The Oil Change That Crashed A Passenger Jet | The Crash Of Smartlynx 9001

A320 Image: www.flickr.com...
Video Made With FlightControlReplay (Not Sponsored): secure.simmark...
This is the story of smartlynx 9001. On the 28th of february 2018, a smartlynx A320 was over the skies of tallin, estonia. It wasnt flying anywhere it was just a flight to train pilots for the airline.
On that day the plane had a safety pilot, 4 training pilots and a safety inspector onboard and before the flight they had a safety briefing, today each training pilot would get to fly several touch and gos and go-around before doing a full stop landing and switching seats, for a touch and go you approach the runway as you would for a normal landing and then right after touch down you apply full power and take the plane back into the air, it's great practice for pilots as they’d get some great manual flight experience.
The first flight was uneventful, they did the touch and go as normal. But after finishing the takeoff phase of the first touch and go a warning popped up on the screen. ELAC1 + ELAC2 PITCH FAULTs. The ELAC or the elevator aileron computer had an issue, the ELACs as the name suggests controls the elevators and the ailerons, the ELACs take in the sidestick inputs and then move the control surfaces as the pilot commanded. Think of it as a middle man in the chain. The pilot who was monitoring the flight didnt think much of it, read the instructions on the screen to turn the ELAC off and on again and so they did what we all do, they reset the computer to clear the error. Weve all done it, turing it off and then back on fixes a lot of problems and it cleared all caution messages so they continued with their touch and go training
The first student pilot landed his plane and brought the plane to a stop and now the second student pilot assumed control of the plane. As the student number 2 took the plane to the skies the ELAC1 pitch fault error popped up again, and like last time the system was reset as per standard operating procedure at that time. The same happened during the next touch and go. On the fifth touch and go an ELAC2 pitch fault error popped up. This happened again for the 4th students second touch and go, these pitch fault errors kept coming and the resets cleared them but they obviously kept coming back.
It was time for The third touch and go the plane touched down and it was rolling down the runway. The plane then accelerated to 130 knots to lift back into the skies. The instructor asked the student to rotate, he said “rotate rotate” but the plane wouldn't budge, it was stuck to the ground firmly. The pilot replied that was trying his best to raise the nose of the plane into the air. He had his sidestick all the way back demanding a full up elevator to lift the nose up but there was no pitch response at all.
At this point a new warning was shown on the primary flight display a red “manual pitch trim only” the pilots would have to manage the pitch trim on their own. When you trim an airplane the airplane stays in a particular state without any inputs, so let's say you trim a plane up, then the plane will nose up on its own and then stabilize at a certain airspeed. Normally the fly by wire of the A320 does this fully automatically but now the pilots had to manage that on its own and the red “MAN PITCH ONLY” instructed them to do so.

[@Jet-Pack]

13:15 Thanks, I'm glad I could help! Awesome episode!

[@n7__]

5th🙀🙀

[@arctictiger8690]

@@e3xgamer900 we have evidence of time travel!!!!!!

[@Jet-Pack]

@@arctictiger8690 Oh no my secret has been exposed

[@arctictiger8690]

@@e3xgamer900 Look! He/she admitted to it! WE HAVE PROOF! PROOF I TELL YOU!!!!

[@cjswa6473]

Way too much automation...

[@parrotraiser6541]

A first warning might be an instrument glitch. The second time it shows up, the machine's trying to tell you something. Listen to it.

[@deaf2819]

Lol I have so many Acft I’d love to introduce you to

[@evarwilliams]

Completely agree. This was too much dependence on automation.

[@laser31415]

I see this in every industry, everyone puts up with "turn it off and back on" to fix stuff. It's amazingly scary how fragile our technology is. Just good enough to work most of the time. I'm guilty of it too, "1st time is a fluke, 2nd time is a pattern" is my take on the old phrase.

[@portanav]

@@laser31415 When it comes to flight controls you don't mess with it, it's so important that it is part of an emergency brief for an abort below 80 knots. Especially as this was a training flight having an instructor and inspector on board, they should have returned to the ramp after the second assumed glitch. Very poor demonstration of airmanship from the trainers.

[@millomweb]

"A first warning might be an instrument glitch." Right. So in this case, land and find out.

[@MikeBrown-ex9nh]

Here's an idea, when you keep getting multiple warnings park the damn plane for inspection. Their training exercise wasn't worth risking their lives.

[@killman369547]

Yep. After the second reset the best thing to do would've been to get the plane on the ground asap and report the problems to engineering.

[@jimgraham6722]

Not the first time probably not last. Irresponsible to ignore repeated warnings.. Land and fix.

[@ihateemael]

not half obvious is it!

[@martynh5410]

I was thinking the same thing!! Seems crazy to keep flying and doing touch and goes with multiple faults and resets.

[@podgee7507]

what was the safety pilot doing? he should know better, and keep the plane flying.

