Well yes exactly, there have been far too many accidents where even the most rudimentary scan of the instruments would have told the crews that the autopilot was not doing what they presumed it was doing.One is reminded of the old joke: 'How many airline pilots does it take to change a light bulb?' It certainly doesn't take three, and yet, unbelievably that is what caused the crash.Al

...Unless the elevators were jammed (no evidence of that as yet)...

On the contrary it seems the THS was commanding a 13 degree up position during the most part of their last minutes. Maybe due to pilot input or not...A lot in this interim report needs further context to make any sense for aviation savvy people. I for one is eagerly waiting for the complete transcript of the cockpit dialog to understand the crew train of thought and what decisions were made.Cheers,

Yup, until we know more info, most of this thread is speculation. Not that the speculation has been uninteresting, but that's mostly what it is from the info we have.Al

That L-1011 crash over a light bulb again shows how "automation dependency" causes pilots from doing the most fundamental task "fly the airplane". For both pilots to stop the "scan" turned out to be tragic, and that crash shows how pilots get fixated on a problem and let automation totally fly the aircraft. Another example how little events caused that crash. As luck would have it, in the L-10ll the altitude alert chime was located under the engineer's workstation. During the slow decent the engineer went down below to visually check the nose gear and it was at that point when the alert chime sounded. Was that alert chime loud enough for the other pilots to hear? Guess not because they didn't hear it, most likely because they were so focused on the bulb issue. Had the engineer been at his workstation, would he have heard the alert chime? Maybe, maybe not..

I for one is eagerly waiting for the complete transcript of the cockpit dialog to understand the crew train of thought and what decisions were made.Cheers,

DittoRJ

I don't know why the L-1011 crash is so important. That was back in 1972!!!You definitely can't compare cockpit ergonomics, CRM or pilot training from todays aviation with 40 year old technology and training!The Aviation Industry usually learns from such accidents

I don't know why the L-1011 crash is so important. That was back in 1972!!!You definitely can't compare cockpit ergonomics, CRM or pilot training from todays aviation with 40 year old technology and training!The Aviation Industry usually learns from such accidents

You're right, the aviation industry will always be learning from such accidents, because technology will always change.As for history one should always learn from it, so you don't repeat the same mistake...RJ

There are many glass cockpits where the standby instruments are not in the best position to be easily read. Bring back steam!!!! :( There was some months ago an Alitalia Airbus flight returning across the Atlantic that had a complete cockpit blackout for 15 minutes. Fortunately for them there was a supernumary in the cockpit so one pilot "flew" the aeroplane while the others got out their torches and paged through the manuel to sort the problem out.If it is confirmed that the AF crew did have visual rate of descent and reliable AH information then they should have identified the stall. Rightly or wrongly there is the impression that the greater the descent rate the more they pulled up on the stick. for that we will have to wait to be confirmed or not. Sadly there have been too many accidents where the pilots have done completely the opposite to their training. To list just two:-G-ARPI: on 18 June 1972 A HS Trident departing LHR super stalled into the ground. The stick shaker and pusher activated three times because the slats had been retracted instead of the flaps causing the a/c to go immediately into a stall. What did the crew do? The disactivated the stick shaker/pusher!!! They were at FL60 with absolutely no chance of a recovery.5N-ABD 20th November 1969 A Nigerian Aiways VC10 flew into the ground at a descent rate of 1,600fpm The pilot paid no attention to his descent rate and his height. Nor did the rest of the crew pay attention to what the pilot was not doing.vololiberista

I don't know why the L-1011 crash is so important. That was back in 1972!!!

Doesn't matter if it was from 1972 or 1922, I brought it up in response to a comment about accusing the crew being inappropriate, so it was an example to demonstrate that if anyone thinks a flight deck crew is always beyond reproach and that their actions should not be questioned in an investigation, then they are mistaken. It's not about trying to blame the crew, it's about trying to determine the cause.In an air accident investigation, one should question everything, and especially the crew's actions, because if there is one thing humans can do, it is make mistakes, and should that prove to be the case and some mistakes were made, then there can be measures taken to try and determine why they made the mistake in order to prevent it from happening again.Look at the Air Inter Flight 148 accident in 1992 as an example of a crew error leading to a system redesign following the accident investigation. Air Inter 148 was an Airbus A320 which flew into the side of a hill whilst under autopilot control. It was a mistake made by the crew which caused that, but an understandable one; the MCP was not well understood by the crew, nor fully clear in the way it displayed info. As a result, the crew left it in Vertical Speed mode instead of FPA mode, causing them to select a descent rate of 3,300 feet per minute when they thought they were selecting a flight path descent angle of 3.3 degrees. That mistake caused Airbus to modify the way the A320 displays vertical speed, so the accident would not occur again.It was the actions of the crew that caused the crash in the case of Flight 148, but nobody holds them fully responsible for it.Al

