Comments by "Mikko Rantalainen" (@MikkoRantalainen) on "Mentour Pilot" channel.

  1. I was aware of the change blindness / selective attention (the original test with gorilla) so I did notice the tail fin. However, trying to be aware of the surroundings caused me to be off-by-one for the juggling so I failed the main task by trying to be aware of the surroundings. It's easy to imagine that under stress you would need focus on the main task so much that you would fail to notice anything abnormal nearby.
    560
  2. It's interesting that Asiana had company policy to use autopilot at all times possible but still the training didn't result in full understanding of the various autopilot modes. I would have assumed that with such a heavy emphasis on autopilot, fully understanding the autopilot behavior in all modes would be critical.
    410
  3. Because every thing in airplane can cause failure that might result in crash or fire unless you can turn the thing off.
    244
  4. This should be the pinned comment.
    96
  5. Thank you for sharing your thoughts. Could you elaborate a bit on the culture that you think was important background for this accident?
    87
  6. 6:00 Did I understand this correctly? The red stop bar lighting was diagnosed broken in 1998 and it was still not fixed at the time of the accident? I would understand such problems to exists for a couple weeks in worst case. And in case of broken lights, I would expect the airport to be automatically closed in case of fog because of being unable to execute the tasl safely. If the airport were closed every time there was fog, maybe they would have considered fixing those lights a bit more important?
    82
  7. 9:20 This makes me think that pilot training should contain tasks where the input data is intentionally so incorrect that the students should be able to figure it out and ask for clarification. If students learn that all input given by somebody else is always correct, they will never learn to always verify if any input they receive seems to make sense in big picture. Of course, such a sanity check cannot fix small problems but those are probably not dangerous enough to create a safety issue, unlike flying in totally incorrect direction and potentially running out of fuel even if they notice the problem later. And if something this happens in mountain range, then it could turn fatal very fast.
    78
  8. The only reasonable explanation I can see is the "once fuel leak has been confirmed" part above that. If they were only suspecting fuel leak by that point, then you don't need to follow the steps intended for confirmed case. However, when the computer system later confirmed it (even when all the signs already pointed to it) the crew failing to follow the official procedure makes no sense.
    71
  9. It seems that the problem was within Spirit which is even scarier because it means all Boeing aircraft and some other brands, too.
    55
  10. Also very stupid: removing automatic safety measures from the plane before trying this.
    37
  11. If all she got was being blacklisted from future flights, she should consider herself very lucky. That stunt must have caused pretty major costs for the airline.
    37
  12.  @gumby2241  See the link to official Final Report of the accident investigation in the description of the video. As you probably know, commerial planes have CVR so we have full recording of the conversations in the cockpit. Full transcript is in the Appendix 3 of the report. Among other things the report says this: "More generally, the CVR shows that at no time was there any dialogue or briefing related to a possible anomaly during takeoff, as required in the Air Algérie procedures" which quite clearly points the blame to the captain who should be responsible for procedures to be followed. If we didn't have CVR, then we would need to speculate more.
    35
  13. I agree. Also, I think that the worst low altitude situation would be system accidentally deploying reverse thrust on the left immediately after the plane has flied over the end of the runway. In that situation engine power is high and altitude is very low. Even if you couldn't figure out that high altitude situation would be bad, the takeoff configuration should have been included in testing at very minimum. (Or even worse situation would be deploying reverse thrust on all engines after the takeoff but that's definitely not possible to fix and shouldn't happen with true redundancy anyway.)
    35
  14.  @Sergioosh  I would agree with your reasoning in general. However, Boeing has done this enough many times that it appears to be systematic problem with their culture nowadays (and I mention nowadays here because Boeing used to be very safety oriented company maybe 40 years ago but things have gone bad since). It appears that there's great internal pressure for get-there-itis for the design of any new subsystem within Boeing. That doesn't work for single pilot aviation and it doesn't work for big company such as Boeing either. This accident and the whole MAX fiasco would be prime examples of this culture.
    35
  15.  @hbh3144  After the oxygen masks drop in the cabin you can start a clock. If pilots haven't contacted the crew within 10 minutes, you have MAJOR problems. Waiting for hours before trying to contact the cockpit or entering the cockpit (the locked doors were implemented only ater 9/11) has no sense whatsoever. People will start dying or at least getting brain damage within next 5 minutes once your 10 minute timer goes out so you cannot wait any longer.
    34
  16.  @liloruf2838  Lack of oxygen causes brain damage rapidly but hypothermia reduces the use of oxygen and other chemical reactions inside the brain so much that the brain damage is caused much more slowly. Of course, hypothermia may also kill you but that's much slower process again.
    32
  17. That was the captain's way to inform the FO that he's too drunk to fly.
    31
  18.  @Flightsimmovies  99.9 is same as 50% for most people. Either something is absolutely sure either way or it has 50% change of being either one. If you really understand math, you are not part of the common people nowadays.
