Go Back  PPRuNe Forums > Flight Deck Forums > Tech Log
Reload this Page >

AF447 final crew conversation - Thread No. 1

Tech Log The very best in practical technical discussion on the web

AF447 final crew conversation - Thread No. 1

Old 25th Oct 2011, 03:40
  #381 (permalink)  
 
Join Date: Aug 2008
Location: Sydney
Age: 60
Posts: 87
Likes: 0
Received 0 Likes on 0 Posts
'DNA of flying' (i.e. 'regular aircraft'), training and FBW must be an issue.

Originally Posted by HPbleed
Renegade man - good post, however I do feel it's a bit of an airbus bashing. You have to remember that even though the airbus has loads of protections, the control surfaces still work in a normal sense. Stick forward, nose goes down. Stick back, nose comes up. There is no need to put in opposit control to stop or control the movement either, the aircraft just trims for you, so you can set the exact attitude you want then release the stick. So the stall recovery is still the same as for a light aircraft, with the exception that you may actually have to reduce thrust to remove the pitch power couple. I hope that helps you understand the airbus a bit more.
Thanks HPbleed. Not trying to AB bash. Have travelled on most of the AB range and all the Boeings (and few others too). Having been in the IT industry for 30 years I've seen every type of software human interface failure you can think of. And new generations of developers/engineers keep coming through with even bigger convictions that technology will "save us all and take us to the promised land"....Human performance improvement through the art of listening, understanding, acknowledging, valuing and respecting PLUS the development and use of technology is where we'll see the most benefits achieved, but most industries and businesses are still very slow to understand this and the technocrats still push their philosophy without due regard for their and technology's inherent fallibility.

Auto trim sounds just great, but I can imagine even a simple concept like that could have you confused in amongst a systems failure in trying conditions and it's the overall response of the AB's 'ecosystem' (to put it one way) I'm wondering whether has been fully thought through.

Some of the 'protections' built-in to the systems must also be contributing to some deep seated misunderstandings about operations within that 1% area where the machine is no longer 'protecting' a pilot's inputs.

Originally Posted by RenegadeMan. Response by iceman50
"can't possibly have been trained well enough to deal with the conditions they found themselves in that night (dark, IMC, turbulence), the complexities of the flight data systems being compromised by the pitot malfunction and responses of the aircraft to the extremity of the a/p & auto throttle disconnect."
Wrong I am afraid as the Instrument rating is designed for flight in the dark or heaven forbid IMC! The A/C was NOT at any "extremity" when the AP and A/THR disconnected they were virtually STRAIGHT and LEVEL, one of the first things we were taught as PILOTS!
Hey iceman50, I wasn't for a moment saying the crew wasn't trained to handle the dark or IMC, I'm simply saying that these conditions PLUS whatever occurred at AP and A/THR disconnect (and yes I take your point that at that moment it was not 'extreme') PLUS the AB's system's response(s) PLUS the potential poor crew initial response PLUS potentially other things none of us can be clear on yet, facilitated a complex mix that the crew couldn't understand and this must be partly a training issue as well as the man-machine interface psychology issue that makes me concerned this type of accident could readily happen again.
Originally Posted by RenegadeMan. Response by Clandestino
"but if the side stick of an Airbus behaves (sometimes, always or only occasionally such as perhaps when the aircraft is fully stalled) in a manner not dissimilar or even a little bit like the old video game I’ve mentioned above (i.e. the pilot makes an input such as ‘stick fully back’ and a substantial forward stick counteracting input is required to negate the state that the first input leaves the aircraft in)
It does not. Not in any control law. Rest of your post is based on this assupmption and is therefore not true.
Thanks Clandestino; glad you've cleared that up for me (although given I was pretty good at that old video game [which someone reminded me was actually called 'Asteroid'] I was thinking I'd be in a good position to head on up to the cockpit of an AB if the entire crew became disabled and a call went out to the passengers "can anyone onboard fly?"!) Please don't just write off the rest of my post. One of the challenges on forums like this is that if you display some ignorance, everything else you've said just gets flicked away and the potential for the debate to morph into something that might help can be lost. I (and all other crew/SLF) want to be able to travel on these airliners confident the crews flying them are trained well enough to cope with the complex scenarios and multi-faceted problems systems failures can introduce.