[@JA-ux7dd]

Imagine, a high tech aircraft keeps telling you that there is a serious problem, you do several landings and you still ignore the warnings.

[@protonneutron9046]

Darwin Award Candidates

[@K7DFA]

@Proton Neutron: The problem with the "Darwin Award" is that you have to be a non-contribuor to the "gene pool". Lots of pilots have contributed (some more than once 😢☹️), to the gene pool, and are thus "ineligible" for the "Darwin Award"☹️😢

[@tomstravels520]

Not true. It was actually the pilot not resetting the he computer that caused the problem. If he’d kept resetting every time this wouldn’t have happened

[@protonneutron9046]

@@tomstravels520 Really? Please site the accident report saying that. Or, admit you are making sh!t up.

[@tomstravels520]

@@protonneutron9046 when the pilot reset the computer it restored both roll and pitch channel of the ELAC fo full working order as there was no problems with controlling the stabiliser. Then when he landed and held on to the THS, because of the issue with the OVM and PTA’s this caused the pitch channel of the ELAC in command to stop working and transfer to the other. The pilot then reset the computer and then both ELACS are then fully working again. On the last one the pilot didn’t reset the partially failed ELAC so when they landed and the ELAC detected a “fault” it couldn’t transfer to the other ELAC as it had been doing all this time so then went to the SEC’s but they then suffered from the design flaw www.ojk.ee/et/system/files/fail/manus/ee0180_es_san_investigation_report.pdf

[@dobiedude7479]

My favorite story is about one time that we were having a problem with a helicopter. We had to call Germany to talk to an engineer. He insisted that we didn’t have a problem. “It was not designed to fail in that mode”. Well it did. We figured it out on our own. We sent a detailed report to the US based technical representative for the company.

[@atzuras]

Germans do not do mistakes. Don't try to argue with them. But humbly asking for help, then they may listen to.

[@ted.angell7609]

Something tells me the engineer was French, not German.

[@pascalcoole2725]

@@atzuras I do NOT agree on this... my experience, Germans always where verry kind and humble when i came in the picture and they always said to their coleagues, don't argue with him or he will hit you with the right page on the flight procedure manuals as wel as with the maintenance manual. Still the helicopter story sounds funny and i'm sure it is genuine ;)

[@justarandomtechpriest1578]

@@atzuras it's not designed to fucking fail

[@mata2723]

it is like debugging a program. When you have eliminated all the other possibilities, the impossible may be possible (I had it once where True was False because I was importing files where someone had the breautiful idea of redefining False.... But with plane software, I would expect the conception to try to account for these unexpected failures and reconverge when possible to a good state or at least give human the right info and control....not so easy , we have seen so many wrong cases

[@bigdogbandal]

Automation is great, to a point. When it is made to lock an operator out of the controls upon failure, it goes too far.

[@rusticbox9908]

I don't think the pilots were ever 'locked out' of the controls, they still had full control of the trim wheels and engine throttles to fly the plane as third redundancy in an anticipated two computer failure like this video.

[@oakld]

That is totally misinterpreted. The balance of automation here was very good, all they needed to do is to react to the failures and land. And when the system gave up, they should have follow ECAS message MANUAL TRIM ONLY, so there was a lack of crew action, not the automation problem. However, don't get me wrong, I don't totally blame the crew, it wasn't that easy to figure the situation out. Still maintenance problem in the first place.

[@pascalcoole2725]

@@oakld I do agree about the procedure you proposed. Not doing so makes it a repeated pilot failure ! A dunky doesn't hit the same stone twice..... It's a blamable accident !

[@Rocco-tb9ih]

But it wasn't made to lock the operator out on failure, that just happened to be the unintended consequence

[@justingrey6008]

Automation is an amazing aid when done properly. Automation oversight is amazing when done properly. And here lies the issue. Large complicated machines need the automation and should have it, but they also need good over sight systems. And when those fail only highly trained and competent can recover them without any warning. So yes, more advanced automation is great and we need more of it but training in that automation is just as critical. (I work in a factory with automated equipment, sometimes something simple and easy brings everything to a stop because of lack of training and nothing else)

[@rexbentley8332]

Rube Goldberg machines.

[@bestboy138]

one time when i got my oil changed Jiffy Lube didn’t tighten the filter and oil got everywhere but nobody died, yet.

[@howardjohnson6584]

There was another oil change incident where maintenance left three o-rings off the drain plugs on all three engines. Yep. All three engines failed but fortunately they got one restated and were able to land. So, if the computers don't crash the plane, and the pilots don't crash the plane, and mech failure doesn't crash the plane, maintenance will crash the plane.

