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

AF 447 Thread No. 12

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

AF 447 Thread No. 12

Old 15th Oct 2014, 19:10
  #661 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by BOAC
- don't Boeings have it too?
The FBW models (777/787) do, but I believe it operates differently. Tdracer's probably the best bet to go into detail there.

Originally Posted by rudderrudderrat
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.
"Even" me? Fie, sirrah - I'm just trying to be even-handed and not get carried away. What you say may be true, but I don't have a list of all incidents where Alternate Law was activated to make a call there...

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 ... the airplane's reaction to maintain a neutral-sidestick command of 1.0g is up elevator, followed by nose-up stabilizer. ooops."
From what I've read it's not a straight g-load setup, but I'll defer to those who know better. Your A320 seems to have a safeguard against NU trim when the Stall Warning sounds, and why this isn't the case for the later widebodies is absolutely a question that should be put to Airbus.

However - in a stall recovery scenario (as opposed to the avoidance scenario), there's no way the SS should be neutral - it should be commanding ND in order to effect a recovery.
DozyWannabe is offline  
Old 15th Oct 2014, 19:37
  #662 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 65
Posts: 7,357
Received 541 Likes on 342 Posts
Dozy:
Without getting into the point of "don't stall a passenger liner, that isn't what the passengers paid for with their ticket" ...
However - in a stall recovery scenario (as opposed to the avoidance scenario), there's no way the SS should be neutral - it should be commanding ND in order to effect a recovery.
Or maybe the pilots should be doing that. So long as HAL doesn't obstruct that action, HAL will at worst be neutral, and more probably helpful with all of the features in his kit.
Lonewolf_50 is offline  
Old 15th Oct 2014, 20:06
  #663 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Lonewolf_50
Or maybe the pilots should be doing that.
Indeed - and it's the only way the SS will give that order is if the pilot commands it. I'd hoped that'd be obvious!

PS. Gums will know all about this, but I've always tended to wince a little at the "HAL" analogy. HAL was a state-of-the-art artificial intelligence (and a fictional one at that), whereas the computers used in the FBW Airbii are roughly as "intelligent" as the controller chips in your washing machine.
DozyWannabe is offline  
Old 15th Oct 2014, 20:26
  #664 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 65
Posts: 7,357
Received 541 Likes on 342 Posts
Dozy:
I find your objection rather empty.

HAL (a fictional computer meant to work on a fictional spacecraft) served the same function as an autopilot does, which is to relieve the crew of X amount of tedious functions in pursuit of their mission being. HAL is used metaphorically, as a symbol of the dangers of overreliance on automation. (Who the hell is flying the plane: the pilots or HAL?)

Early autopilots were often referred to as "George" for a reason that I think goes to slang of the time: "Let George do it" was a throwaway phrase from about the time my dad was in college, 40's - 50's. Are you going to object to that as well?

Note: overreliance on automation has come to the attention of the FAA, in a negative sense, of late ... I'll suggest to you that AF 447 is a fruit from the tree of the overreilance on automation, which seems to have at its root a non trivial number of airline company SOPs.
Lonewolf_50 is offline  
Old 15th Oct 2014, 20:46
  #665 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Lonewolf_50
HAL (a fictional computer meant to work on a fictional spacecraft) served the same function as an autopilot does...
Sure - but I thought we were talking about the EFCS, which is a very distinct concept from autoflight...

(Though autoflight computers aren't that smart either...)
DozyWannabe is offline  
Old 15th Oct 2014, 21:04
  #666 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Originally Posted by Gretchenfrage
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).
There's nothing wrong with that statement, but in the case of AF447, Bonin was aware the A/T had dropped out, and he used the thrust as and when he wanted.

The 'Elephant' in the cockpit was in fact the biggest piece of control equipment on the flight deck - the THS Wheel, which was designed to perform two functions,

(1) Give a visual feedback of the THS position, and
(2) Provide a direct means of manual control of the THS, if and when required.

I venture that (1) was not noticed, because the THS isn't/wasn't a normal part of the crew's scan, and secondly I have the distinct impression that in the cruise its movement was small and only becomes active in the non cruise sectors of a flight. Under NL that is a given, and the THS just moves to neutralize elevator demand as required (the simplistic explanation).

This brings me back to Bonin's actions with the SS, and his apparent disregard for the NU demands he was making and the NU attitude that he must have seen on the PFD. I venture that he was schooled to believe that under NL the aircraft was unstallable, and SS ANU commands would be modified by the UNoverridable protections. That being the case, it probably goes without saying that (2) never featured in his training.