The rest of my post was an effort to highlight "(the) man-machine interface psychology, economics, politics and big business needing to come clean and invest more dollars into research and training rather than just about this particular crew's lack of ability, perceived or otherwise". I think that is still the major concern here and how crews can ensure they're ahead of the curve on comprehending what the machine is attempting to do. These highly automated systems are a miracle of ingenuity and technical brilliance but what appears to be missing is an overarching realisation that it is simply not possible to cover every combination of potential system responses to incorrect data input and the likelihood of a subsequent incorrect human interpretation under the extreme stress of an inflight emergency (and being in an stalled airliner [regardless of how the stall was entered, who was responsible or what warning horns were sounding either continuously or intermittently] travelling downwards at 10K FPM with just a few minutes left is an environment that makes the phrase 'extreme stress' sound like an understatement) is very real and likely to occur again and with increasing regularity given lowering of crew standards.

Last edited by RenegadeMan; 25th Oct 2011 at 03:43. Reason: subject truncated (fixed it)
RenegadeMan is offline  
Old 25th Oct 2011, 08:39
  #382 (permalink)  
 
Join Date: Nov 2010
Location: FR
Posts: 477
Likes: 0
Received 0 Likes on 0 Posts
Hi ChristiaanJ

I'm not an 'aerodynamic engineering oriented' people, but I do agree with your post #373 and thank you for it, as it lifted the doubts I still had regarding the terminology.
AlphaZuluRomeo is offline  
Old 25th Oct 2011, 09:13
  #383 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Just have a look at this graphic to confirm what ChristiaanJ has posited.
mm43 is offline  
Old 25th Oct 2011, 09:45
  #384 (permalink)  
 
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes on 0 Posts
Don't forget that the THS at 14 degrees nose up "blanks" the elevators, and since the HS is TRIM, the PITCH is effectively "blanked". When flying with a loaded (unloaded v/v its purpose) tail, PITCH can get "squirrelly" Especially ND. The THS is trimmed ND for aft cg, so 447 was off to a risky start in her climb. Trimming it up, whether auto or manual, is not a good idea. There is a case to be made for her climbing in default with the THS at ND set. This a/c should have been descending, whether computer, or manual. imho. Temps, aft cg, etc.
Lyman is offline  
Old 25th Oct 2011, 10:26
  #385 (permalink)  
 
Join Date: Jun 2009
Location: Germany
Age: 71
Posts: 776
Received 3 Likes on 1 Post
mm43
Just have a look at this graphic to confirm what ChristiaanJ has posited.
Some points i like to remember in viewing this graph:

The AOA was considerably higher than 30°, BEA states that it never went below 35°. In a former thread the AOA was computed (GS and descent rate) at something like 60°.

The graph assumes, that the FPA (flightpath angle) would decrease with the THS and elevator full ND. Out of my practical expierience and my theoretical knowledge with full stalls and spins the FPA would increase considerably in the process of reducing AOA due to increase in descent rate. That leads initially to a chasing of the AOA. If stopped too early, secondary stall is preprogrammed.

Although the graph only shows the THS and the elevators, the behaviour of the complete aircraft including the point when power can be applied again in the individual phases of the recovery is decisive in the validation of a successfull recovery attempt and the altitude required to accomplish safe recovery.

Another point validating the effectiveness and the time/ altitude required for recovery is the rate, with which the FCPC´s would change the position of the elevators in the short term and the THS in the long term. As shown before, the max deflection of the elevators the crew did achieve (with limitied SS down) was a reduction from 30° nose up to 15° nose up. And this reduction didn´t achieve anything noticable to the crew. Manual THS trim (nowhere mentioned in the books except in direct law, and never trained) and prolonged ND input would have been mandatory anyway.

