Air France 447 Documentary

[goforbroke]

I did a quick search, but couldnt find a thread for this, I apologise if this is a repost.

I watched that documentary the other week, and they talked about the aircraft possibly losing all methods of tracking airspeed. I understand that the aircraft collect this information from sensors in the Pitot tubes, but do aircraft not use GPS at all? Why doesnt an aircraft have a GPS backup incase of a failure like this.

[ninja-lewis]

goforbroke said:

I did a quick search, but couldnt find a thread for this, I apologise if this is a repost.

I watched that documentary the other week, and they talked about the aircraft possibly losing all methods of tracking airspeed. I understand that the aircraft collect this information from sensors in the Pitot tubes, but do aircraft not use GPS at all? Why doesnt an aircraft have a GPS backup incase of a failure like this.

GPS would only measure your ground speed. Much more important is your air speed (speed relative to the air) as that determines whether you have lift or stall or indeed are exceeding the limitations of the airframe.

Edited by ninja-lewis on Monday 7th June 12:13

[Nicholas Blair]

There is a thread somewhere from last week, but I missed the show. Don't seem to see it repeated anywhere.

[goforbroke]

Hmm, so I guess seeing as you are working in 3 dimensions that the GPS could not accuratly measure your speed if you were flying anything other than a level flight path e.g. diving towards the ground then. I never thought about it like that.

I did have a quick look through a few previous pages for a topic on this, and a quick search found nothing. I watched it recorded on the PVR, but as it's been a couple of weeks now since it was aired I assume it wont even be on 4OD anymore. I think the best bet is to wait to see if it's repeated on More4 in the near future.

[ninja-lewis]

goforbroke said:

Hmm, so I guess seeing as you are working in 3 dimensions that the GPS could not accuratly measure your speed if you were flying anything other than a level flight path e.g. diving towards the ground then. I never thought about it like that.

I did have a quick look through a few previous pages for a topic on this, and a quick search found nothing. I watched it recorded on the PVR, but as it's been a couple of weeks now since it was aired I assume it wont even be on 4OD anymore. I think the best bet is to wait to see if it's repeated on More4 in the near future.

It's more that GPS obviously can't measure the flow of air over the wing. Wind will obviously affect the flow of air (if the headwind is 100 knots and your airspeed is 100 knots your ground speed would be 0 - you're effectively hovering over a point on the ground) but there's also the angle of attack of the wing. If the AoA gets too high, the flow over the wing is reduced and eventually the aircraft stalls - it is no longer generating sufficient lift. But the flight envelope is also bounded at the other end - go too fast and you may exceed the aircraft's critical mach - the maximum speed at which air can travel over the wing without losing lift (due to shockwaves and flow separation).

Edited by ninja-lewis on Monday 7th June 16:30

[Geneve]

Ground speed is irrelevant to aerodynamics - a pilot, and many aircraft systems, need to know the speed of the airflow over the aircraft.

Those pitot tubes measure dynamic pressure and are crucial to safe flight. You'll often see them protected with a cover saying 'Remove Before Flight'. If you don't remove them it's the same as being blocked - there's no airspeed information - probably the most important detail the pilot needs.

[SamHH]

goforbroke said:

I did have a quick look through a few previous pages for a topic on this, and a quick search found nothing. I watched it recorded on the PVR, but as it's been a couple of weeks now since it was aired I assume it wont even be on 4OD anymore. I think the best bet is to wait to see if it's repeated on More4 in the near future.

Is this what you're looking for?
http://www.bbc.co.uk/iplayer/episode/b00sndh5/Lost...

[goforbroke]

SamHH said:

goforbroke said:

I did have a quick look through a few previous pages for a topic on this, and a quick search found nothing. I watched it recorded on the PVR, but as it's been a couple of weeks now since it was aired I assume it wont even be on 4OD anymore. I think the best bet is to wait to see if it's repeated on More4 in the near future.

Is this what you're looking for?
http://www.bbc.co.uk/iplayer/episode/b00sndh5/Lost...

Not me, but that's a handy link for anyone that wants to see it. I was sure it was on C4 though, obviously not.

Thanks all for the info regarding airspeed, you've broadened my knowledge a little bit more now.

[cpufreak]

the thing that annoyed me with that program, is it didn't give any new information that wasn't in the public domain.

It also didn't touch on *why* the pilots might have chosen not to change course more drastically (There was conjecture last year that it could have meant a diversion to another airport due to insufficient fuel being carried onboard to make such a detour, due to pilots being awarded bonuses on cost savings...)