However, the 'Mammoth' in the cockpit was ALT2b, where a hand on the THS at the appropriate time could have prevented most of what happened. Though, when you don't appreciate how to manage the aircraft energy at FL350+, then there is little left to be said.
mm43 is offline  
Old 15th Oct 2014, 21:08
  #667 (permalink)  
 
Join Date: Jun 2011
Location: West of Offa's dyke
Age: 88
Posts: 476
Likes: 0
Received 0 Likes on 0 Posts
Your A320 seems to have a safeguard against NU trim when the Stall Warning sounds

Dozy I have an apology and a confession to make:
You have mentioned several times recently that the A320 has a feature where NU THS movement is inhibited in Alternate when Stall Warning is triggered. Since you got that from me I thought I should check my source,and to be blunt I can't trace where I got it from. There IS a normal law feature on both A320 and A330/340 where NU THS movement is limited to the existing value when alpha protection is triggered (and THS limited to 2 deg ND); i.e. no further NU trim is allowed when there is a potential incidence problem looming, but I can't find any reference to a similar feature in Alternate (possibly worked off stall warning).


That being said, it seems to me that such a function would fill what BOAC is looking for rather well, and I can't see any particular reason or difficulty to prevent it.


Sorry!
Owain Glyndwr is offline  
Old 15th Oct 2014, 21:15
  #668 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
@OG:

No apology necessary. Regardless of where your info came from, my experience in the A320 sim (presuming that the systems behaviour was accurate) showed that NU trim was indeed inhibited shortly after we began the NU SS input. Whether that was linked to the SW or a feature of that control law on the A320, the trim nevertheless stopped of its own accord.

Originally Posted by mm43
However, the 'Mammoth' in the cockpit was ALT2b, where a hand on the THS at the appropriate time could have prevented most of what happened.
With respect, how does that work? Regardless of the THS position, Bonin was still pulling NU fairly consistently all the way down. They're at about 28,000ft at the point he releases the SS from the back stop - even with that much to play with, it doesn't leave much time for recovery...

Last edited by DozyWannabe; 15th Oct 2014 at 21:32.
DozyWannabe is offline  
Old 15th Oct 2014, 22:20
  #669 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Originally Posted by DozyWannabe
With respect, how does that work?
Well, to start with the THS wouldn't have added the extra 10° NU and any ND on the SS would have been more effective. Possibly the change may have been noticed - in time?

Ideally, the hand should have been on the THS Wheel from the moment ALT2b was established, which would mean that the implications of Alt2b and the THS behaviour were known. In this case, they weren't.

Last edited by mm43; 15th Oct 2014 at 23:17. Reason: arm not long enough to reach the THS! - added Wheel
mm43 is offline  
Old 15th Oct 2014, 23:23
  #670 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by mm43
Well, to start with the THS wouldn't have added the extra 10° NU and any ND on the SS would have been more effective.
In terms of recovery, the THS position wouldn't have made that much of a difference until sufficient forward airspeed was gained. Even if our experiment (in which the SS was held against the forward stop and only relaxed when 10deg ND attitude was attained) represents a "best-case" scenario for SS input rolling the THS back to neutral (i.e. in under 10 seconds), the situation might still have been recoverable. Even if it was only possible to recover by rolling the trim wheel forward manually, it would still take several seconds to stabilise the attitude at 10deg ND with the SS.

Possibly the change may have been noticed - in time?
Debatable (see above)...

Ideally, the hand should have been on the THS from the moment ALT2b was established.
Why? The THS position was OK at AP disconnect - it didn't start rolling back significantly until the stall was developed...
DozyWannabe is offline  
Old 15th Oct 2014, 23:43
  #671 (permalink)  
 
Join Date: Jun 2011
Location: Devonshire
Age: 96
Posts: 297
Likes: 0
Received 0 Likes on 0 Posts
The "elephant", the THS, was not noticed because it was not in the normal scan. It appears to be painted in black and white segments.

If the colours could be replaced for those areas which are not seen in cruising flight, this movement might have been noticed by one of the pilots.

I recall having seen this done on one type of aircraft, where a bright RED was used.

Last edited by Linktrained; 15th Oct 2014 at 23:46. Reason: spelng
Linktrained is offline  
Old 16th Oct 2014, 00:16
  #672 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Originally Posted by DozyWannabe
Why? The THS position was OK at AP disconnect . .
I'm not arguing that, but trying to make the point that once in ALT2b the FCS will seek to maintain 1g, and if the THS Wheel starts to move backwards (NU) the likely reason is the IAS is dropping, and/or you are holding the SS in NU position. Take your pick.
mm43 is offline  
Old 16th Oct 2014, 00:38
  #673 (permalink)  
 
