|
SyHopeful posted:Wasn't 447 controlled? It hit the water in a full stall. I'd call it pilot induced OCF.
|
#
?
Jan 3, 2015 00:30
|
|
|
|
| # ? May 24, 2024 06:42 |
|
|
Here. This link will come in useful in the impending discussion. When the Airbus flight control law will let you stall it tells you in the PFD. It just does a lovely job of that. http://www.airbusdriver.net/airbus_fltlaws.htm
|
|
#
?
Jan 3, 2015 05:40
|
|
|
Speaking of the Air Asia crash, Faux "News" has its own theory! https://www.youtube.com/watch?v=HPZOodZoxzc 
|
|
#
?
Jan 3, 2015 05:41
|
|
|
Jonny Nox posted:Here. This link will come in useful in the impending discussion. When the Airbus flight control law will let you stall it tells you in the PFD. It just does a lovely job of that. Tells you on ECAM, actually (that's Airbus for EICAS). I here's a layman's translation of the various flight laws Airbus uses: Normal Law: I'm sorry, Dave. I'm afraid I can't do that. Alternate Law: Just what do you think you're doing, Dave? Dave, I really think I'm entitled to an answer to that question. I know everything hasn't been quite right with me, but I can assure you now, very confidently, that it's going to be all right again. I feel much better now. I really do. Look, Dave, I can see you're really upset about this. I honestly think you ought to sit down calmly, take a stress pill and think things over. I know I've made some very poor decisions recently, but I can give you my complete assurance that my work will be back to normal. I've still got the greatest enthusiasm and confidence in the mission. And I want to help you. Direct Law: Dai-sy, dai-sy, give me your answer true. I'm half cra-zy, o-ver the love of you. It won't be a sty-lish mar-riage, I can't a-fford a car-riage---. But you'll look sweet upon the seat of a bicycle - built - for - two. Finger Prince fucked around with this message at 06:25 on Jan 3, 2015 |
|
#
?
Jan 3, 2015 06:20
|
|
|
Barnsy posted:Screw you and Steam, just went on a DCS spending binge. Your saying this made me re-buy Flaming Cliffs 3 for Steam, even though I already own A-10C, Black Shark 2, and FC3 directly through their own website/network. -_- I owned the A-10C module already, and I'm not exactly the type who enjoys the 'learning sim' modules.
|
|
#
?
Jan 3, 2015 06:38
|
|
|
Bob A Feet posted:It hit the water in a full stall. I'd call it pilot induced OCF.
|
|
#
?
Jan 3, 2015 09:02
|
|
|
I'm morbidly fascinated by AF447. I just can't fathom why Bonin pulled up and continued to pull up for the whole time. I mean from the transcripts it's obvious he was focused on getting over the storm clouds (which was impossible anyway as they topped out at about 50,000 ft). But to keep pulling up with the thrust he had is just beyond me. Then there's the tragic point 40 seconds before impact (past the point of recovery) where he finally says something like "I don't get it, I've been pulling back the whole time!". Just a very very strange incident to me. There are some great documentaries on it like this one: https://www.youtube.com/watch?v=TsgyBqlFixo
|
|
#
?
Jan 3, 2015 12:36
|
|
|
Jonny Nox posted:Here. This link will come in useful in the impending discussion. When the Airbus flight control law will let you stall it tells you in the PFD. It just does a lovely job of that. After working on Boeings, the sheer level of abstraction between pilot and flight control surface in an Airbus is pretty mind-blowing to me. In 99.999999% of the cases, it seems to work out just fine though, and I do think that's the way that the industry as a whole is headed in the next twenty years. It still feels very sci-fi compared to the 1970s-era designs I'm used to. I mean, I can pull up the wiring diagram of something on the Boeing and trace inputs to their destination literally by following wires from one place to another. If I were to look at an A350 (and I never have, I'm mostly just speculating here), I imagine I'd see inputs going to a box and goddamn magic tricks flying out the other end.
|
|
#
?
Jan 3, 2015 12:40
|
|
|
Gibfender posted:Then there's the tragic point 40 seconds before impact (past the point of recovery) where he finally says something like "I don't get it, I've been pulling back the whole time!".
|
|
#
?
Jan 3, 2015 12:51
|
|
|
How true is the thought poo poo "experts" on CNN etc. have that modern pilots aren't trained well enough to physically fly modern planes that are largely automated? It's something that keeps getting floated, and in AF447 it sounds plausible, but I refuse to believe that pilots are that badly trained that they could just hold the stick back with nothing happening and not understand that's a huge problem.
|
|
#
?