[@MiniAirCrashInvestigation]

You got anymore details on that for me Howard? Maybe a flight number? It sounds interesting

[@howardjohnson6584]

@@MiniAirCrashInvestigation Well, I've put three links up for you but they don't show up. Enter this phrase into youtubes search bar. eastern airlines flight 855

[@chiefdenis]

@@MiniAirCrashInvestigation i believe you may have covered that incident, i think he's referring to the lockheed l1011 incident

[@algrayson8965]

@@howardjohnson6584 Then the guys who designed the seals say, “No go! Too cold!” but the managers say, “No more delays. Lift Off!”

[@TheTreegodfather]

I'd still rather be the one with ultimate authority over a computer.

[@tomstravels520]

This has nothing to do with having authority. It was basically computers not able to carry out the commands

[@TheTreegodfather]

@@tomstravels520 Did you not hear the question at the end of the video?

[@EstorilEm]

I’d need to do some research, but I’d love to know what flight mode the aircraft was in during this? Was it possible to switch into direct law and regain control over the elevator? It’s my understanding that this is always the case.

[@jjaus]

Airbus a/c do have this function.

[@kerucutgaming2216]

@@EstorilEm problem with modern airliner, yes Boeing included, is that they do away with the cables. So the sidestick/yoke have no direct connection to the flight controls. This done mostly to reduce weight and conserve fuel. The difference is that Boeing try a lot to make the plane feel like conventional aircraft.

[@LeonelEBD]

It seems like doing the sequence on the simulator to somehow do as this flight did was an enormous job.

[@yabattler]

My man, I've been watching your episodes since very early on, and I'm stoked that they are getting the props (no pun intended) they deserve. Do you release these as podcasts (or is that something you would consider)? I'd love to be able to listen to these on my drive to work. Stunning work mate!

[@kyuuteevee]

As an aviation geek, and not knowing until 3 years later that this happened on my 14th birthday, I have no words.

[@RidingSigning]

I know nothing about aviation, I am a natural scientist but your videos are so educational and I enjoy watching them. Keep up the good work. Thanks

[@peterspindley5965]

I want 0 and 10. A plane that can fly itself, but also can be flown manually when situations dictate. Excellent video.

[@manojbala6870]

Irrespective of the fault with multiple resets, the trainer showed great presence of mind. Remember this case from avherald years back. Not many will understand. He saved the fuselage and lives. Lovely channel Edit: automation is a wonderful tool, but only as long as it's monitored

[@escaperoomleander1948]

Computer One: Yay, we're flying! Computer Two: No, we're on the ground. One: Flying! Two: Ground! Both: I quit. Let the pilots figure it out.

[@sabatiniontech7256]

Perhaps the most famous oil incident in a jet was in 1983 when an Eastern Airlines jet was being serviced as part of a class. The instructor failed to replace the "O" ring around the oil sensor as he serviced each of the engines. The plane was returned to service and traveled 100 miles before enough oil escaped to cause the first engine to fail. Fortunately the flight crew managed to turn the plane around and the last engine managed to hang on despite no oil pressure until they were on the ground.

[@ellisandrews440]

Excellent episode guys ! A combination of a lot of automation and pilots having an complete understanding of the automation of the aircraft type they are flying. Unfortunately in the Max 8 case the new automation wasn’t conveyed to the pilots. Re this episode the wrong oil used that set up a scenario. So important that mechanics double & triple check what parts and lubricants are to be installed. Thanks heavens that they all survived.

[@truberthefighter9256]

I like your videos! They're not too complex for my lazy brain; but they also tell the basic things to know🙂

[@pascalcoole2725]

I realy like your last question, here my answer: There is a golden rule in aviation: "Whenever a system fails and the checklist, QRH, Manual, whatever states RESET-IT you do this only once !" Therefore i considder this as a pilot failure (as well as a bad example for the students). As far as automation concerns: I don't mind automating stuff as long as the crew is fully aware of this automation. Which by experience not always is the case. On the airbus automation was mainly done to make it more efficient and being so more profitable (almost the same as the 737-MAX). Personaly i don't really mind as long as you realise that this results in more complicated systems which might result in a system that is not fully understand by the crew (This is probarly the only time in my live that i said this such a polite and politicaly correct way). A slightly different approach which i prefer but is less efficient is a system that assists the pilot for normal operations or gently fights the pilot when doing something which is really unadvisable. An example could be reducing trottles when IAS goes beyond Vne for the given configuration... In this case the pilot knows "I'm doing something that i really shouldn't but this is an emergency, so I go for it" and manualy advance the trottles again. Full automation is nice if everything goes well. If everything would go well all the time we would not need a pilot at all. So i would prefer a system that would be much closer to 'flying by the seat of my pants' than 'How fast are you on the FMC'. This is verry well described by Warren vandenBurg in "The children of the magenta line" (here on youtube)

[@ericlozen9631]

Thank you for the feedback.