It is easy on the paper, but it is already challenge in the aircraft in pure VMC with all gadgets working and it might have taken a wonder to do it in the Cockpit of AF447.

Last edited by RetiredF4; 25th Oct 2011 at 10:38.
RetiredF4 is offline  
Old 25th Oct 2011, 11:23
  #386 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
mm43 - what is the meaning of 'AoI' in your diargam? If it is 'incidence' I think we need to be careful on definitions, but I cannot see that it has much relevance anyway.

To follow Franzl, some while ago in discussions somewhere in this maze about 'recovery' I said that the pitch change required to unstall could have been as high as 50 degrees nose down - well past the limits most pilots would contemplate. This would indeed have drastically increased the rate of descent until sufficient manoeuvre capability was there for pitching nose up, which is why my subjective assessment of min recovery altitude from THIS stall was around 20k.

I don't really want to be dragged in to the SS/yoke battle except to say that if I had been dragged out of my pit in my silk pyjamas and arrived in a cockpit at 35,000' with 15 degrees nose up, 10,000fpm+ and a co-pilot with the yoke back in his groin, I might have had a clue as to why the nose was up that high - I hope so.
BOAC is offline  
Old 25th Oct 2011, 12:32
  #387 (permalink)  
 
Join Date: Jan 2005
Location: France
Posts: 2,315
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by stepwilk
Horizontal stabilizer.
Thanks, now corrected.
Wrote that after mid-night... should have checked it better.
ChristiaanJ is offline  
Old 25th Oct 2011, 17:05
  #388 (permalink)  
 
Join Date: Jan 2008
Location: UK
Posts: 116
Likes: 0
Received 0 Likes on 0 Posts
Lord Spandex Masher
AD, just curious. How far above the main wing is the 330 elevator? Is it a similar distance as, for example, a BAc 111?

You simply can't confine deep stalls or super stalls to T-Tail only types. It is entirely possible to get a deep/super stall on a conventional (read not a T-Tail) type. A swept wing is actually more prone to a deep stall than a T-Tail but a T-Tail is harder to recover, hence pushers and all that jazz. Remember swept wing pitch up?!
Sorry for the delayed reply. I don't know how far above the main wing the 330 horiz-stab is, but I would guess that the designers would have ensured that, at a main wing stall alpha, the horiz-stab would still be flying.

The BAC 1-11 was a T-tail aeroplane.

I do not believe that a swept-wing aircraft is prone, at all, to a deep stall, as the elevator would normally be available for recovery (provided the PH is not holding the stick hard back!).

Regarding swept-wing pitch up at the stall, I understand that, with washout, the wingtip stalls last, retaining roll-control at the point of the stall. Therefore, the tendency of a swept-wing aircraft is to pitch down, not up, unless someone is holding the stick hard back.

ChristiaanJ - Sorry, while I was composing my rant on deep-stalls, you had covered the point very well.
Aileron Drag is offline  
Old 25th Oct 2011, 17:44
  #389 (permalink)  
 
Join Date: Jul 2009
Location: DFW
Age: 61
Posts: 221
Likes: 0
Received 0 Likes on 0 Posts
Regarding swept-wing pitch up at the stall, I understand that, with washout, the wingtip stalls last, retaining roll-control at the point of the stall. Therefore, the tendency of a swept-wing aircraft is to pitch down, not up, unless someone is holding the stick hard back.
Tips stall first on swept wings, not last. Washout reduces the effect.

edit: tips stall first on rear swept wings. The X29 featured foward sweep to get around this as does the HansaJet.