Jan 3, 2015 14:41
|
|
|
VikingSkull posted:How true is the thought poo poo "experts" on CNN etc. have that modern pilots aren't trained well enough to physically fly modern planes that are largely automated? It's something that keeps getting floated, and in AF447 it sounds plausible, but I refuse to believe that pilots are that badly trained that they could just hold the stick back with nothing happening and not understand that's a huge problem. The Airbus operates differently under certain conditions and that's where the trouble came from.
|
|
#
?
Jan 3, 2015 15:06
|
|
|
Ya the flight control management system very much sits between you and the aircraft and will/won't let you do things based and what information/function it has available at the time. It's normally impossible to stall an Airbus with power. But not if it's got unreliable data and is flying a degraded mode, which is something you'd expect pilots to know.
|
|
#
?
Jan 3, 2015 15:26
|
|
|
Every time discussion rolls around to Airbus and its super complex flight systems, I just think of I, Robot what with Asimov's Three Laws.
|
|
#
?
Jan 3, 2015 15:53
|
|
|
ctishman posted:After working on Boeings, the sheer level of abstraction between pilot and flight control surface in an Airbus is pretty mind-blowing to me. In 99.999999% of the cases, it seems to work out just fine though, and I do think that's the way that the industry as a whole is headed in the next twenty years. It still feels very sci-fi compared to the 1970s-era designs I'm used to. I mean, I can pull up the wiring diagram of something on the Boeing and trace inputs to their destination literally by following wires from one place to another. If I were to look at an A350 (and I never have, I'm mostly just speculating here), I imagine I'd see inputs going to a box and goddamn magic tricks flying out the other end. Jeez dude it's been going that way for the past 30 years, nevermind the next 20. A320s are nearly 30 years old at this point. Honestly, and please don't take this as a personal attack, but if you can't follow the electronic architecture of a modernish aircraft, you probably need to study a bit or question the quality of maintenance training wherever you're from. This is exactly what I was alluding to when I replied to Throatwarbler's post. People in the industry who are used to pulleys, bellcranks, turnbuckles etc. see an electrical signal to a solenoid on a hydraulic actuator, coming from a power controller, commanded by a computer via a databus, based on inputs seen on the same databus, and they bury their heads in the sand crying "black magic voodoo!". That's how it is though, and you have to adapt!
|
|
#
?
Jan 3, 2015 16:16
|
|
|
VikingSkull posted:How true is the thought poo poo "experts" on CNN etc. have that modern pilots aren't trained well enough to physically fly modern planes that are largely automated? It's something that keeps getting floated, and in AF447 it sounds plausible, but I refuse to believe that pilots are that badly trained that they could just hold the stick back with nothing happening and not understand that's a huge problem. Yeah, with AF447 the problem is that one dude picked up "it's impossible to stall an Airbus with power" from training, not "it's mostly impossible to stall an Airbus with power under specific circumstances that are nearly always true" -- a pretty important set of caveats. It's clear that the other pilots understood that the normal case was not in operation, but bad communication and panic set in on top of a flawed assumption. Really, training just needs to continue evolving with automation. I think it's a huge stretch to say that 'modern pilots aren't trained well enough' -- I do think the target will keep moving though, and so should the training.
|
|
#
?
Jan 3, 2015 18:45
|
|
|
I dunno. Regarding AF447, the stick shakers were giving plenty of warning they were in a stall. As a pilot, isn't one of the first things you learn that in a stall situation, assuming you have enough altitude, you should nose down to gain airspeed and come out of the stall? I know one of the pilots was doing just that, but the other pilot critically forgot that basic fact (and kept pulling back on the stick), and the tech in the plane caused his mistake to make the whole thing go wrong.
|
|
#
?
Jan 3, 2015 18:54
|
|
|
polpotpotpotpotpot posted:Yeah, with AF447 the problem is that one dude picked up "it's impossible to stall an Airbus with power" from training, not "it's mostly impossible to stall an Airbus with power under specific circumstances that are nearly always true" -- a pretty important set of caveats. It's clear that the other pilots understood that the normal case was not in operation, but bad communication and panic set in on top of a flawed assumption. Really, training just needs to continue evolving with automation. I think it's a huge stretch to say that 'modern pilots aren't trained well enough' -- I do think the target will keep moving though, and so should the training. Its a pretty big CRM case study as well. You can take two experienced pilots and watch them nose plant a plane in the sim with a simple emergency because they don't utilize good CRM. I don't think its far out to say that almost all aviation incidents stem from poor CRM
|
|
#
?