[@tomstravels520]

Regarding your first paragraph this was not clearly stated for training flight though at the time in the manuals. If this was an A-B flight the caution would either never have appeared or would appear once as the pilots only ever set manual trim on the ground unless in DCT Law

[@pascalcoole2725]

@@tomstravels520 From a training point of few i do understand your response. We could have a discussion about this. For instance that failure like this should be trained in a simulator. If it happens in real live and you don't know for sure what the problem is, assume trouble.

[@tomstravels520]

@@pascalcoole2725 well the manuals were changed to give more clearer direction in training circuit flights and how often you can reset the computers

[@pascalcoole2725]

@@tomstravels520 Ahh ok.. Of course i was not aware of that. Beside aviation for me is 20 years ago. Still miss it.

[@tannerwolf1267]

Again, great content! Thank you! I really appreciate what you bring to KZbin.

[@caladanian]

Precise, logical and good wrap up - including a nearly philosophical ending. :D I think every grade of automation has it’s own challenges. In the end the combined quality of construction, maintenance, education, tests, experience and learning from faults sums up to higher security. Reality is always complex.

[@susanbrettdavis8839]

Most awesome video ever and top notch ex-plane-ation! I think there is a happy medium with automation and pilot control. Pilots have to be computer nerds as well and never stop monitoring what is going on. I recently was on a flight where the pilot announced he was letting the plane land itself. 😬 was my reaction, but that landing was the most gentle I have ever experienced😁

[@FJVP74]

Hi. First of all, I would like to thank you for all your videos which I follow. I guess it is not easy to get all the information you use and put it together, so congratulations for your work. In response to your question, I believe that automation is very important to relieve a pilot workload, however I firmly think that pilots should fly the airplane more, specially in take off and landings. I follow many channels related to the subject and once the computers give up or get to many error messages, pilots often get overwhelmed and forget to really fly the airplane. Maybe training should focus more in flying visually and using analogic instruments and fabricants include an easy fast override switch to allow pilots to take control and diagnose problems after insuring a stable flight. So I guess maybe an 8 🙂

[@jeffrey8885]

6:51 there are 2 hydraulic motors (Green and Yellow system) & 3 electric trim motors in the THS Actuator.

[@michaelhawthorne8696]

Great detail in this vid about the multi failure of the computers and why, nice bit of research, thank you ! I suspect the pilots should have landed on the first computer malfunction warning.

[@fredorman2429]

My opinion of manual vs automation: a manufacturer built a fully automated plane, but wanted to retain the emergency manual option. The system was reviewed by a committee and the result was: fully automated control plus a pilot and an attack dog in the cockpit. If the automation failed the pilot could take over. The dog was there to bite the pilot if he touched the controls.

[@aaltvandenham]

I love automation, as long as it says what it does and why (or why not). My car is an Opel Omega build 2002 (Cadillac Compact in US). Downhill acceleration, sharp curve ahead, kept going, hit the brakes with force. Forgot I put the cruise-control on 15 minutes before. There's no light "on" and/or "giving gasoline to maintain set speed". This could kill you. Question: how does the Tesla autopilot evade potholes in the road?

[@nancysherburne7445]

I have been enjoying videos from TheFlightChannel, Mentour Pilot, Green Dot Aviation, and Mini Air Crash Investigations for only a few months. How sad when crashes kill everyone on board but at least more improvements are made to lessen the chances of each happening again. It seems there are two hosts for MACI and I like the one who hosts this video better. That said, I must comment on something that has nothing to do with aviation, namely grammar. However helpful it is to know what ELAC, SEC, THS, THS-OVM, and PTA mean, most are not acronyms. All acronyms, such as LASER and SCUBA, must be able to be pronounced as words. If the letters cannot be pronounced as words, such as THS and PTA, they are initialisms. I will look forward to more videos of this nature as it helps me understand what a stressful job piloting an aircraft can be.

[@markmark5269]

Pilot Training 101: Ignoring those stupid computer warnings.

[@billthomas635]

Lucky??? These guys continued to fly an unserviceable airplane.

[@howardjohnson6584]

Exactly! Idiots!

[@Thesa4mppa]

There's a video showing the landing. Saw it on Facebook or some site like that. Pretty rough looking...

[@jrb_sland5066]

The key to my opinion about automation is in your phrase "...a problem that no-one could have anticipated..." i.e. no-one spent enough time thinking through all the possible [not just probable] failure modes. My opinion - automation features should be introduced in stages, separated by long enough intervals [years, in some cases] that all parties involved can be convinced that this part of the automation is stable & safe. Only then should the next level of automation be enabled, with conscious awareness that new "unanticipated" complex failures may manifest themselves. Never forget that very simple math equations can produce chaos, and as complexity goes up, so does the likelihood of trouble. We learn best from failures, so failures should be accorded our respect, not our fear.