[crisisjez]

Most modern pilots are encouraged to allow the automatic systems to operate the Aircraft for almost the entirety of any flight.
This has resulted in a general degredation of Piloting skills.
There will always be doubt about what actually brought AF447 down but IMO none of the failures would have done so.

[anonymous-user]

crisisjez said:

There will always be doubt about what actually brought AF447 down but IMO none of the failures would have done so.

You could say that about almost any failure on an aircraft with the exception of a bomb or some other nasty event leading to a complete structural failure. Unreliable airspeed in the dark over the atlantic with some convective weather about is almost as bad as it gets. In my list of "what I wouldn't want to go wrong on an aircraft" It's right up there at the top with fire in the flight deck/cabin.

All the instruments reading differently, a huge list of conflicting failure messages and aural warnings in the dark in or near a thunderstorm. Good luck with that. Sure that alone doesn't make the aircraft explode but it's a bit lazy to suggest the crash was due to the pilots skills.

[crisisjez]

el stovey said:

You could say that about almost any failure on an aircraft with the exception of a bomb or some other nasty event leading to a complete structural failure.

Yes....I would.

el stovey said:

Sure that alone doesn't make the aircraft explode but it's a bit lazy to suggest the crash was due to the pilots skills.

Well as there will never be a true account of this accident then all that leaves is conjecture.
But based on personal experience the role of the modern Pilot is very much centered around the Automated operation of the Aircraft, adherence rigidly to SOP's and knowing where to find thing's in the books.
In this instance that Automation was no longer available and I would hazard a guess that the last time either of the 2 poor guys up front hand flew was for a few minutes during their last LPC/OPC. Which means their scan would be poor at best, certainly not up to the task of flying limited panel due to unreliable airspeed and possibly in Direct Law to boot.
That, coupled with the conditions they faced, stack's the odds well against them.
This is not a critisism of them, but of the Industry BTW.

Edited by crisisjez on Tuesday 8th June 12:14

[Turbodiesel1690]

I find it hard to believe that these pilots were incapable of flying the aircraft without instruments but with all other systems OK (engines / power / flight controls). Assuming this was the case of course. Surely they would have some idea of their speed based on the throttle settings and would therefore at the very least be able to keep the plane from slowing to a stall or going too fast just by 'flying the plane'??

[ninja-lewis]

Turbodiesel1690 said:

I find it hard to believe that these pilots were incapable of flying the aircraft without instruments but with all other systems OK (engines / power / flight controls). Assuming this was the case of course. Surely they would have some idea of their speed based on the throttle settings and would therefore at the very least be able to keep the plane from slowing to a stall or going too fast just by 'flying the plane'??

Two points.

1) Most modern airliners have autothrust. Pilots set the desired range of thrust on a dial and then the computers automatically adjust the thrust within that range - e.g. reducing thrust when encountering turbulence for a more comfortable flight. But when the pitot tubes failed, the computers no longer knew the aircraft's air speed and thus the autothrust tripped off. Procedure at this point is typically to set a specific level of throttle and pitch up - which gives you fairly good estimation of what your air speed will be. When the autothrust tripped off, the BBC speculate that it was set at a lower level than the actual thrust levers would indicate - because on Airbus aircraft the thrust levers do not move with the autothrust. Thus the pilots would look at their thrust levers (their minds would likely focus on limited visual inputs directly in front of them under stress) and believe they already had the required thrust level set. One of the investigators on the BBC programme found evidence of several incidents where the pilots took more than 60 seconds to correct their thrust when autothrust tripped off.

2) The other issue with flying the plane is that the flight envelope at altitude and in a thunderstorm at night is not the same as in the circuit on a calm day. The window between stall speed and critical mach speed gets narrower as altitude increases - for example the U2 at it's operational ceiling has a 10 knot window - until you reach a point called coffin corner where the two meet and stable flight is impossible. The accurate flying this area requires would be extremely difficult without air speed indicators, let alone in the middle of a severe thunderstorm. It's quite possible that the aircraft was out of control before the pilots had a chance to make sense of the error messages they were seeing or that the thunderstorm simply made it impossible to keep the aircraft within the narrow window.

[mattdaniels]

Turbodiesel1690 said:

Surely they would have some idea of their speed based on the throttle settings and would therefore at the very least be able to keep the plane from slowing to a stall or going too fast just by 'flying the plane'??

You're thinking about the throttle like in a car (got my foot flat to the floor means I must be going fast).
Having aircraft throttle rammed forward does not necessarily equate to having enough airspeed (and thus lift) to prevent the aircraft stalling.