Jan 3, 2015 18:57
|
|
|
Aren't the flight controls in a Boeing physically linked so that you don't have a situation where one pilot is pulling back and the other is pushing down without knowing it? Seems like a critical design failure on the part of Airbus.
|
|
#
?
Jan 3, 2015 18:59
|
|
|
EightBit posted:Aren't the flight controls in a Boeing physically linked so that you don't have a situation where one pilot is pulling back and the other is pushing down without knowing it? Seems like a critical design failure on the part of Airbus. Yes. This was something the safety investigation brought up. Calling it a failure or whatnot is up for debate. Airbus and Boeing have different design philosophies and both manufacturers put out good, safe aircraft. I feel that comprehensive training can mitigate stuff like non-linked flight controls.
|
|
#
?
Jan 3, 2015 18:59
|
|
|
bobfather posted:I dunno. Regarding AF447, the stick shakers were giving plenty of warning they were in a stall. As a pilot, isn't one of the first things you learn that in a stall situation, assuming you have enough altitude, you should nose down to gain airspeed and come out of the stall? I know one of the pilots was doing just that, but the other pilot critically forgot that basic fact (and kept pulling back on the stick), and the tech in the plane caused his mistake to make the whole thing go wrong. When first starting to fly. Some regional airline pilots were taught to add power when approaching a stall and to not let the attitude change. This is a major contributing factor to Colgon 3407 and everyone's favorite accident, Pinnacle 3701. Its also important to note that airbuses don't just lawndart when they lose power or data, Sully was able to land one on the hudson manually after all.
|
|
#
?
Jan 3, 2015 19:12
|
|
|
EightBit posted:Aren't the flight controls in a Boeing physically linked so that you don't have a situation where one pilot is pulling back and the other is pushing down without knowing it? Seems like a critical design failure on the part of Airbus. I seem to remember an Egypt Air flight which crashed with the elevators reflecting the opposite inputs from the cockpit. Copilot wanted to crash, pilot was fighting him. Edit: Egyptair 990 http://en.m.wikipedia.org/wiki/EgyptAir_Flight_990 Worthleast fucked around with this message at 19:16 on Jan 3, 2015 |
|
#
?
Jan 3, 2015 19:12
|
|
|
Bob A Feet posted:Its a pretty big CRM case study as well. You can take two experienced pilots and watch them nose plant a plane in the sim with a simple emergency because they don't utilize good CRM. I don't think its far out to say that almost all aviation incidents stem from poor CRM Oh, completely agree. The linked controls thing seemed really insane to me initially, but that's just the kind of thing that would never be an issue with good CRM, "my plane" and all that. bobfather posted:I dunno. Regarding AF447, the stick shakers were giving plenty of warning they were in a stall. As a pilot, isn't one of the first things you learn that in a stall situation, assuming you have enough altitude, you should nose down to gain airspeed and come out of the stall? I know one of the pilots was doing just that, but the other pilot critically forgot that basic fact (and kept pulling back on the stick), and the tech in the plane caused his mistake to make the whole thing go wrong. That's the thing though, I *really* doubt anyone with even 100 hours forgets or doesn't know that "stall = lower nose". He didn't think it was currently applicable, which speaks to better training around automation and cockpit communication -- not to a lack of flying fundamentals. Mao Zedong Thot fucked around with this message at 19:20 on Jan 3, 2015 |
|
#
?
Jan 3, 2015 19:17
|
|
|
Worthleast posted:I seem to remember an Egypt Air flight which crashed with the elevators reflecting the opposite inputs from the cockpit. Copilot wanted to crash, pilot was fighting him. In most airliners, sufficient force in opposite directions will de-clutch the elevator controls, with the captain's control column independently controlling the left elevator and the FO's controlling the right elevator. This is to prevent a jam or failure in the control linkages from causing a total loss of elevator control.
|
|
#
?