[@rtrThanos]

I drive a sports car with a manual transmission, so I’m obviously against automation as I like having full control of what the car is doing. However I think automation is still crucial to the extent that it can save your butt in a bad situation. I was going downhill into what I thought was a sweeping turn, but it turned out to be a decreasing radius turn. By the time I realized I went into the turn too fast it was too late. I hit the brakes so hard that anti-lock didn’t kick in fast enough to prevent my rear end from kicking out. I thought fo sho that I was spinning off the pavement and flipping my car, then I saw lights blinking on the dashboard. The active stability control was intervening and controlling how hard each of my brakes were engaging so that wheels with the most grip got more brake and wheels with least grip got less brake. The car straightened out, I pulled over for a minute until my hands stopped shaking, and I thanked my lucky stars that automation saved my ass. But I will never, ever, own a self-driving car. I’m a computer technician that sees bad programming and failing microchips all the time. I wouldn’t have a job if computers worked perfectly 100% of the time, so I can never agree with giving computers more authority than the humans operating them. Plus my chances of survival against zombies are better than they are against Cyberdyne Systems Model 101 T-800 Terminators.

[@danielschoch9604]

The issue is not about how much automation. The issue is the pilot's awareness who is flying the plane.

[@ernestmachpro3341]

No injuries, no deaths, good plane.

[@fuzzymath6240]

It really seems right now that could be the Airbus vs Boeing tug of war. I used to think that anyways. Really though with fuel cost as just one example, large passenger jets will continue to get more computerized. That doesn't mean less control for the pilot. When there isn't built in redundancy, that I think like you have pointed out also, is really when there is less control for the pilot. And I'm not a pilot so really what I think is irrelevant in the long run I guess but man it is just so damn fascinating. Your video's started out great and have gotten even better I love to see when a new one is ready and just wanted you to know that!!

[@ronniewall1481]

GREAT SHOWS. THESE ARE FUN TO LISTEN TO WHEN DOING HOUSE WORK AND OTHER STUFF.

[@TimwiTerby]

Whether to use automation or not is not a matter of putting it on a scale from 1 to 10. For any single thing that could be manual or automated, the correct choice is to use the one that is safer. Use automation exactly in those places where the machine performs more reliably than humans, and vice versa. Anything else would just be insane.

[@Matticitt]

The ideal and the goal is for automation to fly the plane completely and for a human to be there to just supervise. This should be designed in a way to keep the pilot as aware of what's going on as possible. The biggest problems with automation so far were improperly designed systems which did something as designed, or in error, without pilots noticing causing confusion or lack of proper reaction from them.

[@jamysalmeida18]

As a flight engineer of c130. I vote 5 for automation of the planes. Automation is good for human errors. But aways we had to consider a eletronic error.

[@chrish5791]

It’s just amazing that hundreds of thousands of multi engine flights took place during WW ll without even one computer and this plane has enough computers to fill a college computer lab. The one thing that I know for sure after building and servicing computers for years is that they fail and the degree of the consequence of this failure is directly proportional to the importance of what they’re controlling the moment they fail. It would seem logical to have a one time override removing them from the control circuit if this problem cascades through the redundant computers.

[@Ronin4614]

Great video with superb detail. I believe in automation and support FBW, but; we quickly forget what makes things fail and lose some control as a result. Thank you.

[@nilamelody]

Manually trimming those trim wheels is a very muscle heavy task :d

[@etherraichu]

I propose we call horizontal stabilizers "Hor-Stables" Because apparently I have the sense of humor of a child and find that funny.

[@kwikbit]

Lots of good comments below - interesting that most people are talking about the pilot repeatedly ignoring/resetting the alarm. Automation is brilliant - but this incident was really, fundamentally, caused by the "wrong" oil surely ? As with all automation , it's only as good/reliable as the ongoing maintenance . Personally, I think that all designers should consider their software/hardware to be as reliable as a windows based PC. Maybe all planes should have a massive big red "I HAVE CONTROL" button on top of the instrument panel. This puts the plane's computer(s) in a Windows-equivalent Safe mode. 🧐

[@Hedgy327]

Automation is rarely a problem unless, as in this case, pilots choose to ignore repeated warnings that there was a problem. It's not unusual for one of those annoying little red lights to come on ... so switch it off and switch it on again (as it were) and usually all is well. When it happens again and again it is likely to be more significant. Why did the pilots keep resetting the warning??

[@ACR909]

What a legendary crew.

[@henrybowden9907]

I think a lot of the problems come when the automated systems fail to inform the pilots clearly what is wrong. It's all very well having this stuff with all its bells and whistles but the pilots MUST be kept informed properly by the equipment. The so-called glass cockpit might have been able to solve this problem but the aircraft manufacturers have failed dismally in this respect.

