Volume

Sorry for spoiling the discussion with facts from the FDR...
For the first 30 Seconds of the event, nose up eleveator never exceeded 1° nose up. (full would be 30°)
For the first Minute of the event, nose up eleveator never exceeded 5° nose up.
Only when allready in full stall and THS already at almost full nose up, the elevator for the first time was fully deflected nose up (and then left there for around 45 seconds with the THS at the full nose up stop as well by that time).
During the first more than 90 seconds the aircraft never "ran out of nose up elevator", it was using less than 30% of the available elevator deflection. It was perfectly possible to fully stall the aircraft using less than 30% of available elevator deflection, but a lot of THS.

tartare

G109 - that is a brilliant article.
How terrifying.
I just read through this:
http://aerosociety.com/Assets/Docs/P...y_Training.pdf
Have followed the whole `children of the magenta line' loss of basic stick and rudder skills debacle for some time now.
A layman's question.
To what extent do airlines practise jet upset recovery training or jet upset scenarios in simulator check-rides that would mitigate the AF447 circumstances?
Is it standard practise / has it been for a long time?
Does it vary from carrier to carrier?
Are there any today who are considered best practise in simulating extreme jet-upsets.
I had just assumed that it'd be standard practise - even allowing for iced pitots (aka partial panel failure exrecises in steam cockpits).
Nasty old check-ride Captain puts the 30 degrees nose in a bank up right near coffin-corner and then smiles and says "Open your eyes - you have control..."

roulishollandais

Boeing's Synthetic airspeed, Airbus BUSS, safe USN's AoA, or Klopfstein's old inertial HUD.

Volume

Can a simulator accurately "simulate" what no test pilot ever dared to fly? Post Stall behaviour of aircraft is very hard to predict. Detached airflow is beyond any CFD capability. Reynolds numbers will make wind tunnel model testing invalid as well. And which combinations of system failure and upset do you want to simulate?
Philosophy today ist to avoid getting there, not how to survive once being there.

jcjeant

I think this philosophy is in force since the flight of Wright brothers

RetiredF4

This statement is esentially misleading and not completely correct. Correct would be the airplane will maintain a loadfactor of 1g. Correct would be if you add ....... if the autothrottle takes care of the speed. What's the difference some might ask.

Otherwise without the autothrottle (that droppped out at AF447 right at the beginning) the aircraft will decelerate in a climb or accelerate in a descent, and the computers would increase the pitch ( if in a climb) or decrease the pitch (if in a descent) to maintain the one g flightpath.

That mindset "set the pitch and it will stay there (and go there) might have influenced Bonins actions with the SS.

Gysbreght

Does it have to be 'accurate'? As long as it indicates to the pilot in a reasonably realistic manner that the airplane is stalled, and elicits the proper response from the trainee, it would perform its intended function. No need to go beyond conditions that 'no test pilot ever dared to fly'.

tartare

It's an interesting point.
Philosophically the more one thinks about swept wing high subsonic airliners the more one realises they are designed to climb benignly, cruise in a stable configuration and descend gently.
Big, vulnerable ocean liners of the sky that are pretty much optimised to point directly into a high mach airflow, and stay there.
A great over simplication of course, but makes you realise how perilous things get very quickly when they are outside that envelope.

Jwscud

I thought it might be an interesting addition to the discussion to note that Gulfstream has just launched the new G500 and 600 with full FBW akin to the 650, but with linked and fully back driven side sticks.

bubbers44

How dare Gulfstream take all the challenge out of FBW by letting both pilots know what the PF is doing with the SS?

The next thing Gulfstream will do is put a big yoke in front of the pilots so even the jump seater can see what they are doing.

CONF iture

And they call them Active control sidesticks :
Active control sidesticks replace the control columns, creating more space and better comfort for pilots. The sidesticks are digitally linked to provide the same response and control of a traditional pedestal-mounted yoke to ensure that both pilots see and feel every maneuver and control input the other pilot makes.

Gretchenfrage

I wouldn't ridicule this bubbers. I think it's great.

Small manufacturers or those not depending on solely profit orientated, cynical airline managers, can change design to reach the modern optimum. A optimum defined by experience and listening to research. For such manufacturers the slogan "we strive for utmost safety" is not only a hollow commercial.

I welcome a moving side-stick, just as i welcomed a moving auto-throttle.
Both systems allow the pilots to instantly be part of the modern fly-by-wire system, to act in a symbiotic partnership and, if needed, to act as last authority (at least if minimally trained, but that's another story).

The two biggies in the airline market are cornered in their geriatric philosophy due to silly commercial statements of their CEO's, or due to looming lawsuits if they would change only the slightest small design flaw (thanks to the cynical lawyer brigade).