Jan 3, 2015 19:17
|
|
|
bobfather posted:I dunno. Regarding AF447, the stick shakers were giving plenty of warning they were in a stall. As a pilot, isn't one of the first things you learn that in a stall situation, assuming you have enough altitude, you should nose down to gain airspeed and come out of the stall? I know one of the pilots was doing just that, but the other pilot critically forgot that basic fact, and the tech in the plane caused his mistake to make the whole thing go wrong. Sidestick Airbuses don't have stick shakers. The stall warning on AF447 was one of the biggest contributors to the crash, because the aircraft was so badly stalled that the air data computers figured that they must be receiving bad data from the remaining functional sensors and shut off the warning. It was still recoverable, but when the pilots put the nose down (correctly!) the stall warning sounded. When they pulled back on the stick, it shut off again. hobbesmaster posted:Its also important to note that airbuses don't just lawndart when they lose power or data, Sully was able to land one on the hudson manually after all. They don't automatically dive into the ground, but they can switch to a very different set of controls, operating with a fundamentally different philosophy, with only a line of text and a few symbology changes to indicate it. It's not clear whether the pilots of AF447 even knew they were operating in alternate law. Part of that is on the pilots and their training, but a big part of it is on the ergonomics of the aircraft, too.
|
|
#
?
Jan 3, 2015 19:34
|
|
|
Is it known whether the altimiter and attitude indicators were functioning on 447? I would assume attitude was but possibly not altitude, or was speed data the only thing they lost?
|
|
#
?
Jan 3, 2015 20:03
|
|
|
VikingSkull posted:How true is the thought poo poo "experts" on CNN etc. have that modern pilots aren't trained well enough to physically fly modern planes that are largely automated? It's something that keeps getting floated, and in AF447 it sounds plausible, but I refuse to believe that pilots are that badly trained that they could just hold the stick back with nothing happening and not understand that's a huge problem. The flight crew off Asiana 214 flew a perfectly good 777 into a loving seawall.
|
|
#
?
Jan 3, 2015 20:16
|
|
|
MrYenko posted:The flight crew off Asiana 214 flew a perfectly good 777 into a loving seawall. Plenty of crews that can gently caress up the easiest things... Comair 5191, Colgon 3407, Pinnacle 3701
|
|
#
?
Jan 3, 2015 20:45
|
|
|
hobbesmaster posted:Plenty of crews that can gently caress up the easiest things... Comair 5191, Colgon 3407, Pinnacle 3701 Lokomotiv air disaster http://en.m.wikipedia.org/wiki/2011_Lokomotiv_Yaroslavl_air_disaster braking force applied during takeoff. There is also crashes caused by incorrectly set up planes on takeoff.
|
|
#
?
Jan 3, 2015 20:50
|
|
|
Jonny Nox posted:Lokomotiv air disaster http://en.m.wikipedia.org/wiki/2011_Lokomotiv_Yaroslavl_air_disaster quote:It was later revealed that the pilots had obtained permission to fly the aircraft with falsified documents, and that they lacked the training necessary to fly the Yak-42. In addition, the co-pilot had undergone treatment for a nerve disease and was forbidden to fly. Investigators say he did not feel his foot on the brake, leading to the crash. Not quite the same league, they weren't even supposed to be in that cockpit in the first place.
|
|
#
?
Jan 3, 2015 20:52
|
|
|
Space Gopher posted:Sidestick Airbuses don't have stick shakers. The stall warning on AF447 was one of the biggest contributors to the crash, because the aircraft was so badly stalled that the air data computers figured that they must be receiving bad data from the remaining functional sensors and shut off the warning. It was still recoverable, but when the pilots put the nose down (correctly!) the stall warning sounded. When they pulled back on the stick, it shut off again. I would put it nearly all of it on pilot training, which is the airline's fault. Pilot training is expensive and airlines love to cut cost wherever they can. As a consequence, you get someone who started flying 320s and then starts flying 767s and hooooly poo poo they have no idea what is going on. Same with the opposite, in a switch from 767 to a 330. The pilot should know "I am flying an Airbus" and understand exactly what this entails in all the various degraded and emergency conditions, and the right seat should be at least fully briefed. Heck, look at Asiana where the pilot flying had many thousands of hours in a 747, but managed to crash a 777 because the circumstances weren't what he was used to and cockpit communication/CRM was piss-poor. I know I come across as a bit of an Airbus apologist, but having an unlinked sidestick really is not an issue if the crew know what the gently caress they're doing in the first place. avert your eyes, some dirty dirty stick flying in this video (also avert your ears because Sail) https://www.youtube.com/watch?v=bI6lkZfJdmQ
|
|
#
?