But I understand your point about pitching.
TTex600 is offline  
Old 25th Oct 2011, 18:14
  #390 (permalink)  
Per Ardua ad Astraeus
 
Join Date: Mar 2000
Location: UK
Posts: 18,579
Likes: 0
Received 0 Likes on 0 Posts
Guys - with the estimated AoA's we have in this accident you can forget which bit 'stalls first' or 'tail blanking' at 60 deg AoA the whole wing is so stalled that it matters not, and the tailplane would be well in the airstream.
BOAC is offline  
Old 25th Oct 2011, 18:45
  #391 (permalink)  
 
Join Date: Jun 2009
Location: Bedford, UK
Age: 70
Posts: 1,319
Received 16 Likes on 10 Posts
From the old days, AOI = Angle Of Incidence.
Mr Optimistic is offline  
Old 25th Oct 2011, 19:18
  #392 (permalink)  
 
Join Date: Aug 2009
Location: Germany
Age: 67
Posts: 1,777
Likes: 0
Received 0 Likes on 0 Posts
The new Air France board of directors after the Stalinist purges

Air France - Corporate : Board of directors
No more Gourgeon and Schramm .. the (bad) arrogants spokespeople after AF447 event
jcjeant is offline  
Old 25th Oct 2011, 20:58
  #393 (permalink)  
 
Join Date: Feb 2005
Location: Correr es mi destino por no llevar papel
Posts: 1,420
Received 0 Likes on 0 Posts
Originally Posted by OK 465
The canard configured Velocity had a deep stall problem initially.

No T-tail.
Nope. It did not.

First, regarding the deep stall, we were discussing airliners, not GA types.

Second, GA canards like: VariEze, LongEZ or SpeedCanard have low wing loading, relatively high speed range (relative to min speed, that is) which necessitates use of large foreplane - compared to wing. Now this foreplane absolutely must be designed so it stalls before mainplane. If it were otherwise, loss of lift at mainplane, combined with still working foreplane woud introduce such a pitching momentum it could tumble the aeroplane end-over-end and that's the issue with GA canards. This tumble is very possibly irrecoverable but is not deep stall.

Third, I am sure that there were numerous witnesses who saw F-GZCP taxi out to her fateful flight and I have strong reasons to believe that not one of them will testify she was canard equipped.

Originally Posted by Lord Spandex Masher
How far above the main wing is the 330 elevator? Is it a similar distance as, for example, a BAc 111?
No. Lower. Further back.

Originally Posted by Lord Spandex Masher
You simply can't confine deep stalls or super stalls to T-Tail only types. It is entirely possible to get a deep/super stall on a conventional (read not a T-Tail) type.
Please provide reference to this statement of yours. I really wonder how is it possible to not put the conventional mainplane below the wing's wake at stall and beyond.

Originally Posted by Lord Spandex Masher
Remember swept wing pitch up?!
Only in the tales of yesteryear, when swept wing was largely an unknown quantity. Since then, wise aerodynamicists came up with something called "washout". When measuredly applied, it assures that roots stall first, thereby: 1)eliminating pitch-up associated with tips that stall first 2) reducing rolling momentum from asymmetric flow separation - separation now occurs closer to CG in lateral sense 3) keeping the outboard ailerons effective. Statement that:
Originally Posted by TTex600
Tips stall first on swept wings, not last. Washout reduces the effect.
was true when Huns were en vogue. Nowadays, on swept wing transports, washout reduces the effect to the point of elimination, as was tragically proven by certain crew of certain airliner that fell flatly into Atlantic without gyrations that have to be expected with tips stalling first


Originally Posted by TTex600
edit: tips stall first on rear swept wings. The X29 featured foward sweep to get around this as does the HansaJet.
In both cases primary consideration was not elimination of wingtips stalling first.

X-29's forward swept wing was intended to direct spanwise airflow towards the fuselage instead of towards the wingtips. While this does reduce the effective AoA of wingtips, with forward swept wing the roots that stall first will cause pitch-up so exacerbating the problem which FSW was supposedly intended to solve.

HansaJet's wing was mid-mounted so there would be minimum fuselage diameter and minimum interference drag from wing to fuselage joint. It got forward sweep so wing centre box would not impede on cabin space. Hansa Jet is T-tailed and stick pusher equipped, therefore her stalling characteristics are of academic interest only.