One of them could easily revert to the originally intended design and relieve us from the stumbling block between our legs for 16 hours, the other could finally admit that there are still humans at controls and such creatures work better with all their senses than if deprived by some.

Again, i welcome the design by Gulfstream, their cockpit reflects what should be in any Airbus or Boeing.

BOAC

I saw it, OK, before you were 'trashed'

To cut a long story short - has anyone - FAA/CAA/EASA/Airbus/Boeing come up with a solution to what appears to be a burgeoning 'blackman in the woodpile' - Autotrim?

DozyWannabe

Until some tired spanners wires them up the wrong way - then you have one side electronically forcing opposite motion into the correctly-wired side.

As has been stated before (and not just by me) there are pros and cons to all the existing control philosophies - obviously pilots of the old-school will prefer the closest possible mimicking of traditional behaviour, but that does not necessarily mean traditional is either better or better for everyone.

@BOAC - I still think the autotrim situation is being rather over-emphasised by those who want to redirect blame to the control philosophy and the manufacturer more than they want to do anything else (including solving the actual problems that led to this accident).

BOAC

- it is not a 're-direction'. It is a contributary factor. Without it one hopes this deficient crew (and THY at AMS, and TOM at BOH and........and....) would have noticed the need to keep trimming?

DozyWannabe

I'm not going to say the crew were "deficient", in this particular instance some bad choices were made and, as with a lot of CRM-centric accidents, they never managed to work as a team. They were let down much earlier in the chain by their airline and the regulator, however, and yes - there are some questions about the aircraft systems behaviour.

Autotrim is a fundamental aspect of how the Airbus FBW system operates normally, and causing it to drop out in an abnormal situation when there's nothing stopping it from working is probably overkill. The way it behaved is a contributory factor in this instance, however I have trouble accepting this behaviour:



as "insidious". It is a clear and definite response to the SS input, going from -3deg to -13.5deg in around 45 seconds.

Theoretically there may be an issue with a small climb component causing the THS to trim NU in a less obviously-perceptible way, but that's not what happened here. Re-instating the limits apparent in the A320's system would probably help matters - either limiting to a smaller NU trim in Alternate or halting NU trim entirely with the stall warning (which you can see coming on just as the THS starts rolling NU).

BOAC

- precisely. It does, however, hide from a crew that extra cue that the speed is reducing when it should not be. In 'the old days', you would need to keep applying back stick until you ran out of it, and trim the THS to restore elevator margin. This system allows the concierge to blythely point the aeroplane at the stars and not worry about a thing while the speed heads rapidly south - without any cues.

In order to 'dumb down' the art of flying an aeroplane the manufacturers have created large blinkers for crew to wear.

DozyWannabe

The design is not about "dumb[ing] down", and never was. If you look at things with that assumption then aspects of the design will look that way - but that assumption was applied after the fact by pilots who took too much notice of BZ and not enough of Gordon Corps.

Autotrim exists in this system primarily for the very reason you describe - namely the passive sticks. Before we get away from ourselves, it's worth bearing in mind that it has worked just fine for millions of flights over 26 years in service. We are looking at a single anomaly here, and one in which the controls as applied were *way* outside what should have been attempted even in an emergency situation.

Perhaps it should become an early "must-know" aspect of Airbus FBW conversion training (if it isn't already) that when you're using the stick to fly the aircraft manually, you're affecting trim as well as elevator position.

Circling back to the initial point, the design was not an exercise in "dumbing down", it was simply one possible solution for aircraft control when given the opportunity of a "clean room" design. In a hypothetical scenario where aviation began in 1983 rather than 1903 (and materials/computing tech were unaffected), what the '80s Wright Brothers may have come up with would not look like the 1903 Flyer - it'd probably be a carbon-fibre job festooned with digital technology and operated in an entirely alien manner. With autotrim we're looking at an aspect of the Airbus design (specific to the A330/340 in fact) that *may* have made things a little more difficult, but has done so only once in however many million hours of flying the type has done. Seems a little excessive to say the overall design was a mistake based on that single incident, does it not?

BOAC

- don't Boeings have it too?

rudderrudderrat

Hi DW,
Agreed. Normal Law with its protections is very good.

But even you may be persuaded that ALT Law (without protections but with auto trim continuing beyond the stall warning) doesn't have a good record.

As Bpalmer says in post #773 "G-load demand is a crappy flight control law to be in for stall recover. When the airplane starts to fall (accelerate downward G <1.0) the airplane's reaction to maintain a neutral-sidestick command of 1.0g is up elevator, followed by nose-up stabilizer. ooops."