Join Date: Oct 2007
Location: A better place.
Posts: 2,326
Received 26 Likes on 18 Posts
If I read that Vanity Fair article correctly, it suggested that at certain points the EFIS did not display correct values because the jet's pitch, and/or roll exceeded what it could measure.
I had assumed that like a fast jet panel, an airliner EFIS would display values even if they were way outside the recommended operating envelope.
So if you are 60 degrees nose up (extremely high alpha) will the EFIS in an airliner simply peg out at around 40 degrees nose up?
If - apologies to Tex Johnson - you flew a barrel roll in an A320 for example, or were able to fly a loop would the EFIS display flip and display inverted pitch or roll values as it does in a jet fighter, or is it simply not programmed to do this, because it's not expected, and therefore would be an unnecessary complication/and or cost to add?
tartare is offline  
Old 16th Oct 2014, 00:47
  #674 (permalink)  
 
Join Date: Dec 2012
Location: Ramona, CA
Age: 66
Posts: 38
Likes: 0
Received 0 Likes on 0 Posts
From what I've read it's not a straight g-load setup, but I'll defer to those who know better. Your A320 seems to have a safeguard against NU trim when the Stall Warning sounds, and why this isn't the case for the later widebodies is absolutely a question that should be put to Airbus.
yes, correct. It is normally a mix of g-load and pitch rate. More pitch rate at slower airspeeds, more g-load at higher airspeeds, with a crossover point where both influences are equal at an unpublished speed - probably somewhere around the greendot–250 kts range.

However - in a stall recovery scenario (as opposed to the avoidance scenario), there's no way the SS should be neutral - it should be commanding ND in order to effect a recovery.
And in Normal law it does. But this was a degraded flight control law with no protections (except g-load) left. Normal law would prevent the need for a "recovery". Due to some questionable design choices, AOA was deemed unusable with indicated airspeed below 60KIAS—so, there's nothing left to drive that pitch down except a pilot's training. (but when that pilot's training for stall recover consists only of full power application during stall training at 5,000 ft—there's not much there to help either!)
Bpalmer is offline  
Old 16th Oct 2014, 01:02
  #675 (permalink)  
 
Join Date: Dec 2012
Location: Ramona, CA
Age: 66
Posts: 38
Likes: 0
Received 0 Likes on 0 Posts
From RetiredF4:
BPalmer
The airplane is essentially point-and-go. If you pull back on the stick for a few seconds and let go, the airplane pitch will stay there.
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..
Well, yes, of course —"point and go" wasn't meant as a technically complete description as you must know. But this is still essentially correct and opposite of the behavior one would expect from an airplane that is not essentially neutrally stable.

And you are correct about the need for a pitch change to maintain that trajectory with speed change, and that's what I meant. A conventionally controlled airplane (or one with C*U) will change not only its pitch but also its trajectory to achieve speed stability vs. the Airbus trajectory/g-load stability. This is evident to Airbus pilots when flap settings are changed- the pitch will self adjust to maintain the trajectory/g-load.

But my main point with the "point and go" comment was that if you displace the ss and let go, the pitch is reasonably stable. UNLIKE speed stable aircraft which would in rather short order begin to pitch down for a speed-seeking fugoid.
Bpalmer is offline  
Old 16th Oct 2014, 01:04
  #676 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by mm43
I'm not arguing that, but trying to make the point that once in ALT2b the FCS will seek to maintain 1g, and if the THS Wheel starts to move backwards (NU) the likely reason is the IAS is dropping, and/or you are holding the SS in NU position. Take your pick.
I'll be careful in an attempt not to put my foot in my mouth, but the "maintaining 1G" (or thereabouts) applies in both Normal and Alternate laws. If I recall correctly, the A320's EFCS would interpret that more-or-less literally, and manual S&L flight would actually induce an (extremely) gradual altitude gain (on the level of a few feet per hour) under manual control. This was compensated for and corrected in the A330/340.

I digress - anyway, as I understand it, the deal with ALTN2B is that the "low speed stability" function (i.e. "soft" protection) doesn't work, which means that theoretically:
Originally Posted by Owain Glyndwr
Below 1.3Vs the aircraft is speed unstable and this is where the BEA comment to the effect that the aircraft might drift to stall even with zero stick input becomes valid.
This has nothing to do with autotrim or THS. Every point made in this regard thus far appears to be educated guesswork. The only way to know would be to take an A330-200 up and try the theory out. A level-D sim might get close to it, presuming that the emulation is correct.

Originally Posted by tartare
...will the EFIS in an airliner simply peg out at around 40 degrees nose up?
No. Not in the least. Which part of the article are you referring to?