Originally Posted by RenegadeMan
(the) man-machine interface psychology, economics, politics and big business needing to come clean and invest more dollars into research and training rather than just about this particular crew's lack of ability, perceived or otherwise"
The amount they have already invested in research and training is closer to gigabucks than megabucks. That so far we have no official word that Airbus MMI is sub-optimal or even dangerous boils down to two options a) while imperfect, it really is acceptable b) there's large worldwide conspiracy covering the fact that it really is sub-par and kills people.

Which one to believe? Tough one, ain't it?


Originally Posted by RetiredF4
Manual THS trim (nowhere mentioned in the books except in direct law, and never trained)
Rest of youth post is spot-on, I just have to comment on this. It is not trained because it doesn't work. You can forcibly move the wheel to position of your choice but the FBW will return it to the position it deems to be necessary once you loose your grip and all the while will try to combat the trim with elevator to satisfy G demand.

There's no use and no need for manual trim while auto-trim is working.

Originally Posted by DC-ATE
Why can't we all just admit that if this had been a "conventional" airplane [cable-operated, not computer/electric-operated] that we would not even have a thread on this?!
It is one of the factors but given the contents of the threads we had dealing with AF447 so far, it tells more about ignorance of the posters than alleged complexity of FBW Airbuses.

I suspect that main reasons the AF447 tragedy got so much attention on PPRuNe are a) it involves large body count b) it involves western built and operated widebody c) it involves relatively new flight control technology that is easier to deride than to understand d) most of the posters just can't dismiss the whole affair by using some of their deeply rooted prejudices. If it happened in CIS, they'd conclude it was vodka-caused. If it happened in Africa, it would be corrupt CAAs and shoddy maintenance that doomed it. If it happened in Asia, well it must have been a strong cockpit authority gradient or some similar cultural thing. South America - machismo. Regional airlines - inexperience and poor training. Turkish - exmilguy not well adapted to civvie world.

Well, we're out of prejudices here as it happened to long established western flag carrier. Most of the posters feel that it struck close to home and they can not write it off easily as something-that-can't-happen-to-me. Mental gymnastic they perform to deal with their shock and horror is very interesting thing to behold but its results are not particularly informative.
Clandestino is offline  
Old 25th Oct 2011, 21:20
  #394 (permalink)  
 
Join Date: Jun 2009
Location: Germany
Age: 71
Posts: 776
Received 3 Likes on 1 Post
Quote:
Originally Posted by RetiredF4
Manual THS trim (nowhere mentioned in the books except in direct law, and never trained)
Clandestino
It is not trained because it doesn't work. You can forcibly move the wheel to position of your choice but the FBW will return it to the position it deems to be necessary once you loose your grip and all the while will try to combat the trim with elevator to satisfy G demand.

There's no use and no need for manual trim while auto-trim is working.
A very important point, and i´m glad you mentioned it. I raised that question (how would the FCPC´s behave during manual trim when not in direct law) some moons ago. Nobody followed my lead there.
RetiredF4 is offline  
Old 25th Oct 2011, 23:43
  #395 (permalink)  
 
Join Date: May 2011
Location: BOQ
Age: 79
Posts: 545
Likes: 0
Received 1 Like on 1 Post
This tumble is very possibly irrecoverable but is not deep stall.
There are two documented incidences where deep stalls were encountered. Both of the experienced test pilots indicated that the airplanes stalled at an angle of attack of about 40 to 60 degrees, and "locked" in a non-rotating descent at an almost flat angle to the horizon. Despite efforts to push the nose down by the control stick, rocking the wings and even manually attempting to manipulate the center of gravity forward by shifting their body weight forward, neither pilot was...
Clandy: The first quote is yours and the second is from an NTSB Velocity accident report. You can do a search on the NTSB site and find it in one of the 8 reports for the Velocity. You'll know which one. PAGE 1E. (No tumbling mentioned here.)

Your post was discussing 'aeroplanes'. The Velocity is an 'aeroplane'. Your claim was that 'only' T-tailed 'aeroplanes' could deep stall.

He knows that the only known way to achieve it is to include T-tail in aeroplane's design.
Followed by a claim that no Velocity aircraft ever deep stalled. I believe your words were: "Nope. It did not."

