn’t use the American grade system, second off I never made the original comment. please be more respectful. This is KZfaq not English class so you “gotta” cope.

@@alfi_csgo6011 Wow, really?! This isn’t from 1998?! Man, had me. I never would have guessed this was a new game and not the one from 1998! Next you’ll tell me I don’t need my Commodore 64 to play Sid Meier’s Pirates.

I thought it was called"normal law".

@@bobgillis1137 You're right, It's normal law. I made a typo, thanks for letting me know!

@@huskkyy No worries; fascinating topic (and very alarming!). Good on you for following your dreams. I could never be an airliner pilot.

This is what I was taught also (and have seen in flight sims.) Additionally, the 'takeover' button he refers to is simply the autopilot disconnect. Press once to disconnect AP and twice (or subsequent) to silence the AP disconnect warning. It doesn't matter which side presses it, both sticks are active.

@@noway9880 That’s not entirely true, yes it’s an A/P disconnect button but it is also a Priority Takeover button. When the AP is engaged you can press it to of course disengage the A/P, as there are locks to prevent a light bump from disconnecting the autopilot. However when Autopilot is not engaged, it is a priority takeover button, you can hold it, and for the duration that it’s pressed the opposite side stick will become inactive, if you hold it down for 40 seconds or longer it’ll become latched, meaning the other sidestick becomes inactive in perpetuity, until either pilots press the priority takeover button again.

Yes, but the problems are not so different if it’s a pilot at the controls, interpreting the very same sensors. And certainly humans have some “bugs” of their own, too.

The software is technically free of bugs. Even in this incident, it wasn't any bug that happened, the software did exactly what it was programmed to do. The critical software on planes is mathematically verified for correctness, giving 100% confidence that it doesn't contain any bugs.

​@ronicelyvsaid binhoodwinker8621

elon musk saw the “Fly by wire is a great idea” part and ignored the rest when making the cybertruck

And while there is no computer that could misinterpret inputs or behave different from the pilots expect, there is also nothing that helps them make safe inputs.

spanish airports seem to be horrible

@@AlAl-en9ce Actually they are not. Most are fine and according to international standards.

@@FranciscoCaminoDon’t put national pride above safety.

@@The_ZeroLine I am not.

​@@AlAl-en9ce Ciudad Real International Airport is excellent and only cost €1.1 billion,shame it was never used. 😂

Thanks I’ll look into this one!

Glad you enjoyed it, more on the way ✈️

Fr.. As much as I love the PMDG 737, this one has a much more detailed interior..

@@cancelanime1507 So far the PMDG DC-6 is still my favorite. But the Fenix A320 is right up there and, yeah, beats the 737-700 hands down IMO.

7:07 Airbus: Me money!

Thank you for the kind words and for the video suggestion, I’ll have a look into this one :)

Glad you’re enjoying it! Much more to come 😎

It was too good of an opportunity to pass up! Glad you enjoyed it and thanks for signing up on Patreon :)

@@GreenDotAviation Id love to get a close look in the avionics bay where the AI overlords actually live!

Thanks! I'm glad you appreciate the details :)

Yep, this isn’t the first time the independent sidesticks have caused issues.

@@GreenDotAviation I feel if the sidesticks were mechanically interconnected (to provide a level of fedback) but operated the FBW system it would be so much better. I think airbus missed a trick.

@@allancopland1768 - I've read many folks expressing similar opinions over the years, however a thorough dive into the historical aspects indicate that interconnected controls have caused as many problems as non-interconnected - a famous one being EgyptAir 990, where one pilot's nose down input combined with the other's nose-up inputs caused a split-elevator condition exacerbating the loss of control. In the case of AF447, the override button *was* pressed - unfortunately, because the failure in CRM meant that the pilot in the right-hand seat wasn't being monitored, he simply pushed his override button shortly afterwards. No amount of technology can reliably overcome an uncoordinated crew.

Thanks!

Thanks! Glad you’re enjoying them, more on the way 😎

Glad you like it! :)

Some fly by wire aircraft do have “active” side sticks which move based on inputs from the other pilot (or even the autopilot), I can only guess they decided the cheaper option was “good enough”

Thank you! 🙏🏼🙏🏼

Blaming a person is much more comforting than shouting at a computer . That's what I chalk it up to. That way, we can all feel like, "Hm, I wouldn't have made the mistakes that pilot made." That's human nature.

Cathartic is the word I was looking for. A computer system causing a crash is like... being cheated out of life. It's not supposed to happen. We like human accountability and the ability that we have to adapt in ways that computers can't (yet).

Yeah it's something that's taught during the very first flight with a CFI. There's a pilot flying (PF) and pilot monitoring (PM) and airlines have very detailed SOPs regarding this. It doesn't excuse the poor software algorithm but the dual inputs shouldn't have happened. Especially if there was a line trainer on jump seat.

737max was an IQ issue. Mentor would not support this statement, but he clearly demonstrated what needed to be followed to avert the 737 max issue.

@@Menstral First, i believe that flying a plane is not primarily an iq-centered thing. In my opinion, training, experience, coordination and professionalism is what makes a pilot good. Second, the aoa sensor on 737 max was indeed broken and triggered a severe misbehavior of the planes automation/mcas system. Sure, you could solve those situations it if you quickly found the solution under high pressure. But clearly its not that simple given that two professional crews did not manage to do so. Also Boeing got significant criticism for the rushed and flawed implementation of the system so I still believe my argument has merit. By contrast, in this incident/video the pilots made double inputs which to me sounds like a very basic cockpit coordination mistake. To be clear, i also agree that the automation did not really act intuitively and clearly airbus agreed and adressed the automation behavior. Anyway, in the end i was only trying to feedback on the thumbnail. But into the rabbit hole I go 😁

I think the question is: why the inputs from both sticks can be additive? The video doesn't seem to explain this at all.

Glad you enjoyed it! I run it on pretty high settings, and for the ‘showcase’ views I turn the camera exposure up so that more of the scene is visible in darkness. Might that be what you noticed, or is there something else that stuck out?

@@GreenDotAviation mostly the lighting so the exposure makes sense, much appreciated. The way you explain the events and go through the errors is second to none, by far the best Air Crash re-enactments I’ve seen… and I watch a lot of these ;)

8