[@epapa737]

5 at most I do appreciate automation removing a big burden on the pilots but I love how computers disconnect when they know somethings up

[@calyodelphi124]

I'm ambivalent about fully automated aircraft, but in the same breath automation significantly reduces the pilot workload and allows the pilots to focus more on all three of the primary tasks of their job. The more automation you introduce into an aircraft, the less hard work the pilots have to actually do in the course of a flight. But this necessitates that the pilots must compensate by doing even more hard work on the ground, training for the very, very real probability that that automation is GOING to fail or throw its hands up and Just Quit on them, and then it's entirely upon the pilots' shoulders to take over for what the computers had been doing for them. That requires the airlines to have to more thoroughly train their pilots on heavily automated aircraft, and if the airline falls short of all that training, well... you get pilots who're just as clueless to what's happening to the aircraft as the computers that just threw the problem onto the pilots' laps.

[@dextermorgan1]

When an airplane gives you multiple warnings, listen to it...

[@jeremydennis6988]

I think what you do is great thanks very much...

[@praetorprime]

You're asking the wrong question. 1-10 automation is irrelevant if the QA applied to that automation is inadequate. Testing is ultimately the hardest challenge facing automation integration into any mechanical system where lives are on the line. This is entirely the fault of the QA team responsible for designing the test plans for these redundant systems, especially given the fact that they knew they were working with asynchronous state/data streams. Competent QA would have known to run this test, or to build a system that throws random component failures/out-of-spec responses into their scenarios. Regardless, my answer is N/A, none of the above. The pilot should, at any point, be able to fully assume surface control of their vehicle. For anything less complicated than a supersonic stealth bomber, at least.

[@Baldorcete]

Automation 9.5, being the 0.5 remaining pilot taking control of the entire aircraft, or selected systems when he needs to.

[@scottwright8354]

Over the years I've seen and read many reports of problems with the automation of Airbus aircraft. A few of those, like a Quantus flight that injured numerous passengers and crew, was never fully explained. As many have commented, a lot of pilots much prefer having a system providing direct mechanical connection to flight controls when all else fails. I'm not an aeronautical engineer, but I'd assume that Airbus's philosophy is that using one hundred percent fly-by-wire saves the weight of extra hydraulic and mechanical push-pull tubes. Yet, I'm one of those professional pilots that would have some reluctance flying such an airplane every day.

[@toshaheritalvinen6426]

4 for automation. The human touch is a beautiful thing

[@YouTuber-ep5xx]

You spelled is wrong. 280 lives lost.

[@garystafford3354]

5 on automaton. Plus I know in USA any flight control issues/problems is a no go condition or red X in the aircraft forms. They had indications of a problem with the 1st trainee and the kept flying and the 2nd trainee flew his time. That 1st time they should have aborted the flight training after the first problem that arose. They are very lucky they walked away.

[@EternityForest]

It's amazing that the microswitch piston was so sensitive that the wrong oil would cause that. As a non-pilot, the whole thing seems unnecessarily full of fussy mechanical bits. But I guess it is good that it failed then, rather than slowly wearing out and causing a real tradgedy somehow later. This doesn't really seem like an automation failure though. The software really should not have been vulnerable to race conditions like that, and I hope that kind of thing is fixed now, but it seems like what really failed was probably human error, putting the wrong oil in the first place, and possibly the microswitches not being robust enough. I'm not exactly sure how or why anyone still likes microswitches. They are reliable, but not as reliable as you would expect given how absolutely confident people seem to feel about them. But I guess that is why they had 3.

[@BillyAlabama]

Another excellent presentation.

[@nikolaospeterson2495]

Weel. I like SOMEdegree of automation, however I do believe that manual conrtol should heve immediate proprity at any moment something doesn't seem right.

[@oldstyleanalog6459]

i trust to fly with these students.They saved the day

[@d.mckormic]

In all honesty its less a game of more or less automation but ensuring the automation put in place doesn't manage sabotaging the entire system crashing the plane due to improper data perception. While more complex commercial aircraft automation can be compared to the cruise control of a car, it'll do exactly what the can says, but if you hit the brakes or accelerate you're still in control as the driver/pilot.

[@ianr]

This channel is superb, so why are there 5 dislikes? 🤔

[@Miginyon]

I think automation is a great thing, provided pilots understand not only what the automation will automate, but also they need basic understanding of software engineering. For me I can code so I understand the limitations better and the fact that it isn’t something as reliable as magic

[@bloodlos]

I think we should have automation that assists the pilot but the human should still be 100% in control of the plane. A computer fault should not be able to bring down a plane.

[@kathypratt5399]

I learned to fly on a 1943 Stearman trainer all stick and rudder. No BS just listen and feel the plane,. It would let you know if something was wrong. Ever have your computer crash, too many on the new planes and new cars. Worked as an A & P, never trusted the computers, better to have a good pilot

[@crazimathx4905]

I think automation and manual both should be there....like the pitch one.... pilot can override but if something is going wrong by pilots then automation backup

[@Doggy-B]

Balance of automation and human input should be a mutual 50/50 imo.. never ever remove the human equation, yes it can flawed, but so can tech.