Originally Posted by Bpalmer
Due to some questionable design choices, AOA was deemed unusable with indicated airspeed below 60KIAS
Well, hold on a minute there. If the manufacturer of the AoA vanes specifies that their output is unreliable below 60kts, how else are you supposed to implement the system? Remember that the A330 and 340 were originally certified with the Goodrich pitot tubes (never known to suffer a dual/triple failure), and the Thales AA fit was an option later in the type's lifecycle.

Last edited by DozyWannabe; 16th Oct 2014 at 01:18.
DozyWannabe is offline  
Old 16th Oct 2014, 01:33
  #677 (permalink)  
 
Join Date: Dec 2012
Location: Ramona, CA
Age: 66
Posts: 38
Likes: 0
Received 0 Likes on 0 Posts
If I read that Vanity Fair article correctly, it suggested that at certain points the EFIS did not display correct values because the jet's pitch, and/or roll exceeded what it could measure.
I had assumed that like a fast jet panel, an airliner EFIS would display values even if they were way outside the recommended operating envelope.
So if you are 60 degrees nose up (extremely high alpha) will the EFIS in an airliner simply peg out at around 40 degrees nose up?
If - apologies to Tex Johnson - you flew a barrel roll in an A320 for example, or were able to fly a loop would the EFIS display flip and display inverted pitch or roll values as it does in a jet fighter, or is it simply not programmed to do this, because it's not expected, and therefore would be an unnecessary complication/and or cost to add?
AF447 did not reach an unusual attitude at any point. The only parameter that could have triggered the unusual attitude law would have been the 45° AOA.

The EFIS is certainly capable of displaying the full range of attitudes (having done them in the simulator).

However, there is some anecdotal evidence of loss attitude display in similar incidents, detailed in my book. AF447: "we have not more usable displays" (certainly open to interpretation)—which AB says refers to airspeed and vertical speed indications since there is no good explanation why attitude should be lost in such a situation, and a Mihin Lanka flight whose pilots "reported the loss of both attitude displays on their Primary Flight Displays (PFDs) for up to 20 seconds." It is for this reason that the BEA recommended that there be some visual recording of what is actually displayed on the EFIS, instead of only recording the raw data.
Bpalmer is offline  
Old 16th Oct 2014, 03:57
  #678 (permalink)  
 
Join Date: Aug 2014
Location: Montreal
Age: 53
Posts: 71
Likes: 0
Received 0 Likes on 0 Posts
Why? The THS position was OK at AP disconnect - it didn't start rolling back significantly until the stall was developed...
Apart of SS position, why THS did start rolling back while Stall was online? What's the logic in this, if any?
_Phoenix_ is offline  
Old 16th Oct 2014, 04:39
  #679 (permalink)  
 
Join Date: Aug 2009
Location: Germany
Age: 67
Posts: 1,777
Likes: 0
Received 0 Likes on 0 Posts
If the manufacturer of the AoA vanes specifies that their output is unreliable below 60kts, how else are you supposed to implement the system?
Ok that's fine
But why stop the alarm when stall speed is well over 60 knots and certainly under 60 knots !!!!
This does not correlate with the vanes certification
What would be the point that they are certified to 30 knots or 80 knots
The fact is that 80 or 30 knots forward speed .. A330 aircraft is no longer in normal flight .. and the first thing that comes to mind .. is that this airplane is stall
Again and again .. why stop the stall alarm at 60 knots ????
Same question here:
What's the logic in this, if any?
At 61 knots .. stall alarm warn the pilot .. your plane is in stall
At 59 knots and under .. no more stall .. that's the Airbus magic .....

Last edited by jcjeant; 16th Oct 2014 at 04:54.
jcjeant is offline  
Old 16th Oct 2014, 06:29
  #680 (permalink)  
 
Join Date: Oct 2009
Location: UK
Posts: 1,270
Likes: 0
Received 0 Likes on 0 Posts
Hi DozyWannabe,
If the manufacturer of the AoA vanes specifies that their output is unreliable below 60kts, how else are you supposed to implement the system?
You simply make the stall warning invalid below 60 kts AND on the ground (to prevent false nuisance stall warnings during the early take off roll).

However, when you are airborne, if the angle of AoA sensors think you are stalled and even if you have less than 60 kts indicated forward airspeed - then you are definitely stalled. Therefore remove the 60kt logic once airborne.

@ Owain Glyndwr
but I can't find any reference to a similar feature in Alternate (possibly worked off stall warning).
A320 FCOM OP-020, Flight Controls, Alternate Law
"At the flight envelope limit, the aircraft is not protected, i.e.:
In high speed, natural aircraft static stability is restored with an overspeed warning
In low speed (at a speed threshold that is below VLS), the automatic pitch trim stops and natural longitudinal static stability is restored, with a stall warning at 1.03 VS1G."
rudderrudderrat 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.