Which comes back to not flying the aircraft,not picking up on a 3300 fpm decent,or in other words, a unstabilized approach. Basic airmanship.The same as flying thru level 5 thunderstorms.Or in other words, fear of overriding dispatch.

Which comes back to not flying the aircraft,not picking up on a 3300 fpm decent,or in other words, a unstabilized approach. Basic airmanship.The same as flying thru level 5 thunderstorms.Or in other words, fear of overriding dispatch.

Which part of:"At 2 h 12 min 02, the PF said "I don’t have any more indications", and the PNF said "we haveno valid indications".(p. 3 of the BEA Report) is not clear to you?You get the -3,300 fom indication by opening a window and sticking your finger out?Cheers,- jahman.

I wonder just what proportion of flight cockpit voice recorder transcripts conclude with the words, "Oh sh*t!"

I am referring to the Air Inter crash,which you brought up, where they should have gone missed with a 3300 fpm decent rate on the approach.That is basic airman ship,not being on speed and stabilized below 1000 in IMC is cause for a missed approach.

I don't know why the L-1011 crash is so important. That was back in 1972!!!You definitely can't compare cockpit ergonomics, CRM or pilot training from todays aviation with 40 year old technology and training!The Aviation Industry usually learns from such accidents

In '72 humans were flying airplanes. The year is 2011 and humans, people like you an I, are still in the cockpit. The "human factor" is at much at work today as it was back in 1972.For instance:Air Florida Flight 90- 1982Comair Flight 191- 2006Colgan Air Flight 3407- 2009All crashes caused by stupid mistakes. :( Unfortunately all with considerable loss of life. I should mention I do not believe this AF crew made a "stupid" mistake, but rather an understandable one given the circumstances.

I just had an experience last night that made me think of what the Air France crew had gone through.It was my leg and we were coming back home around midnight. We knew that there were cells around home from the weather reports and forecasts. Our nexrad was unfortunately down so we relied on just the radar. We were painting the sky -5° down to +5° up and 45° left and right of center. The radar was picking up sporadic cells but nothing along our flight path. Nearing our TOD point center called out a cell to our 11 o'clock 10 miles away. We began to focus the radar in this area and nothing was showing up. About 1 minute later we hit severe turbulence and the autopilot clicked off. Aircraft was banking up to 60° in either direction and we we began deviating to the right. This didn't help so we climbed up out of it and flew for about 20 miles with the interior and exterior lights off so we could visually find the clouds.I don't know the situation of the Air France aircraft like some of you think you do but I can say this was a real eyeopener on how things could really go south fast when you think you are clear of the danger.

I'm curious about the Airbus FBW pitch control. It's said that in both normal and alt law, that pitch is controlled by load factor demand, with neutral stick meaning 1.0g load factor. It's suggested that in alt law the feedback and gains are modified, with no explanation of how or why. I've seen reference that accelerometers are used to measure g-force, which is reasonable, but I wonder if there are other inputs? I guess in normal law maybe AoA is also considered but not sure, and not for alt law either (aside from protections, which we know aren't enforced in alt law). Also not sure how elevator and THS are coordinated to manage load factor.I've seen it suggested that at cruise without AP on, the normal reaction with neutral stick is to gradually induce a +300 FPM climb.Edit: I suppose CG location would impact that?I haven't seen where loss of AP would create a sudden out of trim condition, and it's hard to see how this works if in fact load factor is being monitored/controlled. But I suppose in a high AoA situation, measuring load factor might be problematic. I've seen many pilots say that in this sort of general scenario, to maintain a set pitch attitude. As a non-pilot, ISTM that this would be a challenge when your input is for load factor, not pitch. You would have to have an intuitive understanding of the relationship, or look for visual feedback on your instruments, right?scott s..