[@ABFox]

I'd go for around a 7. I was lucky enough to take the controls of a 707 type aircraft, under the condition that I could keep it more or less in track and + or - 1,000ft. We were at cruise in uncongested airspace and I had an IP in the right seat so it was as safe as one could expect. While a wonderful experience, it gave me a whole new appreciation for what pilots have to go through on a good day (minus autopilot of course). If automation means that a pilot can free up some brain bites for what's important, I'm all for it, given the condition that they can take back full control if needed. Caveat, I like the idea of airbus not allowing the plane to be put in unsafe configurations, but I'm not to keen on the boeing 737-max type systems, especially with their initial lack of redundancy.

[@BerraLJ]

I would say a 5 for automation, some stuff is likely good but also the more automation the less the pilots fly manually and that is a skill they need, and we seen that tech can go terribly wrong at times.

[@plannine6542]

If pilots need to be able to override the automation, the automation should also have the ability to override the pilot. The warning systems in this case repeatedly identified the failing components- it was the crew who ignored the warnings and did almost everything possible to avoid taking the appropriate action.

[@tomstravels520]

The appropriate action was to reset the computer. This was the official procedure and every time a reset was done it restored the computer. If this was a normal flight they wouldn’t have seen this warning again

[@YanDaOne_QC]

to me, the amount of automation in planes right now is perfect. No more, no less that's where I stand

[@DaMetaEX]

i'm an 8. anything that is a finicky as an aircraft needs enough automation to be able to control the aircraft easily should something happen to the pilots. but also know that something has gone horribly wrong with the aircraft and properly inform the pilots. a repeating error is not one you take lightly. after 3rd or 4th alert the safety pilot should have known the aircraft was in a problematic spot. and that unlucky bounce sealed the aircraft's fate. thankfully it wasn't a serious accident. could you imagine if this happened during a fully fueled and loaded flight.

[@freddyhollingsworth5945]

These $100,000,000 airplanes can't handle sensors and computer logic, but we're supposed to trust $45,000 self driving cars with our lives?

[@geekychannel2543]

I think a good pilot/automation ballence is around 5 or 6 Computers and people both fail regularly and the computer needs to be thought of as an assistant that can take over if something happens to the pilots (like if they fall asleep) but the pilots need to be able to override the computer when something like this happens.

[@josephmassaro]

I think planes should be fully automated, but pilot's should be trained as if they aren't. Automation is only going to become more complex so pilot's need to know how to deal with problems that will inevitably arise with it. What is it Scotty from Star Trek said? "The more they overthink the plumbing, the easier it is to stop up the drain."

[@specialopsdave]

FBW systems should have emergency switches to bypass the computer and instead directly control the angle and position of the control surfaces, but pilots would have to be trained to work without the feel of a cable-pulley system or the computer's assistance

[@mrbmp09]

There should always be a mechanical connection to the flight controls. It's just a matter of time.

[@reggier2343]

I gained a lot of plane knowledge.

[@44R0Ndin]

I'm of the opinion that the more automation is put into the plane, the easier the pilot's job becomes. So on a scale of 1 to 10, with 1 being no automation and 10 being as much automation as possible, I'd go with a 9, meaning I want as much automation as possible but I still want a human pilot able to override the automation if in their judgement they need to. The reason is that with increasing automation, the number of "edge cases" where you get a perfect storm of conditions that send the automation into an undesired state increases. The flight computers must get increasingly complex and the programming must get increasingly hard to design to compensate for this, and the pilots must be kept aware of and trained on all the automation present in the plane. As the 737 Max groundings made us learn, there is no such thing as "the pilots don't need to know about this particular bit of automation because XYZ" where XYZ is any reason you can think of, because there's the counter argument of "what if it's telling the aircraft to do something CLEARLY WRONG" such as commanding the nose down for no good reason, with no training the pilots won't know what to do and you end up with a lot of dead people. I do not think it is possible to automate an aircraft that carries humans to the point that you no longer need a pilot. Not only is it psychologically undesirable to do so for many (most?) humans, there are always going to be situations that you did not program the computer to handle such as specific combinations of systems failures and maintenance failures. While a human pilot may not be specifically trained to handle every combination of systems failure and/or maintenance failure, provided they know what all the automation does on the aircraft it has been shown by several cases that a human pilot will more often than not have enough tools at their disposal to adapt to the situation and use what IS working on the aircraft to land it safely with as little loss of life as possible. I do not think that is something you can program a computer to do without at least doubling the amount of time it takes to design the flight computer software.