    28
  19.  @hbh3144  If the pilots really understood that the cabin pressure was incorrect they would either fix the pressurization or taken the plane below 10000 ft. They did neither. Nor did they use their own oxygen masks either. It seems pretty clear that pilots didn't have a clue but crew should be able to figure out if the plane took the descent or not. In the end, nobody can know for sure because there are no recordings. However, at least one crew member did understand that he needs extra oxygen so he did also understand that they're flying high without pressurization. And to figure that out, he must used oxygen for the whole flight because otherwise he couldn't have figured that out and would have passed out once first oxygen source went empty.
    26
  20. I did watch the sequence a couple of extra times and I can see the fin, the pole getting shorter (imagine that being the fuel indicator!) and one of the street signs changing a couple of times. I still cannot see the kitty anywhere – can somebody give a timestamp and color to look for? Update: found it: bottom right corner around 18:05 but it's rendered under the Youtube user interace in case you have the controls visible on the screen. In addition, the graphics is pretty dim so it cannot be seen if you have lots of screen glare.
    25
  21. I find it really surprising that recommendations didn't include training to trust the attitude indicator. The changes of three separate attitude sensors all failing at the same time are so close to zero that it would be sensible to question every other sensor data you have before that – especially when attitude sensors do not require any input from external sources unlike e.g. airspeed or angle of attack indicators. Had the pilots trusted the attitude indicators, the situation would have been under control pretty fast. Also, I would have expected recommendation to automatically always declare pan pan if an engine is lost at high altitude. Update: loved the "it's always better than your vestibular system"!
    23
  22.  @CrowPal  I agree that the plane should not be considered safe to fly with this many faults on a single flight even if engineer reports "no fault found". That should be pretty clear indicator that the engineer simply doesn't undestand the issue and shouldn't be making any claims about the safety of the plane. The fact that previous intermittent problems were "fixed" by simply replacing random parts also suggests that the root fault wasn't found.
    22
  23. @James Thompson By the time they run out of predefined waypoints, every display would have something abnormal on the screen. They weren't even glancing.
    21
  24. I agree. I expected at least some head damage and maybe permanent brain damage due lack of oxygen because it sure is not easy to breath in such a wind. However, cold environment does reduce the damage due lack of oxygen in brain so it might have helped here.
    20
  25. Props to the artist that create the comic panels. Or if it was AI, props to the prompter because getting that even results with current AI is really hard.
    20
  26. You should put your seat belt properly even for the car ride into town (that includes the lower belt not going across your stomach as many overweight people too often use it). Most car accidents happen when the driver is too familiar with the surroundings.
    18
  27. I agree. I think this captain would have needed a hard lesson about this and the way to proceed with that would have been the FO to say "your controls" after they received the takeoff clearance but the plane was still standing still. There's no way the captain could have taken the flight at that point and they would had have to taxi off the runway to discuss about why things got that far without proper preparations for the flight. Unfortunately, this didn't happen and over a hundred people got killed.
    17
  28. I too have previously only heard the explanation that load shifted and caused center of gravity to change. Nobody told that it was actually more probable that cargo crashed the back of the plane breaking stuff. And the parts found on the airway clearly point out that the plane suffered huge hardware failure already before the crash. Great content from you as always!
    16
  29. My thoughts exactly, too. This video made me to think how important it's that the legislation doesn't make rules that cause get-there-itis on the expense of the safety. The actual reasoning behind the change was to make sure that captains do not regularly get too much work without enough rest but if the law is poorly written, it cannot handle special cases like the bombing behind this disaster. Had the law been that if the rest time cannot be followed, the captain is forced to have one day off following the flight, there wouldn't be such a pressure for quick departure.
    16
  30.  @HenryLoenwind  The problem here was that a plug was signed as door. You cannot open a plug, simply detach it.
    13
  31. 24:03 I'm not a pilot but when you're flying in clouds and hear "Bank Angle" warning, shouldn't you look at the attitude indicator and maybe backup attitude indicator to figure out what to do, instead of looking or feeling the yoke? When you consider what could possibly fail in a modern jet, attitude indicator is maybe the most reliable data input you can have.
    12
  32.  @Lovedbychrist1  I agree that the most astounishing part was that nobody noticed extra bolts after putting the door/plug in place. I would be pretty unhappy to find 4 unattached bolts after maintaining a car, let alone an airplane. Did somebody notice those bolts and simply pocket them instead of raising an issue that they found extra bolts but do not know where they belong to?
    11
  33.  @charleshammer2928  I agree that they should have grounded that plane but they were following Boeing instructions. When a repair was attempted that resolved the issue even for a short time, the 500 hour timer would be reset. I totally disagree with the timer reset – if the fault affects the same system time after time, it's clearly faulty and timer reset must not be done. This problem was caused by Boeing on so many levels, including all the effort to dodge the blame after the accident which required Lauda himself to be really persistant with the investigation.