[crisisjez]

In an Aircraft power controls performance (Level,climb or descend)
Attitude controls airspeed.
Bottom line is that very few Modern pilots could tell you the correct attitude and thrust settings without looking them up in a book.
Which could easily be too late, and may well have been

[crisisjez]

ninja-lewis said:

When the autothrust tripped off, the BBC speculate that it was set at a lower level than the actual thrust levers would indicate - because on Airbus aircraft the thrust levers do not move with the autothrust.

Regardless of wether the autothrust computer was signalling a thrust reduction or not, at the time when the Autothrust disconnected the thrust levers were physically in the CLM gate so the engines would have immediately spooled up to a climb thrust setting. Now as they were trying to maintain level flight this would have resulted in a significant increase in airspeed and could quite easily have reached an overspeed condition.
The Pilots would have had to physically moved the thrust levers out of the climb gate to prevent this from happening, something that should be instinctive, but sadly there are a few major Airlines out there who do not actually allow their Pilots to ever manually control thrust.
For these Pilots manual thrust would no longer be instinctive and would not form part of their thought process untill some indication of the lack (or excess) of thrust prompted a delayed reaction.

Like I said...all speculation, but if it does prove to be Pilot error, which it almost certainly will be....then its really Airline and Manufacturer error for turning us into monkeys

[tank slapper]

Turbodiesel1690 said:

I find it hard to believe that these pilots were incapable of flying the aircraft without instruments but with all other systems OK (engines / power / flight controls). Assuming this was the case of course. Surely they would have some idea of their speed based on the throttle settings and would therefore at the very least be able to keep the plane from slowing to a stall or going too fast just by 'flying the plane'??

It isn't as simple as that unfortunately. I am sure a pilot will be along shortly to clarify, but you can't fly an aircraft by feel without visual reference, as the feeling you get from the aircraft are not always an accurate indication of what it is doing. In bad weather or at night flying is done by referring to the instruments, and if these are giving inaccurate data or no data then it would not take much for the aircraft to end up in an unrecoverable situation.

[crisisjez]

tank slapper said:

Turbodiesel1690 said:

I find it hard to believe that these pilots were incapable of flying the aircraft without instruments but with all other systems OK (engines / power / flight controls). Assuming this was the case of course. Surely they would have some idea of their speed based on the throttle settings and would therefore at the very least be able to keep the plane from slowing to a stall or going too fast just by 'flying the plane'??

It isn't as simple as that unfortunately. I am sure a pilot will be along shortly to clarify, but you can't fly an aircraft by feel without visual reference, as the feeling you get from the aircraft are not always an accurate indication of what it is doing. In bad weather or at night flying is done by referring to the instruments, and if these are giving inaccurate data or no data then it would not take much for the aircraft to end up in an unrecoverable situation.

You do still have attitude information as this is not dependant on Pitot Tubes, you would not have altitude or vertical speed data, but the GPS nav page on the navigation computer will give you sufficient information to maintain level flight.
It would have been a stty place to be that's no doubt, but any bell, whistle, light or naked Hosty takes a far distant second to maintaining control of the Aircraft. (Well maybe not the Hosty)

[ninja-lewis]

crisisjez said:

ninja-lewis said:

When the autothrust tripped off, the BBC speculate that it was set at a lower level than the actual thrust levers would indicate - because on Airbus aircraft the thrust levers do not move with the autothrust.

Regardless of wether the autothrust computer was signalling a thrust reduction or not, at the time when the Autothrust disconnected the thrust levers were physically in the CLM gate so the engines would have immediately spooled up to a climb thrust setting. Now as they were trying to maintain level flight this would have resulted in a significant increase in airspeed and could quite easily have reached an overspeed condition.
The Pilots would have had to physically moved the thrust levers out of the climb gate to prevent this from happening, something that should be instinctive, but sadly there are a few major Airlines out there who do not actually allow their Pilots to ever manually control thrust.
For these Pilots manual thrust would no longer be instinctive and would not form part of their thought process untill some indication of the lack (or excess) of thrust prompted a delayed reaction.

Like I said...all speculation, but if it does prove to be Pilot error, which it almost certainly will be....then its really Airline and Manufacturer error for turning us into monkeys

The argument made in the BBC programme was the opposite. When autothrust disengaged, the engines were left at their current thrust level rather than matching the level set by the levers - apparently this is the way Airbus do it whereas Boeing move the levers with autothrust. Hence looking at the levers (rather than the checking the engine thrust level on the panel) would fool you into thinking that the aircraft was already at the required thrust level. You would need to move the levers out of the gate and back in to bring the engines back to the thrust level of the levers. Failure to recognise this quickly enough would lead to a stall, rather than overspeed.

That's the conclusion the BBC were presenting.