[@omaryak]

I would want an 8/10 level of automation. Just enough control when it’s needed, but lots of systems to prevent common errors. I’ve watched enough plane crash videos to see why those systems are needed

[@emergencylowmaneuvering7350]

They did a Turnback to opposite runway. Didnt have engines to circle around to same runway they took off from. Smart desicion. Pakistan airways 5383 last year messed the engines when a no gear landing the captain did and then went on a go around. Took off, but instead of doing the turnback to opposite runway like they did on this video, he decided to go around the airport, with failing engines, Idit captain. Of course, the engines failed like here but on the far downwind leg he dummy tried to do- and they crashed short of airport killing many on airplane and ground. When engines are impacted like that is stupid to take the longest route to land of a Turnaround the airport. Do a TURNBACK to the closest runway instead. Geee!! Some guys lose common sense when get nervous. I taught turnbacks and turnArounds to many pilots with partial power to let them see the difference in those 2 maneuvers...

[@100Lesemaus100]

Automation level 1-10: I think I’d like to have as much automation as possible really. I would always want there to be a crew that can step in at any second and take over control however. So 9 on the automation maybe? For it to be 10, I think there needs to be stricter/more frequent controls and automation still needs to develop more.

[@allenmorseiii295]

I think a balance between automation and pilot control is necessary. So, a 5. Computers run on sensors, some of which have their own delicacies. After a misreading causes a problem, there should be an immediate and simple method of bypassing the problem by the pilot and flying the plane with good ol' stick & rudder. There should NEVER be an automated system pilots do not know about, (MCAS on the Boeing 737 Max) and are not familiar with!

[@JamesW81]

I think automation should be somewhere around level 5. Enough to help, but not enough to "go rogue" so to speak.

[@slaheddinechiboub8247]

Good evening 1/ This is a training flight and not a flight test From the first failure the instructor should take over make a full stop landing and cancel the training flight 2/ A double failure with gear down leads to direct law witch is different from alternate law with the gear up and this is very important for the trim And nobody talks about it Conclusion: The captain in charge of the training failed in his mission Rgda

[@tomstravels520]

The aircraft was in mechanical law for the pitch which is below direct law

[@slaheddinechiboub8247]

THATS EVEN WORSE ANOTHER REASON TO STOP THE TRAINING

[@jjaus]

Automation is a different question. If the damn aircraft gives you failure warnings and you are not hours from land, then land and have it checked out.

[@roo1314]

It isn't really a question of how much automation but what/who ultimately can control the plane. When automation fails it would be much safer if the pilots could just shut all the failed systems down and be able to control the plane manually. So many times when automation fails the pilots are forced to search for and try to resolve the computer problem. This distracts them from what the plane is actually doing. If they could just switch the automation off but still fly the plane manually that would be very, very nice.

[@mikemorgan5015]

Yeah........NO. This isn't rebooting your laptop. WTF were these "pilots" thinking? Remind me to stay the hell away from Smartlynx, if this is the mentality of their air crews. WOW! I'd have been happy to get it on the ground after the first one and get maintenance on it immediately.

[@chrisco7]

Given this was for training, I guess this is good, you need pilots that know how the automation works and what to do when it goes wrong, but provided that is available then more automation is better.

[@DGTelevsionNetwork]

Imagine dousing a microswitch into oil. Was an accelerometer too hard to make or something?

[@rushfan3]

I believe we should have automation set to level 6

[@LamSH1109]

Why did Eng2 catch fire, and the eventually both engines failed?

[@Singularity24601]

Not a pilot, but when it comes to automation in computers and other tech, I personally want 10/10 automation but retaining the option to disable various elements to 0/10 as I see fit. Reality is complex. Edge cases and exceptions are the rule rather than the exception.

[@simu31]

I disagree, the automation didn't fail as such. Humans forced the automation to not be able to do it's job. Humans didn't use the correct oil. Humans didn't listen *repeatedly* to the aircraft when it was giving multiple warnings. Subtract either of those two human errors, there would never have been a problem. Out of interest, as I don't remember you mentioning it, why were the pilots not using the TOGA setting, why were they forcing the trim wheels? That's the point of the TOGA setting

[@markwelch1836]

I engineer plant automtion. We like to say we just need a guard dog to growl at the operator when he touches the controls he shouldn't. :)

[@Powertampa]

Airbus, requiring pilots to have a computer science degree since 1998

[@patrickbuick5459]

It isn't such a simple question to answer. One reason being system integration of other systems added when lessons were learned from previous accidents. However the primary reason is changes in aircraft design. Some of the more modern designs can't fly without computer integration because the design is inherently unstable. However, all systems require sensors, which can be unreliable as shown in many of these videos and that can give false information to both the computer and the pilot, creating a major issue, especially if in IFR conditions.

[@donnafromnyc]

If anyone here is familiar with the phrase, "Children of the Magenta", you will know why you need 2 pilots who are experienced in hand flying the plane up front. Too many countries are training too many pilots in their schools to be bus drivers who rely too heavily on automation and cockpit readouts, but can't control their aircraft safely if systems fail. So my vote on automation? About a 6.