    11
  34. Awesome video, as usual! In the end, I can only wonder how on earth this airport was allowed to operate this way? It clearly had lots of things wrong and still was used for commerial flights. I would understand this kind of problems for hobby use and IFR conditions.
    11
  35.  @Robbedem  Perhaps companies are pushing FAA little to high to lower the bar for new pilots? I understand that companies would like to have more pilots on market to lower the salaries but lowering the bar to entry for the skill required is absolutely the wrong answer.
    11
  36. Yes, and understanding that humans are very prone to confirmation bias. If things start to go out of expected state, re-evaluate the situation to see if you have misunderstood something seriously wrong.
    11
  37.  @hairyairey  If this happened today, pretty much every passanger has a GPS + GLONASS enabled device in their pocket which can automatically report location down to 5–10 meters when they call 112 (that's the 911 for Europeans).
    10
  38. Great video as usual! In addition to the things you mentioned in the video, I think that the tech log should always include constant notice about potentially misbehaving systems in the aircraft. And I think that 500 hour counter reset is not safe thing to do. If the issue cannot be fixed in 500 flight hours, then the plane should be grounded until the fault can be fixed for real. At the minimum the tech log should contain static item saying "Potential reverse thrust deployment in mid-air due unknown fault, maintenance hasn't been able to solve the problem since 17 Sep 1990.". 500 flight hours is roughly equivalent to 25000 km or 15000 miles in a car. If I had a car that had constantly re-appearing fault for so long, I wouldn't simply keep resetting some kind of counter and pretend that maybe it's now okay unless the actual cause has been fixed. The car analog would be ABS system failure or airbag failure randomly appearing every now and then. For sure, if I was lending such a car to somebody, I would tell that it may have some unknown failure of those parts instead of saying that the car is all-ok. And maybe figure out what's the expected end result (can it have total brake failure? is it possible that the airbag explodes on your face while driving?) and how to deal with the situation if that happens even before starting to use it.
    10
  39. Why wasn't LRD designed to instantly shut down the bleed air valve? If the known side-effect of LDR activation was always dumping 13 litres of oil into the engine which will then get into bleed air, surely the engineers creating this system could have figured out that it's not a good idea to breath the resulting mist??
    10
  40. That part was interesting to me, too. I guess either (1) they were so stressed by that time that they accidentally pressed the wrong button or (2) it seemed like a solution to start covering their tracks. The fact that cockpit voice recorder breaker was not pulled open after the landing seems fishy, too.
    10
  41. Landing at speed of 124 nautical miles per hour (about 230 km/h) and trying to brake on slippery road is going to be sketchy no matter how you do it. Even in a passanger car with winter tires, that could require maybe 2000 meters in very slippery conditions.
    10
  42. It's really interesting that any pilot would even thought that it's okay to fly without fuel gauge. It's not unheard to get fuel leak in the air and without a fuel gauge, there's no way to figure that out in time.
    9
  43.  @bookcat123  You should explain at least once why you're repeating the instructions. It's a systematic method to avoid accidents caused by yet unknown future miscommunication. And since you cannot know when the future miscommuncation will happen, you have to make it a habit to always repeat the instructions to verify the communcation has been understood correctly. Computer communication systems use checksums for this but humans cannot use that method effectively so repeating the full message or at least the key parts is the only way to verify that the communcation succeeded. The key point is that communication errors happen even between computers so you should never ever assume that communcation between humans would automatically succeed.
    9
  44. ​@James Thompson It's hard to say for sure because it seems that the pilots were not telling everything but it seems probable that they would have noticed any loud enough sound because they reacted to cabin crew call, too. The question is if cabin crew wouldn't have done anything, would they have kept flying until low fuel sound would have woken them up?
    9
  45.  @DeneF  Maybe he made clothes from the extra parts?
    9
  46. If the auto-mechanic floors the engine to diagnose it and the original failure is failing oil pump, that floor it / hammer it diagnosis method will rapidly convert a cheap fix to a very expensive fix.
    9
  47. 15:50 How on earth tower control or ground control wouldn't be aware of all markings on the taxi roads and runway? Is this assumed to be normal today on other airports?
    9
  48. I think the first officer should have said "your controls" when they got takeoff clearance. They would then had to taxi back to discuss about why captain had zero preparations done for the flight.
    8
  49.  @russcampiond05  What I would like to know is did the 3rd pilot notice that the FO started to press buttons and pull speed brake without saying anything? That would have been valueable information to the captain to better understand the situation. Especially when the captain had already taken control you wouldn't expect FO to press autopilot or touch the speed brakes – why didn't the 3rd pilot say anything?
    8
  50.  @souledgar  737 MAX is a great example why any automation able to control flight surfaces in aircraft should be based on redundant hardware and still be speficially trained for the pilots. Boeing pretended that you can use single point of failure and skip training – only to reduce costs to the airlines and increase profit margins. 346 casualties tell otherwise.
    8