Does the Velocity stall description bring anything to mind?

Just curious, when was the last time you positioned the trim manually in a stalled A330?
OK465 is offline  
Old 25th Oct 2011, 23:51
  #396 (permalink)  
BarbiesBoyfriend
Guest
 
Posts: n/a
You know guys, if you get that THS out of position, you are going to have one hell of a job to do.

If the AP (or someone else) trims it nose up (or down) and then hands you it. What are you going to do?

I remember an incident where an inexperienced pilot got a ERJ-145 out of trim THS wise. Luckily it was nose UP. Come to think of it, 447 was nose up too. Glad the Barbie-jet pilots got out of it.

Anyways. My main point is this:

That tail needs to be pointing where you're going.

If it isn't- you're in a world of **** until it is.

Last edited by BarbiesBoyfriend; 26th Oct 2011 at 00:08.
 
Old 26th Oct 2011, 01:51
  #397 (permalink)  
 
Join Date: Sep 2010
Location: England
Posts: 1,955
Likes: 0
Received 0 Likes on 0 Posts
Clandestino,

Only in the tales of yesteryear, when swept wing was largely an unknown quantity. Since then, wise aerodynamicists came up with something called "washout". When measuredly applied, it assures that roots stall first, thereby: 1)eliminating pitch-up associated with tips that stall first 2) reducing rolling momentum from asymmetric flow separation - separation now occurs closer to CG in lateral sense 3) keeping the outboard ailerons effective.
Not quite. Even if the entire length of a wing, swept or straight, stalls at the same time the separation of the airflow occurs towards the rear of the wing causing the centre of lift to move forward. Unchecked that may cause pitch up.

Please provide reference to this statement of yours.
I can't, I have paper copies of several flight test documents on controls and stability.

Nowadays, on swept wing transports, washout reduces the effect to the point of elimination, as was tragically proven by certain crew of certain airliner that fell flatly into Atlantic without gyrations that have to be expected with tips stalling first
Did you not spot the part about several roll excursions, some up to 40 odd degrees and the 270 heading change?!

It is not the tendancy of a T-Tail to deep stall it is the recovery that can be a problem. Note that the King Air, although a T-Tail, is not prone to deep stalls, straight wing you see!
Lord Spandex Masher is offline  
Old 26th Oct 2011, 04:44
  #398 (permalink)  
 
Join Date: Jun 2001
Location: UK
Posts: 3,982
Likes: 0
Received 1 Like on 1 Post
Not quite. Even if the entire length of a wing, swept or straight, stalls at the same time the separation of the airflow occurs towards the rear of the wing causing the centre of lift to move forward. Unchecked that may cause pitch up.
Isn't this why some types have a stick nudger to ensure pitch down at the stall?
fireflybob is offline  
Old 26th Oct 2011, 05:07
  #399 (permalink)  
 
Join Date: Mar 2002
Location: Seat 1A
Posts: 8,532
Received 72 Likes on 41 Posts
Originally Posted by Lord Spandex
Even if the entire length of a wing, swept or straight, stalls at the same time the separation of the airflow occurs towards the rear of the wing causing the centre of lift to move forward. Unchecked that may cause pitch up.
The purpose of washout (look at the outboard 1/4 of a 747 wing from the rear) is to prevent this entire-wing-stall-at-the-same-time. The entire wing won't stall "at the same time", surely?
Capn Bloggs is offline  
Old 26th Oct 2011, 08:06
  #400 (permalink)  
 
Join Date: Nov 2008
Location: Crew lounge
Posts: 87
Likes: 0
Received 0 Likes on 0 Posts
Again, somewhere on the Tech Log threads at least one Airbus FBW pilot stated that F/Os train on both sides of the flight deck, so left/right-hand orientation should not have been a problem.
Sorry. Once again : prior to june 1st 2009 Air France F/Os did NOT get any, on purpose, left-seat flight training.

Some sim sessions ("true" captain not compulsory) just happenned to be with a two F/O crew but the purpose was not to train F/Os in the left seat.
GerardC is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.