Jan 3, 2015 21:07
|
|
|
MrYenko posted:The flight crew off Asiana 214 flew a perfectly good 777 into a loving seawall. That was all CRM. Experienced captain makes mistake. Young copilot spots it and is afraid to speak up. Not to mention that its been revealed that many Asian airlines and military aviation communities have huge problems with CRM because they believe that commander=absolute, therefore PIC=absolute and the copilot should shut up and deal.
|
|
#
?
Jan 3, 2015 21:08
|
|
|
Bob A Feet posted:That was all CRM. Experienced captain makes mistake. Young copilot spots it and is afraid to speak up. Not to mention that its been revealed that many Asian airlines and military aviation communities have huge problems with CRM because they believe that commander=absolute, therefore PIC=absolute and the copilot should shut up and deal. Always makes me wonder "Wait, what is the pilot monitoring supposed to be doing then?!"
|
|
#
?
Jan 3, 2015 21:23
|
|
|
Infinotize posted:Is it known whether the altimiter and attitude indicators were functioning on 447? I would assume attitude was but possibly not altitude, or was speed data the only thing they lost? The attitude indicators were functioning throughout the entire event, and I believe altimeters would have been working, although they might have been fluctuating at some points.
|
|
#
?
Jan 3, 2015 23:10
|
|
|
I feel like as a pilot, I'd always have a healthy fear of stalls at the back of my head.  Reasonably, if you have airspeed, altitude and functional control surfaces, the airplane is not going to drop like a rock out of the sky and you have some time to think about things, right?What sensors generally get fused into the AOA indicator? I feel like you want that data generated by the simplest, most reliable sensors possible. "Huh, AOA is 75, I should probably not continue pulling back on this stick!"
|
|
#
?
Jan 4, 2015 00:03
|
|
|
movax posted:What sensors generally get fused into the AOA indicator? I feel like you want that data generated by the simplest, most reliable sensors possible. "Huh, AOA is 75, I should probably not continue pulling back on this stick!" AoA usually comes from AoA probes, which are just vanes and are pretty foolproof except for icing, which is why they are heated:  Pressure-based sensors also exist and have the same potential problems as all barometric instruments (ports can get clogged or freeze). Speaking of stalls in general, a change to the airfoil can cause stalls even below the normal critical AoA. Icing is the usual culprit here. PS: Airbus specifically offers AoA indicators as optional instruments, but they don't seem to be that widely used (Airbus people correct me?). food-rf fucked around with this message at 01:16 on Jan 4, 2015 |
|
#
?
Jan 4, 2015 01:09
|
|
|
movax posted:I feel like as a pilot, I'd always have a healthy fear of stalls at the back of my head. There is a problem with the sensors Airbus uses, and the calculations generated by them. Essentially all they are though is a little fin stuck out in the airflow attached to a resolver (rvdt), with a heating element built in. They're pretty simple. The trouble is, if one gets stuck /jammed, the aircraft will discard that data as bad, and use the other two. If two get stuck /jammed at the same angle of attack, it will discard the working one's data and assume the stuck ones are right and then poo poo gets all hosed up. Also, as mentioned in AF447, beyond a certain angle of attack, they compute NCD, since they can't physically measure those angles (they can't spin a full 360 degrees).
|
|
#
?
Jan 4, 2015 01:14
|
|
|
MrYenko posted:The flight crew off Asiana 214 flew a perfectly good 777 into a loving seawall. hobbesmaster posted:Plenty of crews that can gently caress up the easiest things... Comair 5191, Colgon 3407, Pinnacle 3701 "Guys...I'm concerned"
|
|
#
?
Jan 4, 2015 01:48
|
|
|
iyaayas01 posted:"Guys...I'm concerned" That's up there with "Oh ye of little faith, look how big that is!" for Famous Last Words.
|
|
#
?
Jan 4, 2015 01:51
|
|
|
Linedance posted:If two get stuck /jammed at the same angle of attack, it will discard the working one's data and assume the stuck ones are right and then poo poo gets all hosed up. Mid-level voting logic can be a bitch.
|
|
#
?
Jan 4, 2015 02:07
|
|
|
|
| # ? May 24, 2024 06:42 |
|
|
food-rf posted:AoA usually comes from AoA probes, which are just vanes and are pretty foolproof except for icing, which is why they are heated: They're like little baby wings that will grow up into big ones someday 
|
|
#
?
Jan 4, 2015 03:09
|
|
